Skip to main content

Full text of "The New Oxford History of Music"

See other formats


Ihe New ч = -a Е 
Oxford Hi istory of | 
Music % 


е 


THEMODERN АСЕ - 
1890-1960 "e 


THE NEW T: 
OXFORD HISTORY OF MUSIC | 


Editorial Board 
SIR JACK WESTRUP (Chairman) 
DR. GERALD ABRAHAM (Secretary) 3 
MARTIN COOPER DOM ANSELM HUGHES 
DR. EGON WELLESZ 


------ 


Published 
l. ANCIENT AND ORIENTAL MUSIC 
П. EARLY MEDIEVAL MUSIC UP TO 1300 
ПТ. ARS NOVA AND THE RENAISSANCE (1300-1540) 
IV. THE AGE OF HUMANISM (1540-1630) 
ҮП. THE AGE OF Е NLIGHTENMENT (1745-1790) 
X: THE MODERN AGE (1890—1960) 
іп preparation 

V. OPERA AND CHURCH MUSIC (1630-1750) 


УІ. THE GROWTH OF INSTRUMENTAL MUSIC 
(1630-1750) 


VIII. THE AGE OF BEETHOVEN (1790-1830) 
IX. ROMANTICISM (1830-1890) 


ХІ. CHRONOLOGICAL TABLES, BIBLIOGRAPHIES, 
AND INDEX 


In conjunction with Тле New Oxford History of Music, E.M.I. Records Ltd. have 
issued a series of gramophone records under the title of The History of Music 
in Sound. The Handbooks to this series are published by the Oxford University 
Press. Each volume of the series corresponds to a volume of the New Oxford History, 
the editors of equivalent volumes are in most cases the same, and Dr. Abraham is 
the General Editor. There are numerous cross-references from one series to the 
other, and the student of music will find The History of Music in Sound a valuable 
complement to The New Oxford History of Music. 


ISBN 0 19 316310 1 


NEW OXFORD HISTORY OF MUSIC 
VOLUME X 


. - 


EDITORIAL BOARD 


J. A. WESTRUP (Chairman) 
GERALD ABRAHAM (Secretary) 
DOM ANSELM HUGHES 
EGON WELLESZ 
MARTIN COOPER 


THE VOLUMES OF THE 
NEW OXFORD HISTORY OF MUSIC 


Ancient and Oriental Music 


. Early Medieval Music up to 1300 
. Ars Nova and the Renaissance (1300-1540) 
. The Age of Humanism (1540-1630) 


Opera and Church Music (1630-1750) 


. The Growth of Instrumental Music (1630-1750) 
. The Age of Enlightenment (1745-1790) 

. The Age of Beethoven (1790-1830) 

. Romanticism (1830-1890) 

. The Modern Age (1890-1960) 

. Chronological Tables, Bibliographies, and Index 


к 
У. 


СКРЯБИНА A SRRJABIN: 


rlLPOMETE?R® соч60 ор PROMETHEUS” 


% 


NAA ZA 798 
d УЗСНЕВ M 


URLIN: MOSKAU 


SKRYABIN'S PROMETHEUS, 1911 (see p. 35) 


The front cover of the score designed by the theosophist poet Jean Delville. It is 
orange, the colour of fire, and shows a huge flaming sun with an androgynous face 
(symbolizing the union of male and female) enclosed іп a *world-lyre' and sur- 
rounded by magical and cosmic symbols—stars, comets, and spiralling clouds. The 
piercing eyes ‘express the will’, according to Skryabin, and the face is surrounded 
by primeval chaos ‘from which the world-will calls everything to life’, first to 
material life, after which matter is dissolved into spirit and in an orgiastic dance 
united with God. 


THE MODERN 
AGE 


1890-1960 


EDITED BY 
MARTIN COOPER 


LONDON 


OXFORD UNIVERSITY PRESS 
NEW YORK TORONTO 
1974 


Oxford University Press, Ely House, London W.1 


GLASGOW NEW YORK TORONTO MELBOURNE WELLINGTON 
CAPE TOWN SALISBURY IBADAN NAIROBI LUSAKA ADDIS ABABA 
BOMBAY CALCUTTA MADRAS KARACHI LAHORE DACCA 
KUALA LUMPUR SINGAPORE HONG KONG TOKYO 


ISBN O 19 316310 1 


(O Oxford University Press 1974 


АП rights reserved. No part of this publication may be reproduced, 

stored in a retrieval system, or transmitted, in any form or by any 

means, electronic, mechanical, photocopying, recording or other- 
wise, without the prior permission of Oxford University Press 


PRINTED IN GREAT BRITAIN BY 
EBENEZER BAYLIS AND SON LTD 
THE TRINITY PRESS, WORCESTER, AND LONDON 


297525 


GENERAL INTRODUCTION 


ТнЕ New Oxford History of Music is not a revision of the older Oxford 
History of Music, first published in six volumes under the general 
editorship of Sir Henry Hadow between 1901 and 1905. It has been 
planned as an entirely new survey of music from the earliest times down 
to comparatively recent years, including not only the achievements of 
the Western world but also the contributions made by eastern civiliza- 
tions and primitive socicties. The examination of this immense field 
is the work of a large number of contributors, English and foreign. 
The attempt has been made to achieve uniformity without any loss of 
individuality. If this attempt has been successful, the result is due largely 
to the patience and co-operation shown by the contributors themselves. 
Overlapping has to some extent been avoided by the use of frequent 
cross-references; but we have not thought it proper to prevent different 
authors from expressing different views about the same subject, where 
it could legitimately be regarded as falling into more than one category. 

The scope of the work is sufficiently indicated by the titles of the 
several volumes. Our object throughout has been to present music, 
not as an isolated phenomenon or the work of outstanding composers, 
but as an art developing in constant association with every form 
of human culture and activity. The biographies of individuals are 
therefore merely incidental to the main plan of the history, and those 
who want detailed information of this kind must seek it elsewhere. 
No hard and fast system of division into chapters has been attempted. 
The treatment is sometimes by forms, sometimes by periods, sometimes 
also by countries, according to the importance which one element or 
another may assume. The division into volumes has to some extent 
been determined by practical considerations; but pains have been taken 
to ensure that the breaks occur at points which are logically and 
historically justifiable. The result may be that the work of a single 
composer who lived to a ripe age is divided between two volumes. The 
later operas of Monteverdi, for example, belong to the history of 
Venetian opera and hence find their natural place in volume v, not with 
the discussion of his earlier operas to be found in volume iv. On the 
other hand, we have not insisted on a rigid chronological division where 
the result would be illogical or confusing. If a subject finds its natural 
conclusion some ten years after the date assigned for the end of a 
period, it is obviously preferable to complete it within the limits of one 


vi GENERAL INTRODUCTION 


volume rather than to allow it to overflow into a second. An exception 
to the general scheme of continuous chronology is to be found in 
volumes v and vi, which deal with different aspects of the same period 
and so are complementary to each other. 

The history as a whole is intended to be useful to the professed 
student of music, for whom the documentation of sources and the 
bibliographies are particularly designed. But the growing interest in the 
music of all periods shown by music-lovers in general has encouraged 
us to bear their interests also in mind. It is inevitable that a work of 
this kind should employ a large number of technical terms and deal 
with highly specialized matters. We have, however, tried to ensure 
that the technical terms are intelligible to the ordinary reader and that 
what is specialized is not necessarily wrapped in obscurity. Finally, 
since music must be heard to be fuily appreciated, we have given 
references throughout to the records issued by His Master's Voice 
(R.C.A. Victor) under the general title The History of Music in Sound. 
These records are collected in a series of albums which correspond to 
the volumes of the present work, and have been designed to be used 
with it. 

J. A. WESTRUP 
GERALD ABRAHAM 
ANSELM HUGHES 
EGON WELLESZ 
MARTIN COOPER 


CONTENTS 


GENERAL INTRODUCTION 


INTRODUCTION TO VOLUME X 


I. THE APOGEE AND DECLINE OF ROMANTICISM: 
1890-1914. Ву GERALD ABRAHAM, formerly Professor of 


Music, University of Liverpool 
Introduction 
Central Europe 
France 
Spain and Italy 
Russia and Poland 
Scandinavia 
Britain and the United States 
The Musical Language of Late Romanticism 
The Textures of Late Romanticism 
Problems of Structure 
End of an Epoch 


II. THE REACTION AGAINST ROMANTICISM: 
1914. By GERALD ABRAHAM 


Introduction 

The Impact of Bach 

Busoni and Reger 

Non-functional Harmony 

The Evolution of Debussy’s Style 
Debussy and Ravel 

The Wider Influence of Debussy 
The New Harmony in New Contexts 
Schoenberg 


III. STAGE WORKS: 1890-1918. By MARTIN COOPER 
Central Europe 
Italy and the New Realism 
The Musical Language of Verismo 
France and the Reaction against Realism 
Russia and the Slavonic Periphery 
Bohemia 
After Verismo: Art nouveau and the Classical Reaction 
Towards New Shores 
Ballet 
Symbolist Drama 


1890- 


101 
105 
127 
140 


145 
146 
153 
159 
164 
174 
177 
182 
196 
196 
202 


viii CONTENTS 
IV. MUSIC IN THE MAINLAND OF EUROPE: 1918-1939. 


By MOSCO CARNER 208 
General Characteristics 208 
Stravinsky after the Rite 210 
Stravinsky's Neo-classicism А DIT 
The ‘Symphony of Psalms’ and Later Works 223 
Sergey Prokofyev 229 
The Older Generation in France 231 
Erik Satie and ‘Les Six’ 236 
‘La Jeune France’ 254 
Italy: La generazione dell’80 256 
Ottorini Respighi, Giorgio Ghedini, Mario Castelnuovo-Tedesco 268 
Luigi Dallapiccola and Goffredo Petrassi 270 
Béla Bartók 274 
Zoltan Kodaly 299 
Leos Janacek 301 
Other Czech Composers 309 
Karol Szymanowski 311 
Manuel de Falla 315 
Espla and Turina 317 
Later Spanish Composers 318 
Switzerland 319 
Scandinavia and Holland 321 
Richard Strauss's Last Years 322 
Pfitzner and Schreker 326 
Ferrucio Busoni 327) 
Paul Hindemith 327 
Kurt Weill and Ernst Kfenek 338 
The Second Viennese School—Arnold Schoenberg 340 
Alban Berg and Anton Webern 362 


V. MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960. 
By PETER EVANS, Professor of Music, University of Southampton 387 


Igor Stravinsky and the Rapprochement 389 
Paul Hindemith and the German Middle Generation 402 
The Middle Generation Elsewhere 419 
Olivier Messiaen 428 
The New Serialism 434 
Pierre Boulez 441 
Karlheinz Stockhausen 454 
The Avant-garde in Italy and Elsewhere 468 
Music on Tape 487 


Conservatism and Compromise 493 


CONTENTS ix 
VI. MUSIC IN BRITAIN: 1918-1960. By ARTHUR HUTCHINGS, 


formerly Professor of Music, University of Durham 503 
Introduction 503 
Bax and his Romantic Contemporaries 504 
Ralph Vaughan Williams 507 
Gustav Holst ӘЛЕ 
Peter Warlock and his Contemporaries 519 
Alan Bush 523 
Constant Lambert and William Walton 523 
Edmund Rubbra 531 
Lennox Berkeley 536 
Alan Rawsthorne 537 
Gerald Finzi 539 
Michael Tippett 540 
Benjamin Britten 543 
Elisabeth Lutyens 556 
Immigrant Composers | 557 
Humphrey Searle 563 
The Younger Generation 563 


VII. AMERICAN MUSIC: 1918-1960 
(i) MUSIC IN THE UNITED STATES. 
By RICHARD FRANKO GOLDMAN, President, the Peabody 


Institute, Baltimore 569 
The Background 569 
Charles Ives 574 
The Decade 1920-30: Gershwin and Copland 583 
Radicals of the 1920s 596 
Henry Cowell 597 
Virgil Thomson and Others 602 
International Styles—Wallingford Riegger, Roger Sessions, 

Walter Piston 607 
Roy Harris 614 
Тпе 1930s—the Impact of European Immigrants 619 
The Situation in 1950—the Decade 1950—60 622 


(1) MUSIC IN LATIN AMERICA. 

By GERARD BÉHAGUE, Professor of Music, University of Illinois 635 
Nationalism 635 
Modernism and the Avant-garde 638 


VIII. MUSIC IN THE SOVIET UNION. By GERALD ABRAHAM 639 


The Political Background 639 
The Beginnings of Soviet Opera and Ballet 643 
Early Soviet Orchestral Music 648 


Early Soviet Music in Other Fields 657 


CONTENTS 


. B 


Socialist Realism in Opera 

Socialist Realism in the Concert Hall 
Prokofyev's War and Peace 

Instrumental Music, 1941—7 

Vocal Music, 1941-7 ы 
Political Background after 1947 

Opera, 1948-60 

Instrumental Music, 1948—60 

The Avant-garde 

Music in the non-Russian Republics 


BIBLIOGRAPHY 


CONTENTS OF THE HISTORY OF MUSIC IN 
SOUND, VOLUME X 


INDEX 


660 
670 
676 
681 
684 
685 
686 
693 
699 
699 


701 


085 
737 


ІШ 


ПІ. 


ІУ. 


УТ. 


Ум. 


VIII. 


IX. 


ILLUSTRATIONS 


SKRYABIN'S PROMETHEUS (1911) THE FRONT COVER 


OF THE SCORE (see p. 35) Frontispiece 


Designed by Jean Delville. By courtesy of the British Museum. 


facing page 


. STRAUSS'S SALOME (1905). THE DANCE OF THE SEVEN 


MOS IN THE ORIGINAL DRESDEN PRODUCTION (see 
р. 1 
By courtesy of the Raymond Mander and Joe Mitchenson Theatre Collection. 


PUCCINI'S TOSCA (1900). A POSTER ADVERTISING THE 
FIRST PERFORMANCE (see p. 156) 
By courtesy of the Victoria and Albert Museum, London. 


RIMSKY-KORSAKOV'S THE GOLDEN COCKEREL (1909). 
THE STAGE SET OF ACT ONE OF THE ORIGINAL PRO- 
DUCTION (see p. 176) 

Designed by I. Bilibin. Reproduced from ‘Zolotoy Petushok’ (Moscow, 1909), 
the sovenir booklet published after the first performance. 


PALLASI THE THREE-CORNERED НАТ (1919). THE 
OPENING SCENE, SHOWING THE STAGE SET BY 
PICASSO, AND THE CHOREOGRAPHER MASSINE IN 
THE PART OF THE MILLER (ее p. 201) 

By courtesy of the Raymond Mander and Joe Mitchenson Theatre Collection. 


- (а) ИМОМЕССЕК”5 JUDITH (1925). А SCENE FROM THE 


ОВСА ОР СТОЛ АГ ТЕГЕ РЕОРГЕ 5 
THEATRE, DARMSTADT (see p. 250) 


By courtesy of the Raymond Mander and Joe Mitchenson Theatre Collection. 


(D KRENEK'S JONNY SPIELT AUF (1927). THE STATION 
SCENE FROM THE ORIGINAL PRODUCTION AT THE 
NEUES THEATER, LEIPZIG (see p. 340) 

By courtesy of the Theater-Museum, Munich. 


STOCKHAUSEN'S ZYKLUS (1959). A PAGE FROM THE 
SCORE (see p. 465) 
By courtesy of Universal Edition (London) Ltd. 


VAUGHAN WILLIAMS'S JOB (1931). A PAINTING BY 
GWEN RAVERAT OF A DESIGN FOR THE ORIGINAL 
STAGE PRODUCTION AT THE CAMBRIDGE THEATRE, 
LONDON (see p. 512) 

By courtesy of the Fitzwilliam Museum, Cambridge. 


BRITTEN 5 ВШЕХ BUDD (1952). A SCENE FROM THE 
ОСЛЕ РОО ЕЛИСОМ Atl THE ROYAL OPERA 
HOUSE, COVENT GARDEN (see p. 554) 

By courtesy of Roger Wood. 


SHOSTAKOVICH'S THELADYMACBETHOFTHEMTSENSK 
DISTRICT (1934). А SCENE FROM THE ORIGINAL 
MOSCOW PRODUCTION (see p. 663) 

By courtesy of the Raymond Mander and Joe Mitchenson Theatre Collection. 


148 


156 


176 


201 


250 


250 


465 


512 


554 


663 


ACKNOWLEDGEMENTS 


Copyright music examples in this volume are quoted by kind permission of the 
following: 


Associated Music Publishers Inc.: Elliott Carter, String Quartet (1951); Henry 
Cowell, Tiger. 


Boosey & Hawkes Music Publishers Ltd: Béla Bartók, Concerto for Orchestra, 
Microcosmos, Rhapsody for Piano, String Quartet No. 6, Sonata for Solo Violin; 
Benjamin Britten, Holy Sonnets, Midsummer Nights Dream, Peter Grimes, 
Serenade Op. 31, Sinfonia da Requiem, Winter Words; Aaron Copland, Appala- 
chian Spring, Piano Fantasy; Frederick Delius, A Village Romeo and Juliet, In a 
Summer Garden; Гео5 Janácek, Diary of One Who Vanished; Dmitry Kabalevsky, 
Violin Concerto; Vadim Salmanov, The Twelve; Dmitry Shostakovich, 2nd Piano 
Concerto; Richard Strauss, Der Rosenkavalier, Elektra, Salome, Waldseligkeit ; 
Igor Stravinsky, Movements, The Rake's Progress, Rite of Spring, Symphony of 
Psalms, Three Songs from Shakespeare, Threni. 


Bote & Bock Verlag: Francesco Malipiero, Torneo Notturno. 


Breitkopf & Hártel (London) Ltd, on behalf of Breitkopf & Hartel, Wiesbaden: 
Ferruccio Busoni, Toccata and Fugue in C Major, Arlecchino. 


Chappell & Co Ltd: Arnold Bax, Symphonies Nos. 2 & 3; George Gershwin, 
Rhapsody in Blue; Roy Harris, Third Symphony; William Schuman, Symphony 
No. IV; Arnold Schoenberg, Violin Concerto. 


J. & W. Chester Ltd: Manuel de Falla, Harpsichord Concerto; Arthur Honegger, 
Le Roi David; Francesco Malipiero, Rispetti e Strambotti; Arnold Schoenberg, 
Serenade; Igor Stravinsky, L'Histoire du Soldat, Les Noces. 


J. Curwen & Sons Ltd: Ralph Vaughan Williams, Pastoral Symphony. 


Durand & Cie, Paris: Vincent D'Indy, Fervaal, L'Étranger, Symphony No. 2; 
Paul Dukas, Ariane et Barbe-bleu, Piano Sonata (1901), Le Péri; Darius Milhaud, 
Catalogue des fleurs; Olivier Messiaen, Visions de l'Amen; Maurice Ravel, Gaspard 
de la Nuit, L'Heure espagnole; Florent Schmitt, La Tragédie de Salome. 


Editions J. Hamelle, Paris: Gabriel Fauré, Green Op. 58. 
Editions A. Leduc, Paris: Alfred Bruneau, Messidor. 
Edition Peters: Richard Strauss, Ein Heldenleben. 


Editions Salabert: Arthur Honegger, Quartet No. 3; Albert Roussel, Piano Suite; 
Erik Satie, Le Fils des étoiles, Sonneries de la Rose Croix. 


Editio Musica Budapest: Béla Bartók, In Full Flower, Op. 10 No. 1. 

Faber Music Limited on behalf of J. Curwen & Sons Ltd: Gustav Holst, Savitri. 

Franco Colombo Inc., New York: Edgard Varése, Hyperprism. 

d & Cie, Paris: Pierre Boulez, Piano Sonata No. 2; Darius Milhaud, Symphony 
(992 


Alfred Lengnick & Co Ltd: Edmund Rubbra, Symphony No. 5. 
F. E. C. Leuckart Musikverlag : Richard Strauss, A Hero's Life. 
Marks Music Corporation, New York: Roger Sessions, Quintet for Strings (1958). 


xiv ACKNOWLEDGEMENTS 


Mercury Music Corporation: Charles Ives, Anti-Abolitionist Riots, The Housatonic 
at Stockbridge, Three Places in New England, Variations on “Атпегіса” (1891); 
Leon Kirchner, String Quartet (1949). 


Merrymount Music Inc.: Wallingford Riegger, Music for Brass Choir Op. 45. 


Virgil Thomson and Mercury Music Corporation: Virgil Thomson, Four Saints in 
Three Acts. 


Messrs. Novello & Co Ltd: Edward Elgar, Symphony No. 2. 


Oxford University Press: Constant Lambert, Summer’s Last Will and Testament; 
Alan Rawsthorne, Piano Concerto Мо. 1, The Sprat, Symphonic Studies; William 
Walton, Symphony No. 1; Ralph Vaughan Williams, Symphonies Nos. 4 and 5. 


G. Ricordi & Co: Alfredo Catalani, La Wally; Pietro Mascagni, Iris; Italo Monte- 
mezzi, L'Amore dei tre Re; Ildebrando Pizzetti, Debora e Jaele, Messa di Requiem; 
Giacomo Puccini, La Bohéme, Tosca. 


Schlesinger: Alban Berg, Piano Sonata Op. 1. 
Mrs. Gertrude Schoenberg: Arnold Schoenberg, Die Jacobsleiter. 


Schott & Co: Boris Blacher, Orchester-Ornament; Karl-Birger Blomdahl, Aniara; 
Luigi Dallapiccola, Canti di Liberazione; Wolfgang Fortner, Movement for 
Orchestra; Peter Racine Fricker, Elegy, Symphony No. 2; Karl Amadeus Hartmann, 
Concerto for Viola; Hans Werner Henze, Drei Dithyramben, Piano Variations, 
Symphony No. 2; Paul Hindemith, Das Marienleben, Das Unaufhórliche No. 4, 
Kammermusik Op. 24 No. 1, Ludus Tonalis, Mathis der Maler, Philharmonia 
Concerto; Luigi Nono, Canto Sospeso, Diario polacca, Incontri; Igor Stravinsky, 
Concerto for two pianos, Symphony in C; Michael Tippett, String Quartet No. 2. 


Schott & Co on behalf of В. Schotts Söhne: Igor Stravinsky, Firebird, Scherzo 
Fantastique. 


Schott & Co on behalf of Max Eschig, Paris: Darius Milhaud, La Création du monde; 
Maurice Ravel, Alborado del Gracioso; Francis Poulenc, Le Bestiaire; Erik Satie, 
La Mort de Socrate. 


Stainer & Bell Ltd: Holst, Hymn of Jesus. 
Sonzogno, Milan: Umberto Giordano, Andrea Chénier. 


Universal Edition (London) Ltd: Béla Bartók, Dance Suite (1923) Finale; Luciano 
Berio, Circles; Pierre Boulez, Improvisation sur Mallarmé, Le Marteau sans 
maitre, Structures; Olivier Messiaen, Oiseaux Exotiques; Darius Milhaud, 
Christophe Colomb; Bo Nilsson, Zwanzig Gruppen; Nikos Skalkottas, 4th Piano 
Suite; Karlheinz Stockhausen, Gruppen, Kontrapunkte, Klavierstück 1. 


Universal Edition (Alfred A. Kalmus Ltd) Ltd: Alban Berg, Lulu, Lyric Suite, 
Violin Concerto; Alois Haba, String Quartet Op. 12; Leoš Janáček, Cunning Little 
Vixen, Glagolithic Mass, Jenüfa; Arnold Schoenberg, Erwartung, Fünf Orchester- 
stiicke No. 2, Kammersymphonie, Op. 11 No. 1, Op. 11 No. 3, Pierrot Lunaire, 
Pelleas & Melisande, Piano Pieces Op. 33 No. 1, Quartet in A, Sechs Lieder Op. 8 
No. 1, String Quartet No. 1, Variations for Orchestra; Franz Schreker, Der Ferne 
Klang; Karol Szymanowski, King Roger; Anton Webern, Cantata No. 1, Drei 
Gesänge Op. 23, Drei Lieder Op. 18 No. 1, Das Augenlicht, Orchesterstück Ор. 6 
No. 2, Passacaglia, Piano Variations Op. 27, Zwei Lieder Op. 19 No. 1, Symphony 
Op. 21. 


Universal Edition (London) Ltd on behalf of Boosey & Hawkes Inc., copyright for the 
USA: Béla Bartók, Bluebeard's Castle, String Quartet No. 4, Violin Sonata No. 1. 


Wilhelm Hansen, Musik-Forlag, Copenhagen: Carl Nielsen, Maskarade. 


INTRODUCTION TO VOLUMEX 


THE precise dating of any historical period can never be more than 
a convention or a convenience, and in the case of this last volume of the 
New Oxford History of Music the decision has been complicated by a 
further question implicit in the nature of history. If we are to date any 
historical period from the first emergence rather than the full develop- 
ment of its most characteristic features, it seemed reasonable to choose 
1890 as one term of the modern period. The terminus ad quem was more 
difficult to determine. At what remove of time does it become possible 
to see events, personalities, and individual works in that perspective 
which we call historical rather than as isolated points in our own 
experience, or markings on a map of whose exact orientation we are 
still uncertain? Those who interpret the concept of history rigorously 
argue that it is still too early to obtain a balanced and objective view 
of any musical events that have taken place in the last quarter of a 
century; and voices were in fact raised in support of making 1950 the 
end-date of the present volume. This would have been both safe and 
convenient, since it would have avoided the necessity of controversial 
selections among the phenomena of contemporary music and made it 
possible for this last volume to maintain the same strictly objective 
character as the other nine. Against this it was very strongly felt that the 
tempo of change in the third quarter of the twentieth century has been 
such that a gap of some quarter of a century between the latest events 
considered in the New Oxford History of Music and the appearance of 
the last volume would appear unreal and pedantic. Writers dealing 
with the most recent musical phenomena have naturally been obliged 
to content themselves with chronicling facts rather than interpreting or 
evaluating; and even the interval between the writing and the publica- 
tion of these latest chapters has in some cases been long enough to 
alter the perspective. This is the price that must be paid for 
continuing the narrative up to the comparatively recent past, a tribute 
to Time whose payment has fallen particularly heavily on Peter Evans 
and Richard Franko Goldman. 

If the time-span of the present volume is open to question, the sub- 
division of the seventy years under consideration may well seem even 
more arbitrary. It was determined by two main considerations: the 
emergence of an unmistakably new conception of music after the First 
World War, and the very different rate at which this new conception 
became general among composers in different countries. The head- 
waters of the New Music were in Central Europe, that is to say within a 


xvi INTRODUCTION 


quadrilateral bounded by Berlin, Paris, Milan, and Budapest; so that it 
seemed logical to regard the development of music in those countries as 
a European mainstream. On the periphery of this central area neither 
Scandinavia nor Iberia seemed to warrant separate treatment; but 
musical developments in the Soviet Union, the United States, and 
Great Britain have been rich enough in themselves, and different 
enough from each other and from those in countries of the European 
mainland, to demand individual handling. 

The movement in musical thought and practice since 1890 has been 
so fast and so complex, and the movements themselves so multifarious, 
that it is possible to make only the broadest generalizations about the 
period as a whole. Until the last ten years of the period under review, it 
might have been possible to describe the transformation which began 
to show in music after 1890 and became complete before 1920 as the 
superseding of the idea of music as a language by the idea of music as 
an architecture. In panserialism, however, architecture was itself super- 
seded by a concept nearer to that of engineering, which has in its turn 
been superseded by random, indeterminate, and aleatory principles 
which suggest, however misleadingly, a more than cursory glance in the 
direction of the higher mathematics. Even the autonomy of music, once 
regarded as possibly the most hard-won of all the principles underlying 
the New Music, has been surrendered; and the baroque rhetoric of 
Ше composer's statements accompanying Olivier Messiaen's Et vitam 
venturi saeculi or Karlheinz Stockhausen's Setzt die Segel zur Sonne 
far outshine the vague cosmic pretensions of Skryabin's Prometheus or 
Ives's Universal Symphony, which seemed so ludicrous to the generation 
that came to maturity in the heyday of neo-classicism апа 
Gebrauchsmusik. 

If it was possible to trace a definite pattern in the distribution and 
progress of the New Music between the wars, patterns since 1950 have 
changed so quickly into each other that the effect is kaleidoscopic. The 
war of 1939-45 obliterated the neat geographical demarcations which 
allotted spheres of interest—Central Europe to Bartók and Schoenberg, 
the rest of Germany and Northern Europe to Hindemith, Paris, and the 
Latin world to Stravinsky—by sending each of these composers to the 
U.S.A. When the war ended the music of Stravinsky, Hindemith, and 
Bartók was comparatively well known to musicians outside Germany, 
and isolated examples of their works were winning acceptance with the 
public; but ‘serial’ music was still a closed book to all but a tiny 
minority. When the National Socialist ban on the music of Schoenberg 
and his followers was raised by the defeat of Hitler, serialism became 
almost overnight equivalent to a certificate of opposition to National 


INTRODUCTION xvii 


Socialist ideology, and serial music was suddenly in demand in Germany 
on a large scale. But frequent performance revealed in Schoenberg's 
own music a deeply romantic, as well as a drily didactic vein and in 
Berg's a strong post-romantic impulse alien to the mood of the younger 
generation of composers, who turned their attention to the third, most 
radical, and least publicized member of the Second Viennese School, 
Anton von Webern. The discovery of Webern's music was the most 
important of the discoveries made by composers during the 1950s; and 
when Stravinsky himself admitted, on his own terms and very much in 
his own manner, conversion to serialism, it looked for a moment as 
though a firm new general pattern might be about to declare itself. 
Instead of this, panserialism produced a predictable reaction, compli- 
cated by the new presence of electronic music and by the first stirring 
of what has proved to be a strong, though by definition unorganized 
anti-intellectual movement, or ‘servile revolt’, directed against the 
élitism inherent in all mandarin art and indeed in all art that makes 
intellectual demands. 

If composers since 1920 have often seemed confused by the shifting 
interests and attitudes of the pioneers, this whole period has been 
marked by a refusal on the part of the public, at first resolute and all 
but absolute and still only cautiously yielding, to interest itself in the 
New Music. This refusal has been the more marked in that the huge 
expansion and technical improvement in the recording and broadcasting 
of music during the second half of our period has multiplied the demand 
for music to an enormous extent. Up to 1950 the New Music was accepted 
by the public in direct proportion to the number of its links with the old. 
The stumbling-blocks were two: constructivism (music as architecture) 
and atonality. That is to say, the huge majority of music-lovers every- 
where still regarded music as a language for communication and was 
not prepared to accept any idiom that could not be related, clearly even 
if remotely, to the diatonic or modal systems. This double barrier 
explains why Stravinsky, Hindemith, and Bartók achieved partial and 
cautious acceptance before Berg: Berg before Schoenberg: and 
Schoenberg before Webern. 

This reluctance on the part of the public to accept the New Music, 
in conjunction with the new demand, accounts for the extraordinary 
spate of revivals or discoveries which have been a major feature of the 
last half century in every country except the U.S.S.R., where both 
composer and consumer have been guided by authority along lines 
determined by national and ideological exclusiveness. Elsewhere the 
eclipse of Wagner, whose music embodies late nineteenth-century 
romantic ideals, was followed by revivals and reinterpretations of the 


2 


xviii INTRODUCTION 


operas of Mozart, Verdi, Handel, and more recently Bellini and 
Donizetti. The rediscovery of Vivaldi stimulated a new interest in 
baroque music generally, and Monteverdi's music has been more per- 
formed in the last twenty years than in the three centuries which have 
elapsed since his death. Medieval music, though still a minority interest, 
has made considerable progress with the general public and exercised a 
clear influence on young composers. A corollary of these revivals has 
been the rediscovery of unfamiliar or forgotten timbres, most notably that 
of the countertenor voice, the recorder, lute, viol family, and chamber 
organ. Still more remarkable has been the growing popularity of 
Oriental, especially Indian, music whose influence on composers 
(Messiaen and Boulez) began earlier. Even the over-exploited nineteenth 
century has yielded some unexpected novelties, and the few devoted and 
indefatigable champions who sponsored the apparently lost causes of 
Berlioz and Mahler have been rewarded in England by an overwhelming 
public response comparable to that which during the 1930s met Olin 
Downes's campaign for Sibelius in the U.S.A. and Sir Thomas 
Beecham's for Delius in Great Britain. 

This picture of a sharply divided musical world, in which a small 
avant-garde pioneers almost out of sight of the main body of per- 
formers and listeners, who concern themselves with musical archaeology 
and indiscriminate truffle-hunting, is something entirely new. It 
suggests a parallel with Alexandrine historicizing and eclecticism, and an 
apparent falling-off in creative vitality. It should not, however, be 
forgotten that an indeterminable, but certainly large proportion of the 
creative power hitherto employed in the arts now finds an outlet in 
science, and that the triumphs of the human spirit in the twentieth 
century are likely to be found by future generations in its scientific 
rather than its artistic achievements. In the New Music itself the 
rejection of a heteronomous, quasi-literary art in favour of autonomy 
and constructivism represented a clear rapprochement with the scientific 
ideal, and one which has become increasingly close. Meanwhile the 
gap between composer and public closes very slowly. Peter Evans 
writes of ‘the entirely new powers of discrimination’ and ‘new aural 
training’ demanded by the latest music; and although Richard Franko 
Goldman considers that public hostility may be a valuable ‘teething- 
ring’ for young composers, he also emphasizes the significant loss of 
“ап audience that demands and rejects', while Professor Evans speaks of 
works being ‘performed because no one is sure that they should not be’. 

The danger of a ‘composer’s music’ developing in a vacuum (ог a 
studio), which at one time seemed very real, has been countered by the 
appearance of the ‘servile revolt’ against intellectual effort as a form of 


INTRODUCTION xix 


élitism, and by the willingness of a number of composers to abandon 
the idea of music as a discipline and to regard it as a labile, poly- 
morphous activity comprising social protest and black humour, 
emotional 'togetherness', spiritual aspiration, and deliberate triviality 
in proportions chosen by the individual listener, who may well find 
himself also a part-time performer. This new form of 'concert', in 
which the composer provides hardly more than a ground-plan and music 
is chiefly important as a background or atmosphere, certainly owes 
something not only to oriental music and theatre, but also to oriental 
religion. It is perhaps as much a non-liturgy as an anti-concert. 

Of all men the historian should be the least tempted to prophecy; 
and since the disappearance of the old norms by which music was 
formerly judged has not yet been followed by the appearance of new, 
there is no historical precedent to tempt the drawing of a parallel. At 
present there is no sign of any diminishing interest in the music of the 
European past, and interest in non-European music is on the increase. 
The future of the anti-intellectual movement, which I have called ‘the 
servile revolt', is closely bound up with the future development of 
today's younger generation who are its chief supporters. 

Perhaps the greatest threat to music as an art comes from its increased 
availability, a slowly growing inclination to regard as a kind of piped 
Muzak first the small change of the minor baroque composers, then the 
Tafelmusik, the divertimenti, cassations and dance suites of the masters, 
and so by an easy, insensible transition the symphonies of Haydn, 
Mozart, and perhaps Beethoven. The accepted institution of music as 
a background to other activities, and even to no activity at all, plainly 
dulls the listening faculty itself, eventually blurring the line that divides 
passive hearing from active listening. The distinction between noise and 
music, already much less clear than formerly, may easily become as 
indeterminate as that which divides writing from literature, often a 
matter of opinion and nomenclature rather than exact definition. 
Certainly the future of the public concert on its present scale is un- 
certain; and although music-theatre is very much alive, it seems 
improbable that there will be any but sporadic additions to the 
repertory of works suitable for performance in opera-houses as we 
know them today. European music, which once expressed man’s idea of 
what he should be, his aspirations, now reflects what in his own eye he 
is; and it seems likely that this will continue to be true of music in the 
immediate future. 


j N 1 ҒА 
i nodi ^y ia АА: ve та ? 
Ръка сами, ШТ è enia ELM. | ah 
ЕТІЛІП [Жый \ I. THAN “4%; ha wh 


| % А Т! ГУ 
И Инк | И 
| .. 


а. МИ 
NT 


it SM, 
EET 
i нед EDU 
ES n i p {ом И 
iy 4: e КЕ 
А п} ers нис 
> 24 НЕА 

| Du LA у 

ін i dar г 
үз” >. 
coser Desde 
qum 62 
М таш ed 
ui am alive od 
ча , 


- ЗУ 


я уиубашунр а 


enr 


pA 


c. ye 
anl Яни D 
ne ing 


I 


I 


ШЕР АРОСБЕАМО DECLINE OF ROMANTICISM: 
1890-1914 


By GERALD ABRAHAM 


INTRODUCTION 


THE last decade of the nineteenth century and the first of the twentieth 
witnessed the final, though not the finest, efflorescence of all those 
musical conceptions, phenomena, and tendencies that can be subsumed 
in the general idea of romanticism: music as a record of the most 
subtle and intimate personal emotions and impressions, music as a 
rhetorical language addressed to large audiences (and, as in the cases of 
Tchaikovsky’s last works and of Mahler’s in general, the two were by 
no means mutually exclusive), music fertilized more richly than ever 
before by literature and painting, with the vast expansion of tonality, 
the complication of chromatic harmony and texture in general, and 
the richness and rarefaction of orchestration that were developed by 
the striving for wider and more refined expressive power. No doubt 
European music would have reached this condition of highly charged 
romanticism even if Wagner had never existed, but in fact Wagner had 
so completely summed up the various tendencies of musical romanticism 
and carried them to fresh heights—particularly in Tristan, Gétter- 
dümmerung and Parsifal—that it is easy and not altogether erroneous to 
regard as Wagnerian everything that is romantic in the music of the 
period. The lay public of the time certainly did, thanks to his non- 
musical cultural prestige as well as to the fashionableness of his music. 
The romantic artist saw himself as an interpreter of the transcendental, 
and consequently felt himself to be above and apart from the common 
man; the musician was the ideal romantic artist and Wagner the ideal 
romantic musician. Much that was not Wagnerian was tinged with 
Wagner and those who resisted Wagner or reacted against him did so 
very consciously. Even the fresh wind which had begun to blow from 
Russia was contaminated by Wagnerian scents in the 1890s. 

In the years immediately after Wagner's death his ideas were 
commonly misunderstood—for instance, any operatic theme associated 
with a character was regarded as a leitmotive—and the influence of his 


2 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


idiom and technique was very superficial. As Romain Rolland put it, 
Че musiciens frangais traduisaient dans le style de Wagner des pensées 
de Gounod ou de Massenet'.! But as performances of Tristan and the 
Ring outside Bayreuth became more common, filling out the imperfect 
impressions gained from scores and concert excerpts—Parsifal was 
legally restricted to Bayreuth until 1913 though illegal performances 
were given earlier, notably in America—the influence deepened. Such 
typical works as Strauss's Guntram (comp. 1893), d'Indy's Fervaal 
(1895), Schoenberg's Verklärte Nacht (1899), Chausson's Le Roi Arthus 
(1899), and Elgar's Dream of Gerontius (1900), are saturated with it. 
Even composers who fought against it, as Debussy fought against Че 
fantôme du vieux Klingsor,? partially succumbed. The Wagnerian 
empire, the supreme embodiment of musical romanticism, was at its 
heyday; like all empires, artistic and political, it bore within it the seeds 
of decay. Signs of reaction had already begun to appear and their 
development will be described in the following chapter. The present one 
is limited to a rapid survey of the European situation and a discussion 
of late romanticism and its decline. 


CENTRAL EUROPE 

Central Europe— Germany and Austria together with the non-German 
countries ruled by the Habsburgs—had been the heart-land of romantic- 
ism; consequently it was there that the crisis of romanticism was most 
severely felt. Perhaps the most obvious symptom of crisis was the 
exaggeration of antithetic moods: the brashness and exuberant virility 
which it is tempting to relate to the aggressive confidence of Wilhelmine 
Germany, and the autumnal melancholy of a culture that was drifting 
away from its foundations of religious faith and finding nothing to 
replace them. The first is most apparent in the work of Richard Strauss 
(1864-1949) but appears also in Gustav Mahler (1860-1911)--Еог 
instance in the finale of his First Symphony (completed 1888)—and in 
much of the music of their lesser German contemporaries. The second 
is much more characteristic of Mahler but also appears in many pages 
of even the young Strauss of the 1890s and dominates the last works of 
Brahms (d. 1897): the Clarinet Quintet, the piano pieces Op. 116-19, 
and the chorale preludes. (Although Brahms resisted the techniques of 
late romanticism, he was very sensitive to its spirit.) One finds both in 
the third leading representative of the Strauss-Mahler generation, Hugo 
Wolf (1860-1903), whose achievement seems so much less than theirs 

1 Musiciens d'aujourd'hui (Paris, 3rd edition, 1908), p. 243. 


* Letter to Chausson, 2 October 1893, ‘Correspondance inédite de Claude Debussy et 
Ernest Chausson', Revue musicale, vii (1925—6), p. 120. 


CENTRAL EUROPE 3 


only because it was mainly confined to the miniature form of the Lied! 
and terminated untimely by his mental collapse. 

Beside Brahms, other older figures still played an important part in 
the musical scene: his friend, the more eclectic and progressive Dvorak 
(d. 1904), the Wagner-worshipping Bruckner (d. 1896) whose sym- 
phonies are very much less influenced by Wagner than they appeared to 
his contemporaries who knew them only in versions doctored by well- 
meaning admirers, and such representatives of genuine conservatism as 
Josef Rheinberger (d. 1901) and Max Bruch (who lived on till 1920). 
For some time these, even Brahms, seemed to have no artistic progeny 
of any importance; even the other progressive composers of the 
Strauss-Mahler-Wolf generation (Emil Nikolaus von Reznicek, Eugen 
d'Albert, Georg Schumann, Max von Schillings) were very secondary 
musicians, distributaries rather than tributaries of the Wagnerian 
mainstream. Only Hans Pfitzner (1869-1949), a Wagnerian but a very 
conservative Wagnerian who in later years became a pugnacious 
defender of conservatism, and his older contemporary Engelbert 
Humperdinck (1854—1921) stand out of the ruck—the latter mainly on 
the strength of a single work, Hänsel und Gretel. Most of them were 
first and foremost opera-composers and their work is dealt with else- 
where in this volume.? 

In the climate of the period it was natural that every progressive 
musician should think first of writing for the stage and that even his 
instrumental compositions should be saturated with extra-musical ideas. 
Absolute music and its ideal medium, chamber music, were abandoned 
to the conservatives. In the orchestral field the dominant figure was 
Richard Strauss, who had built his reputation on programmatic works— 
Aus Italien, Macbeth, Don Juan, Tod und Verklürung—and extended it 
during the 1890s with four more tone-poems: Till Eulenspiegels lustige 
Streiche (1895) and Also sprach Zarathustra (1896), Don Quixote (1897), 
and Ein Heldenleben (1898).3 With these scores, works of extraordinary 
inventive exuberance and virtuosity of scoring, he established himself 
as the outstanding German composer of the end of the century. The 
thematic orchestral polyphony of the mature Wagner style is not only 
employed with complete, confident mastery but with heightened 
brilliance and with a lightness and vivacity rare in Wagner and the 
Wagnerians. The range and sweep of a melody such as the opening of 
Ein Heldenleben, which does not really draw breath until the seventeenth 


1 Wolf's songs and instrumental compositions, which date mainly from the pre-1890 
period, are discussed in Vol. IX. 

? See chap. 3. 

3 Strauss's symphonic poems, as such, are discussed in Vol. IX. 


4 THE APOGEE AND DECLINE ОЕ ROMANTICISM: 1890-1914 


bar, would be as difficult to match in earlier music as would the incessant 
battering rhythm of the *battle' section, where rhythm for its own sake 
anticipates the admittedly more subtle rhythmic triumphs of a decade 
or so later. If the structures of Zarathustra апа Heldenleben are too 
sectional to be satisfactory, they are more daring than those of any 
earlier symphonic poems, and the conceptions of programmatic 
variations in Don Quixote and condensed symphony in the later 
Symphonia domestica (1903) were equally striking. And music was no 
longer only a language of emotion; it was now, more than with any 
earlier composer, a language not capable of describing concrete objects 
but offering acceptable symbols for them. Only in his normal harmony 
did Strauss usually fail to overtrump his contemporaries or even his 
immediate predecessors: when he did so, as at the end of Zarathustra 
and in some passages of Elektra, it was for a special effect of 
shock: 


Tx 
ELEKTRA, Act I 
a ia > FLI 
y Г = 
ET 4 Е ЕР 
І я | 7 
Wind IT т 
pot cmo Ж Башын > |) n. 1 
ЕЗ == ГР ү = Тее 
Тюз ПІ Рр Ppp 
f FIRST MAID Е КЕР 
/ ста KK ЕЕ ЕН 
Е а 9 = |) be 
(das ewige)Blut des Мог - - des von der Die - le (abspülen) 


dc ee с © 
0 кеу IEEE теле ет 
я: 5 (2 AR.» be Ea ba — RES BET SS) A 
НЕ НЕЕ: Е ЕЛІ 
траш ШЕ P 


——=== 


dei a a ERES 


s. [Ж-а be eat РЕ == 
б=т == ре ра о реа Це = 
p Sf yz 2p { 
аға | Bp — = == 
тр P dc pom 


([wipe the eternal] blood of murder from the floor) 


A typically rich-sounding passage, such as that at figure 24 in Ein 
Heldenleben, consists of four melodic strands: 


CENTRAL EUROPE 5 


Bes 42 
p» | 
el 


embroidered on a solid and, for the period, unadventurous harmonic 
background: 


Doe eea 
x 


[ж ge —— ГЕ 


filled in by clarinet tremolos, horn chords, and the tremolo of divided 
violas, with harp or trombones at points of emphasis. The points where 
the part-writing sometimes momentarily breaks out of the harmonic 
frame, as at the starred B in bar 3 of Ex. 2, pass unnoticed because of the 
difference of timbre. 

During the period of Till, Zarathustra, and Heldenleben, Mahler pro- 
duced three symphonies, no. 2 in C minor (1894), no. 3 in D minor 
(1896) and no. 4 in G (1900), equally involved with extra-musical 
conceptions but in a very different way. Strauss was an extrovert, 
technically self-confident, and fascinated by the musical depiction of 
characters and events; he was a born dramatist and as his powers 
matured he passed in Don Quixote from brilliant and sympathetic 
external characterization to something more profound, that inner sym- 
pathy with another character which is the mark and indispensable 
property of a true dramatic composer. Conversely, when he attempted 
avowed autobiography in Ein Heldenleben and the Symphonia domestica, 
he dramatized himself—and his wife—without revealing his inner 


6 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


self; these works are over-concerned with external circumstances. 
Although his first essays in opera, Guntram (1893) and Feuersnot 
(1901), had not been particularly successful, he must now have recog- 
nized that the dramatic and descriptive powers of purely instrumental 
music could be carried no further and, with the single exception of the 
superficial and pictorial Alpensymphonie (1915), he abandoned it for 
his true métier, musical drama. On the other hand Mahler, despite his 
life-long occupation with the conducting of opera, culminating in his 
directorship of the Vienna Court Opera from 1897 to 1907, never wrote 
an opera at all—other than a youthful Ernst von Schwaben, which he 
destroyed.! His most dramatic work is the cantata Das Klagende Lied, 
which dates substantially from 1880.? A profound and complicated 
introvert, he composed hardly anything that was not autobiographical. 
With all his experience as a conductor, he lacked self-confidence as an 
orchestral composer, as is manifest from his innumerable revisions and 
other changes of mind and by the nervous meticulousness of his detailed 
instructions for performance. 

Mahler's symphonies are no more absolute music than Strauss's 
tone-poems but they are at the same time both more and less explicit. 
Whereas with Strauss the external world nearly always plays a vivid 
part in his conception, with Mahler it exists only as indeterminate 
‘nature’, symbolized by bird-song and cowbells and posthorn calls, as a 
setting for the poet's subjective brooding. For his First Symphony, a 
purely orchestral composition, Mahler had written, and then suppressed, 
a literary programme ;? in the three symphonies of the 1890s he found it 
necessary, as Beethoven had done in the Ninth, to call in a poet's 
words. As he wrote to Arthur Seidl in 1897* *When I conceive a great 
musical creation, I always come to the point where I must draw on the 
“word” as bearer of my musical idea.’ We have it on his own authority? 
that the first movement of the Second Symphony is a funeral ceremony 
for the hero of the First—who 15 perhaps his own youth; that the 
second movement, innocent as Schubert or Dvořák, is a happy recollec- 
tion of a sunny day in this hero's life; that the third represents a return 
to the senseless and repellent hurly-burly of everyday life (it is based on 
his song-setting of ‘Des Antonius von Padua Fischpredigt’ from Des 
Knaben Wunderhorn, just as the First Symphony is partly based on 
themes from his Lieder eines fahrenden Gesellen); and that the finale is 
an answer to ‘the great question “Why have you lived? Why have you 

1 Gustav Mahler, Briefe, 1879—1911 (Berlin, Vienna and Leipzig, 1925), p. 8, n. 1. 

? See Donald Mitchell, Gustav Mahler: The Early Years (London, 1958), pp. 152-3. 

3 Printed іп full in Paul Stefan, .Gustav Mahler (Neue, vermehrte und veränderte 


Ausgabe; Munich, 1920), p. 113. 
* Briefe, p. 228. 5 Ibid., pp. 188-9. 


CENTRAL EUROPE y 


suffered? Is that all merely a huge and frightful joke?” ’. The answer is а 
grandiose choral setting of a slight adaptation of Klopstock’s Resurrec- 
tion ode, *Auferstehen, ja auferstehen’, which Mahler had just heard 
sung at Hans von Bülow's funeral, to which the transition from the 
bitter humour of the third movement is made through another Wunder- 
horn poem, ‘Urlicht’, sung by a solo voice, one of the most Brucknerian 
things Mahler ever wrote. He had already, with this distension of both 
the form and media of the symphony—the Third is scored for an 
enormous orchestra—reached the point when he could say, 'Symphony 
means to me the building up of a world with all the technical means 
available' (“Міг heisst Symphonie: mit allen Mitteln der vorhandenen 
Technik mir eine Welt aufbauen.) Mahler regarded his Third Sym- 
phony, to which the Fourth may be considered an epilogue, as “а 
musical poem embracing all the stages of development step by step. 
It begins with inanimate nature and rises to the love of Сой.” It was 
originally conceived with a title derived from Nietzsche, “Меше 
fróhliche Wissenschaft’; later it became “Рап, not only as the god's 
name but as the idea of ‘allness’; then again it became ‘A Summer 
Morning's Dream', the big initial movement in sonata-form being 
called “Рап awakes; the summer marches іп” and followed by shorter 
ones: ‘What the flowers in the field tell me’, ‘What the animals in the 
wood tell me' (based on one of the earliest of his Wunderhorn songs, 
*Ablósung im Sommer’), ‘What Man tells me’ (the midnight song from 
Nietzsche’s Zarathustra, which Strauss was suggesting in his tone-poem 
almost at the same time, but composed by Mahler for alto solo), ‘What 
the angels tell me’ (“Es sungen drei Engel’ from the Wunderhorn, set for 
alto and boys’ choir), and ‘What love tells me’ (an Adagio for orchestra 
only). But these titles were suppressed before publication, leaving the 
naive listener with an impression of a heterogeneous suite not even 
unified by a key-scheme, since the movements are respectively in D 
minor, A major, C minor, D major, F major, and A major. (Mahler 
once suggested that his unconventional key-schemes were motivated by 
the pursuit of innovation for its own sake; nevertheless they underlie all 
but two, nos. 6 and 8, of his later symphonies.) The exquisitely naive 
finale of the Fourth Symphony, yet another Wunderhorn setting, was 
to have been the penultimate movement of the Third: ‘What the child 
tells me’, and the Symphony is, for Mahler, remarkably free from darker 
shades; only in the second movement does Death with his fiddle (a solo 
violin tuned up a tone) cast his shadow—and by no means over the 
whole of that. 

These three symphonies, naive in almost everything but their irony, 

1 Tbid. p. 161. 


8 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


gangliated in structure, closely akin to song even when they do not 
actually break into it, are much more truly romantic both in musical 
substance and in their programmes than the symphonic poems of 
Strauss, where beside the genuine romanticism of such wonderful 
passages as the death of Don Quixote one finds much that is only 
superficial romantic gesture and the exaggeration of romantic language 
in conjunction with a realism (Quixote's bleating sheep, the battle- 
section in Heldenleben) that, as in opera of the period, is downright 
anti-romantic. And after the turn of the century, when Strauss soon 
withdrew from the symphonic field while his rival went on to produce 
first a trilogy of purely instrumental symphonies, no. 5 (1902), no. 6 
(1904), and no. 7 (1905), then a gigantic totally choral symphony, no. 8 
(1907), and yet another for orchestra alone, no. 9 (1909)—as well as the 
torso of a Tenth—Mahler's position as the leading Central European 
symphonist was uncontested. He had markedly developed. In the three 
symphonies of 1902-5, which are as definitely related to each other as 
the four Wunderhorn symphonies, he not only found it possible to 
dispense with ‘the word’; they are—particularly nos. 6 and 7—much 
less disconcertingly naive, more closely knit; the orchestral polyphony 
rivals Strauss's in its mastery while the massive effects (above all, in the 
finale of the Sixth and opening of the Seventh) recall the granite monu- 
mentality of Bruckner. Only the Adagietto of no. 5 seems to belong 
to the earlier group. But the music is no less romantic in essence—the 
funeral-march first movement of no. 5 with the Wagnerian cello 
cantilena of both this and the slow sections of the following Allegro, the 
autobiographical suggestions of no. 6 (which according to his widow! 
has elements of a symphonia domestica) with the hammer death-blows 
of its finale, the three ‘inside’? movements—the two pieces of ‘night 
music’ and the ‘shadowlike’ scherzo—no less subjective and secretly 
programmatic. The Eighth Symphony, monumental both in proportions 
and in the forces employed, also struck out in a new direction despite 
an obvious affinity with the finale of the earlier giant, no. 2; a symphony 
іп two movements, one а setting of ‘Veni creator spiritus’, the second 
(embracing elements of Adagio, scherzo, and finale) of the closing scene 
of Goethe's Faust, employing eight vocal soloists, double chorus, and 
boys' choir, as well as an orchestra of unprecedented size, demanded 
and was given a more impressive power of coordination than the 
earlier, partly vocal symphonies. This was the last of the colossi of 
musical romanticism, a line stretching from the outdoor compositions 


1 Alma Maria Mahler, Gustav Mahler: Erinnerungen und Briefe (Amsterdam, 1940); 
English trans. by Basil Creighton (London, 1946); enlarged and revised edition (London, 
1968), p. 70. 


CENTRAL EUROPE 9 


of the French Revolution through Berlioz, and Mahler followed it with 
two works which express with exceptional poignancy the autumnal 
mood and sunset colouring of musical romanticism itself: Das Lied von 
der Erde (1908), a song-cycle for alto, tenor, and orchestra that he 
himself styled a ‘symphony’, and the closely related but purely orchestral 
Ninth Symphony. In sharp contrast with the Eighth, these last scores 
are remarkable for the almost chamber-musiclike finesse of the orches- 
tral writing, a finesse already adumbrated in the accompaniment of the 
cycle of Kindertotenlieder (Dirges for children) (1904) and the second 
Nachtmusik of the Seventh Symphony. 

In this as in the influence of Bach, which is apparent in the fugal 
elements in the finale of his Fifth Symphony and the feats of invertible 
counterpoint in the Eighth, Mahler showed himself sensitive to the 
spirit of the decade, for of the Central European composers born in the 
seventies who came to the fore after the turn of the century, the two 
most important—Max Reger (1873-1916) and Arnold Schoenberg 
(1874-1951)—both were under the spell of Bach and both were 
attracted to chamber music. This was a generation of reconcilers; the 
days of partisanship were coming to an end; musicians could, and 
commonly did, admire both Wagner and Brahms. And often Dvorak 
as well: Schoenberg's early D major String Quartet (1897) shows particu- 
larly in its texture and in the opening subject of the finale that he too 
came under the influence of the Czech master. Schoenberg's leaning to 
Brahms was strengthened by his teacher (and later brother-in-law) 
Alexander von Zemlinsky (1872-1942), who was also drawn to chamber 
music though he became better known as a composer and conductor of 
opera. Franz Schreker (1878-1934) began as a Brahmsian and evolved 
into a full-blooded eclectic, though his really notable works, mainly 
operas, appeared only just before the First World War, while the more 
conservative Franz Schmidt (1874—1939) was even slower in finding a 
creative personality. The young Hungarian Brahmsian Ernó Dohnányi 
(1877-1960) and Dvorák's two most talented pupils, Vitézslav Novak 
(1870-1949) and Josef Suk (1874—1935), all developed into complete 
eclectics, sensitive to Strauss or Tchaikovsky, following the will-o’-the- 
wisps of impressionism or nationalism, and thus freely and often 
eloquently employing the /ingua franca of late romanticism without 
extending it. The older and far more original Leoš Janáček (1854-1928) 
remained in obscurity throughout this period; even his most successful 
opera Лей pastorkyňa (better known as Jenůfa), was performed only in 
the provincial city of Brünn (Brno) in 1904, and then forgotten until 
1916, when it was followed by all his major works.! 

1 See pp. 179-82 and 301-4. 


10 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


Some of the slightly younger men, Béla Bartók (1881—1945) and 
Zoltán Kodály (1882-1967) in Hungary, and Schoenberg's disciples 
Anton von Webern (1883-1945) and Alban Berg (1885-1935) in Austria, 
were much more significant figures, still romantics when they first 
appeared on the scene; Bartók's Kossuth Symphony (1903), for instance, 
was very closely modelled on Ein Heldenleben and the Portraits, 
Elegies, and Dirges (1907—10) were autobiographical in essence, even 
though the nature of his musical language was changing. These were the 
composers who, with Schoenberg and Reger, were to succeed Mahler 
when he died and Strauss when he began to repeat himself. It is sympto- 
matic that, unlike most of the true romantics, none of them was in a 
hurry to write his first opera and that, when they did, both Bartók's A 
kékszakállú herceg уйға (Duke  Bluebeard's Castle) (1911) and 
Schoenberg's Die glückliche Hand (1913) were short and highly experi- 
mental works, reaching out in entirely new directions.! Berg and 
Kodály—ultimately more important as a teacher and stimulator in his 
native country than as a creator—were even later in entering this field, 
while Reger and Webern never entered it at all. On the other hand they 
all cultivated chamber music, both in the usual sense of instrumental 
ensemble music and in the wider one of intimate media generally, 
including solo piano music and the solo song with piano. 

A great chapter in the history of chamber music closed in 1895 with 
Brahms's compositions for clarinet and other instruments and Dvorak’s 
last quartets. Wolf might have opened a new one if he had remained sane, 
but Strauss, who turned away from chamber music and piano music 
after his youth, and Mahler, who was never interested in either, were too 
intoxicated with the possibilities of the post-Wagnerian orchestra to 
find modest, more or less monochrome media congenial. It is true that 
Strauss, unlike Mahler, continued to write songs with piano after his 
youth; indeed most of the songs by which һе is best known—‘Cacilie’, 
*Heimliche Aufforderung’, ‘Morgen’, “Traum durch die Dammerung’— 
date from the 1890s, while the songs of the new century, from Op. 48 
onward, include such masterpieces as ‘Blindenklage’, Op. 56, по. 2.2 
That they should show the influence of Wolf in their often declamatory 
voice-parts and their rich, almost more important piano-parts was only 
to be expected. No important German song-composer of the period, 
neither Reger nor Schoenberg nor that gentler talent, Joseph Marx 
(1882-1964), a latter-day Robert Franz, completely escaped Wolf's 
influence. They might begin song-writing in emulation of Brahms or 
the earlier nineteenth-century masters of the Lied, but they inevitably 


1 See pp. 202-3 and 206. 
? Recorded in The History of Music in Sound, x. 


CENTRAL EUROPE 11 


passed deeply through the experience of Wolf before they found their 
true selves, Strauss through the sheer exuberance of his creative per- 
sonality, Reger in the series of Schlichte Weisen begun in 1904 by 
deliberate simplification. Reger's ‘Maienbliiten’, Op. 66, по. 5, and the 
opening of his *Morgen', Op. 66, no. 10 (1902), show his debt, even in 
maturity, to Wolf: 


Ех. 4 espress. 
Ziemlich langsam, mit innigstem Ausdruck [nie schleppend] p= 


() 
ET. : be уктуг ык ар, 
Se SSS ола ааа 
НЕЕ гез "есте сз) 


49 "€ БП в» # 


ер el 
ic odo oc" —sSLA- 1 


-Ж-т 
Ducum 
Bi (езтн РЖ керле 

эи л сызу с age. 77, л. Ии, шш 


(And tomorrow the sun will shine again, and on the way where I shall go.) 


12 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


Reger's greater harmonic sophistication, compared with Strauss's in the 
latter's much better known composition of the same poem (1894), is 
again apparent in the settings of Richard Dehmel’s ‘Waldseligkeit’ 
which both composers made in 1901: 


Ex. 5 (i) STRAUSS (Ор. 49, No. 1) 


ETE um 
ВИ ARR E. Hemd eee Tl 


zart p m 
Фр molto legato 


EXE ЕЕ 


чча 
| 


13 


CENTRAL EUROPE 


> ES 4 MS | 
У el MM i 
| n | INI 


14 
ELLUL 
| 


pl - 4,5: 
I je Msc Ұңғы зы ul 
г n 


pud 
as eee 2228160 177 
rau - schen, 
з] 20 
=. 
К 
ld — 14 
= 
= 
= 
o 
< 


[== 

[ж == 

Ее 
4 


eT 
РЕ В 
pmi 


КҮН cum ue c mcns ------- 2-1 
В 
be-ginnt zu 
=! bs 
р Я \! Ф. 1b «2 
-e 
—— 
FPEM 
та assai Lm 


D) 
sempre con Pedale, 


2p 


Der Wald 
HH 
H 


Ausserst zart, ausdrucksvoll (doch nie schleppend) 


je 
== 


CA ГР д 

tay? D 57 

газове Ria 
f) 

= =m 

[54 155^ 

iy 2-1 


(ii) REGER (Op. 62, No. 2) 
assai 
delicato 


14 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


f) mL Ooo 
S E 
А ИЕ aay eer у E- 
шара ыста а Г. 
I > Б LJ [2—4 Е в 


| 


Bau - men nahi die - Nacht; 


; ЕЕ ERU 
RP је == 5 
14141479 ү» 


(The wood begins to rustle, night draws near the trees;) 


As an instrumental composer Reger, pupil of the musical polymath 
Hugo Riemann, began his career with two violin sonatas, Opp. 1 and 3, 
a trio, Op. 2, and a cello sonata, Op. 5, all published before he was 
twenty. He first attracted wider notice through a Suite in E minor for 
his own instrument, the organ (1895), dedicated “То the Manes of 
Johann Sebastian Bach' which includes a highly ingenious fugue, a 
canonic intermezzo, and a baroque passacaglia. In one fundamental 
respect, his view of music as a craft rather than a language of sub- 
jective expression, Reger was not a romantic at all; he was a leader 
in the reaction against romanticism, and his work will be considered 
under that aspect in the following chapter. But as an artist of his time 
he could not escape romanticism in the form of Wagner's harmonic 
heritage, for he was one of those who could accept Brahms and Wagner 
--апа also Strauss, whose orchestral technique was not without its 
effect on his own. His conception of Bach, common to his contem- 
poraries, was a romantic conception and his own organ compositions 
on chorales, above all the three Phantasien, Op. 52 (1900), are highly 
romantic, not to say programmatic. The C major Violin Sonata (1903), 
the first movement of which spells out his enemies as ‘sheep’ and “аре” 
(Eb =S, C, Bh = H, A, Е, E and А, Е, Е, E), is as frankly auto- 
biographical as Ein Heldenleben. Nor was the composer of such 
orchestral works as the G major Serenade (1906), which so impressed 
the young Prokofyev, the Sinfonischer Prolog zu einer Tragódie (1908) 
and the Vier Tondichtungen nach Arnold Bócklin (1913), a pure crafts- 
man, certainly not a simple one. Even his chamber music, which is 
greater in bulk than his orchestral and organ music put together, is 
sometimes romantic in the broader sense; Guido Bagier! justly speaks of 


! Max Reger (Stuttgart and Berlin, 1923), p. 256. 


CENTRAL EUROPE 15 


the C minor Piano Quintet, Op. 64 (1898) as “а masterpiece of Sturm und 
Drang'. He goes on to describe it as “ап uncompleted symphony in the 
guise of chamber music’. Reger never did write a symphony; the 
Sinfonietta (1905), originally conceived as a serenade and scored for 
what at the beginning of the century was a ‘small’ orchestra, that is, 
one of classical size with the addition of a harp, was a failure and the 
Sinfonischer Prolog, the first movement of a projected symphony, was 
left on its own. As Bagier puts it, ‘each chamber-music work is a 
preliminary step toward this goal'. But chamber music was a much 
more congenial medium? and Reger's intensive cultivation of it accords 
with this basic preference for the forms of absolute music. 

Yet the earliest major work of Schoenberg, a more fundamental 
romantic then Reger, was a bold essay in programmatic chamber 
music, a string sextet inspired by а poem of Richard Dehmel, Verklärte 
Nacht (1899). Schoenberg knew his Wagner really thoroughly at an 
early age, much earlier than Reger had done, and as his next composi- 
tions show he was also familiar with both Mahler and Strauss. The 
Gurrelieder, poems by Jens Peter Jacobsen? set for colossal vocal and 
instrumental forces, were composed by May 1901 though the instru- 
mentation was not completed till ten years later; the work would hardly 
have been planned in that form without the Mahlerian precedents, 
although the Eighth Symphony was still to come. The symphonic 
poem, Pelleas und Melisande (1903), too, was clearly written in emula- 
tion of Strauss, who actually suggested the subject to him, though for 
operatic treatment: it portrays a series of episodes from Maeterlinck's 
play, with themes for Golaud, Mélisande, and Pelléas (the two latter 
combined at against a picturesque background when Mélisande 
lets fall her hair from the tower-window); the dense orchestral poly- 
phony, often bursting out of the harmonic framework, is very 
Straussian and the climax of the love-scene from onward is 
closely related to the love-music of Ein Heldenleben (cf. Exs. 2-3). On the 
other hand, Mélisande’s playing with the ring |6) is painted in a Mah- 
lerian Lündler. The sequence of Schoenberg's early chamber compositions 
is equally significant in a different way. First after Verklürte Nacht came 
the String Quartet in D minor, Op. 7, which was published as “Хо. Г 
(1905), a thematically dense score in which, it has been claimed without 
too much exaggeration,‘ ‘there is not an inside part or a figure that is 
not thematic’, then the Kammersymphonie, Op. 9 (1906) for fifteen solo 

1 Tbid., p. 266. ? Witness the String Trio, op. 776 (1904), of which the slow movement 
is recorded in The History of Music in Sound, x. 

3 Translated from the Danish by R. F. Arnold. 


* Egon Wellesz, Arnold Schónberg (Leipzig, 1921), p. 101; English translation by W. H. 
Kerridge (London, 1925 and 1971). 


16 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


instruments, some of its material again oddly Straussian in outline— 
Strauss distorted by the whole-tone scale. In both of these, as indeed in 
Verklürte Nacht and Pelleas und Melisande, the elements of the four- 
movement sonata are compressed into a single, very long movement, 
scherzo and slow movement being inserted before and after the main 
development, and reprise understood in the freest sense. But whereas this 
was the complete antithesis of Mahler's practice of expansion, in the 
Second Quartet (F sharp minor, Op. 10) (1908)! Schoenberg, like 
Mahler, ‘drew on the “ууога” as bearer of the musical idea’; in the 
third movement, adagio con variazioni, the soprano sings Stefan 
George's ‘Litanei’, in the finale the same poet's ‘Entriickung’ with its 
peculiarly apposite first line, ‘Ich fühle Luft von anderen Planeten’ 
(‘I feel air from other planets’). And, as Dika Newlin has pointed out, 
‘the spirit of Mahler hovers even more persistently over the scherzo, 
with its bizarre quotation (in the trio section) of “Асһ, du lieber 
Augustin”. ? The dedication of the Harmonielehre (1911) speaks for 
itself; in later years too Schoenberg often ‘expressed indebtedness to 
Mahler? and during 1907-8 their personal relationship was at its 
closest. Yet there were profound differences, even greater differences 
than between Schoenberg and Reger, in whose music he also ‘admired 
many things’. Mahler’s art was ап end, Reger's seemingly a dead end, 
Schoenberg’s at once an end and a beginning. The upward-leaping horn 
theme in perfect fourths at the beginning of the Kammersymphonie is, 
taken by itself, more anti-tonal than anything in Mahler or Reger, who 
took wild liberties with tonality but always left its pillars standing. And 
here, too, the Schoenberg passage stands between buttresses of F major 
and E major. But in the finale of the Second Quartet Schoenberg arrived 
at full atonality. 

With the fanatical courage that was the predominant strain in his 
character, he did not shrink from the consequences and turn back. 
Atonal music presented many problems, of which the most difficult 
was that of structure in general and the construction of extended 
instrumental compositions in particular. Hence the works of the next 
few years which lean on drama—the monodrama Erwartung (1909) and 
the music-drama Die gliickliche Hand (1913)—and those with poetic 
texts, the George song-cycle Das Buch der hängenden Gärten (1909), the 
Maeterlinck Herzgewdchse (1911), the melodrama Pierrot lunaire (on 
Otto Erich Hartleben's translation of Albert Giraud's poems) (1912), 
and the four songs with orchestra, Op. 22 (1913-14)—are more assured 


! First movement recorded in The History of Music in Sound, x. 
2 Bruckner—Mahler—Schoenberg (New York, 1947), р. 235. 
3 Ibid., pp. vii and 241-2. abid. p. 275. 


CENTRAL EUROPE 17 


than the purely instrumental pieces, the Drei Klavierstücke, Op. 111, 
and Fünf Orchesterstücke, Op. 16,? of 1909, epoch-making as they were, 
and the Sechs kleine Klavierstücke, Op. 19, of 1911. This group of 
compositions, in which both the romantic conception of music as a 
language of expression and the musical techniques of romanticism are 
carried to their ultimate extremity, was written under the sign of the 
avant-garde movement that was leaving its mark on all forms of 
European art at the time: expressionism, the antithesis of impression- 
ism, the morbidly intense expressive content of concentrated particles of 
speech or line or sound, content with the minimum of matter and form. 
Representational painting was reduced to geometrical relationships 
such as the cube, dramatic characters to dreamlike abstractions from 
the subconscious like “Тһе Man', “Тһе Woman', and “Тһе Gentleman' 
in Die glückliche Hand.? Тп such music as Op. 11, no. 3, each thematic 
idea was self-sufficient, unrepeated, not balanced symmetrically by a 
following phrase; repetition was at all tolerable only in the form of 
constructive, developing variation. With tonality mortally strained by 
late-romantic techniques and undermined by those of impressionism,* 
chords had finally lost their functional significance. With the *emancipa- 
tion of dissonance',* which Schoenberg adumbrated in his Harmonie- 
lehre,® cadence no longer had meaning as a point of relaxed harmonic 
tension. Atonal music was, in fact, also athematic and, apart from the 
supports of text or drama, nearly amorphous.? The crucial element in 
Pierrot lunaire was not the Sprechstimme of the reciter, speaking at 
approximate musical pitches, which attracted so much attention at the 
time—it had already been used in the Gurrelieder and Die glückliche 
Hand—but the means by which Schoenberg attempted to counter 
amorphousness: the devices of canon and inversion which were to lead 
him to the note-row and twelve-note serialism. Yet the decadent, sado- 
masochistic subject of Pierrot lunaire is, as Stuckenschmidt has pointed 
out,? typical of late romanticism, with origins reaching back to Parsifal 
and coming to a climax in Salome and Elektra. Even at this point 
when Schoenberg's music had become totally un-Wagnerian and 


1 Бог an intensive study of Op. 11, and the works leading up to it, see Reinhold Brink- 
mann, Arnold Schénberg: Drei Klavierstücke Op. 11 (Wiesbaden, 1969); on Das Buch der 
hängenden Gärten, see idem, ‘Schönberg und George’, Archiv für Musikwissenschaft, xxvi 
(1969), p. 1. 

2 There is a good study of Op. 16 in Anthony Payne, Schoenberg (London, 1968), pp. 20-8. 

з On Schoenberg and expressionism, see Karl H. Wörner, Neue Musik in der Entscheidung 
(Mainz, 1954), pp. 58-9. 

4 See p. 90. 5 Schoenberg, Style and Idea (New York, 1950), p. 104. 

$ Leipzig and Vienna, 1911, e.g. pp. 370, 433, and elsewhere. 

7 Cf. infra, pp. 140-1. 

8 Hans Heinz Stuckenschmidt, Arnold Schónberg (Zürich, 2nd ed. 1957). English trans- 
lation by E. T. Roberts and Humphrey Searle (London, 1959), p. 65. 


18 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


anti-romantic, the underlying ideas are often romantic and Wagnerian. 
Erwartung and Die glückliche Hand are conceived in the image of the 
Gesamtkunstwerk; one, at least, of the Five Orchestral Pieces had a 
poetic connotation—the changing chord-colours of no. 3 were ‘com- 
parable, as Schoenberg says, with the ever-changing colour-impression 
of the lightly agitated surface of a lake’;! and the funeral bell tolls for 
Mahler in the little piano piece, Op. 19, no. 6. 

Webern had written atonal, non-thematic miniatures, the Fünf Sütze 
for string quartet, Op. 5 (1909) and the Sechs Stiicke for orchestra, 
Op. 6 (1910), before Schoenberg’s Op. 19; but although expressionism 
may be considered the final convulsion of romanticism, and despite the 
alla marcia funebre of Op 6, no. 4, and the suppressed titles of his next 
orchestral work, the Fiinf Stiicke, Op. 10 (1911-13), Webern was not— 
after Im Sommerwind and the other juvenilia of 1904-5--а genuine 
romantic. His earliest acknowledged composition, the orchestral 
Passacaglia of 1908, already manifests a Regerian love of construction 
for its own sake. Its very large orchestra, like that of Op. 6, is character- 
istic of the period, just as the chamber orchestras of Op. 10 and the 
Vier Lieder, Op. 13 (1914-18), are characteristic of the reaction from it. 
Even the extreme condensation of expressionism lasted only until 
Schoenberg showed him a way of escape from it. But if Webern was 
never a romantic, his friend Berg was never—to the last—anything else. 
The majority of the still mostly unpublished early songs? and some of 
the published ones (such as the beautiful *Traumgekrónt" of 1907) are 
in the true tradition of the romantic Lied; the expressionistic Vier Lieder, 
Op. 2 (1909-10)? mark the end of that tradition. Berg’s early instru- 
mental works follow closely the curve of development traced by the 
Schoenberg pieces on which they were to some extent modelled: the 
Piano Sonata, Op. 1 (1908) on the Kammersymphonie, the String 
Quartet, Op. 3 (1910) on Quartets nos. 1 and 2, and the Vier Stücke for 
clarinet and piano Op. 5 (1913) on the aphoristic piano pieces of 
Schoenberg's Op. 19, while the Drei Stücke for large orchestra, Op. 6 
(1914) shows a curious affinity with Mahler.* But the romantic vein in 
Berg remained fairly constantly obvious in the music of his maturity,’ 
and never more so than in the final Violin Concerto. 

The early course of Bartók's evolution was not dissimilar. Turning 
from Schumann and Brahms to Wagner, he was overwhelmed in 1902 


1 Webern, 'Schónbergs Musik’, in the symposium Arnold Schónberg (Munich, 1912), p. 44. 

2 Nicholas Chadwick, ‘Berg’s Unpublished Songs in the Österreichische National- 
bibliothek’, Music and Letters, lii (1971), р. 123. 

3 Nos. 2 and 3 recorded in Тйе History of Music in Sound, x. 

4 Оп this see particularly Hans Е. Redlich, Alban Berg—Versuch einer Würdigung 
(Vienna, 1957), pp. 93ff., and its condensed English version (New York, 1957), pp. 65—72. 

5 See infra, pp. 362 ff. 


FRANCE 19 


by the discovery of Strauss, particularly of Also sprach Zarathustra! and 
Ein Heldenleben. What set Bartók apart from his Austrian contem- 
poraries was his passionate nationalism, conventionally Hungarian in 
the Rhapsody for piano, Op. 1 (1904) and the first Suite for orchestra, 
Op. 3 (1905), but soon given a fresh direction by his discovery, in 
Kodály's footsteps, in 1905-6 of the authentic, un-gypsified folk-music 
of the Hungarian peasants. A belated acquaintance with Debussy's 
music іп 19072 was the second most important factor in guiding him 
away from tonality and the Central European mainstream toward 
simpler, less dense textures and new melodic modes, and his Bagatelles 
for piano, Op. 6 (1908) constitute a miniature thesaurus of the new 
harmonic and tonal devices of the period: counterpoint of chord- 
blocks (no. 4), bitonality (no. 1), fourth chords (no. 11), a foreshadowing 
of 12-note composition (no. 3),? the non-functional side-slipping of 
dissonances.* Schoenberg quoted (inaccurately) from no. 10 to illustrate 
the last of these, side by side with excerpts from Webern and Schreker's 
Der ferne Klang, in his Harmonielehre.? Bartók got to know Schoen- 
berg's music only in 1910; it affected both his harmony and the boldness 
of his line-drawing, and he wrote of it with warmth, though even in 
1937 he had still not heard Pierrot lunaire. Yet the last two of the 
Bagatelles are cryptically autobiographical quite in the Schumann 
manner, belonging to the group of works (including the First Violin 
Concerto and Ше Két arckép (Two Portraits), Op. 5) associated with 
the violinist Stefi Geyer; and in the piano Elegies, Op. 8b (1908-9) and 
the Két kép (Two pictures) for orchestra, Op. 10 (1910), Bartók reverted 
to romantic or impressionistic textures. It was only in 1911 with Blue- 
beard’s Castle and the Allegro barbaro for piano that Bartók achieved 
a satisfactory and highly personal synthesis of these diverse elements. 


FRANCE 


Except in the earlier works of Berlioz and the mature ones of César 
Franck, romanticism had never penetrated French music very deeply. 
Fruitful contacts with literature and the pictorial are characteristic of 
romantic music, but they had been characteristic of French music long 
before the romantic movement. Chopin left his mark. Liszt influenced 
Saint-Saéns, the oldest French composer of any significance during the 


1 Bartók, ‘Selbstbiographie’, Musikbldtrer des Anbruch, iii (1921), p. 88. 

2 Ibid., р. 89. 

? Oliver Neighbour, “Тһе Evolution of Twelve-Note Music', Proceedings of the Royal 
Musical Association, \xxxi (1954—5), р. 53. 

4 See, particularly, Edwin von der Null, Béla Bartók (Halle, 1930), pp. 3-14. 

5 p. 469. 

6 See his article, ‘Arnold Schönbergs Musik in Ungarn’, Musikblatter des Anbruch, ii 
(1920), p. 647. 


20 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


1890s, and Franck and his followers, particularly in their use of theme- 
transformation in major instrumental works. French opera from 
Chabrier’s Gwendoline onward abounds in leitmotives and other 
Wagnerian properties—even Massenet was not totally unaware of them 
— but d'Indy rightly held that the Wagnerism of Gwendoline is ‘more 
apparent than real’ and that Chausson's Roi Arthus, ‘the most complete 
specimen of a work influenced by the Bayreuth master’, is saved by ‘the 
intervention of our French nature’ from ‘exaggerations which no 
German writer would have known how to resist’,1 a judgement which 
is equally true of his own Fervaal. The romanticism of French orchestral 
music is often only picturesque and superficial, as with Dukas’s 
L’Apprenti sorcier (1897). It is not surprising that France began to 
reject romanticism before Germany, but it is typical of the period 
generally that two of the most effective agents of the rejection were 
composers who had fallen heavily under the spell of Wagner: Vincent 
d'Indy (1851-1931), by his foundation in 1896 of the Schola Cantorum 
which had as its main objects “а return to the Gregorian tradition of 
performing plainsong' (i.e. the Solesmes tradition) and ‘the rehabilita- 
tion of the music known as Palestrinian’,? and Claude Debussy (1862- 
1918) by his rejection of functional harmony and his obedience to 
Verlaine's command? ‘Prends l'éloquence et tords-lui son cou!” 

Despite the antagonism of their ideals and temperaments, an antagon- 
ism which did not preclude a limited amount of mutual admiration and 
even influence,’ d'Indy and Debussy had more in common than ап 
early enchantment with Wagner—from which the older man finally 
escaped only much later, after the other's death. Both preferred to work 
with small units of sound rather than long-breathed melody; both had a 
distaste for clear, symmetrical rhythmic patterns, though d'Indy 
vigorously distorted them whereas Debussy gently dissolved them; 
both orchestrated generally with characteristically French restraint and 
clarity. But the differences went much deeper, as deep as the personal 
difference between devout, even bigoted Catholic and free-thinking, 
free-living faun. Debussy was the very ideal of the late-nineteenth- 
century artist, all sensuality and sensibility; d'Indy, like Reger, was 
essentially a cerebral craftsman. Paul Landormy quotes his *peremptory 
affirmation’ that ‘une oeuvre d'art se fait? and the analyses of his own 
compositions in his Cours de composition musicale make amply clear 


1 Richard Wagner et son influence sur l'art musical frangais (Paris, 1930), pp. 69 and 75. 

? Prospectus of La Tribune de Saint-Gervais, i (1895). 

3 т Jadis et naguére (Paris, 1884). 

4 See Léon Vallas, Vincent d’ Indy, ii (Paris, 1950), pp. 195-6, 255, 319-21. 

5 La Musique française de Franck à Debussy (Paris, 1943), р. 77. 

6 Paris, three vols., i, 1903; ii, premiere partie (1909); ii, seconde partie (1933); iii (1950). 


FRANCE 21 


his concern with structural ingenuity for its own sake. Even his attitude 
to harmony was the complete antithesis of Debussy's; he wrote that 
‘chords as combinations of sounds appear only as the effect of a halt in 
the movement of the melodic parts . . . musically chords do not exist and 
harmony is not the science of chords. .. . The study of them for their 
own sake is an absolute aesthetic еггог.’! His approach to texture in his 
mature works was primarily contrapuntal, culminating in the polyphony 
of the Second Symphony (1903).? 

The conflicts of the romantic crisis are much more strongly marked in 
d'Indy's orchestral music than in his operas, in which can be traced a 
gentle decline from the strongly Wagnerian Fervaal, through L'Étranger, 
to the unsuccessful Légende de Saint Christophe to which he devoted 
most of the years 1908-15.? His earlier instrumental music had been 
almost entirely programmatic;* after the death of his master Franck in 
1890 there was a marked turn to absolute or relatively absolute music. 
In that very year he wrote his First String Quartet, and in the Second, 
in E major (1897), written in deliberate emulation of Franck, he carried 
the principle of essential thematic unity almost to its last extreme eight 
years before Schoenberg's Op. 7, all its themes being generated by the 
four-note motive which stands as its ‘epigraph’.* The same thematic 
density and tightness of construction mark the Second Symphony and the 
Piano Sonata in Е (1907).9 Yet the spirit of romanticism was not 
completely extinct even in these. The programmatic Istar of 1896 had 
been cast in the form of fairly strict ‘symphonic variations’; conversely 
the seemingly ‘absolute’ B flat Symphony is a struggle between two 
basic themes, symbols of darkness and light or evil and good, and the 
Piano Sonata, though profoundly different from the programmatic, 
Schumannesquely autobiographical piano compositions of the 1880s— 
the Poéme des montagnes and the Tableaux de voyage—is not without 
remains of romantic rhetoric. Moreover the Symphony was followed in 
1905 by the ‘symphonic triptych’ Jour d’été a la montagne: picturesque, 
romantic in essence and expression, even based on a detailed pro- 
gramme." Like so many of d'Indy's mature compositions Jour d'été 
introduces folk-tunes, which he had collected from the late 1880s 
onward, and is crowned by a melody derived from plainsong, the two 
elements which—particularly plainsong—played in his work a role 

! Cours de composition, i. p. 91. Schoenberg repeats this almost exactly in his Harmonie- 
lehre: in polyphony “Ше Akkorde entstehen nur also 7и/ е der Stimmenführung und sind, 
da die Verantwortung für das Zusammenklangliche vom Melodischen getragen wird, ohne 
Bedeutung für die Konstruktion’ (p. 348). But he adds, ‘That is naturally only half true... .’ 

? See Ex. 46 on pp. 108—9. 

з On the operas, see infra, pp. 169-71 and Cours, iii, pp. 201 ff. 


4 See Vol. IX. 5 Analysis in Cours, И?, pp. 267-70. 
$ Tbid., pp. 175 and 429. ? [bid., pp. 327-30. 


22 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


comparable with that of peasant-song їй Bartók's in freeing it from 
Germanic ways of thought. 

D'Indy was the most doctrinaire and the most influential of the 
Franckists, but his pre-eminence is partly due to the early silence of 
Duparc, who ceased to compose in 1885, the early death of Ernest 
Chausson (1855-1899), and the still more premature death of the 
brilliant young Belgian, Guillaume Lekeu (1870-1894). More individual 
than the other actual pupils of Franck, such as Guy Ropartz (1864- 
1955), were their allies: Albéric Magnard (1864-1914), Paul Dukas 
(1865-1935), and Albert Roussel (1869-1937). The operas of this group, 
above all Chausson's Roi Arthus and Dukas's Ariane её Barbe-bleue 
(1907), were more authentically Wagnerian than those of Massenet's 
so-called 'realis pupils, Alfred Bruneau (1857-1934) and Gustave 
Charpentier (1860-1956) but in Ariane the Wagnerism is strongly 
modified by Debussy's influence and both Dukas and Chausson 
were primarily instrumental composers; it is in their instrumental 
music that the crisis of romanticism is more perceptible, particularly in 
the way they treat their heritage from Franck. Duparc said that, 
*Chausson comes more directly from Franck than any of us'? and his 
В flat Symphony (1890), his Poéme for violin and orchestra (1896), and 
his chamber music are certainly Franckian in style though marked by a 
very individual vein of refined lyrical poetry. Magnard's symphonies, 
particularly No. 3 in B flat minor (1896), are subjectively, even rhetori- 
cally romantic, but reject his master d'Indy's cherished ‘cyclic form’. 
The open fifths that begin the Third Symphony: 

Ex. 6 


Modéré 


А 


dq 
ту 


) (| 


6 
| 


| 
TR 
| 


$ 
(| 


D.B. (div.) ди 


1 On French opera of this period, see infra, pp. 164 ff. 
? Quoted by Landormy, op. cit., p. 89. 


FRANCE 25 


actually seem to stand between the openings of La Damoiselle élue and 
Pelléas. Dukas's Ariane also betrays an awareness of Debussy (see Ex. 7) 


Ex. 7 
[Assez lent] 
Animez insensiblement 


Woodwind Ef 
Bs 


02 m 
| E Осо С с; = 
() „Шаг! ee E e 

| aa A EE шше eee 
Vlas. T Т! ЕЕ ЕЕ: 


poco cresc. 


- tran - ges! Pour-quoi re-cu-lez -vous quand je m’ap-pro-che? 


24 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


+ СА. 6507655. 


А-уе2-уоив peur еп-сог? 


(How strangely you look at me! .. . Why do you draw back when I approach?.. . 
Are you still afraid?) 


in its vocal writing, its harmony, and its texture and of Strauss in its full- 
bodied orchestration. In sharp contrast to the earlier programmatic 
Apprenti sorcier, his C major Symphony (1897), his Piano Sonata in 
E flat minor (1901), and the Variations, interlude et final sur un théme 
de Rameau (1903) are first and foremost musical structures. Although 
the conception is often more academic than romantic, Lisztian or 
Franckian romanticism comes to the surface in the finale of the Sonata: 


Ex.8 
Animé (d = 146) 


‚ „> ЖЯ ый ан кк E 


espress. | == 


SS 
ез —1—. 
E 


FRANCE 25 


and there are picturesque suggestions in the slow movement of the 
Symphony (e.g. where flute and clarinet call ppp, lointain). On the 
contrary, Roussel’s early work, with the exception of some d'Indyish 
chamber music—particularly the Divertissement for piano and wind 
quintet (1906) closely modelled on d'Indy's wind Divertissement of 
1898'—is frankly picturesque romanticism. His so-called First Sym- 
phony, Le Роёте de la forét, is an impression of a specific place, like 
d'Indy's Jour d'été, and consists of four pieces written independently: 
Forét d'hiver (1906), Renouveau (1905), Soir d'été (1904) and Faunes et 
dryades (1906). Stylistically it answers completely to his too modest 
description of all his music of the period 1898-1913: ‘marked by the 
weak influence of Debussy but above all by the struggle with the tech- 
nique learned from d’Indy’.? Evocations for soli and chorus (in the last 
movement) and orchestra (composed 1910-11) are avowedly impres- 
sions of travel in India; the ballet-pantomime Le Festin de l'araignée is 
naturally programmatic; and the opera-ballet Padmdvati, mostly 
composed before the outbreak of the 1914-18 war though the scoring 
was not completed till after it, is in the same vein as Evocations. All 
three reveal an individuality that by no means rests on the Indian 
elements in two of them. Roussel would have been the last to deny their 
romanticism; for him even Debussy was ‘a romantic in the best sense 
of the term'.? Yet he had already in 1906-8, in the Divertissement and 
parts of the D minor Violin Sonata, made essays in a much more spare 
and quasi-classical style and it was to this that he turned more and more, 
finally almost exclusively, in the post-war years. As for Dukas, after his 
ballet La Péri (1912) he remained almost completely silent. 

Chausson has been described as ‘un trait d'union non négligeable entre 
Franck et Debussy'.* Substitute ‘d’Indy’ for ‘Franck’ and the same could 
be said of Dukas and Roussel, but Chausson might be more truly 
described as a link between Franck and Fauré. Gabriel Fauré (1845- 
1924), an older man, was a pupil of Saint-Saéns more original than his 
master, more French than Franck. His music—mostly songs, piano 
music, and chamber music—is finely polished, classically restrained in 
expression yet saturated with a warm tender lyrical poetry which must 
be called romantic unless we are prepared to deny the epithet to Chopin 
as well, though Fauré has more often but less aptly been called the 
French Schumann. Fauré's earliest published works had been songs, 
and in the 1890s he reached his apogee as a song-writer with the 

1 See Basil Deane, A/bert Roussel (London, 1961), pp. 102-4. 

? Octave Séré, Cinquante ans de musique francaise (Paris, 1925) ii, p. 398. 

3 See Arthur Hoérée, Albert Roussel (Paris, 1938), p. 108. 


4 Hoérée, ‘Chausson et la musique française’, in Ernest Chausson, special number of 
La Revue musicale (December 1925), p. 193. 


26 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


Mélodies, Op. 58 (1890), and the cycle La Bonne Chanson (1892); both 
consist entirely of Verlaine settings—indeed nearly all Fauré's Verlaine 
songs date from about this time, as do most of Debussy's. 

Even before 1890 Fauré's role in the history of the French mélodie! 
was comparable with that of Wolf—perhaps one should say “оҒ Brahms 
and Wolf'—in the history of the Lied: the songs of Duparc, Chausson, 
and the young Debussy were all in some degree affected by him. The 
poetry of Verlaine and the other symbolists, which Debussy discovered 
earlier than Fauré, was the catalyst which precipitated some of his best 
work—and of Debussy's in this field. Fauré’s earlier songs are marked 
by a lyrical expansiveness which is almost invariably saved from the 
commonplace or sentimental by some subtle touch of melody or 
harmony or both, as in the first eight bars of ‘Les Présents’.? A suggestion 
of modalism is conveyed less often by actual flattening of a leading-note 
than tacitly by the omission of the seventh altogether, as at the end of 
the same song or at the end of ‘Green’, Op. 58, no. 3: 


Ex. 9 Allegretto con moto (J = 72) sempre dolce 
D 


р Ат ШУГЕ coal 
ME U pe e^ xe || == == da, НИТ е ысынан 
12, күл 


De. p Dd. з 
(And let me sleep awhile, since you are resting.) 


1 On Fauré's songs see particularly Vladimir Jankélévitch, Gabriel Fauré: Ses mélodies— 
Son esthétique (revised edition, Paris, 1951). 
? Recorded in The History of Music in Sound, іх. 


FRANCE 27 


The parallel passage in Debussy's setting of two years earlier has a 
romantic warmth that reminds one of Fauré at his less subtle. 


Ex. 10 


Nee aaa a f 
шаһ шағар 3м е E 
еее / т ут eee E 
кеске pa iS pa жаа ы i a 
- 


But when Debussy came to compose “Еп sourdine’ іп 1892 һе had 
arrived at both a more personal piano style, as at the beginning of the 


Ex. 11 
G) Andante moderato (4 — 63) dolce 


28 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


souf - fleber - ceur 


song, and a much less melodic way of setting words which, even when 
he is closest to Fauré's general style, as at ‘Laissons-nous persuader’ 
(Ех. 1101), is in marked contrast with Fauré's own setting in Op. 58, 
йо. (Ех d 


Réveusement lent 
Intimement doux 


Бартел реа 
Қамы D сата D аштан A жет i et [ut 


FRANCE 29 


(Let us be persuaded by the gentle lulling breeze.) 


There is less lyrical expansiveness in the songs of La bonne chanson and 
an exquisite sensitivity to every detail of the text which makes this cycle 
an outstanding masterpiece. In the later song-cycles, from La Chanson 
4’Ёуе (1907-10) onward, this lyrical restraint is more and more noticeable; 
voice-part, harmony, general texture become ever simpler. The same 
increasing austerity is apparent in Fauré’s later instrumental music. 
In his own way he too turned his back on romanticism. 

Debussy had done so long before. Although ‘old Klingsor’s ghost’ 
was still haunting him in 1893 he had already been touched by two 
influences which played some part in laying it: a Javanese gamelan and 
the music of the Russian nationalists, both heard at the Paris Exposition 
Universelle of 1889, although he had already encountered the Russians 
in 1881-2. He continued to employ some of the techniques and outward 
forms of romanticism—‘cyclic’ form in the String Quartet (1893) and 
leitmotives in Pelléas, programmatic or picturesque orchestral and 
piano music, even a suggestion of Franck's strident urgency in the 
‘Dialogue du vent et de la mer’ of La Mer (1905)—long after his music 
had been drained of romantic emotion and romantic eloquence. This 
was only the Debussyan form of the dualism common to most of his 
musical contemporaries; in the Douze Études and En blanc et noir for 
piano (1915) and the three chamber sonatas (1915-17) he finally 
resolved it. The younger musicians who grew up under Fauré's wing, 
such as Charles Koechlin (1867-1951), Florent Schmitt (1870—1958), 
Roger-Ducasse (1873-1954) and Maurice Ravel (1875-1937) never 
formed a ‘school’ comparable with the d'Indy circle but all shared their 
master's romantic-classic dualism, with classical tendencies gradually 
gaining the upper hand; all came to a greater or less degree under the 
influence of Debussy. The most romantic of them was Schmitt whose 
large-scale Piano Quintet, the composition of which was spread over the 

4 


30 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


years 1901-8, is held together by the cyclic principle and whose Рзаите 
XLVI (1904) for soprano, chorus, organ and orchestra is ‘monumental’ 
in the sense of Strauss and Mahler. (Schmitt was, as Calvocoressi 
remarked, “тоге directly influenced by the German romantics than 
any other French composer of any standing’.) In the case of Ravel, by 
far the most important, one catches him in the very act of ‘wringing the 
neck of eloquence' in his String Quartet (1903); while the picturesque- 
objective dualism of the piano works of 1905-8—the Miroirs, the 
Sonatina, and Gaspard de la nuit (which Ravel himself described as 
‘trois poèmes romantiques")?—is still manifest in the ballet Daphnis et 
Chloé, composed during the period 1906-11. The scales were tipped 
more and more in the direction of absolute music in the Valses nobles et 
sentimentales (1911), the Trois Poémes de Stéphane Mallarmé (1913), the 
Piano Trio (1915), Le Tombeau de Couperin (1917), the Sonata for violin 
and cello (1922), and still later works.? 


SPAIN AND ITALY 

Apart from Italian opera, neither Italy nor Spain had contributed 
significantly to musical romanticism. Indeed Spain had contributed 
nothing but the violin compositions of Pablo Sarasate (1844—1908) and 
the folk melodies that were borrowed or imitated by various Frenchmen 
and Russians. Even her opera had not developed beyond the zarzuela 
and the género chico (the petit genre) exemplified by the Verbena de la 
paloma (1894) of Tomás Bretón (1850-1923) and the Revoltosa (1897) of 
Ruperto Chapí (1851—1909), until Felipe Pedrell (1841-1922) composed 
his ambitious but unsuccessful Catalan trilogy Els Pireneus (1891) 
which characteristically had to wait till 1902 for its first performance—in 
Italian. Pedrell gave a much greater impulse to Spanish musical nationa- 
lism by his writings and musicological publications, though the first 
volume of his Cancionero musical popular español appeared only in 
1919, while Conrado del Campo (1876-1953) belatedly imported 
Central European romanticism in his symphonic poem La divina 
comedia (1904) and his Bocetos fantásticos for string quartet (1908). 
Thus romanticism reached Spain only as it was dying, and mainly from 
France and therefore heavily diluted. It was under the influence of 
contemporary French piano-writing that Pedrell's pupil, Isaac Albéniz 
(1860-1909), crowned an undistinguished creative career with his 
four sets of remarkable piano pieces Jberia (1906-09), and under that of 
Albéniz, modified by Domenico Scarlatti’s, that Enrique Granados 
(1867-1916) wrote his Goyescas (1912-14). Granados lived in Paris 


1 Musicians Gallery (London, 1933), p. 56. 
? *Esquisse autobiographique', Revue musicale, xix (1938), p. 213. See р 235 


SPAIN AND ITALY 31 


during the late 1880s and both Manuel de Falla (1876-1946) and 
Joaquín Turina (1882-1949) were there for six or seven years until the 
1914 war drove them home again. Of these two, Turina always remained 
a romantic in essence but Falla, much the more gifted, evolved from a 
glorified zarzuela, La Vida breve (1905; prod. 1913 in French), through 
the romantic impressionism of the Quatre piéces espagnoles for piano 
(1908): and the much less romantic impressionism of the orchestral 
Noches en los jardines de Езрайа (begun in 1909, completed in 1915), 
El amor brujo (1915), and El sombrero de tres picos (1917), to the classic 
austerity of the Fantasia bética for piano (1919) and the other post-1918 
works.? 

Whereas Wagner left Spanish music totally untouched, at least the 
superficial features of his technique—more or less continuous orchestral 
texture and what were supposed to be leitmotives—were appropriated 
by Giacomo Puccini (1858-1924) and the lesser composers of verismo, 
Ruggiero Leoncavallo (1858-1919), Pietro Mascagni (1863-1945), and 
Umberto Giordano (1867-1948) whose work is discussed in Chapter 3, 
while Italo Montemezzi (1875—1952) penetrated a little more deeply into 
Wagner's style in L'amore dei tre re (1913). Outside the theatre, the 
orchestral, chamber, and piano music of Giovanni Sgambati (1841- 
1914) and Giuseppe Martucci (1856-1909), disciples of Liszt and 
champions of Wagner, neither a very strong artistic personality, is very 
much poorer in genuine romantic sensibility than the scores of Puccini, 
while Leone Sinigaglia (1868—1944), a disciple of Dvořák, was important 
mainly as a collector and utilizer of Piedmontese folk-music and Antonio 
Scontrino (1850-1922) can hardly be considered a romantic at all. 

The only outstanding instrumental composer of the Puccini genera- 
tion, Ferruccio Busoni (1866-1924), elected as early as 1886 to leave his 
native land for Germany, Finland, Russia, and America before finally 
settling in Berlin in 1894. A great pianist and outstanding conductor, 
champion of every variety of contemporary music that needed a 
champion, Busoni seemed a reincarnation of Liszt, whose music he 
greatly admired. A true internationalist, he belonged to Italy no more 
than Liszt to Hungary and his own music has stronger affinities with 
Germany—he had some German blood—than with Italy. But the 
musical Germany that attracted him was not romantic Germany but the 
Germany of Bach (above all), Beethoven, and Brahms, and the Italian 
elements in his make-up which lighten the Violin Concerto (1897) and 
shine out in the Lustspielouvertüre (1897), the fourth movement of the 
Piano Concerto (1904), the suite inspired by Gozzi's Turandot (1904), 


1 No. 3, ‘Montafiesa’ is recorded in The History of Music in Sound, x. 
? On these, see infra pp. 200-1 and 315-17. 


32 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


and the opera Arlecchino (1917) belong to the world of Verdi's Falstaff 
(1893) and have nothing of morbidezza in their melody. Busoni's music 
is by no means devoid of romantic traits, both technical (the Lisztian 
organization of the Violin Concerto; the colossal scale and choral finale 
of the Piano Concerto) and in spirit (the Symphonisches Tongedicht of 
1893, the tragic third movement of the Piano Concerto). His first opera, 
Die Brautwahl (1912) was based on E. T. A. Hoffmann and his final 
masterpiece on the most characteristic of all romantic subjects, Doktor 
Faust (begun 1914; prod. 1925).! But the ideal which he constantly 
strove to realize was that which toward the end of his life he designated 
‘young classicism’ (junge Klassizitdt) and defined as 

the mastering, the sifting and turning to account of all the gains of previous 
experiments and their inclusion in strong and beautiful forms. This art will at 
first be old and new at the same time. The definite departure from what is 
thematic and the return to melody ... as the bearer of the idea and the 
begetter of harmony, in short the most highly developed (not the most 
complicated) polyphony. ... The renunciation of subjectivity ... and the 
re-conquest of serenity. ... Not profundity, and personal feeling and meta- 
physics, but music which is absolute, distilled, and never under the mask of 
figures and ideas borrowed from other spheres. Human feeling, but not 
human affairs ... not assigning to an art tasks which lie outside its nature. 
Description in music, for instance. ? 

He belongs therefore to another chapter as do for one reason or another 
the best of the younger composers of the generazione dell’ 80. 


RUSSIA AND POLAND 


Wagner was relatively little known and less admired by Russian 
musicians until 1889, when Angelo Neumann's opera company brought 
the Ring to St. Petersburg and Moscow. The impact of mature Wagner 
impressed Rimsky-Korsakov and was perhaps responsible for his 
conversion from an occasional to an almost exclusive opera-composer; 
yet, beyond an enrichment of his already extremely colourful orchestral 
palette and a more systematic use of leitmotives, it had no marked 
influence on his style. With the very important exception of Tchaikovsky, 
the romanticism of Russian composers had seldom taken the form of 
expression of exclusively personal emotion. Modelled on Berlioz, Liszt, 
and Glinka's Ruslan (and thus on Weber at second hand instead of 
Wagner at first), it adopted the forms and picturesque gestures of 
romanticism and entered into the romantic cult of past ages and the 
magical and exotic, but seldom with deep personal emotional involve- 


1 See infra, pp. 194—5. 

? Open letter to Paul Bekker, Frankfurter Zeitung, 7 February 1920. Reprinted in Гоп 
der Einheit der Musik (Berlin, 1922) p. 276-9; trs. Rosamond Ley, The Essence of Music 
(London, 1957), pp. 20-2. 


RUSSIA AND POLAND 33 


ment. It was Tchaikovsky who was the supreme Russian romanticist 
and, after his death in 1893, while much in his style and technique 
passed into common usage, only three or four composers of any 
significance followed him in his employment of music as an emotional 
language: Sergey Rakhmaninov (1873-1943) most closely, Sergey 
Taneyev (1856-1915) and Nikolay Metner (1880-1951) whose romanti- 
cism was of the restrained nature of Brahms's, and Aleksandr Skryabin 
(1872-1915) who must be counted among those who attempted to push 
the aesthetic of the Gesamtkunstwerk to its farthest consequence. The 
surviving members of the former Mighty Handful continued and 
extended its tradition of colourful, Russian-flavoured music. Rimsky- 
Korsakov (d. 1908) in a series of operas three of which, Kashchey 
bessmertny (Kashchey the Immortal) (1902), Kitezh (1905; prod. 1907), 
and Zolotoy Petushok (The Golden Cockerel) (1907: prod. 1909), mani- 
fest a further deepening of his veins of mysticism and satire and a 
further development of his pungent but highly artificial harmonic 
vocabulary, Balakirev (d. 1910) creating new works in his old vein. 
(Cui (d. 1918), the third survivor, had always lived mainly on glory 
reflected from his colleagues.) Their numerous pupils, Anatoly Lyadov 
(1855-1914), Sergey Lyapunov (1859-1924), Aleksandr Glazunov 
(1865-1936), and others less talented, produced syntheses of the idiom 
inherited from the Handful with a highly polished, euphonious style 
in which technical fluency often outran the flow of creative ideas. AII 
were overshadowed by Rimsky-Korsakov's last notable pupil, Igor 
Stravinsky (1881-1971), who in three full-length ballets, Zhar-ptitsa 
(The Firebird) (1910), Petrushka (1911), and Vesna svyashchennaya (The 
Rite of Spring) (1913), enriched and extended the old ‘nationalist’ idiom 
with harmonic and rhythmic conceptions drawn partly from late 
Rimsky-Korsakov and Skryabin, partly from the West, and more and 
more from his own fertile inventiveness, but also drained away from it 
the last hint of the romantic. 

The concentration on craftsmanship in the music of the nationalist 
epigones and the emotions and philosophical ideas underlying the 
equally highly polished music of Rakhmaninov and Skryabin are both 
attributable in part to the intellectual climate of Russia at this period. 
The optimistic positivism of the third quarter of the nineteenth century, 
when artistic truth mattered much more than beauty, had long since 
lost its force. Under an increasingly reactionary political regime only 
the revolutionary extremists preserved the energy that comes from 
despair; and in the general atmosphere of futility which Chekhov 
derided, the creative artist tended to take refuge in a sterile cult of 

1 See below pp. 174-7. 


34 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


*Parnassianism', of *pure' beauty, in symbolism and mysticism. Music- 
ally this took the form of the *miévre lyrisme’ of Tchaikovsky’s imita- 
tors, which Stravinsky derided.! Rakhmaninov's pessimism is much less 
anguished and personal than Tchaikovsky's or Mahler's, and often 
shows itself in flaccidity of invention rather than positive expression. 
His most successful essay in actual programme-music, the tone-poem 
inspired by Bócklin's painting Die Toteninsel (1908), is typical of the 
polished superficiality of his elegiac art. Fine craftsmanship is also a 
distinctive quality of Skryabin's music, but almost from the first, in 
the piano Preludes, Opp. 11, 13, 15, 16 and 17 (mostly 1894—5), it 
was informed by a poetry more individual than Rakhmaninov's and 
expressing a much wider range of moods. Quickly absorbing and out- 
growing the influence of Chopin and Schumann, he developed from an 
exquisite piano-miniaturist into a musical mystic,? increasingly obsessed 
with a pathological belief in his own Messiahship. His early piano 
sonatas, nos. 1—3 (1893-8), are normal late-romantic compositions, but 
from 1900 onward he became more and more deeply drawn into a 
mystical philosophy of art and his First Symphony, completed in 1900, 
ends with a setting for mezzo-soprano, tenor, and chorus of a poem of 
his own in which art is nearly equated with religion. Whether he had 
first-hand knowledge of Mahler's first three symphonies is uncertain 
though he must have known of them by hearsay. In any case, he 
followed neither Mahler nor earlier composers of partly choral sym- 
phonies, for in effect his vocal finale is an epilogue in E major, partly 
based on one of the themes of the purely orchestral first movement, a 
lento also in E major: 


Ех. 12 
Lento (4 = 66) 


3 Flutes 


() 8 g 
ГУ uia 
[Fury W wj 
I fan’ 

LN 


Strings 
Vc. pizz. 

cH 

| 


NU] 


! Poétique musicale (Cambridge, Mass., 1942), p. 65. 

2 Martin Cooper, ‘Aleksandr Skryabin and the Russian Renaissance’, Studi musicali, i 
(1972), p. 327, shows the relationship of his ideas to those of the Russian thinkers and poets 
of the day. 


RUSSIA AND POLAND 35 


These two slow movements in E major frame a conventional four- 
movement symphony in E minor. The catastrophic failure of the 
experiment was caused by Skryabin’s inability to devise anything better 
than a square-cut diatonic fugue, of a type more often written by 
students than by full-fledged composers, for the chorus. After the purely 
orchestral Second Symphony (1901), showing the influence of Liszt and 
Wagner (and perhaps also of Nietzschean philosophy, which obsessed 
him for a time) but also drawing nearer in style to his own far more 
individual piano music, he began to dream of “а fusion of all the arts, 
but not a theatrical one like Wagner’s’. ‘Art’, he said, ‘must unite with 
philosophy and religion in an indivisible whole to form a new gospel, 
which will replace the old Gospel we have outlived. I cherish the dream 
of creating such a “mystery”. For it, it would be necessary to build a 
special temple—perhaps here, perhaps far away in India. But mankind 
is not yet ready for 17.1 Skryabin's Third Symphony, Le divin poème 
(1904)—preceded by the harmonically advanced Fourth Sonata (1903) 
—is in three movements, ‘Struggles’, ‘Delights’, ‘Divine Play’, and, 
according to the programme written by his mistress, Tatyana Schlézer, 


represents the evolution of the human spirit which, torn from an entire past 
of beliefs and mysteries which it surmounts and overturns, passes through 
Pantheism and attains to a joyous and intoxicated affirmation of its liberty 
and its unity with the universe (the divine ‘Ego’), 


The one-movement Poem ekstaza (Poem of Ecstasy) (1907), for very 
large orchestra with, for instance, quadruple woodwind, eight horns, 
and organ, and the Fifth Piano Sonata (1908) which quickly followed 
it? elaborate similar ideas with a further intensification of post- 
Wagnerian harmony. In his next and last orchestral work, Prometheus 
(1910), ‘for full orchestra and piano, with organ, [wordless] chorus, and 
clavier à lumiéres’, Skryabin carried his harmonic language a stage 
further still to the very brink of atonality;* this was a preliminary study 
for the long contemplated but never written ‘mystery’, a ‘liturgical act’ 
combining dancing, music, poetry, colours, and scents, which would 
induce ‘a supreme, final ecstasy’ in which ‘the physical plane of our 
consciousness would disappear and a world cataclysm would begin'.* 
The part for the clavier à lumières is in normal musical notation, but 


1 Yuly D. Engel, “А. М. Skryabin. Biografichesky ocherk’, Muzikalny Sovremennik 
(1916), no. 4—5, p. 56. 

? Skryabin published the verbal poem, the first part of which (lines 1—224) provided the 
programme of the orchestral poem, at Geneva in 1906; lines 227-35 were printed as an 
epigraph to the Fifth Sonata. The whole poem (original text and German translation by 
Ernst Moritz Arndt) is printed as an appendix to Clemens-Christoph Johannes von Gleich, 
Die sinfonischen Werke von Alexander Skrjabin (Bilthoven, 1963), p. 113. 

3 See infra pp. 136-7, and the Frontispiece of this volume. 4 Engel, op. cit., p. 89. 


36 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


Skryabin intended that the pressure of each key should produce not a 
musical sound but an intense light, coloured in accordance with his 
associations of sound and colour (e.g. C — red, C sharp-D flat — 
violet, D — yellow, and so on), and flooding the concert-hall; mere 
projection of colours on a screen was a compromise which he rejected as 
‘trivial’. The Sonatas Мо. 6-10 (1912-13) and the other post- 
Promethean piano-compositions probably—Opp. 71-4 (1914) certainly 
—originated as music for the *mystery'. Thus Skryabin's Gesamtkunst- 
werk was never realized and his nearest approach to it, Prometheus, 
fell short of Schoenberg's Die glückliche Hand, even—most ironically— 
of the ballets of the outstanding anti-Wagnerian, Stravinsky. 

The mélomimiques of Vladimir Rebikov (1866—1920), short dramatic 
studies of psychological situations for one or two characters, with piano 
accompaniment, and his ‘musico-psychographic dramas’, attracted a 
good deal of attention during the early part of the century; but like his 
harmonic experiments with chords of superimposed perfect fourths, 
they were foredoomed to failure by the weakness of his musical 
invention. Nevertheless they were known to Janaéek and may have 
suggested the form of his Zápisník zmizelého (Diary of one who 
vanished).? 

Although traces of Skryabin's stylistic influence are perceptible in 
The Firebird and the early works of Myaskovsky and other Soviet 
composers of the early post-Revolution period, even in the young 
Prokofyev, only one important non-Russian composer was deeply 
affected by him, the Pole Karol Szymanowski (1882-1937),?? who 
towered above the other members of the *Mlada Polska' (Young 
Poland) group, Mieczysław Karłowicz (1876-1909), Grzegorz Fitelberg 
(1879-1953), Ludomir Rózycki (1884-1953), and others. The common 
aim of these fyoung Poles' was a renaissance of Polish music to be 
effected by the reconciliation of the national idiom with more or less 
contemporary forms and harmony; all were influenced by Strauss. But 
Szymanowski, after a First Symphony (1907) in which the shadow 
of Reger also appears, a symphonic colossus typical of the period, 
went on to write a Second (1909) and a partly vocal Third (1915-16) 
in which Skryabin and French music played a liberating role. The 
influence of Skryabin's style is still more marked in Szymanowski's 
Métopes (1915) and Masques (1917) for piano and his Mythes (1915) for 
violin and piano, while the blending of the mystic, the erotic, and the 


! Leonid Sabaneyev, Vospominaniya o Skryabine (Moscow, 1925), p. 61. 

? See infra, pp. 307-8. 

? бес particularly Józef Chominski, ‘Szymanowski i Skryabin' in Igor Вега (ed.), 
Russko-polskie muzikalnie svyazi (Moscow, 1963), p. 375. 


SCANDINAVIA 37 


orgiastic in his later opera Król Roger (King Roger) (1926) also smacks 
strongly of Skryabin.! 


SCANDINAVIA 

Those countries which may, more conveniently than accurately, be 
called Nordic and Anglo-Saxon, were most strongly influenced by 
Germanic techniques and idioms, less by French, and hardly at all by 
Slavonic even after Dvorak and Tchaikovsky had become familiar and 
even popular. But there was an affinity with the Slavonic lands in that 
clinging to national idioms which has always been associated with 
musical romanticism although the impersonality of such idioms is in 
direct conflict with the subjectivity that is the essence of romanticism. 
The Norwegian Christian Sinding (1865-1941) and the Swede Wilhelm 
Stenhammar (1871—1927) are typical of the German-orientated com- 
posers of the period who yet managed to infuse national qualities—and 
in Stenhammar's case an element of neoclassicism—into their later 
works. But it was the older Grieg (1843-1907) who still dominated the 
Scandinavian scene, ever a thoroughbred romantic but in some respects 
a new Grieg as in the Haugtussa Sang-Cyclus (1898) and the piano 
arrangements of Norske Folkeviser, Ор. 66 (1896) and Slåtter (1902), 
the harmony of which last is influenced by a folk instrument, the 
hardingfele or ‘Hardanger fiddle'.? 

Grieg was hardly ever completely successful when he attempted 
large-scale composition; even the so-called Symphonic Dances on Nor- 
wegian folk-tunes (1898) are essentially compilations of miniatures. 
But the two Scandinavians who came to the fore during his later years, 
the Finn Jean Sibelius (1865-1957) and the Dane Carl Nielsen (1865— 
1931), were both masters of the larger forms, particularly of the sym- 
phony. Both were markedly individual talents and, although it is not 
difficult to detect the influence of Bruckner's ostinati on Sibelius or to 
trace this or that melodic turn in his earlier works to Borodin or 
Tchaikovsky, his mature idiom is strikingly unlike that of any of his 
predecessors or contemporaries. It is always clearly diatonic, as is 
Nielsen's. Although Sibelius anticipated Mahler in employing the 
chorus in a very large-scale symphonic work (Kullervo, ‘symphonic 
poem for soloists, chorus and orchestra', 1892),? he was the antipode of 
Mahler, reacting in his symphonies—even in the First (1899)—against 
all those procedures in the nineteenth-century symphony which Mahler 


1 On Szymanowski's later works, see infra, pp. 311-15. 

2 See John Horton, ‘Grieg’s “Slaatter” for Pianoforte’, Music and Letters, xxvi (1945), 
p. 229. 

3 See Robert Layton, Sibelius (London, 1965), pp. 107-14. 


38 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


developed further. Structures are tightened; transitions are from the 
first most skilfully made; and the thematic work is severely dialectical 
so that, while theme-transformation goes on continually, a totally 
organic web of sound is woven without need of such mechanical devices 
as the motto-theme to give it a sense of unity. In the Fourth Symphony 
(1911), which the composer himself described as “а protest against the 
compositions of today',! compression is carried so far that the first 
subject consists of only 25 adagio bars, the second of only 8; but this 
compression and the chamber-like treatment of the orchestra, if a 
*protest' against elephantiasis, was very much in line with the other 
tendencies of that date which have already been remarked on. Com- 
pression led to actual fusion of the movements in the Seventh Symphony 
(1924). Although Sibelius's earlier symphonies are romantic in tone, 
they are not programmatic and he protested vigorously against attempts 
to interpret them so.” The strong romantic element in his make-up was, 
as it were, siphoned off in a series of tone-poems from En Saga (original 
version, 1892; revised 1901) to Tapiola (1926) and other works. On the 
other hand Nielsen, after a First Symphony (1892) chiefly remarkable 
for its unconventional treatment of tonality, went on to write a com- 
pletely programmatic Second, De fire Temperamenter (The Four 
Temperaments) (1902),? and others with suggestive titles: no. 3 Sinfonia 
espansiva (1911), no. 4 Det Uudslukkelige (The Unquenchable) (1916), 
no. 6 Sinfonia semplice (1925). His justification for such titles was that 
‘while a short hint or title can illuminate the music in many ways . . . the 
programme or title may indicate a mood or emotion but not an idea or 
concrete action'.* 


BRITAIN AND THE UNITED STATES 


Music in the English-speaking countries remained until the end of the 
nineteenth century, and after, under influences that were not only 
German but rather conservatively German. It was Brahms rather than 
Wagner who attracted Alexander Mackenzie (1847-1935)—whose best 
work, like Sullivan's, was all done before 1890—Hubert Parry (1848- 
1918) and Charles Stanford (1852-1924). The American talents of 
Edward MacDowell (1861—1908) and Horatio Parker (1863-1919) were 
initially fostered by Raff and Rheinberger. But the strongest and most 


! Harold E. Johnson, Sibelius (London, 1960), p. 126. 

? See, for instance, Karl Ekman, Jean Sibelius (Helsinki, 1935; English translation by 
Edward Визе, London, 1936), pp. 191-2 (English edition). 

3 For Nielsen's own explanation see Robert Simpson, Carl Nielsen, Symphonist (London, 
1952), pp. 42-3. 

4 Nielsen, Levende Musik (Copenhagen, 1925), p. 41. 


BRITAIN AND THE UNITED STATES S9 


individual Anglo-Saxon of their period, Edward Elgar (1857-1934), 
after a long slow development in the debilitating climate of the English 
provinces, discovered for himself the real fountainheads of later 
romanticism, Schumann, Liszt and Wagner, and applied their tech- 
niques of harmony, orchestration, and leitmotive first in the dead forms 
of the Victorian cantata and the dying one of the oratorio, most 
successfully in The Dream of Gerontius (1900), and then in a series of 
intensely personal orchestral works—the Enigma Variations (1899), two 
symphonies (1908 and 1910), the symphonic study Falstaff (1913), the 
concertos for violin and for cello (1910 and 1919)—which constitute 
the high-water mark of English musical romanticism. Elgar really stands 
alone; the work of the only comparable figure, Frederick Delius 
(1862-1934), English merely by the accident of birth, has nothing in 
common with any English tradition and little with the mainstream of 
Continental romanticism. Delius was an isolated cosmopolitan like 
Busoni, poetically far more and technically far less gifted. (A compar- 
able, though less individual, character was his almost exact contem- 
porary, the Alsatian-born Charles Martin Loeffler (1861-1935), an 
American from 1881 onward.) He stands nearest to his colleagues not 
in his musical style but in his sources. Like Schoenberg he was attracted 
by the poetry of Jens Peter Jacobsen (the opera Fennimore and Gerda 
(1910), the Arabesk for baritone, chorus and orchestra (1911), and a 
number of songs); like Strauss and Mahler, he came under the spell of 
Nietzsche’s Zarathustra (A Mass of Life, 1905); like his younger English 
contemporaries, he set Walt Whitman (Sea Drift, 1903) and based 
compositions on folk-melodies: Appalachia (1902), Brigg Fair (1907), 
and On hearing the first cuckoo in spring (1912). But his borrowed folk- 
melodies were American Negro and Norwegian as well as English and 
he used them merely as catalysts for his ultra-romantic nostalgia, 
embedding them in his personal dialect of Griegian harmony, whereas 
the best of his younger English contemporaries sought in English folk- 
melody for the basis of a musical style that should be independent of 
the German tradition. 

These were Ralph Vaughan Williams (1872-1958) and Gustav Holst 
(1874-1934), both pupils of Stanford who had himself written a 
symphony and orchestral rhapsodies on Irish tunes. But Stanford 
treated folk-music as ‘material’ for traditional composition, while for 
his pupils it was the principal—though not the sole—instrument of 
emancipation from the heritage of both Brahms and Wagner, from the 
forms and idioms of the European mainstream altogether. Although 


1 See pp. 507-19. 


40 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


there are plentiful traces of romanticism in their extra-musical thinking, 
their suppression of subjective emotion and rhetoric was almost as 
complete as their rejection of chromaticism and rich, complicated 
textures. In Holst's directness of expression and fresh approach to the 
essentials of music, and in his vein of transcendentalism, itis possible to 
find hidden affinities with his American contemporary, the Horatio 
Parker pupil, Charles Ives (1874-1935) 1 but Holst was never as radical 
as Ives and he achieved general recognition with the performance of 
The Planets in 1918; Ives had to wait considerably longer. The more 
conservative Vaughan Williams made his mark more quickly; before 
the choral Sea Symphony (1910), his name was already well known to 
his fellow-countrymen. 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 


It is significant that Vaughan Williams, Holst, and Ives rejected not 
only the idioms of their European contemporaries but their sophis- 
ticated techniques. To say what they wanted to say, more idiosyncratic 
homespun techniques were not so much adequate as necessary. But 
Holst's ostinati and static basses, Vaughan Williams's fumbling with 
the methods һе had learned from Parry and Stanford, Ives’s ‘collisions 
of musical events', seem embarrassingly naive beside the harmonic and 
orchestral refinements of the Continental composers of the period. 
Even the Europeans who were turning back to sound-for-sound's-sake 
and away from the concept of music as a quasi-language were still often 
employing the techniques which the late romantics had evolved in their 
striving to catch and communicate the subtlest nuances of complicated, 
even perverse, emotion and mystical or ironic ideas. The erotic oriental 
mysticism of Szymanowski struggles in the song 'Smutna wiosna’, 
Op. 24, no. 6 (1911) to express simultaneously the sense of spring 
returning and sorrow for a dead love throughtypical romantic harmony 
and a plethora of meticulous markings. ‘Where art thou?’ his lover 
cries (Ex. 13). Three years earlier Holst had set the same words, in 
nearly the same sense, in his oriental-mystical chamber-opera Savitri, 
with the utmost austerity (Ex. 14). And although Holst's voice-part 
outlines an F minor triad, the typical ostinato bass shows that he has 
really parted company with tonality, while the Szymanowski song is in 
D major despite the constant sound of minor sorrow in the harmony 
and despite the fact that only one phrase in the voice-part—the last—is 
in D major and floats down to a piano cadence which perfectly illustrates 
the final stage of functional harmony in dissolution (Ex. 15). 


1 See pp. 574-83. 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 41 


Ex. 13 (Poco meno) 


Andante dolcissimo Holt express. 
---+---< 


е molto espressivo фр 


Moderato 
(Solo) PPP 


Via. 


eT 
-е 0 
ep ПЕТ, 


Vc. 
(2 Soli) ES =A p 


1890-1914 


42 THE APOGEE AND DECLINE OF ROMANTICISM: 
SATYAVAN 


ENTE 
il T | | | — 
| 


шш 3 | M. 
| К (о 1-1 avuc 
а M "Nh 


dim. 


N 
" 


|н 
41 


(+ FI. П) 


ии 
[—— smi 
|- cr md 
are 
a 
ia 
——— 
DD 


Mine eyes 
V 


WE 


morendo 
rallent. 


ay me жан 
aaa 
Ерос 
aes 
Есен 
E 
т 
pe | 
Z 
iz 
ES 
peu] 
ОФ. Бе 
(2 
[Rr 
Ба ар 
ГЫ 


where art thou? 


[———— 1 
үс шект р == ырк зис | шиш) 
ЕЕ ЕЕ ЕЕ ee Ee =] 
Tc {7 V 
= 
© 25 ря 
pem 424 
2 
dolciss. 


| | Е: M 
É - 
+ . 
Jh T ЬЯ + 
ІШСЕ T = 
1 ыш S 
; | 
Ш 0 ДҮ 
| і ( m 82; 
RG со C 55% 4 А < ы 
ML SE eum s c 


(loveliest flower of spring!) 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 43 


This process of dissolution had certainly been protracted. It had already 
reached an advanced stage in the opening of Act Ш of Parsifal which 
was written in 1879, but its progress must be measured not so much 
by its intensity in certain passages as by the extent of its cancer-like 
spread through entire movements and works. lt was essentially a 
harmonic process and fundamentally affected melody only in so far 
as melody is the surface of harmony; much seemingly chromatic 
melody, even in Tristan, is no more than a rippling of the surface with 
chromatic appoggiaturas and passing-notes. But basically diatonic 
melody, even purely diatonic melody, with its tonal associations and 
implications was still an important feature of the late romantic sound- 
language and few composers discarded it entirely. Strauss, Elgar, 
Dukas, Rakhmaninov were but a few of those who often employed 
clear diatonic harmony and melody; these provide the language for 


Ех. 16 
Stürmisch bewegt (4 = 60) 
ен 
[и «ре у] -—— 
п [EE EE 
0 257 
[e 
PERS == == = 
in pj 0t fa 
Cl. in D ЕЕЕ 


(actual 
sounds) 


Б 
| 
; 


СІ. г; Ggs 
Basset 


Horn 
> 
ав м 
ЕД =) 
Я дя 2 
IEEE. лан 


| 
| 


е Ex L 
Р aure Е] р 
А очи. жс еі 5 
dy 
Strings 
Vc. pizz. arco Jf 
Ley dun —— === 
Ус 3 


44 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


фт тт 
жой 
FI. 
tri--—— tr 
- be b bæ 
Ob. д - - = = 
JU Неда 0, 
CL in D = = 
си == 
267 b 
aq = ir i : 
Basset = i 
Horn ie ( C = 
Bn. is = 6 E 
Б) а=. Da 
mf 3 > 3 3 
I 
ІШ 
E > 
НЕН ДЕ 1 
IV ве | 
JP 
o COR REM тысты 
H T a 
= > 7 
RA = > 
Strings 3 
= WIL = Г 
= ; rip ji 


Delius (Songs of Sunset, 1907), Elgar (end of the Second Symphony, 
1910), and Mahler (‘Abschied’ of Das Lied von der Erde, 1908) to sing 
their saddest farewells. Diatonic melody was often treated as the core 
of otherwise deliquescent or tonally uncertain or obscure harmony, or 
deliberately contrasted with tonally disturbed passages, as in Parsifal, 
Ein Heldenleben, The Golden Cockerel, to suggest antitheses of good 
and evil or real and unreal. Even the purely diatonic could be subjected 
to strange juxtapositions and dislocations like the flattening of the 
second and third degrees of the scale in the seventh bar of Ein 
Heldenleben and the expanded ‘Neapolitan’ interpolation of bars 11-12, 
which is logical and almost orthodox by comparison with the Introduc- 
tion to Rosenkavalier where E major is dazzingly expanded by purely 
diatonic means (Ex. 16). 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 45 


There is a close technical parallel to the *Neapolitan interpolation' of 
Heldenleben near the beginning of the Andante of Mahler's Sixth 
Symphony: 


Ex. 17 
Andante moderato 


= eee _| ____ a 
i : Ес жЕ = жс ыы сш] 
Ob.I В 5 
Woodwind 
CI Р--Е-7- 
р ЕН 


Ey, (iS тарда жаманға ы ВЕТ un "a | 
рм ГГ с ж] zm. Еж) 
a руд D <) 
QS Ez c x cA 5 
Во LII P ~ 
Emo San Jus, [m] 
ju NOD x — ad Reese, ш 


Ey cepe 
EL асны ШШ ШЕЕ жеу” 


Еш дао Bea WALD o Е | Ж „ш 
[Ew ы TS p p or wee] 
Dc» 


Strings 
= m. L— Бы 
ить раа рааны ча] 
АЕ raro ЖШ. ЖҰБЫН ШШЕН ГУЧИ гелін eet ге [eme кш e] 
(р ГР УТ тағу д 2221-0-26 oe БЕН. 
BN SS aS Um mgr DE es а ee ви = СІ ыса mE] 


— m! ше 1—5 Ы 
(з= e 
KA CU sr. 
Бес ee] b 
[me — P-] 


Ta an сс“ D? SS re ee ПЕЕВ pa те ТЕН) текше | 
[9 )--105——*9| aay тр ——— 
ре. Hz Se 
— ——————— " = 
ss M = — 
() (tacet) 
El em атту — 4 = 
amb Ne ушы х= x] K | | 
НХ cj ae 1 
my ра DS 
p > Е 
— —À 
f Dp 


Жош: em SPESA 
P = 


1890-1914 


46 THE APOGEE AND DECLINE OF ROMANTICISM: 


ee шеше 


у= == 
Se! ГТ р Лнан пази 


motivated by some extra-musical purpose, 


though this seems to be 


Strauss's by sheer musical exuberance. Even while Mahler was working 
on the Sixth Symphony in 1904, Schoenberg was opening the first 
of his Sechs Lieder, Op. 8, with a succession of common chords equally 


disruptive of the tonic key, E major: 


‚Ех. 18 


Mässige J 


(XX 


| 


ay 


N 


== ©] 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 47 


C 
pom wi 
a 
5 


е z > 
{ew es Re Se 


© 


(Night flows into day, day into night,) 


Thus, even in diatonic contexts, functional key-relationship remained 
no more than an outworn convention which, as we have seen,! Mahler 
had already discarded in his Third Symphony; the E major coda to the 
very G major Fourth Symphony, a master-stroke of imagination, is 
motivated by no law of tonal gravity but by its contrast of key-colour. 
Or rather it is used as a means of suggesting the unearthly radiance of 
angelic music, just as the tonal aberrations of the Rosenkavalier Intro- 
duction convey the orgiastic turbulence of Oktavian and the Marschallin. 
Such passages can often be justified by extra-musical considerations; 
indeed they usually originate in them; but by this time they had become 
part of the normal language of music. 

If rapid harmonic change could weaken key-sense even in essentially 
diatonic music, it could destroy it completely when the harmony is 
chromatic. A mere veil of chromatic appoggiaturas and passing-notes 
can totally disguise the most familiar of chords, as Wolf showed in the 
opening of the twenty-seventh song of the Italienisches Liederbuch, 
‘Schon streckt’ ich aus’ (1896), where he broods on a single chord of the 
diminished seventh for the best part of two bars before revealing at the 
very end of the second bar that the key is A flat: 


Ex. 19 


Sehr langsam (4 = 42) D" 


48 THE APOGEE AND DECLINE OF ROMANTICISM: 1890—1914 


When the basic harmonies are less familiar and succeed each other more 
quickly, tonality may still be discovered by painstaking analysis but it 
has lost any real significance. The following passage from Schoenberg's 
Kammersymphonie, ten years later than the Wolf example, is preceded 
by a G major chord and it cadences in G major (with a parenthesis 
between dominant—x—and tonic) :1 


Ex. 20 


Woodwind 


Strings 
(all except 
Vn. I muted) 


but recognition that it is in a strongly modified G major is of slight 
importance in that context, or in relation to the total architecture of the 
composition. Yet this passage still has a recognizable cadence and 
symmetrical inner structure; recognizable cadences and obvious 


1 The original dynamic markings of the Symphony, given here, were considerably altered 
in later editions. 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 49 


periodicity became ever less frequent in the language of late romantic- 
ism, giving way to a species of opaque musical ‘prose’. At the other 
extreme, Strauss, Mahler, and Reger—to say nothing of such conserva- 
tives as Pfitzner—sometimes revert to a provocatively naive, sym- 
metrical and diatonic type of melody. 

The old drives and tensions of harmony and key having, by 
over stretching, been weakened almost to non-existence, the energy of 
music was mainly concentrated in line: melody, motive, or more often 
something less easily definable. ‘Prose’ lines of great flexibility, angu- 
larity, and sensitiveness are brought into relationship through, or 
sometimes against, somewhat lymphatic harmony of various kinds: 
chromatically modified diatonic, experimental (for instance, based on 
superimposed fourths instead of thirds) or purely empirical. In the 
first decade of the century one can still often detect a family likeness 
among the melodic lines of very different composers, traceable to their 
common filiation from Wagner—which is naturally most obvious in 
the passages from d'Indy and Mahler, the earliest of the seven quota- 
tions grouped in Ex. 21. 


Ex. 21 
(1) D'INDY (1895) 


Un peu moins lent 


Woodwind 


Solo 
Viola 


poco sfz tres soutenu 


50 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


DON = йт. p ж 
Ц 
| Ә.Н ІЛ ТЕР ы SS кете NIS Е аа 
М LI ]-- 18m 2m sema ig и к ш et P» | d 
[7 
| Ob. Sclo - | CO 
Woodwind 2 trés expressif 
стезс. 
2 — 67656 КИЕ я. 
НЕТ НН =— 
Sj ЧЕНЕ 


Gep 
Violins | а 
BAS) So VET типа ЕЕС) 
Г] ку л ene 1 oH Р 
Dp 
Violas 
Cellos 


Woodwind 


Фр poco rit. 
MU CENT mec 
тиди ИШ mc 


Strings 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 9i 


(iii) BARTÓK (1903) 
Allegro moderato 


2р Ob.I = р 


Tay ee Se eT 
майса | 


БОЛ СЕ Т ЕЕ ЕЕ 
2p 


Cellos (div.) 


+ IRA 


1890-1914 


é 


52 THE APOGEE AND DECLINE OF ROMANTICISM: 


Yar 
па Ц 
Ае 


E 
Woodwind| -foe 


| 


А 


| | 


ШШ ГЛ? КЪ. 


Кое 


жууш ee илиш” сос шиш ы Жаы ЕЕ 
. .. 


| 
УЕ 


Fau 


Cellos (div.) =E E 


(+ Ob. I & Ш) 


#1.42 


Cl. I, II & Eb 


| 
RU 
В 


чей 


< TE 


53 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 


(iv) SCHOENBERG (1906) 


ЕЕ 
bd 
b 
= 
— 
Пеене: Е 


x 


E |". 


атасы меу шет уш ск С ти —— g 


mp 


pem 


E в 
С ош Ира Т -- ЭШ 
л ые — —— хш 


Sehr rasch (J = ca. 104) 


Ut — — — Lo ДЕ ОТ И 


a? “ча | 
кс | ill | 
ГІЗІ 
и: ij | FHA чү 
1. db A jj 
Nisi “ТЕЙ MN. STER 


иар ee f 


Е - 1 2 li { | 

3 Е 5 Е 3 3 i X 

E КР, "i Ц, Ж 
ET П 


54 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 55 


v) SKRYABIN (1907 
е) ( ) 4D espressivo SS === 


FL 


Ob. 
СА. 


factual sounds) 


Вп. 


Horns НА 
(actual sounds) 


Cur of mm 


d "PM T 
А gp се E MEILE 


D) 
ланат тад: F oF | Elgi ERAN 


Violins 
0 = b 
an : = 
Ц 


ЕЕ ЕЕ 
CSV 


РЕ ы: в VI ЖЕЕП а а a 1 за 
(div.) 

GE on лае ТЕА 

Е-Е НЕЕ Нн Ее НЕ Еа 

GERE | RS СВЕН CD ee eS 1 -[9^ Se ea) 


1890-1914 


56 THE APOGEE AND DECLINE OF ROMANTICISM: 


ү c 
ПТА LH 

; а 

E 55 


dim. 


4 
Е@5 


Дъ 


-ASD 


a. 
(actual sounds) 


9 
ЕА 
S—-s 
Dp 


(actual sounds) ЕК6 


Horns 


Violins 


57 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 


ER 


sehr zart 


Т ЕЕЕ 
( i [Е | 
| | m 
^g M И 
ШЕ В Ш 


Таен еее леті, ey легі 
[p ЕТ Cen ee 
Е-е: тазға [n атала) 
ves 
Be e] 
ЕЕ cH т. ЧЫГЫ 
eee осла над! 
p] 
44 
Il 
ІП 
a сшкш EIE] 
=ске Sn 
== ст Енен 


Е |: 
US 
+ | $ | 
| К 5 р 
| ‚ || (| 
\ 
al UM |ы Ш T A И 
1 Pa. БШ 
|| ООРУ. | 
М + а пу Б е 
Ф © 
2 5 3 & S пре 
E ага - Е ФЕ БЕ 5 
г S 5 | “5 Е М L| e = [TN 
Ща К | x d 1 теі? Ih 
29 | 8 " 9 | ч H | ЫҢ M 
= Sé 842525 3525 № A 
чии d JH TRY НО MIN т du 
m = 
га 9 
щ Y ай 2 
> 5 E е 2 Е 5 
е E д Е "n Е > 
S 


mit Dampfer 


58 THE 


Woodwind 


Horns 
(actual sounds) 


Harp 


2nd Vins. 


Violas 


Cellos 


APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 
Fi. I. =. + 


| 
№. 
| 
cp 
т 
е) 


и Ц а Du — — В, EEE шен ішін? | ЖЕТ. um 

= E И С О ПБ ЕЕ (Жей 78) ces maf ЕЕ 

ЕЕ era ШЕ а Н И а 
ж-ы] 


| 
| 


Le) т ана | ЕЕ 
MI c xp wx Ко есы т SS тата” 

[nec yum eem: mae onm TREE Rr ЧГ жй БЫН Pe лысы праи чау Ам] 

puce WO [loe ee eni er] D E oo cmm Pe) 


I9 


== 77 
dall Joc Ge Банан ШЕСТ За Еа 
въз! 


"ДЕЛЕ. У ee SSS SS] 
J | ey SSeS ET ee эң! er. E 
[p co ce ее D _---— 
pcm] T qa Lud 


3 3 
б paren = 3 
{нур E-— ———] 
SS 


ЕР” ee Te a (|7 /ТБАУЕ) ey —a- 
— ү 


(+ Ю.В.) pizz. 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 


(vii) STRAUSS (1909) 
Sehr schnell (2 — 176) 


Woodwind 


Violas 


Cellos 


agitato 


OCTAVIAN 
(hesitating): 


Wenn er mich dort 


59 


1890-1914 


60 THE APOGEE AND DECLINE OF ROMANTICISM: 


MARSCHALLIN 
(pleading): 


ПИ 
ШЕТІ 
ИҢ 
шы | 

| g 
| 


The - res? 


wird aus Dir, 


NN 


Was 


er-wischt, 


61 


THE MUSICAL LANGUAGE OF LATE ROMANTICISM 


= 
Е] 
[jum 
ganz 
ye 


Selmer 


(stamping impatiently): 


MARSCHALLIN: 


il 
ah [А 
Sl 
FN 

ЕЕ! 
| Ы 

P (аят ji: 
i i (ЕЕ: 
I| s | 
[| Ian 

ы 

Шш S UR 
PADO p. 

n D i {г 
ші ШӘ 
В 
a 

e 

ea 
e d 

5 Y 


$i 


The - res! 


OCTAVIAN: 


mein Schatz! 


pronum] 
ІТ есен) 
2 а 
- steck Er sich, 


гр" 


62 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


Ob. I, 


пеш 


Же TIE] Ише; 
=== = 
Но. I & И => 
() 
[ =a 
EY «nS ұста eee eee ey eel НЕТ 
Hm (реле атыла ee Е DC cce p eres] УА 
е а 
still! Das 
ы а > ет 
3 ПЕ” à» 
һа Т^ о ара г) 
raam; vs Не Ы 
as ет са 
Е Б = 
CKESC ж м E 
[-3153— —— = мс. 
LECT eee eee. cnr 
IL BI D рас ae & ee c I 3 (сте ДЕ. 
LACER [4 ЕЕ] 
Grese а 28 2 7, 
de. а 
Gey Е Е 190 gl 
г eS SS 
Ea: 5 
RIA 2—7 
Себет" SS а - 
“а Ю.В. pizz. & Титр.) 


(Oct.: And don't move! If he catches me there, what will become of you, Theres? 
Mar.: Hide, my darling! Oct.: Theres! Mar.: Keep quite still!) 


THE TEXTURES OF LATE ROMANTICISM 63 


These passages are characteristic of the musical language of what may 
reasonably be called the main stream of late romanticism, ‘language’ not 
merely in the sense of ‘technical idiom’ but in the sense of conveying— 
essentially in the initial dropping interval with its suggestion of a 
caress—a definite emotional significance. In all but two of them, the 
composers tell us we are correct in understanding his music as an 
expression of love or tenderness: (1) is from the Prologue to Fervaal, 
where the wounded hero first turns his eyes to the princess Guilhen; in 
(ii), from the first movement of Mahler's Fourth Symphony, the Wagner- 
ian melodic language is clear beyond doubt; (iii) opens the section in the 
Kossuth Symphony headed ‘What sorrow weighs on your spirit, dear 
husband?'; (iv) is from the first section of the Kammersymphonie; in 
(v) from the Poem of Ecstasy the theme has just been played by a solo 
violin, with the marking dolce espress. carezzando; in (vi), the thirteenth 
variation of Webern's Passacaglia, the markings also speak for them- 
selves, as does the anxiety of the lovers in (vii) from the opening scene of 
Rosenkavalier. In each case the sharply defined initial motive generates, 
or dissolves into, a more or less indeterminate non-motivic continua- 
tion. A more motivic continuation of the same basic idea occurs in the 
second movement of Roussel’s D minor Violin Sonata, Op. 11(1907-8) 
which comes particularly close in some of its later appearances, such as 
Ex. 22, to the Skryabin version. 


THE TEXTURES OF LATE ROMANTICISM 

These passages also repay study from the point of view of texture: 
the harmonic bases, the warp and woof of the music, and the instru- 
mental colouring of the strands. Although selected in the first place 
solely on the basis of similar emotion expressed melodically, they have a 
number of common characteristics beneath their very obvious differ- 
ences. 

The harmonic pace is moderate and equable, usually changing with 
the bar. And the basic chords, considered simply as agglomerations of 
sounds, are generally unremarkable; even Bartók's ‘Hungarian’ 
melodic line is borne on very ordinary harmonies before it collapses in 
the Wagnerian cadence at the end of bar 4. But these basic chords are 
thickly overlaid with passing-notes, appoggiaturas, and less easily 
explicable ‘foreign’ notes, and they are more easily related to each other 
as a simple concatenation than functionally and with reference to a firm 
tonal centre. ‘Key’ is obvious enough from point to point but is seldom 
orientated to a more distant horizon-mark. Only d'Indy's F sharp 
minor and Mahler's G major have any real meaning; Strauss's B 
minor and the D major to which it leads are only passing incidents, as 


64 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 
Ex. 22 b 


Assez animé (J = 120) 


the one-flat key-signature suggests—yet that one-flat signature is really 
valid only for two passages, each of a dozen bars or so, separated by 
nearly three hundred bars of non-F major. Webern's 8-bar variation is 
a miniature version of the same thing; it begins and ends with tonic 
chords, buttressed by augmented triads which function as substitute 
dominant sevenths (F natural replacing E-G), but the intervening 
harmonies can be related to D major only by ingenious sophistry. 
In itself the chord of superimposed fourths in bar 5, F-B-E flat-A flat, 
is an instrument of tonal dislocation, as are the augmented fourths/ 
diminished fifths of Skryabin's bass. Compared with these passages of 
Webern and Skryabin, that from Rosenkavalier is strikingly conserva- 
tive, but it should be remembered that Rosenkavalier in general is 
harmonically more conservative than its predecessor, Elektra; Strauss 
was an older man, and Rosenkavalier marked the beginning of his 
creative climacteric. 

Despite this contrast of harmonic vocabulary and of Strauss's al 
fresco writing, such as opera demands, with the jewellers’ work of 


THE TEXTURES OF LATE ROMANTICISM 65 


Mahler, Skryabin, and Webern, the basic texture is the same in all 
four—and in the other examples: through or over the harmonic 
background are drawn one or two horizontal strands which may or 
may not be related to each other. But, except in the Schoenberg passage, 
these lines do not interact on each other and propel the music forward, 
any more than the harmonic successions give a sense of progression 
toward a goal. The vitality of the music lies in the nervous energy of the 
individual lines themselves; it is thus easily exhausted and climaxes tend 
to be achieved artificially, not to say mechanically, by *7ristan- 
Steigerungen’ and mere volume of sound and elaboration of texture. 

The finest craftsmanship of the late romantics was expended on the 
dovetailing and instrumental colouring of the horizontal lines, whether 
of primary or secondary importance—Hauptstimmen or Nebenstimmen, 
as Schoenberg was to distinguish them some time after the Kammer- 
symphonie. Indeed dovetailing and colour-shading are often part of the 
same process, as may be heard at its most subtle in Skryabin's wood- 
wind writing in Ex. 21 (v) or in Elgar's shattering tutti in Ex. 27 (on 
pp. 76-7). The amount of line-drawing entrusted primarily to the wind, 
whether or not supported by strings, is very characteristic of the period. 
And not only woodwind or horns: eighteen bars after (v) Skryabin gives 
his principal melodic line to two trumpets in unison, f та dolce (and 
the secondary line to solo horn and cellos). Egon Wellesz has traced this 
way of employing the trumpet melodically, in the later Strauss, early 
Berg, and elsewhere, not only ‘at moments of heightened brilliance, 
power and solemnity’, from the example of Mahler.! He goes оп 
perceptively: 


This is connected with the endeavour to heighten the intensity of the sound 
without drawing on the full orchestra, with the need to give the melodic line 
the highest degree of emotion without interfering with the clarity of the 
line-drawing. In this tendency ultimately one can see a romantic element, 
that of over-playing one's hand (Uberpointierung). The romantic musician— 
as one may already observe in Liszt—is not content to state objectively the 
values latent in the musical substance but, through the instrumentation, adds 
to it from his own subjective—one might almost say /iterary—feeling a 
wish-idea more complete, richer, more fulfilled than the apparent sound. 
With Mahler this conflict between will to represent and what is represented, 
typical of all romantic creativity, is partially resolved by his ability to extend 
the limits of invention through the colour of the orchestra. 


And he illustrates this with a passage from the first movement of 
Mahler's Fifth Symphony: 


! Die neue Instrumentation (Berlin, 1928), i, pp. 106—7. 


66 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


Ex. 23 


Piótzlich schneller. Leidenschaftlich. Wild 


Vins. I only 


Trumpet I 
Trombones 
Violins 


Horns І, И, Ш (+ 3 Ob.) 


gestopft г-3-- 


(+ Vins. II) 


| 
В 
| 
| 


Q 


THE TEXTURES OF LATE ROMANTICISM 67 


PHP o aN (WW. tt 
9—8 
Ро ЕО SS SS ш 
SS лаза; S сло ee ey 
[n.o “Берер ee [ЕЕ I га 
ашт - | irs ШЕ е а 
mE 252 iS БЕЗ М ee es жаа жыны Go eae Бо 
fas b+ срез Sa Se SS SS SS 
<р пата ЕН аланы eR Бо) eer ees ізі шаг 
2 b > 8 Gt - € са 
ri s pee Ee” ee сы р юр ә 
ор № 1-09 72 TA Pia 12 ФЧ 0—1 
SN. Р да —— — os НУ А Ішсе |і ДА T экс [иг [— ВН A 
en R Ea D ames a СЕ rea p ЕЕ ee p ppm. [mp „р етуу 
— ff? > 3 Г 
f) = Иа p ро — —38 3 cul 
= МЕ SSS RE n- Be Ган ГГ ee 6 0 Е 59 
en ы» ЗГ DEI. 
С Ree S ж шй! шнш Sa S18 -4 „у= ши гиа nest ecd рез E eS 
pus [eo To ee ишш) рип e p И ПБ [— [Emp IM 22 Уат 


| ътла 
а! ОПЕРА -------- ------ 
aan S c = ЕРЕ SSS UR USE 
mem УЛ ES Se (Sl [ah ———3а 
m. ms Bm -м. 


Ше а! fresco scoring of which is as far removed as possible from 
Skryabin's exaggerated finesse. But Mahler and Strauss, indeed Sibelius 
and Elgar and the more conservative composers of the period generally, 
thought directly in terms of the orchestra, while their slightly younger 
contemporaries were not only affected by Debussy's pointillisme— 
Skryabin had a score of La Mer at his elbow, as well as Ein Heldenleben, 
when he was orchestrating Prometheus‘—but tended to translate 
pianistic textures. The instrumental refinements of Ex. 21 (v) reflect the 
exquisite refinements of the composer's piano-playing. The lay-out of 
Webern's variation reminds one of some late Brahms piano-piece, and 
that of the Schoenberg passage, Ex. 21 (iv) corresponds to a great deal of 
nineteenth-century piano-music: a vital top and bottom, with a filling-in 
of conventional figuration. 

Nevertheless, although orchestral writing did not always originate in 
purely orchestral thinking, this was perhaps the last great age of 
orchestral composition, distinguished by the hectic colouring of 
decadence. The orchestra—often a giant orchestra, even for tiny pieces 
such as Webern's Op. 6—was the favourite medium of the later 
romantics, including such doubtful romantics as Ravel. It was the 
medium through which they—and their half-brothers the impressionists 
—depicted with the utmost possible realism the adventures of Don 


1 Sabaneyev, Vospominaniya, p. 54. 


68 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


Quixote and Falstaff, painted seascapes and erotic fantasies, expressed 
immortal longings, and exposed the tragic secrets of their tormented 
souls. It dominated their operas! and influenced even their chamber- 
music and piano-music. Their characteristic chamber works are those 
which seem to aspire to the condition of orchestral music: Reger's C 
minor Piano Quintet, Florent Schmitt's Quintet, Verklärte Nacht which 
Schoenberg himself afterwards arranged for string orchestra, just as he 
made two quite different full-orchestral versions (c. 1914 and 1935) of 
the first Kammersymphonie. The string quartet, in spite of its adaptability 
to almost every change of stylistic fashion, denied them not only wide 
range of colour but full, saturated sound. Most of them ignored it and, 
after the César Franck (1889), the Debussy (1893) written to some 
extent under its influence, and Dvorák's last two masterpieces, the 
form was almost totally neglected by significant composers until Bartók 
and the Schoenberg group turned to it—and, as Schoenberg himself 
admitted,? his First Quartet (1905) was written in too ‘thick’ a style, 
from which he found his way back to a true chamber-music texture 
only gradually. (Webern made a string orchestral version of his 
own Fünf Sätze for string quartet (1909).) Even the exceptions, of which 
the most notable is Ravel's, have many passages that strain toward the 
orchestral. 

The attitude to solo piano music was also symptomatic. No other 
composer was as wholesale as Ravel in the later orchestration of his 
own piano pieces, but there was a widespread tendency, stemming more 
from Liszt than from Brahms, toward quasi-orchestral effects. Some- 
times this amounts to little more than a full romantic sound; sometimes 
it shows in dense and complicated textures which, like those of their 
orchestral counterparts, are not always necessitated by dense and 
complicated thought, but which, when examined in detail, often reveal 
more subtlety and acute sensibility than the listening ear can readily 
take in. This tendency toward very full, quasi-orchestral piano-writing 
is the highest common factor of otherwise totally different styles? 
(Ex. 24) and is fatally deleterious when reflected in organ-music. 

It appears also in the overloaded piano-parts of songs, though in this 
field the quasi-orchestral was now often replaced by the really orchestral. 
Wolf orchestrated the accompaniments of three of his Mórike-Lieder in 
1889, at least fifteen songs the following year, and others later. At about 
the same time Rimsky-Korsakov began to orchestrate songs originally 
composed with piano; Grieg followed suit in 1894; and in 1897 Strauss 


Ис Сар. 9: ? Newlin, op. cit., pp. 215-16. 
з Cf. also Exs. 28 and 29 on pp. 82-4. 


THE TEXTURES OF LATE ROMANTICISM 69 


Ex. 24 
(i) SKRYABIN: Satanic Poem, Op. 36 (1903) 


Allegro (4. = 92—108) 
a R 
Е. * x. 


AR г. 
[ege o =e 
-—— Пр 


с=с 
Pg. | ЕЕЕ на 
por neun Д p o - 4 3 


= 234535 = — 
ee 


p es 


ІШ 
[үр 


| 
Mia 


1890-1914 


70 THE APOGEE AND DECLINE OF ROMANTICISM: 


(ii) SCHOENBERG: Op. 11, no. 3 (1909) _ 


ТЕ 
1 ^ 


кектене im 


ып а Ур 


mentre 
таңына | 
р 


Ses 


T ms 
жин ш 


s 


кесер»: 
Wwe wg | #1 
WU те 


ус]; ез) 
Miri 


71 


THE TEXTURES OF LATE ROMANTICISM 


(iii) GRANADOS, 'Coloquio еп la reja’ (Goyescas) (1911) 


Andantino allegretto 


i | 
E 
' 


rinf. e sosten. 


bebe 2 ue 


DE | 
[к= 


72 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


orchestrated ‘Cäcilie’ and “Могреп” and composed the Vier Gesänge, 
Op. 33, the first of his ‘songs with orchestra’, the genre with which he 
took so touching a farewell to life in the Vier letzte Lieder half a century 
later. But it was Mahler who won recognition as the supreme master of 
the Orchesterlied with his Lieder eines fahrenden Gesellen (with piano, 
1884; orchestrated in the early 1890s)!, Lieder aus ‘Des Knaben Wunder- 
horn’ (1888-99), Kindertotenlieder (1904), Fünf Lieder nach Riickert 
(1904), and above all Das Lied von der Erde (1908) in which the song- 
cycle with orchestra is expanded to symphonic proportions and actually 
styled ‘symphony’. In the same year as the Kindertotenlieder and Rückert- 
Lieder Schoenberg wrote his Sechs Orchester-Lieder, Op. 8 (including 
two Wunderhorn settings, one of which, ‘Sehnsucht’, sounds oddly like 
distorted Mahler). It is significant that Ravel, despite his mania for 
orchestration, never scored any of his songs; that Debussy did so only 
once (a late version, in 1907, of a very early song); and that, whereas 
Alban Berg wrote his Altenberg-Lieder (1912) for voice and large 
orchestra, his friend Webern abandoned piano-accompaniment only in 
favour of small chamber ensembles. The Orchesterlied had a romantic 
ancestry in Berlioz and it was essentially a late romantic phenomenon. 


PROBLEMS OF STRUCTURE 


The opulence of sound which is the most characteristic symptom of 
romantic texture in its last phase was paralleled by cultivation of 
grandiose structure, and the organization of compositions on a vast 
scale presented problems even more difficult than usual to a generation 
that was in the process of devaluing by inflation the tonal and harmonic 
currency it had inherited. Coherence of detail was carried further than 
ever before, but it was a coherence which generally lacked that vitalizing 
quality which may be grotesquely compared with the muscular activity 
of the gut: the tiny thrusts and pulls of appoggiaturas and inter- 
dominant chords, the magnetism of local tonics and vestigial cadences. 
The outward appearances of all these remained but they had lost their 
tension-relaxation effect through over-use and through the general 
raising of the norm of dissonance. The effect of dissonance could be 
achieved only by further intensification of dissonance or by ‘illogical’ 
progressions, in which it appeared that any chord might follow 
any other and which consequently disrupted inner cohesion. Infra- 
structure still depended either on the weakened remains of the familiar 
small-scale symmetry of lyrical phrases, which in Skryabin’s totally 
thematic miniatures sound mechanical, or on asymmetrical ‘prose’, 


+ СЕ. Donald Mitchell, introduction to Alma Mahler, op. cit., pp. xviii-xxi. 


PROBLEMS OF STRUCTURE 73 


which in Webern's non-thematic miniatures sounds incoherent. And 
these were the two master-miniaturists of late romanticism. 

АП the same, miniature forms can exist as aurally comprehensible 
forms without key. The weakening of key-sense was much more serious 
in its effect on large-scale forms. Indeed very large forms in themselves 
weaken key-sense so far as the listener is concerned, even if he possesses 
absolute pitch. But a key-plan remained important, or at least useful, 
to composers in laying out music on a vast scale, even if the central key 
no longer exercised any gravitational pull and no normal listener could 
recognize that the composition ended in the key with which it had begun. 
Few listeners notice that Debussy's Nocturnes begin in B minor and end 
in B major, that the Symphonia Domestica is ‘in’ Е major and the Poem of 
Ecstasy ‘in’ C major; that Salome begins in C sharp minor and would 
have ended in C sharp major if Herod had not commanded Salome's 
execution eleven bars before the end. Key-organization of this kind is 
so unreal as to deprive the so-called ‘progressive tonality’ of Mahler 
and Nielsen! of all meaning or impact; but it was a practical con- 
venience in architectural planning rather than conformity to an 
obsolescent convention. So, when all but the last vestiges of key had 
disappeared, a composer would bring back his second subject at a 
different pitch in the recapitulation (as Berg does in his Piano Sonata, 
Op. 1): 

Ex. 25 


Ex. 26 


1 See pp. 7 and 38. 


74 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


though neither stands in any realistic functional relationship to the 
nominal tonic key of the movement (B minor). 

When key had become nothing more than a point of reference on a 
blue-print, the only structural element apparent to the listener was 
theme, and musical architecture became more and more a matter of the 
treatment of thematic material, even of texture. The quasi-spatial 
conception of musical form began to decay, not as the immediate 
consequence of the decay of tonal feeling but as a later consequence. 
In the end form might almost be defined as no more than ‘essential 
relationships of material’. Distinctions of nomenclature, as between 
‘symphony’ and ‘symphonic poem’, often became as meaningless as 
key-names: the Symphonia Domestica, though it includes slow move- 
ment and scherzo, is indistinguishable from a Tondichtung. The real 
difference was now not so much between symphony and symphonic 
poem as between one kind of symphony and another. The very term 
‘symphony’ had long before the twentieth century become as loose a 
term as ‘novel’, which can include A /a recherche du temps perdu as well 
as Madame Bovary; and the basic distinction has little to do with 
general lay-out. A symphony might be as taut and compressed as 
Schoenberg’s Kammersymphonie or as loose and sprawling as one of 
Mahler’s symphonic *worlds'—or as superficially conventional as one of 
Elgar’s. The fundamental distinction between one kind of symphony 
and another lay in the presence, and degree of presence, of a quasi- 
dialectic element—the processes by which musical ideas generate each 
other, not by obviously mechanical transformation, but by seemingly 
organic growth of tissue. These were the processes, sometimes compar- 
able with those of the logical syllogism, sometimes in the nature of 
what Schoenberg called ‘developing variation’ (entwickelnde Variation), 
which had constituted the main tissue of instrumental musical thinking 
since the Viennese classics; its presence in nineteenth-century music had 
been considered a strength, its absence a weakness. In the decade before 
1914 it suffered its final crisis. 

In the hands of the out-and-out romantics musical consequence and 
contrast became less important than emotional, or otherwise extra- 
musical, consequence and contrast. In the vast area of a Mahler 
symphony all kinds of disparate and disconnected material could be 
employed to record and communicate successive states of mind. At the 
other extreme, composers as different as Schoenberg and Sibelius (whom 
Schoenberg recognized as having ‘the breath of a symphonist’)? by 
tightening thematic procedures as the one did in his Third and Fourth 
Symphonies (1907 and 1911) and the String Quartet (Voces intimae) 


1 Style and Idea, p. 185. ? [bid., р. 195. 


PROBLEMS OF STRUCTURE 75) 


(1909) and the other in the Kammersymphonie (1906) and the first two 
numbered string quartets (1905 and 1908), were either consciously or 
unconsciously gradually sacrificing emotional meaning and consequence 
to musical constructivism. At the same time, by packing more and more 
intense significance into confined musical space, they brought instru- 
mental coherence to the edge of a precipice from which Sibelius turned 
back in his Fifth Symphony but from which Schoenberg boldly leapt 
in the K/avierstücke, Op. 11, no. 3 (1909). In these pieces, as Webern 
wrote, ‘Once stated, the theme expresses everything that it has to say; 
it must be followed by something new.’ 

The dilemma posed by intense concentration of expression may be 
illustrated by a passage from an extremely different, intensely romantic, 
work completed eighteen months later: Elgar's Second Symphony. Here 
into two bars of the Larghetto (Ex. 27), sketched seven years earlier 
under the immediate impact of great sorrow, Elgar compressed an 
overwhelming expression of naked and quivering grief. Everything is 
said, musically, in two bars; but Elgar, an expansive and improvisatory 
genius rather than a great musical architect or thinker, can only 
repeat them with different scoring and then let them run down in 
a feeble continuation. The whole passage remains an incident in the 
movement, though an incident repeated later with heightened hysteria 
but no intensification of musical expressiveness. As a large-scale 
instrumental composer, in his symphonies and concertos, Elgar repre- 
sents the typical conservative late-romantic compromise, submitting 
autobiographical expression to the discipline of at least the appearance 
of more or less traditional form and integrating improvisation by the 
Franckist devices of theme-transformation, sometimes very subtly, 
sometimes with mere mechanical ingenuity, as in the transformation 
of the Allegro molto semiquavers of the scherzo of the First Symphony 
into the main theme of the Adagio. 

Elgar’s conservatism is equally apparent in his oratorios, which were 
‘modern’ only in relation to the form itself. Neither his creative gifts nor 
his technical innovations could save oratorio from the death that 
overtakes all outmoded art-forms, as he recognized by abandoning the 
third part of the trilogy begun with The Apostles (1903) and The 
Kingdom (1906). The only comparable works of any importance in this 
period were two nearly contemporary ‘monumental’ compositions for 
vast forces: Delius’s Mass of Life (1905)—the much shorter Sea-Drift 
would have been described as a Ballade half a century earlier—and 
Schoenberg’s Gurrelieder (composed, though not orchestrated, in 1900) 
where even the triple male chorus is not heard until Part III and the 


1 *Schónbergs Musik’ in Arnold Schónberg (symposium), (Munich, 1912), p. 41, 


76 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


Ex. 21 ч 
Nobilmente e semplice a3 


meae Ier E EE Be 
Тре. ш Саъ 4 — 04 po ж —— m ox ша unum е с ы ы онсе и АШ =н 


Tbn ШК раи a eS Eee 
СА И 2528 OESS EAEN, OE eR КЕ c 
rro E 


Tuba |2 


Timp. 264—7 


= 
JG SE? Sn Pe 0 c. x 
Vin. р x24 gua m: 
GN PAS жн ы Ге ЕЛІН аа. елени 


Vc. 


D.B. 


FRYUBDLEMS UF SINUUCTITURE 


tu 


Edu] 
шеті 
=> 
p | жо mex шее шты 
БГ] 22 B [к Жо [сез е Ееее аа 
БАС СІР m — ЦИ FA пе ---- а 
a наа Е я 
O > 
I, IT Eb. IF =>. ж? 
реж 
ЕЕ РЕ Е ето 4 
EZ a S. 


127 A о ЕЕ 

[7 S IL 1-24 
HESY DUS 

ESE Fo 


II 
mf molto cresc. 
P molto cresc. 


------------- 


ЕР rp 
Г’ АТО НЕЕ ыы ссср т т-тү Әр ue - Ro] 
сл ил uM — 3 — .—— а) | 
NU cu m к т ЖЧ 

2. 


[ 477 Ф "1 In M edis 


78 THE APOGEE AND DECLINE OF ROMANTICISM: 1890-1914 


full chorus only at the very end ofthe work. (The two works have а 
certain stylistic affinity; not only the orchestral introduction and final 
chorus of the Gurrelieder but such passages as Tove's “О, wenn des 
Mondes Strahlen leise gleiten’ could easily belong to A Mass of Life.) 
The relative unimportance of the chorus in the.Gurrelieder is character- 
istic of the period; large-scale works involving chorus, with or without 
vocal soloists, tend to be fundamentally orchestral. The chorus is 
incorporated into symphonic or quasi-symphonic frameworks, as in 
Sibelius's Kullervo and Mahler's Second, Third and Eighth Symphonies; 
or, as in the Gurrelieder—following the precedent of Liszt's Faust 
Symphony rather than Beethoven's Ninth—composers bring in the 
chorus to clinch the finale: Skryabin's First Symphony (1900), Delius's 
Appalachia (1902), Busoni's Piano Concerto (1904), the last section of 
Roussel's Evocations (1911). It was perhaps the example of Mahler's 
Eighth (1907) which suggested the employment of voices throughout 
symphonies organized in the conventional four movements: Vaughan 
Williams's Sea Symphony (1910) and Rakhmaninov's Kolokola (The 
Bells) (1913). Even more symptomatic was the more and more frequent 
use, after Debussy's Sirénes (1899), of a wordless chorus simply as 
a part of the orchestra either for the sake of its sound-colour—as in 
Ravel’s Daphnis and Delius’s Song of the High Hills (both 1911) and the 
end of Holst’s Planets (1917)—or merely to add to the volume of sound, 
as seems to be the case in Skryabin’s Prometheus (1910). 


END OF AN EPOCH 


The period that began with Falstaff and Hänsel und Gretel, the New 
World Symphony and the Pathétique, Brahms’s Clarinet Quintet and 
the Spanish and Italian Liederbiicher of Hugo Wolf, and ended with 
Rosenkavalier and Petrushka, Prometheus and Das Lied von der Erde, 
the period which includes practically all the best work of Strauss, Mahler, 
Reger, Elgar, Delius, Puccini, Rakhmaninov, Skryabin, Debussy, 
Busoni, appears heterogeneous enough when one examines it in detail— 
still more heterogeneous if one draws in other names and works—yet, 
seen in historical perspective, shows an over-riding aesthetic unity. 
It was an over-ripe art that, so far as it has been examined in this chapter, 
owed almost everything to the past. Its sumptuous operas and sump- 
tuous orchestral scores were composed for a public that was wealthy 
enough to support large orchestras and great opera-houses, and that 
was also musically intelligent enough to appreciate superb singing 
not only of opera but of polished Lied and mélodie and to enjoy the 
refinements of instrumental chamber-music. It was possible for the 
composer of the period to use music as a language because he could 


END OF AN EPOCH 79 


count on his hearers’ understanding the language evolved particularly 
during the previous century, with all its sound-symbols and associations 
and conventions; he could also count on their making some effort to 
follow him so long as he could be seen to be logically extending and 
enriching that language. Conversely the real innovator knew that he 
must be prepared to face not merely misunderstanding, but the accusa- 
tion that his music was nonsense. 

Yet, for all its close correspondence to the relatively stable and 
prosperous European civilization that nourished it, its decline cannot be 
attributed to the war which dealt that civilization such a staggering blow. 
The social and economic conditions would never again be so favourable 
to music of that kind, but the symptoms of decay and exhaustion were 
already apparent years before 1914. Sumptuousness of sonority, 
hypertrophic harmony, emotional intensity, music-as-a-language could 
be carried no farther. The future of music lay in other directions, which 
other composers—indeed sometimes the late romantics themselves— 
had long been pointing to and exploring. 


П 


THE REACTION AGAINST ROMANTICISM: 
1890-1914 


By GERALD ABRAHAM 


INTRODUCTION 


THE forces which revitalized music during the rich and colourful 
decadence of romanticism were in appearance and in origins totally 
opposed to each other. One was conservative, the other radical. One 
had always held out against romanticism, though romantic composers 
such as Schumann had sometimes been fascinated by it and given it 
play in their music; the other came into existence actually as a form of 
romanticism. One was concerned with musical pattern and structure 
for its own sake, the other with sonority for its own sake, but even this 
was not really a common factor. However, the rejection of music as a 
rhetorical or autobiographical language ultimately brought them 
together; and the musician who wished to be a pure craftsman, applying 
seemingly outworn techniques of composition to the treatment of 
“ршеіу” musical ideas, found that the musician exploring new sonorous 
effects could provide him with new materials for the exercise of his 
craft. The aesthetic of the nineteenth-century conservatives was that 
expounded by Eduard Hanslick in Vom Musikalisch-Schónen:* ‘а 
complete musical idea . . . is an end in itself and by no means medium 
or material for the representation of feelings and thoughts' and composi- 
tion is not “һе translation of an imagined material (eines gedachten 
Stoffs) into sounds; the sounds themselves are the untranslatable 
primitive language (Ursprache) .? 


THE IMPACT OF BACH 

The conservatives were saved from technical stagnation less by the 
borrowing of devices associated with romanticism (such as theme- 
transformation) for purely musical ends than by the revival of baroque 
methods, largely as a result of the cult of Bach. But the impact of Bach, 
as he changed from a rediscovered historical figure to a living influence, 
was ambiguous. When Schumann composed Bachian fugues in his 


! Leipzig, 1854; sixth, enlarged and improved edition, 1881. 
? Sixth edition, pp. 65 and 194. 


BUSONI AND REGER 81 


Opp. 60 and 72, he was temporarily denying his essential romanticism, 
and the fugal ideas he acquired in these technical exercises enriched the 
musical vocabulary of his later years! more than they affected his 
technique. Many composers who were not 'absolute' musicians or 
conservatives in any sense learned technical lessons from Bach and even 
the idea of fugue could be used as a symbol—of science—in Strauss's 
Zarathustra. Indeed Bach himself was claimed as a romantic by the 
Alsatian Wagner-disciple Albert Schweitzer in his J. S. Bach, le 
musicien-poéte,? which, as Charles Marie Widor put it in his preface, 
revealed the chorale-preludes as not merely ‘modéles de contrepoint pur’ 
but a ‘suite de poèmes d'une éloquence, d'une intensité d’émotion sans 
pareilles’. In fact Schweitzer mistook the baroque symbols of Affekte 
for quasi-Wagnerian 'motifs de la démarche, de la quiétude, de la 
douleur . . . de la lassitude, de la terreur’. But it was Bach's demonstra- 
tion that ‘contrepoint pur’ could be living musical tissue, neither empty 
sound-patterns nor precise ‘langage musical’, which did so much to 
revitalize absolute music at this period. 

It was not from Bach but from Wagner, the Wagner of Meistersinger 
and Górterdámmerung, that Strauss and Mahler and the other late 
romantic masters learned their kind of polyphony: the polyphony of 
thematic lines, or thematic and subordinate lines, controlled by 
harmonic progressions. But there is a world of difference between the 
counterpoint which consists of the combination of themes all based on 
the same chord and the 'contrepoint pur' which, equally secure in its 
harmonic basis, propels itself by the thrusts and stresses of suspension 
and friction and resolution, between free polyphony employed to 
romantic ends and disciplined polyphonic thinking for its own sake. 
The practice of fugue or quasi-fugal writing not only lowers the romantic 
temperature—as Mahler demonstrates in the finale of his Fifth Sym- 
phony (1902)—but provides a firm bone-structure in place of flaccid 
tissue, just as the passacaglia technique provides an element of formal 
discipline. Nor was this merely a matter of structure, important only to 
the composer himself; he wished the listener to listen to the details of 
the music instead of allowing himself to be immersed in washes of rich, 
emotional sound. 


BUSONI AND REGER 
The two masters of the generation after Brahms mostly deeply 


1 Wolfgang Boetticher, Robert Schumann: Einführung in Persönlichkeit und Werk (Berlin, 
1941), p. 572. 

2 Leipzig, 1905; later vastly expanded in a German version (Leipzig, 1908) which was in 
turn revised and expanded for the English translation made by Ernest Newman (London, 
1911). 


82 THE REACTION AGAINST ROMANTICISM: 1890-1914 


impressed by Bach were Ferruccio Busoni (1866-1924) and Max Reger 
(1873-1916), the first a Germanized Italian, the second the most 
Teutonic of Germans. Neither was by any means free from romantic 
strains.1 Busoni was a passionate admirer of Liszt; Reger's harmony is 
often hypertrophied to the last degree. Indeed: their Janus-faces have 
given them an historical importance beyond that of their original 
creative powers. They met in 1895, became close friends and ‘exchanged 
their compositions and piano-arrangements of Bach organ-works’,? but 
to what extent there was any real mutual influence, it is difficult to 
determine. Both adopted the contemporary attitude to Bach, as a 
master who must be rescued from the dry-as-dust academics and pre- 
sented to the public in contemporary terms with all the resources of 
contemporary instruments, an attitude which led to such inflated 
transcriptions as the conclusion of Busoni's piano version of Bach's 
organ Toccata in C major (BWV 564): 


= ш 
2542) 


1 See supra, рр. 14-15 and 32. 
2 Fritz Stein, Max Reger (Potsdam, 1939), p. 20. 


BUSONI AND REGER 83 


| 


They naturally carried this monumental style of piano-writing over into 
original compositions in which they paid homage to Bach: Busoni's 
Fantasia contrappuntistica (1910) and Reger's Variationen und Fuge 
über ein Thema von J. S. Bach, Op. 81 (1904). Near the end of the 
latter, a double Monumentalfuge,! Reger presents both first (right hand) 
and second (left hand) subjects not simply combined but with typically 
thick harmonies: 


1 An apt term coined by Emanuel Gatscher, Die Fugentechnik Max Regers in ihrer 
Entwicklung (Stuttgart, 1925). 


84 THE REACTION AGAINST ROMANTICISM: 1890-1914 
Ех. 29 TN 


(4-60 


(4-72-76 poco a poco 


үй sehr breit = А. ЕЕ 
~ >> оты =з d 


sempre ти. Te iz 1 
y ЕР ЕА + кН РЯ 
ГА те 4 s Nea” oe T. i 
=: 2.8242 8424 ЗГЕ 


Adagio (4 = 30) 


Ae) 
ER РЕ 3 


But an excerpt from the exposition will show Reger's genuine, if always 
rather harmony-bound, linear feeling: 


Ex. 30 


BUSONI AND REGER 85 


espress. 


a severe style which is maintained until the third section of the fugue and 
which represents the Reger who influenced his younger contemporaries, 
including Schoenberg. (‘Schoenberg had always been interested in the 
music of Reger and had admired many things in it ;! when Schoenberg 
wrote of ‘the great masters of our time’ he habitually included Reger 
among their number.) Reger’s sets of variations—Op. 81, the Beethoven 
Variations, Op. 86 (1904), the Telemann Variations, Op. 134 (1914) for 
piano, the Hiller and Mozart sets for orchestra, Op. 100 (1907) and 
Op. 132 (1914)—all end with fugues, as do a number of his other works, 
and nearly all, like Op. 81, with fugues whose dynamic plan reflects his 
own way of playing Bach. He would begin a fugue almost inaudibly and 
end with a triple forte; his own fugue in the Bach Variations begins 
pp (una corda) and ends ffff, and those of the Beethoven, Hiller and 
Mozart sets, after a mf or sfz call to attention, begin pp and proceed 
to a massive, heavily reined-in conclusion. There are highly romantic 
changes of tempo, and almost every entry of the subject is marcato in 
accordance with the common way of playing Bach at that period. 

Not all Reger's very numerous fugues lean so heavily toward the 
romantic, though even the less monumental often preserve the basic 
conception of an overall crescendo. Near the very end of his life in the 
Preludes and Fugues, Op. 131a, for solo violin and Drei Duos ( Kanons 
und Fugen im alten Stil) Op. 131b, for two violins (1914), he did achieve 
mastery of pure and classical line-drawing. Yet the lasting significance 
of Reger's copious fugue-writing lay not in his romanticizing of the 
fugue—he had begun, in the organ pieces of Op. 7 (1892) by imitating 
Bach without harmonic inflation—but in the fugal discipline he imposed 
on romantic music. Here the intellect reasserted its power to think in 
music rather than about music; and his later works in general, notably 
the Telemann Variations and the Clarinet Quintet (1916), recapture 
genuine classical feeling as well as classical techniques. This was no 


! Dika Newlin, Bruckner—Mahler—Schoenberg (New York, 1947), p. 275. 


86 THE REACTION AGAINST ROMANTICISM: 1890-1914 


more than a personal achievement, however. It was reached too late to 
affect Reger's contemporaries, among whom classical feeling—balance 
and textural clarity, emotional restraint, and the other qualities com- 
monly subsumed under the idea of Ше classical—was cherished by 
composers, particularly Ravel and his compatriots, who were little 
concerned with the techniques of polyphonic discipline and harshly 
intellectual harmonic schemes, and in any case knew nothing or little 
of Reger. (When Honegger went to Paris in 1911, “ёги de Richard 
Strauss et de Max Reger, he found the latter completely unknown 
there.) On the other hand, those who were so concerned showed little 
sense of classical ideals, despite their increasing preoccupation with 
music and its technical devices for their own sake. 


NON-FUNCTIONAL HARMONY 


The cultivation of composition-techniques for their own sake was 
fecundated only by contact with the cultivation of sonorities for their 
own sake, which originated not in any conception of absolute music 
but in typically romantic attempts to extend the expressive and sugges- 
tive or descriptive language of music. A characteristic example is the 
added note: two notes, a minor or major second apart, are sounded 
together instead of only one of them, as at the end of the voice-part of 
Wagner's song ‘Im Treibhaus’ (In the hothouse) (1858). The piano 
suggests the ‘heavy drops’ by minor seconds, D-E flat, which can be 
explained in terms of tonal harmony in conjunction with the G-B flat 
of the left hand but which the listener irresistibly hears as irrational 
sounds, a blurred Naturlaut as Mahler might have called it: 


Ex. 31 


Langsam und schwer 


(On the green edge of the leaves.) 


1 Је suis compositeur (Paris, 1951, p. 128; English translation by Wilson О. Clough, 
London, 1966). 


NON-FUNCTIONAL HARMONY 87 


And Wagner proceeds to resolve the E flat on to E natural, producing a 
normal added-sixth chord. Similarly, ten years later, in the ninth of his 
Romanzen aus Tiecks * Magelone', “Ки е, Süssliebchen', Brahms evoked 
the magic indistinctness of woods at night by a soft syncopated major 
second, persisting for five slow bars, but immediately explained by its 
context as part of a dominant seventh and soon given an orthodox 
resolution. By an extraordinary coincidence Borodin had already, in 
1867, also composed a song about a fairy-tale princess sleeping at night, 
surrounded by forest, 'Spyashchaya knyazhna’ (The sleeping princess). 
His song also is in A flat and it also begins with the same syncopated pp 
major second, D flat-E flat, casting the same drowsy spell; but the 
major second is not part of dominant seventh harmony and instead of 
being resolved it moves to another major second: 


Ex. 32 
Andantino 


su glu -khom, 


(In the dense forest sleeps,) 


Syncopated major seconds, constituting a single ‘line’ of the piano part, 
continue uninterruptedly for fourteen bars and are only briefly and un- 
conventionally resolved; major or minor seconds continue to play an 
extremely important part throughout the song. This is something 
fundamentally different from Brahms; it is often possible to analyse 


88 THE REACTION AGAINST ROMANTICISM: 1890-1914 


Borodin's individual chords in terms of conventional harmony and the 
first sixteen bars are in the purest A flat, except for a couple of flattened 
leading notes; but it is impossible to hear the chords in that way. The 
harmony is not functional harmony. Every chord is blurred by the 
added note in the highest part; the two G flats neutralize the faintest 
suggestion of a dominant pull. It is sound without conventional musical 
sense, sound relying on its purely sensuous impact to produce a quasi- 
mesmeric effect; the seconds ask not to be resolved but to be savoured. 

In the setting of the next stanza, which tells how the noisy rout of 
witches and wood-demons flies over the princess without wakening her, 
Borodin even more thoroughly destroys the functional property of 
harmony by a powerful whole-tone scale descending through two 
octaves: 


Ex. 33 

f) Я. dim. rallent. 
E e ----- Dre mE == 
И 7 22 [De 


ved'm ile -shikh shum-niy roy i pro-mchalsya nad 


ш a uy PR 
та 


ЕО 

Dr БАРНА a аа 

L—D «p ПЕ ПЕНН Е ТЕЧ 
== | DD dp Аве 90 


although the firmly outlined subdominant chord prevents any feeling of 
atonality and only the first four notes of the scale (and their repetition) 
are harmonized in the whole-tone mode. Here again the intention is 
suggestive: from Glinka's Ruslan onward, the descending whole-tone 
scale has in Russian music been associated with evil magic, harshness, 
and cruelty. But even without whole-tone harmonies, like those in the 
first bar of Ex. 33 or the music of the avenging statue in Dargomizhsky's 
Stone Guest, it destroys the functional sense of harmony, which 
depends ultimately on gravitation toward points defined by the semi- 
tones in the scale and by the rise or fall of the perfect fourth or fifth in 
the bass. 

Throughout the second half of the nineteenth century functional 
harmony was being weakened, through its melodic surface, by modal 


1 See Vol. EX: 


NON-FUNCTIONAL HARMONY 89 


influences from folk-song, as in Russia, and plainsong as in France, to 
say nothing of pentatonic influences from the music of the Far East, 
such as that which impressed Claude Debussy (1862-1918) at the Paris 
Exposition Universelle in 1889. Fauré's flattening or evasion of the 
seventh degree of the scale, very characteristic of his style, has been 
mentioned in the previous chapter.! Chabrier has flattened sevenths; 
and Debussy in his early songs thought nothing of preceding a final 
tonic chord by a chord with no dominant function. In ‘Beau soir’ 
(1878), in E major, the penultimate chord is a G major triad; indeed the 
whole song is marked by juxtapositions of unrelated diatonic chords. 
‘Mandoline’ (1882) is still bolder. The suggestion of strumming on the 
open strings of the instrument, G D A, is unmistakably dominant, but 
the C major triad has hardly been heard when at bar 8 the composer 
breaks into a succession of unrelated triads and inversions: 


Ex. 34 


Allegr etto ---------- ----- 


(Exchange idle chat under Ше whispering boughs.) 


1p. 26 and Ex. 9. 


90 THE REACTION AGAINST ROMANTICISM: 1890-1914 


This is followed by another kind of unconventional succession, held 
together by the common notes, À G, but giving the ear the impression 
of consonances enriched by added notes. Near the end of the song the 
tonic chord alternates constantly with a 4/3 chord the dominant force of 
which is eliminated by the flattening of the B; in the last seven bars the 
two superimposed fifths of the opening become three—C С D A—of 
which the A is finally silent and the D moves to E. Every one of these 
procedures is in total opposition to the doctrine d’Indy was to enunciate 
later: that ‘chords as combinations of sounds appear only as the effect 
of a halt in the movement of the melodic parts е... musically chords do 
not exist . . .^;! Debussy’s chords exist purely as combinations of sounds 
for their sonorous effect, and are related to their neighbours often as 
much by contrast as by connection. The constitution, colour and 
lay-out of the chord are more important than its context; concord or 
theoretical discord, it is treated as a single unit to be manipulated as 
earlier composers had used consecutive thirds or sixths.? (It should be 
added that Debussy employed major seconds, possibly suggested by 
Borodin's, as early as ‘Ге Jet d'eau' (1889) and ‘Les Angélus’ (1891)—as 
indeed d’Indy did later, with no obvious poetic justification, in the 
prologue to Fervaal: Guilhen’s ‘Au nom du soleil, roi du monde’.) The 
fabric of classical harmony was thus being gently eroded on the diatonic 
side at the same time that it was disintegrating under the more violent 
strains of constantly intensified chromaticism. And it was the gentler 
process that was the more revolutionary, for while it originated as a 
symptom of romanticism it was by its very nature a denial of romantic- 
ism. 


THE EVOLUTION OF DEBUSSY’S STYLE 


It is natural to attribute Debussy’s exploitation of sonority for its own 
sake to the influence of current aesthetic ideas. No musician has been 
more sensitive to his cultural ambience;? and Rimbaud’s ‘Alchimie du 
verbe’ (1873)— J'inventai la couleur des voyelles!—A noir, E blanc, 
I rouge, O bleu, U vert.—Je réglai la forme et le mouvement de chaque 
consonne, et, avec des rythmes instinctifs, je me flattai d’inventer un 
verbe poétique accessible un jour ou l'autre à tous les sens. Je réservais 
la traduction'—had opened the door to the Symbolists and their 
poetry of subtly suggestive non-sense, a play of sounds and images and 


! See p. 21. 

2 On Debussy's non-functional harmony, see Ilse Storb, Untersuchungen zur Auflösung 
der funktionaler Harmonik in den Klavierwerken von Claude Debussy (Cologne Diss., 1967). 

? On the relationship of Debussy to the artistic and literary tendencies and theories of his 
day, see Andreas Liess, Claude Debussy: das Werk im Zeitbild (two vols., Strasbourg, 1936) 
and Edward Lockspeiser, Debussy: his Life and Mind (two vols., London, 1962 and 1965). 


THE EVOLUTION OF DEBUSSY'S STYLE ӨЛІ 


rhythms that should act on the reader as music acts on the listener. 
Similarly Debussy rejected those musical procedures in which the 
intelligence intervenes at the expense of simple perception: 


One combines, constructs, imagines themes intended to express ideas; one 
develops them, modifies them in conjunction with other themes which 
represent other ideas; one makes metaphysics but one doesn’t make music. 
Music should be registered by the listener’s ear without his having to discover 
abstract ideas in the maze of a complicated development. 


In 1909 Monet was to acknowledge an affinity between Debussy’s 
aesthetic and his own when he spoke to Roger Marx of his ‘harmonies 
and concords of colours which are sufficient in themselves and which 
succeed in touching us, as a musical phrase or chord touches us . 
without the aid of a more precise or clearly enunciated idea’.? 

АП the same, it is possible to exaggerate Debussy's impressionability 
to cultural fashions. The pentatonic flute theme which opens his 
orchestral Printemps (1887): 


Ex. 35 
'Trés modéré 


PP — poco =_= PP 


was written two years before he heard the gamelan at the International 
Exhibition, and it is very likely that he was introduced to pentatonic 
and other unusual scales by his theory teacher at the Paris Conserva- 
toire, Albert Lavignac, who also introduced him to Wagner.? 

Debussy's continued employment of Wagnerian and Franckian 
techniques, in music very different in substance and ethos from theirs, 
has been mentioned in the previous chapter.* Such dual nature was not 
uncommon in the music of that period; but whereas with most com- 
posers it consisted of real symbiosis, with Debussy such devices as 
theme-transformation were only the least disagreeable means of saving 
his music from deliquescence. Not only is his harmony in quite early 
compositions *weightless', without sense of tonal gravity, but even his 


! Léon Vallas, Les Idées de Claude Debussy, musicien francais (Paris, 1927), р. 30. 

2 Roger Marx, ‘Les Nymphéas de M. Claude Monet’, Gazette des Beaux-Arts, i (1909), 
p. 523; quoted in George H. Hamilton, Painting and Sculpture in Europe: 1880—1940 
(London, 1967), p. 19. 

3 An impressive catalogue of pentatonic motives and segments in Debussy's music has 
been compiled by Constantin Brailou, *Pentatony in Debussy's Music' in Studia Memoriae 
Belae Bartók Sacra (3rd ed., London, 1959), p. 377. But the first and earliest example, from 
the song ‘Fleur des blés' (1877), is only part of a phrase which is not completely pentatonic, 

4 See р. 29. 


92 THE REACTION AGAINST ROMANTICISM: 1890-1914 


melody floats, as Ex. 35 does, and tends to return to the same note, 
beautiful but lymphatic and listless, outlining shapes in which—as in 
the filigree detail of background texture—it is impossible to deny а 
close affinity with what was, a little later, to become known as art 
nouveau.* When he wished to construct an extended piece of absolute 
music in the String Quartet (1893) he was obliged not only to employ 
the romantic technique of theme-transformation but to lean heavily on 
an actual model: Grieg's Quartet in the same key.? He goes through 
many of the motions appropriate to a large-scale work of this kind: 
sequence, pseudo-imitation (раписшапу in the finale, the most clumsily 
made movement of the four), the return of themes in different keys. 
But the motions are meaningless and distract one from that exquisite 
mosaic of motives constantly shown in fresh harmonic lights and from 
fresh tonal angles, which constitutes the real essence of the music. 
Key-organization is of even less importance than in the work Mahler 
was composing at the same time, his Second Symphony. 

In the compositions of the period between the String Quartet and the 
completion of the score of Pelléas et Mélisande in 1902—the Prélude à 
l'aprés-midi d'un faune (1892-4) and Nocturnes (1897-1901) for 
orchestra, the Proses lyriques (1893) and Chansons de Bilitis (1897) for 
voice and piano, ‘Jardins sous la pluie’ (1894) (the other two Estampes 
seemingly came eight or nine years later) and Pour le piano (1894-1902) 
—he completed the establishment of an extraordinarily personal style 
and harmonic vocabulary which he was more or less content to work 
and re-work and refine during the following decade, until it was nearly 
exhausted. In L’aprés-midi there are still traces of Balakirev or Rimsky- 
Korsakov (the flute triplet-figure three bars before fig. 3) and Massenet 
(the section in D flat), but after that the various influences that con- 
tributed to the formation of his idiom were fully absorbed. Some of 
these have been greatly exaggerated. The unorthodoxy of Mussorgsky's 
empirical harmony, for instance, is quite different in nature from 
Debussy's, even if it was known to him earlier than c. 1896; and it 
would be difficult indeed to isolate any element in Debussy's music 
which he could have picked up only from Erik Satie (1866-1925)? and 
not from the common stock. In 1897 he orchestrated two of Satie's 
three Gymnopédies* (1888) for piano but they made no impression on 
his own work except perhaps in Ше unimportant ‘Danse profane’ for 


1 On Debussy and art nouveau, see Lockspeiser, op. cit., i, pp. 116 ff. and infra, pp. 166 ff. 

? See Gerald Abraham, Grieg : a Symposium (London, 1948), p. 8. 

? On the early personal relationship between Debussy and Satie, see Lockspeiser, op. cit., 
i, pp. 145-9; on his early knowledge of Russian music, ibid, pp. 47-52, and André Schaeffner, 
*Debussy et ses rapports avec la musique russe', in Musique russe (ed. Pierre Souvtchinsky) 
(Paris, 1953), i. 

а The Greek Yopyozatóto was a festival marked by the dancing of naked youths. 


THE EVOLUTION OF DEBUSSY'S STYLE 93 


harp and strings (1904); it was Maurice Ravel (1875-1937) who came 
near to plagiarizing them in ‘Les entretiens de la Belle et de la Béte’ in 
Ma mére l'Oye (1908). As for Satie's three Sarabandes (1887), it is 
impossible to see any connection between them and the Sarabande of 
Pour le piano beyond one insignificant motive and the fact that both 
revive an old dance-form. (The revival of old dance-forms was in any 
case fashionable in France at the time, witness Chabrier's Bourrée 
fantasque (1891), Chausson's Quelques danses (including sarabande, 
pavane and forlane) (1896), and Ravel's Menuet antique (1895) and 
Pavane pour une Infante défunte (1899).) Chabrier had introduced series 
of unresolved ninth chords in several passages of Le roi malgré lui 
(1887). (To fill out an impression of French harmony at this period it 
should be remembered that such relatively conservative composers as 
Saint-Saéns and Bruneau wrote passages melodically and harmonically 
in the whole-tone mode, in the Scherzo, Op. 87, for two pianos (1890), 
and Le Réve (1891) respectively.) Satie's Gnossiennes (1890), his three 
preludes for Sar Péladan's Le Fils des étoiles (1891), his Sonneries de la 
Rose+Croix (1892) and Prélude de la Porte Héroique du Ciel (1894), 
all for piano, did open up new ground—unbarred and rhythmless 
successions of unrelated chords—but Debussy did not follow him on to 
it. Admittedly the chords themselves are sometimes interestingly 
experimental, as is the case with the piled up fourths at the beginning 
of the first Fils des étoiles prelude (1); but although the passage of 
common chords that opens the Sonneries (ii) may have sprung from 
mystical emotion, it could hardly communicate it: 


Ex. 36 


En blanc et immobile 


94 THE REACTION AGAINST ROMANTICISM: 1890-1914 


Highly romantic conceptions are reducéd to nullity by extremely anti- 
romantic musical devices. Romantic symptoms—pictorial and literary 
affinities—are neutralized by non-expressive music in Debussy also; but 
Debussy’s music, while it does not seek to express directly the emotion 
aroused by contemplation of an object, is nevertheless a vivid reaction 
to the object in terms of purely sensuous sound, which evokes an 
imprecise but delightful response akin to that produced by Rimbaud’s 
verbal alchemy or Monet’s visual harmonies. 

The apparent paucity of Debussy’s output during this period is easily 
accounted for by his preoccupation with Pelléas, which is discussed in 
the next chapter, and with other projects which never came to fruition. 
Few as they are, these compositions reveal significant tendencies. 
Following the String Quartet, Debussy experimented further with 
‘absolute’ music in the Sarabande and two later pieces of Pour le piano: 
the side-slipping of triads, sometimes augmented (whole-tone) triads, 
and complete whole-tone passages in the Prelude, side-slipping of 
chords of the seventh and added-note chords in the Sarabande, penta- 
tonic patterns in the Toccata (to which ‘Jardins sous la pluie’ is a faintly 
pictorial counterpart). Despite the non-functional nature of the 
harmony, the pieces are given very clear definition by the nature of the 
piano writing, by diatonic sequences, and by the recurrence of patterns 
and sections. In the Proses lyriques and Chansons de Bilitis, settings of 
prose-poems by Debussy himself and Pierre Louys respectively, 
Debussy moved even further from lyrical song than in the Verlaine sets 
of 1891 and 1892 (the Trois mélodies and the first set of Fétes galantes); 
the piano is often the more important partner and the voice only 
comments, though the Bilitis songs approach the exquisite subtlety of 
the near-parlando style of Pelléas. The sense of the texts and the use of 
all-pervading patterns in the piano-part, in the traditional manner, 
brush aside that problem of form which Debussy still handled clumsily 
and somewhat perfunctorily in his instrumental music. It was in the 
orchestral Nocturnes that Debussy first perfected his instrumental 
technique, not in the tarantella-like ‘Fêtes’ with its echoes of Balakirev’s 
Tamara nor even in the drifting, changing cloud-shapes of ‘Nuages’, 
but in ‘Sirénes’ which is a pure sound-mosaic of fragmented, subtly 
coloured chords and tiny motives, moniliform and athematic except 
that some minute pattern may appear briefly as an ostinato and re- 
appear lightly transformed and in ever-changing harmonic and 
orchestral lights: 


THE EVOLUTION OF DEBUSSY'S STYLE 95 


- СО BM И 3 5 М 
ОШ ПОДОВЕ жеке кет Е ен ССД Са 2-2 ағы (а-г ІІ ЕСІГІ ер қ 

А EL Е Иа if Вот? жартас алты „Биша Гр ees... ERE 

ee сек ы Дл fpem је pp eee 

pee he] SO [ш 1 ШЕШ 9 ——— —H 


BEIM — | Г 
КЕ Идора ар | 
CIERRE m mem SS 
ЕЕ TET AM m] 


‘Sirènes’ begins with a ‘dominant’ chord in B major and ends with a В 
major triad, and here and there the ear catches other familiar harmonies; 
but they bring no sense of tonal unity or contrast, of modulation, or of 
tension and relaxation. The distinction of consonance and dissonance 
had been obliterated long before this in Debussy; each chord has its 
isolated value as sound-effect only. 

These techniques were exploited exhaustively in some of the works 
of the next decade, La Mer (1903—5) and the Images for orchestra 
(1906—12) and the two sets of Images (1905 and 1907) and two books of 
Préludes (1910 and 1910-13) for piano. The sustaining of low bass- 
notes on the piano necessitates suspension of the damper action and 
further enriches and blurs the sonorous effect. Equally idiosyncratic is 
the treatment of the orchestra, in which doubling between groups is 
extremely rare: a figure is usually limited to woodwind or horns or 
strings only, and set against an equally complex pattern in another 
instrumental colour. When doubling between groups does occur, it is 
employed to produce effects of great subtlety, as in the section ‘Les 
Parfums de la nuit’ in ‘Ibéria’ (1908), the second orchestral Image 
(p. 69), where a ‘distant’ melody (lointain et expressif) is played by a 
solo violin, muted, and a solo bassoon in its ‘pinched’ highest register; 
or a little later in the same score (p. 72) where a melody characteristically 
thickened out in 6/4 chords is played by two flutes, two piccolos, three 
trumpets (two of them muted), two solo violins, and two solo cellos, the 
main strand being delicately outlined: 


96 THE REACTION AGAINST ROMANTICISM: 1890-1914 
Ex. 38 — dE 


s TN Е 
(d= 80) re. REN 


= ЫШТЫ Sp LS PIANI оо | 
Сред ч ке ко тул aram ux ee [ЕН 

Fl. 027 ала | TERET ASO. MEE NES Ба [ exam [67 ee | 
LZ езе] e лн | кшш |, a | 


Œil, 
(+Bn. 


Зое lower) 


oye 
7—8 gite да 
ТАП 


шаш ЕЕ 


Horns 
ЕР СЕ. ЖӨ T > 
Oe арии Е 
Babe Je 2 
НН пара 
тү 2 
"Tpts НИИ тави es 
- [= == 
=. 


Sere ey = 


шай) te th ME men, A чат а 
“Ой аз 


P oix et soutenu dans lexpression 


() 
ети : Өсе ж-е 
eae іш” шеші ра ана аашаа) че =й Шы дет 

2 Solo Vins. Bie SS еще | = EE 
= eree eee БЕН Шетен | И е | 


P doux et soutenu dans l'expression ———— 
І 
sur la touche vins, sur la touche 
йй: 


Vins Шау Ee) Ен са еу се СӨ с = =ч гү: 
& Vlas. фу, | mfi p Ее 


Уаз. = — 
Jp trés légerpp trés léger 


() a 
ED 48°) it a ЕЕ ae ae 1) 
2 Solo Cellos Cis ETE m 


p 
doux et soutenu dans l'expression 


D.B. = EARN р == 
7225, LZ 7 Hi gpz ----- 


doux mais soutenu F 


sur la touche 


THE EVOLUTION OF DEBUSSY'S STYLE 97 


„Н-0101, 


и apiy aa сев 


FI. 
GI. = 
(+ Bn. еее ее м 
дое lower) |t— Ага ДФ s MeL. —H 
f) 
РУ ан 
а 1 
1 ey ==. 
EAS 
Horns 
га € с. П 
[-9 eerie 07 —) 
| oum НА 
ГИ 
Tps = = | Pu 
(П & Ш LASER ats HS а 5 =a n == Hag же 
muted) ЕЕЗ ELE GE So ae oe E JX E E 
p ENT са " 77-4 >. 
SSS = ae 


2 Solo Vins, ЕЕЕ ЕЕ ЕЕ 


Vins. II div. |99950 о 
Ку. div. | — np 
. өл ЕТ nc Lo ee [95 Е ВАА 


() 
[—4 4 mmu ВИ eem — — — ee 


УТИ HII emm — em T Remp em ADIRE RT] 
2 Solo Cellos уа Е 


An equally subtle blend of sonorities, this time accompanimental as 
background to a horn solo—the same melody which is soon to be heard 
in quicker tempo, lointain et expressif —occurs a little earlier: 


98 THE REACTION AGAINST dd. 1890-1914 


ба a ()=92) LEES wx 


[ Зин Lu Т.а [43 - 22 


к. DS RR RE = Ia 
— еее 


Ib ul Ра EI © 
Solo Horn -Е-ҢЕ-4--- Eg ————————3 


() Т т 
Д-р ер ұғ Oo бағ Е 
Сен ГОА ------- а 
Е еее) - 


еи ЗЕЕ ЕЕЕ ЕЕЕ КЕН 
е Eg ERE == те 
рр doux et léger 
"Ims „ша тъ Lo ras 
Енн [pg e aa] 
Vu Б ИН .— == —— ЕЕ 


Ус. 


= НИ б=——============ : 
d iT ездері = т ee 


p Gy 441—1 4 = аа ий 
н. luere а а 


EA вао == = > Я = га аи ая = 
dier ru IN EE) 
НО Вы rd 
Solo Horn ЕН | E d > 
(53 
< - 
за ж ст. T = y 2 
а-а-а E cA DEDE с 
Pr 4 — rr тс. 
0 Еде = Е af Bocas eis Е Se esses 
РАНЕН EE ве aaa = 
c к= тсе то 


THE EVOLUTION OF DEBUSSY'S STYLE 99 


In all these compositions, orchestral or pianistic, which may justly be 
described as impressionistic or (better) pointilliste, the basic methods are 
the same: whether modal or pentatonic or whole-tone, the melodic 
fragments are often thickened out into parallel seconds, simple or 
augmented triads or higher-powered chords with or without added 
notes, to form variegated patterns. The fabric is held together not by 
thematic logic or rhetoric, though sometimes by relationships between 
the various tesserae, but by pedals or ostinato figures. Rhythmic 
patterns are also used for the same purpose, though rhythmic impulse— 
sometimes vigorous enough, as in the first and third section of ‘Ibéria’ 
or ‘Golliwogg’s cake walk’ from Children’s Corner for piano (1906-8), 
or even quasi-mechanical as in ‘Mouvement’ from the first set of piano 
Images—is frequently feeble to the point of near-extinction. (On the 
other hand, Debussy’s micro-rhythms can be exquisitely subtle.) 
Sometimes whole sound-complexes аге the binding agents as in ‘Voiles’ 
from the first book of Préludes: 


Ex. 40 
Modéré (2) = 88) 


(ИБИ 2 ———— — — — .. ii bars 22-3 
4 > “Ашы: 1 ИР г 


which becomes 


(iii) bars 27—8 os M 7 
LE SS ACE „m енені 


or the right-hand figure of the section Un peu animé et plus clair of 
‘Cloches à travers les feuilles’ from the second set of Images.! 

Yet side by side with his purely impressionistic compositions Debussy 
never ceased to write others, almost equally non-tonal but hovering on 


1 ‘Cloches’, a compendium of Debussyan technique, is recorded in The History of Music 
in Sound, x. 


100 THE REACTION AGAINST ROMANTICISM: 1890-1914 


the edge of the diatonic and employing simpler harmonies and textures 
and more sober colouring: the settings of Charles d'Orléans for voice 
and piano (nos. 1 and 3 of the Trois chansons de France, 1904) and 
unaccompanied chorus (Trois chansons de Charles d'Orléans, 1908) and 
Villon (Trois ballades, 1910), the music for d'Annunzio's Le Martyre 
de saint Sébastien (1911), the Trois Родтез de Stéphane Mallarmé for 
voice and piano (1913), and the Douze Études for piano (1915). АП these 
in their different ways—renunciation of ‘jewellers’ work’, absence of 
extra-musical suggestion, simple modality of melody and harmony, 
economy of texture—suggest, particularly when all these qualities show 
themselves at the same time, an ever-increasing preoccupation with the 
values we describe as 'classical' It was a tendency that reached its 
apogee in Debussy's last works, the three Sonates of 1915—17, for cello 
and piano, for flute, viola and harp, and for piano and violin. It is 
difficult to imagine music more severely classical than the opening of the 
last Sonata: 


Ex. 41 
Allegro vivo (4. = 55) eae 


Е 
Violin 
шаз 


Piano 


po 

D стаи 
еды apum enl 

= 

d COT ET 


© 


1 The second, ‘Ballade que;feit Villon à la requeste de sa теге pour prier Nostre Dame’ is 
recorded in The History of Music in Sound, x. 


DEBUSSY AND RAVEL 101 
DEBUSSY AND RAVEL 


Although Debussy's development after Pelléas was little affected by the 
music of other composers—in 1913 he told Calvocoressi he was rather 
out of touch with contemporary music ‘because he wished to concen- 
trate and had made it a rule to hear as little music as possible"!—there 
was one young contemporary who undoubtedly came under his 
influence and to whom, it seems possible, he was to a slight extent 
indebted. In March 1898 he heard a performance of Ravel's Les Sites 
auriculaires for two pianos (1895-6) and was so impressed by the first of 
them, the ‘Habanera’ afterwards orchestrated as the third movement of 
the Rapsodie espagnole (1907), that he asked the younger composer to 
lend him the score. This led ultimately to charges of plagiarism in his 
own ‘Soirée dans Grenade’ (1903), second of the Estampes for piano, 
where the resemblance is certainly very striking, and his Lindaraja for 
two pianos (1901), where it is limited to the dance-rhythm and a single 
bar of melody common to other habaneras; indeed Ravel’s was itself 
heavily indebted to Chabrier’s piano-piece of 1885, or, rather, the later 
version in D flat. Similar claims have been made for the influence of 
Ravel’s Jeux d’eau (1901) on Debussy’s piano-style; they are counter- 
balanced by the fact that Jeux d'eau is itself marked by Debussyan 
procedures.? And Ravel admired the Sarabande from Pour le piano 
enough to orchestrate it in 1903, conversely arranged Г” Aprés-midi for 
two pianos, and admitted in 1928 that in his Shéhérazade songs of 
1903, Tinfluence, au moins spirituelle, de Debussy est assez visible’.® 
The fact is that under the influences of Saint-Saéns, Fauré, and 
Chabrier French piano-composers had for some time been exploring 
the resources of the instrument; even minor figures such as Déodat de 
Séverac (1873-1921) in his suites Le Chant de la terre (1901) and En 
Languedoc (1904) made highly individual harmonic experiments before 
succumbing to Debussy's influence in the Baigneuses au soleil (1908), so 
it is hardly surprising that the two leading figures found each other 
mutually stimulating. But even in this field of piano music, where in 
his five Miroirs (1905) and the three pieces of Gaspard de la nuit (1908) 
Ravel comes nearest to Debussy—and the Sonatine (1905) may perhaps 
be regarded as a counterpart of Pour le piano—the differences are very 
obvious. Harmonic vocabulary and pianistic figuration are similar yet 


1 M. D. Calvocoressi, Musicians Gallery (London, 1933), p. 122. 

2 On Debussy's piano-style, see Frank Dawes, Debussy: Piano Music (London, 1969) and 
Robert Schmitz, The Рано Works of Claude Debussy (New York, 1950); on Ravel's, Kurt 
Akeret, Studien zum Klavierwerk von Maurice Ravel (Zürich, 1941), and Henri Gil-Marchex, 
*La technique de piano', Revue musicale (special Ravel number), vi (1925), no. 6, p. 38. 

3 *Esquisse autobiographique', Revue musicale, xix (1938), p. 212. On the musical and 
personal relationships of Debussy and Ravel, see Lockspeiser, op. cit., ii, pp. 33-44, and 
Martin Cooper, French Music (London, 1951), pp. 134-40. 


102 THE REACTION AGAINST ROMANTICISM: 1890-1914 


different; Ravel's piano-writing is virtuosic in the sense of Liszt and 
Balakirev (who meant quite as much to him as to Debussy), it is 
rhythmically incisive as Debussy's seldom is, its shapes are generally 
clear rather than blurred and they succeed each other with that logical 
connection which Debussy minimized or spurned. For all the modernity 
of his musical language—the pentatonic and modal elements, the 
complication of individual chords, the parallel seconds and sevenths, 
and so on—tonal feeling is nearer the surface of Ravel's music than 
of Debussy's. Even the added seconds which blunt the edge of Debussy's 
sound seem to sharpen the edge of Ravel’s, as in these passages from 
*Scarbo', the third piece of Gaspard de la nuit: 


Ex.42. 


@ ааа вери. 
oe 
2 amm -—] 


ЗА LH EIU жетіні ің 
Д СИСАРЕ demere жасан) 
к=п ЕАСИ ЕЕ. с=с gu атанғалы) 


IER AC ERE FA Dg 
ЕСЕ ЕМЕА Ееее и Lap ie E 


Again unlike Debussy, Ravel cared more for instrumental line than for 
instrumental colour. His orchestration—and he delighted in orchestrat- 
ing even his own piano-music—is totally different. The lines are not 
often variegated in themselves as in the romantic line derived from 
Wagner's orchestra, which may pass from horns and lower strings to 
upper woodwind (Ex. 27) or begin on flutes, be doubled by clarinet 


DEBUSSY AND RAVEL 103 


for half a bar and then (while flutes pursue a different line) by violas, 
ending on violas alone—all within four bars (Ex. 21 (v) ). Ravel makes 
no effort to avoid doubling and when he contrasts groups of instru- 
mental colour he does so not to obtain subtle, iridescent effects but 
for the sake of brilliant transparency in the Russian tradition. Thus, 
when he scored the following passage from the ‘Alborada del gracioso’, 
the fourth of the Miroirs: 


Ex. 43 


4 А Oe, m > 
34 ағы ыз шс и г 97599 | а 
ери —4— 119 —— 19 1 5.0 
TEE WERE | 


ж uel 


he gave the melody to piccolo, oboes, cor anglais, bassoons, half the 
first violins, half the seconds, half the violas, and half the cellos (in their 
highest register) in octaves, and the ‘inside pedal’ A—always a favourite 
device of Ravel’s from the ‘Habanera? onward—to the remaining 
strings (in octaves), double basses, double bassoon, and tuba; the rich 
harmony is left to flutes, clarinets, harps, and brass, and the second 
chord of each bar (which is just off the second beat) is piquantly 
emphasized, again in the manner of Rimsky-Korsakov, by triangle, 
tambourine, side-drum, and cymbals. Similarly the harmony of the 
opening of the Valses nobles et sentimentales for piano (1911) sounds 
much clearer in the orchestral version. When Ravel wishes to create a 
wash of imprecise sound as in the ‘Lever du jour’ of the ballet Daphnis 
et Chloé (1909-12), it is paradoxically done with extreme precision and 
serves as a background against which firm, clear lines are drawn in 
contrasted instrumental colours. 


104 THE REACTION AGAINST ROMANTICISM: 1890-1914 


When one turns from the two great sets-of piano pieces and the song- 
cycle Histoires naturelles (1906), the most Debussyish of all Ravel's 
mature works, the completely different basis of his aesthetic becomes 
even more apparent. *Mon Quatuor en Ға” (1902-1903), he told Roland- 
Manuel, ‘répond à une volonté de construction musicale ... qui 
apparait beaucoup plus nette que dans mes précédentes compositions'.1 
It was to appear again in the Sonatine for piano and the Introduction 
et Allegro for harp with flute, clarinet, and string quartet (1906). The 
wish to make а ‘musical structure’, tout simple, seldom visited Debussy 
even in the non-impressionistic works of his maturity; it possessed him 
only at the very end of his life in the three Sonates. But, looking back at 
the frankly impressionistic Jeux d'eau after seventeen years, Ravel 
thought it worthwhile to point out not only that it was ‘the starting- 
point of all the pianistic novelties’ of his music but that it is ‘based оп 
two themes in the manner of the first movement of a sonata, though not 
however subjected to the classic key-scheme'. The classical ground-plan 
of sonata-form, rudimentary in Jeux d'eau, is more obvious in the 
String Quartet and Sonatine but they are equally emancipated from the 
classical key-scheme. It is true the first subject of the first movement of 
the Sonatine would be in F sharp minor if its Es were not all natural, 
and the key suggested (but always evaded) by the second-subject 
material is the orthodox relative major; moreover the second subject is 
recapitulated in an identical evasion of F sharp—which actually 
identifies itself only in the final bar of the movement. But these ghosts 
of keys no more exercise pulls of tonal gravity than the themes in which 
they are embodied act and react on each other. The String Quartet had 
evidenced more thematic logic and is held together as a whole by 
familiar cyclic methods; both slow movement and finale refer back to 
the first. But the sonata-form of the first movement is even less orthodox 
than that of the Sonatine; the second subject is not in a contrasting key 
and is recapitulated in the same key; the tonal changes in the recapitu- 
lation, which is otherwise nearly literal, occur within the subjects and 
in the transition-passages—which thus become merely reliefs from 
tonal monotony. But survival of the mere shell of sonata-form, void of 
everything that had given it life and purpose, is a common phenomenon 
of the period.? Without the gravitational pulls of tonality, neither the 
loosely flowing diatonic or modal lines of a Ravel nor the tightly 
woven chromatic fabrics of the Central Europeans could use sonata- 
form in any significant way. 

Although Ravel dispensed with the dynamism of key, his formal 


1 *Esquisse autobiographique’, p. 212. 
? Cf. the remarks on Alban Berg's Piano Sonata on p. 73. 


THE WIDER INFLUENCE OF DEBUSSY 105 


structure is perfectly lucid in ways that place him apart from Debussy 
and at the opposite pole from the Central Europeans, not so much 
because his melodic lines and harmony are free from hyper-chromatic- 
ism, his textures from over-density and his rhythms from flaccidity, but 
mainly because of its clear articulation by cadences. His cadences are 
not corner-stones and key-stones of tonality; nothing gravitates toward 
them; but their frank definition of periods is in striking contrast with 
the practices of impressionism, where they were of no importance, and 
of late romanticism, where every effort was made to nullify or conceal 
them. A *musical structure', as Ravel conceived it, neither an expression 
of emotion nor a play of exquisite sonorities, was a work of high 
artifice and he saw no reason to conceal the fact. Thus in every way he 
found himself congenially assuming eighteenth-century attitudes toward 
music, just as when, in realizing ‘the Greece of his dreams’ in Daphnis, 
he found that his ‘vast musical fresco’ readily harmonized with ‘that 
which had been imagined and depicted by the French artists of the end 
of the eighteenth century’. Ravel was only in a few of his works and 
in a limited sense an ‘impressionist’, but he adapted some of Debussy’s 
techniques, particularly his harmonic language, to a quite different end, 
a renaissance of classicism, just as others borrowed them to enrich the 
palette of exhausted romanticism: for example, the slow strange chords 
evoking the atmosphere of Herod’s palace at the beginning of Florent 
Schmitt’s mimodrame, La Tragédie de Salomé (1907; symphonic version, 
quoted here, 1911), (Ex. 44), or Dukas’s Debussyan orchestration in La 
Péri (1912), (Ex. 45). 

In considering Debussy in the context of history, it is always necessary 
to distinguish between ‘impressionism’ in the narrowest sense, which 
was—at any rate for the time being—a dead end, and his conscious 
‘emancipation of dissonance’? twenty years or more before Schoenberg 
used the phrase in his Harmonielehre. Admittedly, the dissonances 
Debussy set free were a great deal less fierce than Schoenberg’s. 


THE WIDER INFLUENCE OF DEBUSSY 

Considerable as Debussy’s reputation had been before the performance 
of Pelléas (30 April 1902, the first major landmark in the history of 
twentieth-century music), it was much greater afterwards. During the 
next decade Debussy and Strauss were almost universally regarded as 
the two outstanding figures, the leaders of strongly opposed tendencies, 
in contemporary music. Debussy and Strauss, be it noted, not Ravel 
and Mahler. In France Debussy’s influence was naturally very great; 


1 ‘Esquisse’, p. 213. | 
? бее Lockspeiser, op. cit., i, pp. 204-8. 


THE REACTION AGAINST ROMANTICISM: 1890-1914 


106 


Ex. 44 


Vins. 


Ex. 45. 


a 
= 

ж Е да 
бъ ГІ 

= кор 

А 
ET 

3L 

ПЕ! 

<< 

оз 

28 

5% 

Si 

ща 

=) 


р mare. 


а. 


Он 
=: 


------ 


CODE 


с 


ЕЕ 


Е: 


р mare. 


Tpts. 


== 


има 


muted умыл, 


BUT 3 —1 


Vin. I 


o ъ= == PüÓ— Á« 


сте ai 
21.41 


г 


Y 


2 


r 
y 


I pizz Timp. іп Е col I 


muted pizz.p 


Vc. + D.B. 
(both div.) 


107 


r$ 


bis 


"iet 


— ІҢ 


THE WIDER INFLUENCE OF DEBUSSY 


2 ЕІ.,2 Ob., С.А. 
(3 Bn. 8ve lower) 
СІ. 


| 


pur 
р 


M Mit. | 
ш X. x 
Вай КЕЙ ий 
с 
aS о & 
Р са H a3 a 2 Р Гы 
2 E 3 Oe © ы d 
е д ы ЕГІ |e И. 5 щ > 
в га 
Se ра 
сә 
eS 


T 


E 


ия 
НВ 


[Е 
[LN 


Vin. II 
Va. 
Vc. + D.B. 
(both div.) 


108 THE REACTION AGAINST ROMANTICISM: 1890-1914 


composers of widely different character submitted to his influence in 
one way or another, and only those of an older generation—Saint- 
Saéns, Fauré, d'Indy—were immune to it; and in d'Indy's case the 
immunity was not total.! As Léon Vallas put it, ‘Personne vers 1902- 
1905 ne pouvait se défendre complétement de la contagion musicale de 
Pelléas et Mélisande’.? That d'Indy should have been even slightly 
infected is the more remarkable in that his variety of classicism, which 
he imparted to Dukas, was more rigorous than that of Saint-Saéns; 
Ravel could in later years recognize an affinity with Saint-Saéns but the 
elegance of their classicism was quite foreign to d'Indy; the polyphony 
of d'Indy's Second Symphony (1903) is often as harsh and angular as 
Reger's: 


Ex. 46 


ПЕРЕН АЕ i 
Bass Cl. oP hy с=с = а з= ае : 
mf 

0 А А А А А А 
рг: : 
Si py [en] ---- _—_—_ : 
Horns у= == Woo pp tt е ПИ 
C D : 
= oz ' 
Trumpets | Hes —2 я : 
DNA. і 


Tbn. Па | se Е-Е 


С.В. Tbn. | 


() . 
ЖЕР” ЖЕНГО > ШЫ rein ecc sene D. „чї ee Spese EUR. Түз ыз МР TT 
| улса асы кых = ee 
Vins. о т — ШИ 
Е 9p . БИТ a FÉ 


D 
' cresc. . 
0 unis. 


“Же Сы ш зеш 
Ирак а =. EE ou 
Миз. div. | Е EL X I 2 НЕЕ 


5 ise р. 240), Til, а 
2 Vincent d' Indy, ii (Paris, 1950), р. 255. 


THE WIDER INFLUENCE OF DEBUSSY 109 


Bass Cl. 


Horns 

Trumpets Hf 
e я: Li 
Ша LS Se у Е-е = 
С.В. Tbn. Е И тер т та 
V стезс 
f) II 2 
[mU LiPEmeEUnSILSE.c cmm ат. тарлы. ла. ae SS ie E Е ен 
Vins. = = iem: сзсз e um ЕЕ xl. BÉ m —— та 


E --- E E O. — шы 


ИЕ 
Ни pF 


Vlas. div. 


Sempre cresc. 


and the fugal middle section of the third movement of Dukas’s Piano 
Sonata (1901): 


Ex. 47 
(d = 52) 
Фр mystérieusement 


doucement marqué 


110 THE REACTION AGAINST ROMANTICISM: 1890-1914 


comes from the same school. If composers such as these could not 
resist the infection (cf. also Exs. 44 and 45, the trio of the scherzo of 
d'Indy's Violin Sonata (1904) and his Jour d'été à la montagne (1905)), 
it is not surprising that such natural eclectics as Albert Roussel and 
Charles Koechlin succumbed heavily before finding their own artistic 
personalities. 

Roussel’s acknowledgement of ‘the weak influence of Debussy’ has 
already been quoted.! It is naturally most obvious in such early works 
as the first set of Quatre poémes (by Henri de Régnier), Op. 3 (1903), the 
Rustiques for piano, Ор. 5, and First Symphony (Le Poéme de la 
forét), Op. 7 (both 1904—6), and persists until the orchestral triptych of 
Indian impressions Evocations, Op. 15 (1910-11), but even in the 
earliest of these works a distinctive personality begins to show itself; 
the language of Debussy is used in a different sense. And actually at the 
same time Roussel was writing things like the Divertissement, Op. 6, for 
flute, oboe, clarinet, bassoon, horn and piano (1906) (1), and certain 
parts of the D minor Violin Sonata, Op. 11 (1908) (11), which are as 
neo-classical as Ravel without being particularly Ravelian: 


v See р. 25. 


111 


THE WIDER INFLUENCE OF DEBUSSY 


Ex. 48 


г . 
ое epe Шоо г = үүт асс с ааа 


Ше i та нож сим а CIR e сыы ы ыны ae ы л ee (ee pp x] = — um 1 
Я Я z с 


Ент! “ұлың БН жо” а 


BNA Sk anne en Щ eb] БЕ, = ланы... ЕС 


al 


112 THE REACTION AGAINST ROMANTICISM: 1890-1914 


Gi) Plus vite 


Piano 3 № 
only J^ énergique 


P LL V 
Е 7 аъ Ето Е, 
[XE ш ЕЕ = УИНН 


Nor, despite their structural leaning on d'Indy, are the Divertissement 
and Sonata d'Indyesque. Like Debussy and Ravel, Roussel delighted 
not only in the sound of added-note seconds but in the pentatonic 
mode (from the ‘Ode à un jeune gentilhomme’ (1907), second of the 
Deux poémes chinois, onward); exotic scales and the harmonies based 
on them then played a more and more important part in Roussel's 
music.? And, like Florent Schmitt, Koechlin, and others, he employs 
‘scholastic’ chords of the ninth, not obtained by added notes as 
Debussy's usually are, with a freedom of non-resolution and chromatic 
alteration like that with which rather earlier composers had treated 
chords of the seventh. The dissonances set free by Debussy were used 
in music very unlike his, for example the bourrée of Roussel’s Suite for 
piano, Op. 14 (1910): 


1 See Arthur Hoérée, ‘Га Technique', Revue musicale (special Roussel number) x (1929), 
по. 6, p. 84; this article, of which Roussel himself approved, is reprinted in condensed form 
in Hoérée, Albert Roussel (Paris, 1938), p. 87. 


THE WIDER INFLUENCE OF DEBUSSY 113 


к= лет 


4 Ба 
р tres souple 


Unlike Roussel’s eclecticism, Koechlin's was never focused by a 
strong creative personality. During his long life (1867-1951) he com- 
posed in every style of the day—at least in every French style and in 
some not particularly French—very often with individual touches. One 
hears this at once in his Op. 1, the first set of Rondels for voice and 
piano (1890-94), where the delicious setting of Théodore de Banville's 
*Le Thé' could not quite have been written by Fauré. The opening of the 
prelude to Part I of his L'Abbaye (suite religieuse pour choeurs, orchestre 
et orgue), Op. 16:1 


Ex. 50 
Andante 


га 


vi aa СА oA 

[cuam ar ey _1 

ED ЕЕ ШҮ со x el 

6 5 = Фа eec > E [04 Е ааа ы 
ШЕ. А 4 Ze 2 ВЕ И n —Á— — 

р = А ре 

е 

22 

= 


1 Quoted from the composer’s piano reduction, the only published form. 


114 THE REACTION AGAINST ROMANTICISM: 1890-1914 


Molto tranquillo 


is remarkable in that it was composed during 1899-1901, long before 
‘Га Cathédrale engloutie'. Debussyish harmony, Ravelian line-drawing, 
Satie's deliberate naiveté and eccentricity, experiments of every kind, 
often combining to produce music of undeniable beauty, converged 
most satisfactorily in such trifles as the two sets of Esquisses for piano, 
Op. 41, the more substantial Paysages et Marines, Op. 63 (1905-15), 
and the Quatres Sonatines frangaises (Op. 60) for piano duet (1919). 
In so far as Koechlin was influential, which was really only in the post- 
war period, he certainly contributed to the current of French neo- 
classicism. 

It was not only French composers and foreign musicians living in 
Paris who were infected by Debussy and—during this period to a much 
lesser degree—by Ravel. Composers in other lands began to come 
under the spell of the new techniques. But Debussy's influence followed 
his reputation at a distance of five or six years; if Pelléas established 
the one in 1902, it was only about 1907-8 that young foreign composers 
began to succumb to the other. Paris became the magnet that Leipzig 
once had been. In 1907 Falla went to Paris and worked with Dukas; in 
1908 Vaughan Williams went to Paris to study with Ravel. And before 
we dismiss Debussy's influence on Vaughan Williams as minimal, we 
should note Bartók's account of his first acquaintance with Debussy's 
music: 


THE WIDER INFLUENCE OF DEBUSSY 115 


When, urged by Kodály in 1907, Г began to know and study Debussy's 
compositions, I was astonished to find that certain pentatonic motives 
corresponding to ones in our folk-song similarly play a great role in his 
melody.! 


It was not the harmony that struck him first, though it did soon, not the 
piano-writing or the pointilliste technique which never made much 
impression on him. (Bartók's most Debussyish composition is the first 
of the Két kép (Two pictures) for orchestra, Op. 10, no. 1 (1910): cf. 
(Ex. 52 (iv).) As Edwin von der Nüll put it:? 


He escaped from the atmosphere of neo-romanticism and subscribed to 
Debussyism. He did not, it is true, fall under the spell of Debussy's highly 
strung femininity . . . He took from the Frenchman what he could incorporate 
in his own self-willed, angular nature. He gladly accepted the enrichment of 
harmonic means but did not give up his sharply profiled melodic line; thus 
from the combination of these two forces arose far more radical forms than 
was ever the case with Debussy. Nor did he have at his disposal that narcotic 
twilight which made complicated chord-formations tolerable to the ear of 
1908. 


The same might be said of Schoenberg, though it is difficult to determine 
so precisely when and how Schoenberg became acquainted with 
Debussy's music; according to his own statement in the Harmonielehre,* 
it was 'three or four years' after 1902. However, the influence of 
Debussy's piano-writing is unmistakable in much of the piano-part of 
Pierrot lunaire—a work which, in turn, was to exercise through its 
facture a transitory impression on both Stravinsky ((Mazatsumi’ from 
Tri stikhotvoreniya iz yaponskoy liriki (Japanese lyric poems) ) and 
Ravel (‘Placet futile’ from Trois Poèmes de Stéphane Mallarmé) т 1913. 

1908 was also the year of the earliest works in which the young 
Stravinsky reflected the influence of Debussy: the Scherzo fantastique 
and Fireworks for orchestra, followed by the First Act of his opera, 
Solovey (The Nightingale) in 1909. The Lento middle section of Fire- 
works also borrows from the opening of Dukas's Г’ Apprenti sorcier, and 
some of Ше ‘unmistakable Debussy-like ... harmonic combinations’ 
that have been discovered in early Stravinsky* really derive from his 
teacher Rimsky-Korsakov. For more than sixty years Russian com- 
posers had shown a certain interest in artificial scales and the harmonies 
derived from them: not only the whole-tone scale? but the scale of 


1 ‘Önéletrajz’, Magyar Irás (1921); reprinted in Bartók, Válogatott zenei írásai (Budapest, 
1948), р. 9. German version in Musikblätter des Anbruch, iii (1921), p. 89. 

2 Béla Bartók : Ein Beitrag zur Morphologie der neuen Musik (Halle, 1930), p. 2. 

3 p. 438. е.в. by Roman Vlad, Stravinsky (London, 1960), p. 6. 

5 See supra, p. 88. 


116 THE REACTION AGAINST ROMANTICISM: 1890-1914 


alternate tones and semitones invented: by Rimsky-Korsakov and first 
employed by him in his opera Mlada (1890), and not only the whole- 
tone mode with its augmented-triad harmonies, but ingeniously 
fabricated, non-functional chord-progressions founded on the whole- 
tone scale—as in the prologue to Zolotoy Petushok (The Golden 
Cockerel) (1907): 


Ex.51 | 
Moderato assai (J = 80) 
бор abra Wo. RET 4 te le, 
SS ae ea о и В 
Tekin) ЕН 
ЗА A MN Поле даже wig —  — Е т [ERR ЕО 


p Pice. D 
Р „Ба ? 
Glockenspiel СИЕ В ааа есе 


& Нагр LIES 4, — LLL pp pypeccfa yp ВТ Вет ДИ шаны pn qme Р) [fas Ира 
ЕС г... Сва а Бе а LIAE] (amd атна гла 


Gl.: Hp. 8ve higher 


Уы Е ЕЕ ЕЕ 
Vlas. 3 Е ова тения я ЕЕ 


| 
Ге |] 


arr Sa ies SEC ere РЕН 
mre e Ee So жака Se 
EE ИЯНЕ ee Ae ee [n нашаға 


Bn. I tacet а 
Иа 


THE WIDER INFLUENCE OF DEBUSSY 119 


where the alternation of 6/4 and 6/3 chords based on the descending scale 
admirably suggests the magical character of the Astrologer. It was from 
the same toy-box that Stravinsky took the harmonic ingenuities from 
which he fashioned the music of Kashchey and the Fire-Bird in Zhar- 
Ptitsa (The Fire-Bird) (1910). But great as was Stravinsky’s debt to 
Rimsky-Korsakov and other innovators of the period, the greatest (as 
he acknowledged) was to Debussy: “Тһе musicians of my generation 
and I myself owe the most to Debussy.'? 

The technical devices, especially the orchestral devices, of impression- 
ism—the kaleidoscopic open-work of melodic scraps and spots of 
instrumental colour, the additional dimension given to pure line by 
doubling at conventional or unconventional intervals simultaneously 
(often amounting to side-slipping of chords), varied or unvaried 
repetition of motives or mere figures or rhythmic patterns instead of 
thematic development—all tending to the disintegration of the hitherto 
accepted norms of texture and consequence, were adopted or adapted by 
composers whose creative personalities differed markedly from each 
other’s and from Debussy’s, as may be seen in the six excerpts shown in 
Ex. 52. (The contemporary listener had no opportunity to hear the 
Webern piece until March 1913; he would doubtless have been incredu- 
lous of any prophecy that this would prove the most seminal of the six.) 
Not everything that seems ‘impressionistic’ in the music of the period 
stems from Debussy; the passage in Schoenberg’s Pelleas und Melisande 
(1903), pp. 42-5, where Mélisande lets down her hair from the tower, 
probably owes more to passages in ‘Am Strande von Sorrent’ in Strauss's 
Aus Italien than to Debussy; but even composers whose appreciation of 
Debussy was severely limited were indebted to him technically. Such a 
one was Holst, who admired L'Aprés-midi and liked the Nocturnes 
‘but was never very happy about anything else’ and ‘hated’ Pelléas,? 
yet ‘Saturn’ and ‘Neptune’ (composed 1914) in The Planets could 
hardly have been written without the pre-existence of Debussy’s scores. 
It was another side of Debussy that appealed to Vaughan Williams, the 
diatonic or modal vein that runs from La Damoiselle élue (1888) to 
Le Martyre de Saint Sébastien (1911). 


1 See, for instance, André Schaeffner, Strawinsky (Paris, 1931), р. 21, and Edwin Evans, 
‘The Fire-Bird’ апа ‘Petrushka’ (London, 1933), p. 10. 

2 Conversations with Igor Stravinsky (London, 1959), p. 48. 

3 Imogen Holst, Gustav Holst (London, 1938), p. 147. 


118 
Ex. 52 


THE REACTION AGAINST ROMANTICISM: 1890-1914 


(i) DELIUS: In a Summer Garden (1908) 


FI., Ob. 


CA. 


Вп. D. Bn. 


Harp 


Slower and more reposefully 


А ві 2 5 3 3 == 
Ес T іс 


eye БИР 


| 
| 


Se ааа = 
алУ; (= al 
ow OS De EY LD 
3 0—7 
В. CL. Фр -----1 
ПІ 
сулу =====—=—===== ЕЕ: 
jo mro. = 772 
7 ща” -] 
фри 
4 
БЕБЕ == у == сш = == 


реа оо == 


s 


5 
| 4 
| 


| 


119 


THE WIDER INFLUENCE OF DEBUSSY 


BEA S ы ИВА НЕЕ ЕЕК | 
ерте с NOCT... caes a 
Ia E шышы] 
кезді 
a 

| шз! 
РЕБ — 
puteum ШЕСЕ 
tapas 
------ --- ----.| 
eel 


| 1 1 
n Е | | 

N | || 

м , A. 

il Т қ 

ші | 

i 


үү № 
il 
H | 
LIN il 
«ul 
mu 
di 
| | ШЕН! 
| ; | В i 


| Eo Al сё | 
і We 8) ТЕ idi 


----- 
4 
я 


о 
Bn., D.Bn. 
Timp, 


27+ 


14 


D.B. tace 


г. 


THE REACTION AGAINST ROMANTICISM: 1890-1914 


120 


ШОШ 

M 

ЩІ M 
ШШ 

re 
АЦ 88 
T] 


IY [qe — адыра 


Ba., D. Bn.| Уч 


#- 
TERA p md 
ee 


D 
EE 


Hns. 


p 


АА ла ыыы 


p? 


THE WIDER INFLUENCE OF DEBUSSY 121 


(ii) STRAVINSKY: Scherzo fantastique (1908) 
_ (4-89 


en ig al a ms nm ie eo ee 


ПОНЕ 
Е ar ay 
а [m 


Picc. 


Ob. 
СА. 


Вп. 


=) 

$, 

| І 
ІШ 


TAR "лт ш тщ с a rre іне. —— es | 
s n. І 2 Г) 
бб 2 Se Sy i тше E RE EE ааа 
i е p 
nu f f mf iut 
I p= 
НО = 
Е e Е-Е 
му т 122 [gp слан: 
= wr 
п, № 
pem еи 
"Tpts. Да 
т, Низ. III, IV р 


z des 

у a TERA Жы ыс тыш" еы шс — CAD == 
LIS [X lu I nao 
a а SS eae Eee [ ЕП 
> = 


(0 DU = 
= 2 ый dcc 
BS, Ee БЕ ы ы к лес Мыл ы сал... зі 
BS foc Б LD Р ЕИ с ee б 388 № БВ ИИ НОЕ ВААУ RA.: Е И LL LLL —LÉÉ—ÁLLÉÉLÁ— 
И.Р Se eS ЭРЕШЕ ос ее” 


:3. D.B. soli pizz. 


THE REACTION AGAINST ROMANTICISM: 1890-1914 


22 


1 


сшщ. 
[opus тана нам ml 
Це 


70 ag 
шах», 


Fi. 


> 


4-4 


6, по. 3 (1909) 


Ор. 


Zart bewegt Q 


(iii) WEBERN: Orchesterstück, 


lich 


50 zart wie mög 


рр-- = 


THE WIDER INFLUENCE OF DEBUSSY 123 


4-4. 4-4 d-d. 


aber etwas schneller 


[ пра Г. | pus ери EL —Me— - | 
[ 2 ба з Ан „Эй Мыш Se шы Ө" euo £n mE 

Fl. [ TAs тиви ЕР Es ЗЕЕ || 
П jeep] wt: gum cum 


SS сочен 1 
je ре] 
о ) 
a 
I++ Ф. De це а Р це 
EF >, Te | 107 че um pote | 
Hon и | —# БО О п ая 
Eu EE ЕО Sr T 
Dpp———— — = 
ко cer f hi E I Idol a pu 2811 
Glockenspiel CARLES е ның Ен ERES Кы бе 
Ето e 
Solo Vin. ? poo oe 


ти Dampfer 6 


o 8 8 — 


АСК | == 


Se ee e aS o e e S 


кошсо лае 
.B. eee EEE] 

| Ju 
D.B. (div) JES Fa NN 
(Вов lower) |2 Жена eni ume] 


ти Dampfer 


124 


THE REACTION AGAINST ROMANTICISM: 1890-1914 


(iv) BARTOK: In Full Flower, Ор. 10, no. 1 (1910) 


Fi. 


Ob: С.А. 


СІ. 


Vins. 


Vlas. 


mf espress. 

f) II б LII К ооа 

в =Й == =, у an 
21 c "e m 

7—0 аР Ее СУ, or en 
С 4 mm ae nint idu cm | БИШ === == ка элен ини ЕШР Cee. ll as Бр, 

о ои Һ D 
deb —€— == 1... 


сұ Be 
© mre = 
а [7 
as Са 
и! 22% 2 
П 026 Ре: 
єс аа; 5—6 — р. Be 
ЕЕ ЕН: ЕЕ Е, 
Se at ea ee SS ee eee, 
ок lm adum тые а ара 


125 


THE WIDER INFLUENCE OF DEBUSSY 


Веви НТ» 


Jd. dle 
a) 7 
И oí 
ae 


E 


ИТ 


аву 


с 
га 


Я 


|| 
| 


Е 


ЕУ 


10 


THE REACTION AGAINST ROMANTICISM: 1890-1914 


126 


i 
i 


A 
e 
сай 


ға 


THE NEW HARMONY IN NEW CONTEXTS 127 


THE NEW HARMONY IN NEW CONTEXTS 


The general techniques of impressionism lived on after the demise of 
the aesthetic that had given birth to them, and were employed mainly in 
attempts to rejuvenate romanticism (as in Skryabin's later style), but 
gradually exhausted themselves. With the new harmonic resources 
opened up by Debussy and his compatriots it was a different matter. 
Instead of being employed as elements in pointillisme or dissolved into 
‘romantic’ figuration and texture, often softened by appoggiaturas and 
the like, they were commonly used—even by Debussy himself in such 
late works as the Douze études of 1915--а the material of ‘pure music’.! 
Ironically, the added notes which Debussy had employed to blur the 
outlines of sound were now used to harden those outlines; the piled-up 
chords producing a sensuous haze of sonority into which he breathed 
the minimum of rhythmical life now gave dynamic weight and were 
pounded out in vigorous or subtle rhythms which often seemed the most, 
instead of the least, important element in the music, as in the *Russian 
Dance’ of Petrushka (1911) and the final dance of The Rite of Spring 
(1912). By 1908 piled-up chords were becoming common currency, 
but they were seldom introduced as starkly and uncompromisingly as 
by Bartók in his piano Bagatelles. Bartók was a pioneer of the new 
percussiveness on the very instrument Debussy had treated with such 
magical delicacy, but in any case the piano emphasized the ‘block’ 
character of chord agglomerations. The clarification of Ravel’s har- 
mony in the Valses nobles, when he orchestrated them, has already been 
pointed out? and in a piece written at almost exactly the same time as 
Bartók's Bagatelles, the second of the Fünf Orchesterstiicke, Op. 16, 
Schoenberg gave a classic demonstration of prismatic scoring. The first 
sketch? shows the naked harmonic basis (i), though the instrumentation 
is already indicated, but the effect to the ear is very different (ii): 


1 Debussy, letter to Stravinsky, 24 October 1915. Зее р. 108. 
3 Reproduced in Egon Wellesz, Arnold Schoenberg (Leipzig and Vienna, 1921), facing 
р. 122: 


128 THE REACTION AGAINST ROMANTICISM: 1890-1914 


(i) Mássige Viertel 
II 


Ob. 


B. Fl. + Hn. 3 
(muted) 


Bn. 


Tot. 


Tbn. 


Solo Cello 


The importance of fourths in the chord-structure, as in nos. 10 and 
11 of the Bagatelles, was no novelty by this date although the treatment 
of them was. Chords of superimposed fourths had existed as early as 
Satie’s Fils des étoiles (see Ex. 36 (1) )—to say nothing of Beckmesser's 
lute—and had been an occasional component of the impressionist 
vocabulary (Debussy's L'Isle joyeuse, Ravel’s ‘Vallée des cloches’ in 
Miroirs, and elsewhere). Schoenberg himself had introduced them in 
Pelleas und Melisande just before the appearance of the ‘Pelléas’ theme, 
where they sound less daring in their context and threaded through by 
the ‘Mélisande’ theme (Ex. 54) than when quoted in isolation, as they 
are in the Harmonielehre, though he claims there that they occur ‘quite 
isolated ... as the expression of a mood, the strangeness of which 
obliged me against my will to find a new means of expression’. Further- 
more, each is resolved; they аге not ‘emancipated’ chords like Debussy's 
and Ravel’s. But in 1905 both Mahler and Strauss (Ex. 55) employed 
quartal harmony and melodic lines derived from it in the Seventh 
Symphony (i) and Salome (ii) and (iii). 

So the celebrated opening of Schoenberg’s Kammersymphonie was in 

1 p. 450. 


THE NEW HARMONY IN NEW CONTEXTS 129 


Ех. 54 


Woodwind 


Brass 


Mis are a А 
Е ae ae 
== eae 91 a Se 1 11777077 97 Ла 2] 

жена ө e ———— ee кы acm 


Strings 


С === 


130 THE REACTION AGAINST ROMANTICISM: 1890-1914 


Ex. 55 ; 
Рій mosso [than Adagio] 


4 


дай E CRIT 
Woodwind 


FEED 
ERES eo 
Ses Li ECL [ == 
mm — p 


Tpts. 


Vins. 


sempre Р 
x 


ЕЕ Че жю жже е" —— 7 


oo. ee e 
Ее аа ынны + чырыш шш A „a 
à EE EE EE 2 2 
prec. | Ob. à3; Cl. à3 
Он { SP = ИЛ РЕ 
ҰЯ Jae eel к= 
01-5 эө Se g—9 —]——-— 
[пе Mi — ee He SSS 
V Глисти сочи Iz] | 


= О е. =. , . 
o y . Lo” „е 
а” pr Pp | —— — ДИНЫ БЫЛ — Jp IETS Sao HP 
mi tpe а! елате mi.  ———— E eel ere [УЫП [n cani СҮТ ШЫР НО ЧН ui o CM Daun 
Ге“ 1 «СО ШЙНЕН Juni OmU „ый |) БЕРЫ he i) ЭНЕ Tree) 201 В тышсыз рр а uer 
д 5 LE ГЕ LOE LE 
p cresc. . - 
cresc. molto 
prs 8 


Ро molto cresc, 


131 


THE NEW HARMONY IN NEW CONTEXTS 


(i) 


“i EH i \ 


pizz. arco 
Bem ГЫ 


a Pe 
аи t 
А 
puce р Eee 
y 
ЕЕ Se]. Ca 
г Ф |) 
fe 
P prc o el 
2 рел] 
"ВЕЕР pmi 
= 
Е Е ст ЕТ) 


76) 


po 
ua 
| 
aw ЖЫН ЖЫН 
куст нуш с ру 
a y x eb e 
2 ) ) 

IX ] 
= IN L] 
ШЕР ди MET, E 
hlc o d —— анын 

ae Р 

plzz. arco 

Да es БЕП 
7—8 m 
pour 04 ERI 


М | i m m 
НЯ “| Ў s. M I. E 
ЕЕ |Б > WU n ^ J| mes M 
EM а l^ m. = I Ш 
ІШ = МЫ m X 
um Mj : B OR Rm A ts m 
ER PS PR Ж ui E ОА АА МИ. 


Breit (4 


С! 
Bn 
Horns 
Timp 
Small 
Kettle-Drum 
inE 
Tambourine 
Side Drum 
Vins 
Үс 
(Ist desk) 
Vc. 
D.B 


(iii) 


THE REACTION AGAINST ROMANTICISM: 1890-1914 


132 
Ex. 56 


в 


м т 
АЧ [i 
ug 
M 

mul 
n р 
Vii ilii 
Ш CL 
ili 
ПІ 
IN 
Г mul | 
3 i its 
Ша li | 
¢ Wa i 
Benita Сө A 
Зее slimes 
Оз. d 
KE] №. (> 
ПЕЙ" ІЙ 
5 
3 


im TT 
Г S : () à 
4 нео = ръце 
| | | 
ПЕ! | | 
E | 
M БН | 
UN. На 
|| ү Gath EE | 
ГБ | | 
О ^ d П 
А м mN 
li RE The 
ih MC ІҢ 
NS E | 
| all 
ts s il "i 
Ri al "Pu "n! 
We fmi | | 
ia Ш | 
Exo d и di 


THE NEW HARMONY IN NEW CONTEXTS 133 


itself only abreast of other German music of its date (1906); the fourth- 
chordal climax of the Kammersymphonie (Ex. 56), occurs later, at the 
end of the ‘development’. Quartal music on this scale is the furthest 
possible reaction from hyper-chromaticism. What was new about the 
horn-theme was that it was “риге music'. Whereas Mahler and Strauss 
and Schoenberg himself in Pelleas had turned to quartal music to 
express a strange mood, or tense excitement, or to symbolize the 
prophet's inflexibility or suggest the barbaric, it now existed for its 
own sake as it was to do in Berg's Piano Sonata, Bartók's 
Tenth and Eleventh Bagatelles, the end of his First String Quartet 
(1907) and very much of his Second (1915-17), and a great deal of other 
music of the next decade. In the ‘Spring Divinations’ of The Rite of 
Spring (1912) Stravinsky throws in Schoenberg's six-note quartal chord 
(Ex. 56) as the climax to an ostinato pattern just before fig. 18. A similar 
accumulation of perfect fifths occurs half a dozen bars earlier (see 
Ex. 57); 

Like the whole-tone mode, quartal music generally remained an 
adjunct to more normal music. Except the egregious Rebikov,! who 
had certainly employed occasional fourth-chords in undistinguished 
piano-pieces of the late 1890s but whose monstrosities of quartal 
commonplace, the Вейуа pesni (Chansons blanches) Op. 48, and 
Dances, Op. 51, for piano, date from c. 1913, the only composer to 
construct a system of fourth-chords was Skryabin. Unlike almost all 
his contemporaries, Skryabin did not confine himself mainly to the 
perfect or diatonic fourths derived from impressionism though he 
employed them in his early works in passing (cf. Ex. 12, bar 3); nor 
was his basic chord a purely intellectual conception.? When he began 
to compose his orchestral Prometheus in 1908, he consciously based it 
on what he called his ‘synthetic harmony’: the basic chord С, Е sharp, 
B flat, E, A, D, an arrangement in fourths of the scale C, D, E, F sharp, 
A, B flat which corresponds to 8, 9, 10, 11, 13, 14 in the harmonic series 
—or, rather, to the approximation attainable on the piano keyboard— 
and also differs only in one note (A instead of G sharp) from the 
whole-tone scale. According to Leonid Sabaneyev, it was he who 
devised ex post facto the theory of overtone derivation, which was then 
appropriated by the composer ‘almost as a dogma',? but Skryabin's 

1 See p. 36. 

2 For a survey of Skryabin's harmonic evolution see C. С. J. von Gleich, Die sinfonischen 
Werke von Alexander Skrjabin (Bilthoven, 1963), pp. 85-90. Paul Dickenmann, Die Entwick- 


lung der Harmonik bei A. Skrjabin (Bern, 1935) deals only with his pre-Promethean 


harmony. 
3 Sabaneyev, Vospominaniya o Skryabine (Moscow, 1925), p. 227. Schoenberg likewise 
held that *dissonances . . . are merely more remote consonances in the series of overtones’ 


(Structural Functions of Harmony (London, 1954), p. 193). 


1890-1914 


THE REACTION AGAINST ROMANTICISM: 


Ех. 57 


134 


Flutter-tongue 
Le 
lini 


am | 
Ш (ТШ 
like 
$ i 
АП ӨП 
“ill 
А М 
E 
АР 
i ШШ 
Ш 


g 
її = 
тъ Š 
= 
2 + 
3 З 
"Bb 8 
в: 
а % 
alll = 
кг! ї 
ар! Fla. 
Іш ІН 
ue A aa 


con sord. 


| | 


mh 


135 


THE NEW HARMONY IN NEW CONTEXTS 


| 
Де Фе X 


XI 


=> 


136 THE REACTION AGAINST ROMANTICISM: 1890-1914 


piano-playing fingers had already discovered the sound of the ‘Prome- 
thean chord’, no doubt empirically, in the course of natural evolution 
from the post-Chopinesque harmony of his early compositions. Two 
examples occur in the second and third bars of the Prelude, Op. 37, no. 3, 
written in 1903: 


Ex. 58 


Andante ( = 50—54) M 
m; marcato : | 


The technique is, of course, fundamentally different from the block 
treatment of quartal harmony, though it is equally non-functional. 
When Rebikov hit on something surprisingly close to the ‘Promethean 
chord’ in 1907, in his ‘musico-psychological tale’ Bezdna (The Abyss, 
based on Leonid Andreyev’s story), he could do nothing better than 
side-slip it to accompany the moment when the student ‘gazes raptur- 
ously’ at the schoolgirl: 


Ex. 59 


Рій mosso [zkan Па! 56] 
9 


But Skryabin, having taken his basic harmony as a concord, treats it 
freely in any transposition, breaking it up into melodic fragments and 
figuration, and admitting ‘foreign’ notes as appoggiaturas or passing- 
notes. The chromatic parallel ninths which open the Étude, Op. 65, 
no. 1 (1911-12), are drawn across Promethean chords as a nineteenth- 
century composer would have drawn a simple chromatic scale across 
conventional harmonies. In Prometheus itself the complete six-note 
chord is used, but in Skryabin's post-Promethean compositions the last 


THE NEW HARMONY IN NEW CONTEXTS 137 


partial may be omitted, as in ‘Etrangeté’, Op. 63, no. 2 (1911-12)! 
while in the seventh Sonata, Ор. 64 (1911-12) it is flattened. 

The paradox of Skryabin, the arch-romantic with apocalyptic visions of 
an art, or at least an act, embracing all humanity, is that he should have 
invented a musical language of non-communication, a language the extra- 
musical sense of which only he could understand. Sabaneyev tells us? 
that Skryabin considered the passage in Prometheus quoted as Ex. 60 
*the most tragic thing I have written in my whole life', but although 
Sabaneyev was closer to him than any other colleague he was quite 
unable to agree. Skryabin contributed as much to the final destruction 
of functional harmony as any of the Central Europeans, and the 
‘artificial’ constructivism of his last works was not merely symptomatic 
of a tendency, like Rimsky-Korsakov's in the Cockerel and Stravinsky's 
in The Fire-Bird: it pointed a way—which no one exactly followed, 
though Schoenberg was soon to trace a parallel one—through the 
potential chaos of atonality. Both the Promethean chord and the 
twelve-note row provide synthetic nuclei for a composition and since 
the опе can be ‘verticalized’ into chords and the other ‘horizontalized’ 
into melodic lines, they appear superficially to be opposite sides of the 
same coin. But the notes of the Promethean chord are not ‘related to one 
another' in Schoenberg's sense, and the six-note scale deducible from 
the chord (without transposition) is apt to sound like a chromatically 
altered diatonic scale, though it lacks points of tonic, dominant and 
leading-note function. In his last compositions, the five Preludes, Op. 74, 
of 1914, Skryabin introduced such a profusion of ‘foreign’ chromatic 
notes that he reached a kind of twelve-note music, but it is certainly 
not the Schoenbergian kind.? 

A rather younger Russian contemporary of Skryabin's, Nikolay 
Roslavets (1880-1944), did actually (for instance, in his Deux Composi- 
tions pour piano (1915) ) go farther than Skryabin toward twelve-note 
music.* In an autobiographical article published in 1924,5 Roslavets— 
while denying that his system of sintetakkordi of “6 to 8 or more notes' 
owed anything to post-Promethean Skryabin or to Schoenberg— 
admitted that ‘Skryabin (in a musical-formal respect, but in no wise 
ideologically [Roslavets was an anti-romantic constructivist]) is of 

1 Recorded in Тйе History of Music in Sound, x. 2 Ор. cit. р. 228. 

з On Ше analogies and differences between Promethean and twelve-note structures, see 
Zofia Lissa, *Geschichtliche Vorform der Zwólftontechnik', Acta Musicologica, vii (1935), 
p. 15. George Perle, Serial Composition and Atonality (Berkeley and Los Angeles, 1962), 
p. 41, considers Skryabin as a ‘nondodecaphonic serialist’. 

4See Detlef Gojowy, ‘Nikolaj Andreevic Roslavec, ein früher Zwólftonkomponist', 
Die Musikforschung, xxii (1969), p. 22. 


5 *Nik. A. Roslavets o sebe i svoem tvorchestve', Sovremennaya muzika, no. 5 (November- 
December 1924), p. 132. 


138 
Ex. 60 


Ob. & 
CA. 


СІ. 


B. Cl., Bn., 
D. Bn. 


Horns 


Tpts. 


Tbn. 
Tuba 


Vins. 


Vlas. 


THE REACTION AGAINST ROMANTICISM: 1890-1914 


_———— I 
ИТ не: eg 2. ys 
LE—————4— 155498 Арапя те 5&9 
Кес = ДЕР ker а= рано ра H9 9 я 
[xao EE ey И жойи ГЕ = gue wd каналын Е жк грис Т] 
2 a a oe р===ы 
771, е 
III & С.А. ур 7 II cresc 


II ІП 


Қары. ЕСТЕ ат тел 


стезс. 


[-9 )-—7 Dro 
Eas = ше хак те сызгы 
рс эн = eee | 
II III IV 
mf. тр cresc. 
f) ІП 
yp mc e en 


Ц 
ii 


Г 
| 


: 
| 


IV ' con sord. 


div. 
f) unis. h „ ex ШӘ iT 3 
ge Д Г д дет. e eR 
Дай 4 = инь Г ЕР ДТҰ HN P Uds | 
L Г аъ See. ы ay x l L— M [aac cd 


THE NEW HARMONY IN NEW CONTEXTS 139 


dim. 
Р Р subitement trós doux 
1+ Pic. De 7 Ec c M MEE EE 
2- 3 


f) ene 4: I £r = 
[У о ры 


ЕУ АЕ SPS ЕЕ ЕЕ uen mI Ч] 
Fl. Г ШС Е ее, Zeg 
MIU. ЖЕШ 1 а= eee a ГЕННӘН р ЫМ 


L—— 


Ob. & 


— 
к- 
[1 


| 
Б 
c 
L| 
| 


Horns [em mr cuna -— Bee Е 
Eor EE с па и ЕЕ Е к г 


У( 
7 

he 

w 

) 
М 

в. T 


Тр. 


Vins. 


Vc. 
D.B. 


140 THE REACTION AGAINST ROMANTICISM: 1890-1914 


course far nearer to me than Schoenberg, whose work, I confess, I have 
got to know only comparatively recently’. 


SCHOENBERG 

Compared with the brilliantly erratic side-track of Skryabin's Pro- 
methean experiment, Schoenberg's progress from romanticism, through 
expressionism and non-serial atonality,! seems a straightforward, even 
logical process of evolution. For it was he who gathered the harmonic 
fruits of impressionism—we must take note of his claim to have 
discovered them independently of Debussy?—and integrated their free 
dissonance with post-Wagnerian harmony and then with the polyphonic 
devices of ‘absolute music’. In a lecture given in 1941? he described the 
position around 1908, and hence his own historical role, with great 
accuracy: 


[Debussy's] harmonies, without constructive meaning, often served the 
coloristic purpose of expressing moods and pictures [which] . . . thus became 
constructive elements, incorporated in the musical functions; they produced 
asort of emotional comprehensibility. In this way [as in post-Wagnerian 
harmony] tonality was already dethroned in practise, if not in theory. This 
alone would perhaps not have caused a radical change in compositional 
technique. However, such a change became necessary when there occurred 
simultaneously a development which ended in ... the emancipation of the 
dissonance. . . . One no longer expected ... resolutions of Strauss’ discords; 
one was not disturbed by Debussy's non-functional harmonies, or by the 
harsh counterpoint of later composers. . . . 

A style based on this premise [emancipation of ‘Wagner’s, Strauss’, 
Moussorgsky’s, Debussy’s, Mahler’s, Puccini’s, and Reger’s more remote 
dissonances'] treats dissonances like consonances and renounces a tonal 
center. By avoiding the establishment of a key modulation is excluded. . . . 


He then goes on to describe the problems raised by the first composi- 
tions in this style by himself and his pupils Webern and Berg. “Тһе 
foremost characteristics of these pieces in statu nascendi were their 
extreme expressiveness and their extraordinary brevity.' Hitherto, 
harmony “Чай served as a means of distinguishing the features of the 
form'; transitional passages demanded different harmonic treatment 
from those leading to cadence-points; ‘harmonic variation could be 
executed intelligently and logically only with due consideration of the 
fundamental meaning of the harmonies': 

Fulfillment of all these functions ... could scarcely be assured with chords 


whose constructive values had not as yet been explored. Hence, it seemed at 
first impossible to compose pieces of complicated organization or of great 


1 See chapter 1, pp. 15-17. ? Harmonielehre, pp. 435, 438, 450-1. 
з Printed іп Style and Idea (New York, 1950), particularly pp. 104-6. 


SCHOENBERG 141 


length. A little later I discovered how to construct larger forms by following a 
text ога poem. ... 


But it was only ‘after many unsuccessful attempts during a period of 
approximately twelve years’ that Schoenberg ‘laid the foundations for а 
new procedure in musical construction which seemed fitted to replace 
those structural differentiations provided formerly by tonal harmonies’: 
the method of the twelve-note series.! 

At first there was no question of 'absolute music'. Schoenberg 
re-emphasized the ‘extreme expressiveness’, the ‘extreme emotionality’, 
of his pre-serial atonal compositions—though they were no more 
successful than Skryabin’s in communicating their message to a wider 
circle. And for the same reason: the message was in a language com- 
prehensible to few but the composer himself. But the real problem, as 
Schoenberg makes abundantly clear, was not of expression but of 
construction, a purely musical problem, and the search for a solution 
forced Schoenberg to concentrate his thought on the structural devices 
of ‘pure’ music: canon and ostinato, inversion and variation. He had 
very early shown a predilection for canon: for instance in the first 
movement of the early String Quartet in D (1897) at the beginning of the 
development: 


Ex. 61 
Allegro molto | E ж ; si n 


1 See infra, pp. 341 ff. 


142 THE REACTION AGAINST ROMANTICISM: 1890-1914 


and in the finale of the same work (bars.65 ff.) where the close canon 
between first violin and cello is much more ‘classical’ than the canon— 
also between solo violin and solo cello—in Tove's first song in the 
Gurrelieder. Now he began to construct with microscopic ingenuity, as 
in the first piece of Op. 11: 


I 
langsam 


In Op. 19, no. 2, an ingenious analyst! has discovered a foreshadowing 
of serial construction and at the same time (1911) Schoenberg felt the 
need to write a textbook that should ‘thoroughly impart to a student the 
handicraft of our art, as a carpenter always сап’.? The highest point of 
his own craftsmanship at this period was reached in the later numbers 
of Pierrot lunaire (1912), for instance, по. 17, ‘Parodie’, which begins 
with a canon by inversion between viola and clarinet, while the Sprech- 
stimme follows the viola exactly at a bar's distance (Ex. 63). 

At the repeat this theme becomes a normal canon between Sprech- 
stimme and piccolo, to which is now added another canon by inversion 
between viola and clarinet; the ‘accompanying’ piano part is also freely 
varied. In the next number ‘Der Mondfleck’ similar ingenuities are com- 
pounded by retrograde movement. Not all the numbers of Pierrot lunaire 
are as intellectually constructed as this. No. 8, ‘Nacht’ is a passacaglia, as 


1 Rudolf Wille, *Reihentechnik in Schönbergs opus 19, 2, Die Musikforschung, xix 
(1966), p. 42. 
з Harmonielehre, p. 7. 


SCHOENBERG 143 


Ex. 63 
(Ò = 120) . 
grazioso 
ес ESS оо =т=т CO | 
а. pape == aT e | 
Ро ЕЕ = ттын кет== Lj mm л] = Бар 


Eb 
D == p 
mit Dàmpfer 
() RE ED VEN. 

—JA——A4—————————s 4 LI d БЕ ое” Ж. 

Vla. mo ы: к= — еў ее ес f= Lace] 
ГА (3 отъ - лот ре Л 

= - LI 
Sue P stace. spice. 
— À 


() 
a Lu EMI o -- о ДЕТТЕ ЕЕ 
Sprech ГА 2 
summe г КЛ 


() 
pu Дет 6 сс „л 
ВЕУ 4 
= 
D EL mu Е 
> caro 
22 Е ; 
Piano m Е fo JP 5 
га PR IE ыс И. A ИЕ = 
съ: 7 Е SS 


[m Wes „лый |2 “ІСІ 
DES ес Эш сы xm ту а И Ц P=. а 
Se SS А cal [|— —]-aT р — —w 1 NES ЖШ ШЕШ 
== 
ve > Ба zi me | 
SS спин 
é ——OÁ' À ет | 
РЕБ’ ЕЕ Б] ЗЕЕ Sa D rmi 231—377] 
[Ji — со ысыта SS ие аар: АЕС meu. “ыы БЕП 
Ta fas lo DGS un | rcm ae a es ———————[————/———— урра — [3] 
pm ЗИ IN RLS all dme a ЕЕ арасан 
r1 ж = xm n 
---- а > 
—— = Sb 
Е: 
f) = > —о => ——— 
EIS 01027 Е ЕЕ Т Сс" "чу" ТЗН: АЕН ЕЧЕИ NR RINT V 


Mee - -па-деш, blank und blin - kend, in 
С ЕЕ a 


[m IS то 
Гея» Урна Lie emma —— — ГОТ 3.7 Е 
Lv шша ый Е] 
v D р RJ ә рә 
7 b EIU NS SS: 
D Н H h 
va чо НФ ИБ sy —— ——— | 
p I ba Шан — qM тии S U-N-.« sx] 
[7 (eS re eee ee a 2 ee 
ен] EY ДЕННИ жетісті) 
€——— 


144 THE REACTION AGAINST ROMANTICISM: 1890-1914 


C NS sentimental 
() ты 
ЭЕ —=— Е ee ааа [= 
| —— а < 4 A T— 
Au < р dolce 


sentimental 


b at 
e c ame ae oe eee =] 
Де 4 ПТ Бі; ТЕМ БЕРІ шл падна р {== Ее НЕ Бе 
Hoe P Lid ae gs В раа ести 


ih - rem | grau = = en Haar, 


а 


(Knitting-needles, polished and gleaming, in her grey hair,) 


the composer points out. (The first of the Orchestral Pieces, Op. 16, is 
held together by a three-note ostinato figure in the bass.) But despite the 
ultra-romantic decadence of the Giraud-Hartleben poems, a decadence 
subtly suggested rather than expressed by much of the music of the 
*three-times seven melodramas’, ‘Parodie’ and ‘Der Mondfleck? are 
essays in constructivism as ‘purely’ musical as the works of 1923, the 
Klavierstiicke, Op. 23, and the Serenade, Op. 24, in which Schoenberg 
began to ‘compose with twelve notes’. They are not neo-classical, any 
more than Reger’s fugues are neo-classical—or the overture to Ariadne 
auf Naxos or The Rite of Spring. More deserving of that epithet are 
some of the French compositions of the War years: Fauré’s song-cycle 
Le Jardin clos, Ravel’s Trio, Debussy’s piano Etudes and the Sonates 
pour divers instruments, of the first of which—for cello and piano—he 
himself claimed that the proportions and form were ‘presque classique, 
dans le bon sens du mot’.! But all are symptoms of the same turning 
away from romanticism, the same preoccupation with the basic 
materials of music for their own sake, which were to characterize the 
mainstream of Western music from the 1920s for the next half-century. 


1 Quoted by Vallas, op. cit., p. 365. 


Ш 
STAGE WORKS: 1890-1918 


By MARTIN COOPER 


WHEN Wagner died in Venice on 13 February 1883, his music was still 
hotly discussed but the works had been very little staged outside Ger- 
many and Austria. Elsewhere Tristan und Isolde had been given only in 
London, Die Meistersinger von Nürnberg only in Riga, London, Copen- 
hagen, and Holland; and complete performances of Der Ring des Nibe- 
lungen only in London, Brussels, and Amsterdam. Paris did not have a 
performance of Lohengrin until 1887, followed by Die Walküre in 1893 
and the complete Ring cycle only in 1911. Bayreuth and Munich were 
places of pilgrimage for all Wagnerians, and the list of French 
composers who made this pilgrimage during the seventies and eighties 
includes all the most gifted and distinguished of the rising generation. 
Places of pilgrimage, however, cater by definition for the converted; and 
the large majority of opera-goers outside German-speaking countries, 
and a very high proportion of musicians everywhere, had had very little 
opportunity to judge Wagner's works except by isolated passages 
transferred from their context to the concert-hall. When these works 
began to be generally performed and became familiar, it took only 
a single generation to discover that, far from heralding the dawn of 
a new musical era, Wagner's music represented the all but final stage ina 
great musical epoch which had begun in the second half of the sixteenth 
century. Much, too, of what Wagner and his immediate disciples had 
believed to be of universal significance proved to be either personal 
in application, or relevant at most to German composers for the 
stage. 

In spite of this, opera during the years between 1890 and 1918 was 
dominated by Wagner's shadow, and every composer for the stage was 
forced to adopt an attitude to Wagner's music and to Wagner's ideas of 
dramaturgy. It was possible to accept the one and reject the other, as 
happened in France, and outright rejection of both became increas- 
ingly common; but it was impossible to ignore the questions that 
Wagner had asked or the answers which his own works had provided. 


146 STAGE WORKS: 1890-1918 


CENTRAL EUROPE 


Wagner's fundamental contention that symphonic and chamber music 
belonged to the past and that the future of music lay with the music- 
drama rested on twin generalizations, of his own personal gifts on the 
one hand and of German musical aptitudes and conditions on the other. 
His further identification of German music with himself was proved 
unjustified; but not until his influence had done much to weaken the 
strong national traditions which had hitherto divided the operatic world 
into largely self-subsistent compartments. It is not surprising that this 
influence was strongest in Germany and Austria, where Wagner’s 
dramatic theories were linked to literary, intellectual, and even political 
movements, and Musikdrama appeared not only as the latest develop- 
ment in the history of opera, but also as the first fully mature embodi- 
ment of the form in wholly Germanic terms. Wagner's expansion and 
enrichment of the orchestra, and its greater part in defining as well as 
supporting the drama, were universally copied; but his conception of 
opera as symphony proved to have a limited personal validity and was 
ultimately rejected by even his most enthusiastic followers. The most 
gifted of these was Richard Strauss (1864—1949), whose first opera 
Guntram (1894) is a naive continuation of the knightly world of 
Lohengrin, while in Feuersnot (1901) we have an example of that 
specifically German delight in an idealized version of the national past, 
mythical or historical, viewed in the same golden, transfiguring, and 
largely diatonic light that Wagner had created for Die Meistersinger, 
enhanced by a nostalgic counterpoint employed to heighten the illusion 
of the medieval past. Although Tristan continued to be the most 
musically influential of Wagner's works, it was Parsifal and Meister- 
singer that were most copied by opera-composers of this period. 

The Meistersinger world, as we may for convenience call it, proved a 
particularly rich vein. It was exploited successfully by Engelbert 
Humperdinck (1854-1921), who had worked for a time at Bayreuth as 
Wagner's assistant, in two Mdrchenoper, or fairy-story operas, in which 
a Wagnerian harmonic and orchestral style is scaled down so that the 
disproportion between manner and matter is not offensive. Hänsel und 
Gretel (1893) belongs to a class not uncommon in the nineteenth 
century: a work of art conceived ostensibly for children but in fact 
reflecting an adult's sentimental idealization of childhood and very 
little appreciated by children themselves. It is musically inferior to 
Kónigskinder (Royal Children) (1910), where the medieval atmosphere 
is plainly modelled on that of Die Meistersinger, and there is even a 
boys' dance that suggests an obvious parallel with Wagner's apprentices. 


CENTRAL EUROPE 147 


A more important parallel is the personification of the power of music 
in the character of Der Spielmann, whose (offstage) violin solo in Act III 
provides the focal point of the drama, like Walther von Stolzing's Prize 
Song; and the children's music, which Humperdinck introduces in the 
same spirit as the opening chorale of Die Meistersinger, emphasizes the 
naive, folk-like element in the work. 


Ex. 64 


Kinder а. 


Lie - ber Spiel - mann, al - le Kin-der und ich, wir 


ha - ben ge-be - ten, ge - fleht für dich 


(Dear Mr Player, all the children and I have begged and prayed for you) 


The personification of music in a single character! and the sophisticated 
use of simple, popular material as a reassuring feature in intellectually 
demanding or emotionally threatening situations? are common features 
of opera in this period. Hugo Wolf (1860-1903) further exploited the 
‘golden’ Meistersinger atmosphere in Der Corregidor (1896), formally 
weak but musically rewarding and containing a Beckmesser part in the 
amorous Alcalde. A different kind of Wagnerian affiliation appears 
in Der arme Heinrich (1895), the first opera of Hans Pfitzner (1869- 
1949). Here the music owes as much to Weber as to Wagner, but there 
is a clear echo of Parsifal in the harmonies associated with the rela- 
tionship between Agnes and the wounded Heinrich. Pfitzner's 
Marchenoper, Die Rose vom Liebesgarten (1901) is an essay in that 
sophisticated simplicity more successfully exploited by Humperdinck, 
and it was admired and performed by Gustav Mahler, whose own 
music contains many of the same features that distinguish the operas 
written by members of the first generation after Wagner's death— 
idealization of an imaginary past (Des Knaben Wunderhorn songs) and 


1 СЕ. Beppe in Mascagni's L'Amico Fritz (1891), Sadko in Rimsky-Korsakov's Sadko 
(1898), Floria Tosca in Puccini's Tosca (1900), the Jongleur in Massenet's Jongleur de 
Notre Dame (1902), the Dark Fiddler in Delius's Romeo und Julia auf dem Dorfe (1907), 
the Italian Singer in Strauss's Der Rosenkavalier (1911), and Palestrina in Pfitzner’s Pales- 
trina (1917). 

? E.g. the shepherd boy at the opening of Puccini’s Tosca, Act ІП; the child's piano 
practice in Leoncavallo's Хага, Act П; the children's Christmas song in Massenet's 
Werther; and the fisher-girl’s humming a folk-song at the opening of d'Indy's L’Etranger. 


148 STAGE WORKS: 1890-1918 


the innocence of folk- or child-elements contrasted with psychological 
complexity and suffering (Symphonies 3, 4 and 8). In the operas of 
Siegfried Wagner (1869—1930), Wagner's son and Humperdinck's pupil, 
humour is added to the symbolism of the Магсйепорег (Der Bürenhüuter, 
1899); and Austria produced a variety of popular post-Wagnerian 
crossings with folk-music in Der Evangelimann (1895) by Wilhelm 
Kienzl (1857-1941) and Der Musikant (1910) and Das hóllisch Gold 
(1916) by Julius Bittner (1874-1939). 

It was not only the example of Wagner's music, or his professed 
belief that myth and symbol were the proper concern of the music- 
drama, that turned the minds of many composers in this direction 
when searching for a libretto. The Symbolist movement in France, led 
by the poet Stéphane Mallarmé, had itself been influenced by Wagner's 
theories! and was in its turn to furnish composers with the new ideals of 
hermetic reserve and indirect allusiveness characteristic of a mandarin 
art. Symbolism appeared as a reaction against the realism, or natural- 
ism, that marked the development of the French novel in the hands of 
Flaubert, Zola, and Maupassant; and realism and symbolism were the 
dominant elements in opera between 1890 and 1910, at first strongly 
contrasted but later amalgamating in many instances. 

Both the mosaic-like structure of Wagner's musical idiom involving 
an intricate system of cross-references, and the richly poetic and 
evocative resources of the Wagnerian orchestra clearly corresponded 
more closely to the Symbolists’ use of language in poetry than to the 
prose of the naturalist novelist. Wagner's own symbolic conception of 
opera was carried to its logical extreme in the two works that Strauss 
wrote after Feuersnot—Salome (1905) and Elektra (1909). In both of 
these the role of the orchestra is more important than in any of Wagner's 
works, and the structure and even the scale recall those of Strauss's tone- 
poems. Oscar Wilde wrote his Salome, on which Strauss based his opera, 
in French, and the central figure was inspired by Gustave Moreau's 
picture ‘L’Apparition’; but Wilde's reference to ‘refrains whose recur- 
ring motifs make the poem like a piece of music and bind it together as a 
ballad'? almost certainly betrays a superficial acquaintance with 
Wagnerian theory. His description of the work to Sarah Bernhardt 
as ‘quelque chose de curieux et de sensuel’? suggests that, like Debussy, 
he was already interested in the so-called ‘theatre of cruelty’; and the 
perverted eroticism of Salome, which gave the work a succés de scandale 

1 See Mallarmé, ‘Richard Wagner, rêverie d'un poète français’, Revue wagnérienne, no. 7 
(August, 1885) p. 195. For a study of the impact of Wagner's ideas on French literature see 
Eliot Zuckerman, The first 100 years of Wagner's ‘Tristan’ (New York, 1964), pp. 83-122. 


2 Oscar Wilde, De Profundis (London, 1905), 26th ed., p. 66. 
3 Quoted in Norman Del Mar, Richard Strauss (London, 1962), i, р. 240. 


PLATES 


STRAUSS'S SALOME, 1905 (see p. 148) 
The dance of the seven veils in the original Dresden production 


CENTRAL EUROPE 149 


clearly foreseen by the composer, provided a piquant sauce welcome to 
jaded fin de siécle appetites. Although the crucial Dance of the Seven 
Veils! 15 as weak and commonplace as the dance section in Strauss's 
Also sprach Zarathustra, and Jochanaan's music has the flat, oleaginous 
quality of much painting of the Nazarene School, Salome is a brilliant 
tour de force unique in the history of opera. The characterization of 
every member of the Idumaean court is needle-sharp and concise, and 
Strauss applied all his phenomenal orchestral virtuosity and musical 
inventiveness to set his characters in high relief against the general 
background of sultry, acrid hysteria. Psychological motivation is 
supplied by the ingenious cross-references which form the web of the 
music, whose nervous plasticity follows every physical movement or 
gesture by corresponding rhythmic and melodic shapes as skilfully 
chosen as in the finest of the tone-poems. Ex. 65 (i) expresses Salome's 
violence and perversity, (ii) Jochanaan's prophetic conviction: 


. ees от 
—— Á— оне 


Бізге See SEE] Е М = 
jm scan [main oa RETI —— — — ТУР 
Өй ыш кч NT 
акы 707 Е 


In Elektra Strauss collaborated for the first time with the Austrian poet 
Hugo von Hofmannsthal, whose libretto is deeply coloured by 
Nietzsche's insistence on the part played in Greek tragedy by the dark 
Dionysian world as a counterweight to the bright Apollonian air of 
Goethe’s griechische Heiterkeit. The difference between Strauss’s 
thematic vocabulary here and in Salome is immediately noticeable in the 
simple and majestic phrase associated throughout the work with the 
figure of Agamemnon, and providing the Grundgestalt for countless 
psychologically significant derivations and variations (Ex. 66 (1) ). By the 
side of this diatonic fanfare the theme associated with Klytemnestra has 
the same penetrating and concentrated suggestion of perverse evil as is 
to be found in Salome (Ex. 66 (ii) ): 


X Ser iplis Но 


150 STAGE WORKS: 1890-1918 


There is a further parallel between the two operas in the figure of 
Chrysothemis, Elektra's sister who stands, like Jochanaan in Salome, 
outside the main preoccupations of the drama. But whereas Jochanaan's 
austere and unbending ethical idealism prompted only second-hand 
reminiscences in the composer, Strauss's natural sympathies were 
stirred by the impassioned femininity and life-affirming instincts of 
Chrysothemis, the broad lyrical sweep and diatonic strength of whose 
music sets her in harsh contrast to the rest of the characters. It is 
significant that in the scene with Orestes even the obsessed, half-crazed 
Elektra catches some of this vital warmth: 


J — 108 ELEKTRA 


151 


CENTRAL EUROPE 


Au - gen mich sehn, 


bild 


_ mir ge-schenk-tes Traum 


(O let me behold your eyes, vision that has been granted to me, lovelier than 


any dream) 


152 STAGE WORKS: 1890-1918 


It is in such ecstatic, long-breathed passages as these, often forming 
extended cadences on an enormous scale, that Strauss's debt to Wagner 
is most clear; and he was to write them to the end of his life, always for 
one, two, or even three soprano voices. Indeed a number of duets and a 
trio of this kind formed some of the climactic moments in his next opera, 
Der Rosenkavalier (1911). 

The collaboration with Hofmannsthal had led Strauss into a world 
of literary sophistication that he would hardly have entered otherwise, 
and it is to Hofmannsthal’s brilliant recreation of late eighteenth- 
century Viennese society that Der Rosenkavalier owed much of its 
success. This work is in essence a comedy, but so richly complicated 
by sentiment and spectacle that the comic element is largely overlaid. 
The Marschallin's levée in Act I and the presentation of the Silver Rose 
in Act II, on which the composer lavishes all his extraordinary gifts of 
harmonic and orchestral invention for their own sake rather than for 
specific dramatic effect, recall the great tableaux and divertissements of 
Meyerbeer's operas; and it is only in Act III that the comic element, 
represented by Baron Ochs, predominates unequivocally. In the 
Marschallin's music Strauss made himself the mouthpiece of that 
nostalgia that characterizes so much of the art of this period, in which 
Mahler's Sehnsucht nach der Kindheit was only exceeded by Rakhmani- 
nov's more radical and unqualified nostalgia—the Buddhistic ‘longing 
for non-existence’ (toska nebitiya) of Alexander Blok’s poems. The 
Marschallin’s nostalgia is ostensibly for her vanished youth, but her 
desire ‘to put the clocks back in the night’ is in fact a symbol of that 
deeper nostalgia for the past that had already played an important 
part in Hofmannsthal’s early lyric poetry. As early as 1893 he wrote 
in the Prologue to Artur Schnitzler’s Anatol: 


Hohe Gótter, Taxushecken, 

Wappen nimmermehr vergoldet, 

Sphinxe, durch das Dickicht schimmernd, 
Knarrend óffnen sich die Tore. 

Mit verschlagenen Kaskaden 

und verschlafenen Tritonen, 

Rokoko verstaubt und lieblich, 

бет... das Wien des Canaletto, 

Wien von siebzehn hundert sechzig. 


In his case this was not simply a longing for a vanished innocence and 
simplicity, such as we find in Mahler's symphonies and Humperdinck's 
operas, nor even an idealization of the past in the manner of Wagner's 
Meistersinger. It was rather a delight in what he felt to be the richness, 
variety, and vitality of an older society, including its inhumanity and 


ITALY AND THE NEW REALISM 153 


injustice. Baron Ochs represents that society's unregenerate face, the 
selfishness and ruthlessness of feudalism; and although Ochs is defeated, 
neither Hofmannsthal nor Strauss conceals his sympathy with the 
defeated. Beneath its smiling surface Der Rosenkavalier is in fact a 
‘reactionary’ work of art which owes much of its popularity to that 
nostalgia for ‘the good old days’ that since 1900 has been a powerful, if 
not always acknowledged, emotional force in the bourgeoisie of 
Western Europe. 

Again, as in Salome and Elektra, the fascination of Strauss’s score 
lies in the skilful contrasting of the familiar and the novel, an abundant 
diatonic melodiousness (the Marschallin’s soliloquy and the Italian 
Singer’s aria in Act I, Sophie’s music and Ochs’s waltzes in Act II and 
the trio and final duet in Act III) and the more complex, chromatic 
style of the intervening passages in which these form points of rest, 
exactly like the closed forms of the old opera, which are in fact here 
resurrected. In Strauss’s next collaboration with Hofmannsthal, 
Ariadne auf Naxos (1912), the disparate elements are presented with 
more sophistication, first by the scheme of presenting a play (or plays) 
within a play and then even more strikingly by giving a com- 
media dell'arte improvisation and an opera seria simultaneously. 
This device may perhaps be regarded as a type-figuring of the 
schizophrenia which marked much of the art of these years immediately 
before the First World War. Historically it is interesting that Strauss, 
who started his career in the theatre as an enthusiastic Wagnerian, 
should have returned in Der Rosenkavalier to the closed forms and 
even, in the Italian Singer's aria, to the bel canto of nineteenth-century 
Italian opera. In Ariadne auf Naxos his taste for the most mellifluous 
possible combinations (three soprano voices and harps in a D flat 
major trio) is even more marked ; and in Zerbinetta's aria he resurrected 
coloratura as an expression of character. This had not been attempted 
in Western Europe since Delibes's Lakmé (1883), with the possible 
exception of Catalani's La Wally (1892), where La Wally's “Сапгоп del 
Edelweiss’ adds a fantastic, unreal note to a character in other ways 
fiercely realistic. 


ITALY AND THE NEW REALISM 

Although realism might appear to be of all aesthetic attitudes the one 
least compatible with so highly artificial an art-form as opera, it was in 
fact the influence of the French realist, or naturaliste, school of writers— 
and above all Emile Zola, whose often brutal novels of contemporary 
life began to appear in the seventies—that gave rise to the movement 
which carried opera equally far from the historical or exotic world 


154 STAGE WORKS: 1890-1918 


popularized by Meyerbeer and his followers and the world of myth or 
legend which Wagner chose for his music dramas. Just as Zola himself 
had found inspiration for his new approach to the novel in Gustave 
Flaubert’s Madame Bovary (1857), во opera-composers of the next 
generation found for the first time since Verdi's La Traviata (1853) 
something like a window on contemporary life іп Bizet's Carmen 
(1875).* The extraordinary aristocratic finesse of Bizet's music and his 
instinctive distancing, or stylizing, of the popular, *vulgar', element to 
a great extent masked the real originality of Carmen. Unfortunately very 
few of the composers who were tempted to follow his example possessed 
either his purely musical gifts or that instinct, which can only be called 
classical, for the effortless conversion of raw vulgarities into the material 
for a serious work of art. 

Although in France there are realistic elements in Delibes's ГаКкте, 
and a collaboration between Zola himself and Alfred Bruneau (1857- 
1934)? began in 1891, it was in Italy that realism in opera first developed 
as a distinct movement, under the name of verismo. The Sicilian 
village-drama of Cavalleria rusticana (1890) is innocent of any attempt 
at distancing, and the composer Pietro Mascagni (1863-1945), was 
content with the same naive criteria of simple, sensuous, and passionate 
melody that would have been applied in real life by the characters of 
Giovanni Verga's story. Cavalleria rusticana may be regarded as an 
instinctive, unconscious Italian protest against the ‘mandarin’ music- 
drama, the infiltration of even the Italian operatic world by Wagnerian 
theories. Pagliacci (1892) is only slightly more sophisticated musically, 
and the sentimentality of popular melodrama is equally dominant in 
both works. The composer, Ruggiero Leoncavallo (1858-1919), had in 
fact personal experience of the life of a touring company, though on a 
different level from that shown in Pagliacci and nearer to the compara- 
tively polite demi-monde which he used as the background for his 
Хага (1900). In Mascagni's Iris (1898), realism is tempered not only 
by the exotic Japanese setting but by a strong note of poetic fantasy 
introduced by the librettist, Giovanni Illica. In the opening scene, for 
instance, a hidden chorus makes itself the mouthpiece of the rising sun 
with the words ‘Son Io! son Io la Vita! son la Beltà, la Luce e il 
Calor", and this reappears at the end of the work, when the heroine 
has jumped from a window into what proves to be the town's main 
drain, where she encounters the rag-pickers, symbols of social rejection 
whom we are to meet again in the masterpiece of French realistic opera, 
Louise (1900) by Gustave Charpentier (1860—1956). There is an interest- 
ing example of what was to become a common feature in art nouveau 

1 See Vol. IX. ? See below, p. 164. 


ITALY AND THE NEW REALISM 155 


music, not only opera: Ше wordless chorus singing bocca chiusa. Тһе 
geisha chorus opens Act II of /ris thus: 


Ex. 68 


Lentamento Р 
a bocca chiusa, accompagnandosi al suono di samisen e tam-tam delle altre guechas 


This interlarding of an often crudely realistic modern story with 
decorative, and fundamentally inorganic, symbolism became increas- 
ingly common and represents a bridge between veristic and mytho- 
logical or fairytale opera. 

Leoncavallo's La Bohéme (1897), based on Henri Murger's Vie de 
Bohéme, contained none of these unrealistic elements, but was unable 
to support the inevitable comparison with an opera with the same title 
produced a few months earlier by the composer who was to prove the 
only undoubted master among the Italian veristi, Giacomo Puccini 
(1858-1924). Puccini turned for his librettos to a very different class of 
French writer, and no vitalistic symbolism such as we find in Zola was 
to be found in Prévost, Murger, or Sardou. Apart from Verdi, the 
productions of whose last years towered above all other Italian music 
of the day, the strongest influence in Puccini’s musical formation was 
Jules Massenet (1842-1912), whose Manon (1884) prompted Puccini’s 
first mature work, Manon Lescaut (1893). To his librettist, Emilio 
Praga, he observed very truly that ‘Massenet may feel his subject with 
the powder and the minuets. I shall feel it with a despairing passion’? 
and this passione disperata is the keynote of Puccini’s music, a lyrical 
outpouring that puts into the shade the eighteenth-century French 
graces which had formed an important attraction in the older man’s 
work. In two of the three operas which followed—La Bohéme (1896), 
Tosca (1900),3 and Madama Butterfly (1904)—Puccini turned again to 


1 Similar choruses are to be found in Puccini’s Madama Butterfly and Rakhmaninov’s 
Francesca da Rimini, in the Sirénes in the third of Debussy’s Nocturnes, and in Delius 
passim, 

2 Quoted in Mosco Carner, Puccini (London, 1958), p. 57. 3 See pl. II. 


156 STAGE WORKS: 1890-1918 


French sources; but in each case his librettists, Giacomo Giacosa and | 
Luigi Шіса, faithfully presented him with versions of Murger and | 
Sardou that allowed full scope for the despair as well as the passion in 
the composer's personality. If this despairing quality in Puccini's 
lyricism had its immediate origin in his own personal psychology, it 
also corresponded to a more general mood in the public, a vague 
emotional unrest and presentiment of future disaster that found 
poignant expression in the almost funereal melodies, which have often 
escaped notice among Puccini’s innumerable expressions of erotic 
passion. In Act II, scene 5, of Tosca the Andante sostenuto, with 
muted violins playing on the G string, could be the funeral march not 
simply of Scarpia but of a whole civilization, like the slow movement of 
Elgar's Second Symphony written ten years later. 


Ex. 69 
Andante sostenuto 
Vn.1 & Vla. 


O НЕ 
у а [pee 


Бу Астыма еті [s [ee [n comm mn Fun ee 
a aa ЕЕ ЕЕ ЕЕ НЕ 


PLATE H 


7 m 


PUCCINI'S TOSCA, 1900 (see p. 156) 


A poster advertising the first performance 


ITALY AND THE NEW REALISM 157 


The frailty of Puccini's heroines and the fated character of their loves 
had a sentimental and sensual appeal for his listeners similar to that 
which Verdi's La Traviata had enjoyed half a century earlier. Only this 
appeal was now greatly enhanced by the obscure, and often unacknow- 
ledged, sense of moral and social insecurity, of enjoying the last un- 
certain splendours of an age of European civilization that was 
unmistakeably coming to an end. In this sense Puccini's relationship 
to Verdi is exactly paralleled by Strauss’s relationship to Wagner. 
In each case we can observe the characteristic marks of a decadence— 
the replacing of strength by violence, the exclusive concern with the 
subjective interests of the individual and the constant demand for 
emotional extremes, reflected in dynamic and other markings of 
expression formerly reserved for climactic moments. 

The elements of melodrama and sadism in Puccini's works are not 
without parallel in nineteenth-century opera, including Wagner; 
but they had never been so nakedly exposed, even exploited. The 
description of Mimi's physical and emotional distress in Act III of 
La Bohéme has a gloating quality, to which Puccini added a macabre 
note of banality by using a thinly disguised version of the fashionable 
‘hesitation’ waltz (Ex. 70). 

The lengthy and detailed exploitation of Cavaradossi’s torturing in 
Act II of Tosca, and its close linking with Scarpia’s physical desire for 


Ex. 70 


Sostenuto molto 
PPP 
Fi. 
Gh 
Vin. + 
Via. 
div. 


158 STAGE WORKS: 1890-1918 


ч/ 
San-gue son ros - se 


(A terrible cough racks her thin chest, already her hollow cheeks have a hectic 
flush) 


Tosca, is in its less glaring way quite as perverted as the child-princess's 
erotic apostrophizing of Jochanaan's severed head in Salome. For 
Madama Butterfly and La Fanciulla del West (1910) Puccini turned 
to the novelette; but his handling of the orchestra, at its most delicate 
and transparent in La Bohéme, grew increasingly adventurous, while 


THE MUSICAL LANGUAGE OF VERISMO 159 


his harmonic language was increasingly enriched by the absorption 
of the last of many French influences—the elements of Debussy's 
style to be found in whole-tone chords, sequences of block harmonies 
and atmospheric chains of fifths. The last and most powerful of his 
purely veristic operas, И Tabarro, appeared іп 1918 in the Trittico, 
with the convent opera Suor Angelica (women's voices only) and 
Gianni Schicchi, a Dante-based comedy the vicious bitterness of whose 
humour has not always been observed. In Turandot, which he left 
unfinished at his death in 1924, Puccini returned to a type of Italian 
grand opera which he had done more than any other single composer to 
supersede. The exotic magnificence of the spectacle is often matched 
by the music, especially in the central scene of the three riddles; and in 
Liü Puccini created the last of his fragile, fated heroines pitted against 
hopelessly unequal odds, here represented by the pathologically cold 
and revengeful Turandot, a sketch for whose portrait had already 
appeared in the Zia Principessa of Suor Angelica. 


THE MUSICAL LANGUAGE OF VERISMO 


Opera became an increasingly popular art during the nineteenth 
century in Italy, where Verdi's personal devotion to the cause of Italian 
liberation and unity even lent it on occasion a political colouring. Only a 
very small group of Italian musicians and intellectuals, of whom the 
most distinguished was Arrigo Boito, interested themselves in Wagner's 
music and theories. The chief foreign influences were Meyerbeer and 
Massenet, both of whom contributed to that progressive popularization 
of musical style and manner of presentation which followed naturally 
on the musical enfranchisement of a wider section of the public. In the 
operas which Verdi wrote for Paris, and later in Ама, the element of 
orchestral and scenic display, the ambitious use of the chorus in 
large quasi-symphonic tableaux, and a new dramatic style of vocal 
virtuosity reveal Meyerbeer's influence. Following Gounod, and in 
conscious reaction against Meyerbeer, Massenet developed a new, 
intimate type of opera in which scenic display is less important and the 
chief emphasis is on erotically poignant melody, short-breathed and 
presented with the greatest possible emotional impact. The touch of 
realism lent by comparative informality was considerably increased by 
the Italian veristi, whose melodies often show the unmistakably 
melodramatic character of Italian popular speech. Irregular phrase- 
lengths and frequent variations of metre are interspersed with dramatic 
pauses. Triplet figures lend an added urgency and, combined with 
stylized sobs, produce an almost physical effect on the unsophisticated 
listener, as in Tosca, Act III: 


160 STAGE WORKS: 1890-1918 


sostenuto e cresc. con slancio rit. (bursts into tears) 
тз Л 


ПРА В О О. - 
ең iS ее 2. M оч | 
)—H p m i ey созо шп. 

AUR 54 ый IS НЫШ iam [ecu col 


vi-ta, Тап-іо la vi-ta! 


(The hour is past and I am dying in despair! And never have I loved life so much) 


THE MUSICAL LANGUAGE OF VERISMO 161 


Like Diderot and Greuze in the 1750s, the veristi chose convention- 
ally heart-rending subjects, cultivating in their operas a similar 
larmoyant emotionalism aimed at the unsophisticated. It was not only 
erotic scenes that were presented in this perfervid way. In Act I of 
Andrea Chénier (1896) by Umberto Giordano (1867-1948), the hero 
Gérard indignantly commiserates with his old father who has worked 
for sixty years in serf-like conditions for arrogant aristocratic masters. 
The free declamation follows the more formal melody in the orchestra: 


162 STAGE WORKS: 1890-1918 


ETE Eo кыр == 
Ја for- za dei tuo ner - vi 
cu 1_ 

Е At "8 ЕЕ ЕЕ == 


- -Ф- 
ЗЕЕ ЕР е ЕЕ == 


(You have been a servant for sixty years, old man! and have given freely to your 
impudent, arrogant masters your loyalty, the sweat of your brow, and the strength 
of your muscles) 


The doubling of the melodic line in the orchestra, the pauses, the 
appoggiature or stylized sobs and the final unison are well calculated to 
give the scene an intensity that will stir even the dullest listener. Har- 
monic effects were often calculated in the same way to stun the listener, 
as in the opening chords of Тозса; while at the opposite extreme ап 
almost conversational lyrical style often takes on a caressing, childlike 
character, as in Mascagni's Iris (Act II), (Ex. 73). 

If block sequences proved the simplest and most effective weapon 
in the emotional arsenal of the veristi, Puccini in particular makes use 
of altered chords of all kinds in order to extract the last refinements of 
poignancy from a motif often simple in itself, as in Act IV of La Войёте: 


Ex. 73 і 
Larghetto mosso - con molta semplicità 
IRIS 


Vo-glio il giar-di-no ші - o! iovoglioil mio giar-di - по, 


THE MUSICAL LANGUAGE OF VERISMO 163 


delicato poco rit. 


a tempo 


(I want my garden with its surrounding hedge, and my white cottage) 


where the dying Mimi is represented by a mortally sick version of her 
original motif: 
Ex. 74 

Molto meno 


molto espress. 
f) C.A. & Vlas 


As early as 1892 Alfredo Catalani (1854-93) had used in La Wally 
more sophisticated and Wagnerian sequences in order to heighten 
the tension: 

Ex. 75 


co- me  s'a.dora in ciel, Ра- do - re. - те - i! 
stento col canto 


ГИР, с 5 е 
ot РГ a Ра 


(As adoration is known in heaven, so would I adore you) 


164 STAGE WORKS: 1890-1918 


But verismo in Italy was only rarely to follow this path, which was 
subsequently adopted by veristic composers in Germany and Austria. 

As we have seen, French subjects were particularly favoured by the 
Italian veristi, and Prévost, Murger, and Sardou had furnished material 
for three of Puccini's most successful operas. He was to turn to France 
again for the first of the three operas in his Trittico (1918), and Л 
Tabarro may be counted as the last masterpiece of the movement which 
had spent its vital force by 1910. It was not only Puccini who had fallen 
under Massenet's spell. The whole Act III of Leoncavallo's Хага 
(1900), which takes place in a Paris drawing-room, is redolent of 
Massenet, and Milio's elegiac reflections clearly echo des Grieux's 
dream in Act II of Manon. 


FRANCE AND THE REACTION AGAINST REALISM 


Meanwhile Massenet himself repaid the compliment implicit in such 
imitations when he wrote La Navarraise (1894), which might appear to 
be a parody of an Italian veristic opera in its violence and brutality. 
Far more characteristic of Massenet's real musical character, and a 
work of delicate charm and sensibility, was his Werther (1892), in which 
domestic and genre scenes are skilfully used to diversify the unhappy 
love-story in a manner quite foreign to the more simple-minded Italian 
veristi. Sophie's Че gai soleil’, in particular, which distinguishes her 
from her sister Charlotte in exactly the same way as Strauss was to 
distinguish Chrysothemis from Elektra, is one of the last and finest 
examples of a characterization which goes back to the old opéra comique. 
(Soeur Constance in Poulenc’s Dialogues les Carmélites (1957) is 
perhaps the latest avatar of this traditionally French type of character.) 
Thais (1894) is yet another working of the popular theme of the prosti- 
tute redeemed by love, and the ballet, in which a ballroom waltz is 
danced by La Perdition, even harks back to Meyerbeer's Robert le 
Diable. More in harmony with the new tastes that Massenet himself had 
done much to create was Sapho (1897), based on a story of contem- 
porary life by Alphonse Daudet, while in Grisélidis (1901) and Le 
Jongleur de Notre Dame (1902) Massenet moved into the field of 
medieval legend without abandoning his own personal lyrical style. 

If operatic realism was really foreign to Massenet, it was the whole 
inspiration of Alfred Bruneau (1857-1934), who chose Zola's stories 
for five operas written between 1891 and 1905. In Г’ Attaque du moulin, 
which is the most strictly realistic in character, Bruneau used one of 
the Soirées de Médan dealing with an incident in the Franco-Prussian 
war. Messidor (1897) and L’Ouragan (1901), on the other hand, contain 


FRANCE AND THE REACTION AGAINST REALISM 165 


strong veins of symbolism. In Messidor the contrast between the evil 
power of Gold and the beneficent Water dominates, while in L’Ouragan 
weather and scenery are used to symbolize the passions of the characters. 
Bruneau's music, though unmistakably French and considerably in- 
indebted to Massenet, also shows Wagnerian influence in the orchestra- 
tion. The third act of Messidor is an interesting example of the 
apparently incompatible elements that Bruneau combined in his operas. 
The first scene is entitled ‘Га Légende d'Or' and is a full-length ballet, 
full of symbolical figures including the Madonna and Child, set in a 
vast cathedral-like cave. On the other hand the second scene shows a 
newly installed machine for washing the gold brought down from the 
mountains by the river. As the curtain rises, the large wheel of the 
machine is shown in motion. Bruneau plainly turned to Wagner in his 
music for the opening of the scene, and the epic of man's mastering 
natural forces by mechanical power suggested to him a parallel with the 
primitive forces of the Valkyries: 


Ex. 76 
Animé modérément 


Dm 5 р куша 
------за зенита про pelis —— parili 
га 1] Шелли eo a aS as Be ea q ПЕЕ y EE 
aoe 22 FLAG VI —— eet Pg 
ie a E і Бе 


wep — —g- Pei poe de m — rd 
iS ——4 па uem ЕН ни. 
DPI Fou 


De- 
= SSS aa ae ини ЕЕ 
OF ӘН ТІР (ішсе! ee шетен | Беттен ee а UB ШЕ ТЕ 27 Жа шана на 
шаһ ——M—À ——À А 543-2 оГ A ПИВО клен T aae 
WD Бар -————— [LE um asuma 


166 STAGE WORKS: 1890-1918 


It is characteristic of French opera in general, and of Bruneau in 
particular, that the final scene of Messidor shows a return to the 
conventions of Latin rustic life. After the villain, Mathias, has admitted 
the murder of Véronique's husband and committed suicide by jumping 
from a rock, the smiling crops that have sprung up under the beneficent 
influence of the Water are solemnly blessed, in the traditional Latin 
liturgical formulae, by the village priest. This suggests a link with Gou- 
nod rather than with Massenet, 

The same is true of the most remarkable of all French realistic 
operas, Charpentier's Louise (1900). Here the banal story of a Parisian 
working-girl's love-affair and her escape from parental authority is 
set against a romantic vision of the street and factory-life of Paris, 
envisaged as the Ville-Lumiére and itself a product of that art nouveau 
that was subtly modifying the attitude of French artists to their 
materials. Charpentier's musical invention is undistinguished and his 
melodic ideas have a facile sentimental appeal (and even a preference 
for triplet-phrases, as in Louise's ‘Depuis le jour’) similar to that of the 
lesser Italian veristi. The very naiveté and plasticity of his style, however, 
the absence of all academic stiffness and the suggestion of mysterious 
vital forces present beneath the commonplace surface of everyday life 
were novel. 

These features were to find expression of a very different kind in the 
Pelléas et Mélisande (1902) of Claude Debussy (1862-1918), where the 
reaction against realism appears at its most marked. Debussy had 
already shown in his cantata La Damoiselle élue (1888) his sympathy 
with the ideals of the English Pre-Raphaelites, which were at the root of 
art nouveau\—the freely flowing, plant-like arabesque, the sinuous and 
mysterious femininity whose misleadingly ‘chaste’ contours and anae- 
mic complexion barely conceal a sophisticated sensuality, and the hushed 
palette and matt surfaces that give an equally misleading impression of 
monastic tranquillity. In Maeterlinck’s poem Debussy found a perfect 
literary expression of these features, and in addition a scale of moral 
values dominated by compassion in the face of human suffering and 
the acknowledgement of the mystery of human existence, a rejection 
of the too easy optimism of scientific materialism and of Nietzsche’s 
philosophy of power. In retreat from, yet still deeply marked by his 
early Wagnerian fervours, Debussy had come upon a score of Mus- 
sorgsky’s Boris Godunov which had suggested to him rhythms, phrase- 
shapes, and harmonies useful in his search for a new dramatic 
language and a new musical prosody for the French language. A 
systematic use of the whole-tone scale and its resulting harmonies 


1 See below р. 183. 


FRANCE AND THE REACTION AGAINST REALISM 167 


hitherto used only in isolated passages for dramatic or decorative 
purposes by Liszt and his Russian followers, enabled Debussy to avoid 
the emphatic statements and rhetorical cadences conventionally 
associated with the opera. The echo of ecclesiastical modes heard in the 
very opening bars seems to set the action of Pelléas outside chrono- 
logical time and topographical space; and it is followed immediately 
by the pliable, floating motif that suggests the passive movement of 
weed in a gently flowing stream—an image of human powerlessness 
against the flow of events. 


Ex. 77 


Trés modéré Ob., С.А. б 


22 
[ugue — — mu] 
[ee =. 


Debussy's views on dramaturgy and the part to be played in the drama 
by the orchestra are reflected in a conversation, reported by Jean 
Cocteau, where Debussy quoted with approval the advice of Erik Satie. 
‘There is no need for the orchestra to grimace when a character comes 
on to the stage. Do the trees in the scenery grimace? What we have to 
do is to create a musical scenery, a musical atmosphere in which the 
characters of the drama move and talk.”+ 

In spite of many echoes of Wagner (particularly Parsifal) in Debussy's 
harmony, and frequent almost verbatim quotations of Mussorgsky's 
Boris Godunov (mostly Pimen's music)? Debussy's score is wholly 
original in conception and texture. The handling of the unaccented 


1 Quoted in Rollo Myers, Erik Satie (London, 1948), p. 32. 

? For Parsifal see e.g. the Lent section in the orchestral interlude between Act IT, scenes 1 
and 2. For Boris Godunov see Act II, scene 2, the orchestral introduction to Act IV, and 
Act III, scene 2. Debussy's two-bar structures are also of Russian origin. 


168 STAGE WORKS: 1890-1918 


French language, to which the emphatic accentuation proper to both 
German and Italian is completely alien, is faithful throughout to the 
muted understatements and elusiveness of Maeterlinck's text. The 
breathless simplicity of the lovers' declaration is characteristic and 
unique in operatic literature. Having set the scene, the orchestra is 
suddenly silent: 


Ex. 78 32-1 
P EE ms 
Ob. eux ш ла ве та м 
CA Hi e Ef go eR. | 
(ОЁ ”_ m Ў П —— 
ате с сш] 
PELLEAS 
Е Гар SS, SS в т = 
р г р р Г 7 лас ee «РЕ 
: Ти ne sais pas pour-quoi il faut que je т - 


Strings 


- loi -gne.. Tu пе sais pas que c'est par-ce que 


FRANCE AND THE REACTION AGAINST REALISM 169 


MÉLISANDE PELLÉAS 


PELLEAS 


ES 
КА a EE. ar И" er ee ааа Ж A |} 
m ре в НЯ SE Sa S Бел / аты (И ee 
Е == паа) гр o 

8 


e Рф 
Е — 1 14-і 


Те t'ai-me. Je t'aimeaus - si. Oh! qu'as tu dit, Mé-lis- an-de! 
[Orchestra tacet] 


je ne l'ai presque pas en-ten-du!___. 
(P: You don't know why I have to go away . . . you don't know that it is 
because . . . I love you. M: I love you too. P: Oh! what did you say, Mélisande? 
I could hardly hear!) 


Although pastel colours and the pliable arabesques of art nouveau play a 
large part in the score, Golaud's explosions of anger are anything but 
muted and the scene (Act ІП, scene 4) in which he sets the child Yniold 
to spy on the lovers is in its way as terrible a depiction of ungovernable 
jealousy as anything in Verdi's Otello, and represents a heightened form 
of realism beside which the naivetés of Italian verismo seem pale indeed. 

With Pelléas opera moved for the first time unequivocally outside the 
conventions of the nineteenth-century opera-house. Debussy's new 
conception of opera clearly influenced Paul Dukas (1865-1935) in 
another setting of a Maeterlinck play, Ariane et Barbe-Bleue (1907), in 
which choral, descriptive, and even purely decorative scenes represent 
links with the past. Themes are treated symphonically and cyclically, 
and thematic material shaped into a grandiose coda. These musical 
procedures are largely hidden from the ordinary listener by Dukas's 
orchestration, which often has the amplitude and much of the brilliance 
of Strauss's. A less accomplished example of the style is to be found in 
Macbeth (1910), the single opera written by Ernest Bloch (1880-1959) 
where Lady Macbeth in the sleep-walking scene improbably echoes 
Mélisande (Ex. 79). 

During the same years as the realistic operas of Bruneau and Char- 
pentier and the innovatory Pelléas there appeared the last monuments 
of the Wagnerian enthusiasm which had made so deep a mark on 
French music in the eighties. Something has already been said of 
Vincent d'Indy's Parsifal-like Fervaal (1897) and Ernest Chausson's 
posthumously performed Le Roi Arthus (1903), which witnesses to that 
obsession with Tristan traceable as far back as Duparc's earlier songs 
(‘Soupir’, 1868). In the same year as Le Roi Arthus, Vincent d'Indy 
produced L’Ftranger. This is a very free reworking by the composer of 
the Flying Dutchman story. The mysterious Stranger, whose success as a 
fisherman and evangelical behaviour cause havoc in a small French 


170 STAGE WORKS: 1890-1918 


LADY MACBETH en retenant 


un soldat! avoir peur! 


(Fie, my lord, fie! A soldier and afraid!) 


fishing-village, seeks redemption from an unspecified guilt in the love of 
the symbolically-named Vita. She already has a pretender to her hand, 
a dashing excise-man very unlike Wagner’s Erik. Torn between her 
feelings for the two men Vita, like Senta, sacrifices her life when she 
shares the Stranger's attempt to rescue a ship in distress. D'Indy makes 
effective use of folk-song choruses, and a traditional opéra comique 
style of characterization for Vita’s mother; but the music of the Stranger 
and his scenes with Vita have a Wagnerian harmonic density and 
flavour. In bar 4 of the following passage from Act II the Franck 
fingerprint is unmistakable: 


Ex. 80 


Spe = чыш; 
НЕ с ІЗ ee s 


RN SETS 
и aec — a 


FRANCE AND THE REACTION AGAINST REALISM 171 


(Forgive me the rash words that escaped me yesterday. Forgive me, tell me that 
you forgive me and that I can go away absolved) 


It is interesting to find Debussy, in his article for СИ Blas on the first 
performance of L'Étranger (Brussels, Theátre de la Monnaie) observing 
that ‘this work is an admirable lesson for those who believe in the crude 
imported style which consists in crushing music under cartloads of 
realism." 

André Messager (1853-1929), whom Debussy had chosen to conduct 
the first performance of Pelléas, showed an excellent craftsmanship 
and a delightful lyrical gift in a series of operettas which included 
Madame Chrysanthéme (1893), Les p'tites Michu (1897), and Véronique 
(1898). The only other country in which light opera was treated with 
comparable skill was Austria, where Franz Léhar (1870-1948) produced 
in Die lustige Witwe (The Merry Widow) (1905) a work which has 
outlived all but a few of the serious post-Wagnerian or veristic operas 
of the day. Certainly the final numbers of Massenet's long portrait- 
gallery of women, which ended only with his death in 1912, contained 


1 Reprinted in Monsieur Croche the Dilettante- Hater (English ed., London, 1927). 


172 STAGE WORKS: 1890-1918 


nothing more distinguished than the.slight Thérèse (1907); and the 
advent of Sergey Dyagilev's Ballet russe to Paris in 1909 had the effect 
of distracting the attention of the most gifted composers away from the 
lyrical drama. Apart from the musically interesting but dramatically 
weak Pénélope (1912) by Gabriel Fauré (1845-1924) the only opera of 
distinction produced in France between 1908 and 1918 was L'Heure 
espagnole (1911) by Maurice Ravel (1875-1934). Іп this, a deliberately 
frivolous subject, which might seem to belong to the operetta rather 
than to the opera, is handled with a care and skill that at first seem 
disproportionate. Ravel, however, was seeking for a solution of the 
problem facing all French composers of the day: that of avoiding the 
shadows of both Wagner and Debussy and presenting the public with a 
work that made full allowance for the well-developed French sense of 
the ridiculous yet still allowed a place for fine musical craftsmanship. 

Ravel's characters are no more than exquisitely fashioned puppets, 
but they express their conventional sentiments in a perfectly calculated 
and highly polished language, in which Spanish rhythms and melodic 
phrases are presented with French gracefulness and a strong sense of 
parody. The plot has the symmetry of a mouvement de ballet, and Ravel 
and his librettist Franc-Nohain emphasize this fragile artificiality in the 
final quintet which, like the vaudeville in the old opéra comique, points 
the moral. Here Ravel harks back to the clowning of Offenbach's 
Belle Hélène, with his take-off of vocal fioriture. 


Ex. 81 Presse beaucoup rubato 


CONCEPCION 


GONZALVE 5 4— ha Po Fi 
'TORQUEMADA 
RAMIRO 
INIG o: fg mmm 


Orchestra 


173 


FRANCE AND THE REACTION AGAINST REALISM 


se СШ | В 
Ae a МИ 


(Ah! The muleteer has his turn!) 


3 


174 STAGE WORKS: 1890-1918 


Ravel did not write another opera until 1925, when he collaborated 
with Colette in a fantaisie lyrique, L'Enfant et les sortiléges, which is as 
much ballet as opera. Here the divertissements provided by animals and 
inanimate objects far outweigh the slender central theme in importance. 
This was in effect a return to the ideals of the French eighteenth century, 
which had already prompted the most important of the works which 
Ravel wrote for the Ballets russes. Before examining these ballets it will 
be as well to trace the history of opera in Russia itself during the last 
years before it was largely replaced by ballet. 


RUSSIA AND THE SLAVONIC PERIPHERY 

The year 1890 witnessed the production in St. Petersburg of two dif- 
ferent flawed masterpieces, Tchaikovsky's Pikovaya Dama (The Queen of 
Spades) and Borodin's posthumous Knyaz Igor (Prince Igor).1 Tchaikov- 
sky was to write only one more opera before his death in 1893, the 
comparatively feeble fairytale Jolanta (1892), and it was fairytale operas 
that were to predominate among the dozen operas that Nikolay 
Rimsky-Korsakov (1844-1908) was to write during the last fifteen 
years of his life. Already in his Snegurochka (Snowmaiden) (1882) 
there had been a marked accentuation of the fantastic and spectacular 
elements at the expense of character-delineation and human interest in 
general; and this emphasis, all but complete in Mlada (1892), was very 
marked in Noch pered Rozhdestvom (Christmas Eve) (1895) and the 
brilliant Sadko (1898). The decorative scenes and divertissements 
showing the bustling commercial life of Novgorod and the submarine 
kingdom to which Sadko is carried off by the Sea King’s daughter 
quite outweigh the rest of the work in musical interest. Rimsky- 
Korsakov’s preoccupation with fantasy and spectacle, and the banality 
of the love-music in his operas, no doubt reflect a certain lack of 
humanity in his character; but this is counterbalanced by the skill and 
inventiveness displayed in his cold and glittering instrumental combina- 
tions, as in the scene in Sadko where the fish are turned into ingots of 
gold. 

In Mozart i Salieri, which was produced in the same year as Sadko 
(1898), Rimsky-Korsakov imitated Dargomizhsky! in setting a complete 
Pushkin text, but used a pastiche ‘classical’ style in every way different 
from his own and Dargomizhsky’s. Human interest is strong in Tsar- 
skaya Nevesta (The Tsar’s Bride) (1899) and Servilia (1902) but these 
are feeble works compared with the four fairy-stories or legends 
Skazka o Tsare Saltane (The Tale of Tsar Saltan) (1900) Kashchey 


1 бее Vol. IX. 


RUSSIA AND THE SLAVONIC PERIPHERY 175 


Bessmertny (Immortal Kashchey) (1902), Skazanie o nevidimom grade 
Kitezhe i deve Fevronii (The Legend of the Invisible City of Kitezh and 
the Maiden Fevronia) (1907) and Zolotoy Petushok (The Golden 
Cockerel) (1909). 

The first two of these are fantasies, fairy-stories from the Russian 
past that provide the composer with scenes in which he can simply 
exploit to the full his colouristic imagination. In Kitezh and The Golden 
Cockerel, on the other hand, the fairy-story is used as a parable for the 
times, the deeply troubled years of the Russo-Japanese War and the 
1905 revolution. Kitezh reflects the ferment in Russian religious апа 
philosophical thought during these years, when Vladimir Solovyev 
was developing the final stages of his new ideas of Christian mysticism 
and Leo Tolstoy a new version of evangelical simplicity, while theo- 
sophical and anthroposophical speculations were under the influence of 
Rudolf Steiner. Although Rimsky-Korsakov himself was a typical 
nineteenth-century rationalist, he felt the aesthetic attraction of the 
legend, medieval and originally Platonic, of a Holy City conceived in 
terms of Orthodox liturgy and symbolism. His librettist, V. I. Belsky, 
based his text on a thirteenth-century legend and his language is 
deliberately archaic, influenced by that of Wagner’s librettos. The 
composer drew not only on the traditional songs and chants which had 
provided much of the material of his earlier operas, but on Wagnerian 
models. For instance, the opening scene, which shows the maiden 
Fevronia alone in the depth of the forest in summer, recalls the Forest 
Murmurs of Wagner’s Siegfried; and in Act IV there are clear 
reminiscences of the Flower Maidens in Parsifal. The figure of Grishka 
Kuterma, who betrays the city of Kitezh to the Tatars and is never- 
theless forgiven and redeemed by Fevronia, is a cross between the 
holy simpleton of Russian tradition and the tortured psychological 
criminals of Dostoevsky. Throughout the opera the conflation of 
Orthodox ideals and imagery with the pantheistic nature-worship of 
the pre-Christian Slavs is reflected in the music, which alternates 
between a stiff, archaic and fundamentally modal idiom and Wagnerian 
colour and flexibility. There are passages in Kitezh where Rimsky- 
Korsakov denies himself his customary luxury of orchestration and 
adopts, in the interests of the drama, a much more austere manner that 
recalls Mussorgsky. In Act III, scene 2, where Grishka Kuterma’s guilt 
is beginning to send him out of his mind, the insistent expanding 
figure representing the bells of Kitezh ringing in his head is heard 
echoing in the lowest reaches of the orchestra, a psychological trans- 
formation that is very telling. 


176 STAGE WORKS: 1890-1918 


Ц [ | Ц 

ПА я P 4 P m. а 

НАЕ кеш кашкасы E СЫС о 
е л. савана ава реа 


Strings and muted Horns 


e 


(No escape from hellish torments—no life for me on God's fair earth! I shall cast 
myself headlong into the depths) 


Pushkin’s The Golden Cockerel? is a fable which shows a stupid, ease- 
loving king deaf to warnings of imminent disaster and unwilling to 
keep his promises, engaged with a mysterious Eastern enemy with 
whose queen he falls in love. The parallel with the political situation in 
Russia at the time of the war with Japan was close enough for the 
censor to raise obstacles to the performance of the work; but this last 
of Rimsky-Korsakov's fourteen operas is the finest example of his 
powers both as dramatist and orchestrator. The clever musical con- 
trasting of the solid King Dodon and his court with the mysterious trio 
of the Astrologer, the Golden Cockerel, and the Queen of Shemakha 
shows clearly his unusual ability to create atmosphere rather, tha 
human character by musical means. His use of coloratura, as in the 
shimmering roulades of the Queen of Shemakha, anticipates the 
Zerbinetta of Strauss's Ariadne auf Naxos. 


1 See pl. Ш. 


PLATE IH 


шая 71 Ха pousisap uononpoud [eursrio 34} jo IYL 12у JO 198 28015 әчү 


(941 “4 225) 6061 “ТЯЧЯЯООЭ NHUG'IOD FHL 8,.ЛОМУБНОЯ-А ЯБИТА 


LIE, 


BOHEMIA 177 


The Golden Cockerei was later performed in a ballet version by 
Dyagilev's Ballet russe, and in Rimsky-Korsakov's earlier. Kashchey 
the Immortal (1902) the element of ballet was already strong. Every- 
thing in this ‘autumn fairy-story', which is in fact another winter-spring 
myth like the earlier Snegurochka, is magic. The Stormwind suggests a 
dancer rather than a singer, and the Princess Kashcheevna is character- 
ized as much by the orchestra and by her movements as by her actual 
singing. It was, in fact, with a ballet based on another version of the 
same story that Rimsky-Korsakov's most gifted pupil, Igor Stravinsky 
(1882-1971), was to make his name. 

The principles of realism, which had found an early expression in 
parts of Mussorgsky's Boris Godunov and in Serov's The Power of Evil 
during the 1870s, hardly interested Russian composers of the nineties. 
Sergey Rakhmaninov (1873-1943) was influenced by Mascagni's 
example in his A/eko (1893); but he showed his real dramatic powers 
much later in the more ambitious setting of Pushkin's Skupoy Ritsar 
(The Miserly Knight), and an awareness of contemporary fashion in 
the bouche fermée choruses of the more feeble Francesca da Rimini, 
both of which appeared in 1906. 


BOHEMIA 

Russian music and literature played an increasingly important role 
in the artistic developments of Bohemia as dissatisfaction with Austrian 
rule and the centralizing policy of Vienna grew more acute. Dvorak 
had chosen a Russian subject for his Dimitrij as early as 1881, but the 
influence of Wagner is paramount in the most important of his later 
operas, Cert a Káca (The Devil and Kate, 1899) and Rusalka (1901), 
a fairy-story which had inspired Dargomizhsky. Wagner and Smetana 
largely determine the style and character of Sárka (1897) one of the 
handful of stage-works by Zdenék Fibich (1850-1900), which include an 
historically interesting trilogy in which the composer attempted to 
revive the melodrama. An earlier Bohemian composer, ЛЇЇ Benda 
(1722-95), had experimented with some success in setting a spoken text 
against orchestral accompaniment and comment;! but Fibich's musical 
language, in which the already old-fashioned language of Schumann 
is only superficially modernized by a Wagnerian leitmotive technique, 
was not strong enough to bear the weight of the drama in his trilogy of 
heroic Greek legends based on the character of Hippodamia—Námluvy 
Pelopovy (Pelops’ Wooing) (1889), Smir Tantaláv (Tantalus's Atone- 
ment (1890) and Smrt Hippodamie (Hippodamia's Death) (1891). The 
following excerpt from Námluvy Pelopovy, Act IV, scene 2, is typical: 

! See Vol. VII, pp. 76-9. 


1890-1918 


STAGE WORKS: 


178 


Ex. 83 


ve 


Chci přímou cestou 


PEL. Neb to 
nemusíš.! byl Myrtillos! 


ít 


Hippopamia Daleko j 


hledám si! 


toho dosáhnout, já vy - 


ku jeho 


zámyslem. PEL. Kdo ru 


we =... 


узаК nikdy 


Ніррор. Svou rukou jen, 


------ 


kdo tebe 


æ æ a ne wee 


vedl zlotřilou? 
Quasi Adagio 


BOHEMIA 179 


|] 1 
kdo у očích v tvoje patřé à -!v srdci zápalil se blahosti, kdo nemoh’ jinak, néz té! 
smévem a milovat.! 

CES Д 


! 
| p€—M——— a 
! pe———MÓÓ—— 1 


(Pelops: Y mean to plumb the depths of this hideous treachery, leave nothing 
unexplored until I find the traitor. Hippod: You will not have far to seek. P: Then 
it is Myrtillos! H: Only his hand, not his the design. P: Then who guided that 
infamous hand? H: One, Pelops, who loved you from the moment you crossed our 
threshold, with a love more burning than the sun in the heavens. One who looked 
smiling into your eyes, one whose heart blazed in ecstasy and could not but 
love you.) 


The only opera of outstanding quality and originality to appear in 
Bohemia (or rather Moravia) at this time passed unnoticed by the 
outside world, and had to wait twelve years for its first performance in 
Prague. This was Лей pastorkyňa (Her Step-daughter) (1904) by Leos 
Janáček (1854-1928). The story by Gabriela Preissová is one of peasant 
life, involving primitive passions as violent, in a setting as realistic, as 
those of Mascagni's Cavalleria rusticana. In fact this opera, which made 
its reputation in the outside world under the title of Jenufa, may be 
considered as the supreme masterpiece of verismo although the com- 
poser insisted on 'naturism', rather than naturalism, as his guiding 
principle. In this, as in his use of a prose libretto, he followed the 
example of Alfred Bruneau. 

Janácek's musical style is naturalistic in the sense that his vocal line 
was modelled on the rhythms of Moravian peasant speech, just as 
Mussorgsky had shaped his phrases and rhythms on those of the 
Russian language. Unrealistic elements, however, appear in the many 
repetitions of words or single phrases, sometimes emphasizing a single 


180 STAGE WORKS: 1890-1918 


idea but often modifying it slightly; and in Janácek's handling of the 
denouement where the characteristically veristic horror and despair are 
replaced by the catharsis brought about by the Kostelnicka's confession 
of her crime. Janácek's use of cross-rhythms and repetition, for intensi- 
fying and slightly modifying a phrase, often has a quasi-hypnotic effect: 


Ех. 84 Act I, scene 1 


ТНЕ GRANDMOTHER 


3 
И boty = as Lum га ест new] 
(дуст = ^ ж: Hf ро 
WU [mx BÉ eet Cl Gd == mns eir [io mm Vl ma 


Co to máš za ra-dost! Со to máš, dev-Ci-co, za ra-dost! 


ua ete Eee Rin 


ЕТ ЕЕ Е: En 


Ва- те nu jsi na u- ci la ta ке а 


et ot et et ot ef тата: E LL, 
psc Е am = | eFerer er ЕЕ НЕ EE 


BOHEMIA 181 


> PP 


Muzs- ky  ro-zum máš 


(What joy that gives you, my girl, what joy! You taught even the servant-girl to 
read! You have a man's intelligence) 


Or a vocal line unaccompanied but echoed, or interrupted, by the 
orchestra may be given the solemnity of a primitive spell: 


Ex. 85 Act II, scene 1 


Meno mosso А 
KOSTELNICKA 


ОН 
И | Ве ERR 
AG ate Е 
El Cac ten - 173 E) елү ааа 
GS ame Y o Ер peret] 


182 STAGE WORKS: 1890-1918 


pli-zna-la зе svým po-kles-kem. _ 


do hro-bu spro-vo-dit, 


(From the day that I brought you home I noticed your grieving and your misery, 
and when you confessed your guilt I thought that it would be the end of me) 


AFTER VERISMO: ART NOUVEAU AND THE CLASSICAL REACTION 
Wagner’s legacy, though most fruitfully invested by the composers 
of Central Europe, had certainly enriched opera in both France and 
Russia. Realism, too, though primarily an Italian movement in opera, 
proved a fertilizing agent well beyond the borders of Italy, first in 
France and then in Central Europe, where it was combined with 
Wagnerian elements in a number of works which caught the imagination 
of the day, though they proved to possess small lasting power. A third 


AFTER VERISMO: ART NOUVEAU & CLASSICAL REACTION 183 


element which played an important part in modifying both neo- 
Wagnerian and realistic opera has already been mentioned on several 
occasions: the art nouveau, or Jugendstil, movement which originated 
in the visual arts. It was a minority movement, one of the many protests 
against the accepted presuppositions of artistic academies and so, in 
the last resort, against the industrial society of nineteenth-century 
Western Europe. The roots of the movement can be easily traced, 
through the Aesthetic Movement of the eighties in England, back to the 
Pre-Raphaelite painters of the fifties; and they in their turn were 
indebted to the German school of *Nazarene' painters, who owed their 
origin to the secessionist Guild of St. Luke founded in Vienna in 1809 
by Overbeck. The rejection of the Renaissance image of man and the 
return to an imagined Middle Ages was accompanied by a new cult of 
nature, a conscious turning away from the literal representation of 
human life at its fullest and most magnificent to a study of the forms 
and textures of inanimate nature. Man's place in the new art was 
humbler and more mysterious, no longer that of ‘the lord of creation’ 
but suggesting rather an order of being shared with plant and animal 
life and equally at the mercy of destructive forces. During the eighties 
this new attitude found expression in an enthusiasm for Japanese art, 
especially prints and pottery, and in the cultivation of the arabesque or 
stylized plant-forms of oriental art; and it soon spread to feminine 
fashions, which dictated clothes moulded naturally to the figure instead 
of the exaggerations and stylizations of the crinoline and the bustle. 
Floating veils and gauzes, clinging stuffs and floral designs such as 
those elaborated by William Morris replaced stiff brocades and velvets, 
and the bourgeois ideal of opulent display yielded to a subtler and more 
poetic conception of feminine beauty. 

Art nouveau elements in the opera go back as far as Wagner's Rhine- 
maidens and Flowermaidens, and oriental and flower motifs often 
modify the realism of Italian opera in the 1890s (e.g. Puccini's Madama 
Butterfly and Mascagni's Iris and L'Amico Fritz). The sense of human 
fragility, the mysteriousness of existence and an overriding sense of 
pity, which characterize the writing of Maurice Maeterlinck, found 
musical expression in the operas based on his works, and supremely 
in Debussy's Pelléas et Mélisande. In Charpentier's Louise and in 
Dukas's treatment of Maeterlinck's Ariane et Barbe-Bleue there is 
more than an echo of the contemporary struggle to obtain for women a 
position in society if not equal to that of men at least superior to that of 
minors. Outside the Latin countries an amalgam of Wagnerian (or 
later Straussian) harmony and orchestration with typically veristic 
Italian melody is to be found in the operas of Eugen d'Albert (1864— 


184 STAGE WORKS: 1890-1918 


1932) and, in a more personal idiom, Frederick Delius (1862-1934) and 
Franz Schreker (1878-1934). D'Albert's Tiefland (1903) is a skilful 
and effective score almost equally indebted to Puccini and Wagner, and 
the story of violent passions in a Pyrenean village is comparable to that 
of Janaéek’s almost exactly contemporary Лей pastorkyňa (‘Jenufa’). 
Art nouveau elements are much stronger in Delius. His first opera, 
Koanga (1904), is set in Florida and the heavily chromatic post-Wagner- 
ian style of the white planters’ music is happily relieved by the slave 
choruses, whose old-fashioned ‘nigger-minstrel’ style suggests an 
acquaintance with the ballads of Alfred Scott-Gatty. Romeo und Julia 
auf dem Dorfe (А Village Romeo and Juliet) (1907), based on a story 
from Gottfried Keller's Die Leute von Seldwyla, presents a pair of 
child lovers defying the conventions of a small-minded, property- 
conscious society and ends in the Liebestod of a double suicide. The 
Black Fiddler is a personification of music as a sinister power, and 
Delius's melting chromatic harmonies spread a very characteristic art 
nouveau veil over melodic lines which are often of extreme simplicity. 


Ex. 86 
Der schwarze Geiger (in der Ferne) 
The Dark Fiddler (in the distance) 


ein - за - mer Sän-ger, du ei - lest da-hin, durch das 


О pi - per un -fol -lowed, how thou flin-gest on thro the 
Гал 


Dik - kicht der Wàl - der, durch Kraut und Ge-strüpp. Ich 
tan - gle of trees and the wran- gle of shrubs while 


AFTER VERISMO: ART NOUVEAU & CLASSICAL REACTION 185 


ei - le dir nach, dein ver - lass-en - ег Gei-ger: 
I must limp af - ter thy — fidd.-ler for - за - ken, 


sind ja doch Brii-der, du 
are we not com-rades, o 


In Schreker’s Der ferne Klang (The Distant Sound) (1912) the heavy 
unisons and fortissimo sequences of unrelated triads recall verismo, the 
thick texture and opulent orchestration are borrowed from Strauss, and 
much of the harmony shows the superficial influence of Debussy. The 
realistic opening scenes of Act I gradually take on a symbolical, unreal 
character; and Act II takes place in the Casa delle Maschere, a fantastic 
brothel on an island in the Venetian lagoon, which is a thinly disguised 
modernization of Klingsor’s garden in Act II of Parsifal. In this act 
the contrasting of gipsy music on the stage in close antiphony with the 
orchestra (a device not lost on Alban Berg, who made the piano- 
reduction of Schreker’s score) alternates with passages of character- 
istically overblown late romantic harmony, like Grete’s dream. 


186 STAGE WORKS: 1890-1918 


ren, dünkt 


Gipsy music 


D 6 L) 
ЕР 1-1 Қ с ==. | o— " 
; ERI 
——— ШУ ЕЕ 
эфирш ПРЕД Ва ПИН и ната poc 
=== А = 


Е 
— 


Appassionato rubato, sehr rhythmisch, scharf, doch die 
Viertel gleich lang. 
(The count, surprised and intrigued, sends the gipsies away.) 


AFTER VERISMO: ART NOUVEAU & CLASSICAL REACTION 187 


(For many years, it seems to me, I have been dreaming a wild dream.) 


The heroine of Der ferne Klang, Grete, is a minor character in the long 
portrait-gallery of erotic pathology, which stretches from  Verdi's 
Violetta to Alban Berg's Lulu, through Wagners Kundry and 
Massenet's Manon and Thais, Strauss's Salome and Puccini's Turandot. 
The exploitation of sexual elements is an even stronger feature of 
Schreker's Die Gezeichneten (The Branded) (1918). 

Art nouveau in its most innocent, even etiolated form strongly 
influenced two English operas of the period. The mystery of life and 
death and the all-prevailing power of conjugal love are presented in an 
Indian scriptural setting, and in a musical language of extreme chastity 
by Gustav Holst (1874—1934) in his chamber opera Savitri (1916), and a 
Celtic fairy twilight provided the background for The Immortal Hour 
(1914) by Rutland Boughton (1878-1960). The revived interest in 
Celtic mythology, and the example of Wagner, prompted three operas 
by Joseph Holbrooke (1878-1958), The Children of Don (1912), Dylan, 
Son of the Wave (1914), and Bronwen (1929). These stood out as novel 
compared with the well-made but musically characterless operas of Sir 
Charles Stanford (1852-1924), whose Shamus O'Brien (1896) put 
conventional Irish comedy on the operatic stage for the first time. Art 
nouveau elements appear in the wood-spirits of Der Wald (1902) by 
Ethel Smyth (1858-1944), whose Strandrecht (The Wreckers) (1906) 
would have insured her a future as an operatic composer in any country 
but England, where opera at this time was virtually confined to a 
summer season at Covent Garden closely linked with the social life of 
London and provided almost entirely by foreign artists. A later comedy, 
The Boatswain's Mate (1916), contains much shrewd characterization 
and clear references to the still continuing struggle for women's rights. 

In Italy Gabriele d'Annunzio's highly coloured and scented, but 
sterile poetry had a powerful effect on all the arts, prolonging the 


188 STAGE WORKS: 1890-1918 


‘decadence’ of the nineties by at least two decades. As verismo lost its 
novelty, d'Annunzio's influence prompted a wave of spectacular neo- 
romanticism to be felt in Francesca da Rimini (1914) by Riccardo 
Zandonai (1883-1944), a work full of medieval pageantry including a 
stage orchestra of flute, piffero, clarinet in C, viola pomposa, and lute. 
This, like Fedra (1915), the first opera by Ildebrando Pizzetti (1880— 
1968), was based on a play by d'Annunzio, who also provided the 
subject of La Nave (1918) by Italo Montemezzi (1875-1952). 
Montemezzi’s L’Amore dei tre Re (1913) is outstandingly the finest of 
these neo-romantic operas, a powerful story presented in a musical 
language quite free of the more facile theatrical exaggerations and 
stylistic coarseness of verismo, but still plainly indebted to Tristan in 
the love-music which forms the greater part of the work, one way and 
another. 


o Е: а жесе PS. 
Г) кожа В jesse] aE, 
Lcd A eS ЕЕ 52] ppp 


Io non ti vo - glio guar- 


AFTER VERISMO: ART NOUVEAU & CLASSICAL REACTION 189 


сай». 79 atempo 


(I do not want to look at you before you speak, beloved, chosen one, ever-living, 
my soul...) 


At least a decade before verismo had produced its last major flowering 
in Puccini's Л Tabarro (1918) or combined with the neo-romantic 
fashion in I Gioelli della Madonna (The Jewels of the Madonna) (1911) 
by Ermanno Wolf-Ferrari (1876-1948), this gifted half-German, half- 
Italian composer was anticipating what was to become the classical 
reaction in two operas based on Goldoni. Both Die neugierigen Frauen 
(The Inquisitive Women) (1903) and Die vier Grobiane (A School for 
Fathers) (1906) were given at Munich, where the Venetian-born 
composer had studied with Rheinberger, acquiring an excellent and 
robust musico-dramatic technique which he used to evoke the world 
of the Italian eighteenth century. Wolf-Ferrari's musical language is 
traditional, but his use of jt is personal and his strong melodic gift often 
seems to lie midway between Mahler and Puccini, as in this song from 
Act I of Die vier Grobiane: 


Ех. 89 
Tranquillo 
LUCIETA 


Ja, lie-ber Va-ter mein, so sein Sie wie- "s Кеш, bin 


4 -- аа | есі 
Ob = perd жиса к= Ежен) кте) е а 
p ке GE 


190 STAGE WORKS: 1890-1918 


_ dolce 
и 
Б Мю А - мм” оре | 


в» е”, e 

ae SS == ее РЕН 

2 D) Э РОО, 
БЕ 2 


TE тә Fn.) ue 
JE? aerea. жен ша. D e] 
[mmc 


kannauchoh - ne Ball_ von Her - zen 


[== E 
==) = aaa ги БЕБІ 
[-— 2 a = sui] = 


(Be nice to me again, dear father, for after all I am your little daughter, and a 
good and obedient little girl. It’s Carnival too—but of course I can be joyful and 
happy without a mask or a ball.) 


The very topical Susannens Geheimnis (Susanna’s Secret) (1909), 
where the plot turns on a husband’s discovery that his wife smokes 
cigarettes, is a ‘modern-style’ comedy such as Strauss was later to 
write in his autobiographical Intermezzo. 

Neither the mood of the times nor the musical language available 


AFTER VERISMO: ART NOUVEAU & CLASSICAL REACTION 191 


favoured the writing of such broadly human comedies as Die Meister- 
singer or Falstaff during the first two decades of the twentieth century. 
The outstanding exceptions, Der Rosenkavalier and L'Heure espagnole, 
though musically far more ambitious than Wolf-Ferrari's charming and 
slender period pieces, do not stand comparison with Wagner or Verdi. 
Maskarade (1906) by the Danish composer Carl Nielsen, is a comedy 
based, like Wolf-Ferrari's Goldoni operas, on an eighteenth-century 
classic—Holberg’s comedy of 1723—and alternating between a lively 
neo-classical style, partly pastiche and partly Regerish in character 
(overture, the folie d'Espagne and archaic dances in Act I, Hieronymus's 
C major aria in praise of *the good old days', and the three-part madrigal 
in Act III) and memories of nineteenth-century models. These include 
Die Meistersinger (the character of Arv recalls David, and the Night- 
watchman in Act III is an obvious borrowing) and, at a deeper level, 
Berlioz. In an earlier opera, Saul og David (1902), the love-duet between 
David and Saul’s daughter, Michal, already recalls the unique combina- 
tion of sensuality and chastity that distinguishes the love-music in 
Berlioz's Roméo et Juliette and Les Troyens; and the scene between 
Leander and Leonora in Act II of Maskarade confirms this impression: 


LEONORE LEANDER poco rall. 


(kissing her) 


po pee ЕЕ a 2 а] 
ТОЙ GRECO ТЕН E ЖЕП "ЕН ee 


ртс тыл 22 ы ————— 5 
=. E92 Feu] 


192 STAGE WORKS: 1890-1918 


LEONORE 


() Я И о. 

at Б 

= MUI М-С 

Басты ы | CHRGLEN. ee Жанр и Қолы" USE UO тыз стш ше ооо 
291 в 

кшш ee meae uni] 


Buts —H 
24 


(Leander: No, let me drink from rosy beakers the happiness, the life which you 
have given me by that single word. Leonore: Yes, drink on!) 


In the character of the valet Henrik, Nielsen created one of the great 
servant-roles of operatic literature, a worthy Danish successor to 
Mozart's Leporello. 

A very different kind of wit from that of either Nielsen or Wolf- 
Ferrari marks the operas of another half-German, half-Italian composer, 
Ferruccio Busoni (1866-1924). His double nationality found a striking 
parallel in a double indebtedness to J. S. Bach and Liszt, who between 
them dominated his twin careers of composer-theorist and piano 
virtuoso. It is significant that, whereas Wolf-Ferrari turned for the 
subjects of his operas to Goldoni, Busoni's choice among Italian 
writers was the fantastic, even visionary Gozzi, for whose Turandot he 
wrote incidental music in 1911. From this he later constructed an 
opera of the same name, given in a double-bill with his Arlecchino 
in 1917. These and Die Brautwahl (1912) represent the first conscious 
intrusion into the operatic field of the neo-classical spirit—objective, 
quick-witted, dry, and contrived, in the sense of stylistically eclectic— 
which had already appeared in Busoni’s own Fantasia contrappuntistica 
and even earlier in the works of his friend Max Reger. 


1 See above pp. 82 ff. 


AFTER VERISMO: ART NOUVEAU & CLASSICAL REACTION 193 


Mascagni had already attempted, though unsuccessfully, to reclaim 
for the opera the figures of the Italian commedia dell'arte in his Le 
Maschere (The Masks) (1901). Busoni's Arlecchino is altogether 
different from the sentimental and decorative, Watteauesque use of the 
old comedy figures which became fashionable in French literature after 
the publication of Jules and Edmond de Goncourt's L’Art du dix- 
huitiéme siécle (1859—75) and developed into a mannerism among poets 
of the nineties who imitated Verlaine.t The masks in Arlecchino are used 
to lend greater objectivity and more general relevance to Busoni’s very 
shrewd and individual comments on human existence. The musical 
style of Arlecchino is desperately taut and nervous, as full of allusions as 
Busoni’s text, which shows an intimate knowledge of Dante and Goethe. 
Musical references include quotations or parodies of Don Giovanni’s 
‘Finch’ han del vino’ and Beethoven’s Fifth Symphony and the Fidelio 
march, with sidelong glances at Rossini and Wagner, and some 
eighteenth-century pastiche. Much of the music has a nervous, staccato 
glitter that looks forward to Stravinsky’s Rake’s Progress—the last of 
the neo-classical operas, as this is the first—but the sham funeral march 
in Act IV might be a parody of Mahler, and there are passages in which 
the telling awkwardness of the vocal line and the orchestral spacing 
recall those of Dargomizhsky’s Stone Guest, though the more probable 
and immediate source is Strauss’s Elektra: 


1 The composer’s own curiously ambiguous attitude to his Arlecchino, and to opera in 
general, can be found in his essay Entwurf einer neuer Aesthetik der Tonkunst (1907), 
relevant passages of which appear in The Essence of Music (English translation, London, 
1957), pp. 40, 60-70. 


194 STAGE WORKS: 1890-1918 


(I know of cardinals who would deserve to be treated by you.) 


The music of Turandot has little except some oriental pastiche in 
common with the opera that Puccini left unfinished at his death in 1924. 
The fantastic plays an even more important role here than in Arlecchino, 
and the eclectic, consciously fabricated nature of Busoni's music is 
emphasized when he introduces Ше old English tune of ‘Greensleeves’ 
at the opening of Act II. Intellectual contrivance, a vivid musical 
imagination, and vast musical and literary erudition are combined in 
Dr. Faustus, the opera which the composer left unfinished at his death 
in 1924.! Busoni's libretto shows a literary gift very similar, and not 
much inferior, to his musical powers. After the opening symphonia, 
which evokes the spring festival of Easter, two preludes show Faustus 
approached by three mysterious students from Cracow and then 
summoning Mephistopheles. As Faustus signs away his soul, Busoni’s 
‘Credo’ and ‘Gloria’ emphasize the world that he is deliberately 
rejecting; and the extended organ-solo in the following church scene, 
where Gretchen's brother is seen praying for vengeance, is used 
for the same effect, while the vivace Te Deum in 6/8 time in the second 
scene of the Hauptspiel becomes a song to wine and women. The first 
scene of the Hauptspiel, which shows Mephistopheles at the court of 
Parma, contains many characteristics of nineteenth-century grand 
opera—a polonaise, pages duelling to waltz-music, noble girls with 
wreaths, and scenes of magic in which Mephistopheles conjures up 
apparitions of Solomon and Balkis, Samson and Delilah, John the 
Baptist and Salome, before spiriting away the Duchess. In the following 
scene, Mephistopheles appears in a Wittenberg tavern among quarrel- 
ling students (Te Deum pitted against ‘Ein’ feste Burg’ in the manner of 
Meyerbeer's Huguenots). 

In the final scenes the appearance of the Nightwatchman and 


1 It was completed by his pupil Philipp Jarnach and performed in 1925. 


AFTER VERISMO: ART NOUVEAU & CLASSICAL REACTION 195 


Mephistopheles's tempo di minuetto serenade with pizzicato accompani- 
ment inevitably call up the ghosts of Wagner and Berlioz. But the end 
of the opera is a monologue for Faustus, in which Busoni's own 
Faustian spirit of enquiry and *young classicism' find worthy expression. 
Busoni's eclectic historicism with its many literary as well as musical 
overtones from the European past, seemed altogether too contrived and 
intellectual when Dr. Faustus first appeared. Fifty years later the 
perspective has changed as completely as in any conjuring-trick of 
Mephistopheles himself; and what seemed the freakish, if brilliant 
concoction of an intellectual visionary has found an honourable place 
in the mainstream of operatic history. 

If there was something of the Abbé Vogler about Busoni, Hans 
Pfitzner's lonely championship of late romantic musical ideals and 
language in the revolutionary atmosphere of the 1914-18 war suggests 
rather a musical Don Quixote.! His Palestrina (1917) is the latest of the 
operas centred around a character in whom the power of music 
is personified, as earlier opera-composers had chosen Orpheus and 
Wagner Hans Sachs, although Hindemith was to choose artist-heroes 
for his Cardillac (1926) and Mathis der Maler (1938). Pfitzner's musical 
idiom is basically Wagnerian, though extended by harmonic freedoms 
that owe something to Strauss, perhaps more to Reger. The legend of 
Palestrina’s composing the so-called ‘Missa Papae Marcelli! in a single 
night gave Pfitzner, who wrote his own libretto, the opportunity for a 
powerfully conceived choral scene (Act I, scenes 4-6) in which 
composers of earlier ages urge him on and a chorus of angels echoes 
phrases of the Mass as he writes them. In Act II the scenes depicting the 
Council of Trent contain masterly vignettes of the leaders of the 
different national factions and their suites—the Bishop of Budoja 
characterized by brisk neo-classical music, for example, the Assyrian 
Patriarch using oriental melismata and the Master of Ceremonies, 
Ercole Severolus, depicted in austere brass unisons. In the final scene of 
the opera, where Palestrina is shown sitting peacefully at the organ 
while crowds outside hail him as “езсиег of music', Pfitzner clearly 
reflected his idea of his own position as legitimate champion of musical 
traditions which he saw on all sides rejected. His loyalty to the past was 
to become increasingly intense, and blind, during the next thirty years, 
while the musical revolution was to establish its legitimacy often in the 
face of social ostracism and even, in some cases, political persecution. 
Nineteenth-century traditions, whether of Wagnerian music-drama or 
Italian verismo, could not resist the new forces which were transfiguring 
society as well as the arts. Puccini wrote the last Italian grand opera 


1 Pfitzner made a personal attack on Busoni in an essay of 1917, *Futuristengefahr'. 


196 STAGE WORKS: 1890-1918 


with Turandot and died in 1924. Strauss, like Pfitzner, was less fortunate 
in living another quarter of a century and continuing to write, against 
the current of the times, complex and luxuriant fables like Die Frau 
ohne Schatten (The Woman without a Shadow) (1919), Die aegyptische 
Helena (1928) and Die Liebe der Danae (1940) or nostalgic comedies in 
the vein of Der Rosenkavalier, like Arabella (1933). Despite real indivi- 
dual merits these proved as ineffective against the tide of events as his 
own Marschallin's desire to put the clocks back in order to preserve her 
youth. The lyrical drama was far from dead, but its spirit had deserted 
the old forms and was searching for the new. 


TOWARDS NEW SHORES 

Although Strauss and Pfitzner eventually rejected Wagnerian music- 
drama in favour of a return to the traditional opera, we have seen that 
in Russia Rimsky-Korsakov's last operas suggest a quite different way 
out of a situation which most intelligent composers felt to be an impasse. 
Instead of insisting, like the Italian veristi and their followers in France 
and Central Europe, on ignoring the gulf between a highly artificial 
form and naturalistic subject-matter, Rimsky-Korsakov chose exclu- 
sively fantastic subjects; and Debussy, though avoiding all hint of 
fantasy and the supernatural, found in Maeterlinck’s Pelléas et 
Mélisande a largely interior, only half articulate drama. These were in 
effect the two paths explored by the composers who came to maturity 
around 1910, and they led to two quite different solutions: on the one 
hand the ballet and on the other hand the symbolist or ‘expressionist’ 
drama with music, the one essentially an extrovert form and the other 
raising introversion to a principle. 


BALLET 


Although ballet had been cultivated in Russia on quite a different 
footing from any Western European country, with the possible excep- 
tion of Denmark, at least since the days of Noverre, ballet music first 
became a serious art-form in France with Adolphe Adam and, far more 
importantly, Leo Delibes (1836-91). The first great ballet scores in 
Russia were Tchaikovsky's,! much indebted to Delibes in detail but 
incomparably richer and more varied musically. It was the musical 
standard set by Tchaikovsky іп Swan Lake, The Sleeping Beauty, and 
The Nutcracker that made it possible for the young Igor Stravinsky 
(1882-1971) to consider ballet as a possible form, and the example of 


1 See Vol. IX. 


BALLET 197 


his teacher, Rimsky-Korsakov, that showed him how the fundamentally 
plastic language of ballet could be employed in a work which lay on the 
borderline between this and opera. The first act of his opera Тйе 
Nightingale was written in 1908-9, before his three ballets Zhar-ptitsa 
(The Fire-Bird) (1910), Petrushka (1911) and Le Sacre du printemps 
(The Rite of Spring) (1913) were presented in Paris by Sergey Dyagilev’s 
Ballet Russe. In both ТЛе Nightingale and The Fire-Bird Stravinsky is 
still very much indebted to Rimsky-Korsakov’s ‘magic’ orchestration. 
The intermezzo ‘Courants аш” is full of harp and horn glissandos, 
glittering trills and tremolandos from the whole orchestra; and the 
Nightingale's song recalls the coloratura of the Queen of Shemakha in 
The Golden Cockerel: 


Ex. 92 ^ 2. J 
) = 66 (circa) Ee. Гаф ae ROSSIGNOL 


ast c Pe 2 4:. P та 
Hity EHI iml eM 
BUSS внесете e Жш ESAE cual кезе) 
EE i seg ес = жини às. 


0 - те 
Ут НЧ 9-614597 ФФ ti Ot 
E = л рел cum а ЕВ = жору "ek Ре цеви X 
SSS еее ы E Sea at as a 

ХЭР EE es ee | ee | == 


“ын 
de усл сс к= 


pipe EA 


meno mosso accel. 


r) 

ги аа == Т гу E. 2 

“л-ден ша а Рен 
pen 


In The Fire-Bird Stravinsky shows his acquaintance with contemporary 
French music very clearly, and the earliest example of what was to 
become his characteristic mature style, laconic in utterance, with 
irregular ostinato rhythms, and analytical rather than synthetic orches- 
tration is to be found in the ‘Danse infernale’; 


198 STAGE WORKS: 1890-1918 


Tromb., Ob., Fl., Cl. 
mf Cymbal roll 
= 


In Petrushka, a puppet fantasy played against the background of 
a Russian Shrovetide fair, the crude, incisive, spiky harmonies are 
either superimposed to form a giant accordion-like orchestral sound, 
still recognizably diatonic, or detached from this background with 
exaggerated emphasis and high relief that recall those of a Fauviste 
painting. The quotation of deliberately trival music—a French café 
song and a Tyrolese waltz by Lanner—and the embedding of this 
material in a score where echoes of Russian folk-music closely jostle 


BALLET 199 


fragments of a highly sophisticated Konzertstück for piano and orchestra, 
together form a clear portrait of Stravinsky's musical character, in 
which sharp intelligence, curious erudition and the gifts of a colourist 
and a miniaturist rather than those of a musical architect are directed 
by a forcible, even aggressive personality. This aggressive trait was to 
find formidable expression in The Rite of Spring, ‘pictures of pagan 
Russia' presented with a neo-primitive savagery of rhythm and a 
frequent piling-up of disparate harmonic aggregates that seemed at first 
provocatively revolutionary. In fact, however, the substance of Stravin- 
sky's harmony in Le Sacre is seldom bolder than that to be found in the 
last piano works of Aleksandr Skryabin,! and it was the vertiginous 
rhythmic vitality and the enormous percussive impact of the music that 
gave it a unique character. Where Stravinsky's harmonic daring differed 
from Skryabin's was in its application; and here Т/е Rite shows clearly 
the influence of his French friends, Maurice Ravel (1875-1937) and 
Florent Schmitt (1870-1958). In La Tragédie de Salomé (1907-12) 


(3 + 13) 


Schmitt wrote a ‘Danse des éclairs’ with a rhythm of anda 


‘Danse de l'effro, which plainly anticipate The Rite. Already іп 
Petrushka, and even in Zhar-ptitsa, Stravinsky had exploited with bril- 
liant effect the clash of triads whose tonics lie an augmented fourth apart 
(C-F sharp)—a bitonal effect already used by Ravel for the cadenza 
of his Jeux d’eau (1902); in The Rite it is Stravinsky’s methodical 
insistence, at the opening of Ше ‘Cercles mystérieux des adolescentes’, 
on a chord consisting of a dominant seventh on E flat and a chord of 
F flat (E) major, rather than the chord itself, that was novel. On the 
other hand the rhythmic pattern in the final ‘Danse sacrale'—9/8-5/8— 
3/8-2/4-7/4-3/4-7/4-3/8-2/4 and so on—was a unique innovation in 
European music.? The leading French composers of the day were soon 
fired by the new possibilities suggested by Dyagilev's Ballet russe. 
Debussy's Jeux and Ravel’s Daphnis et Chloe, both dating from 1912, 
represent the latest and highest achievements in that Impressionist 
writing for the orchestra which had been developing since Debussy's 
Prélude à l'aprés-midi d'un faune (1894). The pointillistic style already 
developed in painting by Georges Pierre Seurat, who died in 1891, was 
effectively imitated in Jeux, where Debussy breaks down his material 
to the smallest possible units, writing music that is not so much thematic 
as cellular. Ravel's score is perfectly judged and organized to the 


1 See chapter 2, pp. 133-7. 
? Cf. Jean Cocteau, Le Coq et l'Arlequin (Paris, 1918), p. 64 *Le Sacre est encore une 
“oeuvre fauve’’—une oeuvre fauve organisée’. 


200 STAGE WORKS: 1890-1918 


smallest detail, a monument of that refinement of taste and that poetic 
sensibility expressed in terms of colour and rhythm, which were to be 
superseded by the neo-primitive savagery of The Rite and the dry, 
objective prose style of the neo-classical reaction. An elegant, well- 
made ballet-score of the same period is Paul Dukas's La Péri (1912), in 
which Russian influence is again strong. 

Quite different from this Franco-Russian ballet style which developed 
from the mutual exchanges of ideas between Rimsky-Korsakov and 
Stravinsky on the one hand and Debussy and Ravel on the other, was 
Le Martyre de Saint Sébastien (1911), incidental music written by 
Debussy for a mystery-play by Gabriele d'Annunzio. Here the most 
disparate and unexpected sources are drawn upon by the composer— 
Palestrina, the Wagner of Parsifal, and the Liszt of Christus—to form a 
modern version of that fresque musicale at which Gounod had aimed at 
the end of his life. D'Annunzio's hypertrophic and profoundly insensi- 
tive art was at the opposite pole to Debussy's; but Le Martyre lies on 
the same borderline between opera and ballet as the symbolist dramas 
of Bartók and Schoenberg. Much of the music is in a flat-toned fresco 
style that suggests the oratorio rather than the theatre (Ex. 94). 

Debussy and Ravel both obliged Dyagilev by either supplying him 
with new ballet-scores or permitting the use of former works or arrange- 
ments of former works (Prélude à l'aprés-midi d'un faune, Ma Mére 
l'oye, Valses nobles et sentimentales, renamed Adelaide ou le langage des 
fleurs). Manuel de Falla (1876-1946) was the most gifted of the com- 
posers, after Stravinsky, who owed their international reputation to 
music written for the Russian Ballet. In E/ amor brujo (Love the Sorcerer) 


Ex. 94 
Mezzo- - TA : = = | m 
Sop. j ——À z ES = = т 
=, Тар; a 
(à bouche fermée) 
5 IA —— fem 
Al ===: = Е 2 | E 2 5 =| 
Se е. Ш — 


ж о Solo Vin 


ес eae -= 
р 3 x — pi f Е 
SE ТЕТЕ 


| ә- Е : Tajer pex ee ие” 


pum а ет, 
A he ae RET 


e "T ДА 


5 
и 


BURLINGAME 
PUBLIC: 


PLATE IV 


лош ay} Jo ыға oy} ш oursseJA 1эчАел8оэлоцо ou) рив osseotq Ад 195 ә8®]$ au) Зшмоцв ‘эиэ5$ Suruado au 


(102 d 298) 616] ШУН QHNINSOO-ITHH.LL ЯНА S.N'YIVA 


f T 
саз етер I)e ы с Eme. ume г LE Га... m— I md 
ГД Бери eL = ee |на 
ТЕ Co WE en E y 255 Le ee eee 


с E—————— S e 
СУ: + е. O- 
v ы 


Е E Г Г | ВЕЕТ —] 
El, Ae e отэ (ESS U M "ry ee іа ысы 60 ee ee 1 

кү ee SS ES SSeS SS SS a ДЕ es ПН т ee ЧЕ гг л 
p ал eee ны 0] ee ee ПЕЛЕ ee Г ИЙЕ ЭШЕ IT -r- [. 
TI NEIN "ug ЖЕРЕ 

() 
m ES N 
Г Б 

М ЕШ Г „чаш ан? ДГ 

съм? Sree Cog dee S DENE те) ғы) 


(1915) and El sombrero de tres picos (The Three-cornered Hat) (1919)! he 
moved forward from the refined zarzuela style of his early La vida 
breve (1913) to a language in which Spanish dance-rhythms and 
melodies based on Spanish folk-song patterns are combined with a 
French elegance and polish, and the neo-classical element played a 
considerable part. In Е/ sombrero de tres picos, where the influence of 
Petrushka is often felt, the use of a quasi-symphonic style gives the 
*Dance of the Neighbours', which opens Part II, a size and a breadth 
quite new to music based on Spanish folk-songs and folk-dances. 

In El retablo de Maese Pedro (Master Peter's Puppet Show) (1923) 
Falla wrote the finest and most intelligent of puppet-operas, reproducing 
with extraordinary skill the atmosphere of sixteenth-century Spain and 
drawing, like his master Felipe Pedrell (1841-1922), on contemporary 
material as his inspiration. Е! Retablo is as untranslatable as Dargo- 
mizhsky's Stone Guest, but it is a small masterpiece. In Pedrell’s own 
opera, Els Pirineus (The Pyrenees) (published 1894, performed 1902), 
archaic material is combined with music written in a style that owes 
something to Wagner yet remains fundamentally Latin in its graceful 
sentiment and physical verve. There was no flowering of opera, how- 
ever, in Spain; Falla's dance-dramas remain the only serious Iberian 
contribution to the lyrical theatre in the first quarter of the twentieth 
century. 


1 See pl. IV. 


202 STAGE WORKS: 1890-1918 


SYMBOLIST DRAMA 
Music formed an integral part of the Symbolist aesthetic, since it 
represented an ideal, abstract art to which all other arts are continually 
approximating. A Symbolist poem is essentially an act of contemplation, 
a magic evocation of the metaphysical reality behind appearances. 
Action belongs to the other, phenomenal world; and we have seen 
how Maeterlinck's dramas attracted both Debussy and Dukas by their 
interior nature, an intense life of the spirit that is hardly exteriorized in 
gesture or action. ‘Vous avez brisé la glace avec des fers rougis’ says 
Pelléas, after Mélisande's confession of love; and it is characteristic of 
symbolist drama in general that the surface of life is there shown as dis- 
turbed only when a character's emotional life has reached volcanic heat. 
Іп А kékszakállú herceg уйға (Duke Bluebeard’s Castle) (written in 
1911, performed 1918) Béla Bartók (1881—1945) set an Hungarian 
adaptation by Béla Balázs of Maeterlinck's Ariane et Barbe Bleue in a 
style where the tonic accent of the Hungarian language combines with a 
parlando rubato to create vocal lines entirely new to European music. 
There is no action, only the change of mood induced in Judith by the 
successive opening of the seven doors in Bluebeard's castle. These are 
symbolized not only by the changing nature of the orchestral music (a 
46-bar trill, recalling Strauss's Sa/ome, for the torture-chamber: a horn 
theme against string tremolos for the flower-garden; an orchestral tutti, 
with organ and Debussy-like sequences of major triads for Bluebeard's 
wide domains) but also by the play of light and colour. This is an 
important element in the work, which moves out of darkness into light 
and back again into darkness. This association of colour with music, 
which had already interested Skryabin! and was to concern Schoenberg, 
is indicative of the synaesthesia which had been adumbrated by E. T. A. 
Hoffmann and Baudelaire and consciously aimed at by Wagner in his 
ideal of the Gesamtkunstwerk, against which Stravinsky and the whole 
neo-classical school were to protest. Although action is minimal, the 
psychological reflection of the music is clear: Bluebeard's happiness is 
contingent on those he loves, and he can possess them only in memory. 
Judith, by arousing these memories, automatically becomes one of them. 
Although there are momentary echoes of Debussy and Strauss (the 
D flat major 6/4 chord at Bluebeard's protestation of love after the 
opening of the sixth door) Bartók's language is already an individual 
blend of Hungarian and Western European elements, with chains of 
sevenths or minor seconds often built into powerful combinations. 


1 See above p. 36. 


SYMBOLIST DRAMA 203 
Ex. 95 


Sempre molto sostenuto 


Vc., Bn. кр Sa ol 


aa чаша) a See еве Қа стена 
сова ре p EE 
итар EADEM 


е 


n 


kit sze-ret-tél én e - ]ót - tem? 
(Tell me, Bluebeard, whom have you loved before me?) 


Maeterlinck had attracted the attention of other writers and musicians 
quite outside the orbit of the French-speaking world, and Arnold 
Schoenberg (1874-1951) had written his symphonic poem Pelleas und 
Melisande (1902-3) ‘after Maeterlinck', as well as setting Herzgewdchse 
for soprano, celesta, harmonium, and harp (1911). More important in 
determining the character of Schoenberg's work for the theatre, how- 
ever, were his experiences as a painter. These began in 1907, when he 
created portraits and masks which foreshadowed those of the artists 
who, five years later, were to form the Blaue Reiter group in Munich— 
Kandinsky, Klee, and Marc. Expression— not the rendering of the 


204 STAGE WORKS: 1890-1918 


visible, but the rendering visible —was the object of this art; and 
Expressionism had from the outset something of the amorphous quality 
of the human psyche itself, lacking both the geometrical foundations of 
French Cubism and the intellectually defined aims of Italian Futurism. 
Schoenberg's monodrama Erwartung (Expectation) (composed in 1909, 
performed 1924) is a minutely detailed study of erotic tension and 
existential anxiety. Although the speech-like prosody is naturalistic, 
the exaggeration characteristic of all Expressionist art is to be found in 
the enormous melodic intervals, which give the vocal line the nature as 
well as the appearance of a fever chart.! Marie Pappenheim’s text 
presents the monologue of a woman walking through a forest at night 
in search of her lover, whom she eventually finds dead. Every minute 
inflexion of her feelings is reflected in the myriad changes of instru- 
mental colour, pitch, texture, and dynamics; and unacknowledged 
resentments and hostilities are exteriorized with the acuteness of the 
new psychological analysts. Schoenberg's language owes something still 
to the Salome and Elektra of Richard Strauss, but its nervous tension 
and fragmented texture are entirely novel, and the detailed, often 
delicate scoring for the large orchestra achieves a feverish delicacy. 


Ex. 96 

II. Szene (Very dark, wide path; tall, closely planted trees. She gropes her way 
4-52 forward) 
wieder etwas langsamer 


(noch hinter der Szene) 
FRAU D 


ЕЕ = 
Ва ЕД А 
ага ге нед 31 

FH Di" E „с ЛА BC ey (шел Ge Le Lm Р 
D ще 


1 H. Н, Stuckenschmidt, Arnold Schoenberg (Zürich, 1951: English translation, London, 
1959), p. 54. 


SYMBOLIST DRAMA 205 
459 
wieder viel rascher 


z (trembling, tries to look at her 
(stoops, gropes with her hands, then screams) 


own hand) 


Was? Јаб 1051... Eingeklemmt? . . Nein 


HURTS ли, = mie 
Г В Hb = Dd 


a eee =e aaa Ty zd [277 
Li е - Н ра 7 p mcs amen) 22 
И Нм Е Ее. Р716 
е 2 1 еу а 


ТЕР иа 


АР га 
г ИГТ 
bz — 1719057 


206 STAGE WORKS: 1890-1918 


mf Tbn. (con sord.) 


(Is that still the path? . .. Ah! here it is. What's that? Leave go of me! ... hemmed 
іп?... No—there was something creeping . . . and here (оо... who is that touching 
me? Go away—I must go оп...) 


In Die glückliche Hand (The Lucky Hand) (written in 1910-13, per- 
formed 1924) Schoenberg wrote an Expressionist drama in which the 
distinction between dream and reality is abolished. There is no logical 
sequence in the scenes between the Man, the Woman, and the Gentle- 
man, only an often confused rhythmic polyphony and “ап insatiable 
piling up of harmonic complexes as an allegory of the many layers of 
the psychological subject.’! The writing for the chorus of six men and 
six women includes whispering and speaking as well as singing, and 
Schoenberg's insistence on the exact visual representation includes 
crescendos of colour, or shifting colour-schemes which reflect the 
emotions of the characters, as in the 25-bar sequence brown-dirty 
green-dark blue-grey which corresponds to a growing tension in the 
Man, and leads eventually to a bright-yellow climax. 

These two works of Schoenberg's are the first examples in music of 
that exclusive concern with the suppressed elements in human existence 
that Adorno regards as necessarily characteristic of twentieth-century 
art.? “Тп so far as art to-day has any real substance, it reflects without any 
concessions, and brings to the level of consciousness, what people would 
like to forget.' Radicalism such as this, pursued with such total and 
exclusive passion, posed the problem of musical theatre in terms which 
made any substantial continuity with the old opera out of the question. 
It was left to Schoenberg's pupil Alban Berg (1885-1935) to find a 
compromise.? 

Outside Central Europe composers were not attracted by Symbolist 


! Theodor W. Adorno, Philosophie der neuen Musik (Tübingen, 1949), p. 20. 
2 Ibid., p. 8. 3 See below pp. 363 ff. 


SYMBOLIST DRAMA 207 


drama and knew little or nothing of Expressionism in these years. 
Rebikov's *musico-psychological drama’ Alpha i Omega (1911), which 
purports to show the beginning and the end of the human race in terms 
of diabolical seduction, is no more than a curiosity rendered musically 
tedious by an all too thorough and unimaginative use of whole-tone 
harmony. Stravinsky, on the other hand, was to experiment during the 
war years, when the resources of Dyagilev's ballet were no longer 
available, with chamber-music entertainments such as Renard (com- 
posed іп 1916-17, performed 1922) and L’ Histoire du soldat (1918). But 
both these must be considered as substitutes for opera rather than 
extensions of operatic form, and their examination belongs to the next 
chapter.! 


1 See pp. 211-15. 


IV 


MUSIC IN THE MAINLAND 
OF EUROPE: 1918-1939 


By Mosco CARNER 


GENERAL CHARACTERISTICS 


HERACLITUS'S “АП is flux, nothing is stationary’ is true of all epochs 
that have witnessed a radical moral, political, intellectual, and artistic 
transvaluation; but it would appear to apply with particular force to 
the arts in the years between 1918 and 1939. Centripetal and centrifugal 
forces were pulling in opposite directions, school stood against school 
each proclaiming a different aesthetic creed, and the air was full of 
slogans and counter-slogans. It was as though a new Sturm und Drang 
had seized the European mind. In music, rumblings of the coming storm 
had been heard during the decade or so before the First World War. 
Both Debussy's impressionism and Schoenberg's expressionism sprang 
from a new ‘feeling’ about music and both represented far-reaching and 
successful attempts to free musical thinking from the fetters of tradition 
and to advance into new territory. Yet it was not until the beginning of 
the 1920s that the full strength of these powerful forces was felt—in 
Central Europe, where revolutionary political and social changes fol- 
lowed in the wake of a lost war and strongly influenced the arts, 
particularly in Germany, but also among young artists in France and 
Italy, who were profoundly affected by the new ideas that had sprung 
up during and shortly after the war. The general feeling was that one 
epoch had come to an end and another was born which held the 
promise of a regeneration, if not indeed a rebirth, of the human spirit, 
bringing with it a liberation from the obsolescent traditions and con- 
ventions of the romantic past. The millennium, so it seemed, was round 
the corner. If this optimism proved unjustified in the light of subsequent 
developments in the political and social life of most European nations, 
it nevertheless had the effect of creating in the arts a new outlook, a 
spectacular change of orientation among artists. 

In music this transformation manifested itself in two aspects. On the 
one hand, open war was declared on the lingering romanticism and all 
it implied—pronounced subjectivism, unhindered projection of the 
artist’s private ego into his work, the tendency to make music a means 


GENERAL CHARACTERISTICS 209 


to an end by using it as a vehicle for emotional expression and psycho- 
logical exploration or rendering it a reflecting mirror of literary and 
programmatic ideas. The most violent attacks were reserved for the 
Wagnerian music-drama and its progeny as the chief culprit in having 
led music away from its true self by the emphasis on emotional, psycho- 
logical, philosophical, and mythological elements, by the sensuality and 
opulence of its tone colours and by the huge array of instrumental 
forces. More positively artists sought to write music that avoided at all 
costs the subjective, expressive factor and that aimed instead at a 
detached, objective attitude such, it was assumed, as had been cultivated 
in the Baroque period. Composers turned to a new consideration of the 
musical material as such, exploring its immanent laws of texture and 
structure and treating it in an analogous manner to the architect's 
handling of bricks and stone. ‘Construction’ became the alpha and 
omega of this New Music. The work of building in sounds and achieving 
a clear order and organization assumed the significance which the 
former emotional ‘content’ had possessed. Technical craftsmanship was 
placed on a higher plane than inspiration, the sheer métier was often 
exalted over imagination. This attitude scarcely affected the aesthetic 
merits of the works of such masters as Stravinsky, Bartók, or Schoen- 
berg, but with less creative artists it led to a species of dry, cerebral 
academicism. Form, design, and texture were reduced to essentials and 
this explains the characteristic predilection for chamber music and 
chamber orchestras, in which each part of the fabric was set out with 
succinctness and clarity and was of individual musical account. This 
was the antithesis of the large forms and dense overloaded texture shown 
in late-romantic music, particularly the opera and the symphonic poem. 
In search of this ideal of pure construction a return was made to the 
style of Baroque polyphony, yet with a marked difference of emphasis; 
for the vast extension and final attenuation of tonality and the eman- 
cipation of the dissonance to an equal status with the consonance had 
led to the gradual destruction of functional harmony, and thus the new 
contrapuntal style developed largely non-harmonic, linear patterns, in 
which vertical configurations became the accidental result of horizontal 
part-writing. In the same way the Baroque forms of suite, concerto, and 
cantata were revived, as ideal moulds into which to pour a purely 
musical content, whereas the dramatic form of the sonata was corre- 
spondingly neglected. Expression became an incidental by-product of 
construction and even where it was intentional, as in operas and can- 
tatas, it was expression not of subjective ego-bound feeling, but of an 
impersonal, more universal order. The object was to concentrate 
attention not on the individual’s personal feelings, but on types of 


210 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


sentiment which possessed general validity. Whether it was the neo- 
classicism of Stravinsky, the New Objectivity of Hindemith, or the 
mixture of experiment with the urge to ‘épater le bourgeois’ of the 
young French artists known as ‘Les Six'—all shared іп common the 
negation of subjectivism and the concentration on an objective interplay 
of the constituents of the musical material. The sole exceptions in this 
general trend towards emotional detachment were Schoenberg, Berg, 
and, chiefly during his pre-dodecaphonic period, Webern. These com- 
posers adhered essentially to the aesthetic tenets of romanticism, 
pressing construction into the almost exclusive service of espressivo; and 
thus they represented, in a sense, the ‘reaction’. Yet, on the other hand, 
Schoenberg’s establishment of the serial method constituted an entirely 
novel and revolutionary approach to the technique of composition and 
resulted in the creation of what may be termed a true ars nova. 

In short, the years between the German Wars formed a period of 
intense search for means whereby to transform radically the face of 
music and lend it a new physiognomy—a process that is parallel to the 
beginning of the seventeenth century when the old polyphony gave way 
to homophonic thinking and a whole new musical outlook was born. 
The strong impulse to find fresh solutions for old problems led to 
a rage of experiment, some of it issuing in merely ephemeral results 
but leading in other cases to enduring achievements which may be 
ranked with the great masterpieces of past ages. 


STRAVINSKY AFTER THE RITE 


One of the foremost exponents of the movement to break with the 
heritage of romanticism and evolve a new aesthetic outlook was Igor 
Stravinsky! (1882-1971). After the monster score of The Rite of Spring 
(1913) in which primitive ‘barbaric’ expression is combined with the 
highest degree of technical sophistication, especially in rhythm and 
instrumentation, Stravinsky realized that an extension of the boundaries 
of music in this direction was not possible and he changed his course, 
making simplicity and economy of means his guiding principle. Charac- 
teristic of this new development were already three pre-war compositions 
—the vocal Three Japanese Lyrics and Pribautki (written between 1912 
and 1914) and the Three Pieces for String Quartet (1914),? intimate small- 
scale works of which the first two are scored for a chamber orchestra. 
This treatment was probably suggested by the peculiar instrumentation 
of Schoenberg’s Pierrot Lunaire which Stravinsky heard in Berlin in 


1 For Stravinsky's earlier music see Chapter II. 
? No. 1 is recorded in The History of Music in Sound, x. 


STRAVINSKY AFTER THE RITE 211 


1912, but it largely corresponded with the new aim he was setting him- 
self in his artistic aspirations. The great restriction of forces available, 
particularly in the Ballets Russes, during the war years between 1914 and 
1918, also played a purely practical part in determining Stravinsky's 
change of front. 

This departure from his previous style reached its first culmination in 
a series of three chamber works for the stage—Renard (1917), L'Histoire 
du soldat (1918) and Les Noces (1917, orch. 1923)—in which Stravinsky 
evolved novel musico-dramatic forms and a new type of orchestration. 
Though they were written at the period when the composer had 
severed his connexion with his native country and was living in 
Switzerland, they still belong in style to his *Russian' period and their 
texts all derive from Russian sources—in the case of Renard and 
L'Histoire from folk-tales and in Les Noces from records of ancient 
Russian folk customs and wedding rites. Stravinsky, who began as a 
nationalist, inherited his love of Russian folk-material from his teacher 
Rimsky-Korsakov, but the younger composer treated it in a strikingly 
different manner from his master. In Renard and L'Histoire he does 
not anchor the story in a definite time or place, but sets it in the 
timelessness and unreality of a fairy-tale; and as early as Petrushka 
(1911), instead of individual living characters he presents puppet-like 
figures governed by rudimentary primitive instincts. This mechanical, 
automaton-like element appears to correspond with a trait in Stravin- 
sky's artistic make-up and finds its musical expression in the stereotyped, 
continuous repetition of short melodic-rhythmic patterns whose 
machine-like regularity is, however, broken up by the conflict between 
metre and rhythm, resulting in the displacement of the strong beats in 
the bar and in an ideal, if not actual, negation of the bar-line division of 
the phrases (see Ex. 97). In setting the texts of these three works 
Stravinsky was, of course, attracted by the story, by its poetic images 
and metaphors; but what exercised on him an equal, if not indeed 
stronger, fascination was the sequence of words and syllables—in 
short, the purely phonetic element which ‘produces an effect on one's 
sensibility very closely akin to that of music’.! 

Renard is a ‘Burlesque in Song and Dance’, one of whose novel 
features is the division of the cast into singers and actors, with the 
former (two tenors and two basses) placed in the orchestra. This proce- 
dure was adopted in order to ensure, on the one hand, the greatest 
possible concentration on the music and, on the other, the greatest 
liberty in the representation of the stage action; and it sprang from 
Stravinsky's dissatisfaction with the absence of close union between 


1 Трог Stravinsky, Chroniques de ma vie (Paris, 1935; English translation 1936, p. 91.) 


212 MUSIC IN THE MAINLAND CF EUROPE: 1918-1939 


music and action observed in the production of his opera Rossignol (The 
Nightingale) (1914). The plot of Renard unfolds in two dramatic episodes 
of which the second, with the exception of the ending, is a literal repeti- 
tion of the first. The Fox lures the Cock from his perch and seizes him 
but, owing to the intervention of the Cat and the Goat, has to release 
him. The play starts again, but this time ends with the killing of the 
Fox by the Cat, followed by a jubilant dance, and the burlesque closes 
with the mummers' ancient request to the public: 


‘Et si l'histoir' vous a plu, 
Payez-moi c'qui m'est dà! 


The style of the work is Russian in the sense that it is marked by the 
repetition of short melodic phrases. But a typically Stravinskyian 
quality is noticeable in their outline, which is clear-cut and sharp-edged 
as though etched with a stylus, a feature that remains characteristic 
of all his later periods. Some of the melodies possess the flavour of 
Russian folk-song, as the introductory March which is partly Dorian, 
partly Phrygian, or the duet between the Cat and the Goat, sung to the 
accompaniment of a gusli.! The instrumental ensemble consists of flute, 
cor anglais, E flat clarinet, two horns, trumpet, string quintet, percus- 
sion, and cimbalom. This last instrument, for which the composer 
developed a strong liking, is largely used to accompany the Cock, play- 
ing sustained pedals and marking the rhythmic accents, and generally 
serves the purpose of holding the harmonic textures together. The sound 
produced by this unusual combination of instruments, particularly by 
the cimbalom, the various percussion instruments and the strings 
plucked and martellato, is dry, sharp, and very resonant. 

Renard is a genuine burlesque with a strong element of caricature, as 
when, for instance, the Fox, disguised as a monk, sings to the Cock with 
an expression of mock-piety,? or when the Cock replies in an arrogant, 
complacent vein.? Stravinsky exacts from the four singers a large 
measure of vocal virtuosity and special effects, their music frequently 
imitating in a stylized fashion the noises of the farmyard. There is, 
however, no attempt at a fusion of the diverse musical, poetic, and 
dramatic elements, which are complementary rather than blended 
together, and the work as a whole creates a static effect partly due to the 
symmetrical arrangement of the action but more especially to the 
automaton-like, circumscribed, and formalized nature of the music. 

L'Histoire du soldat was conceived in peculiar circumstances. During 
the latter part of the war Stravinsky found himself cut off from Russia, 


1 At 62. 2 At 11. VAHLS 


STRAVINSKY AFTER THE RITE 213 


and after the Russian Revolution of 1917 he decided never to return and 
settled for the time being in Switzerland. This meant that he could 
expect no income from Russia and it occurred to him to improve his 
situation by writing a stage work that would require only two or three 
characters and a handful of instruments and that could be produced in a 
mobile theatre to be transported from place to place and staged in the 
smallest locality. The Swiss poet, C. F. Ramuz, who had provided the 
French version of Renard, suggested a mimed narration, a story to be 
read, acted, and danced. Stravinsky's choice fell on Afanasyev's collec- 
tion of Russian tales, which contains a cycle of stories dealing with the 
adventures of a deserter and his compact with the devil. 

L'Histoire has four characters: the Soldier and the Devil who have 
speaking parts, the Princess who is a dancer, and the Narrator who 
combines several functions: he recounts the tale, sometimes comments 
on the action as an outsider rather in the manner of a Greek chorus, 
and at other times participates in it, expressing the Soldier's thoughts and 
feelings and occasionally addressing him in direct speech. The orchestra 
consists of a violin and a double bass, a clarinet (on account of its 
possessing the widest compass among the woodwind) and bassoon, 
a cornet-à-pistons, trumpet and trombone, and finally a variety of 
eight percussion to be managed by a single player, the whole to be 
directed by a conductor. 

Stravinsky was partly influenced in his choice of this combination by 
contemporary jazz, a selection of which had been brought to him from 
America in 1918 by the conductor, Ernest Ansermet. Jazz was bound 
to appeal to a musician with such interest in rhythmic-percussive experi- 
ments as Stravinsky, and indeed he included a tango and a ragtime 
among the Princess's dances and followed this up by his Ragtime for 
eleven instruments (1918) and the Piano Rag-Music (1919). 

The instrumental ensemble is to be placed on the stage, for it was the 
composer's view that seeing the movements of the players facilitates 
Ше spectator’s auditory perception. On another ‘drum’ at the opposite 
side of the stage sits the Narrator, before a little table ‘with a pint of 
white wine', while the centre is occupied by the actors. For the thirteen 
musical numbers, to be played simultaneously with the action or inter- 
polated in the narration, Stravinsky drew on a great variety of sources: 
Russian folk-song, American ragtime, Argentine tango, Viennese 
waltz, Swiss brass band, and Bach chorale.! Yet all these heterogeneous 
elements are fused together in the alembic of Stravinsky’s imagination, 
resulting in a perfect homogeneity of style. The melodic and contra- 


1The Royal March, recorded in The History of Music in Sound, x, is pseudo- 
Andalusian. 


.214 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


puntal lines are invented in terms of the particular nature of the 
instruments producing textures of sharply defined sonorities, and the 
fact that the part-writing is linear rather than harmonically determined 
accounts for the astringently dissonant character of the music, as illu- 
strated in the following example which also shows the pull of rhythmic 
phrases against the underlying metre: x 


Ех. 97 


Clar. 
inA 


Bassoon 


y Ll я Г... Е ей 
Cornet ГД 9 || wv фе [LJ ғ А теулер) 
: AE е? av oi eae Ss ЕЕ PL T Se CT YES E А IR Ы ЕСУ ay sl 
in B flat А = «2 "RN, [np EY (AE A ЕР p 


Stravinsky, it should be observed, pretends to no sympathy, no com- 
passion for the Soldier; he is not emotionally involved and illustrates 
from outside, dispassionate and objective. This tallies with the view he 
expounded in his autobiography! where he wrote that 


music, by its very nature, is essentially powerless to express anything at all, 
whether a feeling, an attitude of mind, a psychological mood, a phenomenon 
of nature, etc. . . . Expression has never been an inherent property of music. 
That is by no means the purpose of its existence. If, as is nearly always the 
case, music appears simply to express something, this is only an illusion and 
not a reality... The phenomenon of music is given to us with the sole 
purpose of establishing an order in things, including particularly the co- 
ordination between тап and time. To be put into practice, its indispensable 
and single requirement is construction. Construction once completed, this 
order has been attained, and there is nothing more to be said... One 
could not better define the sensation produced by music than by saying that it 


! Chronicle of my Life, p. 91. 


STRAVINSKY AFTER THE RITE 215 


is identical with that evoked by contemplation of the interplay of architectural 
forms. 

Stravinsky's theoretical axiom, however, does not entirely square with 
his actual practice which fell somewhat short of his strict demands for 
non-expressive music and pure construction. Even in his neo-classical 
works emotive impulses press to the surface, especially in the stage 


Bae BCC ы) 
a SS Se Sa == Y BI ERO АТ RN. 
ТАЗА И epu <a ee I _ 


works where something of the characters' feelings and sentiments is 
projected into the music, and it was not until Stravinsky embraced 
serial technique that he came near his ideal of pure construction. 
Nevertheless, it was left to the dodecaphonic Webern and the succeeding 
generation of advanced composers to realize (if this is not a contradic- 
tion in terms) the Platonic idea of construction. Les Noces, which was 
completed before L'Histoire du soldat but not orchestrated until 1923, 
expresses, together with the joys and exultation of bride and bridegroom, 
the awe and trepidation felt by man at the renewal of life. Stravinsky 
adapted the text from Kireyevsky's collection of Russian folk poems 
which contains a number bearing on the subject of peasant wedding 
customs, but he used this material in a very free manner, paying little 
heed to ethnographical considerations and aiming to reproduce not the 
actual character of the wedding ritual, but its essential spirit and feeling. 
Though entitled *Russian Choreographic Scenes', the work is more in 


216 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


the nature of a stage cantata, with four.scenes which follow one another 
without a break and employ four soloists (soprano, mezzo soprano, 
tenor and bass), chorus, four pianos, and percussion. There is no real 
plot, the action unfolding in four static tableaux which describe the 
preparations and the actual wedding of the couple. 

The work displays an extraordinary unity of conception and treat- 
ment. Virtually all the material is derived from a single cell—the interval 
of the fourth divided into a major second and minor third and vice versa, 
and there are also other permutations. This basic motive is heard at the 
very beginning: 


Ex. 98 


Soprano solo 


П D 
p 955.189 525.1738 fen 17329: —]13—2 —] 
у NE Е-е е ee ema у = [eps [e ГО СД dii Qn. 60] 
f? Сезе ae ЦИШ БЕ > ое 
Ng ppp Ij cw Е el ааа Дъ ессе mmn | zum Lor 
VY 


Tres-se, tres - se, mà. — ma tres-se à moi 


The motive is pentatonic, being modelled on Russian folk-song, but in 
its elaborations it frequently takes on a modal character, and the 
presence in it of the major second leads, harmonically, to bitonal 
chords at the distance of that interval. In addition to the themes derived 
from the germ cell, there are several other ideas which are first intro- 
duced as counterpoints and subsequently achieve independence. The 
entire melodic material is of Stravinsky’s own invention, with the 
exception of three themes which are, however, so closely integrated 
with the composer’s individual language that they all but lose their 
original identity. The setting of the Russian text is syllabic almost 
throughout and emphasizes the sonorous (phonetic) quality of the words. 
It must be admitted that this peculiar treatment, in combination with 
the melodic-rhythmic repetitions, results in a certain monotony which 
was probably intentional, but it is a monotony which has an extra- 
ordinary mesmeric effect comparable to that of primitive Asian and 
African music. 

Stravinsky's neo-classical phase, which began after Les Noces, has 
sometimes been dismissed as ‘time travelling’, ‘musical impersonation’, 
‘music about music’, or simply as a ‘pastiche’ of older styles. Yet 
Stravinsky’s neo-classicism reflects quite as much his own sharply 
defined individuality as the acute historical sense which he developed in 
order to fuse the Russian elements in his musical personality with the 
wide range of Western musical thought of the last three centuries. To 
possess a sense of the past is to enrich the present; and by his references 


STRAVINSKY'S NEO-CLASSICISM 217 


to the styles of former composers Stravinsky has reconstituted a 
picture of Western musical tradition seen through the prism of his own 
imagination. The original music, observed with the sensibility and taste 
of an artist of a later age, is so metamorphosed as to acquire an aito- 
gether new meaning or, to put it differently, the old material is so re- 
thought and commented upon in the light of Stravinsky's own artistic 
experience that the result is a genuine individual creation. Closely 
linked with this neo-classicizing tendency was Stravinsky's pre- 
occupation with the Nietzschean dialectic of Apollonian and Dionysian 
art. The Apollonian artist has for his supreme object the creation of 
order by means of self-restriction and self-imposed laws. The Dionysian 
artist, on the other hand, is ruled by irrational urges and a pronounced 
tendency towards the frenzied and chaotic. After The Rite of Spring 
Stravinsky gradually turned from Dionysian to Apollonian art, moving 
on to a more or less abstract plane, where the invention appears at 
times cold, cerebral, and even arid. Even in those of his works which 
are coloured Бу a certain expressive element—Oedipus Rex, the Sym- 
phony of Psalms, and Perséphone—the emotions are of a generic, 
universal order finding expression in highly stylized and formalized 
music from which all suggestions of an anecdotal, picturesque, psycho- 
logical, or philosophical nature are absent. 


STRAVINSKY'S NEO-CLASSICISM 


Stravinsky's neo-classical period was initiated by the ballet Pulcinella 
(1920), after music by Pergolesi, but the process of remoulding and re- 
thinking the old material is here seen still in its initial and tentative stage. 
Pergolesi's themes are taken over intact, and while Stravinsky adheres 
to their original melodic outlines it is chiefly in the inner harmony and 
the rhythm that he begins to press his own stamp on the music; the 
harmonic texture is stiffened and given more violent impact by astrin- 
gent chords, the symmetry of the phrases is broken and their regular 
accent often displaced by syncopation. Where Stravinsky is entirely 
himself is in the orchestration, in which he adapts the eighteenth- 
century concertante manner to his own brilliant purposes. The subse- 
quent Symphonies of Wind Instruments (1920) and the Octet (1923) both 
employ an instrumental combination which is the outward expression 
of the impersonal, objective nature of the music. The absence from both 
works of the strings is significant. 

The Octet, scored for flute, clarinet, two bassoons, two trumpets, and 
a tenor and bass trombone, is modelled partly on the Venetian renais- 
sance style (Gabrieli) and partly on Bach. It consists of three compact 
movements—a Sinfonia (overture), an Air with five variations, and a 


218 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Finale. If the Symphonies of Wind Instruments already contained many 
passages of non-harmonic, linear counterpoint—to be distinguished 
from the technique of short ostinatos and heterophony characteristic 
of Stravinsky's Russian period—the Octet shows the total application 
of this typically twentieth-century Western device. 

Between the Symphonies of Wind Instruments and the Octet Stravin- 
sky composed the one-act opera Мауга (1922) which is in a sense a 
throwback to his Russian style, yet at the same time is influenced by 
his neo-classicistic preoccupation. In it he adopts and adapts to his own 
manner of speech the language of Glinka, Dargomizhsky, and Tchai- 
kovsky and writes in the convention of Russo-Italian opera. The libretto 
is based on a rhymed story, The Little House in Kolomna, by Pushkin— 
Pushkin who in the composer's own words *was the most perfect 
representation of that wonderful line which began with Peter the Great 
and which, by a fortunate alloy, has united the most characteristically 
Russian elements with the spiritual riches of the West . . . As for myself, 
I had always been aware that I had in me the germs of this same 
mentality only needing development, and I subsequently deliberately 
cultivated it'.* This precise statement helps to explain one of the guiding 
aims which Stravinsky pursued after his ‘Russian’ period, and Mavra 
is a characteristic example of this in its resuscitation of the shades of 
Glinka and Tchaikovsky, whose Europeanized nationalism the composer 
compares favourably with the 'doctrinaire aestheticism" of the pure 
Russian nationalists. The opera, dedicated to the memory of Pushkin, 
Glinka, and Tchaikovsky, conveys, despite its comic action, a feeling 
of nostalgia and even bitterness which no doubt reflects Stravinsky's 
mixed feelings about Russia, ancient as well as modern. In style the 
work is an opera buffa consisting of arias, duets, and ensembles, but it 
has no recitatives. 

Stravinsky's strong attraction to the wind ensemble, demonstrated 
in the Symphonies of Wind Instruments, the Octet and Mavra, was 
followed by an equally keen interest in the piano as a solo instrument. 
True, during his previous years he had written some piano pieces, such 
as the Five Easy Pieces (1917) and The Five Fingers (1921), but these 
were intended for his children; at the opposite end stand the Piano Rag- 
Music and the Three Movements from Petrushka (1921) (written for 
Artur Rubinstein) which are virtuoso pieces. In his use of the piano 
as an orchestral instrument he strongly emphasized its percussive aspect 
(Petrushka) developing this to a ne plus ultra in the martellato style of 
the four pianos in Les Noces. In the works of his neo-classical period he 
still exploits the percussive quality, but he takes advantage also of the 

1 Chronicle of my Life, p. 159. 


STRAVINSKY'S NEO-CLASSICISM 219 


legato and cantabile potentialities of the keyboard. This preoccupation 
with the piano led first to the Concerto for piano, wind orchestra, double 
basses and timpani (1924), then the Piano Sonata (1924), the Serenade in 
А (1925) and, after a lapse of four years, the Capriccio for piano and 
orchestra (1929). The characteristic common to these four works is the 
propulsive ‘motoric’ and ‘open’ writing for the piano and, in the first 
three works, the influence of the Baroque harpsichord style, while the 
Capriccio emulated the brilliant virtuoso writing of Weber or Mendels- 
sohn. 

The Piano Concerto was the first work that suggested Stravinsky's 
‘return to Bach’. The first movement is in the style of a Bach toccata 
whose introductory Largo recalls in its measured dotted rhythm Lully 
and the seventeenth-century French overture. The middle movement, a 
Larghissimo, unfolds with a stately tread somewhat reminiscent of 
Vivaldi, while the middle section contains a cantabile of incantatory 
effect, the orchestral melody in 2/4 being accompanied in the treble of 
the piano and the double basses by an implied 3/8 rhythm. The finale 
combines the flavour of eighteenth-century music with fierce syncopa- 
tions which recall remotely the ‘barbaric’ rhythms of Petrushka and 
The Rite of Spring, proving that, despite the adoption of the Apollonian 
principle, Stravinsky's old Dionysian urge could not be altogether sup- 
pressed. The Piano Sonata is noteworthy for the sparse, attenuated 
texture of the first and last movements, mostly in two parts, with the 
treble of the opening Allegro at times thickened out with euphonious 
thirds and sixths. Yet the music as a whole strikes the ear as cold and 
calculated, and almost the only relief is provided by the lyrical charm of 
the middle movement, an Adagietto that is reminiscent, in its rich 
ornamentation and the ostinato-like appoggiaturas of the middle sec- 
tion, of early Beethoven. Though Stravinsky studied Beethoven's piano 
sonatas before his own composition, the latter is entirely different 
in form and in development of thought. The title ‘sonata’ is intended in 
its pre-classical, Baroque, sense denoting music to be played as against 
*cantata', 1.е. music to be sung. The Serenade in А is a far more genial 
and immediately attractive work emulating the style of eighteenth- 
century ‘night music’ and consisting of four short, well-contrasted 
movements. It provides a characteristic example of his concept of ‘polar’ 
or extended tonality, the note A acting as the point of harmonic 
attraction in the completely free flow of keys, and at various times 
assuming the function of tonic, dominant or major and minor third. 

Stravinsky’s neo-classicism assumed a profounder, more spiritual 
aspect after his discovery of the world of Greek antiquity. ‘Discovery’ 
is perhaps not quite the appropriate term since two of his very early 


220 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


works, the song cycle Faun and Shepherdess (1906) and the wordless 
song Pastoral (1908), already proclaim a certain affinity with the 
spiritual landscape of the Hellenic world. But it was not until Stravinsky 
came to live in Paris and felt the influence of Cocteau, Gide, and other 
writers that classical antiquity with its idealistic yet essentially tragic 
view of life—man living in a realm of purity and serenity but pursued 
by a remorseless Fate—exercised a fascination on him, and gave rise to 
a series of important stage-works—the opera-oratorio Oedipus Rex 
(1927), the ballet Apollo Musagetes (1928), the melodrama Perséphone 
(1934), and the ballet Orpheus (1948). 

The choice of the Sophoclean Oedipus as a subject for an opera- 
oratorio was determined by two facts. One was Stravinsky’s desire to 
use a plot known to the generality of spectators, so that their undivided 
attention could be given to the musical treatment. The second and, 
perhaps, more important fact is to be found in an affinity between 
something in Stravinsky’s own personality and the spirit of the Greek 
play, which shows man as the plaything of higher, supernatural forces, 
entangled in the meshes of a snare that gradually closes on him and 
eventually destroys him. This idea of a tragic destiny suspended like the 
sword of Damocles over the human being had in the past given rise to 
three of Stravinsky’s works—Petrushka, The Rite of Spring and 
L'Histoire du soldat—but Oedipus Rex projects it in the most un- 
equivocal, direct terms yet with a striking economy of dramatic gestures. 
What there is of external action is chiefly confined to the latter part of 
the drama—the revelation of the Messenger and Shepherd, the suicide 
of Jocasta, and the self-mutilation of Oedipus. For the greater part 
Stravinsky concentrates on the inner, psychological, action—the fall 
of Oedipus from his height as the proud and self-confident King of 
Thebes into an abyss of misery and tragic guilt. Cocteau’s libretto, which 
is masterly in the compression of the ancient tragedy and impressive in 
the simple dignity of the language, was translated into Latin (by Jean 
Daniélou) which was Stravinsky’s idea. Reading Joergensen’s book St. 
Francis of Assisi he discovered that the saint (whose native language was 
Italian) used French on all solemn occasions; and by analogy he 
decided to have a Latin text, which had the advantage of being ‘a 
medium not dead but turned to stone and so monumentalized as to have 
become immune from all risk of vulgarization'.! 

The static, hieratic, and impersonal character of the drama is already 
emphasized in the ‘alienation’ effect created by its method of produc- 
tion: the stage has no depth, the whole action taking place on one level 
suggesting a bas-relief. With the exception of Tiresias, the Shepherd, and 

1 Chronicle of my Life, p. 210. 


STRAVINSKY'S NEO-CLASSICISM 221 


the Messenger, the characters appear in built-up costumes and masks, 
moving only their heads and arms and producing the effect of living 
statues; their entries and exits are made through trap doors which are 
alternately veiled and unveiled by special curtains; the chorus sits on 
three tiers. To assist the spectators in recalling the detail of the plot, a 
Speaker in evening dress explains, in the vernacular, the action in an 
impersonal, detached way like a lecturer. In his musical design Stravin- 
sky follows the scheme of eighteenth-century opera and oratorio: self- 
contained arias, duets, and choruses which have no thematic con- 
nection and are arranged like stark contiguous blocks of granite, the 
cumulative effect of which is one of awe-inspiring monumentality. 
The idiom is for the most part diatonic, another means by which the 
composer achieves directness, immediacy, and simplicity of expression. 
The pervasive rigidity of the rhythmic language corresponds with the 
ritualistic character of the tragedy, and it 15 only at the peripeteia—the 
highly dramatic scene between the Messenger and the Shepherd—that 
the cast-iron regularity of the metrical scheme is abandoned in favour of 
a more irregular treatment, with asymmetrical phrases reminiscent of the 
composer's ‘Russian’ style. 

The musical material of Oedipus Rex is derived from widely differing 
sources—Handel and Bach, Verdi and the early nineteenth-century 
Italian opera, and Russian music. But these heterogeneous materials 
are moulded together into a new and entirely original synthesis in which 
not a single bar could have been written by any other composer but 
Stravinsky. Oedipus is the sole character who is shown in a consistent 
psychological development—from the first aria *Liberi, vos liberabo', 
expressing his self-assurance and supreme pride in the stiffness of the 
vocal writing which is marked by a high tessitura and oriental melismas, 
to the final *Natus sum quo nefastum est' where at last he recognizes the 
enormity of his guilt and reaches the point of utter humiliation. His 
thrice-repeated phrase rotating round the constituent notes of the B 
minor chord (accompanied by shuddering strings and woodwind) and 
his last ‘Lux facta est’ are characteristic examples of Stravinsky's use 
of the simplest harmonic formulae in the service of an intense dramatic 
expression. These and other instances in this work demonstrate the 
extent to which he revitalizes seemingly outworn harmonic devices and 
achieves a strikingly fresh effect by divesting them of their original tonal 
function and grafting them on the living tissues of his personal style. 

If Oedipus Rex was an illustration of the tragic aspect of Greek 
mythology, the ballet Аройо Musagetes, written a year after, represents 
the opposite world—the world of Arcadian bliss, unruffled calm, and 
supreme serenity and beauty. It may be taken as an allegory on the 


16 


227 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


theme of artistic creation: Apollo inspires the muses and leads them to a 
final apotheosis on Mount Parnassus. The composer intended the work 
as a ballet blanc, devoid of all psychological, narrative and expressive 
interest, with an abstract choreography based on the traditional 
classical forms of pas d'action, pas de deux, and variation and using 
simple scenery and monochrome antique costumes. Musically this 
finds its counterpart in a predominantly diatonic style from which all 
harmonic tension and conflict are banished, and the effect of extreme 
calm and purity is heightened by the use of a string orchestra to which 
the composer returned after twelve years of comparative neglect. In his 
Chronicle of my Life! he writes of the utter delight he experienced in 
employing the string medium whose original purpose, he says, was the 
cultivation of canto, and he wished to compose music in which “еуегу- 
thing evolved round the melodic principle’ making ‘the multi-sonorous 
euphony of the strings penetrate even the furthest fibre of the poly- 
phonic web'. 

Perséphone, for which André Gide supplied the text, bears the sub- 
title ‘melodrama’, a description not to be understood in the traditional 
sense of music accompanying a spoken text but as indicating a hybrid 
between opera and ballet, which combines music, singing, dancing, 
miming, and spoken recitative. The work is divided into three parts— 
The Abduction of Perséphone, Persephone in the Underworld and The 
Rebirth of Persephone. In the treatment of the text Stravinsky is seen to 
apply his by now customary method of repeating words and whole 
sentences in order to build up his musical phrases, a striking instance of 
which is the four-part chorus of Nymphs, Danaids, and Shades in the 
second part. Similarly, to suit his rhythmic patterns he broke up the 
words into syllables, to the discomfiture of Gide who, rightly, felt that 
this method did violence to his prosody. This aerated style of writing, 
which Stravinsky had already used in Oedipus Rex, lends the music an 
extraordinary limpidity and transparency and the idiom is in an even 
more pronounced diatonicism than in Apollo Musagetes, while the 
rhythms are remarkably fluent and subtle but frequently of a regular 
pattern, entire sections adhering to the identical metre. The melodic 
invention is distinguished by a natural, unforced flow and almost the 
whole of the work is marked by a remarkable tenderness of expression. 

Stravinsky's love and admiration for Tchaikovsky, to which he first 
gave expression in the opera Mavra, prompted him six years later to a 
ballet Le Baiser de la Fée (The Fairy's Kiss) (1928), composed in com- 
memoration of the thirty-fifth anniversary of the composer's death; the 
choreographic action is taken from Hans Andersen's The Ice Maiden. 

1 pp. 221-2. 


THE 'SYMPHONY OF PSALMS' AND LATER WORKS 223 


While in Mavra it is only the style which is reminiscent of Tchaikovsky, 
the themes being Stravinsky's own, in the ballet the material is borrowed 
in toto from the older composer and treated in something like the 
manner of Pulcinella yet without the sense of parody and caricature 
characteristic of that earlier score. ТЛе Fairy's Kiss foreshadows another 
and far more successful work written by Stravinsky eight years later. 
This was Jeu de cartes (Card Game) (1936), a ballet in *Three Deals', 
whose choreographic plot revolves round a game of poker, with the 
various characters represented by the chief cards. In the first two deals 
the Joker, because of his ability to become any desired card, stirs 
up all sorts of trouble, but in the last deal he is beaten by a Royal Flush 
of Hearts which puts an end to his malice and knavery. In a sense this 
ballet forms a counterpart to L’ Histoire du soldat which is also con- 
cerned with the force of evil but closes with the triumph of the devil, 
while Jeu de cartes ends with his defeat. Each of the three ‘deals’, which 
follow without a break, is prefaced by the same 'processional' music, 
intended to suggest the shuffling of the card pack; each of the main 
movements, however, bears a distinct character of its own and although 
the musical ingredients derive from a great variety of sources—Haydn 
and Weber, Delibes and Tchaikovsky, Johann Strauss and Ravel—the 
result is unmistakably original. 


THE ‘SYMPHONY OF PSALMS’ AND LATER WORKS 


We must now retrace our steps to the year 1930 to examine the first 
major work in which Stravinsky gave eloquent expression to his deep- 
seated religious feelings. The Symphony of Psalms originated in a 
request by Sergey Kusevitsky to compose a work in celebration of the 
fiftieth anniversary of the foundation of the Boston Symphony Orches- 
tra; but a profounder and more exalted impulse is revealed in the 
inscription ‘composed to the Glory of God’. In his Poetics of Music! 
the composer ascribed to music the aim of promoting ‘a communion 
with our fellow-men and with the Supreme Being’ which gives a 
clear indication of the spiritual significance he perceived in music in 
general and in sacred music in particular. 

In choosing three of the Psalms from the Vulgate (38: 13-14; 39: 2-4; 
150) Stravinsky was concerned with man’s relation to God in the spirit 
of the Old Testament which, in contrast to the Christian doctrine, knows 
of no redemption but only offers the consolation of hope and comfort in 
the struggle and sufferings of life on this earth, in response to man’s 
humble prayer and obedience. It is to this conception, compounded of 
utter humility and severity, that the work gives unique expression, 

1 Harvard, 1947, p. 18. 


224 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


avoiding all undue dramatization of the text, toning down the impulse 
of all subjective feeling, aspiring to a devout contemplation of divine 
mystery and thus raising its message to the plane of universality. The 
work is of an extraordinary compactness and terseness, lasting about 
twenty minutes; and although there is a marked contrast in the structure 
and texture of the three linked movements—‘Prayer’ (Prelude), ‘Thanks- 
giving’ (Double Fugue) and ‘Hymn of Praise’ (Symphonic Allegro)—it 
yet conveys the distinct feeling of a higher, spiritual, unity which is 
largely achieved by the application of Stravinsky’s principle of simi- 
larity. What contributes to this impression is a motive of two interlinked 
minor thirds which in various permutations recurs in all three move- 
ments, yet does not possess the role of germ cell from which, as in Les 
Noces, all the essential material is evolved. In the opening movement 
this basic motif is to be found mainly as an accompanying figure, in the 
second movement it forms part of the subject of the instrumental fugue, 
with the second minor third now inverted to a major sixth, and in the 
finale it appears in the opening chorus ‘Laudate Dominum’: 


Ex. 99 
(1) 


In the ‘Prayer’ the contrapuntal burden is carried Бу Ше orchestra 
while the choral writing is predominantly homophonic and is marked, 
first, by small intervals, as in the subdued *Exaudi orationem meum, 
Domine’, sung on the two adjacent notes E and Е; but, as the prayer 
grows more fervent, the intervallic steps become wider. The *Thanks- 
giving’ is fugal, showing the composer at the height of his polyphonic 
ingenuity, especially in the manner in which he achieves a close yet 


THE ‘SYMPHONY OF PSALMS’ AND LATER WORKS 225 


entirely effortless interlocking of the instrumental and choral subjects. 
The most extensive piece is the ‘Hymn of Praise’ which is less con- 
centrated in the structure than the two preceding movements but of a 
more varied texture. The gulf that separates Stravinsky from musicians 
of the past centuries is seen in the way in which he sets the opening 
line, ‘Alleluia. Laudate Dominum" (Ex. 99(iii) ). Where other composers 
would have burst forth in loud jubilations he preserves an attitude of 
utter calm and humility and even where the text invites a pictorial 
image, as in ‘Laudate Eum in cymbalis', the music remains subdued 
and inward-glancing. This is not to say that the ‘Hymn of Praise’ is 
entirely devoid of a feeling of joyful exultation and ecstatic fervour, but 
its general tendency is towards the expression of tranquil, sublimated 
emotions, and, significantly, it closes with the same ethereal choral 
phrase with which it began. 

After the Symphony of Psalms Stravinsky turned to a medium which, 
except for a passing excursion in L’Histoire du soldat, he had not 
attempted before—music for the solo violin. The immediate cause for 
this awakening of a new interest was his acquaintance in early 1931 
with the violinist Samuel Dushkin, for whom he composed the Violin 
Concerto (1931) and the Duo Concertante for violin and piano (1932). 
This task afforded the composer the opportunity for a more serious 
study of the expressive and virtuosic possibilities of the violin, while 
Dushkin collaborated with him in the matter of technical details. 
The Concerto which, following established precedents, is in the 
‘violinistic’ key of D, and shows in general character a certain affinity with 
the Capriccio for piano and orchestra. Of the four movements, the 
opening Toccata and the closing Capriccio are modelled on eighteenth- 
century patterns (Bach and Mozart) and represent Spie/musik in the 
true sense, displaying the composer’s delight in the sheer manipulation 
of form, design, and texture for their own, purely musical sakes. In 
the Toccata soloist and orchestra have interchangeable material. By 
contrast, the Aria I and II are in an expressive lyrical vein, with much 
sustained cantabile writing, and in the Aria II the melodic line dissolves 
into rich yet most supple filigrees (runs and arpeggios). Roman Vlad! 
perceives in the two arias Stravinsky's wistful longing for the full 
expressiveness that music once possessed, an abandonment to a 
nostalgic ‘recherche du temps perdu’. In the Duo Stravinsky explores 
the potentialities of the violin in conjunction with the piano. The work 
is a small-scale illustration of Stravinsky’s attraction to the ancient 
Hellenic world, for in spirit and formal structure it is intended as a 
musical counterpart to the pastoral poetry and scholarly art of antiquity, 


1 Stravinsky (Turin, 1958; English translation 1960), pp. 115-16. 


226 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


with its enforcement of strict rules and an iron discipline on lyrical 
composition. A common theme runs through all the five movements, 
lending them coherence and unity; it first occurs in the fifteenth bar of 
the opening Cantilena and is given its most articulate shape in the slow 
Eclogue II, the central piece of the work round which the other move- 
ments are symmetrically grouped. The tranquil, serene lyricism of this 
Eclogue is admirably matched by the bucolic charm of Eclogue I, which 
opens with a delightful bagpipe episode, the bass of the piano playing a 
patterned drone-like ostinato while the treble echoes in canon the chanter 
tune of the violin, the note A being sustained as an inner pedal. The 
brilliant Gigue is a perpetuo mobile and the final Dithyrambe is note- 
worthy for its gently undulating piano phrases in three and four parts 
and an expressive violin cantilena which reaches its climax in a passage 
of truly ecstatic fervour. 

In the Concerto for Two Pianos (1935) Stravinsky displays the same 
delight in making music for its own sake, the same enjoyment in 
solving specific stylistic and technical problems as characterized the 
Violin Concerto and the Duo Concertante. It is on an imposing scale 
consisting of four movements, of which the first is remarkable for the 
trenchant force of its rhythmic invention and its extraordinary dynamic 
impulse. The Notturno excels in delicate lyrical writing, the ensuing 
Four Variations are a locus classicus of Ше composer’s unremitting 
search for novel construction, and in the four-part fugue all existing 
contrapuntal devices are brought into vivid play. As in Mozart’s 
Sonata for Two Pianos (K.448), which served Stravinsky as a point of 
departure, the two instruments share the same material, but the 
twentieth-century musician goes a step further in achieving a rare 


Ex. 100 
() 


[^ v] 


THE ‘SYMPHONY OF PSALMS’ AND LATER WORKS 227 


degree of taut, close-knit texture, frequently cutting up a thematic idea 
and dividing it between the two instruments, a style of writing that 
creates the impression of a contest between two wrestlers (Ex. 100). 

A feature of the Four Variations is the absence of an explicitly stated 
theme, the first variation starting with a zigzagging motive of wide 
intervals which only gradually takes on a firmer outline until it achieves 
its fully developed shape as the vigorous subject of the final fugue. And 
for all the wide range of keys touched in the course of the music there is, 
nevertheless, a strong pull towards the note E as the tonal pole of the 
entire work, which opens and closes in a clear E major. 


228 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


In the Dumbarton Oaks Concerto in,E flat (1938), so called after the 
place in America where it was composed, Stravinsky aimed at a work 
which, on his own admission, follows the style of Bach's Brandenburg 
Concertos. Yet his indebtedness to Bach is largely confined to emulating 
the general shapes of themes and to the choice of a chamber orchestra 
of fifteen players. The strings and wind are treated in concertante style 
but, unlike his model, Stravinsky makes no attempt at writing a proper 
concertino. 

In concluding this section we may attempt a brief definition of the 
general significance of Stravinsky's achievement during the period of 
the inter-war years. A number of points emerge. In the first place, 
Stravinsky was one of the first to exploit Debussy's rediscovery of the 
primary nature of music, i.e. the meaningful organization of sound, and 
liberate it from the extra-musical superstructure of psychological, 
philosophical, and metaphysical notions and ideas imposed on it by the 
romantics. He saw his role in terms of the medieval master-craftsman 
for whom his material was something to be worked upon according to 
its inherent properties—hence Stravinsky's concentration on texture, 
design, harmony, counterpoint, and instrumentation. Іп this he was 
guided by a most highly developed sense of order, discipline, and 
adjustment of means to ends—the principal characteristics of the 
Apollonian spirit at whose shrine Stravinsky worshipped after his 
Dionysian ‘Russian’ period. Few modern composers have displayed 
greater technical ingenuity, skili, and resourcefulness in handling and 
moulding the raw material of music. If, for a time, Stravinsky adopted 
an almost inhuman, coldly objective detachment from all emotion and 
feeling, this was only a passing phase. While it lasted Stravinsky showed 
a marked predilection for an orchestral style sec, trenchant and 
coruscating in which, significantly, wind instruments, piano, and 
percussion occupied the forefront while the expressive strings were 
relegated to the background. Later works such as Oedipus Rex, the 
Symphony of Psalms, and Perséphone, while ostensibly aiming at an 
*alienation', in fact refute Stravinsky's theoretical axiom that music 
was incapable of expressing anything at all and that construction was 
both its meaning and scope. 

Neo-classicism appears, in retrospect, to have sprung from a deep- 
rooted urge to fuse the East with the West by bringing Western styles 
and techniques to bear upon an intrinsically Eastern or Russian mode 
of thought. (Bartók pursued a similar aim after his discovery of Hun- 
garian folk-song, see p. 275.) For example, the constant reiteration of 
short melodic-rhythmic fragments, though much attenuated in his neo- 
classical works, derives ultimately from Stravinsky's Russian heritage 


SERGEY PROKOFYEV 229 


and was grafted on to devices common to European music of several 
centuries. As with Bartók, this process of assimilation resulted in 
compositions of a highly original order, which present a panoramic 
view of musical history reinterpreted and used as a potent fertilizer. 
Within the neo-classical framework Stravinsky developed an astonishing 
diversity of styles and manners of projection, due to his particular 
approach which considers each new work as a new adventure, a novel 
and isolated creation demanding a novel and appropriate mode of 
treatment. Stravinsky is the perfect stylist, in the sense that to each new 
work he brought a wholly apt and commensurate technical apparatus. 
In spite of the kaleidoscopic changes in his style, changes comparable 
to Picasso's zigzagging evolution, a basic unity can be discerned in 
Stravinsky's music of the inter-war years, the corollary of a fully 
integrated artistic personality which pervades and holds together the 
complex and variegated character of his utterance. Lastly, he stands 
almost unrivalled in the harnessing of a crystal-clear, penetrating, and 
highly analytical intellect strictly controlling the conscious part of his 
creative processes, to a rich, vivid, and most versatile imagination. 
A whole generation of younger composers benefited from his widening 
of the horizon, and this was particularly true of the French musicians 
of the 1920s whose country Stravinsky had adopted as his physical and 
spiritual home and where his personality and influence were, therefore, 
most strongly felt. 


SERGEY PROKOFYEV 

Born a decade later than Stravinsky and in many ways his opposite 
in musical character, Sergey Prokofyev was the enfant terrible of his 
generation at the St. Petersburg Conservatoire, where he studied com- 
position with Lyadov and orchestration with Rimsky-Korsakov and 
became, under Anna Esipova's tuition, a brilliant pianist of a kind 
hitherto unknown, an athlete rather than a poet of the keyboard. An 
interest in neo-classicism was first prompted by the visit of Max Reger 
to St. Petersburg in 1906, when he conducted a programme of his own 
works including the Serenade in G major, Op. 95. Prokofyev's Ten 
Pieces for Piano (1908—13) reflect this interest, and so even more clearly 
does the eighteenth-century pastiche of his Classical Symphony (1916- 
1917). He also admired the orchestral music of Richard Strauss and 
Aleksandr Skryabin, making a piano transcription of the first movement 
of Le Divin Poéme; and both admirations are reflected in his own First 
Piano Concerto (1912). The music of Debussy and the early work of 
Stravinsky had little attraction for the young Prokofyev, whose extra- 
ordinary musical fertility and rough, irreverent high spirits resembled 


230 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


those of the young Hindemith and made him all but totally unacceptable 
to the cultured and sophisticated members of the Dyagilev set. Even so 
Prokofyev's Scythian Suite (Ala and Lolly) (1914-15) was clearly 
prompted by the same contemporary movement of interest in the 
Russian past (Skifstvo) as Stravinsky's Rite of Spring, which it rivals in 
aggressive ‘modernism’ of harmony and rhythm. There was to be little 
in Prokofyev's music comparable with the French sophistication, and 
later the erudite historicism, which played civilizing, westernizing roles 
in Stravinsky's musical development. 

The role of iconoclast and rebel was indeed an important part of the 
young Prokofyev's image of himself, and his reputation as composer 
and performer was predominantly one of ruthlessness and physical 
energy. Certainly the public found these qualities in the first three piano 
concertos and the 7occata, overshadowing the delicate and very per- 
sonal Sarcasms and Visions fugitives and the almost Mussorgskian 
Tales of the Old Grandmother, all piano works written between 1912 and 
1918. His first ballet Shut (The Buffoon, 1920), the cantata Semero ikh 
(Seven, they are seven, 1918) and two of his first three operas reveal 
how deeply Prokofyev was rooted in the literary and artistic world of 
pre-revolutionary Russia. The cantata is the setting of a hermetic 
text by Konstantin Balmont, while the libretto of Lyubov k trem apel- 
sinam (Love for Three Oranges, 1921) is taken from the sardonic and 
fantastic Gozzi and that of Ognenniy Angel (The Flaming Angel, 
composed 1919—27 but first performed 1955) from the 'decadent 
Symbolist poet Valery Bryusov. Only his /grok (The Gambler, 1915-16) 
was based on a realistic story by Dostoevsky. 

Prokofyev's musical language during the years which he spent outside 
Russia (1917—32), though even more strongly diatonic in basic character 
than Stravinsky's, was marked by jagged tonal shifts, which often 
replaced traditional modulation; wide-spaced and deliberately angular 
melodic patterns; aggressive harmonies formed by chromatic added 
notes often with distinct bitonal implications; and most notably by 
rhythms of whirlwind force and motoric regularity and persistence. The 
devil-may-care humour, mordant satire and sheer physical excitement 
of this music have much in common with those of the young Hindemith, 
and also of the younger generation of French composers (Milhaud, 
Poulenc) who were among his friends during the years he spent in Paris 
(1923-32). Much later in life he was to write “Тһе cardinal virtue (or 
sin, if you like) of my life has been the search for an original musical 
language, a musical language of my own. I detest imitation; I detest 
hackneyed methods. I always want to be myself.’' That the lyrical gift 

! Quoted in T. V. Nestyev, Prokofyev (Moscow, 1957, Eng. trans. 1961, p. 466). 


THE OLDER GENERATION IN FRANCE 231 


which he cultivated consciously after his return to Russia in 1932 was 
from the beginning part of his musical character is clear from the First 
Violin Concerto (1916-17) and the Third Piano Concerto (1917-21) 
among many other works. But it was obscured by the consciously harsh, 
‘constructivist’ character of many of the works of Prokofyev's Parisian 
days—the Second Symphony and the ballet Le Pas d'acier (both 1925) 
and the Third Symphony (1929) in which the composer reworked themes 
from the hysterical Flaming Angel. Prokofyev was in touch with Russian 
musical life for almost ten years before he returned finally to live there. 
There were performances of his works in Russia as early as 1923 and he 
paid a visit there in 1927. As he became increasingly disillusioned with 
what seemed to him the artificial nature and narrowly restricted appeal 
of contemporary music in Western Europe, it was natural that he should 
become more aware of the ties that bound him to his native country and 
the possibilities that it promised him as a composer.! 


THE OLDER GENERATION IN FRANCE 


The musical scene in France between the two wars presented a confusing 
but extremely vivid picture. There was the old guard consisting of 
Fauré, Roussel, and Ravel, who were the last representatives of 
Че troisième аре d’or’ in French music and, with the exception of 
Fauré, belonged to the same generation as Debussy. In strong opposi- 
tion ‘Les Nouveaux Jeunes’ were inspired by new ideals and, under the 
collective name of ‘Les Six’, at first rallied round the figure of the ageing 
Satie making him their involuntary chef d'école. Lastly, in the mid- 
1930s a new group, “Га Jeune France’, in its turn reacted against the 
supposedly excessive cosmopolitanism of the previous generation and 
sought a regeneration of French music on national lines. 

Gabriel Fauré (1845-1924) was seventy-three when the war ended, 
but in his few remaining years he composed six chamber works, chamber 
music having always been a particular interest, a few piano composi- 
tions, and three song cycles. Ап exclusive and reticent musician, 
Fauré had always been inclined to write for himself and a select circle 
of connoisseurs rather than for the large public, and during his last years 
he became even more discreet, subtle, and elliptical in his utterance. 
This latest development was partly due to his total deafness which may 
account for the fact—as it does in the case of Beethoven—that Fauré's 
last works reveal a marked attitude of self-communion and a tendency 
to explore to its furthest limits an esoteric and ascetic manner of 
expression. Аз his English biographer says,? in the chamber works ‘the 


1 On his career in the Soviet Union, see Chapter VIII. 
2 Norman Suckling, Fauré (London, 1946), p. 118. 


232 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


opening movements are vigorous without protestation, their finales 
without effervescence; and the intervening andantes in particular distil 
an atmosphere of peaceful intensity where every vibration is significant 
without being insistent, by means of a sensitiveness of Ше mind's ear’. 
Fauré's textures become extraordinarily sparse and transparent, the 
harmony achieves a new simplicity, and the thematic material shows a 
remarkable condensation. True, the long flowing melodic lines so 
characteristic of his previous style still occur here and there, as in the 
slow movements of the First Cello Sonata (1918) or the opening 
movement of the Second Cello Sonata (1921), but they are more often 
than not replaced by cellular melodic units, as in the second movement of 
the Piano Quintet (1921) and the String Quartet (1924), a transcendental 
work and, like his late songs, tenuous and elusive. In the works for 
piano, such as the Fantaisie (1919), the keyboard writing is so designed 
as to achieve a measure of coherence by means of such typically 
Fauréan devices as scale passages, crossing and recrossing of the melodic 
lines and harp-like arpeggios; but there are exceptions as seen in the 
antiphonal treatment of piano and strings in the Trio (1924) which for 
a composer approaching his eightieth year is remarkably fresh and 
vivid. In the two song cycles, Mirages (1919) and L'Horizon chimérique 
(1922), a setting of four poems by Jean de la Ville de Mirmont, Fauré 
captures the evocative power of the text and the significance of individual 
sentences with an unexampled artistry. These songs exhale an air of 
aristocratic refinement and discretion, a civilized tranquillity in which 
the underlying emotions are but vaguely sensed. Together with Roussel, 
Fauré was the last great traditionalist of French music, embodying the 
spirit of French culture in its purest form and combining it with a 
classical serenity of an almost Hellenic quality. 

Albert Roussel (1869-1937), who was a pupil of Vincent d'Indy and 
for many years teacher at the Schola Cantorum, allied the serious out- 
look of his master and the technical solidity and discipline of that 
School to a sensibility that, though increasingly affected by the new 
post-war spirit, never lost its refined and poetic qualities. Roussel was a 
fastidious and highly self-critical artist who aimed “о achieve music 
which is self-contained, music which is divorced from any illustrative 
and descriptive elements and is free from any localization in place... 
Far from wishing to write descriptive music, I constantly try to obliter- 
ate from my mind the memory of objects and forms capable of being 
translated into musical terms. I want my music to be nothing but 
music.’ Whether Rousse] adopted this austere attitude under the 
influence of the neo-classical Stravinsky or followed his own inclination 

1 Quoted in Arthur Hoérée, Albert Roussel (Paris, 1937), p. 66. 


THE OLDER GENERATION IN FRANCE 233 


the fact is that, with advancing maturity, he was more and more given 
to an abstract conception rare in a French musician, yet his individual 
style remained strongly marked and manifested itself, above all, in a 
highly developed rhythmic sense and a natural ability for polyphonic 
writing. The majority of his post-war works reveal a rough-hewn, 
granite-like quality which is in sharp contrast to the subtle and seductive 
textures of Debussy or the highly ornate patterns of Ravel; and they 
possess a directness of expression that makes its points without circum- 
locution. Roussel was an unsentimental, powerful figure in whom there 
was an element of the peasant and the long, piercing glance of the sailor. 

His descriptive Pour une Féte de printemps (1920), with its romantic 
and impressionistic features, was the last work of Roussel's to look back 
to his previous style, and it is only the bitonal clash of the opening 
chord that gives an indication of the direction in which the composer 
was to progress. With the Second Symphony (1921) begins what has 
been termed his ‘classical’ period when he discarded earlier influences, 
chiefly Debussy's. This was followed by the Suite in F (1926) and 
the Concerto for Small Orchestra (1927), both of which are examples 
of Roussel's growing realization of his mature self which found com- 
plete and masterful expression in the Third Symphony (1930) and the 
Fourth Symphony (1934). The Third manipulates a basic five-note 
theme with an extraordinary resourcefulness, and the entire work is 
remarkable for its extreme formal compression, the taut, clear-cut 
texture and its tremendous rhythmic force—the embodiment, it would 
seem, of sheer energy. The Fourth Symphony is more detached in out- 
look and its salient stylistic features are to be seen in a sinewy counter- 
point and bare astringent harmonies (tritone, major sevenths, and the 
chord of the eleventh) which are the result of the uncompromising part- 
writing. The robust vitality of the opening movement is offset by the 
poetic quality of the slow movement for which is reserved the main 
emotional climax of the work, while the finale brings the crowning 
affirmation of the music's predominant spirit. 

If Roussel’s later orchestral works are characteristic of the classical 
bent of his art, his stage works pursue a more varied stylistic course. 
Раатйуай, an opera-ballet in two acts (1918), to a libretto by Louis 
Laloy, deals with a barbaric episode in Indian history of the thirteenth 
century and is based on Hindu scales employed both in the melody and 
harmony and organically integrated into Roussel's own individual 
style, which is expanded and enriched through his free conception of 
modality and a more varied and more supple application of rhythmic 
patterns. In the classical ballet, La Naissance de la lyre (1924), which has 
a scenario after Sophocles, the chief basis of the material is the ancient 


234 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Greek modes. A second ballet with а classical subject, Bacchus et 
Ariane (1930), is characterized by a vigorous rhythmic style and clear, 
sharply defined melodic lines, while the orchestration echoes at times 
the exquisite lights and shades of Debussy's impressionism. Aeneas 
(1935), the composer's third classical ballet, is exceptional in that it is 
more choral than orchestral music, and contains a percussion part for 
piano in which Roussel pays homage to Stravinsky. He always wanted 
to write an opera in a light and humorous vein which he achieved in the 
delightful though perhaps rather risqué comedy, Le Testament de la 
Tante Caroline (1933), which stands apart from the canon of his works. 

Roussel showed no natural feeling for the piano. His Piano Concerto 
(1927) is an introspective and technically forbidding work in which the 
solo instrument largely explores the lower registers. The Cello Concerto 
(1936), his last orchestral composition, resembles a concerto grosso in 
that the orchestra is more individually treated than is usual in a con- 
certo. Of Roussel's post-war chamber music, the String Quartet (1932) 
and String Trio (1937)! stand out for the combination of a taut logical 
thought with a fastidious taste and a highly refined sensibility. 

Roussel’s chief achievement lies in the solidity and distinction of his 
craftmanship with which he upheld the precepts of the Schola Can- 
torum; in his purely musical approach to matters of form, design, and 
texture, and in the way in which, emerging from the impressionism of 
his early and middle periods, he fashioned a powerful personal style 
that is deeply rooted in the time-honoured traditions of French culture 
and art. Like Fauré, he belonged to what has been aptly called ‘the 
silver age of Latin civilization',? but being a younger man than Fauré, 
he survived into a time which saw the dissociation and even disruption 
of traditional values. 

Maurice Ravel (1875-1937), the youngest member of the older 
generation, continued to enjoy the world-wide reputation which he had 
established for himself by his pre-war works,? but in his own country he 
was entirely out of favour with the young group of “Les Six’ who re- 
proached his music for excessive refinement and ornateness, regarding 
him as an exponent of obsolete aesthetics. Ravel felt a strong resentment 
at losing his position among the leaders of musical fashion and strove 
hard towards a transformation and even rejuvenation of his style and 
technical methods. The results, however, were not altogether convincing. 
Ravel had done his best work before 1920 and, although his extra- 
ordinary refinement of taste, his perfect sense of form and his most 


1 From which the first movement is recorded in The History of Music in Sound, x. 
? Martin Cooper, French Music. From the death of Berlioz to the death of Fauré (London, 
1951), p. 214. 3 See Chapter II, particularly pp. 101—5. 


THE OLDER GENERATION IN FRANCE 235 


fastidious and brilliant workmanship were as much in evidence as of 
old, his creative energy now began to show tell-tale signs of impatience, 
strain, and fatigue. As one of his French biographers implies,! Ravel 
was touching the boundaries of his nature—boundaries around which 
the very perfection of his technique had created all but impassable 
barriers. This is already noticeable in the first of his post-war composi- 
tions, the choreographic poem La Valse (1920), an evocation of the 
atmosphere of a Second Empire ballroom in which an apotheosis of 
the Viennese waltz is linked with the suggestion of a fantastic whirl of 
destiny. Yet the febrile character of this danse macabre really sprang 
from Ravel’s dissatisfaction with himself and the work is hardly more 
than a brilliant technical essay in veiled pastiche, of which he was a 
master unrivalled in skill and resourcefulness. In the Sonata for Violin 
and Cello (1922), which is a cyclic work, the music is reduced to 
essentials, the composer almost completely abandoning tbe fascination 
of harmony and concentrating on purely melodic writing, the effect 
of which is often harsh and angular. In the Sonata for Violin and 
Piano (1927) the two instruments share some of the material, but Ravel 
emphasizes their fundamental incompatibility by making them not co- 
operate, but contrast with each other. The second movement is a blues 
and shows the composer following the contemporary fashion for jazz. 
The popular Bolero (1928), originally conceived as a ballet for Ida 
Rubinstein, is a tour de force in the achievement of a long orchestral 
crescendo which becomes more highly coloured with each return of 
the two themes; there is no modulation and no development but a 
persistent reiteration of the same ideas. The Piano Concerto in G 
major and the Piano Concerto for the Left Hand (both 1931) show a 
distinct contrast in conception. The former is in the style of a magnified 
chamber concerto (Ravel originally intended to call it a divertissement) 
while the latter, almost in defiance of the limitation imposed on him by 
having to write a solo part for the left hand only (for the one-armed 
Viennese pianist, Paul Wittgenstein), is marked by a highly elaborate 
virtuoso element. In both works the influence of jazz is conspicuous, 
but through the light-hearted G major Concerto there also stalk the 
ghosts of Couperin, Scarlatti, Mozart, and Fauré. In the left-hand 
Concerto, which is a far more inspired and more forceful work, it is 
Liszt and the sombre Spain of Mérimée whose echoes are heard. 

It was largely in the opera, L'Enfant et les sortiléges (1925), to 
Colette's fantastic and exquisite children's tale, that Ravel succeeded 
in recapturing something of his former self and rediscovered something 
of the enchanting world of his first opera L'Heure espagnole. In the 

1 Roland Manuel, Maurice Ravel (Paris, 1938; Eng. ed. London, 1947), p. 86. 


236 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


music of the Princess there appears again that note of shy, tender, 
yet sophisticated lyricism by which the composer always attempted 
to conquer his innate reluctance to express natural, full-blooded 
sentiments; and each of the characters—animals and pieces of furni- 
ture—is succincdy and most aptly delineated: the sardonic humour 
of the Teapot in ironic jazz rhythms, the malevolence of the Little Old 
Man in short acidulated phrases and the mewing of the Two Cats and 
the croaking of the Frogs, in imitative music of the most brilliant inven- 
tion. The vocal writing in this opera is of a remarkable flexibility and 
predominates over the orchestra which, for all its virtuosity, is chiefly 
characterized by light, aerated sonorities. Thematic development is all 
but abandoned and the continuity of the recitative is frequently broken 
up by arioso passages. Except for the music of the Princess, L’Enfant 
is perhaps Ravel’s furthest attempt to rid himself of all subjective feeling 
and portray the world of things as they appear in objective reality. 


ERIK SATIE AND ‘LES SIX’ 


While Ravel tried to adapt himself to the spirit of the 1920s, no such 
adjustment was needed by Erik Satie (1866-1925), the only member of 
Debussy’s generation who all his life had held high the flag of heterodoxy 
and rebellion. As a composer Satie was a minor figure, the intrinsic value 
of whose work is far outstripped by his significance as a prompter of 
tendencies which younger composers moulded into an artistic creed.! 
He was an eccentric, an intellectual clown, a deliberate hoaxer and 
mystifier—in short, an enfant terrible who under his clown’s mask con- 
cealed, however, a certain child-like tenderness and gentle melancholy. 
Throughout his life he had a horror of the style of the Establishment, 
which was synonymous for him with stagnant and staid academic 
authority. Satie thus fitted ideally into those Parisian circles before and 
after the First World War in which "l'esprit frondeur'—the Frenchman's 
instinctive resistance to all authority—was assiduously cultivated. 

In 1915 Satie met the young Cocteau who combined in his person the 
roles of poet, dramatist, painter, and animator and leader of the 
literary avant-garde, and this encounter marked the beginning of 
Satie's belated fame. For Cocteau found in his music all those elements 
that corresponded with his own novel aesthetic ideals—simplicity, 
brevity, ironic humour, unromantic melancholy, and a complete absence 
of those ‘half-lights, muslin, enervating charms and scents’ with which 
he reproached the symbolists and impressionists. Cocteau decided to 
collaborate with Satie and the fruit of their united efforts was the 
ballet Parade (1916), for which the poet wrote the scenario and Picasso 


! For Satie's earlier music see p. 93. 


ERIK SATIE AND ‘LES SIX" 237 


designed the sets. The scene is a booth at a fair before which perform a 
Chinese juggler, a couple of acrobats, a young American girl who is a 
parody of the type then made popular by transatlantic films, and three 
showmen or managers whose antics give a foretaste of what the public 
are to see inside. The audience and the majority of critics were shocked 
by what they supposed to be a manifestation of Cubism in the choreo- 
graphy and the music; yet in reality the latter is disconcertingly simple 
and naive, consisting mostly of two-bar phrases which are monotonously 
repeated over and over again. The acrobats perform to circus music, the 
first manager is given a theme evocative of the bustle of a fairground, 
and jazz music accompanies the incredible feats executed by the 
American girl. The orchestra contains such ‘instruments’ as а type- 
writer, a revolver, a ship's siren, a Morse apparatus and the like, which 
were added by Satie partly to give the work an exaggerated realistic 
touch and partly to raise the hackles of the audience. Whatever the 
intrinsic value of Satie's music, Parade was certainly a landmark in the 
history of the modern theatre. The ballet Mercure (1924), subtitled 
‘Poses plastiques', for which Picasso again provided the décor, deals 
with an adventure of the Greek god and Satie's contribution repre- 
sents only a subordinate part of the spectacle—the work was in fact 
called “а painter's ballet'. The music is rather academic in treatment, 
though it recalls the atmosphere of the circus and the music-hall which 
was in deliberate contrast to the high degree of ingeniousness and 
sophistication shown by Picasso's scenic construction. The surrealistic 
ballet Reláche (1924), to a scenario by Francis Picabia, carries to what 
then seemed the limit, the spirit of aesthetic nihilism, the incon- 
sequential and crazy plot being seemingly devoid of all intelligible 
meaning. For this Satie solemnly wrote a series of serious and dry 
dance movements which are, however, occasionally relieved by popular 
themes selected for their “еуосайуе” quality. 

One of Satie's principal aims was to achieve authenticity and simpli- 
city of expression and to cultivate *dépouillement' —the stripping down 
of the music to bare essentials—and this led him in 1920 to compose his 
‘musique d’ameublement’, ‘furnishing music’, which originated in a 
remark by Matisse that “һе visualized an art without distracting subject- 
matter and comparable to a good armchair’. The occasion was provided 
by an exhibition of paintings at which a play by Max Jacob was staged 
for which Satie wrote accompanying music. This consisted of well- 
known fragments from Thomas's Mignon and Saint-Saéns's Danse 
macabre juxtaposed with simple isolated themes of Satie's own inven- 
tion, in stereotyped repetition and reminiscent of the patterns of a wall- 
paper or carpet. It was intended as mere background music and the 

17 


238 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


composer went around urging the visitors to talk and not to take any 
notice of what was being played. Satie applied the same kind of writing 
to the film music he wrote for the entr'acte іп Re/áche. 

Satie always maintained that in the setting of a text music should 
never usurp to itself the dramatic and narrative elements but furnish 
only a suitably coloured background, and he provided the best illustra- 
tion of this self-denying ordinance in the cantata Socrate (1919), a 
setting for four female voices and chamber orchestra of excerpts taken 
from Victor Cousin's French translation of passages from the Platonic 
dialogues. The voices progress in continuous chant-like lines, with a 
monotony that is intentional, while the instrumental parts proceed in 
often repeated figures (Ex. 101). 

The quiet uneventful flow of this music is occasionally punctuated by 
the expression of muted emotion, particularly in the last part—'The 
Death of Socrates’. The work, which in Satie's words is ‘an act of 
piety . . . a humble homage’, is his masterpiece and the apotheosis of 
the style he had developed in reaction to the luscious harmonic 
language and the enervating melodic refinement of impressionism. 
It may be said to have paved the way for the neo-classicism of Stravin- 


Ex. 101 


SOS, ретин dM жз ax 
^ В Е E 


ou bien je pour -rais 


ERIK SATIE AND ‘LES SIX’ 239 


(Or I could also say that she fell from the rock of the Areopagus, for it is to that 
spot that some transfer the scene.) 


sky and other composers in the following years. Although Satie 
belonged to no group, he had always manifested enthusiasm for what 
was novel and adventurous in the music of his time, and this explains 
why in the last four years of his life he became the champion of four 
composers who were brought to him by Milhaud and who became 
known as ‘L’Ecole d’Arcueil’—so called after the Parisian suburb where 
Satie lived. The most distinguished member of this group was Henri 
Sauguet (b. 1901) whose music showed the influence of Satie in its easy 
fluency and a certain wit. Spontaneous and unpretentious, it eschews 
profundity, seeking to please by its gracefulness and polished charm. 

Far more important than ‘L’Ecole d’Arcueil’ was the group of six 
young musicians who after the first performance of Parade had 
gathered around Satie as the embodiment of the new spirit, a spirit of 
invigoration and rejuvenation at which they all aimed in their different 
ways. In 1926 the critic Henri Collet dubbed them ‘Les Six’, an analogy 
to the Russian ‘Five’. ‘Les Six’ consisted of Louis Durey (b. 1888) who 
had already left the group, Germaine Tailleferre (b. 1892), Darius 
Milhaud (b. 1892), Arthur Honegger (1892-1955), Francis Poulenc 
(1899-1963) and Georges Auric (b. 1899). If they formed a group at 
all, with common aims and aspirations, this was only true in the early 
1920s, after which each composer developed in his own individual 
manner. Moreover much of the aesthetic doctrine considered as charac- 
teristic of ‘Les Six’ belonged more properly to Poulenc and Auric 
who, unlike the older members of the group, came to Satie without 
being bound by any previous loyalties. The literary mouthpiece of 
‘Les Six’ was Cocteau who provided them with a brilliant manifesto, 
Le Coq et l'Harlequin (1918), in which he proclaimed their artistic 
tenets and elevated the reluctant Satie to the position of a chef 
d'école. The Cock of the title represented the new progressive forces, 
while the Harlequin stood for everything that in Cocteau's view was 

1 English translation by Rollo Myers (London, 1921). 


240 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


antiquated and reactionary in the arts. What united ‘Les Six’ for a brief 
space of time was their declared antagonism to romanticism and 
impressionism, their emphasis on simplicity, clarity and terseness of 
expression, and the avoidance of all pretentiousness and boredom in 
music. They aimed at a masculine art and, stimulated by the enthu- 
siasms for Negro art among contemporary painters, they discovered in 
jazz what they perceived to be a potent antidote to the vague, static 
rhythms and the sophisticated harmonic refinement of impressionism. 
Poulenc and Auric in particular manifested a strong predilection for the 
music of the circus, of the fairground and the music-hall in which, 
according to Cocteau, they looked ‘not for the charm of clowns and 
negroes but for a lesson in equilibrium. This school which teaches hard 
work, strength, the exact use of force and a functional elegance is a real 
haute école’. ‘Les Six’ deliberately cultivated the banal, the common- 
place, and the brutal, yet at the same time they loved the prettiness of 
musiquette, writing pastiches in the style of eighteenth-century French 
ballet music and of French opera of the mid-nineteenth century, and 
frequently all these disparate elements were amusingly but incon- 
gruously brought together in one and the same work. 

Durey and Tailleferre are the least significant members of the group. 
Durey, after initially embracing the new aesthetic tenets, soon dis- 
sociated himself from them and his later style showed an affinity with 
that of Debussy and Ravel. Tailleferre wrote for the most part small- 
scale, unpretentious, and rather short-winded works possessing, how- 
ever, a typically Parisian chic and elegance. The two leading figures of 
‘Les Six’ were Milhaud and Honegger. Milhaud, who was born at 
Aix-en-Provence, was, as he suggests in his autobiography, conditioned 
by Latin-Mediterranean culture, and as a young man he displayed 
a strong aversion to German music, notably Wagner’s. Latin clarity of 
form and Latin elegance and conciseness of melody combine in him 
with an inveterate love for harmonic and contrapuntal experiments, and 
he made polytonality, both in the harmonic sense and in the sense of 
several diatonic melodies being superimposed on one another, a 
characteristic feature of his style. Milhaud’s approach to his material is 
purely musical, and one would be tempted to describe him as a ‘natural’, 
were it not that he possesses an extraordinarily sharp musical intelli- 
gence which permits him to essay an immense variety of genres with 
great self-confidence, though he lacks the self-criticism characteristic of 
less fertile composers. Writing with immense speed he has produced 
an oeuvre enormous in volume but very uneven in invention, although 
in technical craftsmanship he proves himself unfailingly resourceful 

1 Darius Milhaud, Notes sans musique (Paris, 1949; English translation, 1952, p. 21). 


ERIK SATIE AND 'LES SIX" 241 


and accomplished. During the 1920s and early 1930s Milhaud was 
attracted by extraordinarily various ideas and trends which account for 
the curious zigzag course his style pursued—neo-classicism and South 
American music, jazz and African ballet, ritual cult and classical-drama. 

From 1917 to 1919 Milhaud was attached to the French Embassy at 
Rio de Janeiro and this afforded him the opportunity to study Brazilian 
folk-music on the spot. The creative result of these studies is seen in the 
orchestral piece, Le Beuf sur le toit (1920)—the title derives from a 
popular Brazilian song used in it—in which Milhaud transcribed 
tangos, maxixes, sambas and a Portuguese fado linking them with a 
theme of his own invention which recurs, rondo-like, a number of times. 
Cocteau later wrote for it a choreographic scenario that is enacted in an 
American bar during the Prohibition period. In the ballet La Création du 
monde (1923), with choreography based on African mythology about 
the creation of the world, the jazz element is pervasive: 


Ex. 102 


Тері. 


Tromb. 


Saxophone 
іп Е 


| ) 


| 


D.B. 


ІШЕ чыз 


() кэй, на 
Еа ЖЕКС A 
[4 таи пее ни 


242 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Milhaud, who wrote it after a visit to Harlem during his American 
tour in 1922, adopted for it the identical orchestra of seventeen instru- 
ments found in the Negro jazz bands of the time. Cing Symphonies 
for small orchestra (1917-23), Machines agricoles (1919), and Catalogue 
des fleurs (1920) are conceived in a predominantly neo-classic style. The 
symphonies are marked by a rudimentary kind of counterpoint often 
approaching heterophony and by polytonal passages in several simul- 
taneous keys as, for instance, in the third movement of Symphony 
No. 2: 


Ex. 103 


Г — Ди «f 
EC 
үе mus 
2... БЕНЕН - 


ІНЕ 


һе” | 


С.А. 


Вп, 


Celli 


E 
| | 


ERIK SATIE AND ‘LES SIX’ 243 


The harmonic idiom is rather acid but the melodic writing is frequent- 
ly punctuated by simple folk-song-like themes of a Mediterranean 
character. Machines agricoles and Catalogue des fleurs are settings for 
voice and chamber orchestra of descriptions, respectively, of agricul- 
tural implements and flowers from a seedsman’s catalogue, and they 
combine a pastoral vein with an impersonal objective expression. 
Milhaud intended them not as a joke but in the spirit of Virgil’s 
Georgics; yet an element of incongruity cannot be denied to exist 
between the dry, factual nature of the text and the inventive musical 
treatment (Ex. 104). 

One of Milhaud’s finest works is the opera Les Malheurs d’Orphée 
(1925), a modern adaptation of the classical myth in which Orpheus is a 
healer in the Camargue and Eurydice a gypsy girl; the music has the 
luminosity as well as the intensity of the Provençal landscape and the 
economy of means employed is most remarkable. In Le Pauvre Matelot 
(1926), ‘une complainte en trois actes’, which is Milhaud's best-known 
opera, Cocteau, who was the librettist, derived the subject from an 
incident reported in a newspaper, and treated in a realistic, almost 
veristic manner. А sailor returning home incognito passes himself off 
to his wife as a rich friend of her husband and is murdered by her for 
his money. In strong contrast to the character of the action Cocteau's 
language imitates the detached and impersonal style of sailors' tales, and 
this is matched by Milhaud's setting which is completely stylized and 
objective and frequently creates an ‘alienation’ effect such as Stravinsky 
achieved by similar means іп L’ Histoire du soldat and Oedipus Rex. In 
the three miniature operas, L'Enlévement d'Europe, L'Abandon d' Ariane, 
and La Déliverance de Thésée (1927), which together last about thirty 
minutes, Milhaud followed the vogue fashionable at that time in 
Germany for short one-act operas. The antique myths are treated in a 
half poetic, half ironic, and off-hand vein, with a modern tango in the 
second and jazz rhythms in the third opera. Christophe Colomb (1928), 
to a libretto by his friend Paul Claudel, is Milhaud's most ambitious and 
spiritually most profound stage-work of the inter-war period, in which 
he achieved a remarkable fusion of a medieval mystery play with a 
modern opera-oratorio, a fusion deserving the epithet *unique'. The 
action concerns the life of Christopher Columbus as seen from the point 
of view of posterity; and to enhance and intensify the stage drama, a 
cinematographic film duplicates the stage scenes on a screen, presenting 
images which vary in their degree of verisimilitude to the action, 
ranging from a more or less accurate reproduction of the stage scenes to 
unreal, dream-like pictures which concentrate on the poetic essence of 
the subject. Accompanied by percussion, a Narrator, reading from an 


1918-1939 


MUSIC IN THE MAINLAND OF EUROPE: 


Le Begonia 


Ex. 104 


244 


T 
| | E 
ШЕП У WD 


ei 


| 
| 


fleur trés double, 


Au-ro - ra, 
rail, 


de co- 


Be-go -ni - a 


| 1 
M в a 


|) 
al 


© 
> 


: 


Voice 
Vin. 
con sord. 
Vd 
con sord. 

Voice 
Vin 
Vcl. 
D.B 
D.B 


(Begonia Aurora, very double flower, apricot mixed with coral, very attractive 


Shades, unusual and interesting.) 


ERIK SATIE AND 'LES SIX' 245 


old chronicle, explains and links the twenty-seven scenes in a strictly 
rhythmical recitation in which a speaking chorus frequently joins. There 
is a numerous cast of singers but, with the exception of Columbus and 
Isabella of Spain, these parts are all subsidiary and the singing chorus is 
given the most important musical sections. Milhaud set Ше text to 
grave, sombre, and exalted music, rather in the manner of a classical 
oratorio, and effectively contrasts lyrical and epic elements with the 
dramatic. The vocal style oscillates between recitative-like declamation 
and arioso, while the predominantly diatonic character of the instru- 
mental themes is concealed behind a bitonal and polytonal treatment. 
Thus, in the nineteenth scene, Че Rédempteur', one orchestral part 
consists of no fewer than seven superimposed thirds, another part 
moves chromatically in block harmonies and a third part has a bell-like 
ostinato to play; the first chorus sings an independent part in octaves 
while the second chorus is given a five-part organum (Ex. 105). 

Such tremendous complexities—and they are also to be found in 
many of his later works—are difficult to reconcile with Milhaud's de- 
clared aversion to German music. In fact Christophe Colomb and some 
of his subsequent operas such as Maximilien (1930) and Medée (1938) 
show an affinity with the ideals of the German expressionism rather 
than with Milhaud's own ideals of clarity, simplicity, and economy of 
means and attempt ‘not to render the visible but to render visible’ the 
individuals deep-seated irrational instincts and drives. Yet in his 
chamber music and song cycles with orchestra of the same period 
(String Quartets nos. 5-9, 1920-35; Trois chansons de troubadour, 1936; 
Les quatre éléments and Trois élégies, both 1938) Milhaud produced 
music which, despite harmonic astringencies and complicated polytonal 
passages, demonstrates that aspect of his artistic make-up which finds 
expression in a delicate and often alluring Latin lyricism.? 

Arthur Honegger, the son of Swiss parents but born and brought up 
in France, shows in both his musical personality and in his aesthetic 
outlook a mixture of German and French traits. It is significant that 
he displayed no great sympathy for the anti-emotional, ‘objective’ 
tendencies of “Тез Six' to which he adhered only during the early 
1920s, nor did he share the interest in the music of the fairground and 
music-hall so characteristic of Poulenc and Auric. Unlike the other 
members of this group, he favoured large-scale architectural designs 
and linear polyphony; and, so far as his stage works are concerned, he 
showed a strong predilection for grave, tragic subjects treating them in a 
statuesque monumental fashion, yet at the same time revealing a highly 
developed sense of dramatic gesture and theatrical effectiveness. 


1 As in the slow movement of no. 6, recorded in The History of Music in Sound, x. 


MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


246 


Ex. 105 


Chor. I 


an 


ai 


tus! 


Sanc 


= 
D 
в 


REY 
Ви 


Chor. П 


tus! 


Sanc - 


SH 
8 е 8 


Orch. 


а "n 
ШШ, 

(y QI КҮ ШІ | 
LLL in ІІ LLLI ІШІ 

x 

Ф 

Nn 

ш 

а N 

A 

Z 

< 

B 

— 

< 

ж 

= 

5-4 

Q 


e 
8 
> 
D 
о 
Я 
e 
Nn 


TH 


248 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Honegger was comparatively slow in developing a personal style. 
In his early works (First String Quartet, 1917, and two Violin Sonatas, 
1918 and 1919, respectively) he is seen to oscillate between homophonic 
writing and a complex contrapuntal manner, a dichotomy which was to 
persist though the demarcation between the two styles was to grow 
less clear-cut in his maturity. A more personal note is revealed in his 
Pastoral d'été for chamber orchestra (1920) which is unpretentious and 
direct in its appeal and thus embodies one of the chief tenets of “ез 
Six'. It is based on two themes, one lyrical and the other bucolic, and 
displays a relatively simple texture in which a rhythmic ostinato serves 
to generate the atmosphere of a pastoral idyll. Another aspect of 
Honegger's musical character is shown in Pacific 231 (1923) and Rugby 
(1928) both of which are symphonic movements with an illustrative 
tendency, though the composer disclaimed any descriptive intentions. 
Pacific 231 resembles in its form a figured chorale and one of its salient 
features lies in the fact that, while the metronomical speed progressively 
diminishes, the metric units are filled with increasingly more notes. The 
piece translates the exciting visual impression created by a train engine 
of 300 tons hurtling down the track at 75 miles per hour and captures 
the sheer physical exhilaration aroused Бу this spectacle. Though 
Honegger was probably the first to have attempted it in music, this 
intoxication with speed and the impersonal beauty of mechanical objects 
had been extolled as early as 1911 by the Italian Futurists. In Rugby, 
sport and physical movement are made the subject for musical treatment 
though in this Honegger was anticipated by Martin in his Half Time 
(1925). The work hardly varies in speed and uses a swinging theme as its 
main idea. Yet another and more intrinsic side of Honegger's per- 
sonality is shown in the Symphony in Three Movements (1930)— 
music of great muscular strength, rhythmic vigour and lean texture. 
The String Quartets Nos. 2 and 3 (both 1936) display the same charac- 
teristics as the Symphony but combine with their close formal control 
and cogent thematic thinking a large measure of contrapuntal skill. 
Honegger prefers to set the various themes against one another (mostly 
in the recapitulation) rather than exploit any motivic kinship between 
them. In the finale of Quartet No. 3 he appears to flirt with dode- 
caphony, as illustrated in its opening: 


Ex. 106 
Allegro 


ERIK SATIE AND ‘LES SIX’ 249 


but he was too deeply anchored in tonality to go further in his experi- 
mentation with serialism, and the movement closes in an unambiguous 
E major. 

Honegger wrote numerous stage works in the majority of which he 
aimed at combining an appeal to large unsophisticated audiences with 
an interest for the educated music lover and professional musician. They 
are large frescos of a static, monumental character, whose subjects are 
drawn from the Bible, Greek tragedy and French medieval history. 
The best known of these is Le Roi David (1921) which was originally 
written as incidental music to a play by René Morax but later trans- 
formed into a Symphonic Psalm in three parts—in fact, a dramatic 
oratorio in which a Narrator describes the intervening stage action. 
Honegger had not yet quite found himself in this work and the stylistic 
sources on which he draws are of the most heterogeneous kind: 
Handelian counterpoint and folk-song-like melodies, exotic, oriental- 
sounding music and Bachian chorale, simple diatonic harmonies and 
polytonality. But whatever Honegger’s borrowings, the whole is more 
than its constituent parts and the composer succeeds in re-creating the 
spirit and atmosphere of the Old Testament with an impressive vivid- 
ness and directness of expression. The music alternates between harsh, 
barbaric grandeur and lyrical simplicity best exemplified in the various 
choral psalms and solo numbers. Thus, the ‘Song of David the Shepherd" 
employs a vocal melody of a simple diatonic character—combined, 
however, with chromatic harmonies in the orchestra, while in the 
*Lament of Gilboa' oriental arabesques and melismas are used for an 
expressive purpose: 


Ex. 107 
Largamente 


250 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


The ‘Invocation of the Witch of Endor' shows the composer as a 
master of suggestive instrumental evocation, while the *Dance before 
the Ark’ is a choral movement of tremendous vitality and primitive 
corybantic power. Judith (1925)! was also originally conceived as inci- 
dental music for a play by Morax, though subsequently Honegger 
turned it into a three-act opera. The subject, which is taken from the 
well-known biblical story of Judith and Holofernes, is set to music 
remarkable for its evocation of a stark primitive atmosphere, notably 
by means of highly dramatic choruses, and among the chief devices 
used for characterization of the protagonists are oriental melismas and 
psalmodic melodies with a marked modal tendency. 

Compared with Le Roi David, the opera shows a far more homo- 
geneous stylistic character. Honegger's most dramatic opera is un- 
questionably Antigone (1927) whose glowing intensity of expression 
stands at the opposite end to the hieratic grandeur of Stravinsky’s 
Oedipus Rex. Cocteau's libretto, freely adapted from the tragedv of 
Sophocles, strictly observes the three Aristotelian unities and is most 
direct in its language. In character with this tragic subject the music is 
marked by violent dissonances and atonal-sounding polyphonic 
writing; the individual scenes are firmly held together by a close-knit 
symphonic structure. The treatment of the text avoids the traditional 
recitative which is replaced by a word-inspired melodic line such as the 
composer had already attempted in Judith and which is comparable to 
Jandéek’s realistic speech-song and Pizzetti’s declamatory arioso. 
Jeanne d'Arc au bücher (Joan of Arc at the stake) (1935) is a vast and 
spectacular stage oratorio in which Paul Claudel aimed at a synthesis 
of ancient mystery play, chanson de geste and modern drama. Honegger's 
music possesses the immediate graphic quality of a film but it more 
often accompanies than interprets the action. 

Francis Poulenc and Georges Auric were the youngest members of 
the ‘Les Six’ and stood, as has been said, closest to the aesthetics of 
Satie which they even carried a step further. Both composers gained an 
early reputation as musical clowns who wrote with tongue in cheek. 
Both were soon dubbed ‘les sportifs de la musique’, owing to the robust- 

! See pl. V(a). 


РГАТЕ V 


HONEGGER'S JUDITH, 1925 (see p. 250) 


A scene from the original production at the People's Theatre, Darmstadt 


KRENEK’S JONNY SPIELT AUF, 1927 (see p. 340) 


The station scene from the original production at the Neues Theater, Leipzig 


ERIK SATIE AND 'LES SIX' 251 


ness of the sentiments expressed and a sense of physical hustle and 
bustle which emanates from their compositions. Poulenc felt the 
pull of several influences—in the first place Satie's and in varying 
degrees those of Chabrier, Ravel, and Stravinsky. Yet, as he matured, 
his style developed an unmistakable personal cachet compounded of 
playfulness and gravity, of simplicity and sophistication, while some of 
his compositions exhale the flavour of Ча vieille France' lending them 
a certain nostalgic charm. In his music spontaneous melodic invention 
and a well-developed sense of pointed rhythmic structure are allied to a 
neat and deft craftsmanship. Poulenc's preference was for the smaller 
forms. He was eighteen when in the wake of Europe's ‘discovery’ of 
American jazz he wrote the Rhapsodie négre (1917) and this was soon 
followed by the Mouvements perpétuels (1918), three pieces for piano 
which derive their title from the fact that each is based on a rhythmic 
ostinato figure which is almost incessantly repeated. They are simple 
two-part inventions of an unpretentious but vivid charm, with bitonality 
as a characteristic feature. An effective interpreter of his own piano 
music, Poulenc wrote a great number of pieces for the instrument 
among which the Sept Nocturnes (1935) must be reckoned as the most 
important; a tender and delicate melancholy (nos. 1 and 4) and calm 
serenity (no. 6) are matched by the expression of ironic humour (no. 2) 
and the exhilaration engendered by rapid physical movement (no. 5). 
Some of his vocal chamber music reveals perhaps best the two pro- 
minent aspects of Poulenc's artistic personality. Іп Cocarde (1919), 
to poems by Cocteau which are intended as parodies of the style of the 
French Symbolists, the music perfectly matches the spirit of the words 
in a humorously inconsequential manner, the scoring for violin, cornet, 
trombone, bass drum and triangle imitating the fortuitous ensembles 
found in Parisian music-halls and dance-halls. On the other hand, Ле 
Bestiaire for voice, string quartet, flute, clarinet, and bassoon (1919), 
six settings of animal verses by Guillaume Apollinaire, is remarkable 
for the blend of gentle irony and prettiness and for the delicate and 
pointed texture in which each note seems to stand in its right place 
(Ex. 108). 

The individual songs are of the utmost brevity, with ‘Га Sauterelle’ 
extending to no more than four bars. In the ballet Les Biches (1923) 
Poulenc conjures up partly the vieille France of the eighteenth century 
(Rondeau and Adagietto), partly the spirit of the jazz age (Rag- 
Mazurka), while the exquisitely wrought Concert champétre for 
harpsichord and orchestra (1928) echoes the world of Chambonniéres 
and Couperin. Outstanding among Poulenc's religious works is the 
Mass for unaccompanied chorus (1937), which in its marked purity of 


MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


252 
"Ex. 108 


L'Ecrevisse 


есігін! 


Jp 


| 


О! 


bien chanté 
- cer -ti- tude, 


ux, 


do 
In 


ERIK SATIE AND ‘LES SIX" 


ПЕЙ 


4 


| 


| 
Is 
u 


LT 


| 
| 


en 


nous nous 


| 
| 


PNY 


s doux 


trè 


cre - visses 


me s'en vont 


com 


(Uncertainty, O! my delight, you and I go away as crayfish do.) 


8 


254 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


style and serenity of spirit provides a demonstration of the fact that this 
composer was at times able to discard his tongue-in-cheek approach and 
probe into more profound emotions. The Litanies à la Vierge Noire de 
Rocamadour (1936), and Quatre Motets pour un temps de pénitence 
(1938-39) belong to this same serious and often impressive vein. 

Like Poulenc, Auric showed a precocious talent, composing, it has 
been stated, more than 300 songs and piano pieces between the ages of 
twelve and sixteen. His early admiration of Ravel led him first to 
imitate the former's style but subsequently Auric came under the 
influence of Satie and, partly, Stravinsky when he adopted Satie's 
precept that the largest measure of audacity lies in simplicity. Thus, in 
the Huit Poémes (1920), settings of wittily ironic verses by Cocteau, the 
voice part is formed by a straightforward diatonic melodic line, rather 
in the style of a French folk-song or popular tune, and is embedded in 
an equally simple but very pliant piano accompaniment that often 
serves to point the witticism of the text. But this phase of deliberate 
simplicity soon passed and Auric, developing a more personal note, 
harnessed a muscular rhythm and violent discordant harmonies to a 
plain melodic statement. For instance, in Les Joues en feu (1921), a 
setting of three poems by Raymond Radiguet, the voice part of the 
first song modulates no further than the subdominant but the piano 
indulges in the most dissonant clashes between E flat and B major while 
the third song is characterized by first successive and then simultaneous 
bitonality. Technically Auric avoids phrases of any length and elabora- 
tion and, like Poulenc, he juxtaposes sections of the most heterogeneous 
character in order to produce a violent shock of contrasts. Thus, in the 
ballet Les Fácheux (1925), after Moliére, a highly adorned dance marked 
*trés lente et expressive' is offset by a fast staccato movement suggesting 
sheer motor energy. Since the great success of René Clair's film, А nous 
la liberté, for which he wrote the music, Auric has been much in demand 
as a screen composer. In his general artistic make-up a sense of comedy, 
irony, and scepticism and a sharp intelligence are perhaps the most 
salient features. 


“ТА JEUNE FRANCE’ 


Since the time of the querelle des bouffons Parisian artists have shown a 
strong inclination to group themselves into separatist bodies, each 
representing a particular aesthetic creed and engaging in heated 
polemics with their antagonists. This phenomenon is the manifestation 
of a typically Gallic spirit of individuality and heterodoxy not en- 
countered to the same extent in any other European country. We have 
seen how in the years shortly after the First World War the short-lived 


‘LA JEUNE FRANCE’ 255 


group of 'Les Six' was formed in reaction against impressionism, and 
sixteen years later, in 1936, four members of the rising young generation 
grouped themselves into 'La Jeune France' whose chief aim was to 
achieve yet another rejuvenation of French music by attempting to re- 
instate those deeper and more permanent values which they found 
lacking in the creations of their elders. At the same time they en- 
deavoured, like the young Italian composers of the 1920s, to revive a 
national spirit neglected by the older generation in favour of an excessive 
internationalism, and in general they were opposed both to revolution- 
ary tendencies and to faint-hearted academicism. АП four members of 
“Га Jeune France' were born in the first decade of the twentieth century 
—André Jolivet (1905), Yves Baudrier (1906), Daniel Lesur (1908), and 
Olivier Messiaen (1908). The mature works of these composers do not 
fall into the purview of this chapter but their early style demands a 
brief consideration here. 

From the first Jolivet and Messiaen were the outstanding members of 
the group, a fact amply confirmed by their later achievements. Jolivet, 
who was a pupil of Edgard Varése, first adopted Ше ‘hermetic’ style of 
his teacher, with its constant search for extraordinary instrumental 
timbres and novel effects in the field of rhythm. During his early period 
Jolivet was much preoccupied with ideas about such primitive forms of 
religion as magic and the appropriation of cosmic forces, believing in 
vibrations from unseen psychical ‘fluids’ which he thought he could 
detect in all manifestations of life and even in inanimate objects. 
Already the titles of those works intended as an expression of these 
psychic phenomena are indicative of Jolivet's ideas and his concept of 
music as a ‘cosmic’ force endowed with magical properties!—the piano 
suite Mana (1935), Incantation: Pour que l'image devienne symbole for 
ondes martenot or flute (1937), the orchestral prelude Cosmogonie 
(1938), and Cinq Danses rituelles (1939). The declared purpose of this 
music was to cast a magic spell, and in pursuing this object Jolivet 
evolved a personal style of great harmonic freedom and marked by a 
wide range of novel rhythmic and dynamic effects, all with the aim of 
generating in the listener an appropriate state of mind. After the Second 
World War, however, Jolivet began to turn to music which is more 
closely concerned with man and his natural feelings. 

Messiaen, who was a pupil of Paul Dukas and the organist Marcel 
Dupré, showed very early a strong leaning towards a particular form of 
Catholic mysticism, a subject which, according to his note for the first 
London performance (1938) of his La Nativité du Seigneur, ‘is the best 
subject, for it encompasses all subjects. And the abundance of technical 

1 Cf, the case of Skryabin, p. 34. 


256 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


means allows the heart to open up freely.’ It was largely as a composer of 
organ music that Messiaen made his name in the 1930s, and particu- 
larly by £a Nativité du Seigneur (1935), a series of nine meditations on 
Christian themes, and Le Corps glorieux (1939). These and some other 
works of that period form the transition to Messiaen's later style, in 
which plainsong, elements from Hindu music and a novel conception 
of rhythmic structure play an important part. Yet what remained con- 
stant in his music, whether vocal or instrumental, is the expression of 
intense emotional sincerity, religious ecstasy and mystical contemplation. 
Composers who remained more or less independent of the aesthetic 
doctrines advanced by the group of “Гез Six' апа “Та Jeune France' 
include Jacques Ibert (1890-1962), and Jean Frangaix (b. 1912) whose 
prodigal fertility and extreme facility are matched by elegance and 
brilliance of workmanship rather than any profounder qualities. 


ITALY: LA GENERAZIONE DELL'80 


In Italy the roots from which the new post-war ideas were to spring 
were planted in the first decade of the present century and showed a 
pronouncedly national bias. Under the intellectual leadership of the 
musicologist Fausto Torrefranca, who published a fiercely polemical 
book on Puccini,! the young composers of that time—chief among them 
Ildebrando Pizzetti (1880-1968), Gian Francesco Malipiero (1882-1973), 
and Alfredo Casella (1883-1947) representing the generation after 
Puccini’s—gathered together. Theso-called generazione dell’80 shared the 
conviction that native opera had run its course and become effete and 
that a risorgimento, a renascence of Italian music, could only be brought 
about by the assimilation of the spirit and style of the great instrumental 
masters of the past: Frescobaldi, Vivaldi, Corelli, Veracini, and others. 
The monopoly that opera occupied in the musical life of the nation was 
deeply resented, vocalism was condemned and finally the stage was 
reached when the young iconoclasts clamoured for a ban on all those 
composers who had exclusively devoted themselves to opera (Puccini, 
Mascagni, Leoncavallo). By diverting the creative energies from melo- 
dramma to instrumental music the avant-garde hoped to achieve that 
ristabilimento dell? equilibrio which it so much desired and which, 
indeed, was restored by the large amount of instrumental works 
composed in Italy during the inter-war period. But it is to be noted 
that this process had already been begun, though in isolation, in 
the 1880s by such composers as Martucci, Bossi, Sinigaglia, and 
Sgambati. These however, had carried out their reform largely by imitat- 


! Giacomo Puccini e l'opera internazionale (Turin, 1912). 


ITALY: LA GENERAZIONE DELL'80 257 


ing the German symphonists, whereas the new movement sought to 
achieve the same object by a return to the instrumental forms of 
seventeenth- and eighteenth-century Italian music. The new anti- 
operatic tendencies were both idealistic and nationalistic. Idealistic, 
because they set out to turn away from the bourgeois mentality of 
realist opera, with its lack of spirituality and high moral values; 
nationalistic, because they demanded an art exclusively nurtured in old 
Italian soil and freed from the influence of both later German romanti- 
cism and French impressionism. In a thoughtful book,! which includes a 
fairly balanced study of Puccini, Pizzetti reproached the entire impres- 
sionistic school for its over-refinement, its growing exclusion of the life 
of the emotions and ‘its prodigious faculty of stifling the will to live’.? 
On the other hand, he accused Puccini and the rest of the Italian 
realists of having sinned in the opposite direction by their emotional 
excesses and their superabundance of crude vitality which, according to 
Pizzetti, defied full translation into satisfactory aesthetic expression.? 
Yet the extreme position taken up by these young firebrands was 
gradually relinquished with growing maturity when both Pizzetti and 
Malipiero devoted their attention to opera, although guided by aesthetic 
and stylistic principles totally different from those of verismo and 
realist opera in general. 

While the hostility towards this kind of opera, particularly as 
represented by Puccini, was at its height, Puccini himself had been 
undergoing, from about 1912, a certain change of outlook and began to 
detach himself from the line he had pursued in Tosca, Madama Butterfly, 
and La fanciulla del West. Тпе first milestone of this changed course was 
Il trittico (1918) of which only the first episode, П tabarro, after a 
French play La Houppelande, contained a strong touch of verismo in 
its brutal ending. А piéce noire, this opera is a masterpiece of dramatic 
concentration and in its evocation of a most unusual atmosphere— 
the wretched life of bargees on the River Seine at the beginning of the 
present century, which forms the background to a triangular drama 
between husband, wife and lover. The second episode, Suor Angelica, 
shared with the first an uncommon subject and uncommon setting— 
the tragic frustration of maternal love and the suicide of a nun in an 
Italian convent at the end of the seventeenth century. The opera, in 
which all the characters are women, suffers from the quietism and 
passivity inseparable from the monastic atmosphere. The final miracle, 
too, remains only a stage spectacle, Puccini lacking the power to convey 
mystic ecstasy and the cathartic force of Divine Grace. But the music 


1 Ildebrando Pizzetti, Musicisti contemporanei. Saggi critici (Milan, 1914). 
2 [bid., p. 130. 3 Ibid., p. 51. 


258 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


is of impeccable craftmanship and particularly successful in its por- 
trayal of the cruel aunt and of the monastic milieu. The third panel of 
Il trittico, Gianni Schicchi, was Puccini's sole comic opera and shows 
a most remarkable extension of his creative range. The subject, which 
was drawn from a brief anecdote in Dante's Inferno, concerns the duping 
of scheming heirs by the impersonator of the dead testator and is 
turned into a riotous comedy of situation. The music, reviving the 
spirit of eighteenth-century opera buffa, is brilliantly witty and ironic; 
the only slight blemish in it is Lauretta's sentimental air “О mio babbino 
caro'. Compared with Verdi's Falstaff, Gianni Schicchi shows a brand 
of humour which is bitter, mordant, even harshly cynical. Turandot 
(1924), after the play of the same name by Carlo Gozzi, is Puccini's 
greatest masterpiece. Its larger theme is the liberating and uniting power 
of true love which lends the work a universal significance absent from 
the rest of Puccini's operas. Moreover, with the exception of the final 
two scenes, left only in sketch by the composer and completed by Franco 
Alfano, the dramatic structure is masterly, and the music represents a 
synthesis and, with it, a consummation of the four separate aspects of 
Puccini's previous style: the lyric-sentimental (Liü), the heroic-grandiose 
(Turandot and Calaf), the comic-grotesque (Masks), and the exotic. 
By the time Turandot was first produced (1926) both Pizzetti and 
Malipiero had launched into opera; but their stage conception, condi- 
tioned and guided by new operatic aesthetics, was entirely different 
from those of both Verdi and the bourgeois dramatists of verismo. The 
main point in Pizzetti’s theory of reform was the argument that if 
opera is musical drama, then everything must be subordinate to the 
tying of the dramatic knot, to the surmounting of the emotional crisis 
and to the unfolding of the characters’ psychology. In such a scheme 
lyricism must never be allowed to interrupt the flow of the inner action, 
no lyrical halts should interrupt the development of the drama. “Еог 
five centuries,’ Pizzetti wrote with dubious accuracy,! ‘from the four- 
teenth to the nineteenth centuries, there was an uninterrupted tendency 
in opera to lyricism at all costs. . . . It is not the characters who have 
sung, or rather lived; the poets have spoken and the musicians have 
sung.' Unless lyricism arises directly out of the dramatic context it has 
no justification. In consequence of this theory Pizzetti developed a 
style of operatic writing in which declamation is the paramount feature, 
a kind of declamation encompassing both dramatic recitative and arioso 
and thus providing a solution of the antithesis between drama and 
lyricism. This heightened declamation, which combines drama and song 
in a higher unity, follows the inflexions of the Italian language so 


! Musica e dramma (Rome, 1945), p. 41. 


ITALY: LA GENERAZIONE DELL‘80 259 


accurately that it renders the translation of the text into other languages, 
without distorting alteration of the vocal line, almost impossible, as for 
instance in this passage from Debora e Jaele: 


Ex. 109 


Debora 


Е ЕЕ NEIGE] SOEUR] pem) SU] рано CE A. Cee қүлы Г тт exa 

га кукук hoi) dii атары 

ST ПЛЕТЕНИЕ SST QU rt ту a Фе рез 
[mue em mr ж 21 


- tez-ze faran сап-рі di ru - i -ne pei gu - fi gli av-vol - 


2 a и: 
£ 5 D 21—249 
Е Ара ЗЕ а Gal седи ұстары їй ТЫШ! 
оре ela +. Нева ЕР НИ [escapan rmt pe с 


toi Ein  ma-no del mio po-pol sa-rà da - to — Si-se-ra, 


(And they shall slay the slayers, and prey on the predators, the vultures shall 
make of the towns and fortresses fields of ruins for the owls. And Sisera shall be 
given into the hands of my people.) 


This unity of words and music—different from Wagner's Wortvers- 
melodie but similar to Debussy's word-inspired declamation—was in 
Pizzetti's case greatly aided by the fact that he was his own librettist, 
who created text and main musical themes and motives simul- 
taneously. A further corollary of his theory is seen in the stripping 
from the text of all verbiage, and in the projecting of the action without 
superfluous episodes in which music could only play the part of an 
adjunct, whereas it should be the vital and direct translation of the inner 
drama. Although by these means Pizzetti succeeded in achieving a 
remarkable continuity and pliancy of musical structure and expression, 
it must still be admitted that his exclusive use of the declamatory prin- 
ciple often creates a distinct feeling of monotony. 

Fedra (1912), to a libretto by d'Annunzio, was the first opera in which 
Pizzetti applied his theory, though as yet in a tentative manner; it was 
not until he wrote the trilogy Debora e Jaele (1921), Lo straniero (1925), 
and Fra Gherardo (1927), that he succeeded in putting it into full effect. 
Pizzetti, a deeply religious man, went to the Bible for the subject of the 


260 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


first two operas, while for the later part of the trilogy he chose a histori- 
cal subject drawn from a Parma chronicle of the thirteenth century, 
though the ultimate tragedy of Fra Gherardo was wholly invented by the 
composer. The underlying theme of the trilogy is redemption through 
the power of self-sacrificing love and the devotion of human beings 
surrounded by an uncompromising, hostile, and fiercely dogmatic 
world. Though in theory Pizzetti abjured the inclusion of purely lyrical 
episodes, in practice he did not altogether banish them: Lo straniero, 
for example, is marked by more static, sustained, and lyrically expressive 
sections, and Fra Gherardo contains even self-contained vocal music 
of a strophic character. In all these three operas as also in Orséolo 
(1935) and L'Oro (1942), the chorus plays an extremely active role. With 
the function of the chorus in Greek tragedy at the back of his mind, 
Pizzetti treats it as a dramatis persona in its own right and with a 
psychology of its own, assigning to it some of his most powerful music. 

Hardly less important than the stage works which form the nucleus 
of his output are Pizzetti's choral compositions. These represent a 
synthesis of his own style with influences from Gregorian chant and 
fifteenth- and sixteenth-century Italian vocal polyphony. Outstanding 
here is Ше unaccompanied Messa da Requiem (1922), a markedly lyrical 
work in the Dorian mode and representing a serene meditation on 
the theme of death which eschews all dramatic accents. Even the ‘Dies 
Irae', the most extensive and most elaborate of the five movements, is 
in a reticent, subdued vein: 


ITALY: LA GENERAZIONE DELL'80 261 


| 


| 


к= Tes- te— Па-\!4 сит Si - byl - la. 


The magnificent ‘Sanctus’ is for three choruses, one of which is for 
women's voices and the other two for male voices, which alternate 
between antiphony and singing in concert. Most of Pizzetti's instru- 
mental works are implicitly programmatic and conform to his aesthetic 
tenet that music must stand in close relation to life and nature. On this 
theory there is no room for ‘pure’ or ‘objective’ music, music that is not 
evocative of nature or expressive of human emotions; and a whole 
series of Pizzetti's instrumental compositions provides testimony to this 
view. Thus, the Concerto dell'estate for orchestra (1928) consists of three 
movements, ‘Mattutino’, ‘Notturno’, and ‘Gagliarda e finale’, which 
are expressive, respectively, of the pristine freshness of a summer 
morning, of the many voices with which nocturnal nature speaks to 


262 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


man, and of the joys and the zest for life summer inspires in human 
beings. In the Rondo veneziano (1929) the ritornello symbolizes the 
pride and splendour of ancient Venice while the three episodes depict 
scenes from Venetian life. Characteristic of these and other works, 
such as the Canti della stagione alta for piano and orchestra (1930), is 
the essentially vocal nature of the themes though their elaboration 
conforms to symphonic principles. 

Malipiero, as uneven and over-prolific in his production as Milhaud, 
is nevertheless the most eminent among the generazione dell'80. 
Profoundly steeped in the music of the Italian Baroque,! he derived 
from it the austerity and nobility of his general artistic aspirations. To 
baroque music he also owes something of his well-developed sense of 
melody, while his studies of Gregorian chant bore fruit in the modality 
and free structure of his melodic idiom. In his early work Malipiero was 
subject to influences from German and, notably, French music 
(Debussy); and although the latter influence diminished, his harmonic 
style retained parallel shifts of triads, fauxburdons, and chords of 
superimposed fourths and fifths. French impressionism is partly 
responsible, too, for his marked inclination to evoke characteristic 
states of mind and emotional attitudes aroused by the contemplation 
of nature, people and places and by a nostalgic longing for the past. 
Thus, in Pause del silenzio (1917) he describes in seven 'symphonic 
expressions’ the different moods he experienced during the war when 
the tension and turmoil of external life made it very difficult to achieve 
peace of mind and spiritual serenity. In this work he applied for the 
first time the principle of thematic non-development, themes being 
stated and repeated in slight harmonic and rhythmic alterations, after 
which they make room for fresh ideas. Each of the seven sections 
has its own theme but, with no development, the texture is fragmentary 
and kaleidoscopic, though an element of unity is provided by the 
opening horn fanfare which recurs like a ritornello after each section. 
This method of construction, for which a hint is to be found in the 
instrumental works of the ancient Italian masters and also in Debussy, 
may be compared to the psychological process of ‘association by 
contrast’. The application of this method is seen at its clearest in 
Malipiero’s first three string quartets—Rispetti e strambotti (1920), 
Stornelli e ballate (1923) and Cantari alla madrigalesca (1931). The 
titles of the first two works are derived from old Italian verse- 
forms consisting of a series of short poems of different content and 
mood to which unity is given by the verbal style. The music displays 


1 Malipiero edited a complete edition of Monteverdi’s works and collaborated in the 
collected edition of Vivaldi’s music. 


ITALY: LA GENERAZIONE DELL’80 263 


an analogous structure in that there are twenty sections in the first 
quartet and fourteen in the second, all of the most diverse character— 
grave and comic, tender and ironical, fantastic and bucolic. In both 
works a unifying element is generated by a kind of ritornello, which in 
the first quartet takes the form of a succession of quadruple and 
double stops on the open strings (first violin and viola) evoking the 
characteristic sound made by the tuning of strings. 


Ex. 111 


Un poco ritenuto 
Ex 


Ist Уо. еқ 
МЗ! 


Malipiero arrived comparatively late at pure symphonic composition 
in his Prima sinfonia, in quattro tempi come le quattro stagioni (1933) and 
Seconda sinfonia (Elegiaca) (1936) to which he later added seven more 
symphonies. The First Symphony was largely inspired by a nostalgia 
for Venice's great past and partly by Stagione, poems by X. Lamberti, 
and is modelled on the Italian sinfonia of the seventeenth and eighteenth 
centuries, with a characteristic concertante treatment of woodwind and 
horns, while the structure shows a return to the traditional symphonic 
technique abjured by Malipiero in his previous period. In the Second 
Symphony this feature is even more pronounced, and the work demons- 
trates his art of achieving a close fusion between musical thought and 
emotional expression. 

In his operas Malipiero evolved a style diametrically opposed to that 
of Pizzetti. While Pizzetti aimed at dramatic development, continuity 
and flexibility by means of a half-declamatory, half-arioso recitative, 
Malipiero's operatic aesthetics exclude all dramatic dialectics: action 
and characters are not developed but are static like a bas-relief; there is 
no before and no after. The recitative is abandoned as a naturalistic 
device and as an obstacle to lyricism; in other words, lyrical melos is the 
prime feature of Malipiero's operas. Despite the great variety of his out- 
put, the stage works represent his most characteristic and, undoubtedly, 
his most significant creations. It is in them that his northern romanticism, 


264 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


his love of the fantastic, of the supernatural and of nocturnal mystery, 
finds its most eloquent expression. This trait showed from his earliest 
years and persisted through the radical changes which Malipiero's 
artistic outlook underwent, changes which must be interpreted as a 
rebellion against nineteenth-century music, notably German sym- 
phonism and Italian melodramma. His operatic ethos is at once idea- 
listic, romantic, and poetic—‘the classical ideal of an unquiet romantic 
spirit’, as one of the composer's Italian biographers defined it!—and 
this is already seen in Malipiero's choice of subjects from ancient 
Italian poetry, Goldoni, Euripides, Shakespeare, E. T. A. Hoffmann, 
and Pirandello. 

Most typical of Malipiero's novel conception of opera are L'Orfeide 
(1922) and Torneo notturno (1929). The first work is a trilogy of which 
the central piece, Sette canzoni, written in 1919, is perhaps the most 
noteworthy. The text of these ‘seven dramatic expressions’ is drawn 
from verses by Lorenzo de' Medici, Poliziano, and Jacopone da Todi, to 
which Malipiero invented a series of short incidents (some inspired by 
personal experiences) which form independent tableaux or ‘panels’ and 
follow one another in quick succession like a cinematographic sequence, 
with exposition of the action and peripeteia unfolding in a few pages. 
The seven episodes are realistic in essential character but tinged 
with fantastic, dream-like colours, alternating between the tragic, the 
macabre, and the comic-grotesque. Unity is achieved by the character 
of the music which is almost exclusively lyrical and more or less inde- 
pendent of the external stage action, but penetrates to the heart of the 
poetic and psychological character of each episode. Malipiero made a 
significant comment on this when he said that for him “һе dramatic is 
what one sees while the music expresses that which one does not see’. 
The musical centre and core of each episode is formed by a canzone 
which is a simple straightforward song in the ancient Italian manner, 
recalling the airs of native peasants and fishermen. 

The first part of the trilogy, La morte delle maschere, though 
written three years after Sette canzoni, is intended as a kind of prelude 
and is in the style of the old opera buffa. The action is an allegory, 
implying caricature and ultimate condemnation, of conventional atti- 
tudes towards art, which are personified in seven figures from the 
commedia dell'arte who, in a symbolical act at the end, are shut away 
in a huge wardrobe. Then Orpheus appears embodying the idea of pure 
idealistic art and introducing the characters who are to play in Sette 
canzoni. The musical delineation of the Masks and of Orpheus is 


! Guido M. Gatti, in the symposium L’Opera di Gian Francesco Malipiero (Treviso, 
1952), p. IX. 


ITALY: LA GENERAZIONE DELL’80 265 


sharply contrasted in style and expression. The third and last part of the 
trilogy, Orfeo, ovvero l'ottava canzone, is intended as a satire on the 
indifference, incomprehension, or sterile enthusiasm which were shown 
by the large public towards Sette canzoni. The opera presents a play 
within a play, spectators on the stage watching the performance of a 
puppet-show, with the Emperor Nero as an insane and ludicrous hero. 
At the end Orpheus appears singing an impassioned air about his hap- 
less fate which touches the Queen's heart, while the rest of the spectators 
are fast asleep from boredom. The music is a most skilful parody of the 
style of Italian grand opera and of Puccinian lyricism. 

Torneo notturno shows the same dramatic and musical pattern as 
Sette canzoni. As in the earlier work, the text is drawn from ancient 
Italian poetry to which Malipiero invented incidents that are arranged 
into seven night scenes, but the characters are far more static and 
puppet-like than in the earlier opera. They symbolize elemental human 
passions in conflict with one another, and the incidents unfold in the 
unreal, fantastic sphere of some timeless myth or legend. In the two 
main characters, the Disperato and the Spensierato, who appear in all 
seven episodes, the composer portrays two fundamental and dia- 
metrically opposed attitudes to life—despair and hedonism; the final 
murder of the Spensierato by the Disperato is a symbolic act implying 
the victory of the life-destroying over the life-asserting forces. The 
ultimate scene shows a funeral procession in a shadowy distance 
suggesting that the Disperato still continues his hopeless search for 
happiness. Like Sette canzoni, Torneo notturno points to a profound 
pessimism—life seen as a vale of tears and as the mirror of death. 
As in the earlier work, a pronounced lyricism serves to create unity and 
coherence, while a musical link between the individual episodes is 
provided by the recurrence of the Spensierato's Canzone del tempo, 
which, according to the composer, represents the dramatic centre: 


Ex. 112 
Andante 


Chi һа tem - poe tem-po as - pet-ta, il 


tem - po per-dei  tem-po fug - ge 


266 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


sei nel tem-po ver - de 


(He who has time and awaits time, wastes time and time flies like an arrow from 
a bow: thus even while you are in your youth . . .) 


Although Malipiero's style in this opera is eclectic in the sense that 
it draws on diverse sources—Gregorian chant, the ancient arioso, old 
dance forms (ballate), Debussyan harmony and Stravinskyan rhythm— 
these heterogeneous elements coalesce into an organic whole which 
shows a distinct individual physiognomy. 

In Те commedie goldoniane (1922), whose spiritual origin is to be 
traced back to his nostalgia for the ancient splendour of his native 
Venice, Malipiero aims at an illustration of the characteristic atmosphere 
of Venetian life in the past: the streets and piazzette, in Bottega de caffé; 
domestic life with its petty intrigues and complications, in Sior Todoro 
Brontolon; and the hustle and bustle in the port and the lagoons, іп Le 
baruffe chiozzotte. The subjects are taken from three of Goldoni’s plays 
but Malipiero greatly simplified the dramatic intrigues, reduced the 
number of characters to a minimum and, in the last opera, combined 
incidents from several other Goldonian comedies. The three works 
revive the spirit of the eighteenth-century comic opera on the basis of a 
more or less symphonic treatment of the orchestra, and whereas the 
recitative is all but completely banished from L’Orfeide, in Tre com- 
medie it plays a more conspicuous role, while the number of lyrical 
arias is restricted. In Le baruffe chiozzotte there is also a song sung in 
Venetian dialect and the imitation of the cries of Venetian street- 
vendors. Another notable opera is La favola del figlio cambiato (1933) 
for which Pirandello provided the libretto. The ancient fairy-tale of the 
changeling is so altered as to make the child of a poor mother, a young 
boy sane and ambitious, become the son of the King while the King’s 
true son is a poor demented creature. The story is to be interpreted as an 
allegory demonstrating that absolute truth is unascertainable and that 
what people believe passes as truth. The most impressive part of the 
drama is the second act which plays in a port tavern, with harlots, 
sailors, and street urchins as a vivid and colourful background to the 
action. The musical treatment is on the whole less happy than in 


ITALY: LA GENERAZIONE DELL'80 267 


Malipiero's earlier operas, but there is sharp delineation of the chief 
characters and some remarkable vocal effects, such as an ostinato 
chorus of sailors and guttersnipes in Act II and internal voices singing 
on one note in Act III. 

Casella's works before the first German War displayed a wide range 
of the most disparate influences—from Mahler and Richard Strauss to 
Debussy, Ravel, and the *Russian' Stravinsky; and though he later 
settled to a more personal style he remained a more eclectic composer 
than Pizzetti or Malipiero. His essential strength lay in a neat and very 
resourceful craftmanship applied either to straightforward pastiche 
(Scarlattiana, 1926; Paganiniana, 1942) or to the resuscitation of older 
music on the basis of modern technical devices. The neo-classicism of 
the inter-war period found in Casella’s cool and speculative mind a 
responsive echo (Concerto for strings, 1927; Concerto for Orchestra, 
1937; Sinfonia, 1940)—all the more so since he was an assiduous student 
of the music of the ancient Italian masters. Apart from his work as a 
composer, Casella was very active in the propagation of music by 
contemporary foreign composers and combined in his person the roles 
of conductor, pianist, writer and lecturer. 

With his orchestral compositions Casella was instrumental in 
establishing a school of Italian symphonic music. The theatre also 
held a great attraction for him and he wrote a number of works all 
marked by a shrewd instinct for theatrical effectiveness. His ballet 
La Giara (1924), to his own scenario, based on a novel by Pirandello, 
aims at a synthesis of ancient opera buffa and rustic comedy and deftly 
employs some Sicilian folk tunes. His most ambitious stage-work is 
the three-act opera, La donna serpente (1932), after Carlo Gozzi’s 
dramatic fable of the same name, and this represents a summary of 
Casella’s mature stylistic development. True, elements from Baroque 
opera, comic opera of the eighteenth century (especially Pergolesi’s), 
Spontini and neo-classicism are juxtaposed, but they are most skilfully 
used to interpret and accompany the dramatic action. The work is set 
in a legendary and spectacular Orient and combines the atmosphere 
of a fairytale with the simple human emotions displayed by the 
characters of the King and the fairy Miranda. The orchestral score 
presents an almost uninterrupted succession of vivid, brilliant, and 
rhythmically inventive ideas though Casella’s imagination remains 
fundamentally cold and dry. In the same year in which he completed 
La donna serpente, Casella wrote the one-act chamber opera, La favola 
d’Orfeo (1932), to a text by Poliziano which follows with slight devia- 
tions the antique myth. The bulk of the work consists of laments, sung 
by Aristeus and Orpheus, which lean towards the manner of Caccini and 


268 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Monteverdi. An epic and objective. character predominates, the 
exception being the ferocious and rhythmically agitated dance of the 
Bacchantes with which the opera closes. 


OTTORINO RESPIGHI, GIORGIO GHEDINI, MARIO 
CASTELNUOVO-TEDESCO : 

Of the composers who belonged to the generation of Pizzetti, 
Malipiero, and Casella and who contributed their share to the national 
renascence of Italian music but show a less sharply defined musical 
profile, three are noteworthy—Ottorino Respighi (1879-1936), Giorgio 
Federico Ghedini (1892-1965), and Mario Castelnuovo-Tedesco (1895— 
1968). Respighi's musical pedigree can be traced to Rimsky-Korsakov 
(whose pupil he was at St. Petersburg), Strauss, Debussy, and, after the 
war, Stravinsky; at the same time he was much attracted to ancient 
Italian music and Gregorian chant. Italian critics divide Respighi's 
mature period into a ‘Roman’ and a ‘Gregorian’ phase, though the two 
overlapped in reality. The first phase, lasting roughly from 1916 to 1926, 
is characterized by sumptuousness and a highly coloured sensuality of 
expression which make Respighi the musical counterpart of d'Annunzio. 
Into those years fall the symphonic poems Fontane di Roma (1916), 
Pini di Roma (1924), and the Trittico botticelliano (1927), which 
represent a combination of mood-pictures, nature impressions, and 
descriptive music. In these works the passages of more intrinsic value 
occur in those parts that are in a muted poetic vein and tinged with a 
slight melancholy, such as in ‘Fontana di Villa Medici’ and ‘Pini 
presso una catacomba’, and ‘L’Adorazione dei Magi’ of the Trittico. 
The Concerto gregoriano for violin (1922) and the Concerto in modo 
misolidico for piano (1924) anticipate Respighi’s later concentration on 
Gregorian chant and modality. Of his eight operas, at least three claim 
attention. Belfagor (1922), to a diverting libretto after the play of E. L. 
Morselli, displays great vivacity and a marked sense of comedy, with 
vocal writing that owes something to Puccini and an orchestral brilliance 
that echoes Richard Strauss. The one-act Maria egiziaca—trittico per 
concerto (1931), which belongs to Respighi’s ‘Gregorian’ phase, has no 
proper action but represents tableaux vivants showing incidents from 
the conversion of St. Mary of Egypt. The music is in a predominantly 
static, lyrical vein and distinguished by a rare transparence of orchestral 
colours. The scene of the sombre and starkly dramatic La Fiamma 
(1933), after The Witch by G. Wiers-Jenssen, is laid in Ravenna in the 
seventh century and makes use of both Gregorian and Byzantine 
liturgical chant, notably in the choral sections of the first act and in the 
music of the Exarch Basilio and of Eudoxia. In Lucrezia (1935), which 


RESPIGHI, GHEDINI, CASTELNUOVO-TEDESCO 269 


was his last opera, the composer reverts to the dramatic recitative of 
the oldest Italian opera, reducing the role of the orchestra to a minimum 
and entrusting the narration of the plot to a singer who is placed in the 
orchestra pit. 

Ghedini's chief distinction lies in a finely controlled sense of form and 
texture and in his extremely solid workmanship best seen in the deft 
manipulation of a close-knit polyphony. These qualities tend, however, 
to outweigh his imaginative power which is rather academic and drily 
abstract. Ghedini has devoted himself to opera, chamber music, and 
choral and orchestral works of which the Symphony (1938) and 
Archittetture for orchestra (1940) are characteristic examples of his 
general style during the inter-war years. 

Castelnuovo-Tedesco, a pupil of Pizzetti, began to develop a more 
personal style during the First World War. His song cycles, Stelle 
cadenti and Coplas (both 1915), are marked by extreme melodic fluency 
and refinement of the technical means of expression, against which must 
be set a lack of emotional depth and intellectual vigour. In his vocal 
compositions Castelnuovo-Tedesco is not merely content to throw into 
relief the imagery of the text but attempts to reproduce its poetic essence 
by the creation of a distinct musical atmosphere. He has set all the songs 
from the Shakespeare plays in the original English (1926), a notable 
undertaking for a foreign composer and all the more remarkable as his 
treatment of the prosody is impeccable. Of the three operas which he 
wrote during the inter-war period, La Mandragola (1926), after Machia- 
velli’s comedy, is perhaps the most successful, and by employment of 
actual folk-songs skilfully re-creates the atmosphere of Renaissance 
Florence. The Overtures to seven Shakespeare plays (1931-42) are 
noteworthy examples of Castelnuovo-Tedesco's orchestral style while 
the Concerto italiano for violin (1924) is a work of great euphony in 
which the earlier arioso is applied to modern instrumental music; but 
it suffers from a certain tonal monotony, all three movements being 
cast in a modal G minor. Castelnuovo-Tedesco's awareness of his 
Jewish origin can be traced in the Second Violin Concerto (subtitled 
The Prophets) (1938), which combines the concerto form with elements 
from the symphonic poem, the three movements characterizing Isaiah, 
Jeremiah, and Elijah. In the Guitar Concerto (1939), written for 
Andrés Segovia, the temptation to imitate a Spanish manner is success- 
fully avoided and much play is made of the contrast between the 
fragile and subtle sonority of the solo instrument and the more compact 
sounds of the orchestra which is, appropriately, of a modest size 


270 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


LUIGI DALLAPICCOLA AND GOFFREDO PETRASSI 


The next generation of Italian composers, who came to full maturity 
in the post-war years, did not ignore the stylistic and spiritual premises 
of old Italian music on which their predecessors had built, but they 
proved equally receptive to many ideas and ideals common among their 
contemporaries outside Italy. Two musicians stand out here— Luigi 
Dallapiccola (b.1904) and Goffredo Petrassi (b.1904). Dallapiccola, who 
shows the most sharply defined artistic profile among the composers of 
this later generation, began with the exploration of sixteenth-century 
polyphony whose technique and spirit he assimilated in a highly indivi- 
dual manner. Characteristic of this phase of his development are the 
three sets, Cori di Michelangelo Buonarotti il Giovane (1933—36), of which 
the first set is for unaccompanied mixed chorus while the two remaining 
sets are, respectively, for a female chorus and seventeen instruments and 
mixed chorus with large orchestra. In the choral writing Italian vocalism 
is paramount, though allied to a strongly marked rhythmic feeling and 
to a texture which displays the signs of a mind steeped in counterpoint: 


Ex. 113 
Vigoroso ; 


Wee с 
Г. е жн тота Ses Sd 
TERLAMAR. IANUSSEREEEE.I.E 1 - аған С | 
Е Е ОЕ ПИН 
EE NR OMNEM ссп ae) 


di - ciam che chi 1а to - glie, da - toa- 


e di - ciam che chi 1а to - glie da - toa- 


271 


DALLAPICCOLA AND PETRASSI 


ve - dràin du-o. ai 


da - to a-ver 


to - glie, 


la 


asd 
[NT 


-ver уе --ага іп duo 


da - to a.-|ver ve 


ve-dra in duo аі, ve - 


duo di, 


drà 


-уег ve- 


Бо На = in-fer-na - 


dia 


=~ VO 


- па dia 


i 


Ми - 


fer - 


= 05а ІП - 


na dia 


ж 


MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


272 


vo-la in-fer 


vo-la in:fer 


dia - vo-la 


- fer 


-Ja in 


vo 


| 


У 


іші 


vo-la in - fer. 


FE 


- fer 


la in 


ТЕ 


(And we say that he who takes her away shall see her become, іп two days, а 


hellish devil.) 


DALLAPICCOLA AND PETRASSI 273 


Тһе more or less straightforward polyphony of the first set (see Ex. 113) 
is offset by the complex vocal and instrumental density of the later sets. 
In the mid-1930s Dallapiccola came into contact with the music of the 
Viennese dodecaphonic school and this was of decisive importance for 
his subsequent development, though in adopting the twelve-note 
method he did not go against his native temperament, which is out of 
sympathy with decadent romanticism and leans towards a Mediter- 
ranean brand of neo-classicism, such as was advocated by Busoni. An 
instructive instance of this are the three Canti di prigonia (1938-41), 
written for mixed chorus, two pianos, and percussion—a combination 
evidently suggested by Stravinsky’s Les Noces—in which Dallapiccola 
handles the serial technique with considerable freedom, combining it 
with pure diatonic writing. Despite their ancient texts (Mary Queen of 
Scots, Boethius, and Girolamo Savonarola), these choral songs are a 
testimony to the composer’s profound social conscience and that com- 
passionate response to the sufferings of political victims in totalitarian 
countries—a compassion to which his second opera, П prigioniero, 
written after the war, gives even more eloquent expression. As a com- 
poser for the stage Dallapiccola displays a remarkable dramatic instinct, 
especially in suggesting states of nightmare and anxiety which he 
achieves partly by the characteristic nature of his thematic material, 
partly by vocal and orchestral devices. This is already seen in his first 
opera, Volo di notte (1939), after Antoine de Saint-Exupéry’s novel. This 
is the story of a South American airline company, prepared to sacrifice 
the life of a pilot in order to inaugurate night flights, thus hoping to 
serve the advancement of modern technology. The action is symbolic of 
the relation between man and machine, the machine taking full posses- 
sion of the human mind. Like Canti di prigionia, the opera is written 
in a free serial style mixing tonality with atonality. Thus, it opens in 
B major and ends in E major but a very large part of the work resorts 
to dodecaphony, the music being based on the following note-row: 


Ex. 114 


Like Berg in Wozzeck and Lulu, Dallapiccola harnesses the individual 
scenes to ‘closed’ musical forms such as “Tempo di blues’, ‘Pezzo 
ritmico', Chorale and Variations. Pure singing is interspersed with the 
Schoenbergian Sprechstimme to which Dallapiccola, following Berg, 
adds a rhythmic declamation half-way between speech and singing. 


274 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Petrassi's development was chiefly determined by his early friendship 
with Casella, who not only influenced him directly but guided his 
artistic outlook towards contemporary movements outside Italy. 
Hindemith and Stravinsky are the two composers who, apart from 
Casella, have contributed most to the formation of Petrassi's style—the 
first with his contrapuntal complexity and ‘motoric’ rhythm and the 
second, more generally, with his non-emotional, objective approach to 
musical composition. The first work demonstrating Petrassi’s essential 
manner was the early Partita for orchestra (1932) which showed a 
remarkable technical maturity, especially in the assurance of the instru- 
mental treatment, firmness of structure and unity of expression. The 
thematic material is sharply defined and this, combined with rhythmic 
vigour, lends the music a sinewy, athletic quality. While here Petrassi 
still leans on the form of Baroque dances (‘Gagliarda’, ‘Ciaconna’, 
“Сіса”), in the Concerto for Orchestra (1934) he penetrates into the 
intrinsic spirit of Baroque music, which is that of an antithesis. Apart 
from the broad contrast between the individual sections, the concerto 
principle is carried further in that the thematic ideas are allotted to 
different instrumental groups, and thus stand out with great clarity of 
outline while, as a corollary, the scoring avoids the blending of colours. 
As in the Partita, the tonal style is that of Hindemith’s “diatonicized’ 
chromaticism. The Piano Concerto (1939), however, shows simplifica- 
tion of both tonality and texture, particularly in the expressive second 
movement—an air with variations in B flat major. For a number of 
years, beginning with 1934, Petrassi concentrated his chief energies on 
choral composition, such as the Ninth Psalm (1936), the Magnificat 
(1940) and the inspired Cori di Morti (1941) which is a dramatic madrigal 
set to a text by Leopardi. The first two works are expressive of a deeply 
religious vein and distantly echo the style of Italian vocal polyphony of 
the sixteenth century, with which Petrassi had come into direct contact 
as a young Roman choirboy. 


BELA BARTOK 

As we have seen, Stravinsky began his career as a nationalist and after 
the First World War shed the national traits in order to evolve a style 
on Western lines. But there were contemporary composers in Eastern 
Europe and in Spain who cultivated and sustained a latter-day nationa- 
lism half a century after the first outbreak of national consciousness in 
music in Russia, Czechoslovakia, and Scandinavia. The case of these 
twentieth-century nationalists was put in a nutshell by Manuel de Falla 
who in speaking of Debussy said that ‘folkmusic is most satisfactorily 
treated by the cultivated musician, not by using authentic tunes but by 


BÉLA BARTÓK 275 


"feeling" them, by realizing the foundations on which they rest and 
conveying the essence of them in music which is all his own.’! Bartók, 
Kodály, Janácek, Szymanowski, Falla, and others have all done this in a 
deliberate manner; theirs is a nationalism which is more a matter of 
essential thought and general aesthetic attitude than a manifestation of 
concrete, tangible features though all have occasionally drawn on native 
folk material. 

Of these composers Béla Bartók? (1881—1945) is the personality who 
combined nationalism and a highly original personal style in the most 
remarkable manner. It is true that Bartók's early training was Western, 
and throughout his career he availed himself of Western forms and 
techniques, especially sonata and variation form, and linear counter- 
point; but the substance of his creative achievement is to be found in the 
close amalgamation of Hungarian elements with his own way of musical 
thinking. His researches into Hungarian folk-song, begun with his 
friend Kodály in 1905, revealed to him rich untapped sources which 
increasingly fertilized his own style and taught him a number of funda- 
mental differences from Western art music. In his Selbst-Biographie,* 
Bartók has enumerated some of these differences: 


The study of this peasant music was for me of decisive importance, for 
the reason that it revealed to me the possibility of a total emancipation from 
the hegemony of the major-minor system. For the largest and, indeed, the 
more valuable part of this treasure-house of melodies lies in the old church 
modes, in ancient Greek and certain still more primitive scales (notably the 
pentatonic), and also shows the most varied and free rhythms and time- 
changes in both rubato and tempo giusto performance. This was evidence that 
the old scales, which are no longer in use in out art-music, have by no means 
lost their vitality. Their application also makes novel harmonic combinations 
possible. This treatment of the diatonic scale also led to a liberation from 
the petrified (erstarrt) major-minor scale and, as an ultimate result, to a 
completely free handling of each single note of our chromatic twelve-note 
system. 


To this must be added the characteristic syncopation (quaver-dotted 
crotchet and semiquaver-dotted quaver, comparable to the 'Scottish 
snap") which is so prevalent in Bartók's and Kodály's rhythmic style and 
derives from the peculiarity of the Hungarian language, in which the 


17. B. Trend, Manuel de Falla and Spanish Music (London, 1925), p. 55. In the Debussy 
number of the Revue musicale (December 1920) p. 209, Falla wrote: ‘Mais tandis que le 
compositeur espagnol [Felipe Pedrell] fait emploi, dans une grande partie de sa musique, 
du document populaire authentique, on dirait que le maítre frangais s'en est écarté pour 
créer une musique à lui, ne portant de celle qui l'a inspiré que l'essence des éléments 
fondamentaux. 

? For Bartók's early works see Chapters I-III. 

3 Musikblütter des Anbruchs, iii (1921), p. 89. 


276 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


tonic accent falls invariably on the first syllable of the word. On the 
other hand, Bartók's technique of repeating short melodic-rhythmic 
units with the effect of creating a ‘strange feverish excitement’ may have 
originated in his studies of Arab music (see pp. 281-2); it forms one of 
the primitive traits of his style comparable to Stravinsky's melodic 
structure in his *Russian' period. 

Bartók began with simple transcriptions of authentic Hungarian, 
Rumanian, and Slovakian folk-songs and went on to use these tunes as 
*material'—varying, breaking them up, and developing them according 
to Western technique. He finally arrived at the invention of themes 
bearing the melodic and rhythmic characteristics of genuine folk-songs; 
in other words, he wrote what have been called ‘imaginary folk-songs’. 
To put it differently, in this last stage the folk element entered the very 
fibre of Bartók's own creative thought and was sublimated into a very 
personal utterance. In order to appreciate fully this fusing of East and 
West in his music, it will be necessary to look more closely at this 
process of gradual amalgamation. 

Up to 1905, the year in which he started to collect and to research into 
Hungarian folklore, his nationalism expressed itself in the use of what 
was then considered to be authentic folk music but was in reality greatly 
distorted, westernized music. Such were the tunes employed by Liszt 
in his Hungarian Rhapsodies and Brahms in his Hungarian Dances. 
These tunes had been adapted to the major-minor system and, since 
they were mostly played by gypsy bands, they were decked out with rich 
embellishments, trills, and grace-notes and often made to conform to 
the so-called ‘gypsy scale’, ‘natural’ minor but with the fourth and 
seventh degrees raised. Bartók himself used such a melody, complete 
with ornaments and augmented intervals, in his Rhapsody for piano of 
1904: 


With the discovery of Hungarian peasant music, which up to 1905 had 
been entirely unknown to professional musicians, a new chapter opened 
in Bartók's creative career. He profoundly appreciated these long- 


BÉLA BARTÓK 277 


forgotten melodies: 'In their small proportions [the Hungarian folk- 
songs] are as perfect as the large-scale masterpieces of musical art. They 
are, indeed, classical models of the way in which a musical idea—in all 
its freshness and shapeliness—can be expressed in the most concise form 
and with the most modest means: in short, in the most perfect way 
possible." As established by Bartok, there are two fundamental types 
of Hungarian folk-song: the first, the more ancient type, which was on 
the point of dying out by the time he and Kodály began their studies, 
shows different features in different regions, though certain traits are 
common to all. By contrast, the more *modern' type, flourishing in the 
nineteenth century, was identical in all regions of the country. What 
are the characteristics of the old songs? First, they are isometric, 
i.e. each line of the verse consists of the same number of syllables and, 
since most of the settings are syllabic, of the same number of notes. 
Secondly, they all show a quaternary structure, one section to one line 
of the quatrain; in other words, a chain-like arrangement patterned 
ABCD, ABBC, ABAB, or AABC in which A never occurs in the last 
section. Thirdly, the majority are pentatonic, others again are modal. 
Fourthly, they invariably show a downward tendency and never start 
with an anacrusis, in conformity with the first-syllable accentuation of 
Hungarian words. Here are two examples to show a pentatonic and a 
modal tune, respectively :? 


Ex. 116 
P Parlando 


1 Das ungarishe Volkslied (Berlin and Leipzig, 1925), p. 4; English translation, London, 
1931. 
* [bid., Exs. 15 and 45. 


278 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


As to the general character of folk-song, Bartók distinguishes two 
major kinds. The first is a slow song in parlando-rubato, free in rhythm 
and of an improvisatory nature: 


[На] Кош -  lotttestim а Ъо-Ког - ba, 
=. 
Piros vérem hull - а hó-ba; 


(I lie wounded in the thicket, 

My red blood trickles on the snow.) 

Of Bartók's own music it is particularly the quartets where we find 
movements in the style of these slow improvisatory songs, e.g. the third 
movement of Quartet No. 4 (Ex. 133) and the ‘night music’ of the 
Andante of Quartet No. 5. Occasionally these slow songs repeat (like 
the related Rumanian дота, see below) the same note, as illustrated 
by the Andante of Quartet No. 5, the Più adagio section of the Second 
movement of the Piano Concerto No. 3, and the opening of the Adagio 
in the Music for strings, percussion, and celesta. 

The second type of authentic Hungarian peasant music is represented 
by tempo giusto songs in strict time and dance rhythm:? 


which like the previous example shows the characteristic Hungarian 
‘snap’. There are instances in which Bartók uses tempo giusto and 
rubato in alternation, notably in works of his early period, an instructive 
example of which is the Introduction to the finale of his Quartet No. 1 
of 1908. 

As to the more *modern' folk-song, Bartók sees its main difference 
from the authentic type in that the first section of its quaternary struc- 


ТЫ Ех. 21. * Ibid., Ех. 61. 


BÉLA BARTÓK 279 


ture is always repeated in the last section (АААА, AABA, ABBA, or 
ABCA), and the song employs modality more frequently than the old 
Кіпа 


Ех. 119 E EN 


Otherwise, the ‘modern’ type is identical with the authentic, having 
absorbed most of its features. 

Bartók also refers to a third type, the ‘mixed’ folk-song which 
possesses certain traits of the authentic song, such as the quaternary 
structure, but reveals the influence of the West in its major-minor 
tonality :? 

Ex. 120 


This ‘mixed’ type, commonly described in the nineteenth century as airs 
favoris, was especially popular with the upper classes of Hungarian 
society—it is the kind used by Liszt and Brahms—and Bartók himself 
occasionally resorts to it, as in the Improvisations for piano (1920)? and 
even as late as 1944, in the fourth movement of the Concerto for 
Orchestra: 


Ex. 121 


ibid), ЕХ 105. 2 Ibid Ех оте 
3 Nos. 3, 4, 5, and 6 are recorded in The History of Music in Sound, x. 


280 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Inevitably, Hungarian folk music engaged Bartók's main interest. But 
he also made valuable studies of the musical folklore of Rumania, 
Slovakia, and the Arabs round Biskra, and occupied himself with the 
folk-songs of Bulgaria, Yugoslavia, the Ukraine, and Turkey. In 
Rumania he found four different kinds: the colindă or Christmas song, 
the dirge, instrumental dance music, and music not associated with any 
particular occasions. To this last variety belongs the doind or horă 
lungă preponderant in the regions of Maramures and Ugocsa, which is, 
perhaps, together with the dance music, the most important product of 
Rumanian folk music. Like the slow Hungarian song, the doind is in 
parlando-rubato style, free in rhythm and improvisatory but, in addition, 
marked by chant-like, incantatory passages on the same note! and the 
fairly frequent use of the downward third :? 


Ex. 122 
Parlando, rubato 


hei - hai - hai Te-ai te-mut de me-ste - rit. 


(Hey, my sweetheart, when I talked to you you were afraid I might bewitch you.) 


How Bartók uses Ше doind іп a stylized, highly individual form may be 
seen, for instance, from the first movement of his Violin Sonata No. 2 
and, particularly, the Trio in the Marcia of the Quartet No. 6, in which 
the falling third is very prominent: 


1 These are sometimes also found in the Rumanian dirge. 
2 Bartok, Die Volksmusik der Rumänen von Maramures (Munich, 1923), Ex. 23. 


BÉLA BARTÓK 281 


Ex. 123 
@ Molto moderato DN => ж 
f) m mM A y З E 


In his tempo giusto movements Bartók resorts very frequently to the 
characteristic dactylic-trochaic pattern of the Rumanian folk dance.! 


while his other favourite rhythmic pattern originates in Slovakian 
music: 


Ex. 125 


24. 4 JU - E 


In the summer of 1913 Bartók made a journey to the region of Biskra, 
in Algeria, to study Arab music which he found to be very limited in 
compass, mostly moving between two or three neighbouring notes and 
constantly repeating tiny melodic fragments. This primitive structure, 
however, is strongly counterbalanced by the complex polyrhythmic 
character of Arab music as seen, for instance, in a woman-kneja or 
nuptial song:? 


! Tbid., Ex. 166. 
? Bartók, Die Volksmusik der Araber von Biskra und Umgebung (Leipzig, 1920), Ex. 26. 


282 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 
Ex. 126 


Voice 


Percussion 


Bartók fastened on this polyrhythmic aspect and applied it, for instance, 
in the Sonata for two pianos, notably the second movement, while the 
perpetual reiteration of short melodic-rhythmic figures of narrow 
compass is found in many of his works and is strikingly illustrated by 
the opening of the finale of his Dance Suite of 1923: 


Ex. 121 


Bartók's researches into the folk music of the Ukraine, Yugoslavia, 
Bulgaria, and Turkey were not carried far enough to allow him to arrive 
at any definite conclusions. What he wrote, for instance, on the irregular 
*Bulgarian rhythm' has now been shown to be true also of the music 
of the Greeks, Turks, Armenians, Berbers and Hindus.! Here is an 
example of a Bulgarian folk-song in an ‘irrational’ rhythm:? 

1 John W. Downey, La Musique populaire dans l’œuvre de Béla Bartók (Paris, 1964), 


ӘЗІРЛЕ 
? Quoted p. 154, from Vasil Stoin, Narodni pesni ot Timok ао Vit (Sofia, 1928). 


BÉLA BARTÓK 283 


Bartok introduces such rhythmic patterns into the Six Pieces in Bulga- 
rian Rhythmin Vol. VI of his Microcosmos, in the Scherzo and Trio of his 


Quartet No. 5 E and et and the finale of Contrasts 


(8 +5). 

8 

As regards Serbo-Croat folk-song, Bartók began to study it system- 
atically only in 1941-2, when he examined various collections by 
other investigators. There he found one type, particularly widespread in 
Dalmatia, which is in two parts, the second voice holding a pedal or 
moving, heterophonically, in seconds and sevenths with the first voice. 
The following example opens with two sopile, double-reed wind instru- 
ments, which are always used in pairs:! 


Ex. 129 


2 Sopile 


Bartók had used seconds and sevenths to thicken out a line and thus 
lend it greater pungency long before these studies, in fact as early as 
1907, in his 14 Bagatelles for piano. But his late occupation with Serbo- 
Croat folk music bore fresh fruit in the two-part writing in major 
sevenths and minor seconds in the Giuoco delle coppie of his Concerto 
for Orchestra. 

1 Quoted in Bartók and A. Lord, Serbo-Croatian Folk Songs (New York, 1951), p. 63. 


284 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Another fingerprint of Bartók's harmonic style also had its origin in 
folk music. In the folk-songs of the Balkan peninsula there is no clear 
major and minor third but a ‘neutral’ interval higher or lower in pitch 
than the Western interval. This is the natural coroliary of the absence 
of a definite feeling for a major or minor key. Bartók absorbed this 
feature into his own music in two ways. He either combines the major 
and minor third in a single chord as, for instance, in Хо. 143 of his 
Microcosmos and the finale of his Quartet No. 5 (Ех. 130, i and ii); or he 
extends this ‘neutralization’ by casting one part into major and the 
other into minor, thus arriving at a simple bitonality, as in No. 59 of 
Microcosmos (Ex. 130, iii): 


TEE PORE RES 
hihi р mw 

EI А o 2 ee м НЕ 
ть зи 


АЕ ре rec 


To sum up, Bartók's amalgamation of East European folk-song 
with his own style was a gradual process, beginning with simple 
transcriptions and arrangements in which a peasant tune is taken 


BÉLA BARTÓK 285 


either unchanged or only slightly modified, a piano accompaniment 
and, perhaps, a few bars of introduction and postlude added. Bartók 
himself compared this to Bach's treatment of the Protestant chorales 
some of which were originally popular songs.! In a later stage he 
took the native tune as 'material to be altered, developed, and 
elaborated; and the final phase was the invention of ‘imaginary’ 
folk tunes, ie. melodies springing from the composer's mind and 
bearing characteristics of both his own individual style and that of 
authentic folk-song fused into an organic, indivisible unity. Pointers in 
this direction can be clearly discerned in his Quartet No. 2 and the ballet 
The Miraculous Mandarin (1919); but it was not until the First and 
Second Violin Sonatas (1921 and 1922 respectively) that Bartók's 
advance towards a realization of his new ideal became manifest. The 
more obvious Hungarianisms are confined to the last movements which 
are in the nature of primitive folk dances, while in the remaining move- 
ments he applies a novel expressionistic manner. The First Sonata is 
larger in scope than the Second, and has an opening movement bursting 
with a passionate and sweeping melodic line in the violin and with 
savagely percussive chords in the piano. The obscuring of the underlying 
tonality—an infinitely extended C sharp minor—is almost complete and, 
availing himself of the advanced harmonic devices of contemporary 
Western music, Bartók acquired a tonal sense very different from 
that shown in his previous compositions. The individual notes of some 
themes imply a kind of ‘diatonicized’ chromaticism which in appearance 
is almost identical with serial writing and greatly differs from the 
intense, Wagnerian chromaticism encountered in such an early work 
as the first movement of the First String Quartet (1908). The Second 
Sonata is more concentrated and more economical, consisting of 
two parts only, which stand to each other in something like the relation 
of lassú and friss of the verbunkos style—popular Hungarian music 
of the eighteenth and early nineteenth centuries which was played at 
army recruiting (German, Werbung). The first part is in tempo rubato, 
slow and recitative-like, while the second is in fact tempo giusto and 
shows a rondo-like arrangement of varied dance sections. The two 
parts have the main thematic material in common; but in contrast to 
the practice of the classical and romantic duo sonatas, the violin (which 
is leading) and the piano of Bartók's two sonatas are allotted for the 
most part separate material, which emphasizes the individual character 
of the two instruments. Instead of interweaving them Bartók makes 
them complementary to each other, which has the effect of apparent 
contrast but ideal unity of thought. Both works show classical sonata 
1 The Influence of Peasant Music on Modern Music (New York, 1959), p. 71. 


20 


286 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


form, but Bartók varies the expository thematic material in the recapitu- 
lation almost beyond recognition and thus achieves a psychological 
rather than actual reprise. 

In the Dance Suite (1923), which consists of five dances with a 
ritornello and a finale, Bartók reverts to the half authentic, half ima- 
ginary type of Hungarian folk-music, but the following Piano Sonata 
(1926), the suite Out of Doors for piano (1926), and the First Piano 
Concerto (1926) show a modification of the style initiated in the 
violin sonatas. The predominant feature of the Piano Sonata is its 
extreme percussiveness, with explosive chord-clusters compounded of 
seconds, fourths, sevenths, and ninths, and the structure built up by the 
ostinato-like repetition of short melodic phrases of a limited compass, 
whose connexion with Hungarian folk-music is so attenuated as to be 
almost non-existent. The dynamic thrust and power of the music leaves 
no room for lyricism and even the meditative slow movement avoids all 
sustained melody; in spirit the work is akin to Ше Allegro barbaro of 
fifteen years earlier, but the writing is far more concentrated and un- 
compromising. Although Out of Doors shows the same application of 
percussive and repetitive devices as the Piano Sonata, it is a more 
ingratiating composition and programmatic in character. The five 
pieces of which it consists are mood-pictures of which the fourth, 
‘Night Music’, is the first fully developed example of Bartók's extra- 
ordinary sensitiveness to the sounds of a nocturnal world, with the 
chirpings, twitterings, and calls of night animals and insects reproduced 
in an impressionistic manner that owes almost nothing to Debussy but 
represents the composer's individual contribution to evocative music. 
The First Piano Concerto continues in the percussive martellato style 
of the Piano Sonata, but the sounds produced are less harsh and spiky 
than in the former work. This is due partly to the chordal texture being 
largely colouristic, and partly to the fact that a great deal of writing is 
dissolved into contrapuntal lines, whose clashes refuse to be interpreted 
in a vertical sense and pass very quickly. The individual strands of the 
polyphonic fabric are drawn succinctly and stand out with great clarity, 
features which may be traced back to Bartók's intense study of Bach 
and to his editing of Baroque music (Couperin, Frescobaldi, Scarlatti, 
and other composers). The influence of the Baroque concerto can be 
seen in the relation between the piano and the orchestra in which there 
is no clear-cut division and, though there are passages (as in the Baroque 
concerto) where the piano becomes a true solo instrument, especially in 
the slow movement, the general effect remains one of a competition 
between equals, the piano playing the role of a primus inter pares. As in 
the Piano Sonata, the material is fragmentary, with a predominance of 


BÉLA BARTÓK 287 


scale motives that are very much alike, and the feeling for a definite 
tonality is very vague: to say that the Concerto is “іп” E minor is merely 
to indicate a general tendency of the music towards a tonal pole. 
During this period, initiated by the two violin sonatas and lasting till 
about 1930, Bartók advanced to the utmost limits of his intellectual 
exploration, cultivating an expressionistic, abstract, and esoteric style 
of utterance. He appears during these years to be almost exclusively 
concerned with the projection of an inner world in which the dividing 
line between the conscious (rational) and the instinctive (irrational) is 
virtually non-existent. This is perhaps best seen in the Third String 
Quartet (1927) which in style represents a ne plus ultra in concentra- 
tion and subtilization of thought. Like the Beethoven of the late 
quartets and piano sonatas, Bartók in this work seems to be communing 
with himself rather than attempting communication with the outside 
world. This is the main reason why this string quartet is the least 
accessible of his six, making extraordinary demands on the listener's 
perceptive powers; yet it 15 no less a masterpiece than the subsequent 
quartets, differing from them only in the inaccessibility of its aesthetic 
and technical premises. The six quartets occupy a central position in 
Bartók's creative career and may be likened to a diary to which the artist 
confided his most intimate intellectual and emotional experiences and 
adventures. Each of the six quartets stands at the culmination of a 
different phase of Barótk's artistic growth, summing up the essential 
problems, tendencies, and aspirations characteristic of each stylistic 
stage. Just as Beethoven's seventeen quartets represent the apogee of the 
classical form, so does the series of six Bartók quartets mark the con- 
summation of the modern genre. For profundity of thought, imaginative 
power, structural logic, diversity of formal and textural features, and 
enlargement of the technical scope, they stand unrivalled. This quality 
of uniqueness is further enhanced by the following consideration: one 
of the mainsprings of Bartók's art was the attempt to achieve a perfect 
synthesis between East and West, and in his quartets this synthesis may 
be said to have been most nearly achieved. We can realize the full 
measure of this imaginative feat if we bear in mind that this fusion 
between two different musical cultures was brought about in a medium 
which has come to be regarded as the purest and most subtle manifesta- 
tion of Western musical thinking. To have harnessed the instinctive 
primitive forces residing in Hungarian music to the most intellectual of 
Western musical forms—therein lies the historic significance of Bartók's 
quartets. This fusion, however, was achieved at a price. The criticism 
that must be levelled against Bartók's quartet style, as it may be also 
advanced against Beethoven's from the Razumovsky quartets onwards, 


288 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


is that it frequently bursts the framework of the medium. In Beethoven's 
case this was due to a powerful symphonic urge, while with Bartók it is 
the dynamic percussiveness of his harmonic and rhythmic language 
that contradicts the intimacy and ‘inwardness’ associated with the 
quartet medium. 

The Third String Quartet is the shortest and, in expression, most 
intense of the six, consisting of one continuous movement of a highly 
unusual formal design. It is divided into four sections which follow one 
another in the simple pattern slow-fast-slow-fast and these are themati- 
cally closely related in a manner best illustrated by two interlocking 
arches: 


Part 1 Part 2 Recapitulation Coda 
of Part 1 


Part 1 furnishes a striking example of Bartók's technique of developing 
virtually an entire movement from a single germ-cell—here a pentatonic 
figure consisting of a rising fourth and descending minor third. As the 
last section of the ternary Part 1 proves, this basic idea derives from a 
broad sustained theme of Hungarian flavour which was probably 
Bartók's initial idea, but, instead of stating it at the beginning, he seizes 
on its most characteristic motive and evolves from it the major part of 
the movement before revealing the source from which it sprang. The 
immense regenerative force inherent in this germ-cell may be seen from 
the following examples: | 


ал 
dii) : (iv) = 


BÉLA BARTÓK 289 


The recapitulation of Part 1 condenses the expository material to such 
an extent that a few bars have to do duty for a dozen bars or so of the 
exposition (the recapitulation is seventy bars long, as against 122 bars 
of Part 1). Part 2 clearly springs from native dances, with the main 
theme harmonized in ‘primitive’ parallel triads, and the music is 
driven along by an elemental rhythm in which percussive chords and 
syncopations predominate. Its form is that of a sonata-cum-variations. 
The treatment of the main subject proceeds by imitation and stretto; 
and while the melodic structure undergoes comparatively minor altera- 
tions, there are frequent changes of the metric units (2/4, 3/4, 3/8, 5/8, 
and 6/8). The second subject provides the material for the ‘development’ 
which takes the form of a series of variations culminating in a charac- 
teristic Bartókian fugue—a fast scurrying piece in /eggiero style and with 
subdued dynamics. (Unlike the fugues of Beethoven’s last works, 
Bartók's do not serve a dramatic purpose nor are they intended as an 
intellectual catharsis; they are motoric in character and serve to 
intensify the rhythmic drive of the movement in which they occur.) The 
coda 15 a much transformed reprise of Part 2, changing from Allegro to 
Allegro molto, with the contrapuntal texture becoming denser still 
(canons and inversions in stretto), until towards the end the music 
takes on a savagely aggressive character through the use of percussive 
note-repetitions, incessant ostinatos, double-stops, glissandi, and rising 
and falling arpeggios. The work closes on a chord of three superimposed 
fifths based on the note C sharp—the ‘tonic’, which, except for the very 
beginning and a brief passage in Part 1, is never in evidence until the 
concluding bars. 

If the Third String Quartet represents Bartók's farthest advance in 
the direction of intellectual severity, uncompromising harshness of 
emotional expression and formal experiment, the Fourth (1928) marks 
a certain retreat from this extreme position, though it still has some hard 
things to say. Not that Bartók's advanced style undergoes an intrinsic 
change. The writing remains predominantly linear, without regard 
for the violent dissonances produced in the vertical texture; and 


290 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


there is a predilection for scholastic ‘automatic’ devices (canon, imita- 
tion, stretto, inversion and retrograde motion) which at times creates 
the impression of excess, a feeling that Bartók is straining these devices 
and making too self-conscious a use of them. But he does handle them 
with a remarkable freedom and it is not often that he sacrifices the 
shapeliness and equipoise of his melodic lines to Ше demands of strict 
contrapuntal logic. Moreover, the gains acquired in the Third Quartet 
are now consolidated on a broader formal basis, and such sections as 
the wonderful elegy of the third movement and the delightful serenade 
of the fourth suggest that Bartók's personality was maturing and 
becoming accessible to gentler and more relaxed moods. But the quest 
for unity and for correspondences between the movements by novel 
means continues. In the Third Quartet Bartók interlocks what are in 
fact two different movements. In the Fourth Quartet he resorts to the 
so-called Bogen or arch form in that its five movements follow in the 
pattern A-B-C-B'-A' in which the two outer and the two inner move- 
ments, respectively, mirror one another while the central piece C stands 
by itself. In other words, the music progresses from A to C and then 
retraces its steps back via B' to A': 


Allegro — Prestissimo — Non troppo lento - Allegretto — Allegro molto 


(A) ni (C) LP (А) 


The links between Ше respective outer and inner movements are по! 
simply confined to thematic correspondences but also embrace the 
general structure and the character of the music. The first and last 
movements derive their entire material from a germ-cell of six notes 
in the seventh bar of the opening Allegro: 


Ex. 132 
1st movement bar 16 


BÉLA BARTÓK 291 


Ex. 132 (i) looks an unpromising enough motif but it is the measure of 
Bartók's art of organic thinking that from it he evolves a wealth of 
thematic ideas with which he builds up the constantly changing texture 
of the first and last movements. This he does by stretching and contract- 
ing the intervals, forming diatonic or chromatic versions of the same 
figure, subjecting it to fragmentation, inversion and retrograde motion, 
and rhythmic variation. It would be extremely hard to detect the 
inner connexion between the original idea and some of its meta- 
morphoses, without hearing the intervening stages. As one of his recent 
biographers remarks,! it is because Bartók allows us to share his 
thought-processes rather than leaping from the basic motive to its 
furthest transformation that his music carries with it such utter 
conviction. The first movement is an abstract piece wholly concerned 
with problems of form, design, and texture; the finale, on the other hand, 
is a ferocious Hungarian dance which at times oversteps the proper 
string quartet medium by producing an almost orchestral sonority. The 
relationship between the second and the fourth movements is on 
several levels: both share the same thematic material, both are light- 
weight scherzos, and both are played in a special manner—the Prestis- 
simo with muted strings and the Allegretto pizzicato. Yet there is 
a marked contrast of mood between these two movements: while the 
former is a kind of moto perpetuo to which the muted strings lend 
a strange shimmer (not unlike the effect produced by the Allegro 
misterioso of Berg's Lyric Suite), the latter has a delicate, guitar-like 
accompaniment in the manner of a serenade. It 15 in this fourth 
movement that Bartók first uses the ‘snap’ pizzicato in which the 
strings of the instrument are to be plucked with such a force that they 
rebound off the fingerboard with a percussive sound. The only point of 


1 Halsey Stevens, The Life and Music of Béla Bartók (New York, 1953), p. 188. 


292 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


repose in this dynamic quartet is provided by the lyrical espressivo of the 
slow central movement, with its grave, wistful cello recitative of Hun- 
garian character in the first section. Here Bartók recalls music peculiar 
to the tárogató, a woodwind instrument of ancient (Eastern) origin 
whose dark colour is akin to that of the chalumeau register of our 
clarinet: s 


Ex. 133 
Non troppo lento 


() 
Ist & 2nd EAE 
Vins. 


The middle section is an exquisite atmospheric study of the sounds of 
nocturnal nature. The Fifth String Quartet (1934) followed the fourth 
after an interval of six years, yet intrinsically they are sister works, 
though in the later composition the process of intellectual relaxation 


BÉLA BARTÓK 293 


and lyrical expansion is carried a stage further. Significantly, it has two 
slow movements whereas the previous quartet had only one; and its 
melodic lines show a more clear-cut articulation and grow in amplitude 
almost in the manner of themes in the established sense, while the 
harmonic style is on the whole less astringent. Yet the formal organiza- 
tion is the same as in the Fourth Quartet—again an arch form, 
A-B-C-B'-A', in which C is now a scherzo flanked by two slow 
movements in which Bartók achieves a new height of evocative lyrical 
poetry. The correspondences between the opening movement and the 
finale comprise the sharing of the same thematic material, the casting of 
both movements in sonata form, and a clear definition of tonality (B 
flat). Both movements show an abundance of inversion, canons, stretti, 
and passages in invertible counterpoint—the recapitulation of the 
Allegro, for instance, reintroducing the expository material in its 
*mirror' form. The finale is notable for the inclusion of a fugue which in 
its percussive note-reiteration on the two violins, col legno, and a 
drone-like ostinato on the cello, constitutes an illuminating example of 
Bartók's fusion of an Eastern and Western style. It also contains a 
strange episode, marked 'Allegretto con indifferenza', with a trivial 
tune in А major and harmonized in deliberately commonplace fashion 
with alternating tonic and dominant chords, the whole to be played 
*meccanico'. The barrel-organ effect thus produced is like a grimace or 
mocking sneer, recalling Mahler's use of banal tunes to suggest the 
commonplace aspect of life. The second and the fourth movements— 
an Adagio and Andante—share the same ground plan (A-B-A plus 
coda) and the same material. Both open in an atmospheric manner, the 
fragmentary motives and trills of the Adagio corresponding to the re- 
peated pizzicato notes, slurs and gruppetti of the Andante, and both 
movements contain a chorale, a feature almost invariably associated 
with Bartók's *night music'. The central piece is an engaging scherzo in 
alla bulgarese rhythm for which Bartók showed a special predilection in 
his late period. The asymmetrical rhythmic pattern of the scherzo proper, 
(4--2--3) 
8 
feature of the latter is a delicate arabesque formed by а ten-note 
ostinato which is repeated fifty-nine times, with an enchanting melody 
in the vein of Hungarian children's song set against it. With the Sixth 
String Quartet (1939), written five years after the Fifth, we reach the 
‘classical’ stage of Bartók's development, which also includes such 
works as the Music for Strings and Percussion, the Sonata for Two 
Pianos and the Violin Concerto. This last period is characterized by a 
greater simplicity of form and technical devices, by the invention of 


, grows more complex in the trio and a most noteworthy 


294 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


themes of a broader, more sustained character, by a marked reduction 
of grating dissonances, by a greater transparency of texture and a 
sharper tonal perspective. These changes, which were first noticeable 
in the Fifth Quartet, reach their consummation in the Sixth. Thus it is 
significant that Bartók reverts to the classical scheme of four contrasting 
movements, that he anchors the work in a D major-minor tonality 
and that he employs a simpler, less intellectual device than the arch 
form in order to achieve formal unity between the movements. He 
resorts to a motto theme first announced at the very beginning by the 
viola which is a mournful melody of a marked drooping tendency, most 
beautifully shaped and balanced and (like the variation theme of the 
Violin Concerto) constituting an exquisite example of Bartók's skilful 
fusion of the melodic and rhythmic inflexions of a Hungarian melody 
with his own personal manner: 


Ex. 134 


Each of the first three movements is prefaced by this motto theme, but 
on each successive occasion it appears in a slightly different and tex- 
turally richer form, until in the finale it becomes the actual material of 
the movement, expanded and subjected to elaboration and at last 
revealing its full emotional significance. At the other extreme stands the 
gay, even exuberant opening movement. With the Marcia and Burletta, 
however, Bartók returns to the style of his expressionistic period; the 
march is harsh and aggressive and the Burletta's humour is grim and 
sardonic, with the first violin playing quarter-tones against the 'true' 
notes of the second violin and vice versa, thus creating a deliberate ‘out 
of tune’ effect. In the Burletta Bartók's ‘barbaric’ tendency is particu- 
larly emphasized in the fierce harmonic clashes, in glissandos and 
percussive chords to be played ‘at the heel’ of the bow. Indeed it is hard 


BÉLA BARTÓK 295 


to resist the impression that the work possesses an extra-musical 
significance or an implicit programme. 

The Second Piano Concerto (1931) is permeated by a feeling of 
extrovert exuberance comparatively rare with Bartók, its sinewy 
athletic strength proclaiming the composer at the height of his creative 
powers. Its tonic is an.unambiguous С major and though the themes are 
still short and fragmentary, their outline is clear-cut and readily re- 
tained in the listener's mind. The formal structure is more articulate 
than in the First Concerto, the first movement being in sonata form, 
with the themes inverted in the recapitulation, while the finale consists 
essentially of the varied alternation of two ideas, recalling Haydn's 
favourite principle of construction. The slow movement combines an 
Adagio with а Scherzo in the form А-В-А, section А being an example 
of the composer’s ‘night music’ style in which a string chorale, harmon- 
ized in five superimposed fifths, alternates with recitative-like passages 
of an improvisatory character on the piano. The work may be said to 
stand halfway between the Baroque type of the First Concerto and the 
classical layout of the Violin Concerto, with the solo part frequently 
rising from an obbligato to full individual status, though the allocation 
of separate material to the piano and the orchestra still points to the 
eighteenth-century model. 

Bartók's interest in the rhythmic and colouristic possibilities of a 
percussion ensemble in combination with other instruments first 
showed itself in the two piano concertos; but it was not until the second 
half of the 1930s that he composed two works in which he explored to 
the full the potentialities of such a combination. Although the Music 
for Strings, Percussion and Celesta (1936) and the Sonata for Two 
Pianos and Percussion (1937) belong together as bold experiments 
with novel and fascinating sonorities, in the first work the percussion is 
intimately integrated with the rest of the orchestra, while in the second 
the percussive instruments are frequently treated in opposition to the 
two pianos. The Music for Strings is in effect a chamber symphony in 
four movements and is the more highly organized of the two works, 
demonstrating Bartók's ingenious use of Lisztian cyclic form and 
theme-transformation. The entire composition springs virtually from a 
single theme which in its narrow, mostly chromatic, intervals and its 
emotional expression is similar to the motto theme of the Sixth String 
Quartet. In the first movement it forms the subject of a five-part fugue 
and the successive entries of the fugal subject are on alternating rising 
and falling fifths from the initial A to the final E flat, the 'key' in which 
the climax is reached after being approached simultaneously from 
both the higher and the lower circle of fifths. The process is then 


296 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


reversed in a condensed form, with the fugal subject now turned 
upside down, and in the coda both its original and inverted versions 
appear in combination in the *key' of A, the tonal pole of the movement. 
Dynamically, the movement represents a long crescendo-decrescendo, 
blazing into white heat at the climax and mirroring on a large scale the 
arch described by the fugal subject in four bars. There are no counter- 
subjects, no episodes, and no free contrapuntal lines—the whole 
fabric is woven into a single closely-knit organic structure. The remain- 
ing three movements, though introducing some fresh material, are 
essentially an exploration of the implications of the fugal subject in 
different directions. 

The choice of a double string orchestra in combination with piano, 
harp, celesta and percussion instruments of definite and indefinite pitch 
provided Bartók with an immensely wide and varied range of timbres 
which he exploits with uncanny resourcefulness. The two string orches- 
tras are used both in combination and antiphonally, with and without 
the percussion, which is employed to produce both traditional and novel 
sound effects. Just as the Music for Strings exploits both the expressive 
and percussive qualities of a chamber orchestra, so does the Sonata 
exploit these qualities within the more confined medium of two pianos 
and percussion. Bartók was himself a distinguished pianist and a wealth 
of characteristic piano devices are applied to the two instruments, which 
are treated on a completely equal footing. The percussion comprises 
three kettle-drums, a bass drum, two side drums, two cymbals, triangle, 
tam-tam, and, as the only melodic instrument, a xylophone, the whole 
to be handled by two players. In the opening movement the percussion 
is largely used for underlining and emphasizing the piano rhythm, 
while in the two remaining movements it acquires a solo role, the 
xylophone being entrusted with numerous passages of thematic impor- 
tance. In 1940 Bartók transcribed the work as a Concerto for Two 
Pianos and Orchestra, and in this version the role of the percussion is 
less conspicuous, with a corresponding loss of the particular sonorous 
quality characteristic of the original version. 

From the same year as the Sonata for Two Pianos and Percussion 
dates the completion of Microcosmos (1926-37) which consists of 153 
piano pieces of varying length and of progressive technical difficulty. 
Microcosmos is in the first place a didactic work intended to develop a 
pianist's facility at specific stages of his technique, but it is also a com- 
pendium of the essential devices of modern music as seen in the light of 
Bartók's own style—chords of the fourth, major and minor seconds, 
cluster harmonies, modality, bitonality, whole-tone and other scales, 
contrapuntal techniques such as canon and inversion, and special 


BÉLA BARTÓK 297 


rhythmic patterns such as syncopation and asymmetrical rhythms, of 
which the Dances in Bulgarian Rhythm are the finest examples. 

The Violin Concerto (1938) illustrates the perfect equilibrium of all 
that is characteristic of Bartók's mature style. Thus, the Hungarian 
elements are subtilized to such an extent that they become intangible; 
they are like the scent that clings to a plant transplanted from an exotic 
soil. And while the music is immensely virile and shows that intellectual 
passion peculiar to Bartók, it is punctuated by a powerful expressive 
emotional note; this amalgamation of the dramatic with the lyrical is 
perhaps the most remarkable feature of the concerto. Bartók's mellow- 
ness is also evident in a number of technical features. Instead of short 
fragmentary ideas we now find song-like themes of a sustained character; 
the formal design is spacious, yet it creates the impression of the utmost 
concentration on account of the extraordinary vitality and cogency of 
the composer's musical thinking; there is a clear feeling of tonality (B 
minor), a less pronounced use of contrapuntal devices, and the soloist 
assumes a dominating role in a part marked by great virtuosity. The 
orchestral texture shows a noticeable increase in transparency, Bartók's 
former style of block instrumentation being replaced by a more indivi- 
dual treatment of single instruments with much subtle blending of 
colours, especially in the second movement. The work is conceived as a 
kind of variation; in addition to the second movement, which consists of 
theme and six variations, all the thematic material of the opening move- 
ment is repeated in varied guise in the finale, and this correspondence 
extends also to the architecture, the finale duplicating the formal design 
of the first movement (which is in sonata form) almost section by section. 

In America Bartók wrote four major works—the Concerto for 
Orchestra (1943), the Sonata for Solo Violin (1944), the Third Piano 
Concerto (1945) and the Viola Concerto (1945), which remained un- 
finished. The Concerto for Orchestra was written for the Boston 
Symphony Orchestra which explains the extraordinary brilliance of the 
writing, single instruments and groups of homogeneous instruments 
(strings, woodwind, and brass) being treated in a virtuoso concertante 
style. The work is in fact a symphony or, better, a symphonic suite in 
five movements. Like the Violin Concerto, it represents a crystallization 
of all the features of Bartók's maturity—lucid textures in which contra- 
puntal devices are manipulated to vivid effect, trenchant rhythms and a 
strongly affirmed, though still widely extended, tonality. If anything, 
the Hungarian-inspired character of certain themes is more pronounced 
than in the previous work; the chain-like arrangement of the second and 
third movements also follows the pattern of multiple structure charac- 
teristic of old Hungarian folk-songs (p. 277). Thus, in the second 


298 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


movement, *Giuoco delle coppie' (Play of the Couples), which takes the 
place of a symphonic scherzo, the chain structureis seen in the successive 
introduction two by two, each pair with its own theme, of bassoons in 
sixths, oboes in thirds, clarinets in sevenths, flutes in fifths, and muted 
trumpets in major seconds. The Sonata for Solo Violin stands apart 
from Bartók's ‘American’ work in that it reverts to Ше ‘difficult’ 
expressionistic style of the Third and Fourth String Quartets, at any 
rate so far as its first two movements are concerned. The work is severe 
both in emotional expression and in intellectual content, and to this 
severity must be added the deliberate angularity and austerity of the 
style. This is partly the corollary of this particular medium, which is 
restricted by the absence of a bass to provide tonal balance, the impos- 
sibility of sounding more than two notes together and the consequent 
necessity of breaking chords in triple and quadruple stops. The whole 
work represents a technical tour de force of extraordinary Ди сийу. 
Extremes of range are employed in all four movements and much use is 
made of simultaneous arco and pizzicato and of harmonics and 
glissando. The Fugue (second movement), whose subject, Ex. 135, with 


Ex. 135 
Risoluto, non troppo vivo 


its narrow compass and chromatic steps, is most characteristic of Bartók, 
makes perhaps the greatest demands on the player though the first 
movement, in ‘Tempo di ciaccona’ but in sonata form, is scarcely less 
exacting. It is in the Melodia, whose long-spun chromatic line is 
subsequently subjected to variation treatment, and in the final Presto, a 
rondo with three episodes, that a more “ореп” style of writing and a 
more relaxed mood are discernible. The influence of Bach's solo violin 
sonatas is felt not only in the technical treatment of the instrument but 
to some extent also in the neo-classical conception of the Tempo di 
ciaccona and the Fugue. Hungarian memories affect the melodic and 
rhythmic invention of the last two movements in a most subtle way. 

The Third Piano Concerto was the last work Bartók was able virtually 
to complete before his death. It illustrates a further move towards 
structural and tonal simplification characteristic of Bartók's late style, 


ZOLTAN KODÁLY 299 


the texture being light and the orchestration of great transparency, 
especially in the chorale and the ‘night music’ of the slow movement. 
The solo part is far less percussive and less exacting than in the first two 
piano сопсегіов yet it is of sufficient brilliance to stamp the work as a 
true virtuoso concerto. 

Looking back at Bartók's total achievement, one sees that several 
features stand out with great clarity. First, like Stravinsky but to a 
higher degree, he succeeded in an organic fusion of Western art-music 
with Eastern folk music, bringing all the technical resources of the West 
to bear upon native material. In his technique of composition Bartók 
was one of the most sophisticated Western composers, but his spiritual 
allegiance belonged to the ‘primitive’ East, whence he derived much of 
the freshness and seeming spontaneity of his melodic style and the 
immense vitality of his rhythmic invention. Secondly, while adhering 
to the traditional forms of sonata, rondo, and variation, he displayed 
supreme mastery in designing novel schemes and patterns best illus- 
trated in his six string quartets, the Sonata for two pianos, the Music for 
strings, celesta and percussion, the Concerto for orchestra and the 
Sonata for solo violin. The same consummate command is shown in 
his management of the rhythmic and instrumental texture and his use of 
orchestral timbre, notably in the genre of his impressionist ‘night 
music’ which sprang from Debussy, but which Bartok developed into a 
highly original utterance. And thirdly, of the three musicians who 
dominated the musical scene during the first half of the twentieth 
century—Stravinsky, Schoenberg, and Bartók—it is the Hungarian 
master who, despite his immense intellectual control, remained nearest 
to the instinctual, the irrational in music and thus to the Dionysian 
spirit in art. He is the supreme example of the artist who, in the dialectic 
between emotional ‘primitivism’ and intellectual sophistication, never 
allowed the second ascendancy over the first. 


ZOLTAN KODALY 

Zoltan Kodaly (1882-1967) was intimately associated with Bartok in 
his researches into Hungarian folk-music and also absorbed its charac- 
teristic melodic, rhythmic and tonal inflexions into his own idiom; but he 
shows a greater dependence in his melodic invention on folk material 
and is, altogether, a composer of less marked individuality than his 
compatriot. Less intellectual than Bartók, Kodaly was a more direct and 
more sensuous musician whose early works show some allegiance to 
French impressionism. His strength lay in his lyricism, in a sustained, 
expressive line of melody; and it is, therefore, no accident that his best 
work is to be found in vocal music, particularly songs and choral 


300 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


compositions. In these he seems strongly influenced by Palestrina's 
ideal of transparent contrapuntal texture and clear part-writing. His 
outstanding choral work is the Psalmus Hungaricus for tenor solo, 
chorus and orchestra (1923), with a text based on Psalm 55, in which the 
sixteenth-century author identified his own sufferings and sorrows with 
those of King David and, indirectly, with those of the Hungarian people. 
The spirit of this archaic text is recaptured in lyrico-dramatic music of 
extraordinary power and vision, showing a tense melodic style, harmonic 
directness and great ingenuity in the choral writing. Rich in incidents 
and episodes, the work consists of a choral prelude and choral interludes 
which use identical pentatonic material modelled on the style of 
sixteenth-century Hungarian minstrel songs! and with the Hungarian 
fourth prominent in the cadences. 

These interludes constitute the refrain for the various solo passages 
employing different themes, so that the form represents a freely treated 
rondo on a large scale; the vocal parts are reinforced but never over- 
whelmed by the orchestral accompaniment. By contrast, the Te Deum 
(1936), written in commemoration of the 250th anniversary of the 
relief of Buda from the Turkish occupation, is conceived in a triumphant 
vein and with a judicious balance between contrapuntal and more homo- 
phonic choral writing. And, while in the Рзатиз Hungaricus the 
Hungarian element is conspicuous in the general melodic style, in the Te 
Deum it is largely confined to the subjects of the two fugues, and there 
are also references to plainsong melodies. 

With Háry János (1926) and Székelyfonó (The Székely Spinning 
Room) (1932) Kodály made important contributions to national Hun- 
garian opera. Both works employ native subjects, the first the legendary 
exploits of a kind of Hungarian Baron Munchausen and the second 
scenes from the life in a Transylvanian village. In contrast to Bartók, 
who adopted in his opera Duke Bluebeard's Castle the Wagnerian 
principle of musical continuity, Kodaly adheres to the older ‘number’ 
design, with solos, duets, and choral ensembles separated by spoken 
dialogue, as in the Singspiel of Háry János, or orchestral bridge passages 
linking the various numbers, as in The Spinning Room. The music of the 
latter consists mainly of a succession of Transylvanian folk-tunes whose 
texts are so arranged as to imply an action, though the whole character 
of the work is rather that of a scenic cantata. Both operas are true folk- 
operas of a distinctly popular nature and with a strong Hungarian flavour, 
but in Háry János Kodály mixes the native element with ingredients 
from other sources: in the Hungarian scenes, a song from the Bukovina, 


1 See article ‘Kodaly’ in Grove's Dictionary of Music and Musicians (Sth ed. London, 
1954), iv, p. 801. 


LEOS JANÁCEK 301 


verbunkos, and gypsy music; in the Austrian scenes, a Viennese carillon 
tune, an eighteenth-century minuet and Austrian marches. Kodály's 
orchestral music shows a highly developed sense of brilliant colouring, 
particularly in the concertante treatment of woodwind and the richly 
decorated style of the string writing. Characteristic examples of this 
are the Dances of Marosszék (1930) and Dances of Galánta (1933), the 
former based on Hungarian peasant tunes and the latter on verbunkos 
music. The Variations on a Hungarian Folk-song (1939) uses a penta- 
tonic tune, *The Peacock', which provides the theme for sixteen varia- 
tions and a finale that combines a wide range of contrasting expression 
with orchestral virtuosity. The Concerto for Orchestra (1939) shows a 
contest between the different instrumental groups that emulates the 
style of the eighteenth-century concerto grosso. 


LEOS ТАМАСЕК 


Turning to Czechoslovakia, the outstanding figure after the death of 
Smetana and Dvořák was Leoš Janáček (1854-1928) who cultivated a 
nationalism more intrinsic, more ‘intensive’ than that of his two 
predecessors. Like Bartók and Kodály in Hungary, he modelled 
his style on the characteristics of native (chiefly Moravian) folk material, 
and showed a natural gift for inventing melodies in genuine folk style. 
In his vocal compositions he assimilated with immense skill the verbal 
rhythms and tonal inflexions of his native tongue, imitating Mussorg- 
sky's similar musical treatment of Russian speech—as in this passage 
from Příhody Lišky Bystroušky (The experiences of the little vixen 
Sharp-Ears, generally known as The Cunning Little Vixen): 


Ex. 136 


Ze sta- ry Spa-Cek,  ne-sty da-ty zá - let - nik, 


ЕЕ 


pro - vá - dí  ne-prí-stojno-sü v ko - ги - né bu - ku. 


(They accused each other of horrible things, indecent and immoral—that the old 
starling was a shameless philanderer who did unseemly things in the crown of the 
beech) 


2r 


302 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Janáček was the first Czech composer to emancipate himself from the 
prevailing Western influences and he gradually arrived at an entirely 
personal and, in some aspects, highly original style that opened up new 
musical territory. The characteristic features of Janácek's language are 
an aphoristic melodic utterance and a kaleidoscopic change of short 
themes and motives; a continual variation of the melodic and rhythmic 
material; an elliptical harmony in which linking modulatory chords are 
abolished; a fluid tonality which is largely modal and avoids tonic- 
dominant relationships and the leading-note; and a rhythmic idiom in 
which the employment of the smallest metric units (2/8, 3/8) is very 
prominent. Janáček, who belonged to the generation of Mahler, 
Strauss, and Debussy, matured very slowly and did not reach the 
height of his creative career until he was in his sixties. The peak of his 
later achievement is to be found in his five operas, each of which 
treats an unusual subject and provides testimony to the composer's 
interest in psychological conflicts. 

The great success of his early opera Jenůfa (1903)! at the Prague 
production of 1916 appears to have released pent-up forces in Janáček 
and in the next twelve years he composed no less than five works for the 
stage. The first was Výlety Páné Brouckovy (The Excursions of Mr. 
Brouček) (1917), to a libretto by Frantisek Prochazka (after Svatopluk 
Cech), a good-humoured but sharply observed satire on the petit 
bourgeois mentality, with its narrow, smug, materialistic outlook, 
personified in the Prague landlord, Matéj Broucek. Apart from word- 
inspired motives in both parts, in the first (on the moon) it is the lyrical 
element that prevails, while the second part (in the fifteenth century) is 
characterized by a dramatic and ejaculatory style, and the thematic bur- 
den is carried by the orchestra, short and clear-cut phrases being repeated 
or varied on different degrees of the scale. Some phrases play the role 
of leitmotives and are transformed according to the external and psycho- 
logical changes of the action; but Janácek employs them in such a free 
manner that they stand simultaneously for different characters and 
different situations. Katya Kabanova (1921), to Janaéek’s own libretto, 
derived from A. N. Ostrovsky's play The Storm, is the first of his two 
operas with Russian subjects. The story illustrates the conflict between 
the rights and the freedom of the individual, and the rigour of tradition 
in old Russia as seen in the despotism of family life, the servile status of 
married women and the iron grip of moral and social customs, Katya 
representing the urge of the individual to escape from the fetters 
of antiquated modes of life. In spite of the title of the opera, the ideo- 
logical, if not actual, heroine is her mother-in-law Kabanicha, the 

1 See рр. 179-82. 


LEOS JANÁCEK 303 


embodiment of conservatism, and around her are grouped two contrast- 
ing pairs of lovers: Katya and Boris, Kudrjá$ and Varvara. The 
psychological differences between these pairs are clearly reflected in the 
musical characterization, the ill-starred lovers being depicted in broad 
flowing themes with, frequently, wide intervals in the vocal parts while the 
bright, happy pair are portrayed in short, dance-like figures mostly in 2/4 
and 3/8 metres of a distinctly Slavonic flavour. Janácek's mature genius 
is seen not only in his succinct delineation of these characters but in his 
rich invention and the absolute technical mastery of a work which, 
despite its predominantly objective treatment, soars at moments into 
tense dramatic life, as for instance in the last duet between Katya and 
Boris. The interplay of human drama with nature, which is a salient 
characteristic of this opera, affords the composer opportunities for 
atmospheric music, such as the magnificent storm of Act III or the 
wordless humming chorus of the last scene personifying the River Volga, 
which forms the background to the tragedy and is to be taken as the 
symbol of eternal unchangeable life. 

Nature in both its human and animal manifestations is also the 
theme of The Cunning Little Vixen (1923) in which Janácek's pantheism 
and his idea of the essential unity of creation finds full expression. 'My 
themes,’ he once wrote, ‘grow out of the earth, out of animals, out of 
people.’ The truth of this dictum is seen not only in the music which 
transmutes human-speech inflexions, the calls and cries of animals and 
nature sounds into imaginative utterances, but also in the action which 
unfolds at two levels—the actual and the symbolical. The opera repre- 
sents a kind of modern fairy-tale for grown-ups and some scenes have 
the naive air of a Christmas charade; but the Gamekeeper's apostrophe 
to Nature in the final scene is one of the finest lyrical passages in 
twentieth-century opera. Véc Makropulos (The Makropulos Affair) 
(1924), after Karel Capek's comedy, is a strange mixture of everyday 
reality with fantasy. The fantastic element is represented by Elena 
Makropulos who, thanks to an elixir given her by her father, an 
alchemist at the Prague Court of Rudolf II of Habsburg, survives for 
three hundred years into modern times when we meet her as the opera 
singer, Emilia Marty. The opera suffers from too involved a libretto, 
with a complicated legacy-lawsuit in the centre and much dialectics 
arising from the heroine's long past, which impair the conciseness of 
the dramatic action. The music, however, shows an advance on the 
previous stage-works in the greater suppleness and plasticity of the 
orchestral comment accompanying the declamatory phrases of the 
singers, and, significantly, the traditional operatic cantilena is all but 
completely absent. All this heralds the style of Janaéek’s last opera, 


304 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Z mrtvého domu (From the House of the Dead) (1928), his most 
powerful realistic drama. The subject, adapted by the composer from 
Dostoevsky's autobiographical novel, has no coherent story but sets 
out to show various scenes from life in a Siberian prison, in all its 
terror, brutality, and desperation relieved only by occasional touches 
of humour and comedy. Instead of a real plot we are given three 
great narratives told by prisoners about the crimes that brought them to 
Siberia. Grouped around these three tales are two episodes—the arrival 
(first act) and the release (third act) of a political prisoner, and an inter- 
lude which takes the form of a pantomime Don Juan, performed by some 
prisoners for the amusement of the rest. The cast is, with one exception, 
all male. Most characteristic of the music are stark, bare motives of a 
violently dissonant nature produced by clashes of minor and major 
seconds, parallel shifts of chords and long metrical units (6/4, 9/4, and 
4/2). The orchestral style is marked by sharply drawn instrumental lines 
and shows an unusual feature in that treble and bass instruments are 
set against each other at the extremes of their range, with no inter- 
mediate texture (anticipated in Тйе Makropulos Affair), which serves to 
intensify the starkly dramatic character of this opera. The title-page 
bears the inscription: “Їп every creature there is a spark of God'. The 
opera is a most eloquent summing-up of Janaéek’s deep-rooted belief 
in the essential goodness of man and an expression of his profound 
compassion for the sufferings of humanity. 

A word or two must be said about the state of incompletion in which 
Janáček left From the House of the Dead. He had worked directly from 
Dostoevsky's Russian text and had prepared from it a synopsis of 
the action on four loose pages from which he wrote the libretto straight 
into the score—some lines in Russian, others in Czech, and yet others 
in Lachian dialect. After his death this had to be unified in Czech and 
adapted to the vocal line. Moreover, Janáček did not live to complete 
all the orchestration, often writing only the top line and bass which he 
intended to fill in later. This task was performed by two Brno musicians, 
the composer, Osvald Chlubna, and the conductor, Břetislav Bakala, 
both intimately acquainted with Janácek's late orchestral style; the latter 
conducted the first performance of the opera at Brno on 12 April 1930. 

The majority of Janácek's choral music dates from his early, pre-war 
period, but he did not write his greatest work in that genre until 1926, 
two years before his death. This was the Glagolska mše (Glagolitic Mass) 
for soli, chorus, orchestra, and organ: that is, а Mass using, instead of the 
Latin text of the Ordinary, the equivalent text in Church Slavonic.! The 


1 In Croatia Glagolitic Masses have been written by such composers as Božidar Sirola 
(1889-1956), Albe Vidakovié (1914-1964), Krsto Odak (1888-1965). 


LEOS JANÁCEK 305 


liturgical use of Old Slavonic had been banned by the Roman Church in 
the eleventh century, although permitted again temporarily under 
Charles IV (1316—78). In 1920 an edict was issued allowing the Slavonic 
Mass to be celebrated in Bohemia on the Feasts of St. Cyril and St. 
Methodius and other patron saints, and it was soon after this that 
Janáček first contemplated a Glagolitic Mass. He took the text from the 
church-music periodical Cyril, xlvi (1920) which reproduced it in a 
linguistically inaccurate version. Moreover, he ignored the characteristic 
semi-vowels and nasals of Church Slavonic and set the words according 
to the accentuation of modern Czech. It consists of the traditional five 
choral sections (the equivalent of Kyrie, Gloria, Credo, Sanctus, and 
Agnus Dei) and of three instrumental numbers—an Introduction, an 
organ solo after the ‘Agneée Вой] (Agnus Dei), and a final Intrada. 
Both the Introduction and the Intrada have a fanfare character, and 
Janaéek visualized the performance of his Mass іп the open, with the 
congregation returning to the church in solemn procession to the music 
of the Intrada. The style of the work is boldly simple and at times 
picturesquely dramatic—in a word, entirely original. The vocal melodies 
frequently spring from the verbal rhythm and in the ejaculatory passages 
for soloists and chorus the tessitura for the solo soprano and solo tenor 
is often high, as for instance in the Kyrie: 


Ex. 137 


[yn e be | 
Sop. Solo К ісіме.) 


Go- spo -di po - mi - luj! 


Go - spo-di po-mi - №)! 


MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


306 


Go - spo - di 


[ + 


Go- spo-di 


Xu 
—ч 
1 
..“ 
Е 
1 
о 
OR 
44 
m 
o 
A 
Pus 
ON 
o 


(Lord have mercy!) 


The essential themes are entrusted to the orchestra on which the vocal 
superstructure rests, sometimes based on the instrumental material, 


at 


other times set against it chordally. As in Stravinsky's Les Noces, a few 
tiny motives pervade the Mass in various metamorphoses. As Hans 
Hollander has shown,! they are marked by some of the characteristics 


1 Leos Janácek (London, 1963), pp. 99-100. 


LEOS JANÁCEK 307 


of Moravian folk-song. Moreover, Slavonic dance rhythms are as 
peculiar to the Credo and the Intrada as are incisive aphoristic themes 
to the remaining movements. The organ solo is an agitated piece in 
ostinato technique symbolizing the Last Judgement, while the ensuing 
Intrada is intended to conjure up the vision of the assembly of the 
Czech people before the throne of God. There is nothing comparable 
to the Glagolitic Mass in the whole of Western religious music and, 
though it must be accounted a Christian work, in its expression of 
primitive elemental joy and jubilation it sounds an unmistakably 
pantheistic note. Of Janácek's orchestral compositions written during 
his late period, the Rhapsody Taras Виа (1918), inspired by Gogol's 
novel, and the ballad B/anik (1920) bear witness to his ardent Slav 
patriotism. More intrinsically characteristic of his late-style, however, 
is the Sinfonietta (1926) which grew out of a fanfare for thirteen 
brass instruments intended for the Prague congress of the Sokols. 
Despite its title, the work is hardly symphonic in the established 
sense, themes being rarely developed but mostly repeated with slight 
variations, and no sooner is an idea introduced than it is discarded for 
another. The texture is mosaic-like and much use is made of pedals and 
ostinatos, as for instance in the fourth movement which is almost 
exclusively based on the incessant reiteration of a theme of folk-dance 
character. The progress of the five movements shows a suite-like order 
of sequence, and the form is rhapsodically loose, often creating the 
impression of an improvisation. There are no thematic connexions 
between the movements, but in order to provide an effective rounding- 
off Janácek repeats the opening fanfare in the finale. The orchestration 
is marked by novel and, in some instances, extraordinary sound-effects. 
The first movement, for example, is scored for eleven trumpets, two 
bass tubas, and drums, while elsewhere the composer aims at the trans- 
parent style of chamber music, with much concertante writing for wood- 
wind and brass. The Sinfonietta is a typical illustration of Janácek's 
laconic and deliberately ‘primitive’ manner and is опе of his few works 
which have no explicit or implicit programme. 

Among Janácek's chamber works for vocal ensemble the Zápisník 
zmizelého (Diary of One Who Vanished) (1918) occupies a unique place. 
The anonymous verses, which were published in the Brno newspaper 
Lidové noviny in 1916, are in the vein of genuine folk-poetry telling of a 
young peasant, Jan, who yields to his love for the gypsy girl Zefka with 
whom he deserts his native village. The subject fascinated Janáček, 
partly because of its social implication—the lifting of the barriers that 
separate an inferior race (gypsies) from the rest of the community— 
and partly because its ardent passion and sensuality strongly echoed 


308 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


a relationship in his own private life. The verses, which are set for 
tenor, contralto, three women's voices, and piano, present sharp- 
edged dramatic vignettes in which the entire gamut of a young man's 
erotic sentiments are blended with nature moods and descriptive 
passages, for all of which Janácek found an immense variety of musical 
expressions. Yet the work, not unlike the Glagolitic Mass, is virtually 
monothematic, being based on a single brief idea, consisting of a 
rising or falling fourth and a major second, which undergoes permuta- 
tions according to the dramatic and emotional progress of the text: 


Ex. 138 
(i) Original motive 


Ne - Се-Ка) 


ne-vyj-du 


(v) no. 17 1 3 


Со ko-mu sti-ze-no__ to-mu ne-u - te-Ce. 
(vi) no. 21 


де в Ж о ~ Ze-nim, — ke-rá mi zvo - lí - te. 

( (ii) she carried herself like a hind; (iii) Don't wait, I’m not going out; (iv) Now 
already dawn appears in the sky; (v) If something is one's fate, there's no escape; 
(vi) that I marry her whom you chose for me.) 

There are twenty-two songs,! the majority of which are for the tenor, 
mostly short in length and oscillating between a declamatory style and a 
more expressive arioso. Small metric units (2/8, 3/8, 4/16, 6/16) combine 
with short incisive figures to lend the vocal phrase a high degree of 
plasticity while the piano points and comments on the moods and 
incidents of the verses. Of Janácek's purely instrumental chamber music 
the First String Quartet (1923) (which owes its inspiration to Tolstoy's 


1 Nos. 15, 16, 17, and 18 are recorded in The History of Music in Sound, x. 


OTHER CZECH COMPOSERS 309 


The Kreutzer Sonata) is rhapsodic in form but creates a feeling of organic 
coherence largely due to the derivation of its main material from the 
opening dance theme and the use of a motto that itself presents a fore- 
shortening of the initial dance theme. The Second String Quartet (1928), 
written six months before his death, is, like Smetana's From My Life 
Quartet, of an autobiographical nature, its subtitle, Intimate Letters, 
referring to the composer's intimate friendship with Kamila Stósslová. 
Characteristic of the work are the abrupt changes of mood, from 
a tranquil and inward lyrical expression to dramatic outburst and im- 
passioned ecstasies. The writing, though not devoid of contrapuntal 
passages, tends towards a more homophonic treatment. 


OTHER CZECH COMPOSERS 


If Janácek's prime significance lay in a conscious and deliberate 
nationalism, the next generation of Czech composers stood in a less 
pronounced, looser relationship to native musical lore and took their 
chief bearings from movements in contemporary Western music. The 
most important figures here are Bohuslav Martinü (1890-1959) and 
Alois Hába (1893-1972). Martinü was a pupil of Roussel in Paris 
at a time (1923) when Milhaud, Honegger, and Poulenc were emerging 
as individual personalities. He soon came under their influence, notably 
Honegger's. Stravinsky's neo-classicism also exercised a considerable 
attraction on him. He was essentially a musical ‘natural’, writing with 
immense facility and great technical skill, but he did not avoid the com- 
monplace and the trivial, with the result that his output shows great 
unevenness. Martinü was a craftsman who did not aim at a settled style 
but was primarily concerned with the potentialities inherent in his 
musical material, notably structure, formal design and rhythmic move- 
ment. The majority of his Parisian works (before 1940, when he emi- 
grated to America) reveal an unemotional, objective attitude of mind 
from which vitality and a certain degree of originality are not absent. A 
composer of great versatility, Martinů devoted himself to the cultivation 
of all types of composition—opera, ballet, choral and orchestral works, 
chamber music, and music for violin and piano. He first made his name 
with an orchestral work, Half-Time (1925), an impression of a soccer 
game with which he anticipated Honegger's Rugby (1928); while noisy 
and even brutal in sound, it has an invigorating, sinewy rhythm. This 
was followed by La Bagarre (1927), commemorating the occasion of the 
landing of the American aviator Lindbergh at Le Bourget and is charged 
with the atmosphere of the movement of big crowds. In both these 
works the mere physical excitement created by a shifting mass of people 
was the primary incentive for the music. The Partita (1931) and the more 


310 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


ambitious Sinfonia Concertante (1932) show Martinů pursuing a neo- 
classical style into which he assimilated the melodic characteristics of 
Czech folk-song, which lend his works freshness and earthy vigour. 
Among his most successful compositions in the neo-classical vein are 
his Concerto Grosso (1938), a virile and dynamic piece in which the 
*concertino' is the piano, and the Double Concerto for two string 
orchestras, piano, and timpani (1938) where the manipulation of the 
two groups of strings—now in opposition, now in concert and now 
alternating with each other—shows remarkable resourcefulness. 

For Martinü opera does not present philosophical or psychological 
problems or a fragment of real life, but is a theatrical spectacle trans- 
formed by music into something sui generis, in all of which one strongly 
senses the potent influence of Stravinsky's operatic aesthetics. His most 
ambitious opera, illustrating Martinü's attitude in a startling manner, is 
Juliette (1937), to a libretto adapted from a play of the same name by 
Georges Neveux, the theme of which is the conflict between reality and 
the illusion of dreams. The work is in the nature of a lyrical poem that 
avoids all dramatic effect. The one-act radio opera, Comedy on the 
Bridge (1937), is written in an attractive light-hearted vein which com- 
bines a pastiche of eighteenth-century comic opera, notably Mozart's, 
with melodic-rhythmic elements from Czech folk-music. The trans- 
parent scoring guarantees the clear audibility of the singers’ words. 

Alois Hába, who was a pupil of Franz Schreker in Vienna and, later 
Berlin, was not a Czech nationalist in the proper sense, but belonged 
rather to the wider orbit formed by Central European music, and was in 
particular associated with an athematic style of writing and with micro- 
tonal experiments. Convinced that Western music must be liberated 
from the principle of thematic development, he developed a new manner 
in which he gradually rejected all traditional means of formal and melo- 
dic design and harmonic structure, such as sequence, repetition, recapi- 
tulation, and the polarity of harmonic zones. In his athematic works 
(string quartets and piano music) Hába aimed at a vigorous and con- 
tinuous flow of free invention, though frequently at the expense of 
musical logic and structural coherence. As for his microtonality, his first 
ideas were suggested by his observations as a violinist, when he noticed 
the slight variations in pitch characteristic of the solo playing on stringed 
instruments. Moreover, his studies of Czech folk-song, notably of his 
native Moravia, made him aware of microtonal deviations in the 
singing of this music. These facts, combined with examples of 
oriental music and with the influence of certain of Busoni's aesthetic 
theories (which advocated the division of the semitone into third- 
and sixth-tones) induced Hába to experiment with the possibilities of 


KAROL SZYMANOWSKI 311 


microtonality. As a result, some of his works are based on quarter- 
tones, such as the opera Matka (The Mother) (1929), the String 
Quartets, opp. 7, 12 (Ex. 139), and 14, and the various Fantasies for 
piano, ор. 25 to op. 31, for which the Czech piano firm of Förster con- 
structed special instruments; other compositions use the sixth-tone, such 


| 


Hog; RF Fa 


TE perp раа | 
раста EET п и ar E — а г В 
1 EMSS Баш EE | ЕЕ ЕЕ ЕАН ШП E нен pa [mme co ES} 
Ше сее В се t я exem аса ста сіре шта аа ғы 


L = quarter tone up, В = three-quarter tone up,d = quarter tone down 


as the opera Přijď království Туё (Thy kingdom come) (1942) and the 
String Quartet, op. 15. Yet, interesting and ingenious as are Hába's 
attempts to introduce novel notions of composition, neither his 
athematic style nor his microtonal experimentation has proved a signifi- 
cant pathway to the future. 


KAROL SZYMANOWSKI 

Unlike Hungary and Czechoslovakia, Poland did not develop an 
advanced national movement until after the First War. Its acknow- 
ledged leader was Karol Szymanowski (1883-1937), the most eminent 
composer that Poland had produced since Chopin. Beginning his career 


312 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


under the influence of Skryabin, he subsequently modelled his style on 
Reger and Richard Strauss, whose thematic complexity he took over 
until acquaintance with the music of Debussy and Ravel led to the 
emergence of a personal manner. A native Polish element is already 
manifest in the nostalgia and brooding melancholy of the works written 
before and during the war,! but it was not until the early 1920s that 
Szymanowski began to take a keen interest in authentic Polish folk- 
music, which was of decisive consequence for the further development 
of his style. Yet, notwithstanding this acquisition of national traits, 
Szymanowski preserved a wide European outlook and availed himself 
of features of contemporary Western music, skilfully absorbing them 
into his own language. А stay at Zakopane in the Tatra Mountains in 
1921 brought him into first direct contact with native folk-music, the 
songs and dances of the highlanders, which reflected their wild un- 
inhibited temperament and the rugged beauty of their mountains. This 
inspired him to write the ballet Harnasie (1926) whose title refers to the 
legendary robbers of the Tatra Mountains. Into this work he introduced 
a number of highland dance and song tunes and, more importantly, 
allowed his personal idiom to be coloured by a harsh barbaric element 
and by the remarkable irregularities of rhythm and accentuation 
characteristic of Tatra folk-music (Ex. 140). 

In the same year (1926) Szymanowski wrote a Stabat Mater for 
soloists, chorus, and orchestra whose masterly polyphonic writing owes 
something to his study of Palestrina; it is significant that its idiom 
appears more simplified in comparison with that of the works of his 
previous period. Archaic harmonic progressions (parallel motion of 
triads) alternate with genuine three- and four-part writing. There is 


Ex. 140 


1 See Chapter I, p. 36. 


KAROL SZYMANOWSKI 313 


() 
ILLO есеге ж шыл НЕ шыт. 2 аре ШУ. 
Те еее ы а із 
eS a * 


SS eee re 
Шы іс G 
ее р 
а ди 


ee ай ee лаг тюше РЕ] 
т” —ы ГА 
Т =ч ше. И ТИ а Res eee 
РН re F Н® Ца 40705" 2 6 ЕЕ ЕЕ LER 


b 
f) 
——— ЕЕЕ ER EAM 


ee р — s —7-w |]. —4—] 
LÁ ERAI (ВО р 
| Бати” ЗЕ парене. 2] с авайс а / ору Ic stas 


314 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


perhaps an excess of pure lyricism in the Stabat Mater but, with this 
qualification, the work is an impressive example of the individual 
projection of a profound religious emotion. 

In King Roger (1924), a three-act opera to a libretto by Jaroslaw 
Iwaszkiewicz, Szymanowski produced one of his most distinguished 
works. The subject is the conflict between Christianity and Hellenic 
paganism and the action takes place in twelfth-century Sicily centring 
on King Roger (the historical Roger II lived in the eleventh century) and 
a shepherd who casts a spell on Queen Roxane and in the last act reveals 
himself as the god Dionysus, to whom Roger, in a symbolic act, 
sacrifices his own soul. The opera contains overtones of Eastern 
mysticism and philosophy which, combined with an erotic sensualism, 
inspired the composer to some of his most characteristic and imaginative 
music. Тһе sensual element is associated with the shepherd and Roxane 
and reaches its climax in the second act, in Roxane's melismatic song: 


Ex. 141 
Andante 


and the orgiastic dance in 7/8 of the shepherd and his musicians. This 
is contrasted with the hieratic Byzantine grandeur of other portions, 
especially the music characterizing Roger and the court atmosphere. 
The style of the work is intensely dramatic, often taking on a starkly 
barbaric expression in which the role of the chorus is most important, 
but there are other episodes in which an ecstatic lyricism prevails. 
The organum-like progressions in much of the choral writing, thematic 
complexity, bitonality, and a tense involuted chromaticism are as 
typical of the music as are the chord-derived melodies of an impres- 
sionistic character. 

The Symphony concertante for piano and orchestra (1936), frequently 
referred to as the Fourth Symphony, shows a change of manner. Instead 
of the romantic lyricism that informs Szymanowski’s general style, most 


MANUEL DE FALLA 315 


of the music is brittle, dry, and brilliant, with the piano treated in a 
percussive manner which owes something to Stravinsky. But the second 
of the three movements, with its dreamy arabesque for the solo flute 
and an ornate, slightly overladen piano part, recalls the composer's 
earlier style. Of the two string quartets, the First Quartet (1917) is 
musically perhaps the more substantial, showing themes of great 
plasticity and a varied contrapuntal texture in the first two movements. 
The mood of the opening movement, which begins with an expressive 
Lento introduction, fluctuates between the exhilaration of the dance- 
like first subject and the melancholy lyricism of the second while the 
working-out section turns to a vein of grotesque mockery. The finale, a 
*Scherzando alla burlesca’, is an example of polytonal writing before 
Milhaud exploited this technique, the four parts being written in the 
respective keys of C, E flat, F sharp, and A major. The Second Quartet 
(1927) is of a more complex texture than the first, containing much linear 
counterpoint, with two fugato passages in the finale, but there are also 
passages of an impressionistic (chordal) character, and bitonality and 
polytonality play a conspicuous part. 


MANUEL DE FALLA 


Nationalism in Spain developed in a continuous line from Felipe 
Pedrell and Isaac Albéniz in the second half of the nineteenth century to 
Enrique Granados and Manuel de Falla (1876-1946) in the first half of 
the present century. It was Falla, the outstanding personality among 
those composers, who put Spain once again on the musical map of 
Europe, with works which are intrinsically as well as extrinsically 
Spanish and yet transcend the purely national element, by their 
combination of imaginative power and a clear-sighted, penetrating 
technical acumen. The basic feature of Falla's style is the brilliant 
application of an essentially impressionistic technique, which he learned 
before the war from Debussy, to the treatment of strongly nationalistic 
material, notably of his native Andalusia. Yet, like Janácek and the 
later Bartók, he scarcely ever introduced authentic folk themes into his 
music; it is rather that his essential thought 15 moulded by the spirit of 
Spanish folk-music and by some of its technical features. Apart from a 
handful of folk-song arrangements, Siete canciones populares españolas 
(1922), it was only on very rare occasions that Falla employed actual 
folk tunes: a few phrases associated with the Miller and his Wife in 
the ballet E/ Sombrero de tres picos (The Three-Cornered Hat) (1919), 
and two melodies in the puppet opera E/ Retablo de Maese Pedro 
(Master Peter's Puppet Show) (1923), based on an incident in Cervantes’ 
Don Quixote—the Catalan dance when Don Quixote smashes the 


316 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


puppets to pieces, and the song of Melisendra in the tower, whose 
tune derives from an old Spanish ballad sung at the time of Cervantes. 
In this last work and in the Harpsichord Concerto (1926) Falla shed 
virtually all superficial nationalist traits, but conveyed its essence in 
music that bears his own entirely individual stamp. His chief means of 
expression are not so much harmony as melody and rhythm; short 
incisive phrases are presented in contrasting and unblended instru- 
mental colours while the juxtaposition of accent and cross-accent 
results in conflicting rhythms of a fascinating effect (Ex. 142). 

The guitar conditioned toa large extent Falla's harmonic idiom and his 
piano writing, as in the Fantasía bética (1919), and it is responsible for 
the internal pedals and the clear, percussive style of the Harpsichord 
Concerto. There is in this latter work something of Domenico 
Scarlatti's Italian clarity and coruscating brilliance, and it is significant 


Ex. 142 
^ 


Ob. 


Clar. 
inA 


Harpsichord 


ESPLÀ AND TURINA 317 


that Falla made prolonged studies of Scarlatti's music before he com- 
posed the Concerto. During the last twenty years of his life Falla was 
occupied with a vast choral work, At/ántida, whose text deals with the 
story of the lost continent, Atlantis, sunk in the sea beyond the Western 
coast of Spain. This was performed at Milan in 1962 (completed by 
Ernesto Halffter) and proved to contain music of a high order of inven- 
tion, though nothing that added to Falla's reputation. 


ESPLÀ AND TURINA 


Falla’s relationship to Andalusian folk-music was paralleled by 
Oscar Espla’s (b. 1886) to the music of his native Alicante. Like Falla, 
Espla made only rare use of authentic folk tunes—to the exceptions 
belongs his cantata, La noche buena del diablo (The Devil’s Christmas 
Eve) (1931), based on traditional Spanish children’s songs. A composer 
of a pronounced intellectual cast of mind—Espla was originally an 
engineer who later turned to music, musicology and philosophy—he 
cultivated the symphonic form in the tone-poems Don Quijote velando 
lus armas (Quixote keeping vigil by his arms) (1925) and Fiesta (1931). 
After his studies in Germany he adopted a scale of his own invention 
(C-D flat-E flat-E-F-G flat-A flat-B flat) on which he built a har- 
monic system that lent his music a regional, Spanish-Mediterranean 
flavour, without incorporating any genuine folk-song material, as for 
instance іп the symphonic poem, Ciclopes de Ifach (1937), one of 
Espla’s most important works. Joaquin Turina (1882-1949) was 
largely inspired by Andalusian scenes (Danzas fantásticas for orchestra, 
1920, and Sinfonia sevillana, 1921). A musician of limited range, 
endowed with a facile and rather superficial pen, Turina is best known 
outside his country for his many songs and piano pieces, notably the 


22 


318 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Cuentos de Espafía (Spanish Tales) (Set I, 1918; Set II, 1928) in which 
picturesqueness and charm are evenly balanced. With the exception of 
Esplà and Gerhard, the composers of Falla's generation, including Falla 
himself, considered Paris the Mecca of Music to which they made their 
pilgrimage after completing their studies in their native land. 


LATER SPANISH COMPOSERS 


Among the musicians of the next generation, Federico Mompou (b. 
1893) and Roberto Gerhard (1896-1970) have attracted most attention. 
Mompou, born in Barcelona, is an exponent of Catalan nationalism, 
with an output largely confined to the miniature form of piano pieces 
and songs, the main characteristics of which are intimacy of feeling and 
refined, subtle workmanship (6 Impresiones intimas, 1914; Cants magics, 
1919; Quatre Mélodies, 1926). Mompou has been called a primitivista 
on account of the marked simplicity of his technical style. Thematic 
development, foreign anyway to a miniaturist, and harmonic complexity 
he avoids in favour of a static lyricism of an evocative impressionist 
nature. Gerhard, who was of Swiss origin and after the end of the 
Spanish Civil War settled in England in 1939, was first а pupil of 
Pedrell, the founder of documentary Spanish nationalism, and then 
studied with Schoenberg in Berlin ftom 1923 to 1928, a fact that would 
partly account for Gerhard's remarkable technical savoir faire. Despite 
his adoption of the 12-note method his ties with his native Catalonia 
remained at first very close, as in the cantata, L'alta naixença del Rei 
En Jaume (The high birth of King James) (1931) and Аааа, interludi y 
аапса for orchestra (1936). In contrast to the predominantly contra- 
puntal character of dodecaphonic music, Gerhard initially stressed the 
vertical (harmonic) element, as in his Wind Quintet (1928) which adopts 
the same combination as Schoenberg's Op. 26, but with an instrumental 
timbre tinged with a Spanish (Catalan) flavour. Also in the music 
written after taking up residence in Britain, Gerhard continued to treat 
serialism in such a way as to allow a pronounced Spanish note to make 
itself felt as, for instance, in his ballet, A4/egr'as (Festivals) (1942) and the 
opera, The Duenna, to a libretto after Sheridan (1948; rev. 1950). 
Gerhard was the first Spanish composer to harness thinking in 12-note 
terms to genuine Spanish feeling. Joaquin Rodrigo (b. 1902), who was 
blind from the age of three, is a prolific composer of minor stature who 
follows Falla yet cultivates a more advanced harmonic idiom. Outside 
Spain Rodrigo is perhaps best known by his Concierto de Aranjuez for 
guitar (1939). There remains a group of Spanish composers who, like 
Les Six in Paris, banded themselves together and under the intellectual 
leadership of the influential critic and writer, Adolfo Salazar (1890- 


SWITZERLAND 319 


1958), formed what became known as e/ grupo de los ocho or el grupo de 
Madrid where this association was founded in the spring of 1930. They 
were: Ernesto Halffter (b. 1905), its most eminent member; his older 
brother, Rodolfo Halffter (b. 1900); Salvador Bacarisse (1898-1963); 
Juan José Mantecón (b. 1896); Fernando Remacha (b. 1898); Gustavo 
Pittaluga (b. 1906); and the woman composer, Rosa Maria Ascot (b. 
1906). (Mantecón, the oldest member, was like Salazar in the first place 
a writer.) What united this group of eight post-Falla musicians was a 
bond of far stronger fibre than that which had originally brought the 
Parisian six together. This was the total rejection of all nationalism and 
impressionism. In a lecture given in Madrid on the aims of e/ grupo de 
los ocho, Pittaluga, the youngest of this group, took a completely nega- 
tive view of Spanish musical folklore and stressed the necessity of 
composing ‘authentic’ music, i.e. music free from all ethnic roots and all 
association with literary, philosophical and metaphysical ideas. They set 
out to write works the sole criterion of which was their intrinsic musical 
and technical qualities. For the rest fno romanticism, no chromaticism, 
no divagations—and no chord of the diminished seventh! 

This advocacy of a kind of New Objectivity in Spanish music demon- 
strated the distance these younger composers had travelled from their 
nationalist predecessors (Pedrell, Falla, Esplà, and Turina), and had 
become receptive to the main trends in European music, notably to the 
neo-classicism of Stravinsky and Hindemith. 


SWITZERLAND 


Owing to its peculiar geographical position, Switzerland has for many 
centuries been subject to the cultural influence from its great neighbours 
in the North-East, South, and West and it is therefore not surprising 
that its musical history should be closely bound up with that of Ger- 
many, France, and, to a lesser degree, Italy. 

Of the German-Swiss composers born between 1880 and 1900, 
Othmar Schoeck (1886-1957) was the outstanding artistic personality 
and occupies an intermediate position between the late romanticism of 
Wolf and Reger (his master) and the modern world. It is as a song 
writer and musical dramatist that Schoeck made his most important 
contribution to Swiss music. His strong lyrical vein manifests itself in a 
prodigious number of songs in which the choice of poets, such as Goethe, 
Eichendorff, Uhland, Hebbel, Keller, and Lenau, proclaims his affinity 
with the great German song composers, notably Wolf. In his various 
operas (Venus, 1920; Penthesilea, 1925; Vom Fischer un syner Fru (The 

1 Quoted in Gilbert Chase, The Music of Spain, (New York, 1959). p. 203. 


320 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Fisherman and his wife), 1930; Massimila Doni, 1935, and Das Schloss 
Dürande, 1939), he achieved, through the exploitation of the con- 
siderable emotional tension inherent in his subjects, a rare combination 
of dramatic power with terse lyrical expression. If there is a common 
theme running through his operas it is that of man's better self in conflict 
with his evil, demoniac powers, and, owing to the particular psychologi- 
cal problems posed in his libretti, Schoeck's musical language in his 
operas appears to be more austere and more uncompromising than in 
his songs. 

Schoeck was the acknowledged leader of the younger generation of 
German-Swiss composers which includes Albert Moeschinger (b. 1897), 
Conrad Beck (b. 1901) and Willy Burkhard (1900—55) whose oratorio, 
Das Gesicht Jesajas (The Vision of Isaiah) (1935) impresses by its direct 
forcefulness and the woodcut simplicity of its utterance. With these and 
other composers the influence of Busoni, who lived in Zürich from 1915 
to 1920, as well as of Hindemith is noticeable, but their common 
denominator is their typically Swiss quality of ruggedness, severity, and 
economy of expression. 

Of the musicians of French Switzerland, Frank Martin (b. 1890) pre- 
sents the sharpest and most individual profile. Martin's early works 
show the strong influence of Franck and Fauré, and some leaning 
towards Ravel. During the inter-war years, however, he gradually 
developed an individual style and grappled with the technical problems 
raised by the New Music. А teacher for many years at the Jaques- 
Dalcroze Institute in Geneva, Martin was in close contact with eurhyth- 
mics which accounts for the fact that his chief interest was at first 
concentrated on rhythmic experiments, of which the Trio sur des chants 
populaires irlandais for violin, cello, and piano (1925) and the orchestral 
Rhythmes (1926) are notable examples. In 1930 Martin began to study 
Schoenberg's twelve-note technique, freely using this method with tonal 
implications while rejecting the aesthetic premises of the dodecaphonic 
school. In an article! he defined his attitude to Schoenberg's aesthetics 
by saying that ‘every rule has for its sole aim the enrichment of style... 
the observation of rules is nothing but an elegance, a pleasure of the 
mind which is independent of any (aesthetic) value'. Characteristic of 
this stage in Martin's development are the Piano Concerto (1934), the 
Symphony (1937), and the first four Ballades for various solo instru- 
ments and orchestra (1938-40). The outstanding work of this period is 
the dramatic oratorio, Le Vin herbé (The magic potion) (1941), based on 
the Roman de Tristan et Yseult by Joseph Bédier, which made Martin's 
name known beyond the confines of Switzerland and in which he applied 

1 Schweizerische Musikzeitung, lxxxii (1942), no. 3. 


SCANDINAVIA AND HOLLAND 321 


the serial technique in conjunction with tonal and homophonic effects 
of a novel kind. The legendary atmosphere is recaptured by means of a 
chorus of twelve reciting voices, from which the two protagonists detach 
themselves as singing characters. The accompaniment consisting of a 
small orchestra of seven strings and piano shows a great variety in the 
thematic and rhythmic treatment. In two subsequent works— Der Cornet 
(1943), after Rilke, and the Six Monologues (1943) from Hofmannsthal's 
Jedermann—Martin achieves a still closer synthesis of serial technique 
with tonal writing in the service of an intense lyrical expression. 


SCANDINAVIA AND HOLLAND 


Music in Scandinavia and Holland during the inter-war years showed 
the same dichotomy as in other countries between conservative ten- 
dencies (which meant adherence, chiefly, to German romanticism) and a 
strong movement towards the assimilation of stylistic and technical 
features of contemporary music in Germany (Hindemith, Schoenberg) 
and France (Debussy, Ravel, and Stravinsky). Of the composers, who 
adopted technical devices and, to some extent, also the aesthetic creeds 
of the New Music, the Norwegian Fartein Valen (1887-1952) is note- 
worthy for his atonal style which he began to evolve in 1924 indepen- 
dently of Schoenberg's tone-rows. While occasionally employing a strict 
serial technique as, for instance, in the Piano Variations Op. 23 (1936), 
Valen most cultivated a texture characterized by combination of various 
independent lines in which the same note series occurs constantly; in 
some works an identical ‘theme’ pervades each of the parts. 

The Dutchman Willem Pijper (1894—1947) was first influenced by 
Debussy but in his subsequent works he achieved a measure of indivi- 
duality, making use of polytonality and polyrhythms and evolving a 
complex, elaborate style of writing which owes something to Schoen- 
berg. In his later period, however, Pijper aimed at simplification and 
economy of means as in the Trio for flute, clarinet and bassoon (1927), 
the String Quartet no. 4 (1928) and the Six Symphonic Epigrams (1928). 
He also devoted himself to the study of Dutch folk-song, the chief 
creative fruits of which are two unaccompanied eight-part choruses, 
Heer Halewijn (1920) and Heer Danielken (1925). Henk Badings (b. 
1907), a pupil of Pijper, is a prolific composer who has written orchestral 
works, concertos, and chamber music in which a neo-classicism, largely 
based on Bach's style, prevails and which combine experiments in form 
and tonality. 


322 MUSIC IN THE MAINLAND ОЕ EUROPE: 1918-1939 


RICHARD STRAUSS'S LAST YEARS 

The musical scene in Central Europe between the two wars presented a 
picture more complex and varied in its aesthetic and stylistic aspects 
than that of post-war France and Italy. Three main currents may be 
distinguished. There was the older generation, led by Richard Strauss 
and Hans Pfitzner, which in varying degrees continued the romanticism 
of the pre-war period. There was the avant-garde under the leadership of 
Hindemith, violently anti-romantic in its attitude, avoiding all sub- 
jectivity and cultivating *New Objectivity' with a concentration on the 
technical and structural potentialities of the musical material per se and, 
so far as opera was concerned, seeking a close link with the political and 
social climate of its time. And there was, lastly, the Second Viennese 
School, essentially romantic in its aesthetic outlook and aims but work- 
ing with Schoenberg's twelve-note system, a revolutionary method 
of composition that radically altered the face of music. 

Richard Strauss (1864-1949)! had written his most significant 
works before the First World War. With the audacities of Salome 
and Elektra he had proved himself the most advanced German com- 
poser of his time, a position from which, partly under the influence 
of his librettist, Hugo von Hofmannsthal, he began to retreat in 
Der Rosenkavalier and Ariadne auf Naxos. In his subsequent operas 
he completely shut himself off from all contact with the modernism 
of the post-war period. At the same time an increasing falling-off 
became noticeable in his earlier vitality and invention, though his 
immense fertility and sovereign technical command continued un- 
diminished to the end of his life. What also remained intact was Strauss's 
innate and restless intellectual curiosity and his ready response to the 
challenge of problems of style, form, and dramatic treatment. This 
explains the curious zigzag course of both his pre-war and post-war 
operas. Die Frau ohne Schatten (Тһе Woman without a Shadow) (1919) 
is an oriental fairy-tale not wholly unlike Mozart's Die Zauberflóte, but 
Hofmannsthal's libretto is packed with a number of obscure symbolisms 
and allegories—the ‘Shadow’ of the title stands for female fertility 
—and the language, for all its poetic imagination, is involuted and full 
of esoteric semantic subtleties. Strauss treated this subject in the style 
of spectacular grand opera, with big ensembles and finales and with 
broad symphonic interludes, and the best part of the music lies in the 
refined, technically resourceful evocation of the magic element. Ever 
since Der Rosenkavalier Strauss had been exercised by the problem of 
word-setting in opera, and in Intermezzo, ‘A Bourgeois Comedy with 


! For Strauss's early and middle years see chapters I and III. 


RICHARD STRAUSS'S LAST YEARS 323 


Symphonic Interludes' (1924), he found for it an original solution. 
The singing parts are largely treated in a light conversational manner 
imitating everyday speech and are accompanied by a remarkably 
transparent orchestra. It is only in the finales of the two acts that 
the vocal writing turns to sustained cantilena, with the lyrical ele- 
ment and also symphonic elaboration being mainly confined to the 
orchestral interludes. Despite the utterly banal nature of the libretto, 
which was Strauss's own, the work constitutes a brilliant experiment in 
the creation of a new type of German comic opera. In Die aegyptische 
Helena (1928) he returned to the mythological Baroque opera such as 
he had essayed in Ariadne auf Naxos, but with this difference that now 
both the protagonists are human characters. Hofmannsthal places 
in the centre of the action the matrimonial drama of Menelaus and 
Helen seen in the light of modern psychology; but he surrounds it with 
the trappings of the ancient myth, magic potions, supernatural figures, 
and the splendour and opulence of an imaginary Orient. Strauss re- 
sponded to this strange concoction with music of the most luxuriant and 
richly proliferating character, in which a heroic hymn-like element pre- 
dominates; but the score lacks truly significant utterances. In Arabella 
(1933), the libretto of which was the last Hofmannsthal wrote before his 
death in 1929, poet and composer endeavoured to repeat the style of 
Der Rosenkavalier. 'The story has much of the period charm and the 
poetic sentiment of the earlier work, Hofmannsthal returning in it to his 
favourite theme of youth's awakening to full awareness of itself through 
the experience of true love. Yet the psychological richness of Der 
Rosenkavalier is absent, and the work contains no dramatic character 
comparable to that of the Feldmarschallin. Arabella's suitor, a Slovene 
aristocrat, who is conceived as a kind of inverted Ochs von Lerchenau— 
young, noble, and generous—fails to inject into the plot the irresistible 
comic vein of Hofmannsthal’s earlier character. The music is instinct 
with a rich lyricism, with a sensuous vocal melody of a melting, at 
times cloying, sweetness; and, since the action takes place in the 
Vienna of the 1860s, Strauss’s waltzes are not felt to be incongruous 
with the period as they are with Maria Theresa’s Vienna in Der Rosen- 
kavalier. Yet, strangely enough, the composer here favoured the rustic, 
coarse-grained Lündler of Bavarian origin rather than his namesake 
Johann's elegant, town-bred waltzes. On the other hand, the orchestral 
texture of Arabella, if compared with that of Der Rosenkavalier, is more 
refined and more supple, frequently achieving the effect of chamber 
music, and the use of leitmotives is more subtle in the psychological 
delineation of the characters. The composer shows his age in the 
paucity of memorable melodic ideas, but the work has proved the most 


324 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


successful of his later operas. Die schweigsame Frau (The Silent Woman) 
(1935), to a libretto by Stefan Zweig who adapted it from Ben Jonson's 
Epicoene, represents Strauss's excursion into pure opera buffa, with 
vivacious ensembles and an immensely fluent parlando style which are 
testimony to his incomparable technical mastery. Themes from old 
English and Italian masters are most skilfully interwoven to suggest the 
atmosphere and time of the action. 

The one-act Der Friedenstag (The Day of Peace) (1938) has a libretto 
by Joseph Gregor dealing with an episode at the end of the Thirty 
Years’ War; but the subject takes оп a universal human significance 
in the dramatic confrontation of the Commander, bound by his military 
duty to blow up the fortress rather than surrender it, with the population 
craving for peace. Less opera than dramatic oratorio, Der Friedenstag 
has big choral scenes at the beginning and end, and indeed concludes 
with a grandiose hymn for peace that owes something to the last-act 
finale of Fidelio. In Daphne (1938) and Die Liebe der Danae (1940), both 
to libretti by Gregor, Strauss returned for the last time to mythological 
Baroque opera of which he was particularly fond. The first, a one-act 
work, mingles tragedy with idyll, with the ancient myth interpreted in 
terms of modern psychology while the music, though devoid of potent 
melodic invention, is among Strauss's technical masterpieces—subtle, 
refined, rich іп motivic work, and imbued with a feeling of Мед!- 
terranean serenity. Die Liebe der Danae, a‘Cheerful Mythology in Three 
Acts', based on an unfinished scenario by Hofmannsthal, telescopes the 
separate myths of Midas and Danae, the underlying theme being the 
victory of true love against which even a god is powerless. The musico- 
dramatic treatment of Jupiter is strongly reminiscent of that of Wotan 
in Die Walküre; and indeed the opera as a whole echoes the Wagner of 
the Ring, notably in its harmonic and orchestral idiom, though the vocal 
writing for Danae and Midas bears in its sustained cantabile Strauss's 
own individual signature. 

Strauss's last opera, Capriccio (1941), is the culmination of the light- 
handed conversational manner which he cultivated in Ariadne, Inter- 
mezzo, and parts of Arabella. The subject, taken from a libretto by 
the Abbate Casti for Salieri's now forgotten opera, Prima la musica e poi 
le parole, was elaborated by the composer and the conductor Clemens 
Krauss into a play dealing with the problem of words and music in 
opera. The action, which takes place in a cháteau near Paris about 1775, 
centres on a young countess who is to decide which of her two im- 
petuous suitors—the poet Olivier and the musician Flamand-—she is to 
marry. In the end she cannot decide for either, a symbolism implying 
the perfect equality of text and music. According to the dramatic 


RICHARD STRAUSS'S LAST YEARS 325 


exigencies, now the words, now the music must prevail and now again 
both will be poised in equilibrium. Strauss moves with the most natural 
ease between parlando, arioso, and more sustained lyrical styles, while 
the orchestral writing is in the sparse and supple style of chamber 
music. 

Strauss was seventy-eight at the time of Capriccio, which he intended 
to be his last work. But his Indian summer continued and saw the 
composition of seven more works—the Second Horn Concerto (1942), 
the First and Second Sonatina for sixteen wind instruments (1943 and 
1945, respectively), the Metamorphosen for twenty-three strings (1945), 
the Oboe Concerto (1946), the Duet Concertino for clarinet and bassoon 
(1947), and the Vier letzte Lieder (1948). Strauss jestingly referred to 
these works as ‘wrist exercises’ and ‘snippets from my workshop’. With 
the exception of the Metamorphosen, they are all slight in musical 
substance and in a simple unassuming vein, written with the aim of 
keeping his hands busy and of delighting the senses. They are moulded, 
more or less closely, on classical (Mozartian) principles and thus 
Strauss's creative career came full circle: for a classical tendency, mani- 
fest in the young Strauss, showed itself again in the octogenarian, yet 
greatly enriched and mellowed by the artistic and human experiences 
of a life-time. The neo-classical tendency is displayed in a number of 
features: in the turn to pure instrumental music; in the avoidance of an 
emotionally charged expression and the emphasis on exquisitely refined 
and polished workmanship; in the symmetrical cut of thematic ideas 
(mostly in regular four and eight bars) and ‘old-fashioned’ cadences; іп 
the marked preference for simple diatonic writing, and in the transparent 
scoring, whose ѕрагѕепеѕѕ and economy—already to be found in Daphne 
and Capriccio—is in strong contrast with the sumptuousness and lavish- 
ness of Strauss's symphonic poems and the majority of his operas. Meta- 
morphosen, which is the most important of Strauss's late compositions, 
shows all these features at their most characteristic, to say nothing of 
a formidable skill displayed in the polyphonic interweaving of the parts. 
Written at the end of the last war and using as one of its four themes the 
first four bars of the Funeral March in Beethoven's Eroica, the work is a 
moving expression of the composer's sadness and grief at the catas- 
trophe that had overtaken Germany and at the same time a nostalgic 
and melancholy reflection on his past life. Equally autobiographical is 
the cycle of the Four Last Songs, tinged as it is with a gentle melancholy. 
The soprano solo part is closely integrated with the orchestra, singer and 
instruments forming an organic whole and weaving a continuous finely 
drawn tapestry of sound from which the voice stands out merely by its 
different timbre. The cycle represents a miniature emotional drama, 


326 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


moving from the subdued animation of ‘Im Frühling’ to a complete 
withdrawal in the closing ‘Im Abendrot’. 


PFITZNER AND SCHREKER 5 


In Hans Pfitzner (1869-1949) Strauss encountered his one serious 
rival to the claim of being the most representative German operatic 
composer of his time. Pfitzner’s music is little known outside Central 
Europe, where he is generally considered as one of the great masters. 
Pfitzner described himself as ‘the last Romantic’ and advanced a theory 
that Germany's last great period in music was the Romantic era and that 
in the twentieth century the country had outlived its musical hegemony 
over the other nations. Not unexpectedly, this philosophy was accom- 
panied by a declared enmity to all modern tendencies and trends, which 
Pfitzner expressed in two pamphlets—Futuristengefahr (1917) and Die 
Neue Aesthetik der musikalischen Impotenz (1919) which were both 
written in reply to Busoni's Entwurf einer Neuen Aesthetik der Tonkunst 
(1907). Pfitzner's music expresses a profound longing for the German 
past and is pervaded by a melancholy and a pessimism to be fairly 
expected of a composer so deeply steeped in Schopenhauer and Wagner. 
Characteristic of this attitude of mind are two works both of which look 
backward in their sentiment and style. In the ‘romantic cantata’, Von 
deutscher Seele (1921), to poems by Eichendorff, Pfitzner seeks to convey 
the mystical yearnings of the romantic artist for the ineffable and the 
transcendental. The opera Palestrina (1917), to his own libretto, 
presents a Künstlerdrama in which, freely adapting the life-story of the 
great sixteenth-century Italian church composer with whom he 
identifies himself, Pfitzner demonstrates the spiritual solitude and isola- 
tion of the artist amid the worldliness and the intrigues of his surround- 
ings. The work is unquestionably Pfitzner's masterpiece and perhaps 
the only one by which posterity is likely to remember this conscious 
follower of Schumann and Wagner.! 

The chief reputation of Franz Schreker (1878-1934), whose name was 
bracketed with those of Strauss and Schoenberg in the Germany of the 
1920s, rests upon his operas of which the most important are Der ferne 
Klang (The Distant Sound) (1913), Die Gezeichneten (The Branded) 
(1918) and Der Schatzgrüber (The Treasure-Digger) (1920). They are 
eclectic in style, much indebted to Strauss (orchestral style), Puccini 
(melody), and Debussy (harmony), and, while dramatically and sceni- 
cally effective, suffer from an obscure erotic symbolism based on ideas 
from Freud and Wedekind. 


1 See also p. 195. ? See pp. 185—7. 


FERRUCCIO BUSONI 327 


FERRUCCIO BUSONI! 

One of the first champions of the New Music in Germany was 
Ferruccio Busoni (1866-1924), who sought to achieve an individual 
*Young Classicism' by the fusion of Latin clarity of form with German 
technical solidity, a fusion in which form, content, and expression 
should be in equilibrium. The conflict between Italian and German in 
Busoni's heredity was reflected in his double life as performer and 
composer. He was one of the greatest piano virtuosi of his day, and a 
sense of kinship with Liszt found expression in studies, arrangements, 
and, perhaps most notably, іп the Chamber Fantasy оп Bizet’s Carmen, 
the last of six piano ‘sonatinas’ written between 1910 and 1922. The 
fifth of these (‘In signo Johannis Sebastiani Magni’) reflects the other 
major influence in Busoni’s own music, that of J. S. Bach, though the 
contrapuntal writing is unambitious compared with that of his Fantasia 
Contrappuntistica (four versions, the last for two pianos, 1922). This 
started as an attempt to solve the problem of the unfinished fugue in 
Die Kunst der Fuge, but proliferated to include a choral prelude and a 
quadruple fugue on the Bach fragment. Busoni’s design leads gradually 
from the procedures of baroque counterpoint to the most daring poly- 
phonic complexities, and the work represents а twentieth-century 
counterpart to Die Kunst der Fuge. 

The same intellectual curiosity and intensity that prompted Busoni’s 
writings are reflected in the second sonatina (1912), while the fourth 
(‘In die Nativitatis Christi MCMXVII’) shows a rare equilibrium 
between the classicizing formal tendencies and the deeply romantic 
kernel of Busoni’s character as a creative artist. There are echoes here 
of Beethoven’s last manner (especially of the Diabelli Variations), and 
the style is not unlike that of Fauré’s last works in harmony and lay-out. 
The cerebral experimentation, which makes much of Busoni’s music 
interesting rather than satisfying, is almost wholly absent. 


PAUL HINDEMITH 

Paul Hindemith (1895-1964) was outstanding in the generation of 
German composers which came to maturity during the late 1920s and 
whose works display the diverse facets of all that is implied in the term 
‘New Music’. If Hindemith became its acknowledged leader, this was 
for two main reasons: first, because the various modern tendencies in 
post-war German music crystallized in his personality with extra- 
ordinary clarity and force; and second because of his superior musical 
gifts, his tremendous fertility and vitality, his truly extraordinary 
technical resourcefulness and industry. Hindemith provides an illustra- 

! On Busoni's operas, see pp. 192-5. 


328 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


tion of that rare combination of a sheer Musikant—a musical ‘natural’— 
with a seemingly inexhaustible fund of creative energy and a technical 
command that recalls the supreme craftsmanship of the great masters 
of the seventeenth and eighteenth centuries. His own attitude towards 
composition has indeed a great deal in common with the approach of 
the pre-nineteenth-century composers. Like them, Hindemith regarded 
himself as primarily a craftsman who pursues his métier with the utmost 
of his skill and intelligence and with whom purely musical problems— 
invention, form and design, balance and proportion, texture and instru- 
mentation—take precedence over expressive qualities. The latter he 
considered during his early period as merely incidental to his funda- 
mental conception of the musical work as an edifice, a construction in 
sound. In his younger years his artistic personality found its chief 
expression in music characterized by a reckless iconoclastic spirit, an 
irrepressible buoyancy and masculine vigour, a sense of irreverent fun 
and a good-natured cynicism. At the time he was regarded as an enfant 
terrible, thumbing his nose at established traditions and conventions and 
disconcerting the public and the critics by his unpredictable antics. 
With the beginning of the 1930s his more serious side came gradually to 
the fore—a Hindemith high-minded and lofty in his attitude to art and 
concerned with things of the mind, with general human problems and 
with philosophical and metaphysical themes. 

Hindemith started as a spiritual disciple of Brahms and even more 
Reger, adopting the latter's use of Bachian counterpoint as a regenera- 
tive constructive means and soon a marked individuality begins to 
show, as in the Second String Quartet, Op. 10 (1919), and the Kammer- 
musik, Op. 24, no. 1 (1922), which display three salient features. First, 
Hindemith is seen to establish an autonomous melodic principle: the 
lines become energy-laden and take on a dynamic ‘motoric’ character 
in which movement per se is the main concern: 


PAUL HINDEMITH 329 


Secondly, functional harmony is in the process of being destroyed by 
an emphasis on seconds, fourths, and sevenths as constructive melodic 
and harmonic intervals which implied an equality between consonance 
and dissonance, though even in his extreme ‘atonal’ period Hindemith 
never accepted the complete ‘emancipation of the dissonance’ envisaged 
by Schoenberg. And thirdly, he now discarded Reger's harmonic 
counterpoint in favour of a bold linear style, in which the vertical 
factor was almost invariably the incidental result of the part-writing. 
At the same time jazz, which became known in Germany about 1920, 
begins to colour Hindemith's manner, as in the Piano Suite, Op. 26 
(1922), and in the finale of the above-mentioned Kammermusik which is 
scored for a small orchestra, including nine percussion instruments. 
This was partly in imitation of a jazz band, partly to ensure the 
utmost clarity of the melodic and rhythmic patterns. The Kammermusik 
is the first instance of Hindemith's interest in orchestral chamber music, 
largely in reaction against the lavish orchestral style of the romantic 
period. The Third String Quartet, Op. 22 (1922), representsthesumming- 
up of all the stylistic aims pursued by the young Hindemith, yet at the 
same time it is also a transitional work paving the way for the neo- 
classicism which finds its first clear expression in the song cycle Das 
Marienleben, Op. 27, for voice and piano (1923). This setting of poems by 
Rainer Maria Rilke consists of fifteen songs divided into four groups, 
each with a character of its own: idyllic and lyrical in the first group, 
dramatic in the second, growing in tension and emotive expression in the 
third, until the fourth group is marked by an almost abstract treatment of 
the text, the musical structure completely absorbing the dramatis per- 
sonae and the action. In the majority of these songs the voice is handled 
in a rather unvocal, instrumental manner and often forms with the piano 
a two- or three-part polyphonic texture. The various forms employed are 
those of Baroque music: ‘Die Darstellung Мапа im Tempel’ (no. 2) 
for instance is а passacaglia, and “Уот Tode Мапа IP (по. 14) is 
designed as a theme with five contrapuntal variations. The setting, in 
avoiding all descriptive and picturesque elements, expresses the poetic 
essence of the words in a more or less stylized and objective way. Das 
Marienleben is a significant landmark in Hindemith's technical develop- 
ment, and evidence of the importance he attached to it may be seen in 
the fact that in 1948 he subjected the music to a fairly extensive revision, 
smoothing down the harshness and angularity of the voice part and 
clarifying the harmonic and contrapuntal texture, but in the process 
sacrificing youthful boldness and spontaneity.! 

1 No. 12, ‘Stillung Мапа mit dem Auferstandenen’, the only song completely unchanged 


in the new version, is recorded in The History of Music in Sound, x. And see Ex. 189, 
pp. 407-8. 


330 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Hindemith's indebtedness to the spirit of Bach reached its highest 
point in the contrapuntal complexity and compactness of the Fourth 
String Quartet, Op. 22 (1924), with a fugue in the first movement and in 
the finale a passacaglia and fugato, which constitute a ne plus ultra in 
close polyphonic thinking. In subsequent works Hindemith began to 
free himself of this dense contrapuntal writing; texture becomes 
gradually looser, more clearly defined in the disposition of the parts, and 
a decorative element is seen to play an increasingly important role. In his 
orchestral style Hindemith more and more resorts to the concertante 
principle, with a resulting gain of clarity in the general texture; and, 
possibly influenced by Stravinsky, he tends to replace, within the single 
movement, construction-by-contrast with construction-by-similarity. 
Typical of this phase are the Piano Concerto, Op. 29 (1924), the Concerto 
for Orchestra, Op. 38 (1925) and the Kammermusik, Op. 36, nos. 3 and 4 
(1925 and 1927 respectively) which are essentially concertos for violin 
and viola, the latter being a demonstration of Hindemith's concertante 
writing at its most felicitous. 

His temporary interest in the “Neue Sachlichheit’ (New Objectivity) 
prompted Hindemith to devote himself to Gebrauchsmusik—utility’ or 
functional music conceived for special purposes such as the film, the 
radio and the school. His various Kammermusiken are in a sense also 
‘utilitarian’ in that, like so much eighteenth-century music, they were 
written for special players or special occasions. Both he and Kurt 
Weill! composed works which are wholly subservient to those uses and 
also intended to bring music to the home and tempt the amateur to play 
it, thus trying to bridge the ever-widening chasm between the composer 
and the ordinary music lover or between ‘the producer and the con- 
sumer of music’, as Hindemith put it in the language of the economist. 
Works of this kind include the Schulwerk fiir Instrumental-Zusammen- 
spiel (1927), the school opera, Wir bauen eine Stadt (We are building a 
town) (1930) and the Sing-und-Spielmusik ftir Liebhaber und Musik- 
freunde (1928-31). In a preface the composer points out that it is not 
written for the concert hall or the professional musician, but is intended 
to provide modern studies for those ‘who wish, for their own pleasure, 
to sing and play or to perform before a small circle of people who share 
their taste’. Yet, after an initial vogue for this kind of music, which 
found its most concrete expression in the music festival held by and for 
young people at Plón in the summer of 1932, enthusiasm waned and the 
whole movement gradually petered out, though in retrospect it must be 
seen as a genuine attempt to win the interest of the ordinary music lover 
for contemporary music. 

1 On Weill, see infra, p. 339. 


PAUL HINDEMITH 331 


Although opera might seem alien, or, at any rate, not as congenial to 
Hindemith's cast of mind as 'absolute' instrumental music, his versa- 
tility, his strong creative energies, and his intellectual curiosity prompted 
him to a number of operatic works. He started with three one-act 
operas, all written in 1921: Mörder, Hoffnung der Frauen (Murderer, hope 
of women), to an expressionist libretto by the painter Oskar Kokoschka, 
the subject of which is symbol-laden to the point of unintelligibility; 
Sancta Susanna, whose libretto by August Stramm portrays female 
eroticism in conflict with the rigours of monastic life; and Das Nusch- 
Nuschi, a satirical play for Burmese marionettes by Franz Blei, treated 
by Hindemith in buffo style and with a great deal of musical parody. 
With the exception of the last work, there is a very noticeable incon- 
gruity between the involved and obscure libretto and the clear-cut, 
stylized nature of the musical setting. 

His next opera, the three-act Cardillac (1926), to a libretto by 
Ferdinand Lion drawn from E. T. A. Hoffmann's story, Das Früulein 
von Scuderi, demonstrates Hindemith's tendency towards objectivity and 
stylized treatment. All psychological and descriptive elements are 
avoided in favour of a strictly musical response to the drama, whose 
stage action frequently runs on a different plane from that of the music. 
Hindemith here aims at a return to the aesthetic of late baroque opera, 
the music having the first say and the individual scenes being not much 
more than a peg on which to hang such 'closed' autonomous musical 
forms as arias, duets, and ensembles. But it has to be admitted that 
the opera suffers from the immobility of these forms in relation to 
the dramatic happenings recalling the style of an oratorio, and from 
too rigid an application of polyphony that often creates the im- 
pression of being written for its own sake, as in the final chorus—a 
passacaglia of considerable length. Also Hindemith's portrayal of the 
chief dramatis personae as objective types is in conflict with the charac- 
ters of the libretto, especially that of the title-hero—a Parisian goldsmith 
who, in his pathological mania to retrieve by murder the jewels pur- 
chased from him by his customers, is a truly romantic and very indivi- 
dual figure. Moreover, the differentiation in the musical delineation of 
the various characters is not carried far enough, though there is an 
attempt to depict Cardillac and his daughter in particular colours— 
the former by a tortuous, darkly scored theme and the latter by florid 
writing. In 1952 Hindemith subjected the opera to some dramatic re- 
visions: a great deal of the text was entirely rewritten, the vocal lines 
were rendered more singable, parts of the score were amplified to 
match the alterations in the plot, and the orchestra was made to yield 
а richer and more varied quality of sound, in keeping with the 


332 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


composer's later style of instrumentation.! Yet, as in the case of Das 
Marienleben, much of the direct and uncompromising character of the 
original version was lost in this process of improving what Hindemith 
regarded as the indiscretions of his youth. The tragedy of Cardillac 
was followed by two comic operas both of which reflect the contem- 
porary inclination to deal with topical subjects. The first of these was the 
one-act Sketsch [sic] mit Musik, Hin und zurück (There and Back) 
(1927), an example of the miniature type of opera then in vogue as a 
reaction against the ponderous seriousness and the huge dimensions of 
Wagnerian and Straussian music-drama. The action deals with a matri- 
monial tragedy in an amusingly flippant manner and reverses the plot 
halfway through, so that the opera ends as it began; the music moves 
from that point also partly in retrograde motion and makes use of 
elements from revue and cabaret. The three-act Neues vom Tage (News 
of the day) (1929), to a libretto by Marcellus Schiffer, who also wrote 
the book of the previous opera, deals with the sensationalism of modern 
newspapers which pry, recklessly and callously, into the private affairs 
of citizens. It is in the buffa style of Hin und zurück, but is more ambi- 
tiously conceived, with great choral scenes and orchestral interludes, 
and contains parodistic skits on the traditional romantic opera—a ‘hate 
duet’ instead of the customary ‘love duet’, and a ‘divorce ensemble’ 
instead of a ‘wedding ensemble’. Altogether, it is a work that witnesses 
to Hindemith’s high spirits and his good-natured irony, though it is a 
little weighed down by his obsessive urge for elaborate contrapuntal 
writing. 

At the beginning of the 1930s, which also marks the beginning of his 
full maturity, Hindemith’s style underwent a very noticeable change in 
the direction of greater relaxation and mellowness. Technically, this is 
seen in the abandonment of a massive, close-knit texture in favour of a 


Ex. 144 
= DEM GG We 
Sehr ruhig 


Ist Oboe 


2nd Oboe 


Cor Anglais 
іп Е 


1 See Н. L. Schilling, Paul Hindemith’s ‘Cardillac’. Beiträge zu einem Vergleich der beiden 
Opernfassungen (Würzburg, 1962). 


PAUL HINDEMITH 333 


с” 


lighter, more transparent treatment, in a greater differentiation being 
made, so far as polyphony is concerned, between main voices and 
subsidiary parts, and in a more clearly defined feeling for key, as in 
the Philharmonic Concerto of 1932 (Ex. 144). 

Typical of this new phase is not only the Philharmoni€ Concerto, con- 
sisting of six variations on a markedly expressive theme, but the oratorio 
Das Unauf hórliche (The Incessant) for four soloists, mixed chorus, boys’ 
chorus, and orchestra (1931), in which the expressive emotional element, 
that up to now had been allowed into his music through the back door, 
assumes an increasingly important role and lends Hindemith's neo- 
classicism a romantic overtone: 

Ex. 145 
Langsam p 


к=з сш зыш 


trägt die Nacht__ das 


Soprano 
Solo 


Piano 


334 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


и 


C uL s 
е 


(Night bears the end.) 


The oratorio is a setting of a half-mystical, half-philosophical text by 
Gottfried Benn, which contrasts eternity with the transience of human 
life and the perpetuity of creation with the temporary aspirations and 
achievements of mankind. This monumental work is in three parts, of 
which the middle is chiefly allotted to the soloists and is flanked by two 
predominantly choral parts, where the great choruses in contrapuntal- 
imitatory style show Handel’s influence while the melodic writing is 
coloured by Gregorian chant. 

The work that best illustrates Hindemith's mature mode of expression, 
and that in its spirituality and musical vision must be accounted a 
masterpiece, is the opera Mathis der Maler (1934), to which he wrote 
his own libretto. Like Die Meistersinger and Pfitzner's Palestrina, this 
opera is an ‘artist’s drama’ centring on the figure of Dürer's contem- 
porary Mathis Grünewald, who is for the composer a symbol of the 
German artist of his own time. The story deals with significant events in 
the painter's life and is set against the background of the Peasant War 
in South Germany in the sixteenth century. The spiritual tensions 
and conflicts which torment Mathis's mind arise from his participa- 
tion in the politics of his country and from his profound questioning 
of his artistic mission; and they take on a topical meaning when 
understood as a reflection of Hindemith's own problems as artist 
and man in Nazi Germany. The opera is a refutation of the totalitarian 
dogma that the artist must be a political animal, subservient to the 
State, and proclaims as its chief message that for an artist to serve any 
other master but himself is to forfeit his moral and intellectual integrity 
and to deny his true mission. More epic than dramatic in character, it 
approaches the style of a scenic oratorio, with big choral scenes in 
contrapuntal manner and with extensive slow-moving solo numbers, 
duets, and ensembles. Wagnerian leitmotive technique alternates with 
the formal devices of the older opera. Occasional use is made of 


PAUL HINDEMITH 335 


German folk-song such as the medieval tune ‘Es sungen drei Engel’ 
(Ех. 146 (i) ) in the Prelude, Engelkonzert, of Gregorian chant as at the 
end of the seventh scene (ii) and of pentatonic melodies modelled on 
Gregorian chant (iii): 


(Mighty Ruler, true God who controllest the changes of things.) 


These lend the music an archaic flavour suited to the ambience of 
this idealistic and noble work. The symphony Mathis der Maler (1934) 
is based almost entirely on music from the opera. 

Of Hindemith's three pre-war ballets, Nobilissima Visione (1938), to 
a choreographic story by Leonid Massine, shows in eleven scenes the 
life of St. Francis of Assisi. Lofty in conception, it is not primarily 
ballet music in the Stravinsky sense, for the range and variety of pure 
dance pattern is restricted. Other works of Hindemith's pre-war period 
include Der Schwanendreher (1935), a kind of viola concerto whose 
material is largely derived and elaborated from a German folk-song. 
(Hindemith was for a number of years viola player in the Amar Quartet 
and frequently appeared as soloist in his own works for that instrument, 
as he did in Der Schwanendreher.) His three Piano Sonatas (all 1936) 
and the Sonata for Piano Duet (1938) are splendid examples which 
show the synthesis of an essentially neo-classical style with a romantically 
expressive mode of utterance. 


336 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


With a composer of Hindemith's temperament a profound interest 
in the theoretical aspect of his craft is not surprising. His was that rare 
combination of a creative and reproductive artist with a searching and 
lucid theoretical thinker. Evidence of this is provided by his textbook, 
Unterweisung im Tonsatz*, which is an exposition of the theory and 
practice of his own methods, in which the organization of the entire 
melodic and harmonic material is based on a ‘diatonization’ of the 
twelve notes of the chromatic scale—the very opposite of Schoenberg's 
dodecaphonic system. 

Like Schoenberg, Hindemith takes for his basis the twelve chromatic 
notes; but, unlike the Schoenbergian series, the twelve notes are strictly 
organized and stand in clearly defined relationships to both a central 
note, the tonic, and to one another. It is these relationships which 
determine the melodic and harmonic values of the twelve notes. 
Hindemith takes as his point of departure the harmonic or overtone 
series acclimatizing its ‘impure’ notes to the temperate scale, and 
derives from it two rows. In contrast to Schoenberg's complete equaliza- 
tion of the twelve chromatic notes, Row I shows a hierarchical order: 


(Hindemith includes the octave as the note next in importance to the 
first note). 

The nearer a note lies to C, the closer its relationship to it; the 
further away from it the weaker this relationship until in the twelfth 
note, B, it is at its weakest. The tritone Е sharp is considered by Hinde- 
mith an outsider standing in a neutral relationship to the tonic C and 
playing an ambivalent role (see p.337). In this he differs from the 
generally held modern view that the tritone is completely unrelated to 
C and therefore represents an extreme, or zero, in the scale of relation- 
ship. 

The varying degrees of relationships to the central note, or tonic, 
constitute for Hindemith ‘rule and measure for the linking of sounds, 
for the order of harmonic progressions and thus for the further course 
of the musical happenings'.? Row I is what has been called a ‘functional 
mode’,? i.e. it indicates the functional role of the twelve chromatic notes 


1 Mainz, 1937; translated as The Craft of Musical Composition (New York, 1942). 

2 Ibid., р. 74. 

*'This term was first used by Richard Hill, Schoenberg's Tone-Rows and the Tonal 
System of the Future', Musical Quarterly, xxii (1936), p. 14. 


PAUL HINDEMITH 337 


о a centre. There is, to repeat, a world of difference between this row 
and the Schoenbergian note-row. The latter is a melodic model or store 
of motives for practical use in a composition, whereas the Hindemithian 
series has no thematic significance whatsoever and is merely a graph 
showing functional relationships. 

It gives only one set of tonal relationships, i.e. the degree of nearness 
to or distance from the tonic. Another row is needed to show the 
relationship between the twelve notes, namely their intervallic relation- 
ship which determines the harmonic value of an interval, according to 
its degree of consonance and dissonance. This purpose is served by 
Hindemith's Row II which he derived from the so-called combination 
tones: 


(The arrows indicate the respective roots of the intervals). 

Hindemith takes full account of the modern conception of the relativity 
of consonance and dissonance by drawing no strict demarcation line 
between the two; there is merely a gradual transition from simple and 
perfect intervals (the first seven intervals of Row П) to less simple 
intervals of higher harmonic tension (the next four intervals) until ex- 
treme tension is reached in the major seventh. The tritone, on the other 
hand, which has no root, is a neutral interval whose harmonic value 
depends on the context in which it occurs. 

On the basis of Row II Hindemith proceeds to build a system of 
chords comprising all possible combinations of intervals whose har- 
monic value is determined by what kind of simple and less simple 
intervals they contain. If chords of a higher harmonic tension move 
towards chords of lesser tension the result of such a progression is a 
harmonischer Fall or harmonic descent—in the opposite case, a harmon- 
ischer Anstieg or harmonic ascent. In other words, the harmonic 
tension decreases and increases in direct ratio to the progression from a 
wertvoller, or valuable, chord to a less valuable chord and vice versa. 
This crescendo and diminuendo in harmonic tension Hindemith calls 
harmonisches Gefälle or harmonic incline, and the planning of a gradual 
and balanced incline is, according to him, the purpose and test of a 
good chord progression, unless a sudden fall in harmonic tension is 
sought for some specific aesthetic effect. For instance, in the following 
progression of six chords, the harmonic tension is seen to increase 
slowly towards the fourth chord and then diminish towards the last: 


338 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Ex. 149 
Y 


Thus, the harmonic Gefälle of this example is well-balanced. Hindemith 
argues that by assessing the harmonic values of chords with the help 
of Row II, composers are given a means whereby to organize and 
control in a deliberate manner the course of harmonic progressions and, 
indeed, the harmonic disposition of a whole piece. 

A. word must be said about Hindemith's theory of tonality in modern 
music. Since his system, based on the natural laws of musical acoustics, 
constitutes a 'diatonicized' chromaticism in which every one of the 
twelve notes is related to a central note—just as the seven diatonic 
notes of the major-minor system are related to a tonic—he maintains 
that it is possible to establish a tonality in every kind of music, including 
even twelve-note music, as he attempts to show in his analysis of a few 
bars from Schoenberg's Piano Piece, Op. 33a.! His procedure in finding 
the key of a given passage is, roughly, the following: he first looks for 
the roots of a given chord progression, then arranges these roots in a 
line, ascertains with the help of Row II the ‘best’ interval, and the root 
of this interval is the tonic of the passage in question: Thus, the line 
formed by the roots of Ex. 149 is: 


Ex. 150 
a> ice 


X 


Its ‘best’ or simplest interval is the perfect fourth G—C whose root is C 
--һепсе C is the tonic of the above passage. 

Whatever might be thought of the involved way in which Hindemith 
establishes the two fundamental rows on which his whole system rests, 
it must be said that Unterweisung іт Tonsatz represents, even after the 
lapse of a whole generation, a very valuable contribution to composi- 
tional theory and a serious attempt to introduce tonal order and 
organization into the complex phenomena of music in the inter-war 
years. 


KURT WEILL AND ERNST KRENEK 
The two most notable German contemporaries of Hindemith during 


1 Unterweisung, pp. 244-45. 


KURT WEILL AND ERNST KRENEK 339 


the inter-war period were Kurt Weill (1900-50) and Ernst Křenek 
(b. 1900). Weill, a pupil of Busoni, began his career with works in which 
he pursued a boldly experimental, expressionistic and abstract line 
(Divertimento, 1923; String Quartet, 1923, and Frauentanz for soprano 
and five instruments, 1924). He subsequently turned to the stage as his 
main field of activity, adopting a more realistic and topical approach 
and creating a new type of German opera in which such heterogeneous 
elements as those from the old Singspiel, jazz, cabaret and Uberbrettl (a 
German form of intimate revue) are moulded into an individual style. 
For this reason and also on account of the raciness of his melodic idiom 
and the pungency of his much simplified harmonic language, Weill’s 
operas achieved at the time a wide popular success. Like the young 
Hindemith, Krenek, and other contemporary composers (Max Brand 
in Maschinist Hopkins, 1929), he cultivated Zeitoper, operas with a 
topical political and social subject in which bourgeois society was 
pilloried and made responsible for the social injustices of the class 
system, as in Aufstieg und Fall der Stadt Mahagonny (Rise and fall of 
the town of Mahagonny) (1927), Die Dreigroschenoper (Threepenny 
Opera) (1928), and Die Biirgschaft (The Security) (1932) in which a 
strong satirical vein is tempered by an intermittent serious attitude. He 
also followed the short-lived vogue for the Гей ск, or educational 
music, in the school opera Der Jasager (The Yea-sayer) (1930). Signifi- 
cantly, five of Weill’s stage works have libretti by the left-wing poet and 
playwright, Bert Brecht; and their collaboration produced, in Die 
Dreigroschenoper, their best-known and most successful work, the 
model for which was John Gay’s The Beggars’ Opera. This was a thinly 
veiled attack on the corruption of contemporary public characters and 
institutions, and similarly the German work represents a mordant satire 
on modern society and public institutions. It is not an opera in the 
traditional sense but a mixture of Singspiel and cabaret, consisting of set 
numbers (arias, duets, ensembles, and parodistic chorale) with inter- 
spersed spoken dialogue, the whole accompanied by a small orchestra. 
Weill replaces the popular melodies of the English original by jazz-like 
tunes of a most immediate appeal and of such simplicity that they can be 
sung by actors, not only by professional singers; his argument for using 
this device was that jazz represented the folk-music of modern times. 
Whereas Weill’s output was largely devoted to opera, Krenek dis- 
played a Protean versatility; in fact, he may be said to be the German 
counterpart of Milhaud. The fact that Kienek’s intellect has always 
been more potent than his creative instinct possibly explains the lack of 
a firm direction in his stylistic development. His zigzag course also 
reflects the spiritual insecurity and absence of a strong inner purpose 


340 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


characteristic of the German Zeitgeist in its extreme manifestations. 
Like Hindemith, Křenek allowed himself to be influenced by the various 
*isms' of the 1920s, but, unlike Hindemith, he did not succeed in achiev- 
ing a distinct personal mode of expression. Не, began as a radical and 
uncompromising experimentalist in an intransigent atonal style— 
three Symphonies (1921-22); four String Quartets (1921-24); the scenic 
cantata Die Zwingburg (The Citadel) (1922); the farcical detective thriller 
Der Sprung über den Schatten (The Leap over the Shadow) (1923), and 
the expressionist Orpheus und Eurydike (1923), the latter to a libretto by 
Kokoschka which treats the ancient myth in a modern psychological 
manner. Then КїепеК executed a volte face, abandoning the experi- 
mental and abstract and turning to a frankly popular idiom. In Jonny 
spielt auf (Johnny strikes up) (1927) a Negro dance-band leader is the 
hero, and Křenek makes use of jazz and the spectacular devices of 
modern stage production, including the radio and a real locomotive.! 
After this he changed his course again, paying tribute to topical Zeitoper 
in the trilogy Der Diktator (1926), Das geheime Königreich (The Secret 
Kingdom) (1927), and Schwergewicht oder die Ehre der Nation (Dead 
Weight or the Honour of the Nation) (1927), all one-act operas. Then 
came a neo-romantic phase in which he drifted into a diatonic style 
best exemplified in the song cycle, Reisebuch aus den österreichischen 
Alpen (Travel Book from the Austrian Alps) (1929), which was largely 
modelled on Schubert's Winterreise; Fiedellieder (Fiddle Songs) (1930), 
and the tragedy Das Leben des Orest (The Life of Orestes) (1930), 
treated as an intensely passionate human drama in the manner of grand 
opera. Returning from his studies in Berlin with Franz Schreker to 
Vienna (1928) he came into close contact with the works of Schoenberg 
and his school, and after prolonged hesitation decided to embrace the 
twelve-note system with all its theoretical and practical implications.? 
From then onwards the majority of Krenek’s compositions were written 
in serial style, such as the monumental Karl V (1933), which draws оп 
elements from the historical drama, grand opera, dumb-show, and the 
film, and is comparable to Milhaud's Christophe Colomb. 


THE SECOND VIENNESE SCHOOL—ARNOLD SCHOENBERG 


In the flux of currents and cross-currents characterizing European 
music during most of the inter-war period, there was one firm rock 
formed by three composers who are collectively known as the Second 
Viennese School—Arnold Schoenberg (1874-1951) and his two dis- 
ciples Alban Berg (1885-1935) and Anton Webern (1883-1945). This 


! See pl. V (b). 
? See his Über neue Musik (Vienna, 1937) and Studies in Counterpoint (New York, 1940). 


THE SECOND VIENNESE SCHOOL--ARNOLD SCHOENBERG 341 


was the only group of composers which showed a consistent develop- 
ment in a single direction, namely the exploitation of a new method of 
composition that represented a revolutionary departure from all other 
methods known in musical history. In its aesthetics, however, it adhered, 
with the exception of the later Webern, to the chief tenet of romanticism 
—music as the vehicle of subjective emotion—but music subjected to 
the immense pressure of largely unconscious irrational forces of the ego 
and thus becoming expressionist in character; in this sense the Second 
Viennese School may be said to represent a reaction against neo- 
classicism and the New Objectivity. 

In earlier chapters of this volume it has been shown how Schoenberg, 
in developing and intensifying Wagner's chromaticism, arrived in the 
period 1908-1914 at a free chromatic or atonal style. His works written 
in this style—a fluid, amorphous, and invertebrate atonal mass— 
represent the liquidation of all previous means of formal and harmonic 
organization; and they were all very short works (Three Pieces for piano, 
Op. 11, Five Pieces for orchestra, Op. 16, Six Piano Pieces, Op. 19) for 
the relinquishing of all harmonic devices, used hitherto to create formal 
articulation in tonal music, precluded the employment of larger forms; 
while in those compositions which did show a measure of formal exten- 
sion, it was the text that conditioned the form (Das Buch der hüngenden 
Gärten (The Book of the Hanging Gardens), Ор. 15, Erwartung (Expecta- 
tion), Op. 17, Die glückliche Hand (The Lucky Hand), Op. 18, and 
Pierrot lunaire, Op. 21). In these ‘free atonal’ works Schoenberg had 
arrived at an end, he found himself in an impasse; and it was in con- 
sequence of this that he began experimenting with a novel principle of 
construction which, analogous to the formal organization of tonal 
music, would guarantee order and coherence and at the same time 
introduce a unifying principle into the melodic and harmonic material. 
These aims he achieved by the method of what he himself called ‘сот- 
posing with the twelve chromatic notes related only to one another’ and 
which is generally known as the twelve-note or serial method, a method 
foreshadowed in some of Schoenberg's atonal works, such as Pierrot 
lunaire and the Orchestral Songs, Op. 22, where an attempt is made to 
tie together the structure by means of recurrent motives. Similarly, the 
last part of an abortive symphony (1915), which later became the ora- 
torio Die Jakobsleiter (Jacob's Ladder), contained a scherzo with a 
dodecaphonic theme. In the oratorio it appears at the very opening— 
the first six notes in the form of a basso ostinato, the last six notes 
sustained in a chord above it: 


342 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Ex. 151 
Sehr rasch 


The fundamental idea behind this new mode of composition was to 
replace one structural force—tonality—by another force—thematic 
unity;or, seen from the point of view of Schoenburg's pre-dodecaphonic 
works, to substitute for unqualified free atonality organized atonality. 
This was to provide a firm basis for the achievement of a genuine atonal 
style, as in his previous compositions it was frequently hard to prevent 
tonal elements from intruding into the music and thus disrupting the 
carefully arranged atonal structure. 

This search for the unification of the entire material of a composition 
by means of an inclusive system of note relationships had been in the air 
during the years shortly before the First World War. There were, for 
instance, Skryabin's attempts to base whole musical structures on what 
he termed ‘synthetic chords’. The Austrian Josef Mathias Hauer (1883- 
1959) anticipated Schoenberg more directly and his theories had some 
influence on the latter's own thinking. As he admitted, *Hauer's theories, 
even where I think them exaggerated, are profound and original, and 
his compositions, even where I regard them as examples rather than as 
compositions, reveal creative gifts'.! In 1908 Hauer began to write music 
based on a system which he later elaborated in several publications? 
demonstrating that the chromatic scale can be arranged in 479,001,600 


! Harmonielehre (third edition, Vienna, 1922), p. 488, 
? Vom Wesen des Musikalischen: Ein Lehrbuch der Zwólftonmusik (Berlin, 1920), and 
Zwoólftontechuik: Die Lehre von den Tropen (Vienna, 1926). 


THE SECOND VIENNESE SCHOOL—ARNOLD SCHOENBERG 343 


permutations; these are grouped into forty-four Tropen which are 
twelve-note patterns rather similar in character to Schoenberg's note- 
rows. Each of these Tropen can again be divided into two sections of 
six notes which stand in complementary intervallic relationship to 
one another; and, in addition, each Trope can be transposed to any 
other degree of the chromatic scale, with further internal regroupings 
of intervals which result in an astronomical number of permutations. 
As a composer Hauer was negligible, and in the event it was not his 
system but Schoenberg's that established itself and proved technically 
and aesthetically of far-reaching significance. 

The dodecaphonic method was primarily a means of achieving order 
and coherence; but it also fulfilled two other purposes with which 
Schoenberg had been concerned almost from the beginning of his career. 
One was to develop a theme by continuous variation, a principle which 
he gradually evolved from hints found in Brahms as, for example, in the 
finale of the Quartet in A minor, Op. 52, no. 2: 


Ex. 152 
Allegro non assai 


Such hints find in the twelve-note technique their consummation; for 
essentially a dodecaphonic composition represents a perpetual series of 
organically linked variations of a basic note-row and its derivations. 
The other purpose was to achieve complete thematic unity, that is, to 
evolve an entire work from a basic idea which is represented by the 
so-called note-row—a series formed by the twelve notes of the chro- 
matic scale arranged in an individual order of sequence and invented 
by the composer. Yet so far as Schoenberg himself was concerned, the 
construction of a note-row was preceded by the invention of an idea 
which already possessed a thematic character; and it was from this idea 
that he worked back to the basic row—in other words, the creative act 
came first and then the play of the intellect on its product. The unity 
of the musical space is ensured by the use of the note-row in two 


344 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


dimensions, vertical and horizontal, which results in an intrinsic identity 
of the harmonic and melodic elements, in which the temporal-spatial 
elements are only two different aspects of the same musical idea. Each 
composition has to be based on a note-row, and to enlarge its scope 
three variants are to be added to the basic series: its inversion or mirror 
form, its cancrizans or retrograde form, and the inversion of the retro- 
grade as, for instance, shown in Schoenberg's Suite, Op. 25: 


Moreover, the note-row and its derivations can be transposed to the 
other degrees of the chromatic scale, with the result that forty-eight 
possible versions of the basic idea are available to the composer— 
always supposing the truly musical ‘identification’ of note-series with 
their ‘opposites’, an intellectual rather than an aural process. These 
transpositions provide one of the principal means of achieving formal 
design, as moving from one transposition to another fulfils a similar 
function to that fulfilled by modulating from one key to another in 
tonal music. 

The twelve-note row combines some of the functions of scale, key, 
and fundamental theme. Like a scale, it contains all the available notes 
of a given note system—in dodecaphonic music, the twelve notes of the 
chromatic scale are arranged not according to a gradually rising and 
falling pitch, but in an order of intervals freshly invented for each 
composition. Thus the row already possesses a melodic character but is 
not yet a melody or a theme because it is not yet rhythmically articulated. 

Two rules in the early application of the note-row were that (a) the 
original order of its intervals must be maintained throughout a com- 
position, and (5) no note must be repeated until the other eleven notes 
are sounded. After the row has been used up melodically or har- 


THE SECOND VIENNESE SCHOOL—ARNOLD SCHOENBERG 345 


monically, or in both ways simultaneously, this procedure is repeated 
until the end of a piece. As the two chief purposes of the row are to 
ensure thematic unity and to replace the function of key in tonal music, 
it follows that it must be present in a composition at all times in either 
basic form (Grundgestalt) or its derivatives. As for the rule of the non- 
repetition of an individual note before the entire row is employed, this 
sprang from Schoenberg's anxiety during his early dodecaphonic period 
lest a repeated note might tend, by thus assuming undue importance, to 
destroy the equality of the twelve chromatic notes and suggest some- 
thing similar to a tonal centre; for the same reason octave doublings of 
a note were forbidden. In the later stage of twelve-note theory these and 
other strict rules were relaxed—note-repetitions were permitted, dis- 
sonances were no longer employed indiscriminately but according to 
their varying degrees of harmonic tension,’ and even consonances were 
readmitted. The note-row as such does not, of course, exclude tonal 
implications—it can be so designed as to include triads and other tonal 
formations, a procedure adopted in particular by Berg to mollify the 
extremely discordant character of pure twelve-note music, and by 
inventing rows with a tonal slant he may be said to have achieved a kind 
of symbiosis with tonal music. Methods of employing the note-row also 
include its sub-division into two or more segments, with groups of six, 
four, or three notes (see Exx. 151, 159, 162, 174, 177, and 179), without 
thereby affecting the all-pervasive thematic function of the entire row. 
For this reason the term ‘functional mode’ is an apt description of its 
true nature which bears a certain affinity to that of the raga in Indian and 
the maqam in Arabic music.” Since the note-row is a horizontal struc- 
ture, the music based on it is primarily contrapuntal in character and the 
ancient devices of inversion, reversion, imitation, and canon play a 
prominent part. Indeed it is in the use of polyphony and variation that a 
composer may show the extent of his inventive imagination and technical 
skill. On the other hand, it has to be admitted that the twelve-note method 
of composition represents an artificial, intellectually conceived system of 
tone relationships which, unlike the empirically derived theoretical 
basis of tonal music, was invented ad hoc. In this it resembles a language 
like Esperanto, without a natural history. The note-row and its manipu- 
lations, fascinating though they are to the intellect, have no basis in the 
observable natural phenomena of music; they are, as it were, test-tube 
creations, laboratory experiments, very close in character to the 


1 Křenek, in his Studies of Counterpoint, distinguishes between dissonances of a higher 
degree of tension or ‘sharp’ dissonances (minor second, major seventh), and dissonances of 
a lower degree of tension or ‘mild’ dissonances (major second, minor seventh). 

2 See Vol. I, рр. 195 and 421. 


346 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


acrostic and other devices which have at different periods engaged 
the attention of poets. Yet what matters in the last analysis is not the 
nature of a system, but whether, by using that system, the composer is 
able to create works of an original and enduring quality, a question 
which has been answered in the affirmative by Schoenberg, Berg, 
Webern, Dallapiccola, and other musicians. Moreover, twelve-note 
music adheres to the same general aesthetic principles as govern the 
rest of Western music irrespective of style and technique—contrast and 
variety in unity, balance and proportion of the parts in relation to the 
whole, effective distribution of climax and anti-climax, and so on. These 
qualities make themselves felt without the listener being consciously 
aware of the particular technical means by which they are obtained. As 
Schoenberg once said, “Ше twelve-note technique is a purely family 
affair’; and he insisted that in the term *twelve-note composition’ the 
accent was to be laid on the noun. 

The great crisis in Schoenberg's creative career occurred during the 
years 1914—18 when as a composer he went underground, as it were, and 
experimented in a number of abortive works with the possibilities of a 
unifying principle based on the employment of the twelve notes of the 
chromatic scale. The first works which showed this principle in embryo 
were the Five Piano Pieces, Op. 23 and the Serenade, Op. 24, for seven 
instruments and solo voice (both 1923).! The fifth of the piano pieces, а 
waltz, and the fourth movement of the Serenade, which is an impas- 
sioned setting of Petrarch's Sonnet no. 217, are both based on twelve- 
note series; but the technique employed was, as Schoenberg himself 
admitted, comparatively rudimentary. Thus, in the vocal movement the 
baritone sings the identical series of twelve notes throughout, without 
any transpositions or derivatives: 


“Тһе first movement of the Serenade is recorded in Тле History of Music in Sound, X. 


THE SECOND VIENNESE SCHOOL—ARNOLD SCHOENBERG 347 


Ex. 154 р 6 
Казсһ , $i. "us 


О kónntichje der Rach’anihr ge-ne-  - sen, diemich durch 


Blick und Re -  degleich zer-stó-ret, und dann 


(O could I ever recover from the revenge on her who destroys me with glance 
and speech alike, and then . . .) 


The only structurally interesting feature lies in the successive distribu- 
tions of the row: as every line of the poem is hendecasyllabic, the 
second line begins with the 12th note of the row, the third line with the 
11th, the fourth line with the 10th and so on, the row gradually doubling 
back on itself until the thirteenth line is reached when the row begins 
again with its first note. The Waltz of Op. 23 is also based on a complete 
note-row while the other pieces show an approximation to it, in that 
each contains a recurring series of several notes—five for instance in the 
third movement: 


Ex. 155 О 
Langsam 


and twelve plus two in the variation movement of Op. 24. Both these 
works, which are in the composer's relaxed and light-hearted vein, are of 
a transitional character in their manipulation of the dodecaphonic 
method. By contrast, in the following Piano Suite, Op. 25 (1924) 
(though some movements were composed in 1921), and in the Wind 
Quintet, Op. 26 (1924), Schoenberg is seen to employ his new method 
with an inflexibility difficult to reconcile with a composer who, despite 
the marked intellectuality of his technique, always placed construction 
at the almost exclusive service of emotional expression. It was as though 
he wanted in these two works to prove to himself the practical validity of 
the twelve-note technique, its manifold structural and contrapuntal 
potentialities. Each of the two compositions is based in its entirety on a 
single row of a special form: that of the Piano Suite is arranged into three 
groups of four notes each, while in the row of the Wind Quintet the last 


348 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


six notes are an almost exact replica, transposed a fourth down, of the 
first six notes, a structure which corresponds to the divisions of a theme 
in tonal music into first (half) and second (full) close. In op. 26 Schoen- 
berg returned to the classical four-movement scheme abandoned during 
his free atonal period, a return made possible through the achievement 
of large-scale structural organization by means of twelve-note technique. 
The Wind Quintet comprises an Allegro in free sonata form, a Scherzo, 
an Adagio in ternary form, and a Rondo, and is scored for flute, oboe, 
clarinet, bassoon, and horn which frequently produce an organ-like 
effect. It is Schoenberg's first large work in the new method and repre- 
sents a compendium of all available contrapuntal devices, especially in 
the finale. 

Schoenberg next applied his new-found technique to the choral 
medium, with special emphasis on the canonic aspect. The two works 
in question, both for mixed chorus, were Four Pieces, Op. 27, and 
Three Satires, Op. 28 (both 1925), in which, with the exception of the 
last two numbers of Op. 27, the settings are of his own texts. The 
Satires were occasioned by the attacks made on Schoenberg's new 
compositional principles in various quarters, and represent his declara- 
tion of war on all current ‘isms’ of the time—neo-classicism, folklorism, 
‘middle of Ше way’ music and ‘wrong note’ tonal music. And to 
demonstrate that it may be as exacting to write good tonal music as 
dodecaphonic music he added three most complicated tonal canons 
which show the same contrapuntal boldness of writing as the serial 
pieces. The next three works were all instrumental. The Suite, op. 29, 
for seven instruments (1926), proves by its gay, unbuttoned mood, 
which reaches a point of sheer exuberance in the final gigue, that 
dodecaphonic music is not concerned exclusively with states of terror 
or anxiety, e.g. the second movement, Yanzschritte (dance steps) 
which is a witty caricature of jazz. As in the Serenade, the unusual 
combination is handled with superb mastery and insight into the 
peculiar character of the individual instruments, with the three strings 
(violin, viola, and cello), the three woodwind (two clarinets and bass 
clarinet) and the piano being treated both in opposition and in concert; 
the keyboard instrument is conceived partly as a concertante, partly 
as a very plastic accompaniment. Schoenberg's own strict rules are here 
relaxed—there are octave doublings, reiteration, and also omission of 
single notes of the basic row, and tonal formations such as major 
and minor thirds and sixths. The slow movement is a set of four 
well-contrasted variations on the song, ‘Annchen von Tharau’, by 
Friedrich Silcher (1789-1860), which is perfectly integrated into the 
serial texture: 


THE SECOND VIENNESE SCHOOL—ARNOLD SCHOENBERG 349 


Theme 


Ex. 156 is та 


Bass Clar. 


"EEls E Y 


Like the Wind Quintet, the Third String Quartet, Op. 30 (1927), is cast 
in a classical mould—a Moderato in free sonata form, an Adagio 
consisting of theme and variations, an Intermezzo in scherzo form, and 
a Rondo. Yet, compared with that earlier work, Schoenberg's formal 
handling here has acquired greater clarity of articulation, the themes 
have an increased plasticity, the contrapuntal lines are more fluid and 
the texture is more airy. And, as in the Suite, the twelve-note technique 
is treated with considerable freedom as, for instance, in the opening 


24 


350 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


movement, where the initial five-note ostinato (G-E-D sharp-A-C) is 
not only continuous for twelve bars but pervades the entire movement 
in various permutations. With the Third String Quartet Schoenberg 
entered his ‘classical’ period. A year later came the Variations for 
Orchestra, Op. 31 (1928), which represents the crystallization of the 
dodecaphonic method in the exploitation of all the horizontal and 
vertical potentialities and is in a sense comparable to Bach's Kunst der 
Fuge. Although orchestral polyphony is carried to its utmost limits, yet, 
owing to the continuous changes in the distribution of the leading parts 
and the scoring, which frequently achieves the effect of infinitely varied 
chamber music, the general texture shows an extraordinary transparence, 
with a gossamer lightness in certain variations (nos. 2, 4, 6, 7, 9). Simi- 
larly, the colour scheme is marked by immense diversity—from the 
hazy, almost impressionist, introduction and the monochrome of the 
theme itself to the subtle woodwind hues of Variation 2, and from 
the fleeting colours of Variation 7 to the compact and massive sound of 
Variation 8. At no point is the orchestral palette used for the sake of 
extraneous effects but always in the service of the musical thought, of 
the individual structure and the general design. The introduction 
gradually reveals the contours of the basic theme and some of its 
transpositions and in its latter part quotes, in the solo trombone, the 
motive B flat-A-C-B natural. This in German spells BACH and is 
intended as a homage to the master of counterpoint. In fact, this 
introduction may be taken as a variation before the actual appearance 
of the variation theme. The latter shows quaternary form built up succes- 
sively from the original row, its retrograde inversion (transposed), its 
retrograde and the inversion (transposed), and is а /ocus classicus to show 
how twelve-note music permits of the invention of a melody which is 
beautifully sustained, of the utmost purity of line and of a perfect 
equipoise in its rise and fall, to say nothing of its tender lyrical expres- 
sion: 


Ex. 157 
Basic Row: 


@) — === 
>- ho > 


7 в зо оир 


THE SECOND VIENNESE SCHOOL—ARNOLD SCHOENBERG 351 


Molto moderato 


Of the numerous ingenuities which characterize the Variations, one 
deserves special mention. While the theme, in its initial statement, 
derives from the successive unfolding of the original row and its three 
derivatives, in the Adagio section, shortly before the close of the work, it 
appears simultaneously in all the four forms of the row: 


Ex. 158 


Cor 
Anglais 


Clar. in 
E flat 


Violin 
(Solo) 


352 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


be — 


в: [| 

SS 

ae ee лашы сш 

ee Иса E 

a PRR ea a 

е Eee ee) 
* (2 


The Variations were Schoenberg's first full-scale orchestral work in 
serial technique, and the comic Von Heute auf Morgen (From Today 
till Tomorrow) (1929) was his first twelve-note opera. It reflects the 
vogue for opera with a topical subject, such as Hindemith's Neues vom 
Tage and Кїепек°ѕ Jonny spielt auf; and the composer also wanted to 
prove that serial technique and a light-hearted vein of expression were 
not incompatible. The rather insipid libretto by Schoenberg's second 
wife Gertrud who concealed her identity behind the pseudonym “Мах 
Blonda', is a social satire illustrating the inconsistencies of modern 
matrimonial assumptions. There are five characters, one of whom, a 
child, is a speaking part. The opera is based on a single note-row that 
yields an infinite number of vocal themes of varying character—lyrical, 
arioso, and dramatic—which are handled with a considerable flexibility 
and lightness of touch, despite the fact that contrapuntal writing 
prevails. Thus, the quarrel duet between husband and wife is a canon, 
and to the final quartet the orchestra contributes two more parts so that 
the texture contains altogether six independent voices. Yet the instru- 
mental style—the opera is scored for an orchestra of normal size but 
includes a large number of percussion instruments—is for the most part 


THE SECOND VIENNESE SCHOOL--ARNOLD SCHOENBERG 353 


restrained and the audibility of the singers' words generally unimpaired. 
Apart from the Suite, Op. 29, the opera is the only workof Schoenberg's 
to make use of jazz elements. 

The two Piano Pieces, Op. 33a and b (1929), show an advanceonthe 
style of Schoenberg's previous dodecaphonic piano music, in that they 
are more clear-cut in design and that his mastery of the new method 
now enabled him to solve similar problems in a more straightforward 
and more assured manner. Thus in the first piece the note-row is first 
stated in three initial chords, of four notes each, which are immediately 
answered by three chords (transposed inverted cancrizans), after which 
it unfolds in horizontal direction (transposed inverted cancrizans in 
treble and retrograde in bass)—an excellent example, within the shortest 
possible space, of the harmonic and melodic use of the twelve-note 
series : 


Ек. 139 


Basic Row: 


The last works Schoenberg wrote before his emigration to America in 
1933 were Begleitungsmusik zu einer Lichtspielszene (Music to a Film 
Scene), Op. 34 (1930), which is an accompaniment to an imaginary film 


354 MUSIC IN THE MAINLAND, OF EUROPE: 1918-1939 


scene describing a mounting tension from imminent danger to fear and 
the final catastrophe; Six Unaccompanied Pieces, Op. 35, for men's 
chorus (1930), which are in a simpler contrapuntal style than his Opp. 
27 and 28, and show a noticeable relaxation of the twelve-note rules, the 
last piece, ‘Verbundenheit’, even employing tonal triads; and the opera 
Moses und Aron (1932) whose completion occupied the composer during 
the last years of his life. 

The first American composition was the Suite for Strings in G (1934), 
a mainly educational work written for the repertory of American 
school orchestras. This accounts for its being wholly tonal, though its 
contrapuntal writing would seem rather complex for a school work. 
Schoenberg's intermittent return to the tonal idiom in this and a few 
later compositions was at the time wrongly interpreted as a concession 
to American taste. Yet, apart from the fact that this assumption is 
amply disproved by other “Атпегісап” works in strict serial technique, 
Schoenberg gave it as his opinion that there was ‘still a lot of good 
music to be written іп C major’ and in an article, ‘On revient toujours’,! 
he confessed that the wish to return to his earlier style remained con- 
stantly with him and that was ‘how and why I sometimes write tonal 
music’. 

The series of important works written during his last period began 
with the Violin Concerto, Op. 36 (1936), which was his first work in 
concerto form and obliged him to consider the, for him, novel problem 
of opposing a single instrument against a full orchestra. ‘Opposing’ 
here is perhaps not the right term, since soloist and orchestra are close 
collaborators, both being equal partners in the elaboration and develop- 
ment of the thematic material. It is, however, a measure of Schoenberg’s 
skill that the work is both a symphonic concerto, in which even the most 
negligible figure is derived from the original row, and a virtuoso piece 
of the most exacting brilliance, the difficulties of the solo part being 
indeed so tremendous that the composer declared in jest that it needed 
a player ‘with a sixth finger'—enormous intervallic jumps, triple and 
quadruple stops, double-stop harmonics, left-hand pizzicato and 
exploitation of extreme registers. Yet, even more than in the violin 
concertos of Beethoven and Brahms, the virtuoso element serves to 
throw the musical ideas into sharpest relief. An important feature of the 
general style is the prevalence of chromatic formations in the solo part, 
heralded in the first statement of the principal theme where the violin 
plays the semitones of the basic row melodically while the orchestra 
presents the rest of the series in chords: 


1 Reprinted in Style and Idea (New York, 1950), p. 211. 


THE SECOND VIENNESE SCHOOL—ARNOLD SCHOENBERG 355 


Ex. 160 | 
Basic Row: 


ши г важат 0771. 8-*9 19 № 12 


(ii) 


The work, corresponding more or less to the classical concerto form, 
consists of a sonata Allegro, a ternary slow movement and march-like 
finale in rondo form, with a clear-cut and easily recognizable structure 
in every one of the movements. The extraordinary formal mastery of 
the work is matched by a rich flow of melodic invention of an intensely 
romantic kind, which reaches in the first two movements a rare degree 
of passionate intensity. 

Compared with the Third String Quartet, the Fourth (written con- 
currently with the Violin Concerto) marks a great advance towards truly 
classical principles: the shape of themes is more clear-cut and sinewy, 
the rhythmic patterns have become more regular, and the texture shows 
an increased equilibrium between homophonic and contrapuntal writing: 


356 MUSIC IN THE MAINLAND, OF EUROPE: 1918-1939 


Ех. 161 
Allegro.molto 
^ 


ist Vin. 


= 
2nd Vin. 
АЗ 


Vla. 


Е сохан 
Cello 5а 
КУ == vien: 


which stands п marked contrast to some works of Schoenberg's 
earlier style, where the impression of a polyphonic surfeit, of contra- 
puntal complexity for its own sake, cannot be resisted. In the Fourth 
Quartet there does not seem a note too many, everything falls into place; 
and, if the Third Quartet frequently revealed a subservience of the 
imagination to the twelve-note method, the later work shows absolute 
freedom in the way the composer bends his technical means entirely to 
expressive purposes. In the period following the Fourth String Quartet 
Schoenberg wrote a number of works, some frankly tonal, while in 
others the underlying note-row is so invented as to imply quasi-tonal 
formations. To the first group belong the Ko/ Nidre, Op. 39, for speaker, 
mixed chorus, and orchestra (1938), written for a Jewish organization 


THE SECOND VIENNESE SCHOOL—ARNOLD SCHOENBERG 357 


and intended for performance on the eve of the Day of Atonement. The 
work is in О minor and makes use of part of the traditional Kol Nidre 
melody, which seems to be of Spanish-Moorish origin. The Chamber 
Symphony, no. 2, Op. 38, which was begun in 1906, at the time of the 
composition of the Chamber Symphony, no. 1, Op. 9, preserves its 
original tonal language—it is in E flat major—but the use of mirror 
forms, especially in the second movement, is to be traced back to the 
influence of the later twelve-note method. Similar in style are the 
Variations on a Recitative for organ, Op. 40, (1941) and Theme and 
Variations for Wind Orchestra, Op. 43 (1943), the first of which is 
written in a greatly extended D minor and the second in G minor, 
though intense chromaticism tends in each case to weaken the gravita- 
tional pull of the tonic. 

In the second group, written in the twelve-note method, a vague and 
intermittent tonal feeling is perceptible—the Ode for Napoleon, Op. 41, 
for speaker, string quartet, and piano, and the Piano Concerto, op. 42 
(both 1942). The Ode is a setting of Byron's poem written after 
Napoleon's abdication at Fontainebleau on 13 April 1814. It is a poem 
against tyranny, a passionate accusation not only of the French 
Emperor but of any kind of political dictatorship, containing also a 
wealth of historical allusions mingled with passages of mordant irony. 
Byron's anticipation of the advent of modern dictators lends his verses 
a burning actuality and it was this that prompted Schoenberg to set 
them to music. The work is marked by an extraordinarily violent 
and explosive quality, largely engendered by the prevalence of broken 
rhythmic patterns and by a percussive piano part which is set against 
short and sharply defined melodic lines on the four strings. The speaking 
part, in contrast to that of Pierrot lunaire of thirty years before, is 
notated on a single line giving only the rhythm at which it is to be 
declaimed. Despite the fact that the work falls into a number of con- 
trasting sections, some of which are in a meditative vein, the dramatic 
breadth and atmosphere are well sustained. The note-row on which 
the work is based permits tonal chords and cadences, and the close 
is in a clear E flat major. Similar in character is the note-row of the 
Piano Concerto, containing oblique references to tonal chords and 
tonalities as, for instance, in the finale where the F sharp major at the 
beginning is opposed by the C major at the end. As in the Violin 
Concerto, Schoenberg skilfully solves the problem of integrating soloist 
and orchestra and for all the virtuoso treatment—arpeggios, wide 
leaps, octaves, and full chords—the piano shares the elaboration of the 
thematic material with the orchestra. To the Piano Concerto the com- 
poser applied the cyclic one-movement form of the Chamber Symphony, 


358 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


no. l, the work consisting of four short and linked sections—an 
Andante which oscillates between a dream-like Viennese Ldndler and а 
siciliano; a vigorous and, often, stormy Allegro molto; an Adagio 
representing the emotional climax of the work, whose ghost-like, 
sinister-sounding effects, after an initial section for orchestra alone, grow 
increasingly agitated and menacing; and a Giocoso in rondo form which 
completely belies the mood of the preceding section by its gay, unbut- 
toned character. Schoenberg, in a note found in the sketches to the 
Concerto, revealed something of its emotional content writing that ‘life 
was so easy (Andante)—suddenly hatred broke out (Allegro molto)— 
a grave situation was created (Adagio)—but life goes on (Копдо). 

The String Trio, Op 45 (1946), was written after an almost fatal 
illness and it reflects the various states of mind the composer ex- 
perienced during this illness, seeming to spring from an acute awareness 
of the no-man's-land dividing life from death; it belongs to his most 
intense and most imaginative compositions opening up a world of vision 
entirely sui generis. Anxiety, agony and existential sadness, prompted by 
the utter solitude of man in extremis, mingle with a retrospect into a 
happy past. The work falls into three main parts and two linking 
episodes, of which the first and last parts convey a dark, oppressive and 
eschatological mood while the middle part is in a serene vein. The tone 
colour of what is essentially a monochrome medium shows a sur- 
prisingly wide range achieved by the use of extreme registers and by 
an exploitation of special effects, such as harmonics, tremolando, 
pizzicato, ви! ponticello, and bowed and struck col legno; in addition, 
there are sudden contrasts of extreme dynamics. 

A Survivor from Warsaw, Op. 46, for speaker, men's chorus, and 
orchestra (1947), the text of which is Schoenberg's own, is based on an 
account of a young Jew who had escaped from the atrocious battle in 
the Warsaw Ghetto in 1944. The composer, who had previously given 
eloquent expression to his Judaism in Die Jakobsleiter, Moses und Aron 
and Ше Kol Мае, here interprets the heroic fight of the Polish Jews 
against their exterminators. The music has an extraordinarily disturbing 
effect, mounting in tension until the climax is reached, in the last 
pages, in the unison chorus singing the ancient Hebrew prayer, Shema 
Ysrael, the only section of the work which is sung and which achieves 
the effect of an emotional catharsis of remarkable power. The speaker, 
whose part is notated, as in the Ode for Napoleon, on a single line 
occupies four-fifths of the work. The style of the work is athematic 
in the sense that, except for an initial four-note motive recurring 
several times and an early anticipation of the theme of the Jewish 
prayer (horn, bars 18-21), the music unfolds in a succession of fresh 


THE SECOND VIENNESE SCHOOL—ARNOLD SCHOENBERG 359 


thematic material of a lapidary character. The fact, however, that these 
various ideas derive from the basic row and its variants, is the measure 
of the extent to which Schoenberg developed his principle of perpetual 
variation. 

Schoenberg's long preoccupation and increasing interest in the spirit 
and the ideals of Judaism and, beyond that, in religious and philo- 
sophical thought in general, began during the First World War, when 
he conceived the unfinished oratorio, Die Jakobsleiter (1915-17). This 
was followed in 1927 by a prose-play, Der Biblische Weg, the action of 
which centres on a young Jew who wants to found a new theocratic 
Israel in Africa. This play is to be taken as a preliminary study for the 
three-act opera, Moses und Aron, in that two chief characters of the 
latter work show traits which the hero of the play combines in his person. 
The first ideas for what was to be Schoenberg's last work occurred to 
him as early as 1923, but it was not until seven years later that he began 
the composition, completing the first two acts (1930-2). Although 
throughout the following period it was his firm intention to finish the 
last act the opera remained incomplete, with a sketch of only eight bars' 
length for the last act. 

The libretto, which was Schoenberg's own and is complete, is largely 
based on the Book of Exodus—the calling of Moses, the revelation on 
Mount Sinai, the erection of the Golden Calf, and Moses's destruction of 
the Tablets, while between the first and second acts fall the intervening 
events of the massacre of the Egyptian armies, the exodus of the 
Israelites, and their wandering in the desert. Schoenberg interpreted 
these events in a manner which transcends their biblical significance 
and reaches far into the sphere of pure spirituality. His fundamental 
theme is the conflict between the characters of Moses and Aaron, a 
conflict already hinted at in the Old Testament: pure thought versus 
image, idea versus its verbal expression, the Kantian thing-in-itself 
versus its appearance, the abstract versus the concrete, silence versus 
communication, the spiritual versus the material, the numinous and 
unimaginable versus the visible and imaginable, reason versus instinct 
and emotion. In the centre of the drama stand Moses (modelled in 
appearance on Michelangelo's statue), who is the mouthpiece of the 
spirit and the law of the one God, and Aaron, the glib exponent and per- 
verter of Moses's God-inspired message, who in his endeavour to bring 
this message down to the understanding of the people resorts to miracles, 
magic and demagogy. When Aaron asks: “Реор!е, chosen by the one 
God, can you have what you cannot imagine?', Moses replies: *No 
image can give you an image of the unimaginable'—an interchange 
which early in the opera already defines the diametrically opposed views 


360 MUSIC IN THE MAINLAND QF EUROPE: 1918-1939 


of the two characters. The second act ends with the apparent victory of 
Aaron who promises the Israelites to lead them into a land of milk and 
honey while Moses has lost confidence in his power to carry out God's 
will, though he continues in his unshaken belief in God as the eternal, 
infinite, and omnipresent noumenon free of all phenomenal attributes. 
In the last act Moses challenges Aaron and accuses him of distorting 
the true spirit of God. When at Moses's command he is set free, Aaron 
falls dead—pure thought has won over the image. The ultimate message 
of the opera is that real life can only be the life of the spirit, and that 
real freedom can only be achieved in seeking a mystical union with 
the Supreme Being. 

Text, action and music form an indissoluble unity achieved by the 
projection of an all-determining and all-embracing idea, and in this 
respect Moses und Aron stands entirely apart from all other modern 
operas. Its character oscillates between scenic oratorio, with static solo 
and choral ensembles, and grand opera of a tense, dramatic kind, 
reaching its apogee in the scene of the dance round the Golden Calf of 
Act II—a barbaric orgy by turns ritual, sacrificial, sexual, and suicidal. 
The people of Israel are given a most active role to play in the drama, 
recalling the turba of the eighteenth-century Passion, and are shown as 
torn between piety and hatred, lethargy and fanaticism, devotion and 
fickleness. From the mass there emerges a group of a few representative 
figures—the Priest (bass), who is shrewdly conservative and clings to the 
old deities, and, opposed to him, three fanatic believers in Moses's one 
God—a girl (soprano) and two young men (tenor and baritone). The 
Supreme Being is symbolized, first, by six solo voices which open the 
opera with four hovering tensionless chords on the syllable ‘О’ that 
stand for the notion of God's eternity and infinity. When the six voices 
begin to intone the text, an orchestral theme occurs representing the 
notion of God as an articulate Will and Thought. These are the two 
central musical ideas of the opera which undergo the most ingenious 
transformation in the course of the work, the melodic and rhythmic 
metamorphoses of the second theme in the dance round the Golden 
Calf being particularly noteworthy. In addition, there is a third, vocal, 
theme of primary importance, which occurs in the early part of Act I 
and stands for God's promise to his chosen people: 


THE SECOND VIENNESE SCHOOL—ARNOLD SCHOENBERG 361 


Ex. 162 à Sehr langsam 


() 
3 female (AG 


voices ` 


3 male 
voices 


Die - ses Volk ist aus - ег - wählt, — 


Nep) 
Dll] трасата) 


(Әс ж 
uum а Ln cui 
4-2 8 ра 


das Volk. 3 deseinz-gen Gotts zu sein. 


(This people is chosen to be the people of the one God.) 


(Ex. 162 (1) opens with the first and last three notes of the original note- 
row (Ex. 163) telescoped into chords; (ii) uses the notes 4 to 9 of the 
row, and (iii) begins with the first six notes of the retrograde, followed 
by the first six notes of the original). It is significant that the role of 
Moses should be spoken (in strict musical rhythm) which is to be taken 
as symbolic, speech serving as the vehicle for pure thought. Aaron, 
however, is a lyrical tenor whose part is often marked by quasi-oriental 
melismata, and the fact that his is a singing character has an equally 
symbolic significance, for song is the expression of the emotional, 
the instinctive and the sensuous—cardinal traits in Aaron's personality. 
The entire opera is based on a single note-row: 


362 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


An interesting feature of the general use of the row is the fact that the 
further Aaron and the Israelites move away from Moses's idea of one 
God, the more tenuous becomes the row in relation to the themes and 
motives. The formal structure of the opera alternates between freely 
treated sections of recitative and arioso and more or less ‘closed’ forms: 
the opening scene, for instance, is a cantata and the dance round the 
Golden Calf a symphony for solo voices, chorus, and orchestra, con- 
sisting of five movements. The texture is predominantly contrapuntal, 
though polyphony is not made an end in itself but serves the dramatic 
action. Imitations, stretti and canon abound: the choral Hymn of Act I 
utilizes a cantus firmus; and the Interlude before Act II is a complex 
double fugue, with the parts sung, whispered, and rhythmically spoken. 
The orchestral forces employed are very large, with triple woodwind 
(four clarinets), piano, celesta, two mandolines, and a wide selection of 
percussion instruments whose most notable use occurs in the orgy of 
Act II. 

Moses und Aron must be accounted as Schoenberg's greatest achieve- 
ment, for the width and breadth of its musical vision are monumental, 
the stark grandeur of its conception is comparable with that of the Old 
Testament itself, and it possesses a spiritual profundity whose full 
import has not yet been compassed.* 


ALBAN BERG AND ANTON WEBERN 

Except for the general expressionist character of their music and the 
adoption of the twelve-note method, Schoenberg's two most eminent 
pupils have little in common. Anton Webern was an ‘intensive’, hermetic 
composer who aimed at the utmost concentration and compression 
of form and at texture of an atomistic nature. Berg, on the other hand, 
was an 'extensive', expansive musician applying himself, like Schoenberg, 
to larger forms and achieving, alone among the three composers, a 
texture їп which the traditional equilibrium between homophony 
and polyphonic writing is on the whole preserved. Webern's stylistic 
development led to the most attenuated lyricism, with the emotive 
impulse subtilized and sublimated almost out of existence, while Berg's 
lyrical vein is ardent, nervous, sensuous, and feminine. Berg looks 
backwards to the romantic past and, in spite of his whole hearted 
adherence to the twelve-note method, he maintained a link with tonal 
music by using note-rows with tonal implications and by availing 
himself, more or less overtly, of the traditional means of sequence, 
repetition and formal correspondence. 


1 For a comprehensive discussion see Karl Н. Wörner, Gottesdienst und Magie (Heidel- 
berg, 1959; English edition, London, 1963). 


ALBAN BERG AND ANTON WEBERN 363 


The period of Berg's creative career with which this chapter is con- 
cerned opened with his first opera Wozzeck (1922). For a full apprecia- 
tion of the reasons that prompted Berg to choose this subject (as also 
that of Lulu) it is necessary to recall that he possessed a very marked 
social conscience. His compassionate humanity made him most sensible 
of the injustices in the political and social systems of Austria under the 
Habsburg monarchy, and filled him with profound sympathy for the 
underprivileged. It was this trait in his moral character that was in the 
first place responsible for selecting the play Woyzeck by George Büchner 
(1813-37) as material for his opera. The play, based on a real incident, 
is a violent attack on German society and German political authority 
before 1848. The fate of the psychopathic soldier Wozzeck, who mur- 
ders his mistress and finally commits suicide, illustrates Büchner's theory 
that the demoralization and dehumanization of the poor derives from 
the cruel injustices of their social environment. The action of the play un- 
folds in a series of twenty-seven self-contained, fragmentary, and loosely 
connected scenes, not unlike a modern ‘screen story’. Out of these 
Berg constructed a three-act libretto consisting of fifteen scenes, five 
scenes to each act, and for reasons of dramatic economy and condensa- 
tion omitted some of the original scenes and telescoped others into a 
single one. The result is a tense, close-knit, and extremely well balanced 
drama, with exposition (Act I), development (Act П) and catastrophe 
(Act III). Each scene forms not only a dramatic unity but also a musical 
unity, Berg employing 'closed' musical forms to achieve inner cohesion. 
Thus the music of the first act, designed to establish Wozzeck's relation- 
ship to the surrounding characters (the grotesque Captain, the crazy 
Doctor who carries out scientific experiments on him, Wozzeck's mis- 
tress Marie and so on) consists of five character pieces, Suite, Rhapsody, 
Military March and Lullaby, Passacaglia, and Rondo. The musical struc- 
ture of the second act is a five-movement symphony, while the third act is 
built in the form of six ‘inventions’—inventions оп a theme, ona note, on 
a rhythm, on a six-note chord, a key and on a perpetuo mobile movement. 
Wozzeck is, for all its novel stylistic and technical features, essentially 
in the line of the Wagnerian-Straussian music drama, exploring and 
elucidating in masterly fashion the psychology of the various characters, 
enhancing the nightmare unreality of the action with consummate in- 
genuity and adding occasional naturalistic touches in the description 
of external phenomena, such as the croaking of the frogs in the pond and 
the ripples of water in the scene of Wozzeck's drowning. The general 
technical style of the music is that of ‘free atonality', characteristic of 
Schoenberg's works of the period between 1908 and 1912, but it is punc- 
tuated by tonal, polytonal and whole-tone scale features, and for the 


364 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


volkstümlich portions such as Andres's hunting song and Marie's 
lullaby, Berg resorts to a symmetrical arrangement of melodic phrases 
and to harmonies of superimposed thirds and fourths which stand in 
marked contrast to the irregularly built periods and dissonant (diminished 
and augmented) intervals of the major part of the music. In the last 
orchestral Interlude (invention on a key), before scene 5, Act III, the 
composer steps out of his role as psychological interpreter and reflects 
the universal significance of Wozzeck's miserable life and death, in 
music which is in an unambiguous D minor. (This piece derives partly 
from a discarded symphony of 1912 which accounts for its strong 
affinity with Mahler's late style.) On the other hand, serial technique is 
foreshadowed in the passacaglia of scene 4, Act I, where a twelve-note 
theme constitutes the basis of twenty-one variations, Berg using the 
form of a passacaglia to symbolize the Doctor's fixed ideas, as indeed 
the choice of all musical forms in this opera was determined by the 
specific character of its individual scenes. Wozzeck employs for the first 
time Schoenberg's device of the Sprechstimme. 

Berg's second and last opera Lulu (1935) owes its origin to the 
same compassion and sympathy for the social outcast as inspired 
Wozzeck; but the action 1s chiefly concerned with the war between the 
sexes and, besides, shows a considerable ambivalence in the psycho- 
logical treatment of the chief character. Berg adapted the libretto from 
the two plays, Erdgeist and Die Büchse der Pandora, by Frank Wedekind 
(1864-1918), which combine erotic satanism with mordant social satire. 
Lulu, the heroine of both plays, is the personification of the untram- 
melled sexual instinct; every type of man is attracted to her, every man 
takes from her what she has to offer but in so doing they meet their own 
destruction. While Lulu destroys, she is herself destroyed, gradually 
sinking into the gutter, becoming a low prostitute and finally the victim 
of a sexual murder. Though Wedekind takes a cynical view of woman's 
sexual role, it is bourgeois society that he makes ultimately responsible 
for its emergence in demonic form. 

There is, however, a strong discrepancy between the Lulu of the text, 
a cold, inhuman femme fatale, and the Lulu characterized in Berg's 
music through which she is made to appear as a creature of flesh and 
blood capable of great suffering. This has been considered a serious 
dramatic flaw of the opera! which is certainly true in terms of the 
libretto and Wedekind's plays. Yet it is precisely by his musical treat- 
ment that Berg turns Lulu into an operatic heroine and engages the 
spectator's sympathy for her. 

In conflating the two plays into a three-act libretto Berg achieved the 


! СГ. Donald Mitchell, “Тһе Character of Lulu,’ Music Review, ху (1954), p. 268. 


ALBAN BERG AND ANTON WEBERN 365 


same compression and unification of the dramatic action as in Wozzeck. 
The attention is most strongly focused on Lulu and on the decisive 
stages of her rise on the social ladder and her gradual downfall to a 
common prostitute, but she hardly undergoes an inner, psychological 
development in the accepted sense. In Wozzeck Berg counteracted the 
loose, episodic sequence of its fifteen scenes by the imposition of self- 
contained instrumental forms; in Lulu, on the other hand, a closer 
dramatic nexus between its seven scenes is reflected іп the more ‘open’ 
on-running character of the music, in which the vocal element pre- 
dominates. The title-role is for coloratura soprano, and the opera 
contains a number of vocal forms such as aria and arioso, canzonetta, 
cavatina, duet, sextet, and hymn; there are also long stretches of sung 
recitative (almost completely absent from Wozzeck) and Sprechstimme. 
Dr. Schoen and his son Alwa, however, are associated with a sonata 
movement and a rondo respectively, their different psychology being 
suggested by the different formal pattern chosen for each. In some 
instances Berg also resorts to contrapuntal devices—a canon between 
Lulu and the painter and chorale variations in the scene with the prince 
(both in Act Т). As in Wozzeck, yet to a larger extent, use is made of 
‘reminiscences’ and leitmotives, including particular harmonies, or- 
chestral combinations, and sonorities which add to the portrayal of 
characters and the changing atmosphere of the drama— chord clusters 
on the piano for the athlete, pentatonic progressions for the Countess 
Geschwitz, a nonet for woodwind for Schigolch, jazz for the theatre 
ambience, and a solo violin playing one of Wedekind's own lute songs 
for Casti Piani, which becomes the theme of the variations leading to 
the final scene of the opera. Moreover, as in Wozzeck, some of the 
music is based on a rhythmic ostinato, as in the monoritmica of the duet 
between Dr. Schoen and the painter (scene 2, Act I), which opens with 
the pattern: 


Ex. 164 


T Wa 


This rhythm dominates the murder of Lulu at the hands of Jack the 
Ripper in the last scene of the opera. 

The scoring is for a large orchestra including a jazz band off-stage, 
complete with saxophone, sousaphone, and banjo. 

By the time he wrote Lulu Berg had completely assimilated the 
dodecaphonic method, yet his handling of it was always highly indivi- 


dual. The work is based on a principal note-row, but by means of most 
25 


366 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


ingenious permutations he derives four other series from it. The chief 
series is: 


By continuous repetition of this series and by selecting from it the 
seventh note (not counting the first note) he arrived at the following 
note-row which provides the theme of Alwa's rondo: 


Ex. 166 


Continuous notation of the original series in which the distance between 
the selected notes increases and decreases at the ratio of, respectively, 
one, two, and three notes, etc. (again not counting the first note), 
yields this note-row from which emerges the energetic theme of Dr. 
Schoen's sonata movement: 


ALBAN BERG AND ANTON WEBERN 367 


ПЕ ра 
Доза Бес] 


[xd 
ара 


У Ne 
Sect, зуль 5 6 7 891011121 2 
1 


Berg left the orchestration of the final act unfinished but he had all 
but completed the music in short score. According to George Perle, 
who had access to the manuscript, there are no more than twenty-two 
bars unfinished in the second scene of Act III, which is both musically 
and dramatically complete, including a full orchestration of the three 
fifths of scene 2 and about the same amount in scene 1. Perle found 
that the sections not complete in the orchestration contain no new 
material but are entirely based on material used in the previous two acts. 
Berg's intention was that the clients of Lulu in the final sceneof theopera 
should be played by the same performers who enact her victims in Acts I 
and II — dramatic symbolism which is reinforced by the fact that the 
music of these corresponding roles is identical, except for notational 
differences. Moreover, in the Lulu Symphony, written to arouse interest 
in the opera and consisting of five numbers—R ondo, Ostinato, Lied of 
Lulu, Variations and Adagio—Berg had incorporated in its last two 
movements extracts from the Interlude and scene 2 of Act III. Perle 
argues with much justification that, despite Schoenberg's refusal in 1936 
to complete the last act (for reasons entirely unconnected with the task in 
hand), the instrumentation of what was left in short score should not 
prove beyond the ability of a musician intimately conversant with Berg's 
late style and his technical methods. 

Berg's Chamber Concerto for piano, violin, and thirteen wind instru- 
ments (1925) was written in celebration of Schoenberg’s fiftieth anniver- 
sary and is the last of Berg's pre-dodecaphonic works. But the way in 


“А Note оп Act III of Lulu’, Perspectives of New Music ii (1964), no. 2, p. 8. See also 
Н. Е. Redlich, Alban Berg (Vienna, 1957), p. 216. 


368 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


which part of the melodic material is derived from an initial motto is 
very close to true twelve-note technique. The three different motives 
of which the motto is formed represent the musical letters in the names 
‘Arnold Schónberg', ‘Anton Webern’ and ‘Alban Berg’ and are the first 
indication that the number ‘three’ and multiples of it will determine a 
variety of formal, rhythmical and instrumental features of the work, in 
conformity with Berg’s predilection for numerical correspondences. 
Thus the Chamber Concerto consists of three linked movements, each 
of which uses a different instrumental design: in the opening Theme and 
Five Variations, the piano is the solo instrument, in the Adagio, the 
violin, and in the Rondo ritmico, both instruments are combined 
with the accompanying wind ensemble. The writing for the two solo 
instruments displays a most brilliant exploitation of their expressive 
and technical possibilities, the virtuoso element reaching the highest 
pitch in the cadenza introduction for piano and violin to the last move- 
ment; while the interplay between soloists and orchestra is immensely 
varied and is ultimately governed by the concertante principle. Another 
noteworthy feature of the work is that the first two movements are 
brought together and amalgamated in the finale in a sonata rondo. 
Though the Concerto is an abstract composition, Berg’s basic incentive 
was a programmatic one of an autobiographical character, namely to 
allow himself to be inspired by ‘friendship, love and a world of human 
and spiritual references'.! 

Berg's first essay in serial technique was his second setting of Theodor 
Storm's poem, Schliesse mir die Augen beide (1925), which he published 
together with his first setting of 1907, to indicate the distance he had 
travelled from a purely tonal style to twelve-note music. In the following 
Lyric Suite (1926) he applied the dodecaphonic method to several of its 
six movements, using in the first movement the same row as in his second 
Storm song, which was a so-called ‘all-interval’ series invented in 1924 
by Berg's pupil, F. H. Klein. The second and fourth movements are in the 
free atonal manner of Wozzeck and the Chamber Concerto. In contrast 
to the Chamber Concerto, the Lyric Suite is loosely constructed and 
avoids all symphonic elaboration. Its rise in dramatic intensity, seen 
in the tempo indications and dynamic markings of the six successive 
movements, was responsible for its being called a "latent opera'. With his 
love for subtle thematic relationships and correspondences between 
movements Berg employs a melodic fragment or motive from one move- 
ment in another. Thus, the second subject of the opening Allegretto 
(bars 23-25) returns in rhythmic and tempo variation in the first rondo 
episode of the following Andante amoroso (bars 16—23): 

! Dedicatory letter to Schoenberg (9 February 1925). 


ALBAN BERG AND ANTON WEBERN 369 


Fi +. 
fa. 
NP ANE 


Similarly, the exposition of the fourth movement (Adagio appassionato) 
is anticipated in the principal theme of the Trio estatico of the preceding 
movement: 


Ex. 169 


Adagio appassionato Wiz ini 
y | | 


The ground plan of the suite is analogous to an open fan: starting with 
the ‘neutral’ first movement (Allegretto gioviale), on one side are the 
third and fifth movements which grow increasingly faster (Allegro mis- 
terioso, Presto delirando), and on the other side are placed the second, 
fourth, and sixth movements which become increasingly slower (Andante 
amoroso, Adagio appassionato, Largo desolato), a procedure that serves 
to heighten the effect of contrast between the successive movements. 
Lyrical and dramatic elements thus alternate, enhancing each other and 
reaching their respective climaxes in the Adagio appassionato and the 
Presto delirando. A possible clue to the programmatic character of the 
work are the quotations, in the fourth movement, of a vocal phrase from 
the Lyrical Symphony (1923) of Alexander von Zemlinsky (to whom it is 
dedicated), and of the Tristan motive in the finale, to say nothing of the 
role played by Berg’s symbolic figure ‘23’ in the bar numbers of the 
work. 

Berg's last completed composition was the Violin Concerto (1935), 
commissioned by the American violinist, Louis Krasner. At the start 
he was uncertain as to its general character and form until the death 
in April 1935 of the eighteen-year-old Manon Gropius, Alma Maria 


370 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Mahler's daughter by her second marriage, when he decided to write it 
as a kind of Requiem for the young girl, dedicating it to the ‘Memory of 
an Angel’. In the event the concerto became Berg's own Requiem, since 
he died in December 1935, aged fifty. On the emotional level it stands 
as an eloquent testimony to the same compassion and humanity that 
inspired Wozzeck and Lulu. The formal organization displays a wealth 
of subtle correspondences between the four movements, and the 
virtuosic element is fully developed, reaching its acme in the third 
movement's accompanied cadenza for the solo violin. The tremendous 
technical difficulties of the cadenza are wholly subservient to the 
expressive ideas and are most ingeniously integrated with the demands 
of a symphonic concerto. The work may be taken as a symphonic 
poem, in the romantic tradition, whose ‘programme’ is the character of 
Manon, her tragic death and ultimate transfiguration, her specific 
symbol being the solo violin which accounts for the tender lyrical feeling 
that informs the major portion of the solo part. 

The Concerto is based on a note-row so designed as to permit the 
employment of the violin's open strings (notes 1, 3, 5, 7). Furthermore, 
this row contains strong tonal implications, gratifying Berg's frequently 
declared wish to use serial technique without the listener being aware of 
it. The last four notes of the row form a whole-tone progression 
identical with the opening of the Bach chorale, Ex. 172(1), introduced in 
the finale. Although this identity was not intentional, it yet constitutes a 
poetic detail of the utmost emotional significance, implying that the 
idea of ultimate deliverance is present in the work from its very 
beginning: 


Chorale motive 


Ex. 170 я j А minor m aa, | 
ВЕ а be Не 


lo #= = 


5 Ре 4 | 
1 2 3 4 5 6 7 8 9 10 11 12 


The work is in two parts each comprising two linked movements of 
which the first two conjure up the image of the girl's character—gentle 
and dreamy in the Andante, full of youthful vivacity and high spirits in 
the Allegretto; the latter movement is a scherzo with two trios in the 
second of which Berg introduces a Carinthian Lündler whose melody is 
perfectly integrated with the serial music. The second part consists of 
an Allegro and Adagio, the former being of a highly dramatic nature 


ALBAN BERG AND ANTON WEBERN 371 


but interspersed with reminiscences from the genial Allegretto, and, as 
in Wozzeck, Lulu, and the Lyric Suite, Berg here resorts to a recurrent 
(dotted) rhythmic pattern, suggestive of Manon's inexorable fate: 


Ex. 171 


PERS E 


The mounting tension of the Allegro culminates in the catastrophe 
and the way in which afterwards the composer negotiates the link to 
the ensuing Adagio is a perfect example of his most subtle art of gradual 
transition: the first nine notes of the tone-row are telescoped into an 
orchestral chord while its three remaining notes are thrown up as a 
melodic figure played by the violin and the orchestra in octaves; this 
pattern is repeated five times with ever-lessening force creating the effect 
of a gradual collapse: on each repetition the orchestral chord is reduced 
by one note while at the same time the violin starts a whole-tone motive 
growing from one note to four, which revealsitself asananticipation of the 
opening phrase of the chorale, “Ев ist genug’, from Bach’s sixtieth church 
cantata, O Ewigkeit, du Donnerwort. This chorale, already boldly har- 
monized by Bach in А major, Berg transposes a semitone up since his 
finale is in the freely handled tonality of B flat: 


Ex. 172 


Es is ge - nug! Herr, wenn es Dir ge - fällt. 
ТУ 


372 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


(11) Es ist ge - nug! Herr, 


Solo Vin. 


Vla. 


Bn. 


(It is enough! Lord, if it please Thee.) 


With the entry of the chorale Berg moves away from the torn, frag- 
mentary texture of the preceding movement to tonally more anchored 
writing—another symbol—and thus achieves a reconciliation between 
the twelve-note method and traditional music. At the opening statement 
the chorale melody is accompanied in the middle part by a partial canon 
(Ех. 17201)), and, іп the bass, by the note-row (transposed to Е 
sharp) whose last note E is sounded by the chorale. The whole passage is 
an example of Berg's close textural thinking. In the orchestral segments of 
the chorale Berg adheres strictly to Bach's original harmonization. There 
follow two variations, both of which use an abbreviated version of the 
chorale; the second variation is punctuated by pointed references to the 
Carinthian Ldndler. The coda, ending like Mahler’s ‘Abschied’ in Das 
Lied von der Erde with the chord of the added sixth, employs the chorale 
melody shortened still further, and concludes with three repeats of the 
sad final ‘Es ist genug!’. 

Anton Webern's art is the complete antithesis of Berg’s, a multum in 
parvo achieved by the utmost concentration of the musical thought and, 


ALBAN BERG AND ANTON WEBERN 373 


springing from it, of projection and technical means. Webern, in fact, 
shows a striking affinity with certain contemporary painters, notably 
with Paul Klee, in the atomistic nature of his style and the severity of his 
vision, and he represents the most radical aspect of Schoenberg's art. 
With him the disintegration of melody, harmony, rhythm, and tone 
colour and the abolition of the difference between the horizontal and 
the vertical are carried as far as was then thought possible. If Webern's 
name is nowadays frequently juxtaposed with that of Debussy, it is 
because Debussy may be said to have started this process of the dis- 
sociation of the constituent elements of music. 

Already Webern's early, pre-dodecaphonic works show features 
characteristic of his mature style—an extremely sparse texture, a 
pointillistic technique, the employment of silence as a principle of 
musical construction, a marked predilection for canonic writing (the 
most rigorous of the contrapuntal disciplines), and a prevalence of low 
dynamics which at one time earned Webern the description ‘the com- 
poser of the pianissimo espressivo'. If, in addition, these early composi- 
tions were characterized by the utmost brevity of form and laconic 
utterance, this was largely due to the fact that they were in the “тее 
atonal’ style, with its lack of any means of large-scale structure. With 
the adoption of Schoenberg's twelve-note method, which provided 
this means, Webern was enabled to tackle more extensive forms, such 
as sonata and variation; but the tendency to extreme compression 
persisted, the thematic material remained aphoristic and the fabric of 
his music tenuous in the extreme. In fact, Webern's essential manner 
scarcely changed after adopting dodecaphonic technique, which was 
to him a new tool for achieving structural organization and coherence 
but affected the character of his music very little. 

Webern first used serial technique in three vocal compositions— 
Drei Geistliche Volkslieder, Op. 17, for voice, clarinet, bass clarinet, and 
violin (1924), Drei Lieder, Op. 18, for voice, E flat clarinet, and guitar 
(1925), and Zwei Lieder, Op. 19, for mixed chorus and five solo instru- 
ments (1926). If the employment of the serial method is here rather 
simple and even tentative, the texture on the other hand displays great 
contrapuntal complexity, voice and instruments forming an extremely 
close-knit ensemble. What is entirely novel, however, is the vocal treat- 
ment, Webern translating the phonetic tension of the words into musical, 
intervallic, tension; in other words, the mere verbal sound is trans- 
formed into an element of musical structure, a device which was to be 
raised to a principle in certain works of Boulez, Stockhausen, and 
Nono. See for instance these two examples, the first from Drei Lieder, 
Op. 18, no. 1, the second from Zwei Lieder, Op. 19, no. 1: 


374 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Ex. 173 
(i) 


Sehr ruhig 
Dn 


есш ада сисе e Caere 


Ncc I Рт ди се ipm m E 
91; 


Grüntder Ros - ma-rin, .  grübt der Myr ten ыс und der 


Na - gerl-stock blüht im Haus 


(The rosemary flourishes, the myrtle flourishes, and the carnations flower in the 
house.) 


GÙ) Lebhaft, leicht und frei oo 


sree eae ae `” 


So früh - zei-ti-ge Nar-zis - sen ит hen 


при | = Bs 
t E 7 
= W - 
р па Без T 1 
So früh - zei- ti-ge Nar - 
LA гул 8. 
Г | Ж ] L4 , | 
WEE pie perm Е-Е 46-12 
früh - - zei-ti - ge Nar - zis - sen Ый - hen 
г 3 1 
I Pa = 5 К Е2 
ЕЕЕ аа - е НЕ, 
E h [ ; -2 
rei - hen-weis im Gar - ten 


Wa lm A ae a Б. 
- ü-ge Nar -  zis - sen blü-hen rei - hen - 
Г owed | 3 
Са ВЛ а жетелі qe aa 
Se с=ш= шз = =. 


- zis-sen blü-hen rei -hen- weis im Саг - ten. 


ALBAN BERG AND ANTON WEBERN 375 


(So the early narcissi bloom in rows in the garden.) 


The String Trio, op. 20 (1927), was Webern's first instrumental work in 
serial technique and marks a departure from his previous free formal 
writing in that it applies strict classical forms—rondo in the first move- 
ment and sonata in the second of its two movements; the latter even 
contains a double-bar and a repeat of the exposition, though the 
recapitulation 1s extremely well camouflaged. In the Trio it is no longer 
possible to discern with any clarity the individual movement of the 
three parts; the ear perceives, instead, a fluidum in which the musical 
ideas run freely from one voice to the other, grouping and regrouping, 
merging and disintegrating, an effect which is enhanced by the wide 
crossings of the parts, by the alternation of arco and pizzicato and by 
the introduction of numerous rests apparently interrupting the flow of 
the melodic line but actually used as part of its rhythmic structure 
and as important as the tangible, audible sound itself. The Trio marks a 
point of departure in Webern's style. In the Symphony, Op. 21, for nine 
solo instruments (clarinet, bass clarinet, two horns, harp, two violins, 
viola, and cello) (1928) the composer is seen to advance further in the 
direction of what may be defined as the use of structure per se, as a 
Kantian thing-in-itself, no longer serving conscious emotional expression 
but generating 'feeling' as an incidental by-product; expression thus be- 
comes a function of construction. Moreover, minute atom-like motives 
in a seemingly kaleidoscopic arrangement take the place of extended 
thematic configurations. This method, novel in both technical and 
aesthetic respects, demonstrated the distance which increasingly separated 
Webern from Schoenberg and Berg for both of whom the espressivo 
remained a cardinal principle. Thus Webern unwittingly paved the way 


376 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


for the 'geometrical' construction and electronic music of the next 
generation of composers. 

The note-row on which the Symphony is based shows a special feature: 
like the series used by Berg for the opening movement of his Lyric Suite, 
its second half represents the transposed cancrizans of the first half so 
that the original row is identical with its retrograde: 


Such structuring of the row by means of contrapuntal devices (see also 
Exs. 177 and 179) was highly characteristic of Webern's serial thinking 
which was in marked contrast to Berg's whose rows are more ‘normal’, 
less close-knit and who was also intent on evoking quasi-tonal associa- 
tions through their intervallic disposition. There is, moreover, a pro- 
nounced difference between the two composers in the way they exploit 
their series. While the ‘expansive’ Berg sought to enlarge the thematic 
possibilities of a row by deriving from it new rows, as in several 
movements of the Lyric Suite and in Lulu, the ‘thrifty’ Webern kept 
strictly to the same row and even reduced its yield as a thematic reservoir 
by constructing some of his series in segments which represent inversions 
and/or retrogrades and/or transpositions of an initial motive of six, four, 
or three notes. On the other hand, this contrapuntal compression results 
in storing up kinetic energy in the row, an energy which is then dis- 
charged in the composition. In Webern’s hand the series takes on a 
dynamic quality. It is, incidentally, probable that his marked preoccupa- 
tion with the ‘automatic’ devices of counterpoint, notably canon, was 
nurtured by his intimate knowledge of Renaissance music and his 
editorial work on Heinrich Isaac. Indeed, from Op. 21 onwards (if not 
already from Op. 15) the technical models of vocal polyphony of the 
fifteenth and sixteenth centuries begin to emerge in Webern's output 
with increasing clarity leading to the masterly four-part chorale of his 
Second Cantata, Op. 31.! 

In constructing for the Symphony a series (Ex. 174) which has no 
independent retrograde, Webern deprived himself of half of its thematic 


1 See Walter Kolneder, Anton Webern. Einführung in Werk und Stil (Rodenkirchen, 1961; 
English edition, 1968), p. 160; Eng. ed., p. 177. 


ALBAN BERG AND ANTON WEBERN 37 


ч 


possibilities; it was precisely for this reason—'the somewhat mathemati- 
cal form of this series'—that Berg, in the Lyric Suite, altered the original 
row for the third, fifth, and sixth movements. But Webern's deliberate 
reduction of the thematic potential of his series is compensated for 


of the first movement: 


Ex. 175 


т. 
Bass Clar. LI. Е 


== e 
и а Ср te Е Red 9 
к ЕЕ 
- 
z^. 7- 


| 


Нар CA 


Ist and 2nd Yon 


ie eee eed 
=~ 

Violins Неа =. =. 
Д Ел арланы нала | БЕН а ы иШ Е ил фиг РТА _ Е кар с шына um E] 
Та далалы ыы = лы отша петлета 

ЕШТЕНЕ АЕ x есы ee 
ет EE |--- == ————— 
| pom um eee eee, — | = m = 1] 

Vlc. = je д [5 mz (берей! ШЕ "БЕ [pa 1 

[mpm A Ael eR poe D UC Uem eae Sa RE el r hee aia РНН 


378 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


mag gi 
ри , 
Ey. = ЕЕ 


Г] 
| 


tional build—clear-cut and self-contained, and consisting of a proper 
melody and an accompaniment. The seven variations and coda are all 
of the same length (11 bars) but different in texture, rhythmic pattern 
and orchestration. The fourth variation (mirror canon) represents the 
centre piece of the movement after which the remaining variations run 
backwards ‘зо that the whole movement is itself a double canon by retro- 
grade тойоп” (Webern). 

Like the Symphony, the Quartet, Op. 22, for violin, clarinet, tenor 
saxophone, and piano (1930) is in two movements whose contrast 
lies partly in their respective speeds and metres— Sehr mássig' in 3/8 and 
‘Sehr schwungvoll’ in (mainly) 1/2 time—and partly in their texture. It 
is the only work of Webern’s to show, in its particular combination and 
the syncopated rhythm of the first movement, an influence of jazz. This 


ALBAN BERG AND ANTON WEBERN 379 


contrasts with Berg who employed jazz elements more extensively in the 
concert aria Der Wein (1929) and in parts of Lulu. There is scarcely any 
suggestion in it of a preferential treatment of one or the other instrument 
— оп the contrary, they are all handled alike and their individual timbre 
is merely exploited to throw a given part into relief. Thesecond movement, 
more loosely woven than the first, provides a particularly good example 
of Webern's predilection for microscopic motives of one, two and three 
notes. 

The Drei Gesänge, Ор. 23 (1934), and Drei Lieder, Op. 25 (1935), for 
voice and piano are settings of Webern's favourite contemporary Austrian 
poet, Hildegard Jone, and show that the vocal medium made the com- 
poser to some extent relax the extreme application of his pointillist 
method in favour of a more sustained, more markedly lyrical writing, 
though the wide leaps and occasional rests in the voice part point to the 
characteristic fingerprints of his instrumental style: 


es fühlt ihn an dem dunk-len Wur -zel- reich, das an die To-ten rührt. 


([The heart] feels [the spring] in the realm of dark roots which touches the dead.) 


The two song cycles may be said to represent preliminary studies for the 
Cantatas Opp. 29 and 31. 

The Concerto, Op. 24 (1934), for flute, oboe, clarinet, horn, trumpet, 
trombone, violin, viola, and piano, written in celebration of Schoen- 
berg's sixtieth birthday, was originally to have been a solo concerto 
for piano and eight instruments. A trace of this initial scheme is to be 
seen in the continuity of the piano part which is the cement holding 
the other parts together. The series on which the work is based, shows a 
still closer inner relationship than that of Op. 21. It is built of athree-note 
motive, its retrograde inversion, its retrograde and its inversion which is 
like a blueprint for the structure of the Concerto as a whole: 


EST 
о RI R I 
pv pem e р И о [room тті 


380 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


A comparison with the theme of Schoenberg’s Variations for Orchestra, 
Op. 31 (Ex. 157(ii) ), in which each of the four serial forms is stated in 
full (4 x 12 notes) shows the extent to which Webern carried the con- 
centration of his serial thought. Writing about this row to Hildegarde 
Јопе,! he compared it with the Latin tag: 


SATOR 
AREPO 
TENET 
OPERA 
ROTAS 


in which the third word is a palindrome and all the five words, read back- 
wards and up and down, form the sentence 'Sower Arepo controls the 
work’ and its inversion “Тһе work controls sower Arepo’. Just as Ex. 177 
constitutes a micro-variation on the initial three-note motive, so does the 
whole Concerto represent a macro-variation in three movements of that 
motive. It is the most complete realization of Schoenberg's concept of 
total unity. All musical happenings spring from a single germ-cell 
producing an organism which Webern was fond of likening to Goethe's 
Urpflanze or primeval plant—‘the root is . . . no different from the stalk, 
the stalk no different from the leaf, the leaf no different from the flower: 
variations of the same 14еа’.? The work is written in concertante style 
which is the corollary of the use of nine solo instruments. The music is 
thin and clear, extremely supple and sensitive. It was, significantly, 
from the Concerto that the concept of total serialization, i.e. the serial 
predetermination of all musical parameters, arose in the Germany of the 
early 1950s. 

The Piano Variations, Op. 27 (1936), is Webern’s only work for piano 
solo. A paramount feature of it lies in the fact that the composer, in 
contrast to the variation movement of his Symphony, here—not unlike 
Stravinsky in the variations of his Concerto for Two Pianos—dispenses 
with a theme so called and that the variations are not demarcated, but 
merge into one another. Thus the first movement opens with what is 
already a variation, namely a statement in canon of the basic row (right 
hand) and its retrograde (left hand): 


111 March 1931. See Webern, Briefe an Hildegard Jone und Josef Humplik, ed. Josef 
Polnauer (Vienna, 1959; Eng. ed., Bryn Mawr, 1967), p. 17. 

2 idem, Wege zur neuen Musik, ed. Willi Reich (Vienna, 1960; Eng. ed., Bryn Mawr, 
1963), p. 56. 


ALBAN BERG AND ANTON WEBERN 381 


The second movement is similarly built and both movements recall, in 
their pianistic lay-out, a Brahmsian intermezzo. In the third movement 
of this ‘Sonatina’ (as Leibowitz calls it) Webern abandons contrapuntal 
texture in favour of a markedly homophonic writing and comes nearest 
to composing a ‘theme and variations’ in that he opens with the retro- 
grade—the ‘theme’—after which are deployed as ‘variations’ the four 
serial forms in various alternations and transpositions. Like the second 
movement of the Symphony, Op. 27 does not constitute thematic 
variations (in the sense practised by Beethoven and Brahms) but contra- 
puntal variations in which the series and its derivatives are unrolled in 
their totality either in combination (first two movements) or in succes- 
sion (third movement). This treatment of the row as the sole generator of 
musical sense has been criticized as ‘fetishism of the series'.! But this 
leaves out of account such aspects as figuration, dynamics, register, and 
accentuation which contribute to the musical sense—though in an 
admittedly lesser degree than structure—and which are subjected to 
‘real’ variation in Op. 27. 

The String Quartet, Op. 29 (1938), shows Webern’s serial invention at 
its most concentrated. The basic row here consists of three four-note 
motives the first of which is formed of the musical letters in the name 
‘Bach’, the second is its inversion or transposed retrograde and the third 
its transposition. In addition, the second half of the series represents the 
retrograde inversion of the first half: 


1 Theodor W. Adorno, Philosophie der neuen Musik (Tübingen, 1949), р. 74. 
26 


382 MUSIC IN THE MAINLAND- OF EUROPE: 1918-1939 
Ex. 179 


о I-RT oT 
p——————4À [rl 
НЕ та 


Mo —pe— uv eue aec pium p 
Куз Se Gl See OF © Ис түт y 4€ 
far oO раз среде mm ИЖ 
| ETE) 


Ex. 179 is the ne plus ultra in close motivic relationships, relationships 
which in idea are an identity, and it provides another illustration of the 
resemblance of a typical Webern row to the palindrome and inversion in 
the word play of the composer’s Latin tag. As in virtually every work of 
his last period, strict canonic treatment prevails though there is some 
relaxation of it in the third movement. Webern’s last instrumental com- 
position was the Variations for orchestra, Op. 30 (1935), though its 
particular combination—flute, oboe, clarinet, bass clarinet, horn, trum- 
pet, trombone, bass tuba, drum, celesta, harp, and strings—it, with the 
exception of the strings, used in a soloistic way and aspires to she nature 
of chamber music; it excels in the most delicate hues and ranges over 
a remarkably wide variety of tone colours. In Op. 30 the composer 
returns to the more traditional form of theme and variations such as he 
had used in the second movement of the Symphony. The theme is first 
alluded to in the introduction or ‘Overture’ and is stated in full in the 
first of the six variations which are clearly set off from one another by 
double-bars. The work shows an ingenious fusion of variation with 
sonata form. 

Between the Drei Lieder, Op. 25, and the Piano Variations, Op. 27, 
Webern wrote Das Augenlicht, Op. 26, for mixed chorus and orchestra 
(1935), and he continued with his interest in the choral medium in the 
Cantata no. 1, Op. 29, for soprano, mixed chorus, and orchestra (1940), 
and the Cantata no. 2, Op. 31, for soprano and bass solo, mixed chorus, 
and orchestra (1943), the latter being his last and largest work which he 
compared to a Missa brevis. The texts of all three compositions are by 
Hildegard Jone and are mystical in character. As in the song cycles, 
Opp. 23 and 25, the vocal medium prompted him to apply his frag- 
mentary style of writing with markedly less consistency and rigidity, 
treating the voices in a more sustained manner and aiming at a warmer, 
more lyrical mode of expression. Similarly, he returned again to a more 
or less clear distinction between horizontal and vertical thinking, as is 
shown in these examples from Das Augenlicht: 


Ex. 180 


әні 


ALBAN BERG AND ANTON WEBERN 383 
(i) Langsam 
= 
[em en 2 ЕН | ee ысыла се ете ИШИНИ LU [ux m] 
= | 4 Im met Ez La | 
po UM EE 1 ч 
mit so - = ме Ster- nen als die 
4 
y O Dun ИЕ ma] 
B. d. Мне тына) [LC ta 
Е МЕн == рар 
D 0 
mit 50 - viel Ster - nen Яв. 
() 5 ің 
Rhee | [mens AGREE | — wie c Too mau mnt 
фр LL 
ЕЕ a a 
> mit 30 = viel Ster- nen 
ша е 


H E саса 
p——qeem— ____ р 
оа sr] [DH es ee ae” |. | 


die Nacht er – hel- len. 


Ster-nen als die Nacht ег - 


(With as many stars as lighten the night.) 


ae са Е И 
Ма ofS санниши 


pum 
= 
[amc x] 
=a 
2a 
[er 


O Meer des Blik - kes mit der 


384 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


Trà - nen - bran-dung! 


(O sea of vision, with the surf of tears!) 


While Op. 26 is in one continuous movement, the Cantata no. 1 consists 
of three the second of which is for soprano solo. Its basic series is con- 
structed like that of the String Quartet, with the same closeness of inner 
motivic relationships. The first entry of the chorus in the opening move- 
ment unfolds the four serial forms in a chordal texture, as indeed the 
whole of this setting is of a homophonic character. To counterbalance 
this, Webern conceived the final movement as a polyphonic piece, 
a four-part double fugue in which subject and countersubject are related 
to each other ‘like antecedent and consequent in a period’. There is a 
wonderful flow and fluidity in the vocal counterpoint which avoids the 
creation of any recognizable centre of gravity: 


p e smart ERE == 
(ЕЕ GER ee [EBEN L {| Гў 
LETTER cmd y ма (——Hn 
р (ae ЧЕТ a 
blas - se - ren Bil -der zum 


die blas-se-ren Bil-der 
— ——ÓÀÀ син 
ӨТ ЖЕЛІГІ. —H p 


Eu -—— ДИН. D^ cM A Rej 

[1| eg ит Цифрово ре салата фа и е 
PIDEN mo EE ее 
„Ана са а] 08 шкын пе Б ae) eee ae 


auch die blas-se-ren Bil - 


ALBAN BERG AND ANTON WEBERN 385 


— p cresc. 7 


> =: SaaS SRM rv Cha ks 4 1a Sa (7 A ЕЕ 
ЕТ БЕЙШЕ Лос ceci] Ee WIS 387] EE lcm 
Hf? тыңнан ісі c IIS PLE III Iph 
E VL аа с=с гради 


m —————Ó poo nnd 


Sie- gel des Spek-trums ge - schmol - zen. 
f) >” Шел nud 
IE FYE O —— — 1 E Ae ЗЕЕ ИЕ ПЕ E qoae] 
ОЕ ЕТ = ра 
zum Sie-gel des Spek-trums бе - schmol - еп. 
| D cresc. F 
ШЕР JEn i ep ess 
HO-H—8-H—— зезе cree quoe — vu 0 —1— 
zum Sie - gel des Spek-trums ge - schmol - zen. 
р cres. етті DM. 1 
63: ВЕ 9----39-- Ир м c — Tey Da pee Метал == 
- der zum Sie-gel des ЕЕ ge-schmol - теп. 


(and also the fainter image is melted as seal of the spectrum.) 


In this fugue, as also in the finale of Op. 31, the voice parts are—an 
almost unique occurrence in Webern—doubled on the wind instruments. 
Op. 31 is in six markedly contrasted movements, part of this contrast 
lying in the choice of different vocal combinations for the individual 
movements. The music suggests that Webern was entering upon a phase 
of new simplicity. This is seen in the first place in the basic series which 
15 a straightforward one, without any contrapuntal devices in its struc- 
ture. Furthermore, the second, fourth, and sixth movements show a 
markedly lesser degree of textural density and the interplay between 
voices and instruments is almost completely free of complexities. Indeed, 
in the last movement there is no interplay at all, Webern doubling the 
four choral parts on the wind. The vocal score gives the impression 
that the chorus is unaccompanied, with the instruments cued in for 
use in rehearsal only. With its long note-values, though the tempo is 
very fluid, and the ‘mensural’ changes of the metre (2/2, 3/2 and 4/2), 
this movement recalls music by Heinrich Isaac whose Choralis Con- 
stantinus had been the subject of Webern's doctoral thesis in 1906. To 
all intents and purposes the finale of this Cantata is a chorale in the 
form of a four-part canon whose three verses are to be sung to the same 
‘tune’. Who could have foreseen that Webern would ultimately arrive 
at a strophic setting of a text? 

What is the historical significance of the serial Webern? It is in the first 
place to be found in the almost total replacement of the romantic idea of 
music as a vehicle for emotional expression (Schoenberg and Berg) by 


386 MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


the idea of music as pure sound construction, seen in his preoccupation 
with the contrapuntal structuring of a row and with the constructive 
quality of a single note and single interval! which are subjected to per- 
реша! variations in intensity, dynamics, accentuation and tone colour. 
With Webern the dissociation of melody, harmony, rhythm, and timbre 
is carried to its utmost limits, the corollary of which was the extreme 
fragmentation of texture. Moreover, Schoenberg's concept of the 
unity of the musical space finds in its most complete realization in 
Webern's abolition of the distinction between horizontal and vertical 
thinking, for his contrapuntal writing can also be interpreted as the linear 
projection of harmonies built from the series. Lastly, Webern reduced 
the means by which he exteriorized his inner vision to their essential 
minimum and thus achieved the highest possible degree of technical 
economy. In short, he drew the most radical, most uncompromising 
conclusions from Schoenberg's twelve-note method of composition and 
in so doing opened up a new vista for the succeeding generation of musi- 
cians who saw it in the serial predetermination of the totality of musical 
components, including pitch sequence, rhythm, duration, intensity, and 
method of playing. The real father-figure of the New Music of the post- 
Second-War period is not Schoenberg but Webern. 


V 


КИ по кол, EUROPEAN 
MAINSTREAM: 1940-1960 


By PETER EVANS 


IN 1933 Anton Webern described the preceding quarter of a century's 
musical developments as “ап advance greater than has ever taken place 
before in the history of music'.! He was reviewing the achievement of 
twelve-note serial practice as the opening up of territory sought ever 
since music had been written, and he looked forward with confidence 
to the further exploration of this new world of musical order. So 
intrepid a reading of history demanded a visionary reading of the future, 
since historical significances are commonly assessed by what in fact 
achieves generally recognized validity; and the signs of Webern's 
times provided few obvious clues to a less prejudiced observer. Over 
three decades later we need less courage to point to significances in that 
period (and nowhere more certainly than in Webern's work) for these 
emerge in investigating our own, but of this itself little more than a 
record can yet be attempted. A chapter which deals with what is still 
present cannot indulge so freely the historian's privilege of viewing in 
perspective the store of knowledge: its concern must simply be with what 
has occupied more than parochial attention. 

A simple chronological subdivision of the period 1940-60 would 
underline all the radical innovations in musical technique which 
distinguish its second decade; but this would tend unduly to obscure the 
continued activity of a generation already formed in earlier methods. 
A broad division may therefore be made between composers who, 
already mature before the war, lived and worked throughout this 
period or who achieved belated prominence during it (thus the last 
works of pioneers like Schoenberg, Webern, and Bartók have been 
discussed in a previous chapter, while Skalkottas and Valen appear 
here), and those of the generation which completed its apprenticeship 
after the war. 

Of the century's earlier pioneers, Stravinsky and Hindemith demand 
further consideration, the former summing up in his own development 


! Der Weg zur neuen Musik, ed. Willi Reich (Vienna, 1960), p. 17. 


388 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


the momentous swing in the orientation of European music which may 
yet justify Webern's confidence. While note must also be taken of later 
works (for example, of Hindemith) which do not reflect a comparable 
change in a composer's practice, one phenomenon may be found to 
outweigh all others in a general survey of the middle generation—the 
emergence of serial method as no mere shibboleth, but as an attitude to 
musical material that might stimulate atrophied thought processes 
without taking toll of individuality. Political boycotting in Germany, 
and blankly uncomprehending or even virulently hostile notice else- 
where, had succeeded in bringing both the music and the ideas of 
Schoenberg and his pupils into almost general disrepute. Even when 
their music began to find more admirers, especially among composers, 
able to form judgements from the scores when performances were still 
rare, a variety of new critical positions was taken up: the works must 
succeed in spite of the underlying theories, or the theories could have 
no relevance to composers not reared in Schoenberg's Viennese en- 
vironment. Yet the end of the war brought a new wave of interest in 
serialism that affected in varying degree composers of many countries 
and already versed in other styles. In Germany and Austria hostility to 
twelve-note music was identified with Nazi sympathies and this political 
circumstance gave the movement a powerful new impulse, seen most 
clearly in the policy of German radio stations. 

These other styles, many of them related by some common factor of 
neo-classicism, were briefly sampled by a generation that had never 
known the allure they had once presented. Impatient of traditions that 
themselves involved historical counterfeiting in varying degrees, this 
new generation sought no more guidance from the past than would lead 
it towards an independent future The new serialism was founded on 
the one hand in Schoenberg's principle of regulation, as applied by 
Webern in his break with conventional conceptions of musical space 
and motion; and on the other in Messiaen's introduction to the 
European art of an exotic conception of musical time. 

In 1940 the possibility of such a revival of serialism seemed remote; 
indeed it was hard to conceive of any artistic movement that could unite 
the talent of a continent in a common purpose. Few of the pre-eminent 
creative figures were able to remain in Europe, and with their emigration 
the focus of attention moved to America. It must be left to another 
chapter to chart the influences brought to bear on American music by 
the influx of so many of Europe's dominating composers; of reciprocal 
influences there are few signs. (Bartók's simplification of idiom had 
begun before he left Europe, though the diatonic limpidity of the 
Third Piano Concerto's slow movement may owe something to what 


IGOR STRAVINSKY AND THE RAPPROCHEMENT 389 


Stravinsky has called the ‘Appalachian’ style, and another Eastern 
European, Martinu, reflects this more directly in some of his sym- 
phonies.) Whether composers died in America, like Schoenberg and 
Bartók; chose to settle there, like Stravinsky and, much earlier, Varèse; 
or subsequently returned to Europe like Hindemith, their work con- 
tributes essentially to a unique European heritage. 


IGOR STRAVINSKY AND THE RAPPROCHEMENT 


It is clear enough that Stravinsky's changed environment hardly 
affected his idiosyncratic stylistic progress; indeed we may be tempted to 
think that his penetrating assessment of the European avant-garde 
and his characteristic opportunism with regard to its findings were 
helped by the perspective of distance. America did however affect his 
choice of media, as in his settings of English and his use of the jazz band 
(Ebony Concerto, 1945). And it is to the American reverence for the full 
orchestral symphony as the centre-piece of musical activity that we owe 
Stravinsky's two essays in this form.! Together with the Two-Piano 
Sonata (1943-4) and the Concerto in D for string orchestra (1946) they 
are stages in a last review of traditional form and texture that culminated 
in The Rake’s Progress. 

The tag *neo-classical is unavoidably prominent in any documenta- 
tion of Stravinsky's career, yet beyond suggesting his dependence on 
initial stimulus from other music, it brings us little nearer an under- 
standing of a crucial aspect of his activity. His symphonies demand not 
only the background of specific tradition, but a general historical 
awareness as extensive as his own. In both first movements the combina- 
tion of simple repeated-quaver accompaniments and a persuasive 
motivic technique strikes a new balance between static and dynamic 
elements, and the use of classical progression divorced from classical 
timing produces an entirely personal articulation. Nominally recogniz- 
ing the progressive schemes of sonata practice, the Symphony in C 
(1940), with brilliant irony, recaptures the rooted quality that charac- 
terizes so much Stravinsky. Its second group turns to the dominant, but 
adds a further section in the subdominant; the moral of this contradic- 
tion is rightly drawn by a move back to the tonic that renders the 
development merely ornamental. Stravinsky observes the traditional 
transposition of second group material on its (reversed) restatement, 
yet this in fact reinforces even more strongly the tonic. 


Exposition —B1 on dominant of V; B2 in IV (to I) 
Restatement—B2 in I (to V); ВІ on dominant of I 


1 See Stravinsky's comments on the American commissioned symphony in Stravinsky 
and Craft, Memories and Commentaries (London, 1960), pp. 92-93. 


390 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


This simultaneous acknowledgement and denial of classical precedent is 
stressed by hints of tonal instability in the first subject. By constant use 
of the non-committal third E-G in ambiguous harmony (avoidance of 
the root and use of the seventh B), Stravinsky is able to suggest E 
minor as an alternative meaning to the professed C major, even in the 
final bars of the movement, and it is on this delicate point that the 
apparently garrulous subject is able to hover for so long without 
tedium. 


Ex. 182 
(Moderato alla breve) 


S 
прш m "idm — 


l 


(Fls. tacent) 


IGOR STRAVINSKY AND THE RAPPROCHEMENT 391 


The Symphony in Three Movements (1945) abandons this feline subtlety 
for harmony locked so tightly in contradiction as to provoke violent 
rhythmic convulsion; here the tonal plan of classical form no longer has 
even ironic relevance, yet the functions of development and truncated 
restatement are made far more vital. The material is powerful and 
varied, but only in the opening of the Andante is there that vein of 
affectionate reminiscence which сап be traced throughout The Rake's 
Progress (1948-51). Whereas the symphony seeks to pursue the conse- 
quences of its pandiatonic language against a background of classical 
procedure, the opera places a series of classical statements in the fore- 
ground and deftly transfers them into the newer idiom. Thus, it might 
seem a more frivolous undertaking or, in the light of Stravinsky’s 
subsequent work, a mere valedictory divertissement. Even though such 
a work would perhaps be a fitting monument to a public which has 
avidly cultivated a narrow tract of music’s past to the detriment of the 
present, the number of specific models it brings to mind (with Don 
Giovanni prominent in Auden and Kallmann’s moral tailpiece, and 
Cosi in the musical textures) would appear calculated impudence if 
Stravinsky carried us no further than the point of recognition. But his 
‘rare form of kleptomania'! is not a sentimental evocation doing duty 
for a personal expressive code. As a composer he does not discover the 
crux of what is to be expressed through any predetermined ideas, but 
rather discovers parallels of feeling in the course of working what may 
be an outmoded musical convention. The stylistic deviations prompted 
by these parallels are, in fact, the quintessential Stravinsky, so that it is 
irrelevant that the background of expressive convention may change 
from Bellini’s to Mozart’s; and the artificiality of the process is particu- 
larly appropriate to the ‘framing’ which has always characterized opera 
buffa. 

Stravinsky’s word-setting provides a simpler duality. Its apparently 
false accents are merely visual, for the vocal lines themselves move 
naturally and often show a fidelity to the inflexions of speech remarkable 
in one who came so late to the English language. Set against a fixed 
metrical accompaniment, they constantly diverge and converge in an 
accent-scheme whose freedom effaces what might have been stilted in 
the stylistic pastiche. The re-timing of conventional progressions further 
loosens the movement, but rarely leads to any disturbance of tonal 
unity within each set piece.* 


1 Stravinsky and Craft, op. cit., p. 110. 
2 See Robert Craft, ‘Reflections on The Rake's Progress’, The Score, ix (September, 1954), 
p. 24, for a study of the set-piece forms. 


392 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Ex. 183 
(d= 60) 


IGOR STRAVINSKY AND THE RAPPROCHEMENT 393 


In Ex. 183 the vocal line finds its own accent-scheme from the words. 
The initial ambiguity of B flat and G minor persists in similar minor 
keys that rest on first inversions so as often to suggest majors. Having 
exploited the strangely exalted pathos of this device, Stravinsky finally 
reverts to the B flat pedal for a genuine major that has gained a trans- 
figured clarity. 

In his work on this opera, Stravinsky was assisted by a young 
American, Robert Craft,! widely versed in contemporary music, whose 
enthusiasms may have shortened the composer's route to serial proce- 
dure, charted with characteristic precision in the works that followed. 
Only a wisdom after the event notes signs of such a departure, incon- 
ceivable at the time, in some music of the forties: in the spare canon-by- 
inversion of the Two-Piano Sonata, or in the octave transpositions of 
the ten-note row (including repetitions) in the first Interlude of Orpheus 
(1947). The significance of the Mass (1948), one of Stravinsky's most 
accomplished works, is easier to sense than to define. In setting for 
liturgical use the text which, above all others, should resist the limiting 
effect of idiomatic cliché, he dispenses with his customary stimulus of 
familiar idiom. Despite broad parallels with plainsong and medieval 
polyphony, and incongruous echoes of his earlier work (Benedictus, cf. 
Oedipus), this score creates the impression of a fresh approach to simple 
elements of sound, a contemplation of chords and conjunct lines 
purged of both the furious rhythmic unrest of the Rite period and the 
automatic propulsion of pseudo-classical metre. From this attitude to 
Webern's contemplation of the interval and of silence may still be a 
long step, but it is not into an utterly foreign land. 

Yet of the fundamental propositions of serialism there is no sign in 


1 See Robert Craft, “А Personal Preface', Тһе Score, xx (June, 1957), p. 7. 


394 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


the Mass, and little of contrapuntal manipulation. The Cantata (1952) 
activates equally simple sounds by a ritual dance metre in the *Lyke- 
Wake Dirge' verses; but the first Ricercar introduces strictly inverted 
canon, and the second interlocks in one line the four serial forms of a 
sinuous eleven-note phrase (tonal, and with many repetitions), then 
works out numerous possibilities of their canonic relationship.! Though 
he still uses free parts, Stravinsky is clearly able to order the total sound 
of an intricate polyphony by his choice of transposition and rhythmic 
shape. But instead of the serialist’s dissolution of the row into the 
stream of the music, he deliberately stresses its melodic constriction 
and draws from this an ecstasy that reflects the text. In returning to 
instrumental textures with the Septet (1953), Stravinsky does not resist 
the lure of some well tried formulas in the sonata first movement; 
only the persistence of the opening motive and the complexity of the 
fugal development reveal his new concern for a basic unity of material. 
Тһе other two movements share one sixteen-note series, still tonal 
though now exploiting a wide range of octave transpositions. As 
ground in the passacaglia it often supports an upper texture woven 
entirely of the same thread. The binary fugal gigue, inverted in the 
second half, repeats both expositions in double fugue: the simultaneous 
subjects deliver the identical series in different note-values. That such 
mastery of serial resource could be won two years after The Rake’s 
Progress is remarkable enough; but that the Septet should also retain 
Stravinsky’s unique sonority gave evidence that his rapprochement with 
serialism was no irresponsible gamble with fashion, but had a validity 
rich in consequences for music other than his own. 

A variety of techniques, notably those of Bartók and Hindemith, have 
employed the twelve-note repertory as an enriched means of demons- 
trating relationships to centres of attraction. But the traditional rela- 
tionships depended on scales of uneven steps, and their retention, in 
however elaborate a form, was inconsistent with a fundamental material 
of undifferentiated semitonal steps (and in neither Bartók nor Hinde- 
mith is this really implicit). While Schoenberg’s recognition of the 
anomaly encouraged his adoption of the serial relations, Stravinsky 
chose to retain the old diatonic relations, but frequently to explore them 
in more than one orbit at a time. The critical hypothesis which treats 
the two men as poles in the music of their time? is supported by the 
results of these procedures. Whereas Schoenberg’s endless revolution of 
the chromatic total produced a multiplicity of pseudo-leading-notes and 


1See Colin Mason, ‘Serial Procedures in the Ricercar П of Stravinsky’s Cantata’, 


Tempo, 61-2 (1962), p. 6. 
2 See Adorno, Philosophie der neuen Musik (Tübingen, 1948), and Francis Burt, ‘An 


Antithesis’, The Score, xviii and xix (1956-7). 


IGOR STRAVINSKY AND THE RAPPROCHEMENT 395 


therefore a sense of constant forward movement, Stravinsky's super- 
imposed tonal pulls were often so mutually contradictory as to lead to 
the suppression of classical flow and to construction from disjunct 
and rhythmically autonomous segments. Despite relaxations from the 
most extreme positions—Schoenberg's occasional use of petrified 
rhythm! as well as his tonal tendencies; Stravinsky's use of “пео- 
classical’ metrical accompaniment patterns—the search for common 
ground was not actively encouraged while both men were alive. 

In the pre-war years, the political climate of Germany and the 
cultural climate of Europe in general rejected Schoenberg’s school, while 
an elegant imitation of the more superficial features of Stravinsky’s 
style was fashionable. When the stresses of war-time life led to a 
widespread demand for release in artistic experience, audiences of 
limited musical background found satisfaction in the classical orchestral 
repertoire or such modern music as could be most easily related to it. 
In these years, serial music was unknown to the public and confidently 
written off by the critics. But, as Dallapiccola has pointed out, *other 
systems, having once fallen into disuse, have never reappeared, whereas 
this one did reappear, and during the war years at that, in isolation and 
in all countries independently'.? And a post-war generation of composers 
concluded almost unanimously that the renewal they sought would set 
out from serial principle, without necessarily retaining Schoenberg's 
own fundamentally traditional rhetoric. Stravinsky's observation of this 
phenomenon must have led то a profound analysis of the further 
potentialities of his creative methods, and only the most cynical can fail 
to see a decision of great moment and still greater courage in his 
adoption of serial method after Schoenberg's death. In fact Stravinsky 
brought at least as much to serialism as he borrowed. His curiously 
oblique, though by no means careless recognition of tonal meanings, in 
dodecaphony as in his earlier music, sounds an entirely different note 
from Berg's yearning for tonal fulfilment. But even if the anti-tonal 
implications of some of his works had been fully realized,? his typical 
disjunct articulation and hypersensitive disposition of translucent 
sonorities would have preserved the vividly personal character of his 
music. The union of one of the most original creative minds and of the 
most constructive thought in the music of this century led to some of 
Stravinsky's most remarkable achievements; it also demonstrated a 
catholic basis for the technique of composition, and this was to 


1 As Hans Keller has pointed out in the String Trio: see Тйе Score, xx (June, 1957), p. 20. 

2 Luigi Dallapiccola, “Оп Ше Twelve-note Road', Music Survey, iv (1951), p. 331. 

3 Stravinsky and Craft, Memories and Commentaries, р. 107, and see Ех. 187 from 
Movements. 


396 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


be welcomed in circumstances where bigoted partisanship had seemed 
the only alternative to despair. 

Stravinsky's own progress to serialism was refreshingly free of such 
bigotry.! Far from accepting without examination the ultimate implica- 
tions of constructive principles essayed in the Septet, he next explored 
the valuable tension set up between these principles and the promptings 
of a text. A simple example is provided in the first of the Shakespeare 
Songs (1953) where *Musick to heare' is symbolized by revolutions of a 
C major scale fragment, absorbed through octave transpositions, 
instrumentation, and rhythmic irregularity into a texture elsewhere of 
the strictest serial observance. The row may be regarded as of twelve 
notes but is a network of inner relationships (see Ex. 184(i) ), and the 
use of the four-note nucleus y as the basic compositional element is an 
obvious pointer to Stravinsky's study of Webern during these transi- 
tional years. 


Ex. 184 
е X retro 
(i) та gel; mX retro- b — 
| = Е = пи x ITO 


уе уіпу.——4 Та vg 


ев ЖЕ 
К iuge Е 


Flute ток — = ЕРИ ае сұт АЕС y fel 5 
2-03 DEN 


Clarinet 2-4 Е 


(sounding FX 


р та marc. 


! Milton Babbitt, ‘Remarks on the recent Stravinsky’, Perspectives of New Music ii (2) 
(1964), p. 35, covers the period from the Cantata to Movements. 


IGOR STRAVINSKY AND THE RAPPROCHEMENT 397 


subito P 


In his next work, Jn Memoriam Dylan Thomas (1954), the technical 
restriction against which a powerful expressive urge is pitted is a single 
series of five notes, retained throughout every strand of the song and of 
its surrounding dirge-canons (for quartets of trombones and strings). 
Though the concentration of such a tour de force is partially eased by 
Stravinsky's ritornello practice, the emotional concentration (the 
‘fierce tears’ of Thomas's poem) is intensified; just as in earlier works, 
the crux of the feeling is discovered in the working-out of the musical 
material. 

Webern wrote ambitiously of a large-scale work, to last half an hour 
even in its reduced form,! yet his loving research into such small musical 
cells never produced, even with the consummate mastery of his last 
period, a work exceeding half that time. So it is not surprising that 
Stravinsky abandoned the introspective methods of these miniatures in 
undertaking two larger compositions, the Canticum Sacrum (1955) and 
Agon (Contest) (1954—7). Not only did he enlarge the scale of his serial 
working by adopting at last the full twelve-note repertory, but in the very 
moment of recognizing resources associated with atonality, he chose 
also to reinvigorate his old pandiatonic style. The chronology of Ароп? 
partly accounts for the juxtaposition, but in the Canticum the stylistic 
compatibility is deliberately and forcefully demonstrated:? the row 
manipulation of the middle movements yields clear tonal meanings, 
while the outer movements, at times even mono-modal, are united by 
one of the devices of serialism, strict retrogression. This crucial docu- 
ment in Stravinsky's development characteristically asserts present 
significance against a penetrating sense of the past. Conceived in honour 


1 Ір letters to Willi Reich concerning the Second Cantata, Ор. 31; quoted in Der Weg 
zur neuen Musik (Vienna, 1960), pp. 69 and 72. 

2 Roman Vlad, Strawinsky (Turin, 1958; Eng. ed., London, 1960), p. 199; Eric Walter 
White, Stravinsky, the Composer and his Works (London, 1966), p. 450. 

3 Robert Craft, “А Concert for Saint Mark’, The Score, xviii (1956), p. 35; and Roberto 
Gerhard, *Twelve-note technique in Stravinsky’, The Score, xx (1957), p. 38. 


27 


398 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


of Venice's patron, Saint Mark, it draws on instrumental timbres that 
recall the Gabrielis, on ricercare and cantus firmus technique and on 
plainsong affinities hidden in its two rows. Conversely, the serial 
technique marks a notable step forward by harmonic compressions and 
linear exchanges, while a new polyrhythmic independence and a control 
of wide vocal intervals underline the debt to Webern. 

Perhaps only in setting a sacred text can a spiritual unity be achieved 
that suspends disbelief in an ageless music. Certainly the piquant 
archaisms of Agon are intended to be savoured as such. Even more 
pointedly than the French overture which opened Apollon, these 
abstract dance movements re-create the stylizations of seventeenth- 
century ballet de cour without recourse to direct stylistic appropriation. 
Here the transition from the diatonic to the serial, achieved in slower 
stages yet in fact progressing further than in the Canticum, makes a 
fascinating summary of Stravinsky’s serial artifice and its consequences 
for his concept of texture. The listener’s impression is unified by the 
extreme originality of instrumental colour; sounds as arresting as the 
Prelude’s flutes with double-bass harmonics over harp and timpani 
arpeggios, or the canonic opposition of acid mandoline and cloying 
harp in the Gaillarde, testify to an undimmed aural imagination, as 
empirical an arbiter in Stravinsky’s serial as in his earlier music. 

It is no involuntary surrender to the mechanics of a system, therefore, 
that gives to Stravinsky’s first work on a single twelve-note row, Threni 
(Lamentations) (1957-58), its dark labyrinthine sound. A composer may 
choose to stress the obsessive character of strict row derivation,! and 
the smouldering gloom of Jeremiah’s Lamentations is uncannily lit up 
in this way. Even the difficulty of the vocal writing, weaving tortuous 
lines without instrumental support, contributes a sense of strain which 
is an important part of Stravinsky’s conception. (If an ability to pitch 
such lines precisely becomes general—and there are clear signs that this 
will happen—something will disappear from this work comparable 
perhaps to the tension lost in the change from the old natural horn to 
the modern instrument.) The Hebrew initial letters, retained from the 
original verse scheme in the Vulgate text, provide an obvious means of 
musical articulation but also an opportunity for a more detached, orna- 
mental exercise of serial art. In the first movement, and ‘Querimonia’ 
section of the second, Stravinsky is content to let these beautiful sounds 
stand as entities (see Ex. 185(i)) or serially linked repetitions (Ex. 185 (1): 
but in ‘Sensus spei", the central and most expansive section of the whole 
work, the eight Hebrew letters, containing twelve syllables, are equated 


! Analytical studies by Hansjórg Pauli, Tempo, xlix (1958), p. 16, and by John S. Weiss- 
mann, The Musical Quarterly, xlv (1959), p. 104. 


IGOR STRAVINSKY AND THE RAPPROCHEMENT 399 


with the twelve notes, and these are sustained in turn as instrumental 
single or double pedals throughout the succeeding verses. The series 
is thus made a super-ordering such as we shall note in the music of 
the post-war generation, though one which is no secret formula but 


Ex. 185 


Basic Set 


Strings 
(pizz.) 


400 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


sun са) 
Ip 4A 
. WS 


E - go vir vi-dens 


Retro „© ңә: 
transposed: 3—12; 1 4-2 


au - 
Trombones P 


BASS MONODY 
continues, then— 


Trombones 


the dominating sounding element. Tonal meanings are created in 
innumerable ways in Threni:+ as well as the network of relations within 
the row (see Ex. 185), its spanning of a minor third sets up a special 
relation between transposed forms at the minor third above and below 


1 Vlad, ор. cit., Eng. ed., p. 215; also Stravinsky and Craft, Memories and Commentaries, 
p. 107. 


IGOR STRAVINSKY AND THE RAPPROCHEMENT 401 


and at the tritone. ‘Querimonia’ explores the rearrangement of row 
segments (e.g. 5-124-4—1; 3-1--12-4) and ‘Solacium’ reveals the simple 
intervals and scalic effect of two-note groupings preparing the way for 
genuine permutation technique (e.g. 12-10-8-6-4-2-1-3-5-7-9-11). 
The orchestral texture is sparse but incisive, and its conversion to 
Stravinskyan sonority of the pointilliste manner (Ex. 186) clearly fore- 
shadows the next major work. 

Even the cryptic and ambiguous title, Movements for piano and 
orchestra (1958-9), suggests some affinity with works like Stockhausen’s 
Gruppen and Zeitmasse—though the title had already been used for the 
orthodox twelve-note textures of Fortner (cf. p. 413), Gilbert Amy’s 
work is a more likely precedent for it—and Stravinsky’s interest in the 
avant-garde is documented in the conversation volumes.! Its serial 


Ex. 186 
Fl. 
Ob. 
Cor. А 
Cit. 
Alto Clt. 
Bass Cit, 
^ pocosf p sub. 
Horn ку == ај ЕЕЕ 
orns ЦД —2 m ME 
sord. 7 соға, ее В 
тат тъ Р осо D p a 


y 3 
г E 
Е == 

Piano poco 

and Piano й тр 
Нагр = 
а, X 2 + 
[nce Зевин ees | 


0) 

ле | fg = 
үз =; 

Chorus parlando sotto voce zi i | | | | | 7 | | | 


(tutti) Re-cor-da-re Do-mi-ne quid ac -ci- 


1 Stravinsky and Craft, Conversations, pp. 125-33, and Memories, pp. 118-26. 


402 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


f) (Alo) in - tu - e - (re) 
== 
ms Si в +. 


ГТ i 
-de-rit no-bis 


(Remember, O Lord, what is come upon us.) 


usages are intricate, often permuting the row, two hexachords of 
mirrored constitution. But its most significant advances towards terri- 
tory pioneered by a far younger generation are in complexity and 
diversity of rhythmic shape (counterpointing freely the ‘irrational’ 
quintuple and septuple groupings; see Ex. 187) and in the avoidance of 
crystallized thematic material, though not of clear quasi-restatement 
correspondences. Chord formation, at the opposite extreme to wide- 
ranging monody in a constantly shifting textural balance remains 
peculiarly Stravinskyan, as does the orchestral sonority, despite a still 
more fragmented technique than in the previous works; and the scheme 
of dynamics has nothing to do with statistical distribution. 


PAUL HINDEMITH AND THE GERMAN MIDDLE GENERATION 
Stravinsky's cosmopolitan nature found America, like his other earlier 
refuges, a background to creative activity that could be accepted, even if it 
was with a kindly scepticism. It is typical of Hindemith that, while in 
America (1940—53) he associated himself enthusiastically with its musical 


PAUL HINDEMITH—GERMAN MIDDLE GENERATION 403 
Ex. 187 


Piano 


EE 
Cit. 
Bass СИ 
Tone. 
Celesta 
Harp 
Piano 
л == = == a le ЄЗ 
ЕН 
А ре 5---- 
5 $3 КЫ ск rer) eel 
Bas Ct. y H— — Я 18-3 59-6318 
SZO Lt) f mS 
c marc. 


Ты. ЕУ 


= mas шегере ласа 
ee [8 =: 


404 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


activities and pedagogy, notably as a professor at Yale, he was eventually 
drawn back to a European environment. Some direct offshoots of his 
American years were the theoretical works! (developing skills on a 
lower level than in the Unterweisung), his high-minded сгедо,? and the 
English settings, especially of the Whitman Requiem,? a moving tribute 
to the dead of whatever nationality. The generally mild harmony of this 
work betokens no change of style, since Hindemith, after enlisting the 
support of natural law* for the simplification of idiom achieved in 
Mathis, seems to have had little difficulty in accepting his own arguments 
as eternal principles. If finer distinctions can be drawn, it might be held 
that works written for America, and especially for its symphony- 
concert audiences, make a more deliberate bid for popularity. How 
valuably this offset the ponderous textures of much late Hindemith 


1 Elementary Training (New York, 1946); Traditional Harmony Тапа II (New York, 1943 
and 1948). For Hindemith's earlier music see Chapter IV, pp. 327-38. 

2 А Composer's World (1949-50 Charles Eliot Norton Lectures; New York, 1952). Ап 
important paragraph added to the German edition (Zürich, 1959) is translated in Journal 
of Music Theory (April, 1961), pp. 110-12. For some discussion of Hindemith's philosophical 
arguments, see Stuart Hampshire, The Score, vii (1952), pp. 58-62. 

3 When lilacs last in the door-yard bloomed, a Requiem ‘for those we love’ (1946). 

4 Unterweisung im Tonsatz (Mainz, 1937); translated as The Craft of Musical Composition 
(1942); its claims to natural law are investigated by Norman Cazden in ‘Hindemith and 
Nature’, Music Review, xv (1954), p. 288, Victor Landau, ‘Hindemith the System Builder’, 
Music Review, xxii (1961), p. 136, Richard Bobbitt, ‘Hindemith’s Twelve-Tone Scale’, 
Music Review, xxvi (1965), p. 104 and A. Daniels, ‘Hindemith’s Contribution to Music 
Theory’, Journal of Music Theory, ix (1965), p. 52. 


PAUL HINDEMITH—GERMAN MIDDLE GENERATION 405 


is seen in the sparkle of the Metamorphoses on themes of Weber (1943) 
and the lucidity of the Sinfonia Serena (1946). 

These symphonic works do not invite direct comparison with his 
orchestral music of the twenties, written in a concertante manner for 
adventurous groupings. Where comparison is inevitable, as between the 
early quartets and Nos. 5 and 6 (1943 and 1945), the later works dis- 
appoint by an unduly explicit tonal harmony, which constricts the old 
contrapuntal textures and encourages square-phrased forms. Favoured 
movement types—bulky march, lyrical pastorale, and so on—are as 
readily to hand as were the affective symbols of the Baroque composer. 

The interludes of Ludus Tonalis (1942) provide a convenient summary 
of these, executed at a particularly high level of invention. Designed 
as a didactic parallel to Das wohltemperierte Clavier, this is governed 
by an all-embracing tonal organization which rules the succession of the 
twelve keys? in a set of fugues according to their tonal distances from an 
initial point, charted by Hindemith's 'Series Г. But it is too uneasy a 
compromise between an archaic constructive principle and a personal 
hypothesis ever to approach either the topicality or the universality of 
its model, whatever the quality of its craftsmanship. Hindemith's 
mastery of contrapuntal device, rooted in the past, was perhaps intended 
as an object-lesson to the present: his mirror technique (the palindromic 
fugue in F, the totally inverted fugue in D flat, and the reversed and 
inverted relation of prelude and postlude) even embraces in the B flat 
fugue the four serial orders of the subject: 


Ex. 188 


(Moderato scherzando) 


! Heinrich Strobel, Paul Hindemith (Mainz, third edition, 1948), p. 127. 
? See supra, p. 336. The fugue in D is recorded in The History of Music in Sound, x. 


406 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


© = subject (B) = subject in retrograde motion 
(D = subject inverted = subject in retrograde inversion 


— subject augmented 


Far from losing definitive form in the fluid, inextricable mesh charac- 
teristic of serial styles, they affirm tonality in a metrically patterned, 
stratified texture. Movements of less subtlety, like the deft 5/8 fugue in 
G or the beautifully spaced canon in B, are more persuasive testimony 
to Hindemith's retention of his old imaginative powers. 

A curious innovation of this period, and an antidote to the monotony 
which may ensue when instrumental patterns are given their head, is the 
freely rhapsodic setting of words in non-vocal contexts. Thus, in the 
Two-Piano Sonata (1942) a recitative-like line declaims in piano tone 
the medieval poem ‘Wynter wakeneth al my care’. Other examples occur 
in the 1949 Horn Concerto (to words by the composer) and in the ballet 
score Hérodiade, a ‘récitation orchestrale’ after Mallarmé, 1944. In 
another work used as a ballet, The Four Temperaments (1940), Hinde- 
mith’s inventiveness takes a different form: the work consists of four 
vast variations for piano and strings on a tripartite theme, preserved 


PAUL HINDEMITH—GERMAN MIDDLE GENERATION 407 


with the fidelity of a cantus firmus throughout radical changes of tempo 
and rhythmic shape.! 

The opera Die Harmonie der Welt (The Harmony of the World) (1956) 
dominates by sheer size Hindemith's output after his return to Europe. 
Though the orchestral movements (also playable as a symphony, as with 
Mathis) introduce some original and complex sonorities in representing 
the whirling of the spheres, the whole work, on the life of the astronomer 
Kepler, rarely recaptures the urgency with which Hindemith had once 
depicted Grünewald's artistic and moral crisis. Lyrical passages of great 
beauty and some powerful choral scenes are momentarily arresting, but 
nobility of purpose is not in itself sufficient to maintain an audience's 
expectancy through the longueurs of philosophical argument. This opera 
provides no evidence of radically new departures after some twenty 
years’ reliance on the style expounded in Mathis; and Hindemith more 
strikingly affirmed his faith in that style by refashioning two operas 
that had symbolized earlier stages in his development, Cardillac (1926- 
1952) and Neues vom Tage (1929-53). 

Modern opinion unanimously condemns the presumption of editors 
who rewrite the works in their charge, but the action of the composer 
who applies the lessons of maturity to the impetuosity of his own youth 
is almost equally debatable. Because it concerns what is perhaps 
Hindemith's most inspired work, the revision of the song cycle Das 
Marienleben (1923-48) has become a test case, too involved to be argued 
here.? Two short quotations (Ex. 189) from 'Rast auf der Flucht nach 


Ex. 189 


(i) (-с.104-108) 1923 Version 


Voice 


Piano 


1 William Hymanson, ‘Hindemith’s Variations’, Music Review, xiii (1952), p. 20. The 
finale of the Octet (1957-8) provides a later example of this idiosyncratic variation tech- 
nique. 

2 See Hindemith’s introductory remarks to the new version (in English, 1953), and 
Rudolf Stephan, ‘Hindemith’s Marienleben (1922-48), Music Review, xv (1954), p. 275. 


408 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


- tier gan-ze Städ 


Voice 


р ____ _. 
EL 
pirum 


Piano 


- бег gan-ze Städ 


(and now, on their grey mule, they endangered whole cities.) 


Agypten’ show one context in which the composer's distrust of instinct 
has led to regulation according to principles developed later. Many 
changes have far-reaching effect on the structure, and several undeniably 
enhance the vocal flow, but only believers in absolute criteria of artistic 
effectiveness will applaud the transformation without hesitation; and 
performers have not relegated the old versions of the cycle, or of 
Cardillac, to the limbo which should logically be their fate. 


PAUL HINDEMITH—GERMAN MIDDLE GENERATION 409 


Critical practice now reserves the term Kapellmeistermusik as a 
reproach, yet it represents a concept that has often justified itself. Only 
the regional structure of Germany's music has made possible her vast 
creative activity, an assured background of general competence from 
which the figure of pronounced originality or genius can emerge. 
Inevitably, few composers achieve this prominence, but new music needs 
the background also, of essentially derivative, second rank figures who, 
given sufficient stimulus, dilute or synthesize unfamiliar idioms. It was 
the misfortune of the German generation that came to maturity in the 
National Socialist period! that this stimulus was denied by the banning 
of twelve-note music and of Hindemith's works. No doubt the latter 
were more immediately attractive, with their mechanized counterpoint 
and ultimate subservience of dissonance to tonal gravitation, and 
Zilig has pointed out that it was an unacknowledged caricature of 
Hindemith's manner, pruned of its dissonance and therefore utterly 
vacuous in its contrapuntal bustle, which became the official Nazi style 
in those barren wastes of Spielmusik by composers better forgotten.” 

For more adventurous minds the frustrations of aesthetic theories 
based on a dictator's whims? led to a crisis which could only be resolved 
in exile or in self-imposed silence. Krenek chose the first course, Hart- 
mann the second, a courageous gesture of protest from a composer just 
reaching the height of his powers. His teacher, Anton Webern, had 
already withdrawn into an obscurity that was to outlast his life. The one 
progressive influence not officially proscribed, that of Stravinsky, was 
not encouraged by performances. And so with the end of the war and 
the disappearance of aesthetic dictatorship, the composers of a stunted 
generation reached out eagerly for all the sustenance that had been 
denied them. The later music of Hindemith, Stravinsky, and the 
Viennese School suddenly became a part of their artistic environment. 
By now there was a younger generation seeking a model, and its dis- 
covery of Webern is a phenomenon to be discussed later. For many of 
the pre-war generation however, total assimilation was harder, and the 
revelation served to reinvigorate a gift for cautious emulation or 
synthesis. 

Carl Orff (b. 1895)* stands apart from these generalizations, for his 


1 A necessarily artificial but workable division is made between those composers born 
before 1910, who appear here, and the younger generation, discussed later. 

? Winfried Zillig, Variationen über neue Musik (Munich, 1959), pp. 136 and 255. 

3 Zillig, op. cit., p. 134, on Hitler's judgement of Hindemith's Op. 38 Concerto. For an 
obituary devaluation of Hindemith's work, including an entirely conjectural argument that 
his stylistic modifications were intended to secure Nazi approval, see Н. Е. Redlich, “Раш 
Hindemith—a re-assessment', Music Review, xxv (1964), p. 241. 

“К.Н. Wörner, ‘Egk and Orff’, Music Review, xiv (1953), р. 186, and Andreas Liess, 
Carl Orff (Zürich, 1955; Eng. ed., London, 1966). 


410 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


individual style had finished with outside influences (mainly of the 
Stravinsky of Les Noces) by the mid-thirties. After the war his most 
notable advance was in the scope of his dramatic undertakings: the 
two operas on Hólderlin's translation of Sophocles, Antigone (1949) 
and Oedipus (1955), apply to larger spans the numbing reiterative tech- 
nique of his immensely popular scenic cantatas. These hypnotic 
incantations of ancient languages or relentless probings of the primal 
dramatic situations codified in classical mythology have a fascination 
for the German that the outsider may find disturbing. Karl-Heinz Füssl 
claims that fas personified wish-fulfilment of the Germans, Orff stands 
in Germany above all criticism'.! Certainly, his prodigious sense of 
musical theatre, which can dispense with all intricacy of purely musical 
development, convinces audiences not particularly inclined towards 
contemporary idioms that they have satisfied their responsibilities in 
this matter. Orff's place in the larger musical world is won by his contro- 
versial but stimulating Schulwerk (an educational method, revised 
1950-4) and his early use of the large array of percussion instruments 
favoured by the new serialist composers. 

Werner Egk (b. 1901), like Orff a Bavarian and by temperament a 
composer for the stage, has a more orthodox technique with a veneer 
of self-consciously French sophistication. This is most evident in scoring 
of an elegance unusual in Germany, but he does not consistently reveal 
the harmonic sensibility which should accompany it. A typical orches- 
tral work, the Variations on a Caribbean Theme (1960), ranges from 
Gallic flute arabesques to the local colour of bongos, congas, and tom- 
toms, but both remain excrescences in a vulgarized Hindemithian 
texture whose moments of aspiration are nearer Hollywood. Such un- 
certainty of level is far less apparent in his operas, and his ability to find 
a simple but engaging formula for each dramatic situation brought 
widespread success to Der Revisor (after Gogol's Government Inspector). 
Though he has never achieved quite the degree of Orff's popular success, 
the genial and picturesque qualities of his early Zaubergeige commended 
him to a similar public on its 1954 revival. 

As well as writing many of his own libretti, Egk supplied one for 
Blacher's Abstrakte Oper (1953). Whereas Messiaen's nonsense text for 
his Cinq Rechants seems contrived for purely musical suggestibility, and 
the phonetic investigations of the avant-garde (Stockhausen, Berio, 
Kagel) are directed towards a bridge between speech and music, Egk 
uses a free juxtaposition of syllables to give an edge to basic emotional 
situations, dramatically resented. pNot only the chic quality of this 


1 Karl-Heinz Füssl, *Music in Austria and Germany today' in Twentieth-Century Music 
ed. Howard Hartog (London, 1960), p. 130. 


PAUL HINDEMITH—GERMAN MIDDLE GENERATION 411 


experiment but its roots in an unacknowledged sentimentality seem 
typical of the work of Boris Blacher (b. 1903).1 Even his use of swing 
formulas has a wan charm that seems an affectation beside the virile 
licence of jazz. In fact, Blacher has no need for these formulas, or for his 
echoes of Stravinsky, as spurs to a keen rhythmic imagination. Nor does 
he constantly rely on the mechanized arrangement of bar lengths 
which is his most celebrated invention. His claim that ‘this variable 
metre is less a rhythmical problem than a formal one"? is justified only 
when the longer bars can be apprehended as entities. Too often, 
however, subsidiary accents and even simple syncopations destroy the 
effect of progression through a higher unit of time. Comparisons of this 
metrical ordering with that of pitch serialism bear little inspection: 
Blacher's method cannot set up the dense network of palpable conse- 
quence and coincidence implied by serial principle. Yet it can be an 
effective liberating agent in conventional textures (see Ex. 190 from 
Orchester-Ornament, 1953) and has been profitably borrowed by com- 
posers too prone to Germanic stolidity, like Hartmann and the early 
Henze. Blacher's preferred textures are lean, ranging between Orff's 
prodigal economy and Webern's excessive refinement, but often occupy- 
ing a middle territory that allows scope to his fluent command of 
academic device (see e.g. the canonic movements in the Paganini 


Andante У 69 


zi 


Ex. 190 


Flute 


Bassoon 


Metre (х № 


Strings 


1 Cf. articles by С. Е. Kosuszek, The Score, i (1949), J. Rufer, Schweizerische Musik- 
zeitung, lxxxviii (1948), and С. von Einem, Schweizerische Musikzeitung, xc (1950). 

2 Boris Blacher іп a contribution to Josef Rufer, Die Komposition mit zwölf Tönen 
(Berlin, 1952; Eng. ed., 1954), p. 178 of English edition. 


MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


412 


Та > 


@ 

ЕЕ ШЕШ aes 

pom 
ЕЕ 


Fe) amies 
| 


‘oh, Wie ey ие | 
Dn 


SSS 


C 
RY) 


PAUL HINDEMITH—GERMAN MIDDLE GENERATION 413 


Variations, 1947, or the choral writing of the 1959 Requiem). A simple 
serialism is often found next to free dissonances, usually mild; these 
occasionally approach the internal contradiction of Stravinskyan 
harmony without casting off the obligations of progression. A typically 
deft synthesis is that of the final section in the Orchesterfantasie (1956), 
a movement which on the immediate level is characterized by Stravin- 
sky's nervous rhythm and third reiterations, by a climax of melodic 
sevenths and ninths and by a canon four-in-one. On a higher time scale 
it Is organized by the tutti interjection of twelve-note serial forms from 
the opening section, delivered in single notes at intervals reduced 
progressively from twenty-three bars to continuous crotchets and still 
further diminution. These orchestral works found a readier hearing 
thanks to their scoring, which is brilliantly apt though essentially 
traditional; in the theatre, over-anxiety to simulate a wealth of abstruse 
meaning (see e.g. Rosamunde Floris, 1960) jars with a music that is best 
when least pretentious. 

Though the vast palindromic Amen of Blacher's Requiem is executed 
in transparent textures, it suggests a peculiarly German pride of 
craftsmanship (contrast e.g. the retrograde techniques of Stravinsky's 
Canticum or Nono's Incontri). Not surprisingly, the more usual outlet 
for this is a weightier contrapuntal idiom springing from a love of the 
Baroque and from Hindemith, whether or not it recognizes the har- 
monic mandate of the Unterweisung. Few of the innumerable composers 
who command such a style have more than local interest; but that it 
can become compelling by consummate mastery of device is shown by 
the best works of the Austrian, Johann Nepomuk David (b. 1895). 
Ernst Pepping (b. 1901) has produced admirable church music but failed 
to make a distinctive mark with his more ambitious work. 

Wolfgang Fortner (b. 1907), who began from just such a position, 
has achieved international recognition. His earlier music added to the 
routine Hindemithian industry a strong feeling for clean sonorities 
deriving from Stravinsky. After the 1947 Symphony, a testimony to that 
period's hope and fears which played an important role in Germany's 
musical recovery, his assimilation of twelve-note methods (a process 
perhaps aided by his pupil, Henze) led to still wider prominence. 
Although his approach to texture and structure remains rooted in the 
past, he provides another instance (with Stravinsky and Dallapiccola 
as the most striking) of the new imaginative vistas opened to an already 
fluent composer who discovers in serial thought a natural medium. In 
the Mouvements for piano and orchestra (1953)! exemplary workman- 
ship, with much neat dovetailing of rows, is lavished on an entirely 

аа comment by Francis Burt, “Ап Antithesis', The Score, xviii (1956), р. 16. 
2 


1940-1960 


- 


MUSIC OF THE EUROPEAN MAINSTREAM: 


414 


Allegro molto vivace 


Gi) 


= 
prn же 


p 
р» Г 


Е рад 


ГА 
( 
ЕС 


Piano 


. | | 


Strings 


Trombones 


бға------77-"----%-------- 


er. 


PAUL HINDEMITH—GERMAN MIDDLE GENERATION 415 


traditional sequential extension of material. The norm of dissonance is 
low (as the series suggests—cf. Ex. 191(1)) and textures fluctuate 
between neo-classical and neo-romantic, but with a preponderance of 
the patterned figuration that is Hindemith's legacy to a whole German 
generation (see Ex. 191 (п)); the serial boogie on the other hand is a 
conceit that would appeal to Blacher. 

Seven years later, in the Ашофе for oboe and orchestra (1960), 
Fortner's habit of transposing a segment of the row as a pseudo-classical 
developing motive! has led to the use of a single motive of three notes; 
the juxtaposition of its four serial orders can produce a twelve-note 
row. It is not only this refinement that suggests acquaintance with post- 
Schoenbergian developments. The rhythmic interest has increased, 
chord building and scoring are more varied; and the rhetorical scheme, 
in which pastoral solo strains and belligerent assaults are convincingly 
related, is new, although inordinately prolonged. The expressive release 
which Fortner's stylistic exploration afforded is strikingly confirmed іп 
his vocal music; the almost embarrassing fervour of The Creation (1954) 
is stimulated as well as controlled by its technical constraints, and the 
Lorca opera, Die Bluthochzeit (The Blood Wedding) (1953) exhibits a 
range that no merely parochial talent could command. 

As well as his important work as a teacher in Heidelberg and Freiburg, 


! Cf. Fortner on his serial technique in Josef Rufer, op. cit., p. 182. 


416 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Fortner organized series of concerts in those cities under the title 
Musica Viva. In this he followed the lead given in Munich by Karl 
Amadeus Hartmann (1905-63). Within six months of the end of the 
war, Hartmann, whose withdrawal from the German musical scene has 
been noted, launched a campaign to re-educate composers, performers, 
and audiences by programmes ranging from the precursors of the new 
music, through the virtually unknown masterpieces of the pre-war 
generation to the experimental work of the emerging generation. 
Catholic but intelligent programme arrangement and adequately re- 
hearsed performances were accorded the highest importance. The 
results of Hartmann's labours can scarcely be overestimated: not only 
did the Munich audiences grow from a mere thirty to some fifteen 
hundred, but Munich Radio adopted the series and transmitted it to a 
far wider public. Its imitation in other German centres (and further 
afield, including Liverpool and Glasgow) brought modern music to 
people mistrustful of the esoteric atmosphere of Darmstadt and 
Donaueschingen, where the newest music has been methodically per- 
formed and discussed by adepts. 

Hartmann's own works reflect these broad sympathies, and an alert 
ear immediately traces debts to Stravinsky, Bartók, Berg, and indeed, 
almost every important figure of the century. Yet these never deface the 
impression of a strongly personal musical ideal, pursued so constantly 
as to suggest monotony to the not wholly sympathetic listener. This ideal 
seems to spring from a new interpretation of Bruckner's symphonic 
concept that has focused on the profound adagios and bucolic scherzos 
of that master. Whereas the intense, rarefied mood of the former is 
subjected to Hartmann's impulsive but disruptive fantasy, the animal 
vigour of the latter is dissipated in mechanized concertante figurations. 
This paradox is seen most clearly in the two symphonies which under- 
line by their scoring a preoccupation with one of these types—the 
ardent, almost hysterically impassioned Symphony No. 4 (1947) for 
strings, framed by adagios, and the prosaically chattering Symphonie 
Concertante (No. 5, 1950) for wind, cellos, and basses, dominated by 
neo-Baroque allegros. The symphony completed between these two 
(though labelled No. 3, 1949) is more satisfactory than either, not only 
because of its full orchestral resources, with much bright-toned percus- 
sion, but also because Hartmann's use of a single continuous span 
compels him to cultivate some middle ground between his extreme 
moods and encourages him to abandon these when their initial urge is 
exhausted. Framed by the funereal opening and its final recall, an 
emotional pattern which the nineteenth century's abuse has made 
unfashionable is given new life by a centre-piece that demonstrates 


PAUL HINDEMITH—GERMAN MIDDLE GENERATION 417 


aggressive optimism in terms of cumulative fugue and ground bass. 
Fugal technique seems a natural solution here, but it has become a 
mere decorative procedure in the Sixth and Seventh Symphonies of 
1952 and 1959. Hartmann's inclination to cling to mechanisms that 
have no essential relation to the music’s growth is even more clearly seen 
in his fondness for mirrored writing, not pulled forward by canon but 
ranged immobile around a central point. In the Viola Concerto (1955), 
the brass chorale (a common borrowing from Bruckner) is punctuated 
by solo interludes (cf. Berg's Violin Concerto); despite a chromatic 
introversion that recalls Bartók, the strict mirror technique: 


Ex. 192 Andante con moto 


Muted 
Brass 


has a stiffness quite foreign to that composer's symmetrical feats. 
The spirited revival of German music by the generation of Blacher, 
Fortner, and Hartmann was one token of a determined bid for national 
self-respect. Blacher's music in particular, with its urbanity, efficiency, 
and hints of defensive irony, is a perfect counterpart to the faintly 
anxious good taste of the modern German hotel or office block. In 
sharp contrast, Austria's recovery of national autonomy was musically 
celebrated by an affectionate embracing of her past, or of those more 
recent composers (like Franz Schmidt) whose memories of it were 
least clouded by this century's preoccupations. Neither the Nazi ban 
nor their subsequent transformation of the European musical scene 
secured for Schoenberg's methods the championship of an Austrian 
public which had surrendered to its characteristic (and commercially 


418 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


profitable) nostalgia. The generation that should have carried on from 
the three great pioneers is represented only by two of their pupils, 
Apostel and Jelinek, and by Ше emigrant Křenek. 

Though Hanns Jelinek (1901-69) worked indefatigably for a 
Gebrauchsmusik based on twelve-note principles, notably in the nine 
volumes of Zwélftonwerk, Ор. 15,1 he failed to find a convincing 
solution to the problems of his age in merely making dodecaphonic 
textures sound as nearly as possible traditional. Audiences that are 
soothed by what is familiar in such music are no nearer acceptance of 
any further consequences of twelve-note methods. Of course an under- 
standing public is still less likely to be recruited if such consequences are 
pursued so far, and so rapidly, as to reach a private world. The remark- 
ably varied career of Ernst Křenek (b. 1900) may be seen as a virtuosic 
feat of liaison. Gifted with an imagination that set him naturally among 
the explorers, he has sometimes been compelled by a sense of respon- 
sibility to interpret his findings in terms which the less ambitious listener 
can hope to understand. He remained faithful to dodecaphony after 
settling in America in 1938, but drew from its few basic principles 
a wide selection of compositional procedures? with as wide a range of 
stylistic consequences. His choral Lamentations (1942) reinstate a scalic 
norm by deriving all their material from two basic hexachordal ‘modes’ 
and the attendant modes obtained by setting out from degrees other 
than the first; developments of this principle in instrumental works in- 
clude systematized rotations of note-order within smaller segments. In 
the years of these experiments, Křenek sometimes sought wider com- 
munication through a simpler technique. But the post-war emergence of 
young composers determined to conduct such experiments on a far 
broader basis found Křenek ready to learn as well as teach. As his own 
work had always remained traditional in its progressive development 
of the recognizable motive, he was reluctant to see this abandoned in 
favour of apparent chaos, suspicious of serial mechanisms bordering on 
total pre-determination, and frankly critical of electronic phenomena 
*of a considerably lower intellectual level of musical consciousness than 
the aspirations which were associated with the music of the past.’ 
Characteristically, he tested these first reactions by the most practical 
method, venturing not only into schemes of multiple serialization in 
such works as Kette, Kreis und Spiegel (Chain, Circle, and Mirror) 


1 See also his textbook, Anleitung zur Zwélftonkomposition (Vienna, 1952), and its review 
by Křenek, Musical Quarterly, хі (1954), р. 250. 

2 See КТепеК’5 writings on these in Music Review, iv (1943), p. 81; Musical Quarterly, 
xxxix (1953), p. 513; Musical Quarterly, xlvi (1960), p. 210, and in Josef Rufer, op. cit., 
p. 188. 

з Ernst Krenek, ‘A glance over the shoulders of the young’, Die Reihe, i (Vienna, 1955; 
Eng. ed., 1958), p. 14. 


THE MIDDLE GENERATION ELSEWHERE 419 


(1958) but joining company with men of the next generation in 
exploring the mingling of electronic sound with voices in the oratorio 
Spiritus Intelligentiae Sanctus (1956). 

After Kfenek's intellectual curiosity has been acknowledged, it must 
be said that his creative achievement is disappointingly uneven, and bears 
no constant relation to the degree of technical innovation. If early 
works like Jonny or Leben des Orest showed that he dispenses with all 
technical stimulus at his peril, the opera Pallas Athene weint (1955) has 
shown that he also needs stimulus of another kind to reach his most 
impressive level. Here a burning concern for liberty is given didactic yet 
forceful expression through musical symbols that lose no immediacy for 
being derived from a personal interpretation of twelve-note practice. 


THE MIDDLE GENERATION ELSEWHERE 


During the years of Nazi artistic proscription, Switzerland was able 
to assume responsibility for the first performance of works as important 
as Mathis, Lulu, and Webern's Orchestral Variations. Her own com- 
posers remained free to profit by any example they chose, and were by 
national temperament inclined to choose widely. Honegger's stylistic 
amalgam was fused long before this period, but his development as a 
symphonist was touched off by an urge to comment on its dark events 
(in the powerful Second Symphony, for strings and trumpet, 1941); 
three more essays followed. A development still more remarkable was 
that of Frank Martin, born in 1890, but little known before the war. 
Stravinsky's journey from pandiatonicism to dodecaphony began as 
research into serial method, not necessarily demanding twelve-note 
material. Martin, on the other hand, was satisfied by old textural ideals, 
realized with considerable ingenuity and refinement, but he increasingly 
liberated his melodic line by use of twelve-note successions.! Below this 
surface his music often remains entirely diatonic, and the bass line 
regains the guiding role lost in both Schoenberg and Stravinsky. As a 
result, his harmonic textures, however fluid in their accommodation of 
chromatic line, never seethe with the inner tensions that are the source of 
Schoenbergian Angst.” АП is so patently under control that excitement is 
noteasily generated, though ungrudging admiration is often aroused. The 
popularity of the Petite Symphonie Concertante (1945)? is earned by the 
felicitous deployment of its varied string colours, and its counterpart 


1 Roman Vlad, “Егапк Martin' in his Modernità e tradizione nella musica contemporanea 
(Turin, 1955), p. 236. 

? Martin's opposition both to Schoenberg's aesthetic and to subjugation to rule is ex 
pounded in articles in Polyphonie, iv (1948) and Schweizerische Musikzeitung, 1хххїї (1942). 

3 Cf. Jacques de Menasce, “Егапк Martin and his Perite Symphonie Concertante', 
Musical Quarterly, xxxiv (1948), p. 271. 


420 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


in wind tone, the 1949 Concerto, is perhaps even more adroit. But 
in works which demand our deeper engagement, the opera on The 
Tempest (1955) and the series of oratorios! dominated by Golgotha 
(1948) and Le Mystére de la Nativité (1959), not all Martin's concern for 
sonority can hide the limitation of his idiom. The first impact of serialism 
has proved a rich source of creative tension in many men's work, but 
the composer who seeks to maintain so delicate a poise is in danger of 
unwittingly reducing it to a pose. 

On the other hand, a composer whose receptivity to new ideas is at 
once eager and cautious may profit for years from a gradually increasing 
commitment to serial procedure on his own terms, and the career of 
Luigi Dallapiccola (b. 1904) affords an interesting study in such 
assimilation.? Stravinsky's magisterial appropriation was achieved after 
serial music had become widely available, but Fascist Italy in 1940 was 
a less propitious environment. ‘Just at the time when everyone had 
ceased to mention atonality or twelve-note music, I began to be pas- 
sionately interested in such problems, Dallapiccola has written.? In 
fact, Schoenberg's influence dated from a performance of Pierrot lunaire 
in 1924, which made him resolve to study composition.* During the 
following years he mastered a wholly Italian style, re-interpreting the 
neo-classical concepts of Busoni and Casella through a feeling for 
polyphonic texture that reached back to pre-classical times, yet excluding 
neither the Verdian cantabile nor the expressive use of instrumental 
colouring. Having forged a link with Italy's past, he sought one with 
the European present that should preserve the native qualities of his 
music. As he writes, “һе field which most attracted me to the twelve- 
note system was that of melody.'* 

Already his melodic lines had tended to juxtapose diatonic, even 
modal, shapes in the wider context of the chromatic repertory; unlike 
Martin, Dallapiccola explored the implications of such lines in contra- 
puntal engagement, especially through canonic procedure. By the time 
of the opera Volo di notte (Night Flight) (1939) he had arrived at work- 
ing principles that foreshadow serialism without aping the Viennese 
pioneers. Deprived by political prejudices from direct study of their 
scores, he sought inspiration in the literary procedures of Joyce and 
Proust. But the lack of technical guidance was no obstacle to the over- 


1 For Martin's views on sacred music, see Schweizerische Musikzeitung, Ixxxvi (1946). 

2 Cf. Roman Vlad, Dallapiccola (Milan, 1957); also articles by Vlad in Horizon, xx (1949. 
(reprinted in Modernità . . .) and The Score, xv (1956), p. 39. 

? Luigi Dallapiccola, “Оп the Twelve-Note Road', Music Survey, iv (i) (1951), p. 321. 

а Cf. Vlad, Modernità . . ., р. 197. 

5 Dallapiccola, in Josef Rufer, op. cit., p. 180. 

в Dallapiccola, “Оп the Twelve-Note Road’, loc. cit., p. 323. 


THE MIDDLE GENERATION ELSEWHERE 421 


whelming expressive urge of the Canti di prigionia (Prison Songs) (1938- 
1941), for the composer's own experience of political tyranny had 
already nourished that fanatical devotion to the cause of human liberty 
which is at the root of his major works.! A natural affinity with modal 
chant enabled him to make of Dies irae an awesome motto, but serial 
thought and an imaginative ear for icy sonorities (two pianos, two harps, 
and many percussion instruments) combined to produce the strange 
menace of the second movement. 

Dallapiccola has stressed that mere quantitative equality of notes 
cannot prevent the predominance of some, due either to their placing 
in time, or to 'extremely subtle relationships which exist between 
certain notes', polarities which are fundamental characteristics of each 
row.? In the works which followed the Canti di prigionia he explored the 
properties of many rows,? never making a technical advance without 
assuring himself that he was still in control of the total sonority and of 
its immediate expressive potentiality. The seductive textures of the 
Liriche greche (Greek Lyrics) (1942-5) conceal experiments in the 
integration of material which made possible the fluency of И Prigioniero 
(The Prisoner) (1944—8).* Іп an age when any compositional method but 
the most recent is fervently denounced, few works have been more 
salutary reminders than this opera that the one essential vehicle for a 
vital creative impulse remains a technique which, though already 
mastered, is still for the composer an imaginative adventure. Dalla- 
piccola's variety of rows and motives, his evocative use of traditional 
chord structure, his mingled echoes of Verdi, Debussy, and Berg, all 
offend against a priori conceptions of serialism; but his right to profit 
as he chooses from serial discipline is vindicated by the powerful impact 
the work has continued to make. 

The technical premises of this impassioned protest against physical 
and mental torture? could not, however, be systematized into a guaran- 
tee of further success. In Dallapiccola's “засга rappresentazione', Job 
(1950), the constant burden of tribulation presenting itself in many 
guises may have prompted his first large-scale structure from a single 
row.$ Far from inducing monotony, this inspired almost too richly 
varied an imagery, and the spoken narrative tends to disperse the 
musical cumulation. After brilliantly demonstrating contrapuntal device 


1 Cf. В. Smith Brindle, ‘Italian Contemporary Music’ in Howard Hartog (ed.), European 
Music of the Twentieth Century (London, 1957), p. 176. “ Dallapiccola, loc. cit., pp. 325-6. 

з Details of Dallapiccola's twelve-note work from 1942 to 1957 are given in Hans Nathan, 
“Тһе Twelve-Tone Compositions of Luigi Dallapiccola', Musical Quarterly, xliv (1958), 
p. 289. * Cf. Dallapiccola, *Notes sur mon opéra', Polyphonie no. 1 (1948). 

5 Cf. Vlad on the significance of 1/ Prigioniero in The Score, xv, (1956), p. 43. 

6 Including а form achieved by permutation; see Nathan, loc. cit., p. 291, n.3. 


MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Ex. 193 


422 


(4 Sops.) 
О 


ea 


al 


Molto lento (2 


nos - tra 


fi - des 


(4 Sops.) 


E 


Horn 


fra - ter. 


О 


(О brother, were our faith firm, miracles would be wrought in us.) 


THE MIDDLE GENERATION ELSEWHERE 423 


in the diatonic contexts of the violin Divertimento on Themes from 
Tartini (1951), Dallapiccola applied the same skill to a row in the 
‘Contrapuncti’ of the Quaderno musicale di Annalibera (Musical album 
for Annalibera) (1952); like the studies in mood and texture which com- 
plete this set of piano pieces,! they show a new preoccupation with 
rhythmic variety that was to distinguish his masterpiece, the Canti di 
liberazione (Songs of Liberation) for chorus and orchestra (1955). 
Indeed the same series underlies both works and its choral presentation 
in the first canto is taken directly from ‘Fregi’ in the Quaderno (see Ex. 
193 to bar 6—an ‘all-interval series") a reminder of the vocal nature of 
all Dallapiccola's line. From the bass entry in bar 5 the imitations of a 
serial fragment with progressive diminution (but constant proportion) 
of note values, introduce a device used in every movement. The scale 
of rhythm values in use, any one of which may be treated as a basic 
unit, reveals a growing interest in Messiaen and the avant-garde. Though 
the frequent width of melodic interval points to Webern, Ex. 193 shows 


5322422444 


how diatonic are the implications of the segments (although the direct 
octaves of J/ Prigioniero have disappeared), and Dallapiccola's thematic 
use of the row elsewhere recognizes the contrasted expression of near- 
conjunct line. 

If the scoring of this work (for full orchestra) comes near to the 
Bergian espressivo, much of his music shows Dallapiccola to be master 
of a slighter medium more typical of the period—solo voice and chamber 
ensemble. Works like the Goethe-Lieder (1953)? and the Cinque canti on 
Greek poems (1956), have refined instrumental textures that contribute 
subtle points of symbolism without ever challenging the expressive 
supremacy of the voice. The visual symbolism (cf. the madrigalists' 
*eye-music") of note patterns that form in the score a cross in Cinque 
canti or a Christmas tree in the Concerto per la notte di natale (Concerto 
for Christmas Eve) (1957) conveys nothing to the listener, but is an 
illuminating commentary on the almost spiritual involvement of this 
composer in all he writes. These works reveal the completed assimilation 
of Webern's influence in their freedom of wide interval and in textures 
assembled from reflecting segments. But Berg remains the inspiration 
for the contrasting harmonic blocks of the Concerto, and the use of the 

1 Orchestrated as the Variazioni (1954). 

2 The German inflexions (Wagner and Schoenberg) which affect the lines of the Goethe- 


Lieder are shown in Nathan, loc. cit., pp. 294-5 and 299. Nos. 2, 3, 5, and 6 are recorded 
in The History of Music in Sound, x. 


424 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1660 


voice in two movements of this instrumental form confirms that even 
in his most advanced style Dallapiccola works with vocal shapes.! 
Other composers of Dallapiccola's generation observed his serial 
progress with interest: Seiber noted a development that 'far from 
restricting his ideas or cramping his style ...seems to have liberated 
his colourful imagination.”? In Italy, the revelation of native lyricism 
enhanced by the twelve-note method proved an incentive to experiments 
that even touched figures as revered as Malipiero and Ghedini. The most 
thoughtful and individual compromise characterized the later work of 
Goffredo Petrassi (b. 1904).3 Less dependent than Dallapiccola on 
the stimulus of a text, he had achieved a neo-Baroque manner which 
softened the rigidity of Hindemith's instrumental counterpoint but was 
founded in a similar extension of diatonic (and in his case modal) 
principle. Gatti's discovery of “Ше demon of modernistic intellectual- 
ization'* in works like the orchestral concertos is unduly severe on 
Petrassi's cool but not aggressive rejection of a romantic pose he would 
have found uncomfortable. Just as Stravinsky proved repeatedly 
that this attitude need not lead to a desiccated objectivity, so Petrassi 
with his male voice Coro di morti (Chorus of the Dead) (1940-1) drew a 
dramatic power from his treatment of Leopardi’s dark verses that is the 
more compelling for its universal intimations. The hard sonorities 
(brass, three pianos, percussion, and double-basses) appear less Stravin- 
skyan in a context that includes choral writing of an archaic (and Italian) 
flow and instrumental contours betraying an interest in twelve-note 
models. Despite a number of stage works, two ballets and the operas 
Il Cordovano (1948) and Morte dell'aria (1950), the Coro was not sur- 
passed until he worked on a subject of comparable grandeur in the 
cantata Noche oscura (Dark night of the soul) (1951). Vlad aptly cites 
some words of its poet, St. John of the Cross, ‘mysteries are revealed 
through strange shapes and images'," to explain Petrassi’s need to break 
loose from all diatonic ties in setting the erotic symbolism of the poem. 
He achieved this by allowing the serial orders of a chromatic motive to 
roam through musical space so as to cover, though not to regularize, 
the resources of dodecaphony. Serial practices reappear in the Third 
Orchestral Concerto (Récréation Concertante, 1952), and in the Fifth 


1Cf. John C. С. Waterhouse, “Тһе Italian Avant-Garde and National Tradition’, 
Tempo, 68 (1964), p. 14. 

2 Matyas Seiber, ‘Composing with Twelve Notes’, Music Survey, iv (iii) (1952), p. 486. 

3 Cf. John S. Weissmann, ‘Goffredo Petrassi', The Score, iii (1950), р. 49, and his book 
of the same title (Milan, 1957); also Roman Vlad in Modernita e tradizione nella musica 
contemporanea (Turin, 1955), p. 217. 

4 Guido M. Gatti, article on Petrassi in Grove, 5th ed., vi, p. 689. 

Сү ач Корс рт 2202» 6 СГ. Weissmann, The Score, iii (1950), p. 59. 

? Cf. Vlad, “Га Noche Oscura’ in Modernità . . ., p. 234. 


THE MIDDLE GENERATION ELSEWHERE 425 


Concerto (1955)! the free permutation of a six-note row in one move- 
ment and its complement in the other produces homogeneous but 
contrasted material. By the time of the 1957 String Quartet and the 
Serenata (1958) Petrassi had extended his interests to the work of the 
younger generation; the latter work, for flute, viola, double-bass, 
harpsichord, and percussion, still betrays his old neo-classical sympa- 
thies yet manages to absorb them into a style ranging easily from simple 
cantabile line to the mannered percussion writing of the avant-garde. 

In countries more remote from the decisive happenings of modern 
music, the composers best equipped to rise above a crippling pro- 
vincialism were those with first-hand experience of the central lines of 
development. Two pupils of Schoenberg who are comparable in this 
respect are the Spaniard Roberto Gerhard (1896-1969) and the Greek 
Nikos Skalkottas (1904—49), both of whom revert occasionally to a 
simpler, nationalist manner. Skalkottas suffered from a self-imposed 
isolation, in which he pursued his own ideas without reference to con- 
tinuing developments of the Central European technique he had 
acquired.? Yet he also lacked the stimulus or corrective of hearing his 
own works in performance. Opportunities to hear the symphonic and 
other extended works are still rare, which is particularly unfortunate 
since the short pieces available in print, talented applications of an 
individual serialism? to square-cut forms, relapse at times into harmonic 
and textural formulas (often more reminiscent of Hindemith of the 
twenties than of Schoenberg) too jaded to convey a vital expressive 
urge, as in this example from the Fourth Suite for Piano: 


Ex. 194 Tempo di Polka moderato Ж 


1 СЕ Kenneth Gaburo in ‘Current Chronicle’, Musical Quarterly, xlii (1956), p. 530. 
? Cf. John G. Papaioannou, *Nikos Skalkottas' in Hartog, op. cit., p. 320. 
Sibide ра 125% 


426 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


It is difficult not to believe that more rapport with European music in 
Skalkottas' later years would have preserved his most personal charac- 
teristics—lithe energy, clear yet not attenuated sonorities, and a quizzical 
or even sardonic tone—and checked his tendency to let textural pro- 
liferation take control at moments of flagging impulse. 

After the Spanish Civil War, Gerhard! left his own country for 
another territory of the musical periphery, England. His wide musical 
sympathies and penetrating intelligence found him a place there 
both as a masterly composer of incidental music and as a commentator 
on contemporary developments;? these he studied deeply and con- 
verted to his own ends. Аз a result, whatever Spanish inflexions ог 
subtlety of English word-setting may contribute to its individual tone, 
his work reveals a European legacy that demands some mention in the 
present survey. This has so far been a record of composers who struck 
a basically diatonic stylistic balance before the war, but were in many 
cases impelled to modify this by some aspect of Schoenberg's 
thought. Dallapiccola has quoted from Giséle Brelet’s comments on 
atonality: 'trop tót venu, П lui fallait attendre que surgisse chez les 
musiciens la conscience des problémes auxquels il prétendait apporter 
une réponse.'? It is interesting to note that this time-lag was felt neces- 
sary even by Gerhard, a composer steeped in the Schoenbergian 
aesthetic and fluent in its practical consequences. In contrast to many 
composers, his late adoption of fully serial technique followed music 
written against a background of intimate knowledge of its workings; 
thus not only is his serial music entirely without the gaucheries of first 
experiments but it can coalesce easily with a freer style.* The Violin 
Concerto (1942-5) is a beautiful example of this, his comic opera The 
Duenna (1945-7) turns again to a pointed diatonicism,® and in later 


1 Cf. The Score, xvii (1956), a sixtieth birthday tribute to Gerhard, with articles on his 
work and a catalogue. 

2 Cf. Gerhard's articles in The Score, vi, ix, xvi, xvii, xx, and xxiii. 

3 Giséle Brelet, Chances de la musique atonale (Alexandria, 1947) quoted by Dallapiccola 
in ‘On the Twelve-Note Road’, loc. cit., p. 320. 

* Cf. David Drew, ‘The Musical Character’, The Score, xvii (1956), p. 39. 

5 On its scoring cf. Norman del Mar, ‘Gerhard as an orchestral composer’, ibid., p. 17. 

$ Cf. John Gardner, ‘The Duenna’, ibid., pp. 20-6. 


THE MIDDLE GENERATION ELSEWHERE 427 


instrumental works he has used an increasingly ordered serialism. To 
the permutational ‘inversions’ of complementary hexachords which 
regulate tonality (in Gerhard's broad sense)! in the 1953 Symphony is 
added in the Quartet (1955) a series of time-proportions derived from 
the pitch set by a simple process of measuring semitonal distances.? 
With the 1959 Symphony he cast off the dependence on motivic contour 
which he considered tautological in serial working, and in Collages 
(1960) he tackled the problem of fusing electronic and instrumental 
sound. Gerhard’s belief in an intelligently considered ‘system of 
arbitrarily set-up co-ordinates’? as а vital spur to the creative imagina- 
tion was amply justified by his own works. 

Other peripheral figures have been content with narrower horizons. 
In Scandinavia the most notable emergence during these two decades 
was of the Norwegian Fartein Valen (1887-1952), though most of his 
music had been written earlier. Valen's link with a central tradition was 
through his teacher Reger, but the contrapuntal style he developed 
stepped out of the shadow of Bach. Despite coming under Schoenberg's 
influence to the extent of using twelve-note melodic rows, he did not 
allow them consistently to permeate the texture; and the motivic dis- 
cussion which unites the parts sounds aimless once a dominating 
melodic strand is abandoned. Valen's enterprise deserves notice as an 
early venture towards territory that has still attracted few Scandinavians. 
More typical was the fluent but eclectic conservatism of the Dane, Vagn 
Holmboe (b. 1909), as indefatigable a composer of chamber concertos 
as was the Hindemith of the twenties, but of far less brilliance. In 
Sweden, Hilding Rosenberg (b. 1892)* wrote a series of oratorios and 
transmitted a catholic taste to some gifted pupils; Gósta Nystroem 
(1890-1966)5, without modifying his style, produced his most distin- 
guished work in the Sinfonia del mare (Sea Symphony) (1948). 

The younger Scandinavian generation includes several composers 
who will call for discussion later. In Central and Eastern Europe too, 
and here even more surprisingly, the influence of the most radical 
advances in serial style helped to nurture some impressive talent. 
Of the older men, little can be reported which reflects a significant 


1 Cf. Gerhard, ‘Tonality in Twelve-note Music’, The Score, vi (1952), p. 23. For relevant 
discussion of hexachord combinational properties cf. George Rochberg, ‘Harmonic Теп- 
dencies of the Hexachord', Journal of Music Theory, iii (1959), p. 208; for a comparable 
technique cf. Ernst Kfenek’s contribution to Rufer, op. cit., pp. 188-91. 

2 Vlad writes on both works in The Score, xvii (1956), p. 27; cf. also Gerhard, ‘Develop- 
ments in Twelve-Tone Technique', ibid., p. 61. Contrast the more elaborate derivations of 
time proportions from pitch series by Stockhausen. 

3 Gerhard, “Тһе Contemporary Musical Situation’, The Score, xvi (1956). 

4 СЕ, Moses Pergament, ‘Hilding Rosenberg" Music & Letters, xxviii (1947), p. 249. 

5 Cf, Pergament, ‘Gösta Nystroem" Music & Letters, xxvii (1946), p. 66. 


428 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


change of reputation, and the impact of twelve-note music found no 
repercussions comparable to the transformation of a Dallapiccola or a 
Fortner. Bohuslav Магии (1890-1959), a Czech living in America, made 
a determined and largely successful bid for command of distinctive 
extended structure in his six symphonies; but his language remained 
nationalist, tinged by а belated impressionism or neo-classical 
patterning. 


OLIVIER MESSIAEN 


So far, this account of the middle generation’s development during 
and after the war has been unified by one central though far from 
universal phenomenon—the revitalizing of style by serial thought. This 
may seem to invalidate the contention? that the philosophical speculation 
which brought serialism into being, and the cerebration required in its 
operation, were inseparable from a peculiarly Germanic musical ethos. 
Yet one country seems to lend some credibility to the contention: 
France produced no composer of distinction in direct succession to 
Schoenberg’s ideas and technique, and vociferously rejected the 
importation of these into her music. In the new serialism as expounded 
by Boulez, not only did the letter of the old technique become a minor 
and inconstant observance, but values deriving from a specifically 
French tradition were consciously asserted. 

The assimilation of principles originating outside France into a style 
of exclusively native tone is a familiar feature in her musical history. 
But the ability to maintain a character has no virtue in itself, and of the 
oldest French generation at work in this period, it may be said that its 
musical character, if unmistakable, was also weak. The only French 
member of Cocteau’s ill-assorted ‘Six’ to reveal any capacity for 
renewal was Poulenc,? whose two operas, Les Mamelles de Tirésias 
(Tiresias’s Breasts) (1944) and Dialogues des Carmélites (1956), remark- 
ably sustain satirical and lyrical moods familiar from his songs. If the 
piety which rather heavily oppresses Carmélites stems from nineteenth- 
century operatic models, Poulenc’s natural affinity with Gounod? 
ensures that it never becomes a pose. Such music about saintliness is 
acceptable in the theatre but it has little to do with religious music. In 
Olivier Messiaen (b. 1908) France has the only important contemporary 
composer whose religious belief is central to his art. 

1Cf, Milo’ Šafránek, Bohuslav Martini (Prague, 1961; Eng. ed., London, 1964), Peter 
Evans, ‘Martina the Symphonist’, Tempo, lv (1960), p. 19. John Clapham, *Martinü's Instru- 
mental Style’, Music Review, xxiv (1963), р. 158, and Harry Halbreich, Bohuslav Martini: 
Werkvergeichnis, Dokumentation, Biographie (Zürich, 1968). 


СЁ р: 380% ? Cf. Henri Hell, Francis Рошенс (Paris, 1958; London, 1959). 
4 Cf. David Drew, ‘Modern French Music’ in Hartog, op. cit., p. 264. 


OLIVIER MESSIAEN 429 


His association with Jolivet, Baudrier, and Lesur in 1936 under the 
title ‘La Jeune France’ was merely a further declaration against the 
smart, heartless music of the day such as he had already made in his 
early works. Certainly he sought no stimulus in the alliance; for this he 
found spiritually in a symbolical interpretation of Catholic dogma and 
technically in his studies of rhythm (especially Hindu), of plainsong, and 
of bird-song. In 1940 his freedom from the preoccupations of his 
contemporaries was clear, but his significance for a younger generation 
was still unsuspected. Now that this is generally acknowledged, it too 
often provides an excuse for ignoring his own erratic but considerable 
creative achievement.! 

Mode, the prescription of limits within which the music may move 
freely, is fundamental to Messiaen's thought and a concept distinct 
from that of series, which determines the order of movement. By span- 
ning the octave with regular patterns of pitch intervals (see Ex. 195) 
modes are obtained ‘of limited transposition’ (since the patterns repeat) 
which offer a variety of gravitational possibilities in place of the single 
point to which asymmetrical structure inevitably relates the old scales: 


Ex. 195 
Examples of ‘Modes of limited transposition’ 


ero = possible 
= possible 


Е 


€3—-5 transpositions 
possible 
Рад 
5 transpositions 
possible 
Sere eee eee eal 


Within one of these modes tonality may remain elusive; and when two 
modes are at work, the texture may become fully chromatic and tonal 
feeling be dependent on melodic emphases. But since the selection with- 
in a given mode is left to the composer, he is equally at liberty to dwell 
on those notes which form conventional successions and simultaneities. 
Messiaen’s style in fact embraces both extremes, but he has sometimes 


1 The outstanding description and critical estimate of Messiaen’s work is by David Drew: 
*Messiaen—a provisional study’, The Score, x, xiii, xiv (1954-5). 


29 


430 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


irritated his disciples by his fondness for the second,! much as Hinde- 
mith favoured a comparatively small proportion of the possibilities 
latent in his own theory of progression. No such theory is implied by 
modality; Messiaen is dependent on his ear for the regulation of 
harmony. But like Stravinsky and Debussy, who also jettisoned the 
progressive mechanism, he is intent on the aural phenomenon per se 
and, like them, both scrupulous and imaginative in its construction. 

If pitch-modality seems at once too defined and too imprecise to be 
regularly adopted by other composers (though it can be relevant to the 
structure of twelve-note rows), Messiaen's rhythmic principles demand 
consideration as leading to the most radical innovations in this branch 
of Western musical thought since the seventeenth-century triumph of 
the dance as background measurement. Treating rhythm as an absolute 
phenomenon separable from other considerations, he demands acute 
perception of the length of each note as an entity, reckoned as a multiple 
of a very small basic unit (present or imagined) rather than as a fraction 
of a palpable higher unit of time.? Thus, the Hindu rhythm at Ex. 
196(1), one of many such patterns in the percussion writing of Oiseaux 


Ex. 196 
ә тера ыы 
ттш II MM E E 


17 elements, totalling 13 x J 


А 
x Me > > а > 
ү) 2 


“2/2. 1. 2; 


х3. 


exotiques (1956), is not felt as a complex syncopation in the 2/4 measure 
(used for notational convenience) but as the proportions 3.2.3.5.5.2.2. 
In his own rhythms Messiaen avoids the emergence of a background 
metre, often by adroit use of small added values: see e.g. the first dotted 
note in Ex. 196(ii). To amplify such rhythms into structure—repetition 


! Boulez's objection both to Messiaen's F sharp major penchant and to the self-deception 
of disguising it as modality is voiced in Antoine Goléa, Rencontres avec Pierre Boulez 
(Paris, 1958), p. 160. 

2 Cf. Preface to Quatuor pour la Fin du Temps (1941) and, for the most detailed exposition 
of Messiaen's theories, Technique de mon langage musical (Paris, 1944). 


OLIVIER MESSIAEN 431 


in toto is the simplest method: Ex. 196(ii) is the rhythmic ostinato of the 
Quartet's first movement. As in Machaut's isorhythmic technique, 
repetitions may preserve the proportions in different units; Messiaen's 
augmentations and diminutions are not restricted to classical two- 
multiples—see Ex. 196(iii) from the Quartet’s sixth movement.! Rhyth- 
mic canon is а subtler repetition that produces cross-rhythm, and 
Messiaen attaches particular significance to 'rhythmes non rétro- 
gradables' which reverse from a central value. 

Rhythmic ostinato exemplifies Messiaen's view of his material as 
something formed in essence before the process of composition begins; 
ideas may be repeated literally, decorated profusely, or placed together 
in varying relations of time and space. Borrowing again from medieval 
technique, he may choose a melodic-harmonic color and a rhythmic 
talea of a different number of constituents (e.g. Ex. 196(ii) of seventeen 
rhythmic elements is set against a harmonic ostinato of twenty-nine 
chords)? and let their interaction produce constant change from what is 
immutable. Drew points out a still more rigid automatism in the piano 
piece L'Échange (from Vingt Regards): by quoting two bars and a brief 
key to procedure he is able to account for the following twenty-two 
bars.? Twelve-note serialism is no longer commonly held to be a 
Procrustean constraint on the composer's freedom of choice, but in the 
mechanisms of Messiaen's most painfully contrived movements may be 
seen the first surrender to the hazards of inadequately considered pre- 
determination: Webern discovers musical potentialities while Messiaen 
rotates statistical possibilities. The efforts of his pupils to synthesize the 
practices of these two explorers attracted even Messiaen himself into 
a position more radical than he felt willing to maintain; and it would be 
unjust to represent him solely by the works of those years, or to suggest 
that crude formulas dictate the course of his entire output. During the 
troubled times of the forties, his certainty of direction had no parallel 
among his own generation. 

Though the instrumental colouring (determined by resources avail- 
able in a prison-camp, 1941) makes the eight-movement Quatuor pour 
la Fin du Temps wearisome as a whole, it shows Messiaen's style 
fluctuating between essentially conventional harmonic forms (see the 
two gentle ‘Louanges’ in E major); and the wholly original super- 
imposition of blackbird and nightingale flourishes on a framework of 
interacting ostinati (Liturgie de cristal’). But he may combine such 

1 Cf. Dallapiccola's practice in the Canti di liberazione, Ex. 193 on p. 422. 

2 Messiaen's addiction to prime numbers is more systematically revealed in e.g. Меитез 
rhythmiques (1950), the third element of which is a symmetrical rhythmic pattern expanding 


on repetition from a total of 41 semiquavers to 43, 47, and 53. 
? David Drew, The Score, xiv (1955), рр. 46-47. 


432 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Modéré, joyeux (d= 120) 


Ex. 197 бол 2------------------------------------------ 
= Ч : 
te be же жей е е ка 
пън Pul (> ее xw Shee x peh 
A a Ay 4 10 H ы ІП Е AEE UA 
63 2 ен L oL d mcs е = 
ыо „Р carillonnant 


più 


с» 
БЫ s.d secs 2 ТЫЗ NC PPM 
Modéré, joyeux (2 12) 2d Pa. SP. Bd, Pw 
= m = > d > 
b = с se 4 
f) che ple ba НА == DH e 4 s 1 
Enim EE лае EE ии te 
Гол arri seem ДА ЕЕ Lem 
iy Sf сиізге, sonore Жыр NND та 
Yt) ЖЕСІ БЕЙНЕ 
Fa с | F Би: = = ЕЕ Дена 
: oO 
=> 
DJ 


г СИ Uy сее еее 


к" „ш 
ӨПТ кеселі = as | 
D cud с=з C as ПЫШ есет 
= мы 


о 
Sh Фаза аа а о. 
я F e “т => = 
227 Shay. | #2: 
Eig са НЕ 


extremes: in Ше finale of Visions de ГАтеп (1943),1 a rhythmic complex 
(cf. Ex. 196(ii) used again in this later work) in canon at narrowing time 
intervals rotates three-note groups with chromatic organum thickenings, 
creating a maze of sound (Ex. 197, piano I) which enhances a chorale? as 
1 Cf, also Drew's discussion of Атеп des Étoiles, ibid., pp. 41-44. 
2 The fundamentally unchanged character of Messiaen's procedures in his later style may 


be seen by comparing with this the addition in Пе de Feu 2 of the free initial theme to the 
last pair of the serial permutations discussed on p. 440. 


OLIVIER MESSIAEN 433 


square and monotonous in itself as the feeblest effusions of the French 
organ school (Ex. 197, piano II). The circular key scheme (A-C sharp- 
F natural-A: a break with classical precedent rather than principle) of 
the chorale verses is literally repeated in the third of the Petites Liturgies 
de la Présence Divine (1944). This exceptionally direct work did much 
to establish the composer's reputation. А setting for unison female 
choir of his own text, it achieves dramatic compulsion from effects as 
simple as the clangorous reiteration without harmony of a springing 
pentatonic line. The scoring is for celesta, vibraphone, ondes martenot, 
piano, percussion, and strings, and demands a prescribed spatial 
arrangement of the forces. These anticipations of Boulez extend even 
to the use of the vibraphone to sustain vocal notes. 

In the first Liturgie the piano's bird-song introduces atonal chromatic 
bravura into a work which is unequivocally in A. Chromaticism of this 
kind, latent in Messiaen's modal theory, penetrates still further the 
texture of Vingt Regards sur l'Enfant Jésus (1944) and of the Turangalila- 
Symphonie (1946-8).2 This work looks ahead in such details as the 
association of a twelve-note succession with a scale of diminishing 
rhythmic values;? but its true importance is that of a vast compendium 
of Messiaen's musical language at an unusually well sustained level of 
invention. Though rhythmic counterpoints are superimposed in un- 
precedented profusion, and orchestral colourings are kaleidoscopically 
varied, neither thematicism nor tonality is abandoned.* Five years later, 
Messiaen returned to orchestral composition with a piano concerto, 
Le Réveil des Oiseaux, that emphasized his need to escape from the 
ascetic restrictive schemes of some intervening works. In Oiseaux 
exotiques (1956), a fantasy for piano, wind, and percussion, the com- 
poser continued to indulge, to the virtual exclusion of other material, 
his interest in Hindu and Greek rhythms (twenty patterns) and his gift 
for imaginative evocations of bird-song (forty-seven species). This 
enthusiasm has persisted in his later work, and has once again isolated 
the composer from the main forward movement of European music. 
As a result he is no longer burdened by a crippling sense of responsibility 
and can revert to a characteristic textural luxuriance. Oiseaux exotiques 
15 in fact no pot-pourri but, for Messiaen, a remarkably strong arch- 
shaped movement in which the most flamboyant bird calls act as ritor- 
nelli. The texture is accumulated from the counterpoint of numerous 


1 Cf. Boulez, Improvisation sur Mallarmé no. 1. 

2 Cf. Drew, The Score, xiv, pp. 50-56 and Leonard Burkat, Musical Quarterly, xxxvi, 
(1950), p. 259. 

3 Cf. Drew, The Score, x, p. 48. 

* See p. 45, n. 1. The F sharp movement to which Boulez objected was incorporated in 
Turangalila. 


434 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


calls, each having a fixed repertory of pitches, not necessarily adjacent, 
subject to rhythmic manipulation: 


Ex. 198 
Un peu vif = 132) 2 
- P РЕЧ 


(бга) 12, ; be 
Piccolo - т == EE г 


(Lesser Green 
Leafbird) 


Flute; Oboe 
2 Clarinets 
(Baltimore 

Oriole) 


Glockenspicl 
(Redbilled 
Mesia) 


Xylophone 
(California 
"Thrasher) 


This type of modal limitation had been systematized, together with 
other rationalizations of Messiaen's practice, in the piano study Mode 
de valeurs et d'intensités (1949), a work which must be discussed within 
the context of another generation. 


THE NEW SERIALISM 


In the generation that was learning its craft as the war ended there is 
disagreement as to the historical relationship of their musical achieve- 
ments. Luigi Nono has affirmed his willingness to be judged against the 
background of a continuing historical process, but Stockhausen's 


THE NEW SERIALISM 435 


apologist Dieter Schnebel! has argued that the new music implies the 
disappearance of such a tradition. Certainly the historical inevitability 
of these specific developments, though persuasively expounded by their 
apologists, is very debatable. It is clear that their intellectual premises 
came naturally to a sceptical generation which demanded explanations 
for creative procedure that should satisfy the mind as well as the 
sensibilities. To suggest that the particular equipment which has been 
assembled was partly due to a fortuitous interaction of circumstances is 
merely to recognize one of many such recoveries after crucial moments 
in music's history. Musical meanings are conveyed in a man-made 
language, and there are no a priori grounds for assuming that the 
language now evolving, bv a perplexingly devious but ultimately 
empirical process, cannot communicate the aesthetic experience, even 
if its apparent remoteness from natural analogies may delay or prevent 
any popular acceptance. 

Most composers of this generation served at least a brief apprentice- 
ship in orthodox twelve-note techniques. The published expositions of 
the works of the Viennese School by the Polish-French theorist and 
composer René Leibowitz? were among their first guides in a field 
which many of their elders were just beginning to explore, though 
Martin and Dallapiccola would have been able to initiate their sometime 
pupils Stockhausen and Berio into personal interpretations of dodeca- 
phony. At a time when French musicians were still overcome by 
Leibowitz's revelations (in performances as well as commentaries), and 
when there seemed high hopes of a French school of twelve-note 
composers,? Boulez enlisted as his pupil, already equipped with skills 
derived from Messiaen's teaching. But the most momentous revelation 
of serial composition came with Leibowitz's work at the Darmstadt 
Summer School in 1948 and 1949. 

The Kranichstein Institute, founded at Darmstadt in 1946 due to the 
visionary zeal of one man, Wolfgang Steinecke, was intended first as a 
centre where young German composers could study together and hear 
the whole tract of contemporary music which had been denied them by 
Nazi proscription.* Their first mentor, Fortner, was still unversed in 
serial practice, but could demonstrate lessons he had learnt from 
Stravinsky and Hindemith; and in the following year Hindemith him- 
self directed the course. So far this was only to reveal the true origin, 


1 See Luigi Nono, “Тһе Historical Reality of Music Today' (a Darmstadt lecture of 1959) 
in The Score, xxvii (1960); Dieter Schnebel, ‘Karlheinz Stockhausen’, Die Reihe, iv, p. 121. 

2 René Leibowitz, Schönberg et son école (Paris, 1946); Introduction à la musique de douze 
sons (Paris, 1949). 

3 Cf. Leibowitz on the work of his pupil, André Casanova, in Music Survey, П (iii) (1950). 

4 Cf. Antoine Goléa, Rencontres avec Pierre Boulez (Paris, 1958), рр. 67-80. 


436 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


immeasurably more powerful than its miserable progeny, of the 
diluted pseudo-Hindemith of the familiar Spielmusik style. But 
Leibowitz's courses, together with the admission of foreign students 
from 1949, began the transformation of Darmstadt's scope: the aural 
experience of an almost unsuspected body of rich creative achievement 
and the discovery of reassuring technical bases on which they could 
build united the most talented (and many others) of a whole European 
generation. Their pursuit of that style which should satisfy a need for 
the intellectual underpinning of the creative impulse led them, as 
successive courses and private study made their models more familiar, 
to find this most unambiguously in Webern's music. Thus Darmstadt 
became the driving force behind the first consciously international 
movement in the music of our century. 

At so close a range in time it is not possible to determine this century's 
most decisive moment of change.! The new serialism was near enough to 
Schoenberg to stress all that it had rejected of his methods and rhetoric,? 
but it is from Schoenberg that we must trace perpetual variation, the 
systematic unfolding of musical properties (with him limited to the 
pitch content) and as a corollary the association of sound and number. 
Beyond this he chose to rely on inherited formal constituents (expres- 
sive motive, phrase, and paragraph) for proportions in time, and on the 
ability of the row to construct proportions which should recall the 
traditional spanning of musical space; many composers of the middle 
generation devised individual means by which to sustain the same 
compromise. The attraction which the young generation found in 
Webern's music was its new expressive forms, rooted in the clarity and 
correspondence of intervallic and durational proportions themselves,? 
a demonstrably orderly arrangement of sonorous phenomena yielding 
an undemonstrable but compelling beauty. It has often been pointed 
out* that Webern's adoption of twelve-note rows created far less of a 
gulf in his work than had always existed between this and Schoenberg's; 
but it did enable him to achieve the perfect equilibrium of his later 
works by imposing on one idea, the construction of correspondences 
measured by interval, the constraint of another, serial propriety. By 
reducing the dominating status of the Schoenbergian thematic concept 
of the row, Webern was able to break free from the vast chain of conse- 
quences, essentially unrelated to the row as an autonomous proposition, 


1 See e.g. Th. W. Adorno, ‘Modern Music is growing old’, The Score, xviii (1956), p. 18, 
and the reply by H.-K. Metzger in Die Reihe, iv, Eng. ed. p. 63; also the essay by W.-E. 
von Lewinski in the same vol. p. 1, and Zillig, op. cit., p. 199. 

2 Cf. Pierre Boulez, ‘Schoenberg is dead’, The Score, vi (1952), р. 18. 

3 СГ. the analytical essays in ‘Anton Webern’, Die Reihe, ii. 

* E.g. by Pousseur in Die Reihe, ii, p. 51. 


THE NEW SERIALISM 437 


and to order connexions—and ultimately structures—according to its 
inner potentialities. Since these prove to be virtually limitless, his 
characteristic row already demonstrates symmetry within its segments, 
so that the work makes audible ‘horizontal, vertical and “Фагопа!” 
relationships'! stemming from a three- or four-note cell. These are the 
objects of the listener's attention. Although coincidence produces 
‘harmony’ and succession ‘melody’, these qualities are not allowed to 
take on determining functions which, by re-introducing a degree of 
predictable movement, would divert his attention from the innate 
quality (which includes the duration) of the sounding moment. 

Though serialism of all twelve pitches is one of the restrictions which 
brace Webern's style, it does not serve to guarantee the maximum 
separation of identical (or octave-related) notes. Indeed the close 
proximity of a note common to superimposed or adjacent serial forms 
may be pointed,? yet our hearing is so directed to the quality of the 
intervals, their forming and dissolving connexions, often across a wide 
total span, that such a relation is in no danger of establishing a con- 
ventional polarity. And as Webern's basic pitch-material could be 
reduced to the small cell, later composers have not felt bound to work 
with the complete pitch series.? Serial principle orders succession but 
does not of itself specify the number or nature of elements—a function, 
as was noted earlier, of mode. The new serialism sprang from a detailed 
consideration of both concepts, first brought face to face by com- 
posers who had learnt from dodecaphony and from Messiaen's methods 
(Boulez and Stockhausen, and—through Messiaen’s teaching at 
Darmstadt—a far wider circle, including Nono). Messiaen's treatment 
of rhythmic proportions had demanded a large number of durations 
relatable by number, so that a rhythmic mode (a repository of different 
values) was already part of his equipment. Indeed he had already applied 
to this the principle of series in the use of rhythmic ostinato (cf. row 
repetition) and repetitions in various diminution and augmentation 
ratios (cf. row transposition), and in the importance he attached to 
retrograde rhythm. By contrast, in the late works of Webern could be 
found extremely subtle rhythmic relations,* including precisely timed 
silence as a vital contrapuntal element;? but despite ап aural detachment 
from classical metre? and internal symmetries that prevented classical 

1 H.-K. Metzger, Die Reihe, i (Eng. ed.), p. 43. 

? Cf. Christian Wolff, Die Reihe, ii, p. 62. 

* The developments which led to the abandonment of Schoenberg's principle are dis- 
cussed by Gyórgy Ligeti, ‘Wandlungen der musikalischen Form’, Die Reihe, vii, p. 5. 

* E.g. the rhythm of connexions discussed by Pousseur, loc. cit., p. 59. 

5 Cf. Pierre Boulez, Die Reihe, ii, p. 40. 


6 Webern himself, however, appears to have heard his music tensed against such a 
background. Cf. Peter Stadlen, *Serialism Reconsidered', Тйе Score, xxii (1958), p. 15. 


438 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


flow, few signs of a treatment of duration so objectified as to be 
accommodated into serial logic. 

The objective scrutiny of both pitch and time as musical properties 
that could be systematically ordered was turned, by a natural extension 
of the new attitude, on to other properties. Tone colour, once bound to 
the sense of the melodic phrase, had been fragmented in the changing 
chord of Schoenberg's Op. 16 pieces, and Webern's orchestration 
naturally pointed his construction from intervallic cells. Serialization 
instead of free variation of colour was a simple enough step, and as a 
substitute in media of limited colour (e.g. the piano) a serialization of 
methods of attack was thought plausible, thus rationalizing the tendency 
of modern music to make increasingly fine distinctions in this field. 
Webern's wide range of dynamic values, regarded as properties of the 
individual notes, suggested yet another serial ordering. 

In visualizing the perpetual variation, according to a predetermined 
code, of each factor contributing to the distinctive quality of the musical 
note, the new school sought to atone for the extreme predominance 
accorded to pitch in Western tradition; their music should reflect the 
proportions of the series in a constant interplay of all the elements. For 
а just appreciation of the fluctuating activity within such ‘global forms’, 
the listener, brought up on precise recognition of pitch and compre- 
hension of rhythm by the mnemonic of metre, must acquire entirely 
new powers of discrimination. Apart from the acquisition of similar 
powers, the composer's first concern must be with a means of determin- 
ing the mutual relationship of his serial codes. Messiaen, already 
practised in simultaneously unfolding schemes of pitch and time, 
postulated a rudimentary solution in his piano study, Mode de valeurs 
et d'intensités, significantly inscribed ‘Darmstadt, 1949’—5ее Ex. 
199(1). True to his early principles, this uses mode, not series, but now 


Ex. 199 
y Mode, Division I (top register) 
DU тт ы е о пні ie а ! loco 
Tis 
$5 phe be е В oe MeL 
POP PEP ff f mf ff mf 


= дет 
4 = Durational values 1 to 12 (x Ж); 
E : = - = їп Eu II and III the units 
e D and. 2 respectively. 


THE NEW SERIALISM 439 


its 


(Div. I) 
8 


Gi) 


Piano 


Sf mf P 


(Div. Ш) 


prescribes limits in all four fields of pitch, duration, attack, and 
dynamics. His pitch mode fixes the register of all twelve notes across a 
wide range (see Ex. 199(1); two similar divisions correspond to the 
other staves and virtually cover the piano's compass) and to each of 
these is attached,! once for all, a specific attack drawn from a set of 
twelve, a dynamic level drawn from seven, and a duration drawn from a 
‘chromatic’ range of one to twelve values—4^, J and J^ being the units 
of the three planes. 

The rigid constraints of this piece (which incidentally give an almost 
motivic character, or at least a polarizing effect, to certain exposed pitch 
recurrences) do not contribute to its progress, whereas Messiaen's old 
interacting ostinati solved this problem at least. In another study of 
this set, Ле de Feu 2, he again sets up a fixed association of pitch and 
duration (though now entirely 'scalic': B J^ B flat J A J. «с. to C d. 
see Ех. 200(i)) but reduces the earlier, utterly impracticable differences 


1 A similar conjunction, but with a smaller range of constituents, had been attempted in 
Cantéyodjayá (1948), p. 8. 


440 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Ex. 200 
(i): 
Pitch-duration mode 
a2 15 10 etc.) 


т 6 s-5 9 4 төр зе 2 САБИ 


(7 6 6-5 9 4:0 и 7 E. 
10 4 3 9-11 75 2, Эри see (iii) 
255 :8 1? do 9. , 8" 4л ди 

6 9 3.12 1 MN а 8 Фреш 

ап в ие 3 10 40 6] и 
512 3^1 104 9) 6 2 at Ы 


Piano 


PIERRE BOULEZ 441 


in attack to four types and the dynamics to five. Messiaen reconciles 
here the difference between the constantly changing contour of modality 
and the established successions of linear serialism by the principle of 
permutation of all twelve notes, thus establishing at the same time a 
progressive mechanism. His row is the familiar pattern of semitones 
radiating alternately from a central note, see Ex. 200(1), and the 
permutation used applies the identical process (1.e. 7.6.8.5.9.4.10.3.11.2. 
12.1); а few lines of the table are shown at Ex. 200(1), from which the 
reader can reconstruct the course of those sections of the work based 
on this scheme by analogy with the first, Ex. 200(iii), noting that succes- 
sive permutations are coupled in treble and bass. 

Messiaen's work has been used to demonstrate simply an approach 
to organization of material which was also being developed by his 
pupils at this time. Their aim of subjecting all the elements which 
contribute to a musical texture (the term parameters was adopted) 
to serial procedure was extended into a structural determinant by per- 
mutation of the series according to the dictates of a super-series. These 
general principles were given a new interpretation in almost every work; 
for when so much follows from pre-compositional decisions (even if 
relaxations be permitted in their acceptance) they become an important 
field for the exercise of the composer's creative individuality. That this 
field may prove irksomely narrow is to be deduced not so much from 
the superficial impression of similarity listeners have found in multiple- 
serial structures—since the listening technique which these demand is 
both formidable and unfamiliar—as from many composers' subsequent 
abandonment of this extreme position. It will be profitable at this point 
to consider more specifically the work of some prominent representa- 
tives of the new music. 


PIERRE BOULEZ 


Musical history would not have immediately led us to suppose that 
France would provide the two most celebrated documents in the early 
search for strict control of every musical element —Messiaen's Mode de 
valeurs et d'intensités and Boulez's Structures. The pseudo-polyphony 
of the French Baroque, Berlioz's freedom from the tyranny of conven- 
tional progression and phrasing, and Debussy's reliance on a superb 
harmonic and formal instinct іп abandoning orthodox tonality, 
suggest a native antipathy towards rigidity of procedure. Yet if the 
demonstrable organization of German music is often absent, the 


1 This pattern recurs in the Livre d'Orgue (1953). It is also a schematic serial basis much 
favoured by Luigi Nono; cf. p. 476. 


442 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


characteristically French fantasy is governed by a meticulous feeling for 
order. Messiaen's fitful genius is somewhat exceptional in often needing 
the help of the controlling mechanisms noted earlier; but it is significant 
that, even in the Quatre études de rhythme (which include Mode and 
Ile de Feu 2), these appear in conjunction with freely rhapsodic ideas. 
If Pierre Boulez (b. 1925) was driven by intellectual curiosity to discover, 
in incomparably more precise detail, the consequences of a will to 
establish order in each dimension of his music, the powerful quality of 
his imagination has offset these statistical preoccupations.! 

The more rigidly inhibited a piece is, the more satisfyingly account- 
able it becomes to the analyst, until he may even arrive at the position 
of being able to point out ‘errors’ (as distinct from infelicities or mis- 
judgements) in the composing process. When analysis has once achieved 
such a status, it becomes difficult to recall its essentially subsidiary 
relevance to the listening experience, and a work's notoriety may come 
to depend more on the analyst's labours than on the listener's. This is 
not to decry Boulez's renowned essays in total serialization, but to stress 
that their place in his creative achievement has proved to be less than 
their importance for his development. Some of the works which pre- 
ceded them, though disowned at times by Boulez, give a clearer picture 
of the composer's individuality, and it is a picture which more recent 
works have brought into still sharper focus. 

After the Psalmodies for piano (1945), written before he knew any- 
thing of serialism, Boulez mastered orthodox row-manipulation under 
Leibowitz swiftly enough to explore beyond it in his next works. The 
1946 Flute Sonatina uses more elaborate permutations of fully dodeca- 
phonic thematic material than were recognized in Schoenberg's 
practice; but the passages of development adapt Webern's structure 
from much smaller cells, of a few intervals, producing an athematic 
texture in which attention can be concentrated on the minutiae of the 
sounding fabric—notably the subtle balance between rational and 
irrational elements in its rhythmic cells.? Two years later, the Second 
Piano Sonata established a composing process in which the plotting 
of rhythmic relations seems to be given precedence, and the attendant 
pitches develop intervallic cells without reference to any ideal twelve- 
note basic shape: 


1 Among Boulez’s many writings, Penser la musique aujourd'hui (Paris, 1964; English 
translation, Boulez on Music Today, London, 1971), gives the most impressive account 
of the interplay between intellect and imagination that characterizes his work. 

2 Boulez's analysis in Polyphonie (1948), cited in Goléa, op cit., p. 51. 


PIERRE BOULEZ 443 
Plus vif 
^ 


Tempo Г. extrément rapide IN 
— i р-е 
paid 7 f = | 


444 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Extreme dynamic changes help to convey the explosive fury charac- 
teristic of early Boulez (and a hidden menace powerfully sensed in some 
later, more suave, contexts), but they still progress here according to 
‘expressive’ rather than objectified or purely statistical dictates. Indeed 
the demonstrative rhetoric (and, one might add, the length) of this work 
is so far removed from Webern's highly-charged reticence that the 
resulting sound gives no idea of the debt in interval treatment.! Aggres- 
sive timbre and splintered textures dominate the instrumental sonority, 
but the contrasting quiet passages are of an exacting complexity. In the 
Livre pour Quatuor, which includes ideas reworked from the Sonata,” 
Boulez found a more congenial medium for a true counterpoint of 
dynamic values and for the exercise of an easier lyricism. 

These works provided impressive evidence of a compelling new 
voice capable of more than one inflexion, but they gave no clear clues 
to any symbolic life to be re-lived by the listener. The composer who is 
intent on inculcating a new technique of acute listening may be ill- 
advised to provide too soon any form of parallel to the aural pheno- 
menon. But we may suspect one who constantly refuses to recognize 
the part played by analogy in drawing musical experience from 
patterned sound. (Those who find the material of Die Kunst der Fuge 
truly ‘neutral’ must also find it a very dull work.) Just as Webern's 
songs provide a key to his rarefaction of old expressive symbols, so 
Boulez's vocal music interprets the associations he feels in newer musical 
structures, and, as we saw in Stravinsky's work, this desire to wed 
expressive power to an intricately regulated technique can lead to a 
powerful fusion. Le Marteau sans maítre (The Hammer without a 
Master) has proved a most influential work in spreading the new music, 
but Boulez had already revealed a vivid response to René Char's verse 
in the early cantatas, Le Visage nuptial (1946—51) and Le Soleil des eaux 
(1948/50/58). He regards the singing of a poem as a convention seriously 
weakened by stylized imitation of speech rhythms and intervals; and he 


! Drew, *Modern French Music' in Hartog, op. cit., pp. 292-3. 
2 The Livre has subsequently been re-worked in a version for string orchestra. 


PIERRE BOULEZ 445 


aims instead at communicating the expressive message through the 
musical revaluation, rather than through the making intelligible, of 
texts better ‘understood’ simply by reading.! His wide range of vocal 
effect, even regarded purely musically, is notably free from affectation. 
In the first movement of Le Soleil des eaux the soprano monody has a 
suppleness which, with orchestral scoring of a beguiling clarity (even 
when densely contrapuntal), shows Boulez master of a peculiarly French 
lyricism; and the impetuous choral writing of the second movement 
reaches a natural climax by bursting orthodox musical bounds with 
elemental shouts. As the final bars recede into a limpid chamber-music 
texture, the whole work describes a curve of intensity that is familiar yet 
original. 

In the instrumental works which followed, Boulez was to control such 
curves, here emerging from the interpretation of a poetic experience, by 
a Statistical plan designed to embrace every other feature susceptible to 
analysis. The cantatas already show considerable systematization of 
rhythm, interval succession (including quarter-tone relations in Le 
Visage nuptial), and contrapuntal density, but in elaborating the system 
to the exclusion of emotional parallelism Boulez seemed to embark on 
an entirely new sea. The evocative titles, which related Boulez to the 
line of Messiaen and Debussy, were replaced by the cryptic data of 
Polyphonie X for eighteen instruments (1951) and Structures for two 
pianos (from 1952). The crossing movements which explain the former 
title are not merely of ascending and descending pitch patterns, but of 
rhythmic patterns (Messiaen’s use of a rhythmic ‘chromatic’ scale to- 
gether with its cancrizans in Cantéyodjayá affords a very simple prece- 
dent) and of the governing serial processes; Ligeti’s analysis of Structure 
Ia demonstrates a continuing preoccupation in Boulez’s diagonal 
selections from his permutation tables.? Messiaen's Mode study had 
allowed some freedom of movement, once its fourfold but parallel 
restrictions were set out; but Boulez unflinchingly follows truly serial 
paths in each parameter. The fortuitous coincidences in Polyphonie X of 
pitch, duration, instrumental timbre, and dynamic level? inevitably 
result at times in demands which cannot be accurately realized or which 
repel the ear; and it seems clear that the craftsman of the sensitive 
translucent scoring in Le Soleil des eaux has abdicated momentarily in 
favour of the impassioned yet dispassionate researcher. 


1 Boulez оп words and music in Cahier de la Compagnie Madeleine Renaud-Jean-Louis 
Barrault (May, 1958); cited by Goléa, op. cit., p. 111-13. 

2 Gyorgy Ligeti, ‘Pierre Boulez—Decision and Automatism in Structure la’, Die Reihe, 
iv (Eng. ed., 1960), p. 36. 

3 Boulez on the inevitably approximate nature of dynamic markings: Die Reihe, i, p. 26; 
see also his Penser la musique aujourd'hui, pp. 68-69. 


30 


446 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Boulez has described as ‘totalitarian’ this first essay in total serializa- 
tion.’ His Structures rest on as imposing a network of primary decisions, 
but better considered; and the most vexed problem, that of timbre, is 
replaced by varieties of attack in the single tone colour of two pianos. 
He still writes polyphony but, as Ligeti points out, ‘for all its polyphonic 
stratification of series [the music] is formed from discrete elements of 
pitch and duration: because of the piano’s characteristic attack, the 
acoustic events are concentrated into points that are well defined as to 
frequency and time.’? If we were to rate it no higher, the work would 
remain an invaluable contribution to the new aural training required 
by structures in which each note represents, not a fixed association as 
in Messiaen’s piece, but a unique interaction of independently progress- 
ing serial mechanisms. Both his initial pitch series (but not register—a 
vital field for the composer’s free adjustment) and duration series are 
borrowed from the Messiaen model (cf. Ex. 199), and in the first move- 
ment, the exposition of their simple permutations naturally produces 
sections of identical length (1 ШЕ +2 4- 3... 12 = 78 §s). Each begins 
with the simultaneous impact of the first note in every strand in use, the 
texture varying between one and six strands; and the sectional character 
is further stressed in that one value in the series of dynamics and attacks 
holds good for each strand throughout a pitch/duration combination. 
(Boulez writes of dynamics as “а kind of superstructure with a demon- 
strable function rather than a real factor of structural organization.) 


Ex. 202 


PIANO g mp legato - ж 
5 а 5 
Modéré, presque x 24%, 
А ха ло 
Е 


mano те рио | mop 
РЕ Банк тата x == Й ^ EE 
# ; E > 


1 Goléa, op. cit., p. 139. ? Ligeti, loc. cit., p. 42. 


PIERRE BOULEZ 447 


5 = + 
Baas = Хоя элна Бош] «ГУ ae) eae SSeS 
нин оу — ———— —M— — ы — 


4 b 


Trés Modéré 8р art 
te С 


Ex. 202 shows a complete section of three strands and the first bar of a 
four-strand section. The correspondences across the division show how 
carefully planned is the super-ordering. The three pitch series here are 
clearly a transposition of the retrograde inversion (piano I), and two 
transpositions of the retrograde (piano II); the duration series are an 
‘inversion’ and two ‘basic sets’. But this terminology has no meaning 
apart from the permutation tables, and Boulez set these up by 


448 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


numbering transpositions of the pitch series, but uses them as a 
numerical basis for the other elements too. 

The other two movements of the first book (a second book has 
appeared, and the composer's didactic plan comprises twelve move- 
ments) break away from the jolts—predictable and eventually monoton- 
ous, however varied the intervening patterns—caused by synchronized 
permutations of one simple duration series. Far more forceful 
oppositions can be engendered when duration series move out of 
alignment by using different units, when their larger values are sub- 
divided,! and when the degree of activity is further raised both by a less 
leisurely operation of the dynamic series and, paradoxically, by the 
organization of durations of silence. In the formidably difficult second 
piece, Boulez's command of contrasted piano textures, and of their 
balance across a wide span, betrays a shaping impulse that can now 
impose itself on rigorously ordered material. This is no limp “ореп 
form’ thrown up fortuitously by the machinations of a system: its 
composer could turn with assurance in the following work to a dialectic 
founded in less demonstrable premises. 

Characteristically he reverted to the problem set aside in Structures— 
the ordering of timbre, now according to an ideal of delicacy that was 
not among the preassumptions of Polyphonie X. But he also reverted, in 
Le Marteau sans maítre (1953—7), to his earlier search for a new corre- 
spondence between words and music. The varied methods of vocal 
delivery in Wozzeck (for example), from speech to cantabile line, all 
served to underline an expressive meaning literally or implicitly pre- 
served in the text. Boulez's scale? moves between a rare extreme of 
verbal supremacy (parlando and nominal intelligibility) and musical 
supremacy (vocalized sound merging with the instruments) transferring 
‘meaning’ to a plane on which words are an intrusion. In fact the 
Surrealist texts of Char's poems are strangely illuminated in this process. 
The alto voice is identified so closely with the alto flute and viola that 
there is no radical differentiation when it withdraws entirely (see Ex. 
203), and the instrumental ‘commentaries’ on the poems stand in 
essentially the same relation to them as do the word settings. Yet 
however fundamental these transitions, he has not subjected them to the 
strict rota ordering used by some of his contemporaries nor favoured, 
as they do, a deliberate sterilization of words by phonetic dissection.? 

The popular allure of Le Marteau owes much to an unusual 

1 Subdivision is an antidote to the tendency of large values to attract more attention 


than is serially their due. See Ligeti, ‘Wandlungen der musikalischen Form’, Die Reihe, vii, 


р. 12. 
2 Stockhausen, *Musik und Sprache', Die Reihe, vi, р. 36. 
3 Cf. pp. 480 and 491. 


PIERRE BOULEZ 


Ex. 203 
Assez lent (d = 63) 
Ato Жеген cédez - - - 
Flute E 
(sounding ГР 
notes) 


S 


р 
Xylorimba |Ң — — | 


Vibraphone 


ІІ 
ІШІ 
ІҢ 


Percussion 


June niae suis 

4 eu а са ас елена, 
Guitar [pg Nm camera] m 0: oo 00000 nl 
i= Ninus Ж] 


| 
| 


Was ГРЕЯ 
(ato) Б ae 


ЕЗ 
V LI 

| 
| 
| 
| 
у 
|| 


a b 0] 
уша, Hie ---- E 
[ Bi | 
і Ае) a 1 Е ви] 


(muted) 


450 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Tempo; УЛ Ба poco rall. 
- о ы я 
^ % та ет A 


a mf 
Г т-не Е Е ЕТ е. 
Guitar To — ae EEA sr Е-е eS 


с MAYER Y. p ——— pe c nin A] 
) ASA c3 


м ye — — 
БЕКЕТИ а сазара 


4? А d EN. 
| E) А ми | В +] + S 
мы Чек == pp 
Ге) ы [we dS sNAME ст >] 


PIERRE BOULEZ 


451 


Tempo poco рій lento plus 
poco rall. court long 
() т mf fen 
Fl == —& 7а , 5 E — “Ен 
5-2 
3 Bp 3 3 
Xyl M I ко a 


е == у= Ea 
) ; 
РЕ = | 

Guitar ӨШ 7 5 —] 

B 7n 
Voice Е SEE Ars 

E = ма с 
Те EIU - cheur 2 3 
И Е 1521 6 салса 
е. : 
— ari Jy. pizz. Ж рі22 та 


452 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Ancora più lento 1902 mf — 
= 7 : UE (e " ана 
«Гр K ae j ва (ET) 
н. — 9 7- Еж л 
1 8 Miri 
a" m аа 
Xyl. 6 = 
| 
vib. H = 
р 5 A 
pote yuia 
М D 
Guitar И t 7 an re^ пи лена 
py 7 * 
872 — 
А 2. pp 
Voice ==. Е 
Е — d i ЕБС 
; 3 s'est. tu 6 Э К bm али 
> > 
m = = Ha NG 
Viola Hie te = e=; = 
өн 1—4 | p— te = 


PIERRE BOULEZ 453 


instrumentation that excludes all acerbity in a flickering interplay of 
high sounds. Music which never assaults the ears may be an effective 
narcotic; and the oriental quality of sound together with the undue 
length of some movements can easily deaden the listener's awareness of 
anything more specific than a peculiar fragrance. Boulez has pro- 
nounced his music immune against analysis,! but he cannot mean it to 
be heard in so flaccid a manner: if audiences that find this work appeal- 
ing are to learn to listen discriminatingly to new music in general, they 
will have at least to sense the subtlety of the craftsmanship in Le Marteau 
sans maítre. The variation of timbre, far from a crude pointillisme, 
includes newly imagined effects of merging and emerging colour and of 
‘hocketing’.2 Rhythmic activity, which seems desperate at a first 
hearing,? is varied flexibly between an exact timing of proportioned 
cells (see the three commentaries on Bourreaux de solitude with their 
background of precise percussive reiteration) and a superimposition of 
irrational values which produces a controlled rubato. The systematic 
variation of octave register within the line* conceals yet does not entirely 
dissipate their tight nuclear groupings, but the prevailing impression is 
of an improvisatory fantasy that no longer needs Webern's crystalline 
correspondences. 

As the works of innumerable lesser serialists show, the accumulation 
of restrictions must inevitably produce an increasingly arbitrary effect. 5 
To have created the feeling of freedom within the ordered world of Le 
Marteau was a worthier achievement, and Boulez's characteristic 
reaction was to seek in his next works a controlled relation between such 
freedom and a fundamental strictness of technique. The second 
Improvisation sur Mallarmé is the purest example of this kind, but the 
Third Piano Sonata offers another freedom, an opportunity for the 
player to select his own permutation from serial possibilities of move- 
ment, constitution and succession planned by the composer. This 
element of choice, and therefore of controlled chance, had an honour- 
able French precedent (which Boulez was quick to exploit) in Mallarmé's 
Livre,9 but its application to music was demonstrated most radically 


1 Goléa, op. cit., p. 39, but see his own Penser la musique aujourd'hui (Paris, 1964)» 
for a detailed account of some technical methods. 

2 Robert Craft, ‘Boulez and Stockhausen’, The Score, xxiv (1958), p. 55. 

3 Colin Mason discusses some rhythmic problems of Le Marteau in Music & Letters, 
xxxix (1958), pp. 198-9. 

4 That it is still possible to view the texture as linear polyphony has been made the 
starting point of a criticism of this work as stylized and (by implication) reactionary. See 
André Hodeir, Since Debussy (London, 1961), p. 153. 

5 Ligeti, ‘Wandlungen .. .', Die Reihe, vii, p. 6. 

в Hans Rudolf Zeller, *Mallarmé und das serielle Denken', Die Reihe, vi, p. 5; André 
Boucourechliev, ‘Pli selon pli’, La nouvelle revue française, cx (1961), p. 916; Boulez, 
* “Sonate, que me veux-tu?” ’, Perspectives іп New Music, i, 2 (1963), р. 32. 


454 MUSIC OF THE EUROPEAN'MAINSTREAM: 1940-1960 


by the German, Karlheinz Stockhausen (b. 1928); at this point, where 
Boulez's leadership of the new school was first challenged, it is necessary 
to summarize Stockhausen's earlier work. 


KARLHEINZ STOCKHAUSEN 


Boulez and Stockhausen share skills derived from Messiaen’s 
rhythmic teaching, a fervent admiration for Webern’s music! rather 
than for the Viennese serial principle, and a naturally exuberant 
iconoclasm; and they at first retained enough in common for each to 
act as a valuable stimulus to the other’s inventiveness. But this rivalry, 
industriously promoted by their disciples, never concealed fundamental 
divergences in sonorous ideals. These emerge clearly even in works of 
comparable technical complexity, so that the new music, like that of the 
past, is powerfully coloured by the personality of the composer. 

Stockhausen’s pertinacity in a long and varied training is in illuminat- 
ing contrast to Boulez’s brilliant but impatient apprenticeship. After 
acquiring conservatory disciplines under Frank Martin and Hermann 
Schroeder at Cologne, Stockhausen moved to Paris to study with 
Messiaen and Milhaud. At a time when many composers were sampling 
the desultory charms of musique concréte he abandoned them for a more 
systematic investigation of electro-acoustics, phonetics, and information- 
theory under Meyer-Eppler at Bonn. These studies contributed to the 
command of electronic resources displayed in his work at the Cologne 
Studio,? but the passion for analysis and synthesis which they imply has 
equally marked his instrumental writing. Acoustic research has proved 
as strong a stimulus to the imagination for Stockhausen as literary 
sensibilities have been for Boulez. To this more common stimulus the 
German might appear wholly unresponsive, yet the affecting sounds into 
which the biblical words of Gesang der Jiinglinge (Song of the Youths) 
are transmuted spring as much from a contemplation of the text as from 
the wish to demonstrate a hypothesis. 

The hyphen in Stockhausen’s title Kontra-punkte (1953) underlines 
one technical assumption of this first published work (the earlier 
Kreuzspiel (Cross Play) has subsequently been issued in a revised 
version). But strict pointillisme is not the only relation in which he sets 
the tone colours of ten instruments, reduced progressively to solo piano. 
Already there are examples of an opposite extreme—superimpositions 
of rapid passages which fuse into aurally inextricable entities of a given 
pitch width. Interval correspondences rarely reproduce Webern’s sim- 


! Cf. the tributes to Webern by Boulez and Stockhausen in Die Reihe, ii, (Eng.) pp. 40 
and 37. 
2 Cf. pp. 489-92. 


KARLHEINZ STOCKHAUSEN 455 


plicity but are no less persistent: innumerable patterns are derived by 
octave transposition from the addition of a perfect and an augmented 
fourth (see the last two bars of Ex. 204(1), and in Ex. 204(1) the assembly 
of the chromatic total from these intervals). Permuted dodecaphonic 
successions are used, but a less expected link with Schoenberg is pro- 
vided by occasional weighty ‘chromatic’! harmonic texture, especially in 
the piano, as at Ex. 204(i). Accurate realization of the many irrational 
rhythmic groups is made more hazardous by their dispersal among the 
instruments, yet they remain calculable against an essentially rigid 3/8 
beat. 

With the Klavierstiicke I-IV (1954),? experimental pages revealing 
little of Boulez’s natural pianistic élan, Stockhausen abandoned the 
restrictions of the humanly practicable in his notation of simultaneous 


Ex. 204 | 
Фф = с. 126) 
Mini y Iv uc pue. "iniu О а о И? LS 
т ЕМ бє жы, tel 
= 24 2 = Ее TEN 
Ыы кр сс з m PLE E 
ЕССЕ 
Piano 
Instruments 


lie. retaining the tension of chromatic texture such as Schoenberg's Op. 11, по. 3, 
written when tonality was so recently exorcized; contrast the weightless sound characteristic 
of Webern. 

2 The third piece is analysed by Dieter Schnebel in ‘Karlheinz Stockhausen’, Die Reihe, 
iv, (Eng.) pp. 126-31. 


456 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Piano 


Instr. 3 = = 


457 


KARLHEINZ STOCKHAUSEN 


Wl 


гр 


458 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Piano 


Instr. 


(ii) 
Interval constitution 
of the last two bars 


dynamics—see Ex. 205(i)—and duration-proportions. His demands 
(e.g. the doubly irrational groups in Ex. 205(ii)) are so detailed that it is 
disconcerting to learn that the player, after familiarizing himself with 
their complex relation to a metronomic tempo (as fast as possible), may 
reinterpret them as less taxing proportions in varying tempi. Patient 
study of such notational problems might conceivably lead to a control, 
tauter than that of an impromptu rubato, of the discontinuous tempi 
essential to the structure. But Boulez has written in an implied re- 
ference to these pieces, ‘it is better to substitute an alteration of tempo"! 
and the distinction he then makes between concepts of time appropriate 


* Boulez, ‘At the ends of fruitful land’, Die Reihe, i, (Eng.), рр. 23-24; see also comments 
by Nicholas Ruwet and Henri Pousseur in Die Reihe, vi, pp. 67-68 and 77-78. 


KARLHEINZ STOCKHAUSEN 459 


(= chromatic 
segment 


Piano 


to electronic and instrumental music respectively is fundamental to an 
understanding of Stockhausen's development: 


In the first case, an unchanging tempo within which durations may be sub- 
jected to almost unlimited degrees of variation; in the second a tempo which 
is itself subject to the greatest degree of variation but within which there are 
limits to the degree to which values may be varied.! 


This wide-ranging essay by Boulez abounds in such speculations, often 
brilliantly apposite but not always reflected in his own music. Stock- 
hausen prefers to work out the implications of his ideas in both com- 
posing and theorizing. His next instrumental works, Zeitmasse (Measures 
of time) (1955-6) and Gruppen (Groups) (1955-7), apply the concept of 
fluctuating tempo to a small chamber group and to large orchestral 
forces, while his essay on musical time? has become a required text in 
that study. The five wind players of Zeitmasse, although their parts 
present unprecedented difficulties, are not required constantly to inter- 
relate them on the basis of an identical beat. At times each moves at an 


! Boulez, loc. cit., p. 25. 
? Stockhausen, ‘How time passes . . ., Die Reihe, iii, (Eng.) р. 10. 


460 MUSIC OF THE EUROPEAN.MAINSTREAM: 1940-1960 


even but individual tempo, or several parts engage in an independent 
accelerando or ritardando with or without even tempi elsewhere; or one 
part may rely entirely on cues provided by another. Stockhausen's 
theory of time-fields,! degrees of permissible deviation from mathemati- 
cal precision in the realization of complex rhythmic notation or in 
ensemble performance at tempi without common multiples, rationalizes 
the tendency of post-war composers to break free of the old constraint 
imposed on horizontal movements in time by rigid vertical ‘harmonic’ 
alignment. In Webern interval-patterns vary freely between the succes- 
sive and the simultaneous, but it is only when intervals as objects for 
individual perception have been abandoned in favour of complexes of a 
given note-content (see e.g. bar 2 in Ex. 207(iii)) that the composer may 
claim that discrepancies in the progress-through-time of the total 
sound-constituents no longer impairs intelligibility. In fact, Stockhausen 
ensures that his simultaneous tempi do not lose the proximity he intends 
and can resume precise alignment; he has calculated the superimposi- 
tions with the musical craftsman's awareness of instrumental possi- 
bilities. Action-durations,? such as the time which a competent executant 
will take to play a given passage 'as fast as possible' or the time dictated 
by the requirement “аз slowly as possible, in one breath', provide at 
once a feeling of freedom to the player and a measure of some tolerance 
rather than of utter unpredictability. і 

Clearly, such freedom could not be extended far beyond the confines 
of a small chamber group, and Zeitmasse has become legendary for its 
rehearsal demands. In writing for bigger forces, Stockhausen concen- 
trated on freedom of movement between, rather than within, the tempi 
of three orchestras, each with its conductor, and enhanced this for the 
listener by separating them in space.? Gruppen demonstrates the com- 
poser’s view, prompted by his electro-acoustic researches, of time 
relations as a reflection in the *macro-structure' of those existing in the 
*micro-structure' (i.e. as vibration frequencies). As a pitch doubles its 
value in our conventional temperament after twelve increases by the 
twelfth root of two, he experiments with a comparable ‘chromatic’ 
scale of tempi*—contrast Messiaen's additive scale of durations; see 
pp. 439-40 and 446, and Ex. 206(1) and (ii). 

This scale can be arranged serially, and with three orchestras each may 
have rests in which to prepare the next tempo (see Ex. 206(iii)). Having 
equated tempi with chromatic fundamental pitches, Stockhausen then 
equates constituent rhythmic values with the superimposition of 


! Stockhausen, loc. cit., р. 30. ? Stockhausen, loc. cit., pp. 34-35. 
? Stockhausen, *Musik im Raum', Die Reihe, vi, pp. 61-62. 
* Stockhausen, ‘How time passes . . .’, Die Reihe, iii, (Eng.) р. 21. 


KARLHEINZ STOCKHAUSEN 461 


Ex. 206 


(1) 277.2 311.2 349.2 392.0 4400 493.9 


Vibration frequencies 261.6 293.7 329.7 3700 415.4 466.2 5232 
(per second) 
= pitch е efe 


Gi 
Fundamental phases 636 714 801 899 1009 1123 


(per minute) o= 60 
= tempo 674 756 849 952 1069 120 


(Баг 4) J = 120 end of Group 1 


i Ем 


Tempi of opening 


groups in = 95 
МН Пи = 3 Бы 
(4 represents the beat; II г 
for the manner of its silent beats 


subdivision, see Ex. 207 iii) 


IH 


Group 2 begins =127 (from III) 


ПЕЕ 2 S525 сыга о 


Group 3 begins 


X ЕЕ ee = SSS 


silent beats 


harmonic partials that produces tone-colour, overlaying his funda- 
mental tempo phases—e.g. the semibreves in Ex. 206(ii)—with a variety 
of ‘formants’, fractions which coincide to reinforce the beginning of 
each fundamental phase! (see Ех. 207(ii)). Thus the use of irrational 
values foreshadowed by Webern and favoured by Stockhausen's 
generation, is interpreted as a magnification of the natural phenomenon 
of timbre: ‘rhythm of the sound’ is paralleled аз ‘rhythm of Ше Баг’. 
This brings us no nearer aural perception of the analogy, since a wealth 
of formants? suggests no transformation into a higher unity, though it 
does merge like a wealth of partials into an inextricable sound. Yet 
1Tbid., pp. 16-18. 


2 Ibid., рр. 26-28, gives an explanation of the ‘formant spectra’ which regulate the 
selection of formants sounded. 


31 


462 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Stockhausen's theory does enable him to plot the fluctuations in rhyth- 
mic density so typical of his work, and many transitions in Gruppen 
exploit the territory between apprehension of rhythmic activity as tempi 
and as eventful but opaque, and therefore immobile, sound. 

A similar distinction is made between co-existent instrumental 
textures that merge and those with a high degree of 'permeability'.! 
But Stockhausen's spatial lay-out also permits arresting motions of a 
uniform timbre; a quasi-fugal exposition of nine brass entries is dis- 
tributed in space and its climax-chords dynamically manipulated so as 
to swirl round the orchestras. By a clear scheme of timbre predominances 
and transitions almost two hundred small ‘groups’ of material are 
welded into far longer sequences. Together with organization of tempo 


Ex. 207 


PARTIALS 12 
x2 x3 x4 x5 хб хт x8 хо хю Xll Xi? 4. 
ел LE 


Vu а e 


FUNDAMENTAL 
FREQUENCY 


(i) 


‘FORMANTS? 


FUNDAMENTAL 
PHASE 


* Ligeti, ‘Wandlungen der musikalischen Form’, Die Reihe, vii, pp. 8 and 14. 


KARLHEINZ STOCKHAUSEN 463 
(ш) 4-19 


Flute 


Alto 
Flute 


Тр, 


Holz- 
Trommel 


Trommeln 


Marimba 


Keyed 
Glockenspiel 


Harp 


I 
Vins. Il 


IH 
Vins. IV 


Violas 


>. E c 
Dee normal и 


1 ГЕ 
сае тя те 


mf Tutti senza vibrato D mf 


Chromatic segment Fundamental phase: 4 
(‘band width’) Formants: 2. 3. 4. 5. 6. 7. 8. 9. 10. 


464 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


and timbre a third serialism, of ‘band-widths’, functions no less directly. 
Activity within the groups 15 often no more than a constant flickering of 
many formants between pitches selected by a super-row, but these 
present the utmost diversity between a tight cluster of semitones (see 
e.g. the first bar in Ex. 207(iii) and its activation in the next bar) and a 
punctiliously spaced wide distribution of twelve pitches (occasionally 
more—octaves pass unnoticed at moments of densest activity). As 
many of these distinctive sound patterns remain immediately recogniz- 
able despite reorganization of scoring and rhythms, they provide another 
source of the work’s remarkably integrated effect. The formidable 
theoretical concepts of Gruppen are reflected in diversity of detail con- 
tributing to a form of powerful simplicity. The espressivo of its string 
writing and the furious vigour of its jazz-influenced climax provide the 
closest ties with familiar rhetoric in all Stockhausen, while its frank 
emphasis on the spectacular element natural to orchestral performance 
may yet help to establish this score as a valuable bridge between the 
composer and an uninitiated public. 

Stockhausen’s spatial distribution of orchestral sound logically re- 
solved the contradiction posed by the delicate elaboration of post-war 
serialism and the gargantuan apparatus of the nineteenth-century 
orchestra: when composition is stratified the listener can concentrate on 
local detail or broader relationships. His example was followed by 
many of his contemporaries, including Boulez (in Doubles, performed 
in 1958 but subsequently rewritten), Nono, and Pousseur, and has also 
had important consequences for a natural fusion of electronic and 
instrumental sound. In Carré (1960), Stockhausen essayed spatial 
writing for four choral-orchestral ensembles. This comprehensive score 
has many impressive (and some humorous) moments, but seems to lack 
the dramatic cogency of both Gruppen and Gesang der Jiinglinge. Mean- 
while Stockhausen’s pioneering had attracted at least as much emulation 
in another field. 

Klavierstiick XI (1956) has gained a notoriety incommensurate with 
its absolute worth as a prototype of indeterminacy and the musical 
‘mobile’. (The term was used by analogy with the artistic forms of 
Calder and others, in which spatial relations are variable.) Chance modi- 
fied only the synchronization of the strata in Zeitmasse and Gruppen, but 
in writing for a single performer Stockhausen allowed it to regulate the 
succession and quality of events. Nineteen pieces of piano-texture face 
the player, who chooses their order at random but follows instructions 
in each one as to the manner (i.e. tempo, dynamics and mode of attack) 
in which he shall play his next choice. This provision does ensure con- 


1 Ligeti, loc. cit., p. 8. 


РЕАШЕГХ 


КНАЧЅЕМЅ ZYKLUS, 1959 (see p. 465) 


A page from the score 


KARLHEINZ STOCKHAUSEN 465 


trasts of material, but means that pitch and rhythm relations have been 
composed with no fixed idea of the kind of music they represent. From 
this attitude to one which abandons all concern for the precise notes 
played may appear a small step, and it had already been taken by the 
American, John Cage, and in the ‘graphic notation’ used for works like 
Sylvano Bussotti's Piano Pieces for David Tudor. Having recognized the 
vanity of earlier dreams of absolute control through multiple serialism, 
some composers decided to relinquish what has traditionally been 
regarded as their prime function, the construction of exact sound 
relationships. However his disciples may have profiteered from this 
cheapening of their craft, Stockhausen still retained too much of tradi- 
tional German skills to exercise his aural imagination merely in peri- 
pheral matters and to decline responsibility for the sounding substance. 
His Zyklus (1959),? for an enormous battery of percussion instruments 
controlled by one player, explores a field of sound in which pitch con- 
tinuity is naturally exceptional. Here operations of chance and idio- 
syncratic notation both contribute to a tour de force that admits yet 
directs the improvisatory vehemence associated with these instruments 
in their most developed environments—tribal music and jazz. In the 
Refrain for piano, celesta, and vibraphone (1959), the effect of spon- 
taneity does not exclude a keen regulation of sonorities, and this short 
attractive piece suggested that Stockhausen's use of his talents was not 
to be dictated simply by the principle of novelty at any price. 

Boulez first introduced some freedom of performing procedure in 
works that appeared later than Stockhausen's. Whether he was at least 
as early in conceiving the principle? is less significant than his distinctive 
view of it. In allowing an incredibly large number of possible orders of 
events in Klavierstück XI, Stockhausen shed the artist's responsibility 
to create what he believes to be effective order. Even if we regard 
cumulative rhetoric as dependent on tonality and substitute discon- 
tinuous “ореп” forms, the very existence of music as ап art presupposes 
that some sequences of sound are more satisfying than others; and it 
is a perversion of the serial ideal to insist that we should discover these 
only by dint of working through many less engaging permutations. 

Boulez was repelled by the artistic indifference implied by the in- 
finitely variable mobile and ‘indeterminate’ notations, and in his Third 
Piano Sonata he can be seen weighing the effect of permitted 


1 Stockhausen lectured on *Music and Graphic Arts' at Darmstadt in 1959; see report by 
К.Н. Wörner, Musical Quarterly, xlvi (1960), р. 271. Cf. also *Notation-Interpretation' 
by Cornelius Cardew in Tempo, 58 (1961), p. 21. Cardew was responsible for the ‘realization’ 
of Stockhausen's Carré: see his article in Musical Times, cii (1961), pp. 619 and 698. 

2 See pl. VI. 

? See Goléa, op. cit., pp. 228-9. 


466 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


modifications on structures whose broad design remains within his con- 
trol.! The five ‘formants’ may be placed in various orders, but the massive 
‘Constellation’ or its alternative retrograde version, ‘Constellation- 
Miroir', must remain the centrepiece. Within the movements the inter- 
preter must choose from alternative routes according to principles so 
carefully drawn up that no haphazard presentation of the total material 
is countenanced. In the simplest piece, ‘Antiphonie’, each of the two 
sections may be played in either the original version or in a variation 
printed on the opposite page, a degree of liberty scarcely exceeding that 
of many a Baroque structure. “Тгоре” is published in a spiral-backed 
form that allows a variety of starting points in the cycle of its four 
sections; as one of them (‘Commentaire’) appears twice but is to be 
performed on only one of these occasions, a wider range of possible 
orders emerges. In addition, two of the sections are composed of obli- 
gatory and optional fragments in constant alternation; thus, it is 
unlikely that the enterprising player will ever reproduce the same 
pattern, yet the sequence is always one that Boulez has planned. Simi- 
larly the alternation of blocs (dense harmonic textures) and points in 
‘Constellation’ leaves the player free to choose innumerable different 
itineraries from this ‘map of an unknown city’,? but the composer 
ensures that he does not double in his tracks. The remaining formants, 
‘Strophe’ and ‘Séquence’, have not yet achieved their final shape and, 
given this concept of the sonata as ‘a moving, expanding universe,’ it 
is to be expected that every modification to one movement will tend to 
change the composer’s attitude to those he had thought to have per- 
fected. 

The influence of Mallarmé’s Livre on the flexible shape of this 
sonata has been noted. At least four more works by Boulez already 
performed remained for long, or still exist as, ‘work in progress’, 
due to his view of form as intelligent compilation and to his enthusiasm 
for revision. The most ambitious of these, Pli selon pli for soprano and 
orchestra, is a tribute to Mallarmé built around three Jmprovisations on 
his verses. The impression of the whole work is of the composer’s most 
mature achievement, in which his splenetic violence and lyrical sonority 
contribute not so much to bizarre contrasts as to a comprehensive 
pattern. Boulez has acknowledged his debt to a study of Debussy and 
Ravel’s Mallarmé settings,* and the influence of the former is not far to 


1 Ibid., pp. 229-30; Boulez’s views on chance are expounded in ‘Alea’, La Nouvelle 
Revue Française (November 1957) and Perspectives of New Music, iii, 1 (1964), р. 42. On 
the Third Sonata in its first form see Ligeti, ‘Zur III. Klavier-sonate von Boulez’, Die 
Reihe, v, pp. 38-40; оп a later stage, see Boulez, ‘ Sonate, que me veux-tu?” °, Perspectives 
of New Music, i, 2 (1963), p. 32. 

2 Boulez, loc. cit. 3 Ibid. * Goléa, op. cit., p. 251. 


KARLHEINZ STOCKHAUSEN 467 


seek in the evocative, hovering harmonies. The initial poise of the vocal 
line gives way to ecstatic convolutions of multiple grace-notes (produc- 
ing an inevitably free tempo); and though the orchestra in the final 
*Tombeau' reaches a climax of dense improvisatory activity, it is with 
the impassioned return of the voice that the work's great circle is 
completed. Among the provisions for interpretative liberty built into it 
are alternative circuits of procedure selected by the conductor, and 
durational schemes controlled by the singer's breath (cf. Zeitmasse): in 
Ex. 208, from the second Zmprovisation, all the instrumental groups аге 
played freely between the singer's moves. 


Ex. 208 4, Senza tempo 
-%---------------Б.Қ,Ь>>ББББББ5 
Hes 
ы 
Нагр 
Fra 


Vibraphone [As 
BNE, 


Piano 


р-н ? 
pi * 
= ай 

= T 

p сы 


Р: 

та 

ШЕ въ Не) 
BUILT © 


М2 — Д i i 
Jpp————— . | trés rapide 

Celesta presque aussitot 216—4 Ж> А С ТОР 

2: z< y 


ГИ b 


91% —— Ба 


Voice TA | 
me, fo - 


1 dun seul souffle, aussi lent que possible КЕ REED 


468 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


étouffer toujours sur le temps 


o. ON ae 
ma o a 


p chaque fois p 
comme un 
echo de 


8: 
h 
к= 2 ү е тр 
а == 
sec А: = 
mor ne 
SE SS _ aO [4 
© === — 
tres а = 
2 
Ley: m 
= == == 
Е : i 
P тр i 
y ——— = -ło ы 
Ga === 
con - tre 1а vi - tre 
ta (id) —— m ар 


The general transition of the avant-garde from a fanatical ordering of 
every detail in time to such fluidity has sometimes seemed a уойе face 
explained only by capriciousness or despair; but for Boulez this with- 
drawal of attention from the precise construction of each moment has 
never involved the purely fortuitous, and has encouraged a breadth of 
structure so far without parallel among his contemporaries. 


THE AVANT-GARDE IN ITALY AND ELSEWHERE 


Despite differences of musical temperament that confirm national 
traditions, Boulez and Stockhausen showed obvious parallels, even 


THE AVANT-GARDE IN ITALY AND ELSEWHERE 469 


reciprocity, in their early technical development. A third composer 
whose name was often linked with theirs, Luigi Nono (b. 1924), presents 
a further, and very distinctive, national inflexion, but also a single- 
minded direction of technique towards the immediate expressive pur- 
pose, often of a dramatic urgency. Both Nono and another Venetian, 
Bruno Maderna (1920-73), owe much to the teaching of Hermann 
Scherchen, a conductor and profound student of modern music who 
championed Webern and foresaw his importance for the new generation. 
Maderna acquired professional skill as conductor and composer while 
Nono was still reading law, and could therefore guide his early develop- 
ment. Their attendance at the Darmstadt courses in 1950 founded 
contacts between the European advanced movement and the composers 
of Northern Italy which flourished in the succeeding decade and 
produced some of its most distinguished music. These cultural ties, 
altogether closer than may be traced in Petrassi’s neo-classicism ог 
Dallapiccola’s dodecaphonic allegiance, imply no renunciation of 
national predilections. Nono's Intolleranza 1960 was the first opera of 
the avant-garde, but its passionate involvement in the endless protest 
against oppression recalls not only Dallapiccola but the revolutionary 
Verdi. And Dallapiccola’s concern for lucid sonority and lyrical flow 
are reflected, despite constructional devices far more schematic than 
his, in the works of his sometime pupil, Luciano Berio (b. 1925). 

In basing his Canonic Variations (1950) on a twelve-note row by 
Schoenberg, Nono paid tribute to Viennese expressionism and to 
dodecaphonic principle, both of which he was to transmute in forging 
his mature style. The chamber medium of Polifonica-Monodia-Ritmica 
(1951)! prompted an uncharacteristic sharpness of line, but the Epitaph 
for Federico Garcia Lorca (1951) included Nono’s first attempts to pit 
constructive ingenuity against the expressive compulsion generated by 
a text, to ‘live in a state of tension between rational and emotional 
creative thought'.? The Due Espressioni (1953) are important studies in 
the conversion of Schoenberg’s Klangfarbenmelodie to the ambivalent 
textures, immobile yet quivering throughout with restless activity, that 
have become Nono’s distinguishing sound. Such textures, of constantly 
fluctuating colour, dynamics, spacing and density of events, invite the 
description pointilliste since instruments and even voices frequently 
contribute no more than isolated notes. Here too an ambivalence is 
apparent, for the sense of cohesion is stronger than that of isolation. 
Much of Nono’s art lies in his scrupulous dispersal across many 
registers of notes that are firmly tied together in serial durational 


1 Udo Unger, ‘Luigi Nono’, Die Reihe, iv, (Eng. ed.) p. 5. 
2 Ibid., p. 10. 


470 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


successions (several strata operate at once, based on different sub- 
divisions of a fundamental beat) and by the constant, indeed automatic, 
revolution across the events in time of a pitch-series: see Exx. 209(1) 
and 210(iii). By 1955, many composers had abandoned the fixed twelve- 
note succession as a working principle, but Nono's Incontri (Encounters) 
for twenty-four instruments consists of forty-one reiterations of a row 
and the repetition of the entire structure in reverse. Yet the very auto- 
matism of the procedure reveals how far it is from the privileged thema- 
tic status of the Schoenbergian series: here the succession merely ensures 
an even supply of homogeneous material, while the listener's attention 
is occupied by endlessly new engagements between the same notes due 
to activity in the other parameters and to register changes (see Ex. 
209(1)) Repeated-note flourishes in conflicting durational strata 
generate a conventionally ‘rhythmic’ excitement and make a climactic 
epigram of the central bar, in which the first and last notes of the row 
compete before they exchange roles; but, as with Le Marteau, an 
impression of uniformly attractive or intriguing sound may in fact dull 
response to the changing ‘encounters’ produced by this method. 

In Г canto sospeso (The interrupted song) (1956), variation of or- 
chestral and vocal resources in nine short movements ensures a 
continued freshness of sound. Due to the consistency of colour within 
a choral texture even pointilliste distributions tend to cohere as melodic 
strata, while the solo voices still more forcefully reinstate melodic 
criteria; it is perhaps not only for ease of singing that Nono forms near- 
tonal groupings (overshadowed but not concealed by their contexts) in 
setting texts of such simple but harrowing pathos as these last letters of 
Resistance workers condemned to death; see Ex. 210(1). To use such 
potent literary raw material is to risk disrupting the poise, the distillation 
of the experience, without which music becomes inchoate; the more 
powerful the impact of such lines on the composer, the greater his need 
tó discipline his means of communicating it. Nono has denied Stock- 
hausen's claim that his technique of sharing dissected words among the 
voices (see Ex. 210(ii)) deliberately masks their intelligibility,! but the 


Ex. 209 
(i) 
Грета}  [—— ус соо 
ofa = а retro inv. of b 


[2 онт 
бъз о е Шо = с v = 


* Stockhausen, ‘Musik und Sprache’, Die Reihe, vi, p. 42, and see footnote on p. 44, 


THE AVANT-GARDE IN ITALY AND ELSEWHERE 471 


(ii) 2 rallentando 


Ob. 


Cir. 


Fags. 


Hn. 


To. Ha 
as 


Tbne. 


rallentando 


[e 
Cellos 


D.B. 


(all instruments notated at sounding pitch) 


MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


472 


ра $m 


ti 


= 
ЕЕ 


У} 


й 
С 


5422 


| 
v. Ты 
SII 
| г з 
| В B 


|| 
li 
Це 


к 
Бараа ри 
M — €— 


T 
г 


| 4 
| Sy as 
| | 
Ips + 
сыба ma 7 > 
[NY [| - 


THE AVANT-GARDE IN ITALY AND ELSEWHERE 473 


listener finds that it often disperses their immediacy while elevating their 
meaning, a traditional and unique function of music. Stockhausen’s 
discovery of a serialization of vowel sounds in J canto sospeso may reveal 
more of his own cast of thought than of Nono’s; but the latter’s practice 
of flooding the whole musical fabric with the colour of specific vowels 
(see Ex. 210(iii)(c)) identifies him with a general preoccupation of his 
contemporaries with the sounds of language. 

The technique is essentially that of Incontri; though the straight row 
(Ех. 210(ii)(c); cf. Messiaen's row at Ex. 200(1) ) is not preserved for 
every movement, Ex. 206(ii1) shows a typical unfolding of it across four 
durational strata (units J^, J', J'? and «^— ее Ex. 210(iii)(a)) according 
to a proportional series produced by successive additions—1.2.3.5.8.13;! 
dynamics are also serialized, by a dual ordering.? Indeed, of the works 
of this period which have impressed audiences as putting artifice 
to the service of expressive ardour, // canto sospeso is superficially 
nearest to the unsubtle rigidity of totally predetermined texture. Though 
the composer's choice of register has been noted as the key to the 
beauty of the sound, it 1s his choice of different determining mechanisms 
which is elevated here to a decision involving all his artistry. For the 
dramatic contrasts spring directly from this. The overwhelming menace 
of the wedge of choral sound widening from a vast unison at the words 
*Le porte s'aprono. Eccoli i nostri assassini' (The gates open. There 
are our murderers), or the powerful compassion of the crescendo- 
diminuendo arch in the orchestral interlude surrounded by shimmering 
close spectra of string harmonics, realize designs that are in themselves 
triumphs of the creative spirit. 

Whether Nono surpassed this achievement in essaying a still more 
ambitious and overtly dramaticschemein Intolleranza (1960) is debatable. 
The elaborate symbolism of Angelo Maria Ripellino’s libretto has not 
encouraged as unified a vision, and the note of protest sometimes 
rings shrill; but Nono remained the composer most likely to re-create a 
peculiarly Italian tradition in terms of the new art. Other works 
after П Canto show greater intricacy in applying basically unchanged 
methods. The Cori di Didone (1958) to lines by Giuseppe Ungaretti 
exploit fanwise chromatic movement and syllabic dissection still further, 
yet the level of musical intensity directly reflects the words. Nono has 
declared himself against indeterminacy,? but the possibility of an 
entirely accurate performance of this work seems dubious. And in the 

1See analyses by Stockhausen, loc. cit.; R. Smith Brindle in Musical Quarterly, xlvii 
(1961), p. 247, and Unger, loc. cit., p. 10, all are in fact oversimplified accounts of Nono’s 
procedures. 


? Stockhausen, loc. cit., pp. 48-49. 
? Nono, ‘The Historical Reality of Music Today’, The Score, xxvii (1960), p. 41. 


MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


474 


Bocca chiusa 


ne 


Solo Soprano 


va, 


a quasi chiusa 


Bocc 


ter- 


Pu-mi-da. 


nel 


Күр 
DE 
| na | 22 


| 


Al 
yis 


21 


| ex Да 


Timp. 


THE AVANT-GARDE IN ITALY AND ELSEWHERE 475 


A Ррр 


Eep — наара 


poem fe fue c efe ice — [oj Со ———— | 
aa Е Тый = mEROm 0 c a EMI 
Seer 


(a) Durational 
Series 
1. 2. 3. 5. 8. 13. ( & retrograde) 
operating 
across strata 
of units 


как исак Бы ака СЕРА 
NOS Е 
еее === 


MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


476 


ыы 


p-5 ф=—)5== ES 
ТИИ E Dg E T LER eee 


E 
кеш 


3 Hi» Пе / 


l^ ЩО Ще 
ци 


il 


я! Е. 


Ши 


EXTA ANT OS 


(b) Pitch series 


(c) Pitch series distributed across above durational scheme 


| | E bt 
Ыш" E 
MIN E ME if J 
T A as. | 3 & 
Lk RT al a 
a 7 |( | 
з ы. 
% 
та и 
БІН ПБ ШЕ 
SEN hi 
Ба o.h E. 
| PER DR Ц 
Ш} 
is C та Е | 
Г T у Я 
че © Ц ШЕ 
В ei Е < ( ы 4 


- Ја 


nul 


рі 


VER 
eee 


2 


= 
non 


THE AVANT-GARDE IN ITALY AND ELSEWHERE 4 


ч 
ч 


orchestral stereophony of his Composizione 2 (Diario polacca 1958) 
(Polish Diary 1958) he demands of a vast apparatus (such as Stock- 
hausen devised for the flexible relation of tempi), a precision of timing 
across the four orchestral groups which is vital to his purpose yet 
extravagantly taxing. The opening bars, at Ex. 211 (brass only), show 
comparatively simple entry-delays that move the focal point of the 
sound. 

Nono's recognition began in Germany, and for some years his work 
was scarcely known in his own country. In contrast, Luciano Berio 
played a decisive part in stimulating Italian interest in new music, as 
founder of the Milan electronic studio in 1955, of the concert series 
‘Incontri Musicali' in 1956, and as editor of the journal of that title. In 
his early works he excelled in sensitive orchestral colouring and in 
handling variation-forms easily apprehended as such. This might 


за 


478 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 
Ex. 211 Тре. 


(J = ca.60) a uer 
() [pn И Аи 
= — 
ORCH. Hn. ум: Tons C а pee ees 
n E Ud. 2 A 5 
р пт Но. Г 2 1 
ЗЕ А ee 
Vw 
Tone. || а үч 31 га mf—f Р— тр 
4 [ки тті 1 Im 
(ре — 
ГА 
ORCH. Hn. ж Е 7 —— p um 
(brass) p= mf На. сен ен! 
Ley: z ЕЕ. 
EZ ( = 5 Ееее 
3 Tone. 3 OL g y e ЖИН” JH 
() ТИ Пеева Та 
P [—— pure 
ЕКА 
ORCH. Hn. 22.2. Tone. Uem a sn. И 


ш 
(brass) 


ORCH. 
IV 
(brass) 


Pitch series (cf. Ex. 210 iii, b) 


върне RE = 


suggest a merely decorative talent, but the arrangement of material in 
Omaggio a Joyce! is cumulative, and the five variations of the early 
orchestral piece Nones (from material for an oratorio on Auden's poem) 
are formed into a larger structure. Berio deliberately reinstates here 


1 Cf. p. 492. 


THE AVANT-GARDE IN ITALY AND ELSEWHERE 479 


pitch polarities (their serial superstructure permitting spacious articula- 
tion) and the octave—as a unique phenomenon that can be neither 
surreptitiously overlooked nor systematically excluded. By building on a 
symmetrical row (see Ex. 212) that includes two Ds, he is able, dramati- 
cally yet logically, to expose the open octave as the opposite extreme 


Ex. 212 


retro. inv. of x 


to the density of the total chromatic spectrum.! The middle territory 
includes many familiar chordal formations, and the work has more of 
the Schoenbergian espressivo than is implied by its technical premises 
of multiple serialism; the sudden dissolution of climax-texture to quiet 
string harmony is a gesture much used in later works. In Quaderni 
I (1960), it is almost a cliché, and the widening or narrowing chromatic 
spectrum is a very direct means of controlling emotional tension; the 
orchestral invention may have been stimulated by experience with 
electronic processes yet it remains idiomatic and unfailingly attractive. 
Although the five sections may be played in various successions, their 
inner forms retain Berio's characteristic simplicity. The String Quartet 
(1956) seeks to compensate for an essentially monochromatic medium 
by systematic use of vibrato, pizzicato, sordino, and col legno effects, 
influenced in this by Maderna's quartet of the previous year.? Here too 
variation-principle is at work,? on material of great nervous intensity. 
But Berio avoids overtaxing the listener by diluting the stream of 
events at times, and his practice of dividing long durations into repeated 
notes even provides a link with the rhythmic world of Stravinsky and 
Bartók (see Ex. 213). 

Berio's Allelujah II (1956-8) profits from the example of Gruppen in its 
spatial and temporal conception of five instrumental bodies, but the 
scoring and the chain of transformations forged from the initial material 
maintain personal traits. Circles (1960), for soprano, harp, and two 
percussion players, supplements Stockhausen's views on the progression 
between tone and noise, intelligibility and basic speech elements.* The 


1R. Smith Brindle, Musical Quarterly, xliv (1958), рр. 95-101; Piero Santi, ‘Luciano 
Berio', Die Reihe, iv (Eng. ed.) pp. 99-100. 

2 On Maderna’s quartet, see Giacomo Manzoni, ‘Bruno Maderna’, Die Reihe, iv (Eng. 
ed.) pp. 115-18. 

3 Santi, loc. cit., р. 100. 4 Cf. pp. 470-3 and 491. 


480 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Sas 
Ve. аа 


[У.О. = vibrato ordinario] 


disintegration of language is begun here by the poet, E. E. Cummings, 
but taken by the composer to the point at which individual consonants 
are recognized as additional percussion instruments. Similarly the 
musical content is also gradually reduced to ‘noise’, and the entire 
process is then reversed for the second half of the piece. The percussion 
writing derives from Zyklus, each player controlling fifteen types of 
instrument; and its improvisatory practices include the free ‘expression’ 
of words omitted in the singer’s restatement (see Ex. 214(ii)), while the 


THE AVANT-GARDE IN ITALY AND ELSEWHERE 481 


singer is twice required to walk to a new position. Although this work 
was denounced as a farrago of modernism,! in fact it demonstrates a 
notable gift for evocative imagery that has nothing to do with theoretical 
speculation (see Ex. 214(1) ) and an arched form naturally adapted to 
its descent into elemental sound and recovery of musical meaning. 


Ex. 214 
(1) 
| ae с 7-2” ШЕ а Е] 
Voice Нач кшк ишкс SS Ss 4 
ті - ver-ly 15 ү 
- 5 Ia, -— 
Wy 2! кес XI em | 
FD El 
j р sempre 
Harp ; 
ENE de 
wen [TIRE > [— 1o — г 
-- кес E 4 — — — [2 
pues no mue Е Ш Е м | -x— | 
lascia vibrareb 1 
иг = 3—1 2=— 
SSS SS EE E келі h E BESS 
io. == = Eee bee SS ee ро ра. "d j= А m] 
ыы Cf a mır Fu үк - Иа с ааа НЫЕ] 
М.у p 
flow - er — gone so - Шу by tomb 
f) Ц 


Y РР 
то - - - - -(sly) 
^ b te b 
7 — жазала E. EI = =: 
je Р "DD 


1 See the review by К. F. Goldman, Musical Quarterly, xlvii (1961), p. 239. 


482 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 
Gi) 


Bongos таа 
5 
қ Тот-Тот |. — та ағасы! 
IC. 
Pare) 3 Triangles | AREA 
3 susp. Cymbals - lis hushed AL. 5 


Voice Еа a И eS ааа 
5 Ы 2 
-rain night 
where night 
Tamburino basco c ра EUREN 
2 Вопро$ of dreamsend| 7”) 
П 2 Congas = nun 


As two of the most forceful creative personalities of the new music, 
Boulez and Stockhausen strongly influenced the development of 
numerous lesser men. Any well-publicized artistic movement will 
attract untalented imitators, to some of whom the avant-garde implies 
only the pursuit of the dernier cri. The restrictions of total serialism and 
the freedoms of indeterminacy are equally easy to reproduce, and 
innumerable personal foibles can be devised by any composer merely 
intent on writing an impressive score. When the average score is of 
formidable complexity, the number of those capable of mentally 
formulating its sound must be very small; and so the historically un- 
precedented position arises in which works reach performance because 
no one has felt able to pronounce with certainty that they were unworthy 
of it. Fortunately the deception cannot be sustained indefinitely, and 
the barrage of technical information thrown up around many new 
works—another curiosity of this period—is unlikely to conceal from 
the experienced listener evidence of patent ineptitude in the manipula- 
tion of sound relationships, rather than of their symbols on paper. 

The charlatanism of music that simply apes the fashionable must 
not discredit the efforts of composers whose genuinely exploring cast 
of mind attracts them naturally to the new paths. Not all of these 
composers depended slavishly on the enterprise of the two pioneers. 
The Belgian, Henri Pousseur (b. 1929), an associate of Stockhausen 
in the earliest electronic investigations, has also shown the German's 
influence in his instrumental works. But his individual treatment 
of texture has developed from a detailed study of Webern,! and from 


1 Henri Pousseur, ‘Webern’s Organic Chromaticism’, Die Reihe, ii, (Eng. ed.) pp. 51-60; 
also his comparative study of Schoenberg and Webern, Incontri Musicali, i (Milan, 1956). 


THE AVANT-GARDE IN ITALY AND ELSEWHERE 483 


speculation of evident integrity.! His important analysis of Webern's 
*organic chromaticism'—chains of connexion through the semitone 
and its octave-transpositions that dispel the polarizing tendencies of 
other intervals (even including the octave)—has led directly to the 
predominance in his idiosyncratic harmonic system of the major 
seventh and minor ninth. Twelve-note serialism is not relevant to his 
schemes, but the Quintet (1955) in memory of Webern adapts the 
interval succession of the row in that composer's Op. 22 Quartet. Each 
interval is regarded as a chromatic scale segment, all its semitonal 
constituents appearing, but as octave-transpositions; as their order of 
appearance within a given time-span is free, the relations of the various 
strata can be regulated so as to control the by-product of other 
intervals.? In later works, Pousseur has diversified his consistent but 
monotonous harmonic material by acknowledging non-chromatic 
intervals as primary factors, but systematically neutralizing their polar 
tendencies through added chromatic relations. Ex. 215 from his 
Impromptu shows a simple example founded in whole-tone intervals: 


“Ех. 215 


Allegretto (^ c. 96) 


Tritones 


1 Pousseur, ‘Outline of a Method’, Die Reihe, iii, (Eng. ed.) pp. 44-88. 
? See the critical analysis of the Quintet's procedures in G. M. Koenig, ‘Henri Pousseur', 
Die Reihe, iv, (Eng. ed.) pp. 16-28. 


484 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


accel, - - E Ea is Zr PE im E. 


Tritones ) 


It also shows an early attempt to perfect a notation for the ‘qualitative 
relationship’ of durations, freed not merely from complex irrational 
values but from all numerical rigidity. Grace-notes within the square 
brackets are included within the main note- (or rest-) value, and the 
conventions of commas, dashes, and plus-signs allow the player to 
concentrate on immediate subtleties of proportion such as the listener 
can detect. The Mobile for two pianos (1958), in extending this liberty 
to two players, naturally exploits the flexible relation of the strata first 
explored by Stockhausen! and possibilities of their varied arrangement 
such as Pousseur had already provided in his electronic piece Scambi 
(Exchanges) (1957). 

It was among the circle associated with Stockhausen in Cologne, 
mostly of musicians from peripheral countries, that the pursuit of 
innovation sometimes outstripped any demonstrable urge towards new 
expressive media. Bo Nilsson (b. 1937), a Swedish composer of pre- 
cocious talents, in his vocal works Méádchentotenlieder (Dirges for Girls) 
(1958) and Ein irrender Sohn (A straying son) (1958) fluently synthesized 
vocal resources and instrumental colourings reminiscent of Le Marteau 
with the temporal freedom and stereophony of Gruppen, adding some 
distinctive treatment of octaves (as tone colour) and of near-conven- 
tional line. The nervous sensitivity of these works is signally lacking in 

1 Pousseur, ‘Musik, Form und Praxis’, Die Reihe, vi, pp. 82-83. 


THE AVANT-GARDE IN ITALY AND ELSEWHERE 485 


the grotesque Zwanzig Gruppen (Twenty groups) for piccclo, oboe, and 
clarinet. By applying simultaneously the action-durations of Stock- 
hausen's Zeitmasse and the ‘mobile’ principle, Nilsson contrives to 
accumulate up to one hour's sound from the superimpositions of diverse 
unequal lines selected in random order from sets of twenty per player. 
Since no further control is imposed, and indeed entirely unforeseen 
juxtapositions must be the only subjects of the listener's attention, 
it is difficult not to believe that an indistinguishable effect could be 
produced from any suitably complex material; a few random groups 
(Ex. 216) will reveal the contrast between their meticulous performing 
directions and the anarchy of their relations.! 


Ex. 216 
а verlangsamen ( А: )-.- = 
} Фр 
Piccolo (8va) 
Oboe 
. Clarinet 
(sounding notes) 


——————— -——— [d 
a карты 


К===Е= beschleunigen РР = 
D e 


Не: verlangsamen 


Ch. 


1Cf. Hans Rudolf Zeller, ‘Mallarmé und das serielle Denken’, Die Reihe, vi, p. 29, 
n. 11. On Nilsson’s divorce from practicalities in devising an electronic score, see G. M. 
Koenig, ‘Bo Nilsson’, Die Reihe, iv, (Eng. ed.) pp. 85-88. 


486 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Picc. -(р 


Ob. 


L3:24 13-24 


r^ 


Си. 


— пай ЕЕЕ 
г-9:2- verlangsamen 


mE 
Picc.-f? 
МЗ! 


Ob. 


Cit. 


The Argentinian, Mauricio Kagel (b. 1931), has sought constantly to 
enlarge the sonorous possibilities of familiar media, introducing quarter- 
tone harmony as well as familiar indeterminacies in his String Sextet 
(1953-7) and creating some beautiful new piano sounds through the 
percussion player іп Transicion II (John Cage’s influence is more than 
modish here), but his Anagrama (1957-8) extends its play on the 
assonances of a speaking choir to an inordinate length without aspiring 
to the musicality of Berio's Omaggio. 

Behind such projects, however admirably executed, lurks the tempta- 
tion to “épater le bourgeois’. Confidence іп the essential vitality of the 
new movement is sustained by many musicians from all over Europe 
who accept some of its idioms as a natural vehicle for undemonstrative 
craftsmanship of a very high order. The adoption of advanced serial 
techniques by composers of the Eastern bloc, and the contemporary 


MUSIC ON TAPE 487 


music festivals there (Warsaw, from 1956), might be interpreted as 
nostalgia for old artistic alliances or as spirited repudiation of Soviet 
aesthetic ideology, but the development of men like Haubenstock- 
Ramati (b. 1919) and Witold Lutoslawski (b. 1913) suggests a less 
factitious sympathy with new methods. Roman Haubenstock-Ramati, in 
his Symphonies de timbres (1957), finds a convincing role for development 
and (reversed) restatement of material, a fluid rhythmic variety thatstops 
short of insuperable difficulties, and a distinguished range of orchestral 
colourings; evidence of such mastery compelled respectful investigation 
of later, more enigmatic, scores like the Mobile for Shakespeare (1960). 
Haubenstock left Poland in 1950, but Włodzimierz Kotonski (b. 1925) 
took the lessons of Darmstadt back to Warsaw and worked at the 
electronic studio there. His Kammermusik (1958) and Musique en relief 
(1959) show a gifted miniaturist with a remarkable ear for ingratiating 
textures; so gentle an introduction to modern techniques may well win 
them an audience. Krzysztof Penderecki (b. 1933) set out from a similar 
position, and in his subsequent career he has shown a remarkable 
ability to adapt new techniques to ends which quite unsophisticated 
listeners find compelling: the opaque textures (for fifty-two string parts) 
of his Tren—Ofiarum Hiroszimy (Threnody for the victims of Hiroshima) 
(1961) marked an important stage in this development. Lutoslawski's 
Muzyka zatobna (Funeral music) for Bartok (1958) similarly marked the 
beginning of a new style in a development which has shown initiative 
and imagination in the use of new techniques. 


MUSIC ON TAPE 


Тһе rough distinction made earlier! between a French ordering by 
instinct and а German organization by system 15 reinforced by the 
different approaches in these countries towards a music that eliminates 
the interpreter. Musique concréte is above all an art of tasteful arrange- 
ment of existing sound shapes, while electronic music, faced with an 
infinite wealth of sound potentialities, must shape according to rigidly 
restrictive principles if it is to avoid a chaotic prodigality. Their direct 
construction of the means of production is perhaps no more significant 
ultimately than their recognition of finer distinctions in aural pheno- 
mena formerly classed indiscriminately as ‘noise’, distinctions which 
may contribute to the balance and contrast that form musical experience. 
In this respect, electronic music, founded in acoustic synthesis, offers 
incomparably more precise criteria for a new tabulation of sound since 
it can order its inner structure. 


1 Cf, pp. 441-2. 


488 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


The patchwork construction of forms from an arbitrary assembly of 
‘concrete’ pieces of sound (by manipulation of magnetized recording 
tape) offers an obvious parallel to collage methods in the visual arts. 
Both demand of the creator more aesthetic judgement than conven- 
tional craftsmanship, and both may confuse our sensibility to the whole 
by an appeal to sentimental associations with still recognizable con- 
stituent objects. The works which have achieved the widest note since 
Pierre Schaeffer's experiments! began in Paris in 1948 are frankly 
directed towards programmatic or impressionistic ends. In the Syrn- 
phonie pour un homme seul, by Schaeffer and Pierre Henry, a technique 
which has proved admirable for prompting a wealth of suggestions 
during a few moments of incidental sound is used to produce an 
evocative pattern far more ramified than that of L'Aprés-midi d'un faune 
with means often cruder than those of Pacific 231. Ballet, which has so 
often reduced the status of music intended to be self-sufficient, has come 
to the rescue of this type of concrete music by providing a visual 
continuity around which the sound can conjure up its momentary 
fantasies. 

Several French composers, intent on a music that should justify its 
continuity by an inherent logic, experimented with the equipment 
assembled by Schaeffer. Though his stock of sounds was too ad hoc to 
offer possibilities for a permutational treatment of their constitution— 
rather than of their juxtaposition, already practicable in serialized 
instrumental timbre—the techniques of tape montage? did offer a 
guaranteed precision in the execution of complex rhythmic superimposi- 
tions such as seems likely to remain for ever beyond instrumentalists' 
skill. Boulez constructed two Etudes (1951-2) setting out respectively 
from a single sound and six types of sound, but he was satisfied by his 
results only in the field of durations.? Messiaen's essay Timbres-durées 
(1952) merely worked out at greater length some formulas of his con- 
temporary instrumental music. By now most composers whose ideas call 
for direct contact with the sounding material are more attracted by the 
wider and measurable resources of electronic sound-generation. Yet the 
unusually penetrating quality of Stockhausen’s Gesang der Jiinglinge 
and Berio’s Omaggio a Joyce springs from an analytical treatment of the 
most evocative of all ‘concrete’ sources—the sound of the human voice. 

When such existent sources are abandoned and musical composition 


1 Schaeffer, A la recherche d'une musique concrète (Paris, 1952); see also Lowell Cross, 
‘Electronic Music 1948-53’, Perspectives of New Music, vii (1968), р. 32. 

2 On musique concréte see Schaeffer, op. cit., and Goléa, op. cit., pp. 150-3; on electronic 
music, various articles in Die Reihe, i; see also the bibliography іп. Aurelio de la Vega, 
‘Regarding Electronic Music’, Tempo, 75 (1966), and in Journal of Music Theory, vii (1963). 

3 Boulez has stressed that his concern was for organized duration, not for a mere juggling 
with the physically impracticable: Die Reihe, i, (Eng. ed.), p. 24. 


MUSIC ON TAPE 489 


begins from the electro-acoustic fundamentals (the pure sinusoidal tone 
and, at the other extreme, the *white noise' in which all pitches are 
present), the composer can only abandon a rhetoric based on principles 
of physical tension and relaxation that no longer have any relevance 
to a patently ‘effortless’ aural phenomenon. АП sounds—of whatever 
pitch, duration, volume, density, or timbre—are then without associa- 
tions of graded human achievement and are a priori equally feasible. 
This limitless repertoire did not become available until a few years after 
composers writing for conventional resources had begun to attempt a 
music that should order these parameters simultaneously (Webern and 
Messiaen) or make distinctions as fine in the sphere of noise as in that 
of note (Varèse). In fact timbre, never the most satisfactory element in 
the total serialization process, has ceased to be a parameter in the old 
sense and is recognized as the organization at another level (the micro- 
structure) of relations of time (i.e. vibration-frequencies)! and intensity, 
the familiar tone colours being no more than particular examples (with 
their partials in ‘harmonic’ proportions) from a continuously variable 
and thus serially permutable range. 

An appropriate attitude to its control on two levels therefore existed 
before the new medium, but its surrender to such control emphasized 
how much instrumental performance had been fertilized by individual 
unpredictability and idiosyncrasy. The investigation of this latter field 
has been noted ;? but at the same time as ‘live’ music was being made to 
depend increasingly on the unique, unrepeatable quality of the individual 
performance, music for mechanical means of production, instead of 
being the petrifaction of one such performance, was seen to be most 
aptly conceived as a precise correlation of measured sounds. The onus 
remained with the composer-operator to demonstrate in the process 
anything more widely compelling than the appeal of the underlying 
pattern of calculations. Yet he was confronted at each stage of his work 
by the aural implications of his theories,* and so empiricism inevitably 
became an important modifying factor. 

The challenge was taken up with enthusiasm, if not with consistent 
success. Within a few years of the establishment in 1951 of the first 
studio, at Cologne under the direction of Herbert Eimert (1897-1972), 
attempts to synthesize familiar sounds? were rejected as distractions 


1 Cf. p. 460 for some consequences of this attitude in Stockhausen's instrumental music. 

? Cf. pp. 464-8. 

3 See С. M. Koenig, “Во Nilsson', Die Reihe, iv, (Eng. ed.) pp. 85-88, on the folly of 
planning an electronic composition without access to its realization in sound. 

4 Herbert Eimert, loc. cit., p. 5; Stockhausen, ‘Elektronische und instrumentale Musik’, 
Die Reihe, v, p. 50. 

5 E.g. Eimert's Glockenspiel (1953-4), based on an analysis of bell sounds. 


490 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


from the task of relating an external musical order to the internal 
acoustical order: ‘everything is designed not to escape from the nature 
of electronic sound but to go further into it’. Stockhausen's Studie I 
(1953)? was restricted to sinusoidal tones, but their superimpositions 
were determined by a serial plan relevant to the structure of the whole 
piece,? not by a deliberate creation of harmonic series fusing into single 
timbres nor by any restriction to the tempered twelve-note scale. As 
well as some rather characterless chords, i.e. combinations without 
fusion, it uses the less familiar and aurally intriguing phenomena known 
as tone-mixtures, combinations in which the constituent pure tones 
merge to some extent, yet not so as to produce the impression of a single 
note. Eimert's Etüde über Tongemische (Study in tone-mixtures) (1954)4 
deploys five mixtures, each of nine partials and each transposed to nine 
different registers, with results far more congenial to the medium than 
the relapses into instrumental styles of his Fünf Stücke (1954—5). 

Though the new territory of tone-mixtures was the most immediately 
attractive, investigations were made on both sides of it. Paul Gredinger 
in Formanten reached back towards the nature of instrumental sound in 
seeking to regulate the harmonic ordering of partials so as to produce 
a gradually changing timbre. Stockhausen, in Studie II (1954),9 the 
first electronic work to be published as a diagram," varied the widths of 
mixtures and the closeness of their superimpositions, often approaching 
the dense band of adjacent frequencies of ‘coloured noise’ (i.e. a filtered 
segment of white noise, referable only to a pitch area); the durational 
plan of this short piece is heard as a rhythmic life, diverse yet coherent. 
Here the recognition of individual pitches is no longer of prime impor- 
tance and the music is characterized by just that ‘atonality’ which 
Schoenberg scorned as an impossibility. In the Klangfiguren by Gottfried 
Michael Koenig,® the material is restricted to ‘noise’, so diversely 
graded as to suggest a musical potential higher than that of a conven- 
tional percussive apparatus. 

In Stockhausen's Gesang der Jünglinge (1955-6),? electronic sounds 
take on a disturbing ‘otherness’ when set in relief by the humanity of a 
boy's voice, racked at times out of intelligibility, but never out of 

! Eimert, loc. cit., p. 10. ? Recorded on DGG LP 16133. 

3 Stockhausen's analysis in Technische Hausmitteilungen des NWDR 6 (1954) is sum- 


marized by Burt in “Ап Antithesis', ТЛе Score, xix (1957), p. 64. 

4 DGG LP 16132. 

5 Pousseur, *Formal Elements in a New Compositional Material, Die Reihe, i, (Eng. ed.) 
p. 32, and Gredinger, ‘Serial Technique’, ibid., р. 40. 

ê DGG LP 16133. 

? Vienna, 1956. 

з С. M. Koenig, ‘Studio technique’, Die Reihe, i, (Eng. ed.) pp. 52-54; 'Studium in. 
Studio', Die Reihe, v, p. 74; Klangfiguren is recorded on DGG LP 16134. 

9 DGG LP 16133. 


MUSIC ON TAPE 491 


recognition, by the dissection of its speech elements. Effects such as the 
distant murmur of multitudinous identical voices have a dramatic 
impact far more direct than Stockhausen's comments on the work would 
suggest; his concern is to incorporate vocal sounds as natural stages 
(complemented electronically) in the continuum that links tone to noise, 
vowel to consonant.! His vivid imagination for broad effects is further 
revealed in the spatial direction and movement of the sound by multi- 
channel distribution.? This has subsequently become a crucial aspect of 
electronic composition and has helped to combat the faintly ridiculous 
sensation with which an audience concentrates on sounds emanating 
from a single ‘pseudo-instrument’.* Stockhausen’s fanatical devotion 
to this art is sustained by a vision of public music rooms (spherical 
ideally) giving continuous performances of spatial music.* However 
jgminiscent this may seem of some deplorable cinematic techniques, 
complex stereophony is an altogether natural development of machine- 
music and may help it to achieve a persuasive idiom owing nothing to 
instrumental practice. 

Another achievement which lies ahead is that of a basis for the 
criticism of electronic music. Such ‘scores’ as exist are instructions to the 
technician but a dubious boon to the musical, rather than acoustic, 
analyst;? often the composer who has constructed his own work keeps 
no record of his procedures and has no further interest in them. In this 
unprecedented historical situation the critic must rely largely on an 
instinctive perception of formal balance. His recognition of certain 
broad classes of sound may become as automatic as that of fixed instru- 
mental timbres, but there will always remain a degree of uncertainty in 
the means which will compel him to consider more searchingly the end. 
The pointed avoidance of emotional overtones in such titles as Continuo 
or Perspectives should discourage him from seeking raisons d'étre out- 
side the sounds themselves. This quandary may ultimately be beneficial, 
but meanwhile electronic works have found more receptive ears when 
presented less enigmatically. Artikulation, for example, by Gyórgy 
Ligeti (b. 1923) is acknowledged to be humorous in intent.® Its sounds, 
individually suggestive of screams and whistles, pistol-shots and bath- 
water, were long ago drained of risible potential, but are juxtaposed and 


1 Stockhausen, ‘Musik und Sprache’, Die Reihe, vi, p. 52; 'Actualia', Die Reihe, i, 
(Eng. ed.) p. 45. 

2 Stockhausen, ‘Musik im Raum’, Die Reihe, v, p. 60. 

з Boulez, ‘At the ends of fruitful land . . ^, Die Reihe, i, (Eng. ed.) р. 28. 

* Stockhausen, Die Reihe, v, p. 60. 

5 Зее however B. Fennelly, “А Descriptive Language for the Analysis of Electronic 
Music’, Perspectives of New Music, vi (1967), p. 79. 

® Stockhausen, ‘Elektronische und instrumentale Musik’, Die Reihe, v, p. 53; see also 
Ligeti, ‘Wandlungen der musikalischen Form’, Die Reihe, vii, p. 14. 


492 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


counterpointed here with so much wit and musical craftsmanship that 
the piece 15 exhilarating rather than farcical. 

The facilities of the Cologne Studio could not provide for all the 
composers with ideas that called for realization in the new medium. 
Under the direction of Maderna and Berio, a studio was opened at 
Milan in 1955, and Pousseur worked there until facilities were available 
at Brussels in 1958, while a Dutch studio was used by Henk Badings. 
Eastern European interest іп Western musical developments was con- 
firmed in the institution of a studio at Warsaw and, still further afield, 
America and Japan Joined the pioneering movement. Many of the early 
works produced at these centres are no more than studies, but two 
pieces made at Milan call for notice. Pousseur's Scambi (1957) attempts 
to compensate for absence of the interpreter by introducing the element 
of choice so industriously cultivated in instrumental music like his own 
Mobile for two pianos. The individual strata of Scambi are composed by 
a characteristic application of precise filtering-techniques to the 
random phenomena of white noise, and the operator may select from 
a variety of possible schemes his own sequence and superimposition.! 
Berios Omaggio a Joyce (1958) depends more heavily than does 
Stockhausen's Gesang on the concréte medium of a human voice, here 
reading a passage from Ulysses. Having analysed these sounds, Berio 
proceeds to manipulate them into an edifice of disturbing beauty in 
which ‘musical’ values—of colour as well as pitch—become far more 
important than verbal intelligibility. The lyricism of this work and its 
satisfying emotional curve are Italian qualities which remain recogniz- 
able in more orthodox pieces like Berio's Momenti and Maderna's 
Notturno. Both composers have also experimented with the confronta- 
tion of electronic and instrumental sound. 

A pointer towards such synthesis was provided by Edgar Varése 
(1883-1965). After the intrepid acceptance in his early works of ‘noises’ 
as essential extensions of music's basic material —sound (this in itself 
being a concept for which few were ready in the early thirties)?—Varése 
had withdrawn from composition, but not from vigilant enquiry into 
the nature of noise. More than any eccentric of this century's music he 
saw his ideas pass into common currency, for the perfection of tape- 
recording techniques and of electronically generated sound gave them 
an unsuspected validity. Though his early works have been discovered 
with enthusiasm, the example of Déserts (1954) has been more widely 
followed in its combination of instrumental sounds (piano, wind, and 


1 Marc Wilkinson, “Тууо Months in the Studio di Fonologia', Тйе Score, xxii (1958), p. 45. 

? Wilkinson, “Ап Introduction to the music of Edgar Varese’, The Score, xxi (1957), 
р. 5; Milton Babbitt, *Edgard Varése: a few observations of his music,’ Perspectives of 
New Music, iv (1966), p. 14. 


CONSERVATISM AND COMPROMISE 493 


percussion) and a recorded tape (of noise sources); its crude vitality 
proved his musical nerves to be as strong as ever, and his later Poéme 
électronique (1958), dispensing with instruments, is a tour de force con- 
structed for the multi-channel reproducers of Corbusier’s pavilion at 
the Brussels World Fair. Meanwhile the integration of player and 
machine proceeded through forms as various as Maderna’s Musica 
su due dimensioni (1958) in which a solo flute plays against electronic 
background, and Pousseur’s Rimes pour différentes sources sonores 
(1959) which engages three orchestral groups and two loudspeaker 
channels. In this field too, Stockhausen attempted an exhaustive 
exploration: his Kontakte (1959-60) aims once again at a sound con- 
tinuum (electronic sources providing transitional stages between 
instrumental) but he brings a new ingenuity to spatial disposition and 
rotation of sound produced from a circular formation of four loud- 
speakers with the two players (piano and percussion) at its hub.! Though 
these were among the earliest essays in synthesis, they already suggested 
that audiences welcome the guidance of palpable patterns of human 
achievement. This does not invalidate electronic media, and may indeed 
hasten their development and recognition as artistic expression through 
the ordering of sound—in short, as music. 


CONSERVATISM AND COMPROMISE 


Although many writers consistently deplored the manifold innova- 
tions which obsessed some of the most notable creative talent in 
European music during the sixth decade of this century, it was no longer 
possible to envisage the emergence of a serious style which would show 
no trace of the new strategic ordering of musical time, pitch, intensity, 
and colour. There were composers, not belonging to the avant-garde, 
who found that their slower progress had none the less brought them to 
a point where they were able to contribute to some valid synthesis. Such 
men, sometimes intelligently interested in their contemporaries’ 
explorations, found themselves a territory among the diverse technical 
advances made but not fused during the first half of the century. 
Conservatism proved the refuge of scarcely more of the untalented than 
were to be found in esoteric circles, and was by no means automatically 
a dismal academicism. The common sense that views tradition as a 
measure of, not a defence against, the present can often adapt or 
interpret new idioms in forms acceptable to a wider public. 

Whatever the varieties of individual method to be found among the 


1 The theoretical basis of Kontakte is discussed by Stockhausen іп ‘The Concept of 
Unity in Electronic Music,’ Perspectives of New Music, i, 1 (1962), p. 39. 


33 


494 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


new serialists, they could draw confidence from their essential unity of 
purpose; and they present a ‘movement’ that will continue to engage 
historians' attention. Outside their confines no such solidarity exists, and 
the species of craft practised permit no neat classification. Switzerland, 
always unsympathetic to extremist doctrines, provides a typical cross- 
section in the work of three composers of the generations after Frank 
Martin. The operas of Heinrich Sutermeister (b. 1910) have a devoted 
public, for their musical language firmly excludes any trace of recent 
developments. Comparison of this innocuous diatonicism with Britten's! 
flatters a pedestrian talent with none of Britten's gift for the unpredict- 
able reassessment of the apparently obvious. Sutermeister's exact con- 
temporary, Rolf Liebermann, has yoked twelve-note method to an 
elegance that its originator never sought, notably by the use of triadic 
rows;? these provoke none of Berg's unrest but a stream of figurative 
chatter owing something to Hindemith. The Concerto for Jazz Band 
and Orchestra fails to find much common ground between elements so 
disparate as jazz and dodecaphonic music, but a keen sense of musical 
theatre makes his operas apt, if rarely penetrating. The work of Klaus 
Huber (b. 1924) represents the moderation of a later generation. Taught 
by Blacher and familiar with all the resources developed in the first half 
of the century, he has sought a personal synthesis, one that will convey a 
mood rarely explored during those years, of religious mysticism; his 
chamber cantata Auf die ruhige Nachtzeit (At the quiet night-time) 
(1958) captures this in a serene contemplation of musical symmetries. 

Synthesis does not always achieve this poise, and may do no more 
than appropriate technical gains without a distinctive expressive note. 
Karl-Birger Blomdahl (1916-1968) had an impressive command of tech- 
niques that included serialism, and in his chef d'euvre, the opera Aniara 
(1957-9), he also drew on tape montage. This work bravely attempts 
to provide a comprehensive document for an age in which obsessions 
of impending catastrophe are diverted by visions of planetary explora- 
tion. But the score has little of the keen fantasy demanded by its 
subject: its dodecaphonic basis merely produces a monotonously uni- 
form harmonic level in stubbornly traditional textures and rhythms (see 
Ex. 217), and so, instead of an imaginatively prophetic music, it presents 
studies in the conventionally horrific, lascivious, or ecstatic, which rarely 
suspend disbelief. Blomdahl was a pupil of Hilding Rosenberg, as were 
two other Swedish composers who have shown the same determination 
to escape from a paling nationalism through Central European tech- 


1 Winfried Zillig, Variationen über neue Musik (Munich, 1959), p. 221. 
2 See Liebermann’s contribution to Rufer, Composition with Twelve Notes (London, 
1954), pp. 191-3. 


CONSERVATISM AND COMPROMISE 495 


Tran uillo a c. 66) 
q - D 


Jag läng - tar till — lan-det 


(I long for the land that is not) 


niques. Sven Erik Báck (b. 1919) and Ingvar Lidholm (b. 1921) have 
now moved towards multi-dimensional ordering; despite the difficulties 
of his compromise position, Blomdahl's work suggests the strongest 
talent. In Denmark, an unpretentious synthesis of serialism with strong 
inflexions of Stravinskyan choral style is practised in the church music 
of Bernhard Lewkovitch (b. 1927). 

To detail the innumerable personal solutions worked out by European 
composers who have stopped short of complete association with the 
advanced movement would be both tedious and confusing. Men like 
Roman Vlad (b. 1919) in Italy,! Marius Constant (b. 1925) in France or 


! Ronald Stevenson, “Ап Introduction to the Music of Roman Vlad', Music Review 
xxii (1961), p. 124. 


496 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Tadeusz Baird (b. 1928) in Poland, are evidently conversant with the 
recent developments but cautious in profiting from them. The music of 
such composers plays a vital part in the endless process of acclimatizing 
audiences to sounds they would once have found incoherent, and its 
craftsmanship may well surpass that of some who assist in the more 
spectacular extension of the frontiers. Yet only an unusually compelling 
creative power can give it a dominating position in the contemporary 
scene: in the music of Hans Werner Henze (b. 1926), compromise 
achieves its most persuasive justification. 

Germany's role as chief patron of the new serialism soon extended 
beyond the Darmstadt schools and the Donaueschinger Musiktage 
(revived in 1950) to the commissioning of many scores by the various 
radio stations. Their policy was enlightened and their nominations 
catholic; good performances were ensured and audience figures were not 
a determining factor. Thus the composer whose position was secured 
might freely indulge an exploring spirit, seeking no stimulus outside that 
of like-minded colleagues. These hermetic conditions inevitably 
favoured extravagant innovation for its own sake, but the liberal dis- 
semination of experimental music perhaps re-orientated some German 
composers originally inclined towards more moderate idioms. Bernd 
Alois Zimmermann (1918-1970) for example, whose Symphony 
(1953) is an admirable mosaic of expressionist textures, in subsequently 
pruning this opulence added to Webernian interval proportions a 
strict regulation of other dimensions.! Giselher Klebe (b. 1925)? was a 
pupil of Boris Blacher, and the early two-piano sonata suggests no 
aspirations beyond the mixture of Stravinskyan rhythmic acuity, 
chromaticized diatonicism and swing formulas to be found in the work 
of his talented fellow-pupil, Heimo Erbse (b. 1924).3 But Klebe's need 
to control a piquant fantasy by intellectual disciplines led him 
through orthodox twelve-note techniques handled with virtuosic ease— 
as in the String Quartet (1950) or the Rómische Elegien (1952)—to an 
individual treatment of serial durations.* Neither this nor his experience 
of electronic composition? has led him to abandon a refined thematicism 
or the wealth of traditional ‘expressive’ analogy; yet his fastidious con- 
cern for craftsmanship has deprived his works, including the opera Die 
Rüuben (The Robbers), of the wider currency accorded those of the more 
flamboyant Henze. Transition in the opposite direction, though excep- 


! Reinhold Schubert, *Bernd Alois Zimmermann', Die Reihe, iv, (Eng. ed.) pp. 103-13. 

? W.-E. von Lewinski, *Giselher Klebe’, Die Reihe, iv, (Eng. ed.), рр. 89-97, and A. D. 
McCredie, *Giselher Klebe', Music Review, xxvi (1965), p. 220. 

3 Francis Burt, “Ап Antithesis’, The Score, xix (1957), p. 71. 

* Lewinski, loc. cit., pp. 92-94, on the Elegia Appassionata Trio (1955). 

5 Giselher Klebe, *First Practical Work', Die Reihe, i, (Eng. ed.) p. 17. 


CONSERVATISM AND COMPROMISE 497 


tional, may seem а surer guarantee of popular success. Gottfried von 
Einem (b. 1918), an Austrian pupil of Blacher, has shown increasing 
attachment to a diatonicism seasoned with glib references to popular 
idioms; his Dantons Tod (Death of Danton) (1947) was a landmark in 
the post-war recovery of German opera, but a signal decline from the 
Büchner opera it in some respects aped. 

When the Nazi downfall permitted the vigorous resurgence of German 
music, educators like Blacher and Fortner must have visualized a return 
to musical literacy less violent than that which has reshaped the 
language. Yet if they could not foresee a Stockhausen, they might well 
have foreseen a Henze, for in many ways he was to remain faithful to the 
ideals of those years. This is not to say merely that he became the 
supreme exponent of a synthesis composed of all the major influences 
of the first half-century, but that he pointedly dissociated himself from 
his contemporaries' belief in intellectual canons as an arbiter of sound- 
construction. The composer who in 1947 was seeking *widely-ranging 
tender cantilenas'! was still intent on ‘tender beautiful noises’? twelve 
years later, and this sonorous vision was far more fundamental to his art 
than any one of the techniques he sampled in realizing it. Indeed, he 
turned to twelve-note serialism largely because, under the spell of Berg, 
he saw in it a way to a rich lyricism, not a substitute for tonal coherence 
(cf. Schoenberg) nor one factor in predetermining structure (cf. the early 
*post-Webern' experiments). As he soon abandoned its consistent use, 
his typical methods have sometimes seemed to be as dependent as were 
Stravinsky's on the empirical juxtaposition of apt sounds. But his train- 
ing under Fortner and Leibowitz has ensured that spontaneously in- 
vented ideas suggest appropriate, and often subtle, means for their 
structural extension: ‘structural rules emerge from ideas laid down at 
the beginning of a work; their development and variation are subject to 
no arrangement imposed from without'.? 

His earliest works show the assimilation of Hindemith and Stravinsky 
which was then the routine German apprenticeship, and these influences 
persist in works which adhere painstakingly to the serial letter. The 
Piano Variations (1948) strive to reproduce the manner of Schoenberg's 
first serial period (see Ex. 218(1)), yet hesitate to indulge his fearless 
pungency of dissonance; Ше tonal implications of the row, Ex. 218(ii), 
naturally produced harmony nearer Berg, but Hindemith and Stravinsky 


1 Henze on his Violin Concerto, quoted by Reinhold Schubert in ‘Bernd Alois Zimmer- 
mann', Die Reihe, iv, (Eng. ed.) p. 105. 

? Henze's guidance to his librettists, quoted by W. H. Auden and Chester Kallman in 
‘Genesis of a Libretto’, Glyndebourne Festival Programme Book (1961), p. 37. 

3 Henze, lecture at the Braunschweiger Festliche Tage Neuer Kammermusik 1959, 
quoted by Diether de la Motte, Hans Werner Henze—Der Prinz von Homburg (Mainz, 
1960), p. 60. 


498 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


sometimes affect texture and rhythm. In the Second Symphony (1949), 
though the row bristles with cadential possibilities (see Ex. 218(ii)) and 
is progressively transposed, the harmony is already beginning to explore 
the Stravinskyan immobility bred of inner contradictions (Ex.218(iii)(a) ). 
German influences still predominate however: the movement structure 
is indebted to Hartmann, the scherzo's twelve-note ground, Ex. 
218(iii)(b), uses Blacher's variable metres, wittily reversed in the da capo 
after a palindromic trio, and the finale emulates Hindemith's motoric 


Ex. 218 


(i) BS BS transp. 
d=) = lp" "uod 


Piano 


499 


CONSERVATISM AND COMPROMISE 


Basic set of Piano Variations 


Gi) 


еріне p ее 


min. 


мен ене cece eee Bea 


i || ==? 
TD Бре ret 
Жү Mecsas Sa ае 


БЕНЕН occi 


tria 


min. 


гл Triadic or quas 
—— Leading notes 


dic groups 


1- 


.---- Tonal groups 


Basic set of Second Symphony 


W.W. D espressivo 


500 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


counterpoint even to the Mathis-like climax in a chorale. One more 
serial essay, the String Quartet (1952), is noteworthy for the technical 
feat of preserving (free) twelve-note horizontal successions while dis- 
tributing the series vertically,! but such skill only serves to emphasize 
the faint aura of academicism which surrounds Henze's early instru- 
mental music. 

Academicism implies the mastery of a technique without a propor- 
tionate urge to expression through it. Though much avant-garde music 
is academic, that of a composer like Stockhausen reflects a preoccupa- 
tion with the material's possibilities so ardent as to constitute his most 
characteristic expression. For Henze, the nature of what is to be 
expressed is the stimulus to composition, and this assumes more tangible 
form in a verbal equivalent: the sonorous formulas cited provide less 
pointedly the guidance commonly found in a literary text or a scenario, 
and it was inevitable that he would discover his métier in word-setting 
and the musical theatre. Experience in writing ballet music (Jack 
Pudding reveals the flirtation with jazz which few German composers 
can resist), vocal chamber music like the delicate Apollo et Hyacinthus 
(1949, his most distinguished serial work of this period), and operatic 
essays after Cervantes and Kafka, was summed up in the opera Boule- 
vard Solitude (1951), a reworking of Prévost's Manon Lescaut. Having 
found the key to his most personal utterance in these fields, he was 
reluctant to abandon it in writing for instruments, and the Ode to the 
West Wind (1953) for cello and orchestra follows Hindemith's example? 
in using Shelley's poem as a continuous verbal undercurrent to the 
music. 

The extent of literary indebtedness is exceptional here, but a series of 
instrumental works confirmed Henze's dependence on poetic moods 
(Quattro Poemi, 1955; Drei Dithyramben, 1958). An attitude so redolent 
of the previous century demands the qualities it prized: ‘harmony’ as а 
basic concept (not a product of exceptional textural conditions), contri- 
buting with orchestral colour to sounds considered in themselves 
'expressive', though often used as background to one dominating 
melodic line. Henze's employment of complex chords that, because of 
their constitution from several contradictory triads, have piquancy 
rather than acidity, provides an unexpected extension of orthodox 
harmonic theory—see Ex. 219 from Drei Dithyramben. This luxuriance 
stems from Berg yet avoids turgidity thanks to an ear for orchestral 
clarity as scrupulous and inventive as Stravinsky’s. Henze’s melody 


1 See Henze’s note on this in Rufer, op. cit., p. 185; also Rudolph Stephan, “Hans 
Werner Henze’, Die Reihe, iv, (Eng. ed.) pp. 32-34. 
? Cf. p. 406. 


CONSERVATISM AND COMPROMISE 501 


e b 
== S эл ИИО 


Огсһ. 


is less consistently striking, but his admiration of Italy led him to 
emulate the Verdian cantabile, often most effective when least aspiring, 
as in the engaging Five Neapolitan Songs (1956). Yet these revaluations 
of traditional means did not deter him from profiting by the innova- 
tions of his contemporaries. Comparison of the ground from the 
second Dithyramb (Ех. 220) with an earlier ground (see Ex. 218(iii)(b)) 
reveals how much of the new concept of line and rhythm he has 
accepted. The 1959 Piano Sonata suggests a need to demonstrate his 
freedom of movement in pointilliste textures, even to witticism at their 
expense in the final fugue, while the pantomime The Emperor’s 
Nightingale (1960) exploits the bravura flute writing and exotically 
bright chamber sonorities beloved of the avant-garde as symbols 
particularly appropriate to the legend.! 

Such a range of style need not be disconcerting in an opera composer 
who recognizes the principle of set numbers and the value of parody 
techniques. After the prodigality of Kónig Hirsch (King Stag) (1956) 
Henze curbed a fantasy sometimes too facile in meeting the challenge of 


1 Colin Mason, *Hans Werner Henze', The Listener (17 November 1960), p. 913. 


E "S NIGAME 


502 MUSIC OF THE EUROPEAN MAINSTREAM: 1940-1960 


Ex. 220 


a drama so powerful, and apparently so unsusceptible to musical en- 
hancement, as Der Prinz von Homburg (1960). His association of 
intervals with characters is no guarantee of musical cogency, but deve- 
lopments of material such as the battle scene, a typical application of 
serial principle to free thematicism, later pared down to a twelve-note 
row, reveal how effectively his craft reinforces his dramatic instinct.! 
Henze's admiration for Britten, a composer whose sublime simplicity 
and fleetness he does not naturally share but most nearly captured in 
Kammermusik 1958 (dedicated to Britten), may have prompted his 
chamber opera Elegy for young lovers (1961). In its characterizing 
instrumental colourings and its use of idioms ranging from jazz to ‘post- 
Webernian' vocal fioritura, this summed up Henze's peculiar gifts, and 
emphasized the role that synthesis would continue to play in realizing so 
personal a vision of sound. 


! Diether de la Motte, op. cit., pp. 32-38. 


MUSIC IN BRITAIN 1916-1960 


By ARTHUR HUTCHINGS 


INTRODUCTION 


DURING the years immediately after 1918 music in Great Britain passed 
through a tunnel rarely cheered by gleams of the light into which it was 
eventually to emerge. Until nearly 1930 discerning critics had the uneasy 
impression that Delius and Elgar towered head and shoulders above 
their successors. One of the most intelligent representatives of the 
younger generation, Constant Lambert, in his Music Ho! A Study of 
Music in Decline,! showed the effect of war coinciding with the climax 
of a nationalist movement among artists who had little or no commerce 
with the avant-garde of other countries, for the war had prolonged that 
delay of technical stimuli from abroad which has always flattered 
insular conservatism. The last German romantic who had been hailed 
as ultra-modern, Richard Strauss, was still deeply venerated, but there 
was no comparable cult of French music except among pianists and 
singers, some of whom belatedly recognized Fauré’s subtlety and veiled 
power. Debussy and Ravel were not explored, but were represented by 
frequent performances of two or three pieces which provided a Latin 
sauce to the basic German fare. Sibelius’s symphonies enjoyed a popu- 
larity in Great Britain that had no parallel in Germany but the Strauss 
cult affected not only the importing of other central European music but 
also indigenous composition. 

Native new music offered no technical advance upon Strauss’s and 
did not approach the professional invention or design of even his 
poorest works. The so-called ‘nationalists’ at least offered a different 
flavour; but it was several years before they offered more than songs 
and short vocal or instrumental lyrics, mostly pastoral in character. 

In 1919 Holst and Vaughan Williams re-entered the lists. Holst’s 
suite The Planets, which had been played only privately in 1918, was 
given at a public concert in 1920. Though it was immediately acclaimed 
it could not easily be reconciled with the composer of Savitri (1908) and 
some short choral works which Holst had published before the war. 


1 London, 1934. 


504 MUSIC IN BRITAIN 1916-1960 


Vaughan Williams, whose name and appearance were almost symboli- 
cally insular, had secured a reputation for consistent advance by Toward 
the Unknown Region (1907), A Sea Symphony (1910), the song cycle 
On Wenlock Edge (1909) and the Fantasia on a Theme by Tallis (1910); 
but his London and Pastoral symphonies, first heard in 1920 and 1922 
respectively, were more striking leaps forward. They were symptomatic 
of the persistent British obsession with the symphony. 

While the ambitious young French or Italian composer hoped to 
write a successful opera or ballet, his British counterpart lost time in 
straining to produce grandiose symphonies. Before considering those 
who are held to have succeeded we should consider one who preceded 
them. 


BAX AND HIS ROMANTIC CONTEMPORARIES 

In The Garden of Fand (1916), Tintagel (1917), and November Woods 
(1917) Arnold Bax (1883-1953) had demonstrated his rare command 
of the orchestral palette. His imaginative use of other materials is shown 
in certain chamber works, especially his exquisite Nonet and some piano 
sonatas, and Mater ora Filium (1921), a virtuosic motet for double 
choir. The seven symphonies which he composed between 1923 and 
1939 were well received, especially the Third (1929), which was less 
bitter and turbulent than its predecessors; for Bax, shy even among the 
friends who marvelled at the facility with which he translated complex 
orchestral score into keyboard terms, seemed to release in these sym- 
phonies dark moods and passions, their violence offset by dreamy 
lyricism that suggested exhaustion after protest. Both passages quoted 
here illustrate the last point, though they were not chosen to do so. 
Ex. 221, from the Second Symphony (1925), shows how Bax could 
write appealing melodies that fall just short of memorability and 
distinction: 


Andante 
Strings 


ARNOLD BAX 505 


( 89 y 
IM WEIT — T3 — d 
И ӘД yy BR LT 


Ex. 222, from the Third Symphony, how adroitly he improvised what 
was needed at any point in a long work—in this case the coda to a big 
movement: 


Bax disproves the belief that Tristan or Elektra pushed chromatic 
harmony to a point beyond which classical tonality finishes, or that 
continental composers left little more to be drawn from a large romantic 
orchestra. His weakest music is less banal than Strauss’s but his 


506 MUSIC IN BRITAIN 1916-1960 


strongest, rivalling Strauss in imagination, does not equal him in 
coherence and distinction. Bax loved Celtic literature, the Irish land- 
scape and seascape, and though his symphonies are not overtly pro- 
grammatic their movements begin as if they were intended as tone 
poems evoking Celtic legends, or as music for films set in a Celtic 
atmosphere. 

Bax and Cyril Scott (1879-1970) have been the two most inadequately 
assessed among those of Strauss's and Debussy's British contemporaries 
who are not labelled ‘nationalist’. To the surprise of his compatriots 
Scott was thought by many continental musicians to be 'the leading 
representative of modernism in England', perhaps because his compos- 
ing chiefly for the pianoforte made him easily accessible. Moreover his 
appeal is made almost entirely by impressionist harmony, usually more 
complex than that of Debussy or Ravel, and sounding ‘perfumed’ or 
*exotic' to an aesthetic served by Vaughan Williams. 

Few others who aspired to symphonic poems, romantic symphonies, 
music dramas, or works that would have been acceptable before the 
war in France or Germany are remembered today. The ambitious 
attempts by Joseph Holbrooke (1878-1961) and Rutland Boughton 
(1878-1960) at Celtic emulations of Wagner's dramas are virtually for- 
gotten. A few of the many songs and part-songs of Granville Bantock 
(1868-1946) are still heard, but not his tone poems nor his choral 
works, and he is honoured chiefly because he used an official appoint- 
ment for the encouragement of young musicians. The only exceptions 
are two composers who did not publish anything too ambitious for their 
talents. Frank Bridge (1879-1941) might be known to very few concert 
goers if it were not for Britten's acknowledgement of his indebtedness to 
a fine teacher and mentor, yet so notable a *musician's musician’ as 
Bridge deserves mention. He was a fine enough violinist and viola player 
to deputize in the Joachim Quartet, and his name was often the first sug- 
gested to direct a concert or opera at short notice. Though an admirable 
advance can be traced through the major works he composed before and 
after the war, his music has no strong public personality, but is cherished 
among players of chamber music who recognize his remarkable feeling 
for the character of their instruments. 

John Ireland (1879-1962) also advanced his technique without 
fertilization from Teutonic or Latin sources (except Debussy, who 
influenced his keyboard style). Ireland's are among the most satisfying 
of the many settings of verses from A. E. Housman's A Shropshire Гай; 
and although he is best known by a few conservative works—the choral 
and orchestral These things shall be, which is not far removed from the 
cantatas of Parry and Stanford, or the picturesque А London Overture— 


RALPH VAUGHAN WILLIAMS 507 


his true artistic worth is better measured by the songs from poems by 
Thomas Hardy (1925), the much later piano pieces, Sarnia, and fine 
chamber works, such as the Fantasy Sonata for clarinet and piano. 
Obviously Ireland took least risk with the largest forces: the basically 
traditional harmony of his Piano Concerto, for instance, is merely 
disguised by added notes and other decoration. This caution and his 
small, finely wrought output testify to shrewdness, not insincerity. 


RALPH VAUGHAN WILLIAMS 

Most of the next generation are widely regarded as satellites of Ralph 
Vaughan Williams (1872-1958) whose discovery of a distinctively 
English idiom is said to have provided a catalyst for talents which might 
otherwise have remained ineffective, if not inarticulate, attempting 
to do what Strauss and Debussy had done better. 

Vaughan Williams was older than any of the musicians mentioned 
hitherto, each of whom might have described himself, as Bax did, as ‘а 
hopeless romantic. He was therefore more of a pioneer. We shall 
see that he was partly led by a distaste for what he thought weak 
in late-romantic music. The medical or psychological truth in the 
diagnosis 'a late developer' is not easily judged when the subject is an 
artist whose strong will and financial independence enable him to learn 
or do only what seems to suit his purpose. Vaughan Williams was no 
more docile than Beethoven; but it is foolish to pretend that an English- 
man who in 1909 secured the orchestral effects, the adroit modulations 
and witty technical points of Vaughan Williams's incidental music to 
The Wasps was less than clever—although this was an epithet which he 
disliked by temperament and upbringing. The apparent weaknesses of 
the music—coarse scoring, missing climax, vague transition etc.—may 
be no more due to inadvertence or incompetence than Beethoven's 
unorthodox counterpoint and unorthodox treatment of instruments. 

A clue to the solution of this problem is offered by his choice of words 
for six operas, about fifty choral works, and some hundred solo, unison 
and part songs. This friend and admirer of romantics shows distaste for 
several favourite romantic emotions. He shuns words of a kind sought 
by composers from Schubert to Mahler and by Britten and others today. 
The self-pity and despair which cover a large field of the best English 
verse from Shakespeare to Eliot are passed by; grief and anger are 
represented, notably in Riders to the Sea (1937), but only from the 
mouths of the strong or the brave. Erotic material is for mockery 
or satire, and іп Flos Campi (1925) even the Song of Solomon is sub- 
limated to chastity, its luxury a strong man's vision. There is more 
human feeling than private devotion in his treatment of religious words, 


508 MUSIC IN BRITAIN 1916-1960 


though these are surprisingly frequent in the output of a confessed 
agnostic. In fact his choice and interpretation of words plainly reflect the 
liberai humanitarianism of his family and of Parry, together with the 
ethic of his school mentors. Although openly hostile to imperialism and 
insolent privilege, he paradoxically cherished pride as a virtue rather 
than the first of the Seven Deadly Sins, seeing man as “а being darkly 
wise and rudely great', not by the grace and mercy of God but by hisown 
effort and courage—a conception directly opposed to orthodox Chris- 
tian doctrine. He found the best of humanity either in the strong and 
naive ‘naturals’ of folk-song, unspoiled by urban meanness and 
sophistication, or in idealistic optimists like Whitman who exulted 
above sophistication. Endurance—whether in Job, in Scott of the 
Antarctic, in the bereaved heroine of Riders to the Sea, or in the political 
prisoner—always inspired him, and he supported those who showed it 
even when he could not support their cause. 

His music neither complains nor protests, but reflects the endurance 
of men braver than the ‘stout of heart’, for they are past hope or fear and 
know only the duty to stand, not to understand. One might suppose this 
seeming negation of feeling to be beyond music if it were not that such a 
piece as the fourth movement (epilogue) of the Sixth Symphony (1948), 
pianissimo and senza cresc. throughout, finds what music can convey 
and words cannot. 

Vaughan Williams’s limitations and strength came first from his up- 
bringing, which was fortunate enough to prevent his associating the 
words ‘noble’ and ‘manly’ with philistines. Romantic and impressionist, 
he reacted against what he thought unmanly and weak in romantic and 
impressionist music. This fact, not an archaeological interest, led him to 
old music. The trunk and oldest branches of a tree are the strongest and 
so are the earliest growths of an ascendant musical style. The most 
powerful of effects is the unison; the most stark of harmonies are the 
bare fifth and fourth; the modal melodic cadence which rises by a tone 
seems stronger than one which rises by a semitone from a leading note. 
Triads are strong chords; among the ‘weak’ and over-used are dominant 
sevenths, ninths, elevenths, and chromatics other than those of 
chromatic triads; but ‘accidentals’ cease to be chromatic when they 
become the degrees of empirical or exotic modes and scales. Now as 
in the sixteenth century the strength of counterpoint lies less in rhythmic 
contrast than in the thrust of discord. Hence the ‘marching basses’, 
alike in orchestral movements and in hymns or unison songs by Holst 
and Vaughan Williams, a bass note often chosen for no other reason 
than its recalcitrance with the upper harmony. Investigation of the 
development of Vaughan Williams’s technique shows him instinctively 


RALPH VAUGHAN WILLIAMS 509 


seeking a higher degree of dissonance, though not chiefly violent dis- 
sonance. He cannot be reproached for begetting or reaching high 
tension so early in a movement that he cannot secure climax without 
adventitious vulgarity. 

Although it has been truly observed that the ear cannot always tell 
whether Vaughan Williams's melodies are entirely his or whether they 
are from folk-song, he would surely have found a technique for ‘manly’ 
expression if he had known no old music at all. Vaughan Williams took 
what he recognized as his own in old music. That throughout his life he 
gave far more than he took may be observed from two examples chosen 
almost at random from his music. 

The first is the beginning of the Pastoral Symphony (1922) omitting 
the first three bars: 


Solo violin ge T 
бы ш 


Violas & Horns 


34 


510 MUSIC IN BRITAIN 1916-1960 


In it will be recognized the influence of impressionism and folk melody, 
although only short phrases of melody can (unprofitably) be described 
as in this or that mode, which may well be contradicted by the harmony. 
The second quotation takes us forward thirteen yéars to the fourth and 
most aggressively dissonant of his symphonies, but it is chosen from the 
quietest movement: 


Ex. 224 Vns. [8ve higher] 


= == ЛЕНІ рауға 


Cellos & Basses, pizz. 2 | | 


жі ыы A 
-= Dh ы ш 244444122 
ыыы 132222 


Here only the rhythmic contours in the violin melody show апу соп- 
nexion with folk music. The passage is not difficult to grasp yet it is 
neither diatonic, atonal, nor polymodal. It comes beautifully to rest on 
G, yet G is not its tonic; it does not need more counterpoint to be fully 
effective but it badly needs its context. 

Here, as often when his effect is most clearly achieved, the composer 
has patiently exercised his complex intelligence. The first movement of 
the Fifth Symphony (1943), as original in its own way as the strident 
Fourth, will provide a further example: 


Ex. 225 


RALPH VAUGHAN WILLIAMS 511 


Vlas. Cellos к № 


Қа... гене aa T и на г] 
1 Е 
E } е Ее =. 


Vaughan Williams's reputation abroad was established with the 1909 
song-cycle On Wenlock Edge for tenor, piano, and string quartet, several 
features of which do not recur in his many later songs; his repute in 
Great Britain was notably increased by two works first given in the 
following year at provincial festivals. The choral and orchestral effects 
in A Sea Symphony (1910) sometimes attain a splendour (distaste may 
call it grandiosity) that did not recur even in occasional and exultant 
pieces like the Benedicite (1931). In A London Symphony, we encounter 
instead another quality which became all-pervading in Vaughan 
Williams and may be called remoteness. It touches the symphony only 
in a few places, notably the slow movement, *O vast rondure swimming 
in space’, but we find it in the other admired work of 1910, Fantasia 
for Double Orchestra on a Theme of Tallis. 

Vaughan Williams has made his own not only Tallis's long and 
lovely melody but its descant. The Fantasia and much subsequent 
music seemed severe because it stirred the spirit more than the blood. 
‘Remote’ had been used in judgement of romantic artists whose 
subjects included “014, unhappy, far-off things', the remote figures of 
legend, the remote lands of heart's desire, objects of longing and 
aspiration, hills and landscapes that were romantic precisely because 
they were remote, and horizons that were unattainable. Vaughan 
Williams's remoteness is post-romantic and often extra-romantic, for in 
his music even rage, pain, and desire become removed, no longer 
subjective. His aesthetic as much as his ethical beliefs required restraint 
of the immediate cry; man should endure these emotions, contemplate 


512 MUSIC IN BRITAIN 1916-1960 


them and recognize with awe their universal power. Little wonder, then, 
that he was attracted by the story of Job. 

Job: A Masque for Dancing (1931): resembles Petrushka, in that its 
conception embraces two planes which must perforce be separate; and 
just as the same figures in the most corporeal of the arts must here be 
seen on both planes, so the same music makes a direct physical impact 
yet encompasses a drama too vast to be contemplated until it is made 
sufficiently remote. The physical appeal of this music is not just an 
effect of its combination with stage scenes and personages, for it sounds 
as fine when taken out of the stage pit to the concert hall. Here its 
climax of terror—Job’s first vision of heaven with Satan enthroned—is 
enhanced by a sudden crash of the full organ, a grim ‘deadweight’ use of 
this instrument without parallel except in the third movement of 
Sinfonia antartica (1953) where it adds the expression of diabolical 
hostility to that of the vastness and remoteness of the polar regions. 
The music of Job is far from disembodied and does not exclude the 
pulsation of dancing, but the rhythms of pavane, galliard, sarabande, 
and minuet impart a statuesque dignity rarely attempted in the theatre 
after Gluck's tragedies. 

Vaughan Williams supplied for Job a more admirable score than any 
since Flos Campi, a suite for viola solo, small orchestra and wordless 
chorus. Unconventional alliances of instruments, unbalanced or bizarre 
textures, and opacity are here calculated effects. The inexact unison 
between muted brass and reed instruments conveys Satan's baleful 
exultation, the oily saxophone tone the false commiserations of Job's 
comforters; and the thickening of harmonies by divided cellos gives 
what would otherwise be too courtly and human a ‘Galliard of the Sons 
of the Morning' the kind of impression produced by the angels on a 
Byzantine mosaic or Romanesque tympanum. 

There is a wealth of melismatic rhapsody in Job outside the ‘Dance 
of Youth and Beauty', which recalls Тйе Lark Ascending "(1914). 
The composer's rhythmic sense and wit were bent away from classic- 
romantic contours, and posterity may yet praise him for his pioneer work 
in showing that rhythm is not metre. If one of his movements seems to 
lose grip, the fault is rarely attributable to loss of rhythmic interest. 
He possessed the symphonist's ability to compose coherent and large 
movements rich in ideas, to develop with fine rhetoric a single thought, 
and to expand themes by other means than classical metabolism. 

It would be easy to distinguish grades of workmanship if whole move- 
ments could be regarded as failures—the somewhat facile cavatina in 
the Eighth Symphony (1956), for instance, or the movement originally 


1 See pl. VII. 


ШЕКЕ МҮ) 


uopuo7 “әлғәці әдридшед J} је uononpoad 28015 [ешо 3y} тор usisop e Jo 1елэле чэмо Ад sunured y 


(21674 225) 1661 ‘ОГ S.SSINVITIIA NVHONVA 


GUSTAV HOLST 513 


associated, in a film, with seals and penguins in Sinfonia antartica. But 
often the best of Vaughan Williams's music is inextricably entangled in 
floundering growth. The finale of the Fourth Symphony, for example, 
very clearly sets out upon the classical sonata plan, with first and second 
groups of strong personality. Then, where we expect a sonata-coda, 
comes the astounding fugal epilogue upon the chief motto-idea of the 
whole symphony, savagely devouring the sonata themes as counter- 
subjects. 


GUSTAV HOLST 

Gustav Holst (1874—1935), Vaughan Williams's contemporary and 
friend, was born at Cheltenham and sprang from a milieu of profes- 
sional musicians very different from Vaughan Williams's. He was “а 
musician's musician' because he was fascinated by the technique of the 
art, indeed by any musical sounds or music from any race. Though his 
technical explorations were made at the prompting of creative desire 
they are often held to have hindered its realization. He produced no 
series of symphonies, nor indeed any large-scale undertakings which can 
easily be examined as a series. It is significant that Stravinsky fascinated 
Holst, who may well have had genius enough (if he had been born into 
another tradition and granted health and freedom) to compose equi- 
valents of The Fire Bird or even The Rite of Spring, though he rarely 
achieved the effortless coherence found even in smaller Stravinsky works 
like Apollon Musagéte. Could the composer of Ode to Death (1919), and 
The Hymn of Jesus (1920) have achieved equivalents of А Symphony of 
Psalms, Oedipus Rex, and Threni? The question is less vain than most of 
its kind, because we cannot claim outright for Holst those symphonic 
qualifications that were conceded to Vaughan Williams. However great 
our admiration of Holst's achievement, in circumstances that would 
have frustrated most men, it seems clear that he commanded his musical 
materials better than their germinating processes, and could have found 
a convincing synthesis in more than two or three works if he had lived to 
be less interested in fertilizers than in the ground to which they were ap- 
plied—often too liberally for immediate absorption. Vaughan Williams 
evolved large-scale contexts for ‘tune’ (as distinct from prose-like line) 
which was derived from folk-song, though often neither bucolic nor 
deliberately insular. Holst, having found in folk-music and modes, or in 
empirical and eastern scales and rhythms, an escape from Anglo- 
German melody, continued with explorations that produced wholly 
individual flavours. Yet when he wanted a ‘tune’ he continued to the 
end of his life to introduce the contour of folk-song or dance into 
refractory and recalcitrant textures. 


514 MUSIC IN BRITAIN 1916-1960 


It seems fair to illustrate some of these points from The Hymn of 
Jesus. Near the opening the music passes from ‘Pange lingua’ to ‘Vexilla 
regis’ with an easy mastery remarkable from a British composer in 1917; 
the effect at the first performance in 1920 was as electrifying as that of the 
subsequent full chorus: i 


Lento 


4 HE - 345 14.5: 


After a verse from each hymn іп unaccompanied unison the orchestra 
makes its next transition: 


Ex. 221 


[m 


GUSTAV HOLST 515 


к 


(Collies SSS 25 


A short passage of routine imitation leads to the first of many six- 
four chords, then more imitative work on the first half of the motive, the 
harmonies becoming just sufficiently biplanar to avoid commonplace. 
A few bars before the much-quoted biplanar ‘To you who gaze, a lamp 
am Г: 


Ex. 228 
[Orchestral and lower voice parts omitted] 


О 4 
Ga’ 22 таа eem A ЕН 
-Жы-ы-н poe cuum ЭЙ М ие eme ч Блез; cam — ВЕН 


Semi Chorus 
апай ©һо па ш т — ВА 


2nd Choir 


Organ bass 


516 MUSIC IN BRITAIN 1916-1960 


To уоп that know, a 


you who knock, a door кт To -уош who fare, the 
ES. EI 141525: — f БП Юю]? ДГ, 


E 627% сар EINST ЕНЕВ CAL ГАР БТ QC =] 
ЕЕ ЕЕ Е: ЕЕ СЕТЕ НЕЕ Е ЕЕжЕ ЕЕ 
[NU TE A ае Раа 


[63:4 —— Bo = een] 
Е MA. o ee res 
2n аа SS Se 
SS SS eee —— 3 sepes] e e 
go- O 2 Ө Ө 
- = SS 


we find harmony, in itself not despicable, that belongs to Stanford’s 
Anglican church music. 

Analogies with Pelléas et Mélisande and Oedipus Rex suggest that 
Holst’s unification would have been less easy if the text (from the 
apocryphal Acts of St. John) had been too slavishly followed from 
image to image, or if the emotional content had been full-blooded 
instead of allusive. Holst himself, for all his avoidance of publicity, was 
no more the recluse, no more austere than many a man whose work 
nobody has thought frigid; yet people once called him cold, severe, or 
‘mystical’. If a mystic is one who expresses experiences that are incom- 
municable except by a language of symbolism it is difficult to see why 
Holst is more mystical than a host of others. His art is ill-suited to the 
sort of subjects that appealed to Verdi, but he is more often mysterious 
than mystical; and the more forbidding or abstract his theme the 
stronger his brush. ‘Saturn, the Bringer of Old Age’ and ‘Neptune, the 
Mystic’ are the best of The Planets,and Egdon Heathis his finest orchestral 
work. This reflection poses what is surely the ultimate problem about 
Holst, and it concerns the man rather than his technical experiments. 

The Hymn of Jesus was greeted with an enthusiasm that bewildered 


GUSTAV HOLST 517 


the composer. He was unwise enough to concoct his own libretto and 
indulgein elaborate parody of German and Italian opera for Тйе Perfect 
Fool (1923); but in the same year the brilliantly scored ballet music from 
this opera, together with The Planets and А Ғиса! Concerto raised the 
composer to a peak of popularity. Then, during the twelve years left to 
Holst, his new works were treated either with curious respect or frank 
dislike. His daughter! records such press comments as: 'the chilly 
vacillations of its harmonies, where cerebration tamed and bridled in- 
spiration’—referring to the Choral Symphony, for which even Vaughan 
Williams felt only “а cold admiration’. Ву 1931 Holst ‘was always feeling 
exhausted’ and ‘dreaded that his ideas were drying up'.? 

Nevertheless the orchestral piece Egdon Heath (1927), the Lyric 
Movement for viola and orchestra (1933) and Scherzo for orchestra 
(1935) already show such ‘third period’ features as a simpler harmony 
and a new integration of purpose and style. Undoubtedly what we must 
call vision, for lack of another term, seems to have become intense and 
steady in Holst's last major works. Though Egdon Heath evokes Hardy's 
description (in The Return of the Native) of an upland tract of Dorset 
that is ‘mysterious in its swarthy monotony’, the music itself does not 
sprawl or flounder. By the clock this is a short piece that merely suggests 
man’s reaction to the vast, inhospitable landscape. It does not relent to 
entertain the musician with points of technical interest as most of 
Holst’s works certainly do. Imogen Holst says: ‘There is no hint of exile 
in the loneliness of Egdon Heath: it is а home-coming.'? Clear vision 
betokens the genuine mysticism that is less concerned to suggest 
mystery than to pierce it and reveal the truth which it obscures. 

Does the steady vision alone support belief in a ‘third period’? Holst 
defies chronological classification. Few of his published works fit into a 
‘first period’ that reflects as much of the nineteenth century as Beet- 
hoven’s did of the eighteenth; and if penetration into regions (not 
necessarily of the mystic) where nobody can guide him puts a man’s 
music into a ‘third period’ then most of Holst’s major works belong to 
it. Other classifications than the chronological are also unsatisfactory. 
To speak of his ‘Sanskrit period’ is misleading unless we mean only that 
between 1907 and 1912. Holst taught himself to read Sanskrit and 
composed the Hymns from the Rig Veda and the one-act opera Savitri 
(from the Mahabharata) to his own translations. Their musical features 
—the spare texture, the recourse to quintuple and septuple time signa- 
tures, the reticent declamation—are not just of one period. His leaning 
towards the arabesque of eastern melodic lines does not belong only to 


1 Imogen Holst, Gustav Holst (London, 1938), pp. 115-6. 2 Ibid., p. 149. 
3 [dem, The Music of Gustav Holst (London, 1951), р. 101. 


518 MUSIC IN BRITAIN 1916-1960 


one stage in his output or only to works like Beni Mora and Two Eastern 
Pictures. He needed it as he had needed folk-song, because its scales 
and rhythms took him away from convention. (There is no actual rága 
music in Holst's work as there is in Padmávati for he was not, like 
Roussel, a connoisseur of oriental music.) 

William McNaught wrote in 1939:! 


... Holst occupied himself with intricate textural problems that seemed to 
the outer world to be of little importance. A typical instance was a three-part 
vocal canon with a different key-signature to each рагі... Either openly or 
implicitly he was given to the solving of out-of-the-way problems of his 
craft that kept him to the borders of the true road of music. His tremendous 
creative power was partly frittered away by this preoccupation with artifice, 
and it was only because the diminished impetus of that power was so great 


that he made his impression. а 


Some years later Wilfrid Mellers? wrote at length about the debt of 
subsequent musicians to Holst's treatment of the English language. 


Holst did not see in the folk-song cult any wistful reversion to а simpler form 
of existence . . . he was interested іп [folk-song] because words and tune had 
grown up together. . . . His own melodic idiom has affinities with that of 
English folk-song . .. | | more ‘primitive’ but completely unsentimental, 
it arouses expectations which it does not fulfil . . . Holst does sublimate 
speech into a phrase of some considerable extent, but not to the pitch of song. 


The lyricism compatible with polyphony іп Weelkes, Dowland, 
Purcell and many more of Holst’s heroes seems to have been shunned. 
Within Holst’s part-song settings from Robert Bridges are found some 
of his rare foreshadowings of the polyphonic ease assumed by his 
pupil Rubbra. 

This austerity would be understandable in a minor composer for 
whom words meant more than line and harmony, rhythm and texture, 
but Holst was primarily interested in ‘pure’ music, music that did not 
duplicate what words could say, but began where words failed. We are 
left asking ‘Why? after reading Mellers’s perceptive comments :3 


The link with folk-song still prevails—for instance in the pentatonic feeling 
... but both the rhythmic plasticity and the tonality have acquired a sub- 
dued wavering instability which gives to the line its coldly desolate effect . . . 
His unique position, his oddity as a cultural phenomenon, consists precisely 
in his allowing the prose phrase integral expression without trying to 
emotionalize it or to compensate for it with a sensuous harmonic vocabulary. 


And our ‘Why?’ is implied by the severest of all his critics, the composer 


1 Modern Music and Musicians (London, 1946), p. 47. 
2 Wilfrid Mellers, Studies in Contemporary Music (London, 1948) p. 145. 
3 Ibid., р. 147. 


PETER WARLOCK AND HIS CONTEMPORARIES 519 


himself. Imogen Holst tells us that, listening to Schubert's C major 
Quintet in 1930:! 


he realized what he had lost, not only in his music but in his life. He could 
cling to his austerity. He could fill his days with kindliness and good humour. 
He could write music that was neither commonplace, unmeaning, nor tame. 
And he could grope after ideas that were colossal and mysterious. But he had 
missed the warmth of the Schubert Quintet. At the moment, it seemed as if 
this warmth might be the only thing worth having. 


Holst's peculiar psychosis of austerity was purely artistic. It did not 
belong only to the last years when he sat ‘huddled over the бге... as if 
the spirit itself were numb'. Savitri dates from 1908, before the worst 
onset of neuritis or the effects of concussion after a fall, but its lack of 
sensuous appeal has earned it a reputation as “ап opera for spiritual 
and intellectual aristocrats’. 


PETER WARLOCK AND HIS CONTEMPORARIES 


The three or four men who produced distinctive work during the 
decadence of the nationalist movement also refreshed themselves 
directly from folk-music and the works of Tallis, Byrd, Weelkes, Wilbye, 
Dowland, Gibbons and Purcell. The limited achievement of E. J. 
Moeran (1894-1950) was exclusively lyrical, partly because he studied 
the violin as a schoolboy and partly because he had absorbed the 
Norfolk folk-songs which he collected. The best of his original songs 
approach the quality of Warlock’s and his arrangements of Norfolk 
songs are neither timid nor clumsy. In his works more ambitious than 
songs none rises above the commonplace unless the original lyric gift 
is to the fore. His Symphony in G minor (1937) fails to fulfil the promise 
of an eloquently melodic opening; subsequent ideas lose character in 
turgid passages that seem to derive their menacing moods from Sibelius. 
On the other hand Moeran's Violin Concerto of 1942 has few longueurs 
because instead of an imposing façade it offers simple geniality. 
Moeran’s attempts to emulate other men’s complex harmony or 
structure betray the student, not the master; when he was wholly 
himself, as in two pieces of 1931 for small orchestra, he could realize 
forms of quiet beauty that maintain their appeal. The second of these 
two pieces uses the Norfolk song ‘Lonely Waters’ which he had already 
issued for voice; the other, ‘Whythorne’s Shadow’, pays tribute to a 
beautiful song from Thomas Whythorne’s collection of 1571.? 

Twelve of Whythorne’s songs had been published in 1927 by Philip 


1 The Music of Gustav Holst, p. 142. 
? See Vol. IV, pp. 84 and 200. 


520 MUSIC IN BRITAIN 1916-1960 


Heseltine (1894-1930) whose own music was issued under the name 
Peter Warlock. While still a boy he was moved deeply by the music of 
Delius, with whom he became personally acquainted through a relation 
who lived near Delius at Grez-sur-Loing. Later he became an enthusias- 
tic disciple of Bernard Van Dieren (1884-1936) who migrated from 
Holland in 1909 and became 'an enigmatic and portentous figure in the 
background" of London music between the wars. Under his own name 
Heseltine published books on Delius, Gesualdo,? and ТЛе English Ayre. 
Under his pseudonym he brought out editions of Dowland's Lachrimae, 
Locke's and Purcell's fantasies and other consort music, and numerous 
songs and ayres of the sixteenth and seventeenth century. In his own 
compositions archaism and preciosity could sometimes be indulged with 
overt panache and wit, as in the Capriol Suite for string orchestra (1927) 
which uses dance tunes from Thoinot Arbeau's Orchésographie of 1589, 
or in the piquant accompaniments to songs of sly gallantry and tavern 
roistering. He could set a nursery or nonsense verse (as in the albums of 
1923 called Peterisms and Lilligay with the consummate mastery that 
he expended on verses from Shakespeare. 

These veins of his muse are associated with his gnomic pseudonym, 
but from the first songs which he published during the First War to the 
last set of 1929—30 there came from time to time the expression of pity, 
melancholy, or passionate devotion unparalleled in the setting of 
English poetry since Purcell. His tragic expression owes its effect to 
economy, and his taste for exactitude and orderliness is manifest even 
in the frenzied nausea of “Take, О take those lips away’. This short song 
perfectly catches a climactic moment in Shakespeare, yet does so largely 
by a superb use of rich Delian harmonies that sound over-ripe and 
bitter and suggest a passion that has been poisoned. Among the finest 
of his sombre songs are two in his last four, settings of Bruce Blunt's 
poems “Тһе Frostbound Wood’ and “Тһе Fox’. This set also includes 
the exquisitely tranquil ‘Bethlehem Down’. Perhaps his supreme 
achievement lay in works for solo voice with a small group of instru- 
ments. A setting of Yeats's “Тһе Curlew' for tenor voice, flute, cor 
anglais, and string quartet, which foreshadows Britten's skill in dealing 
with a chamber group and solo voice, is surely one of the most beautiful 
evocations of desolate melancholy produced outside opera by any 
musician since Schubert. 

Poverty, malnutrition and tuberculosis, following the effects of 
wounds, gas and shell-shock, deprived first of his reason and then of his 
life Ivor Gurney (1890-1937), a lyric poet of talent and a composer 


1 McNaught, ibid., p. 53. 
? Carlo Gesualdo, Prince of Venosa: Musician and Murderer (London, 1926). 


ARTHUR BLISS 521 


who excelled in the setting of English words. His musical remains are 
insufficient in bulk and maturity to justify speculation as to whether he 
would have reached Warlock's command of other accomplishments 
than a wonderfully sensitive vocal line. Gurney's malady overtook him 
before he had mastered the integration of vocal melody with harmony 
and other points of technique, despite striking passages in which chords 
or modulations perfectly fit verbal imagery. Few of the poems he wrote 
while serving in France were set to music. An exception, 'Severn 
Meadows', made his finest song which, with some of his Housman 
settings, e.g. *Loveliest of trees’, ‘Desire in Spring’, ‘Lights Out’, and 
‘An Epitaph’, justify an echo of Grillparzer’s tribute to Schubert: ‘a rich 
treasure, but still fairer hopes’. 

Gurney’s friend Herbert Howells (b. 1892) was, like him, a Gloucester 
chorister who earned a scholarship at the Royal College of Music. His 
affection for the cathedral atmosphere has elicited a large number of 
anthems, motets, and canticles, some dedicated to particular founda- 
tions, as well as more ambitious works for voices and orchestra, 
such as Hymnus Paradisi (1950) and Missa Sabrinensis (1953). His music 
for the concert-hall is less familiar than his rhapsodic organ pieces and 
his two suites for clavichord. The fastidiousness of his textures, with 
their restless polyphony, makes exacting demands, but both instru- 
mental and vocal effects are calculated with the imagination of a 
craftsman in filigree. 


ARTHUR BLISS 

Among composers who had already shown promise before the end 
of the First War was Arthur Bliss (b. 1891), one of the generation of 
Stanford pupils which included Howells and Gurney. He has been given 
insufficient credit for initiating before 1920 the reaction associated with 
the twenties in works musically superior to many which achieved a 
succés de scandale. Bliss lacks a personal idiom. He has therefore been 
most successful in works where strong personality is less important than 
deft technique, the securing of telling effects and a high order of general 
intelligence—in the films The Shape of Things to Come (1935), The 
Conquest of the Air (1937), and Men of Two Worlds (1945), in the 
incidental music to several plays, and in the ballets Checkmate (1937), 
Miracle in the Gorbals (1944), and Adam Zero (1946). 

His career as a composer has been unusual because, instead of moving 
farther and farther away from his teachers, he seemed by the end of the 
twenties to turn towards their ideals. It was in these works that he 
revealed his full calibre and not just his enviable facility and humour. 
Among them are Pastoral: Lie Strewn the White Flocks (1928) and 


522 MUSIC IN BRITAIN 1916-1960 


the admirably controlled yet rhapsodic Clarinet Quintet of 1931. The 
Quintet is manifestly English, whereas the Pastoral draws most appro- 
priately upon the technique and flavours of the nationalists in settings 
for solo and chorus, with flute, strings, and percussion, of poems 
by Jonson, Fletcher, Theocritus, and others. Bliss lets the singer 
bear the main responsibility for details of interpretation and uses the 
instruments with reticence, as in his background music for films. That 
is probably why his choral symphony Morning Heroes (1930) moves 
without discrepancy from a portion of the Iliad (Hector's farewell to 
Andromache) to poems by Whitman, Li-Po, Wilfred Owen, and Robert 
Nichols. The work requires an orator with chorus and orchestra. 

The best of Bliss's earlier works, Кош for soprano and ten instru- 
ments, using nonsense syllables purely for their sound (without the 
associative subtlety of Edith Sitwell's verses that were issued chiefly 
as studies in rhythm) and his Conversations for three string and three 
wind instruments, both date from 1919 and represent more than 
reaction against the amplitude of the romantic orchestra. The chamber 
groups and the employment of the voice as another instrument were 
means to musical wit in the strict sense of the word; every point, 
humorous or serious, is rooted in the forms and processes of music. 
There is wit not only in “Тһе Committee Meeting’ and ‘In the tube at 
Oxford Circus' (Conversations)! but also in the Colour Symphony (1922) 
which is serious, even romantic, for the four movements evoke the 
symbolic associations of purple, red, blue and green respectively: 
and Bliss’s wit does not forsake him in more sinewy and classicist 
textures like the later Music for Strings (1935) and his two quintets (1931 
and 1937). A piano concerto written for the New York World Fair 
(1939) is a clever attempt to revive the grand manner and opulent scale 
of the late nineteenth century. 

Lord Berners (1883-1950) succeeded to his barony in 1918 but, as 
Gerald Tyrwhitt, had already published such musical facetiae as Trois 
petites marches funébres (for a statesman, a canary, and a rich aunt) 
(1914), and the parodistic Lieder Album (1913) and Valses bourgeoises 
(1917). Later he achieved originality in ballets which include general 
parody of nineteenth-century music and manner, notably The Wedding 
Bouquet (1936) with his own French provincial sets and costumes and 
a chorus to a text devised by Gertrude Stein for its phonetic effect. Тйе 
Triumph of Neptune (1926) was produced by Dyagilev with nautical 
scenes devised by Sacheverell Sitwell from a Victorian toy theatre, and 
Luna Park (1930) was included in one of Cochran’s revues. 


1 Both recorded in The History of Music in Sound, x. 


ALAN BUSH 523 


ALAN BUSH 


Alan Bush (b. 1900), an important influence upon adolescent talent 
and judgement during the twenties and thirties, received his chief 
professional training in Berlin, and his leadership before the war is 
attributable less to his composition than to his strong personality and 
didactic power. He has been compared with Hindemith, sometimes 
ineptly, for though he respects (as fulfilling an obligation to society) 
what he calls “educated music’ which may not be treasured by posterity, 
he has formulated no harmonic-structural system to serve the composer 
who needs premises. Consequently pupils do not need to unlearn Bush’s 
teaching in order to follow examples as diverse as those given by 
Stravinsky, Webern, Blacher, or Martin. 

Bush’s music is polyphonic, and his first notable success was achieved 
with one string quartet (1924) which won a Carnegie award, and another 
(Dialectic, 1929) is still considered his masterpiece. If he sometimes 
achieves only Kapellmeistermusik, of a solidity which suggests the 
church musician rather than the 'people's composer’, it is sometimes 
his commission that has prevented his using strings or voices in a first- 
class demonstration of thematic argument; sometimes, as in the choral 
finale of his Piano Concerto (1937) he cannot rise above an ephemeral or 
naive text. The best of Bush’s instrumental works, such as the C major 
Symphony (1940), Lyric Interlude for violin and piano (1944), and 
Violin Concerto (1948), commended themselves to their first admirers 
on more valid grounds than a technique not quite like that of Raws- 
thorne, Rubbra or any other Englishman. Why, since he admired 
Schoenberg and actually used twelve-note themes in two movements of 
his C major Symphony, did Bush decline to join the radical serialists? 
Plainly for reasons which are bound up with his acceptance of Soviet 
composers’ manifestos concerning proletarian understanding of 
ambitious music. Possibly Bush’s style would have become less complex 
quite apart from his political conscience; but whatever the explanation, 
this simplification furthered the composition of Bush’s operas, whose 
heavily slanted librettos have won them success in East Germany— 
Wat Tyler (1951) and Men of Blackmoor (1955). Musically these works 
are chiefiy distinguished by their strong choral writing, foreshadowed 
by The Winter Journey of 1946, a cantata in which Bush treats the 
Christmas theme non-religiously. 


CONSTANT LAMBERT AND WILLIAM WALTON 


The first publications of three composers of the next generation— 
Constant Lambert (1905-51), William Walton (b. 1902), and Lennox 


524 MUSIC IN BRITAIN 1916-1960 


Berkeley (b. 1903)—associated them with the reaction against patriotism, 
romanticism, and high moral seriousness. Yet these young composers of 
the 1920s were more considerable than any of ‘Les Six’ except Honegger. 
However questionable the opinions in Lambert's Music Ho!, their 
bitterness is passionately serious; and so, beneath sophisticated manner- 
isms, is Walton's poignantly pessimistic lyricism, notably in the slow 
movement of his Sinfonia concertante for orchestra with piano (1927) 
and the outer movements of his Viola Concerto (1929). 

In 1926 Lambert, still a student at the Royal College of Music, was the 
first Englishman whom Dyagilev asked to compose a ballet. Romeo and 
Juliet is set in a dancing school and has a ‘rehearsal’ plot culminating 
in an elopement by aeroplane. Young Lambert's eighteenth-century 
dances, with spicings and occasional jazzings, proved him an adroit 
rather than an original composer, for the work called for parody and 
pastiche. It immediately secured the commission for Pomona, also with 
choreography by Nizhinska, but without farcical elements. To judge only 
from the concert suite, Pomona (produced in Buenos Aires early in 
1927) presented its pastoral myth in a manner as nearly ‘straight’ as 
does, say, Apollon Musagéte; certainly it gave scope for mannered yet 
vital melody, presented less with contrapuntal device than with attrac- 
tive contrapuntal features. Despite the large orchestra required, this 
athletic harmony and instrumentation commended Music for Orchestra 
(1927) to its first audience; its urbanity seemed clean, yet neither sterile 
nor made palatable with pseudo-sentiment from the jazz world. Some 
thematic material from Music for Orchestra (Ex. 229), shown with a 
short specimen of word setting from Summer's Last Will and Testament 
(1935) (Ex. 230), will suggest the quality of Lambert's mind: 


Andante 
Vns. 8ve higher and muted 3 
EUER х 


(i) 


525 


CONSTANT LAMBERT AND WILLIAM WALTON 


Allegro 


es es ДАВА pl 


yp qe я 
D —i 


E ЕЕ лду [ЕРИНО mcr Бесті 


[e$ V | 


Allegretto 


Ex. 230 


thanin a 


be 
и bd | 


less 


va 


Al good things 


> 


4 


fi 
| 
| 


bbs 


2и е0 ре 
at bad 


plea - sure 


plen - ty, 


Peace, 


day, 


35 


526 MUSIC IN BRITAIN 1916-1960 


He was a polymath who occasionally wrote incidental or film music, 
transcribed old music, made arrangements for ballets, and was so fully 
employed as a conductor and writer that his only other considerable 
orchestral work was Aubade héroique, finished in 1942 but suggested by 
an early morning of 1940 when the Vic-Wells Ballet Company was 
escaping from Holland before the German invaders. 

Before appearing as composer and conductor Lambert had been the 
speaker in his friend Walton's Façade (1923), an entertainment devised 
as a Gesamtkunstwerk, in which Edith Sitwell's verses, described as 
*studies in rhythm, sound, and association of 14еа5, were spoken 
through a megaphone in a screen which prevented any distraction caused 
by the appearance of the speaker or of the instrumental ensemble 
which played Walton's pieces. The score, virtuosic both in its original 
form and in the full orchestral dress of its concert suite, and Walton's 
brilliant Sinfonia Concertante (1927), which included brittle rhythms of 
the kind associated with Milhaud's confections and some of jazz 
provenance, were largely responsible for the coupling of Walton's name 
with Lambert’s. These early works have been disparaged as ‘highbrow 
jazz’, a defensive growth against the pathetic quasi-moral Philistinism of 
schools or the drab and even more tyrannical Philistinism of democracy. 
*The perverse highbrowness which indulges in the low, having gone be- 
yond the primitive earnestness of the romantics’,! expressed during the 
twenties the thwarted desires of so large a section of urban society that 
in Germany and Russia government officials and indoctrinated artists 
fulminated against it. Its frivolity, as much as the querulousness of later 
generations, repels balanced minds. But Walton and Lambert were 
convinced of the need to doff the prophetic mantle, to learn from light 
music and jazz, just as the nationalists had learned from primitive and 
folk-music; and Walton and Lambert wrote very little inferior music 
before they passed to a less inhibited revelation of romantic earnestness. 

It is mistaken to suppose that Ғасайе is inferior Walton or to recog- 
nize the ephemeral superficies of the Sinfonia Concertante and not the 
poignant beauty to be found chiefly, but not only, in its slow movement; 
to suppose that Lambert published his Piano Sonata (1929) and his 
Concerto for Piano and Nine Instruments (1931) only to be fashionable, 
or to regard Walton's acid harmonies as *wrong note' effects like those 
in the more vulgar pieces of Prokofyev and ‘Les Six’. Though Walton 
proved more romantic and rhetorical, Lambert as a composer was either 
intellectually more serious, as in Horoscope (1937) with its elaborate 
musical palindrome, or emotionally more serious, as in the Dirge from 


* Chapter on Walton by Colin Mason in British Music of Our Time, ed. Alfred Bacharach 
(London, 1946). 


CONSTANT LAMBERT AND WILLIAM WALTON 527 


Cymbeline (1940) and Summer's Last Will and Testament (1935) with its 
‘King Pest’ rondo and the culminating ‘Sarabande’: 


Queens have died young and fair; 
Dust hath closed Helen’s eye: 

I am sick—I must die. 

Lord have mercy on us! 


Lambert scored one extraordinarily popular success in 1927, with 
a setting of Sacheverell Sitwell’s The Rio Grande for chorus and 
orchestra, ineptly described as symphonic jazz; not even the brilliant 
solo piano part or the ‘damask’ blues chords in the quiet middle 
section (‘The noisy streets are empty and hushed is the air’) keep to jazz 
idiom which, according to the composer, ‘is a more plastic basis than 
folk song or pre-jazz popular song. Jazz, like much exotic music, 
depends more upon rhythmic and melodic inflection than upon a 
square-cut scheme.’ Lambert was aware of the limitations of standard 
jazz syncopations and jazz harmony, not altogether unlike the de- 
ficiencies in all but the best music of the nineteenth-century German 
romantic school, as Wagner himself was aware. Beethoven’s scherzos 
were a serious composer’s development from minuets ‘as rhythmically 
effete as commercial jazz’ and Walton’s Portsmouth Point (1925) in- 
cluded jazz rhythms but produced ‘an atmosphere as far removed from 
Harlem’ as Beethoven’s was from the eighteenth-century dancing room. 
‘The jazz composer is now stagnating, bound to a narrow circle of 
rhythmic and harmonic devices and neglecting the possibilities of form. 
Itis for the highbrow composer to take the next step.’ Lambert shrewdly 
noticed the ineffectiveness of many attempts to bring jazz into the 
concert room; and he particularly disliked composers like Gershwin 
who used ‘only the non-barbaric, non-vital elements . . . Jazz is not 
raw material but half-finished material in which European sophistication 
has been imposed over coloured crudity. There is always danger that 
the highbrow composer шау... leave only the sophisticated trappings 
... asin the Rhapsody in Blue.) 

Perhaps the clearest statement of his belief lies in the words: 'I see 
no reason... why a composer should not be able to rid himself as much 
from the night-club element in jazz as Haydn did from the ballroom 
element in the minuet, and produce the modern equivalent of those 
dance suites of Bach which we treat with as much seriousness as the 
sonatas of Beethoven.’ 

The development of Walton’s own temperament revealed a strain 
of brooding and melancholy at odds with the sharp wit and smart 


1 Music Но!, p. 228. 


528 MUSIC IN BRITAIN 1916-1960 


exterior features of his style. This is seen notably in his deservedly 
popular dramatic oratorio, Belshazzar’s Feast (1929), upon a text skil- 
fully arranged by Osbert Sitwell from the biblical narrative. Walton, like 
Rubbra, Rawsthorne, and other British musicians born between 1900 
and 1905, found in traditional quarries a seam scarcely large enough for 
one man's needs throughout a normal working life. He was put to un- 
usual labour whenever he avoided the obvious recourse to older seams 
with which he made shift in commissioned and ceremonial music, and 
his completion of major work has always been slow; but British 
musicians of Walton's generation were more talented than most of their 
immediate predecessors, and it is mistaken to attribute their restricted 
range entirely to limitations of musical gifts. Without abnormal 
competence they had little chance of a hearing. If they could not 
appear original by adopting foreign styles they could do so only 
by idiosyncracy of style. This Walton commanded without affectation 
and cultivated in his best work by genuine compulsion. He begins 
the slow movement of his First Symphony (1935) with a melody as 
poignant as Ex. 231: 


Ex. 231 


and, despite echoes of its contours, throws it away until it is used as an 
epilogue. He presented impassioned music like: 


529 


CONSTANT LAMBERT AND WILLIAM WALTON 


and vehement music like Exx. 233 and 234 (also from the First Sym- 


phony): 


Ex. 233 


530 MUSIC IN BRITAIN 1916-1960 
Ex. 234 


5 
> => 
Maestoso > >> > ee = 


with detachment, sometimes implied amusement, remains urbane even 
in the haunting close of the Viola Concerto (1929), embedding the 
jaunty first theme of the finale within the soft valedictory texture, 
offsets con malinconia with con malizia, reconciles Stravinskyan subleties 
of accentual rhythm with long-phrased melody never commanded by 
Stravinsky, finds a place in symphonic structure for fastidiously sensuous 
orchestral effects, even to finely judged dynamics for Chinese block, 
castanets, and minor percussion, and was not afraid of the taunt 
that he reflected the taste of a sophisticated intelligentsia. There is no 
insincerity in Walton's style, which is true to the man, known for taci- 
turn good humour and the sardonic wit that informs his scherzos. 
Walton's occasional additions to his musical vocabulary, for example 
in his Cello Concerto (1959) or the passacaglia-finale of his Second 
Symphony (1961), reveal no significant refreshment of his powers, and 
he has not always succeeded in attempts to make his style serve a wider 
range of expression than it did before the war. In his opera Troilus 


EDMUND RUBBRA 531 


and Cressida (1954), for example, the choruses and the witty music for 
Pandarus showed the familiar Walton to his first operatic audience. 
But his music for the protagonists rarely rises to lyricism as poignant 
as in his best purely instrumental music, and hardly ever to ecstasy. 
He often seems rather to be translating into his own idiom the responses 
by which successful opera composers before him met the situations 
requiring different moods and aspects of the grand passion. Aware both 
of increased skill and also of an artistic impasse, Walton has from time 
to time offered a work that resembles a second, perhaps more accom- 
plished, fulfilment of an earlier task. Thus the overture Scapino (1941) is 
better constructed than Portsmouth Point yet does not fully match its 
youthful verve; the great beauty and skill of the Violin Concerto 
(1939) would be admired without reserve if this work (by other points 
than two quiet and lyrical outer movements enclosing an astringently 
witty one) did not invite comparison with the Viola Concerto, which 
is not only among the most attractive specimens of its genre but also 
Walton's most consummate achievement. A similar comment is invited 
by the Second Symphony (1960). 


EDMUND RUBBRA 


At almost every point of personal temperament and musical style 
Rubbra is Walton's antithesis. Despite his wide reading, historical know- 
ledge, and interest in the graphic arts, Rubbra cannot be described as 
urbane. His music almost excludes epigram, impressionistic digressions 
or sudden brilliance; most of its ideas have high lyrical potential and may 
be suffused with lyrical sentiment from the beginning (as at the opening 
of the Fourth Symphony) but they usually need ample time to become 
impassioned. The rhetoric moves steadily to its climaxes, where emo- 
tion glows to full incandescence as in filaments at some distance from 
their source of power. Rubbra's achievement has been to combine the 
classical principle of expanding variation with a continuously poly- 
phonic texture—polyphonic not just in the Wagnerian sense, but in a 
sense that has made people regard some of his movements as gigantic 
motets. Indeed, the claim to have linked renaissance with modern 
musical organisms, falsely made for composers whose texture was at 
least half impressionistic (Vaughan Williams's Mass, for instance, has 
more of Debussy than Byrd in it) can be sensibly made for Rubbra and 
only for him. Rubbra does not seek incidental colour for its own sake. 
Most of his ideas are song-like and grow by polyphonic texture wherein 
the long phrases reach inexorably forward into big paragraphs—not of 
balanced sequences and patterns of figuration, but of asymmetrical 
sentences. 


532 MUSIC IN BRITAIN 1916-1960 


His First Symphony (1938) is fashioned from the materials Rubbra 
has continued to use—the diatonic scales, with a predilection for the 
melodic minor and a sparing use of chromatics except the sixth and 
seventh degrees of the ascending minor scale. Contrary motion of the 
parts, including that between the two directions of this scale, gave a 
measure of discord unremarkable at the time, but the Symphony sounded 
hard-driven because the ideas were hard-driven. They are so in many of 
Rubbra's subsequent movements, across which may be written the 
technical description гета ricercata, but less and less forcing was needed 
as the ideas themselves became more fertile. 

Already in the First Symphony Rubbra had established several 
characteristic forms of tema ricercata, some of them by no means 
motet-like or fugue-like. Only the two most strongly contrasted can 
be described here. The movement corresponding with the minuet or 
scherzo of a classical symphony is called ‘Périgourdine’. A dance-like 
length of tune recurs without episode and with only such variation- 
development as Purcell uses to vary the key in a long ‘ground’. Passing 
harmonies, textures, and counter-themes seem to clutch at the swift 
gliding tune, increasing the effect of unhesitating motion. The second 
movement of Rubbra's Fifth Symphony (1948): 


Ex. 235 


Allegro 
Уп. + 8ve - — — 


EDMUND RUBBRA 533 


shows comparable treatment of a most attractive tune—apparently more 
naive than the *Périgourdine' yet in fact extremely subtle, and full of 
spores which are transplanted for accompanying material but not 
allowed to interfere with the main thread, in which the complete tune 
recurs through all twelve major keys. This pruning back, with the 
suppression of threatened climaxes in deference to the pattern of 
repetitions, produces an effect of intense energy. The corresponding 
first movements show a very different form springing from a song-like 
idea, which gathers counterpoint but does not germinate only by 
counterpoint and is not planned with countersubjects. In the initial 
stages of this type of organism the main stem of melody will normally 
make sense and suggest the play of smaller and larger climax even 
if shorn of accompanying material. Some of these movements with 
а strong ‘main stem’ have elicited the description ‘monistic’ from writers 
who are well aware that its propriety, like that of ‘monothematic’, has 
usually been questioned when applied to movements of symphonic 
dimensions. When Rubbra reached more genial expression than his 
First Symphony, he was not untrue to himself in allowing contrasts that 
deceptively suggest as much relaxation as there is in most classical sym- 
phonies. Thusthe staid melody which opens his Third Symphony (1939) 
is at first paced rather than countered, gathers tension by generating 
polyphony, reaches a distinct cadence (Ex. 236) and passes to what 
seems to be an entirely new, contrasting idea of plain chords which 
resembles the beginning of a classical ‘second group’; its punctuation 
by a four-note figure from the previous polyphony is no stronger a 
means of integration than the corresponding feature in the lyrical 
second subject of Beethoven's Fifth Symphony. Yet when this homo- 
phonic passage reaches its finish in the key from which it began, we 
recognize it as a long extension of the cadence. The polyphony con- 
tinues where it left off, gathering up the opening theme of the movement 
along with the punctuating figure of the episode as if to compensate for 
the relaxation. 


534 


MUSIC IN BRITAIN 1916-1960 


Ex. 236 


=== 


Trb. 


ү D 
5-1 


é 


Ch 
б 
e 
б. 
D М 


EDMUND RUBBRA 535 


Рита 
ени СҮН 2 


If works on a symphonic scale did not take great time and labour, 
we might wonder that Holst's disciple has written little vocal music. 
What he has written achieves distinction (notably in the choral Dark 
Night of the Soul, 1935 and The Morning Watch, 1941) and, when 
liturgical, reconciles scrupulous propriety with imaginative power. 
Moreover, it sometimes serves, like some of his few chamber works, to 
disprove the supposition that he is never inclined towards the im- 
pressionist's love of colourful or alluring effect. As a fine pianist, who as 
a boy loved the late romantics, Rubbra has a connoisseur's appraisal of 
harmony that makes an immediate impact, yet it cannot be denied that 
his best chamber music reflects his symphonic breadth. His Cello 
Sonata (1946) is among the noblest of its kind, his String Quartet, 
no. 2 (1952), within its proper limits, secures emotional intensity 
within imposing forms, and a piece for recorder and piano for Carl 
Dolmetsch, Meditazioni sopra Ceurs désolés, contains within a mere 
five pages an epitome of his art; yet his symphonies are his masterpieces. 

Rubbra's symphonies, of which there are so far eight, stand unique 
in British music. In them are combined inheritances that had been 
thought irreconcilable—polyphony similar in principle to that in which 


536 MUSIC IN BRITAIN 1916-1960 


the greatest British composers of the past excelled, the dimensions of 
romantic symphonies and their scale of climax and tension. 


LENNOX BERKELEY 

Although Berkeley (b. 1903) exemplifies the Parisian and Stravinskyan 
influences of the 1920s, he belongs to no group, French or English, and 
was little known until the 1920s were over. After a normal university 
career, not devoted to music, in 1925 Berkeley went to Paris and worked 
as a pupil of Nadia Boulanger in order to become effectively articulate. 
Many have testified to those parts of Nadia Boulanger's courses that 
are designed to broaden knowledge and taste; but in Berkeley (and in his 
contemporaries under her discipline—Copland and Harris for instance) 
1{ is possible to observe the effect of a complementary concentration 
upon certain composers from Monteverdi to Stravinsky and upon 
composition for prescribed groups and small groups of instruments. 
The first long work which Berkeley sent to England, the oratorio 
Jonah (1929), was scarcely coherent, somewhat congested and blood- 
less; despite inventive and sensitively scored passages it rarely inspired 
fine singing by soloists or chorus. Yet among the mature Berkeley's 
chief merits are his assured clarity of thought and texture and his 
outstandingly attractive vocal writing. Perhaps the most notable 
example of the latter is Lady Nelson's soliloquy at the end of Act II in 
Nelson (1953), but Berkeley's St. Teresa songs for contralto and string 
orchestra (1947) touch an extraordinarily high level of inspiration as well 
as craftsmanship. 

To recognize his cautious but assured growth into a distinct per- 
sonality one should compare his String Trio of 1944 with his Horn Trio 
of 1954, or his two Symphonies of 1940 and 1958. None of these four 
works is imposing or solemn; yet the advance in unpretentious musical 
architecture during the eighteen years which separates them is as clear as 
Mozart's more speedy advance from the serenade-like Salzburg sym- 
phonies and concertos to the Viennese works, with their motivic and 
contrapuntal integration. 

Unlike the preclassical masters, Berkeley was not forced to earn by 
purveying music of serenade gravity; but having abandoned the 
promptings that led to his oratorio, he published between 1935 and 
1950 a series of works which made people regard him as primarily “а 
divertimento composer'. He was content to keep his movements short 
and largely unrhetorical, to hold interest by buoyant rhythm, neat 
harmony, and attractive fragments of melody rather than by the 
musical equivalent of ‘great thoughts’. His Serenade for Strings (1939), 
Divertimento (1945), Concerto for Two Pianos (1943), Nocturne and 


ALAN RAWSTHORNE 537 


Sinfonietta, or his Piano Sonata (1941) and Six Preludes, suggested that 
Berkeley was capable of composition on a grander scale. Berkeley 
would still rather charm than bore, but with the charm there was now 
а concentration occasionally amounting to earnestness. An exception 
is Nelson (1953) which nowhere depends upon ‘good theatre’ to hide 
weak musical seams. In any other European country Nelson, as well as 
Ruth (1956) and the amusing chamber opera The Dinner Engagement 
(1954), would be valued. Admittedly Nelson is dramatically unequal and 
episodic; so also are most of the established favourites of the lyric stage. 


ALAN RAWSTHORNE 


Alan Rawsthorne (1905-71) courted neither public nor minority 
favour. He issued a score of admirable works, almost all instrumental 
and with simple musical titles—quartet, sonata, concerto etc. An ехсер- 
tion is The Creel (1940), a fascinating suite of short pieces for piano 
duet, each corresponding with one of Isaak Walton's ‘characters’ of 
freshwater fish (see Ex. 237(1) ). Rawsthorne's nationality is sometimes 
said to betray itself in turns of melody, especially in lyrical passages. 
Yet there are long stretches of this music, from the Symphonic Sketches 
to the splendid Violin Sonata (1959), which could deceive Swiss, Dutch, 
German, or Scandinavian listeners who did not know Rawsthorne (an 
important condition, for he has a personal style) into believing that they 
were hearing works by their compatriots. The First Violin Concerto, 
one of the works which occupied him during his war service in the army 
(though not played until 1948), was dedicated to Walton; yet here again 
the similarities to Walton are often less striking than affinities with such 
composers as Hindemith and Frank Martin. 

Rawsthorne's music sometimes resembles Hindemith's (in passing, 
not in broad structure) simply because he shares some of his aesthetic 
ideals. He offers athletic design and neat figuration; hence the frequent 
observation that his phrasing, patterning and decorative counterpoint 
recall the baroque masters. But it is the score-reading eye rather than 
the ear which links it with baroque technique. Undoubtedly his “соп- 
ception of variation is not the melodic one of Haydn and Mozart but 
rather that of Bach in the Goldberg Variations'.! Probably the most 
personal element in his idiom is his very fluid harmony; and if he had 
been required to theorize, he would probably have said with Hindemith 
that he avoided atonal techniques because he needed a hierarchy of 
tensions and relaxations among the degrees and intervals of the 
tempered chromatic scale. His use of a twelve-note series has no more 


1 Mellers, op. cit., p. 176. 


538 MUSIC IN BRITAIN 1916-1960 


significance than Walton's or Britten's; in the Quintet for wind instru- 
ments and piano (1962), it merely emphasizes the unchanged elements 
in his style. 

Though Rawsthorne shapes complete movements as assuredly and 
precisely as he does paragraphs and sentences, at least some of his 
persistent characteristics can be observed in short excerpts which, taken 
from very diverse sources, also show how intelligently self-imposed 
limits have not restricted his range of imagination and expression as 
they might have done. Scanning the following excerpts, we see the 
variety of patterning and decorating but do not hear it as neo-baroque 
pastiche: 

“Тһе Sprat', No. 2 of Тйе Cree 
Ех. 237 ) 
O vivace 


кше 
Phe. I 


Pfte. II 


GERALD FINZI 539 


E Symphonic Studies 
Gib Allegro piacevole m Put 0 


Baroque patterning is a superstructure over a thoroughbass; in Raws- 
thorne's texture functions are reversed, for it is the figuration or the 
contrapuntal superstructure that holds together the mercurial har- 
monies—the slippings from augmented triads, the constant major-minor 
alternations, and the elusive tonality. 

If Berkeley had not composed religious songs and a romantic opera, 
his temperamental as well as his technical affinity with Rawsthorne 
might have been more often noticed. The French atmosphere of his 
work should not blind us to several parallels with Rawsthorne's 
methods; and Rawsthorne's music is far from being a bloodless 
exhibition of design and form. There has been no need for Rawsthorne 
to compose a Nobilissima Visione, or even his admirable Street Corner 
Overture (1944), merely to prove the redness of his blood. 


GERALD FINZI 


The songs of Gerald Finzi (1901—56) treat English verse as sensitively 
as Wolf’s did German. Three cycles of songs from Thomas Hardy and 
one from Shakespeare, issued between 1933 and 1949 (especially the 
Hardy cycle of 1936 entitled Earth and Air and Rain) earn Finzi a 
greater distinction than he has secured by music more frequently heard 
in public, such as his cantata Dies Natalis (1939), a setting of five 
Traherne poems for voices and strings, or his Concerto for Clarinet and 
Strings (1949). A light polyphony resembling some of Rubbra's less 
intense orchestral textures gives a quasi-symphonic effect to Finzi's 
accompaniments, | 


540 MUSIC IN BRITAIN.1916-1960 


MICHAEL TIPPETT 

After 1945 only extreme pessimists could fear a dearth of new and 
original native music, for the 1940s confirmed. the impression that 
Michael Tippett (b. 1905) would stimulate controversy with every new 
work; and Benjamin Britten (b. 1913), some of whose instrumental 
works of the late 1930s (especially the concertos) had produced un- 
certainty whether the spate of his ideas would run deep or shallow, now 
produced three fine operas in three successive years—Peter Grimes in 
1945, The Rape of Lucretia in 1946, Albert Herring in 1947. 

Despite the disparity of their ages, Tippett and Britten have tempera- 
mental affinities. They read more widely than most musicians and are 
interested in other arts than music. They differ greatly as composers yet 
are both attracted by subjects (in songs, operas, or cantatas) dominated 
less by direct expression of the primitive passions than by the pity, fear, 
disgust, or amusement with which we contemplate their survival or 
perversion among supposedly civilized men. They would be assured of 
lasting distinction if they had been no more than the first modern 
English composers whose operas aroused international interest, or the 
first whose declamation has challenged the artistry of singers and 
invited a new examination of dramatic word-setting from Purcell to 
Mussorgsky. 

Tippett's deep interest in the polyphonists and in Purcell was not 
merely instilled but grew from inner compulsion. Tippett once confessed 
himself unable to select the greatest composers of this century, but he 
was willing to name the most important ones for himself and other 
British musicians: Stravinsky, Hindemith, Bartók, and Berg. He ad- 
mired Purcell’s emulation of the Italians and esteemed English composers 
of the twentieth century for what they did not share. Bartók, he declared, 
was well served by the asymmetry and fertilizing ‘barbarity’ of the folk- 
music he collected, whereas English composers had been ensnared by 
their lyrical folk-music. 

Tippett's works are few in number and make no bid for the modicum 
of popularity elicited by work that can be allotted to stylistic niches. 
There are few points of style common to Tippett's quartets, cantatas, 
and operas; few indeed can be traced from one quartet to another. Some 
extracts from his Second Quartet may serve to show the originality of his 
instrumental polyphony, indebted to English consort music but 
not finding its integration, as that music did, in the kind of rhythmic 
and melodic symmetry associated with the high baroque and classical 
styles: 


541 


MICHAEL TIPPETT 


Ex. 238 


(i) 2nd. movt. Andante 


(ii) 3rd movt. Presto 


стезс. 


56 


542 MUSIC IN BRITAIN 1916-1960 


The claim that bar lines are purely for the reader's convenience can be 
justly made for a good deal of Tippett's musical thought; but he did not 
reach a consistent and integrated instrumental style. His music can be 
irritatingly insubstantial even when it seems polyphonic and, to the eye, 
suggests rhythmic subtlety. The figuration of his accompaniments (to 
vocal or instrumental lines) seems often to intrude unconvincingly with 
ornamental figuration, not to support or challenge; and even his popular 
Concerto for Double String Orchestra (1939) owes much to an illusion 
of solidity produced by its medium. 

His important vocal works are few and all are set to carefully 
chosen words at the bidding of complex ideas, the symbols without 
which he seems unable to express simple enough truths. The complexity 
of his symbolism, and the fact that it begins before the further symbolism 
of music has been imposed, would have distinguished Tippett from 
Britten even if their music itself had not been so different—Tippett's 
involved, elusive, sometimes inchoate, and Britten's disarmingly direct 
and paradoxically both sophisticated and naive. To be fully articulate 
Tippett needs words and the human voice; and even when he strains the 
voice and sets syllables to melisma without obvious reason, he com- 
municates the anxiety of a high intelligence. Whether spare or elaborate, 
his music is produced by intense mental labour; so is Britten's, but 
Britten rarely betrays labour. 

Two exceptional and daring essays in seemingly incongruous juxta- 
position have been made by Tippett and Britten, the first in 4 Child of 
our Time (1941) by employing Negro spirituals as Bach did chorales, the 
second in his War Requiem (1962) by interposing the operatic declama- 
tion of Wilfred Owen's poems within liturgical treatment of the Latin 
choruses from the Missa pro defunctis. Each caused a sensation at first, 
yet Tippett's cantata is as innocent of vulgarity as Britten's masterpiece. 
In the past, opera has been more effective when we have been caught up 
in the feelings of protagonists than when we have been presented with 
psychological symbolism and moral reflection, but Tippett's cantata 
should be judged as a work sui generis. The difficulty of doing this is 
increased by Tippett's tripartite plan. Each section begins with a chorus 
and ends with a spiritual, suggesting parallels with Bach's Passions. As 
in Bach, both choral and solo sections are used for narration as well as 
for reflection. The intermingling of story and commentary, the quantity 
and brevity of ‘numbers’, and Tippett’s elusively complex rhythms 
enhance the effect of the spirituals which are presented without transla- 
tion to the texture and harmony of other items. 

The Midsummer Marriage (1952) brought into the theatre much of the 
Tippett of A Child of our Time. He wrote his own libretto, giving himself 


BENJAMIN BRITTEN 543 


little scope for ‘characterization’ and much for symbolic representation; 
his music is allied with moral problems which are beyond the bounds 
of musical expression. Yet for its understanding the listener-spectator 
needs no more than the barest outline of the composer's explanations. 
Tippett observes that the classic comic plot presented hindrances to a 
marriage, and that they were largely social; the chief modern hin- 
drances to a good marriage or to any successful relationship are 
ignorance and illusions, especially about ourselves. Therefore in the 
course of his plot the characters discover their true selves, and he 
himself has declared that “һе moral of The Midsummer Marriage is 
enlightenment’ when in a final tableau we see the young lovers trans- 
figured in light, like Hindu deities. A sensitive young man and a hard 
young woman (‘whose illusions are spiritual), a tycoon and his vulgar 
innamorata (‘whose illusions are social’), a mechanic, a clairvoyante, 
ancient Greeks, priests and priestesses, ritual dancers, splendidly and 
diversely caparisoned choruses, a spiral staircase that stops in mid-air 
like something out of Cocteau, a cavern with gates, the play of light 
from dawn to nightfall, the different seasons of nature—here is ample 
mechanism for opera and for a phantasmagoria of symbolism which 
seems to be the necessary catalyst for a Tippett opera; for, as 4 Child 
of our Time shows, Tippett's music does not necessarily become second- 
rate when his text becomes philosophic. 

Music is powerful not only because it intensifies what words and 
spoken drama can convey, but also because it expresses reactions of our 
subconscious and shades of emotion which words can neither describe 
nor even name. Tippett and Britten are modern composers because 
they have sped music on this process of discovery; that their success 
varies, or that their styles may be considered conservative, matters little 
beside the fact that their sensibilities have kept pace with the ideas of 
twentieth-century thinkers, poets and dramatists. Their libretti may be 
symbolical or historical, but produce modern, even ‘problem’ operas 
which disturb comfortably settled minds. 


BENJAMIN BRITTEN 


If Britten had died before setting libretti which tax both mind and 
musical sensibility, he would still be acclaimed as one of the most 
imaginative of composers, distinguished by the degree to which his 
imagination is aural. Few composers so marvellously command the 
acoustic and instrumental materials of music that they always secure, 
by notation and a few spoken directions at rehearsal, the effects they 
imagined before a note was written. This faculty in Britten almost 
disarms criticism. Faced with a level of professional competence reached 


544 MUSIC IN BRITAIN 1916-1960 


by no other British composer since the seventeenth century, one may 
dislike this or that effect and try to defend an objection of mere taste; 
but it is normally dangerous to suggest improvements in the means 
by which the effect is secured. The point may be illustrated from the 
Sunday morning scene in Peter Grimes, during which Ellen's pleadings 
and Grimes's expostulations are heard against evocations of the sea 
and of the service proceeding in the nearby church. An initial suspension 
of disbelief which accepts scenery and music that is only religiously 
picturesque seems less vulgar than a series of adjustments to convention 
at each change from the liturgical to the operatic. Peter Grimes wonder- 
fully captures the style of the English village service from our grand- 
fathers' time to ours; but the parson's intonings, the congregational 
responsory and the incurably intrusive electronic organ reached those 
outside church in spasms. Britten could hardly have flecked his score 
with stray sounds of worship as Debussy did Fétes with distant sounds 
of the fairground; yet had he not overstrained his mind's ear by trying 
to assimilate sections of church-parody into his tense dramatic dialogue? 
No. There is no incongruity between the stylized and the onomatopoeic 
—between Britten and Victorian church chant. 

Almost as disarming is Britten’s professional seriousness, for it has 
often contradicted first impressions. The most conscientious artist may 
occasionally pass what seem the idle thoughts of a busy fellow, and 
when he is extraordinarily gifted as well as conscientious, the listener 
may easily believe that he has been content with ‘facile brilliance’ or 
‘shallow cleverness’. In addition to his share of salutary adverse criti- 
cism Britten has also received adverse criticism that was contradicted 
after subsequent closer acquaintance with the music. The instrumental 
Sinfonia da requiem of 1940 (Ex. 239(1) and (ii) ) illustrates the kind of 
expression to which few people had become accustomed in 1940, when 
Britten was twenty-seven: 


Ex. 239 


Andante molto tranquillo === =ч НЩ 


(i) 


545 


BENJAMIN BRITTEN 


Sul D 


| 
| 


ЗЕ: 


рита 


ED ЕН == а e ame eode Mts г ИР ЕСЕ mt] 


zk Sarre Lae See Gee Е Ц 
i. eee ey Se 4 7 | ] 


546 


Lael 


MUSIC IN BRITAIN 1916-1960 


Andante 


Gii) 


BENJAMIN BRITTEN 547 


Britten perplexed his early critics because he could be related neither 
to a recognized contemporary school nor to a tradition. None of 
Britten's English predecessors, whether nationalists or disciples of con- 
tinental leaders, entirely abandoned criteria of harmony and texture 
formed by ‘the classics’. His major works seemed insubstantial in 
the same sense as Turner's later pictures—all air and light. When 
he decided, as in The Young Person's Guide to the Orchestra (1946), to 
make his peroration with a fugue he sloughed parts or ran them into 
unison to avoid opacity. Similarly chaconnes, canons, and pedals are 
allowed no more complexity than is required for a desired total effect. 

Britten's fugues and his Second String Quartet (1945)—the move- 
ments of which are of more traditional design that most of Britten's— 
suggest that certain designs discover his limitations. He does not 
resemble the English teachers for whom he has expressed admiration 
and affection—Frank Bridge, John Ireland, and Arthur Benjamin 
(1893-1960). It has already been pointed out that these were among 
the least nationalist of their generation and the least attached to any 
school. Unlike Hindemith and Nadia Boulanger, they do not seem to 
have inclined towards the manner of any modern composer or school. 
If we did not know their names not even the title of the most virtuosic 
of Britten's early instrumental works, the string Variations on a Theme 
of Frank Bridge (1937), would lead one to mention their influence, 
although Arthur Benjamin's 77e Devil Take Her (1932), a sparkling and 
splendidly composed one-act opera, is one of the few pre-Britten English 
works that might pass for Britten's, if we could postulate for him an 
operatic ‘first period’. Benjamin wrote five works for the stage: unfortu- 
nately the best music is found in the incomplete Tartuffe, which contains 
spoken parts and is an English adaptation of Moliére's Le Misanthrope. 

Britten is at his best when a plot, text, or vocal line provides the 
thread that helps to make explicit the purport of his texture as a whole 
and also bears the chief concentration of integrating technique. It is 
possible that the revaluation of Sinfonia da requiem has been largely 
the consequence of familiarity with his treatment of the voice and his 
response to dramatic and pictorial situations; the thematic material is 
subtly linked from movement to movement and there are also derivations 
from harmonic formulae; moreover the three movements bear 
titles and could have occurred in an opera or cantata. The examples from 
the third and final movement, “Кедшет aeternam', provide a slight 
anchorage for the listener. The instrumentation of Ex. 239(1) is possibly 
from Stravinsky and Ex. 239(1) is Mahlerian; in both excerpts the 
rhythms are conservative and the harmonies basically so. The flutes clash, 
and several bass notes are discords, yet the sound is gently luminous. 


548 MUSIC IN BRITAIN 1916-1960 


Britten's harmony is as personal as his structure, and so linked with 
structure that we must qualify the label ‘impressionist’. Britten is an 
impressionist in so far as his chords are not conceived within a frame- 
work of symmetrical harmonic rhythm. Singly, in pairs, or in short 
groups they may be sustained, repeated, or varied as long as he needs 
their initial effect. He rarely uses the lush chords of post-romantic 
impressi nists but is fond of triads with added notes, augmented triads, 
and indeed just plain triads divorced from traditional contexts. His 
juxtapositions look simple but are not prefabricated ‘progressions’. The 
lullaby from The Rape of Lucretia quoted in Ex. 239(iii) seems ап excep- 
tion in Britten of four-part harmony with a regular tread, yet it is both 
impressionistic and original in a way which only the dramatic circum- 
stances and the proper instruments fully reveal. Britten’s chords, unlike 
Bartok’s or Stravinsky’s, are rarely built up from fourths; he favours the 
piling of thirds—triad, seventh, and ninth. 

We cannot class Britten with impressionists who continued the refine- 
ment of erotic romanticism, nor with those whose evanescent harmonies 
and colours are mixed with mere wisps and scrappets of melody, for 
the strongest of all his materials of expression is his vocal line. It has 
been imitated but remains unique. Britten’s ‘line’ (vocal or not) gives 
the complete texture a tighter integration than was necessary for former 
impressionists. 

In Peter Grimes several passages of arioso seem to flaunt the ideal of 
‘heightened speech’. Lawyer Swallow’s declamations, of which Ex. 240 
is a specimen, were quoted as examples of perversity: 


As-sign your pret - ti- ness to йе, ІШІ сай ETE 


real рго-рег - ty. 


In his twenties Britten was manifestly attracted by Purcell's inspired 
flamboyance, yet he never emulated Purcell to the extent of sounding 
neo-baroque. Undoubtedly, however, seventeenth-century words in Тйе 
Holy Sonnets of John Donne (1946) have prompted his most Purcellian 


BENJAMIN BRITTEN 549 


settings. The point could be best shown from No. 7, ‘At the round 
earth's imagined corners', but one forbears to excise a quotation from a 
song that 15 unified by a ground-bass technique, and therefore Ex. 241 
is taken from No. 9, ‘Death, be not proud’. The extract is the trans- 
formation of a quiet theme introduced at the words ‘From sleepe and 
rest, which but thy pictures bee’: 


Jf marcato ATN 


And dost with 


а = 


poppie or charms сап make 


550 MUSIC IN BRITAIN 1916-1960 


а 
[7 
[AN D meme ые ПИРДИН ЧЫНЕ ш 


Апа bet-ter than thy  stroake; 


Such direct and fiery Purcellisms now represent a comparatively early 
stage in a composer who could hardly have reached the subtle integra- 
tion of his later settings and yet retained vivid illustrative power if his 
flamboyance had not been strong enough for pruning. He is now rarely 
opulent and never ostentatious, but his refinement is not emaciated, 
timid or affected; it comes from an invention so fertile that it might 
have lost much of its power if it had run riot. 

The stages by which Britten came to mastery in dealing with English 
words invite psychological as well as musical inquiry. According to 
Peter Pears! his thorough study of Purcell was not undertaken until the 
general directions of his style were fixed and his originality recognized. 
The 1936 settings of satirically sententious and sometimes obscure 
verses by Auden in Our Hunting Fathers were by no means a waste of his 
abilities; still less questionable was his determination not to set familiar 
English verses from Elizabethan and Jacobean lyrics, or from later 
sources, until he had met the challenge of French in Les Illuminations 
(Rimbaud) and Italian in Seven Sonnets of Michelangelo (both 1940). 


1 Benjamin Britten: a Commentary on his works, ed. Donald Mitchell and Hans Keller 
(London, 1952), p. 64. 


BENJAMIN BRITTEN 551 


Pears’s comments on the factors which ordained Britten’s course 
need not be echoed here, but we should draw one clear conclusion— 
that, as Byrd declared in a misquoted and misunderstood passage, the 
‘hidden power’ which elicits music at the reading of the right words 
comes not from the words themselves but (Byrd’s phrase) from the 
‘thoughts underlying the words’, and therefore Britten’s self-imposed 
course was neither perverse nor affected. Before returning to his native 
language he wished to kill any temptation to accept without question a 
reciter’s values. 

Britten’s return to English poets was signalized in 1943 by the 
publication of one of his most attractive works, the Serenade for tenor 
solo, horn obbligato, and string orchestra. The subject of evening and 
nightfall links its six texts, chosen from different periods so that an 
anonymous fifteenth-century dirge is flanked by familiar verses from 
Blake and Tennyson, Jonson and Keats; but Britten’s wish to unify the 
cycle precludes any attempt to relate musical styles to literary periods. 
Particularly fine, both in itself and in its effect as a ‘frame’ is the para- 
graph for solo horn that forms the prologue and is repeated distantly 
(‘off stage’) for epilogue. It is to be played on the natural harmonics: 
the feeling for the instrument, the punctuation of the paragraph and 
the extraordinary power of its phrases to suggest both the main theme 
of the cycle and a concentration of associations and memories testify to а 
rare quality of musicianship. 

The Serenade is among the most deservedly popular of Britten's non- 
operatic works, but the key documents to his mature, unostentatious 
integration in arioso are the three Canticles. For subtlety of organiza- 
tion Britten has never surpassed Canticle 3 (1955) a setting of Edith 
Sitwell’s ‘Still falls the rain’. The verses are interspersed with variations 
for horn and piano, and only a long quotation could show the variation 
technique as well as the relation between vocal and instrumental move- 
ments. 

There is hardly a better manual of Britten's later vocal style than a 
set of songs which, unlike the Canticles or Nocturne, are almost domestic 
music, though singer and pianist must be worthy of a modern 
Schubertiad. Winter Words (1954) is an album of eight lyrics from 
Thomas Hardy. The first, ‘At close of day’, accommodates a wealth of 
illustration in variations of an initial idea, yet at the finish leaves us 
aware that the dips of the line throughout depict the background of the 
poem, the swaying of ‘tall trees’. The motor pattern of the second song, 
about a poor lad’s railway journey from home on a bleak night, is a 
traditional integrator; but the Purcellian melisma in the voice part 
produces a total effect by no means traditional. The deliberately 


552 MUSIC IN BRITAIN 1916-1960 


РигсеШап arioso іп No. 5, “Тһе Choirmaster’s Burial’, is self- 
explanatory. The last song, ‘Before Life and After’, is useful for short 
quotation. Its trudging triads offer in themselves little scope for varied 
expression but make a superb background for highly expressive vocal 
inflexion. In Ex. 242 we note how they are halted, Schubert-wise, at key 
points in the poem, the third time with an unemphatic false relation 
which gives “по sense was stung' a vividness which eludes the words 
alone: 


as опе may guess Апа 


toad very smooth 
DIE 


BENJAMIN BRITTEN 


(ii) 


or cross 


er crash 


> 
© 
8 
Е 


None cared 


Brought wrack 


(iii) 


м4 
E: 
це] 
З 
я 
d 
ў 
о 
B 
B 
З 
721 
e 
o 
я 
1 
> 
аж 
50 
ze] 
© 


If 


554 MUSIC IN BRITAIN 1916-1960 


No sense was  stung. 


Britten withdrew his first opera, Paul Bunyan, and no part of it has 
been heard in public. Peter Grimes (1945), comparable in resources and 
musical ingredients with Verdi's Otello rather than any German music- 
drama, remains his most popular work, having been translated into 
many languages and acclaimed in many countries. Тле Rape of Lucretia 
(1946) requires a string and a wind quintet with two percussion players, 
and a cast of three men and three women to whom are added one man 
and one woman who stand on either side of the proscenium as com- 
menting chorus. It is customary to call this work and the ensuing 
Albert Herring (1947) chamber operas, though Lucretia touches levels 
of pathos reached in Britten's full-scale operas and Herring is one of 
the better comic operas of our century. It is possible to deplore the new 
forms of puritanism, manifest in gratuitous moralizing and in an em- 
phasis upon squalid verismo for The Beggar's Opera and in the implica- 
tion that we are wickedly heartless to acquiesce in Gay's idea of “А 
Newgate Pastoral’; yet one may still marvel at Britten’s ability to meet 
Tyrone Guthrie's dialogue and to extend or vary the original music 
without once suggesting academic exercise. The Beggar's Opera (1949) is 
rightly described as ‘opera in three acts’ and wrongly as “а new realiza- 
tion of John Gay's Ballad Opera 1728'. 

The Turn of the Screw (1954) also engages ‘single’ instruments. This 
work, to which ‘chamber’ should never be applied as a diminutive, is 
the finest as well as the most ingenious of Britten's operas. The less 
complex and very directly appealing Billy Budd (1952)! which, marking 
Britten's return to Verdian broadness, full cast, and orchestra, may be 
considered the successor to Grimes. Its naval setting and consequent 
all-male cast seems not to have affected its success. For the coronation 
festa teatrale at Covent Garden in 1953 Britten provided Gloriana, an 
episodic piece upon the relations between Elizabeth I and her favourites; 
the two ballet sections and the incongruous spoken scene at the end are 

1See pl. VIII. 


пәрлегу JUSAOD ‘NOH 2.99 Loy ou) іе иоцопрола [eursrio әцу шол) 20905 


(rec “4 295) 661 д9ПЯ АТИЯ $ МАЛАЯ 


= 
ul 
= 
< 
a 
(е6 


BENJAMIN BRITTEN 559 


по the only features of this work which prove it to be no exception to 
the rule that first-class operas cannot be secured from the most giftep 
composer if he is restricted in choice of subject or given a time-limit. 

Another work in a class by itself is А Midsummer Night’s Dream 
(1960). Abbreviating Shakespeare so that cutting or botching is no- 
where evident, Britten and Pears prepared a libretto which offers even 
more scope than the tragic operas for Britten's characteristic techniques. 
More than a dozen personages are thematically distinguished, their 
musical characterization high-lighted by strong contrasts—Titania and 
the transformed Bottom, the drugged and therefore fawning or scorning 
lovers—and the already wealthy motivic apparatus is further enriched 
by the palpable characterization of such concepts as The Wood, or 
Sleep. Oberon's is a countertenor part and Puck's a speaking part for a 
boy-acrobat; the dances and the onomatopoeic and scenic effects are 
not the only wonderful features of a score without modern rival for the 
variety of suggestive magic that is drawn from the quieter registers of a 
large orchestra. Particularly original is the use of the lighter percussion 
instruments. 

In every one of Britten's operas the music carries the drama, is itself 
the drama. The Rape of Lucretia, based on a play adversely criticized for 
its preciosity, is furnished with an English text in the pseudo-poetry 
practised by teachers and students of prosody. Yet objection is 
irrelevant. If such a sentence as ‘Home is what man leaves to seek’ were 
merely “еі” to the kind of music designed to let words speak for them- 
selves, we should suppose it to be a literal and obscure translation. As it 
comes through an exquisite instrumental texture from the female chorus 
(during the folding of linen) nobody strains to catch its meaning: it is 
simply part of a musical picture of domestic womanhood. The text of 
Lucretia is no more a musical handicap than that of Grimes which, 
based on George Crabbe, contains phrases not easily ‘caught’ as so many 
words, e.g. ‘But when the crowner sits upon it, who can dare to fix the 
guilt?’ It is as a musician that Britten is a consummate dramatist, and 
he would have been so even if he had not composed his operas. 

From the unpretentious, divertimento-like Cello Sonata to 4 Mid- 
summer Night’s Dream Britten’s sheer composition has no parallel in any 
British music by younger composers. And, with all its variety, it is 
remarkably—and unconsciously—consistent. Britten expressed amaze- 
ment when shown the similarity between the chords used in Ех 2431), 
for the setting of Keats’s ‘Sonnet to Sleep’ іп the Serenade, and those 
employed nearly twenty years later to evoke a similar atmosphere in 
A Midsummer Night's Dream (Ex. 243(ii) ). In neither case did the fact 
that four chords utilized all twelve semitones signify any indebtedness 


556 MUSIC IN BRITAIN 1916-1960 
to Schoenberg or his technical successors. Britten's conscious aim was 
to produce equal unexpectedness between the four chords and provide 
a large enough basis for variations. 
Ex. 243 в 
(0) 4 
Adagio 


Tenor НИНА А 
Voice ГЕ 


тшш = 
Violas E o o a uae Hr К зс [sje a д та ЕЕ, 
ШАЛУ, LE 07 5 а j ;: [Ed е 


Brass, muted 


Str. muted 


ELISABETH LUTYENS 


Despite the structural use of twelve-note rows in The Turn of the Screw 
(1954) and the Cantata Academica (1960), celebrating the 500th anni- 
versary of Basle University, these are not twelve-note compositions; the 
music is, as always with Britten, essentially diatonic. 

The first 'twelve-note'" British musician was Britten's senior. То 
Elisabeth Lutyens (b. 1906) must go the credit of first employing in 
England the full Schoenbergian ‘method of composing with twelve 
notes. As a viola player Lutyens had been attracted by Purcell's 
fantasias, and it was her attempt to ‘organize “аіопа!” conceptions in 
accordance with an indigenous tradition’ in a five-part string fantasia 
which 'gave birth unconsciously to a twelve-note theme'.! She turned 
permanently to serial technique in the Concerto for Nine Instruments of 
1940, but she had shown full understanding of Webern in the String 
Trio of the previous year. Until her Sixth String Quartet of 1952, how- 
ever, she seemed curiously limited in rhythmic invention, although she 
was said to be an admirer of Bartók. Her music is most impressive when 
it includes voices, e.g. her setting of Rimbaud's O saisons, O cháteaux 

1 Groves Dictionary (fifth edition) (London, 1954), v, p. 448. 


IMMIGRANT COMPOSERS 557 


for soprano, mandoline, harp, guitar, and strings (1946), The Pit (1947) 
for tenor and bass soloists, with women's chorus and orchestra, her 
Op. 27 Motet (1954) to words selected from Wittgenstein, and her 
chamber opera Infidelio of the same year. Her contemporary, Benjamin 
Frankel (1906—73), and two rather younger women-composers of her 
generation, Elizabeth Maconchy (b. 1907), and Phyllis Tate (b. 1911), 
are more conservative. 


IMMIGRANT COMPOSERS 


Three continental musicians of wide experience and forward-looking 
temperament settled in Britain during the 1930s. Egon Wellesz (b. 1885), 
though esteemed chiefly for his studies of Byzantine chant and of early 
baroque music, has composed eight symphonies and nine string quartets, 
much better known in Central Europe than in the land of his adoption, 
as well as operas and liturgical works. A close friend of Bruno Walter 
and Schoenberg he did not follow Schoenberg’s path but was constantly 
influenced by his historical studies—eastern, medieval, baroque, roman- 
tic, and modern. His music is recognizably Viennese and includes many 
points of experimental interest, a fact which explains his wide influence 
on adventurous young musicians, exercised not by expecting them to 
imitate him but by shrewd understanding of European and English 
traditions and by unfailing optimism and sympathy. 

Ex. 244 is from his Eighth String Quartet (1963). The extended 
Mahlerian character of this passage is clear in the wide intervals of 
the melodic line in the first violin part and the reining-in before the 
irregular metre of the climax. 


Ex. 244 


97 


IMMIGRANT COMPOSERS 559 


Roberto Gerhard (1896-1970) must have been the least docile and the 
least Teutonic among Schoenberg’s pupils. Although his father was 
Swiss and his mother French, he was brought up as a Catalan near 
Barcelona. He came to England at the end of the Spanish Civil War and 
lived in Cambridge. 

In 1945 Gerhard adapted Sheridan's 7he Duenna for an opera in 
which the juxtaposition of techniques is no more incongruous than 
Sheridan's mixture of verses and conversational prose, or his relaxations 
and accelerations of dramatic action. Gathering tensions among the 
characters sometimes produce a Bergian polyphony, and the transitions 
between arias and dialogue-music often engage serial technique. On the 
other hand the charming verses are set to attractive rhythms in a style 
that is often less complex than that of the strophic items in Stravinsky's 
The Rake's Progress. Much in the declamatory and conversational vocal 
line shows both subtlety of characterization and an awareness of 
English speech inflexions that at times actually recalls Britten's. 

An even more powerfully individual artist emerged in the instrumental 
works which Gerhard composed after 1950, when he became an estab- 
lished twelve-note composer, though without losing the love of sensuous 
harmony with which his earliest work showed him to be abnormally 
endowed. Gerhard has provided a clue to the harmonic and orchestral 
allure which he achieves in works like the serial Violin Concerto of 1945, 
Piano Concerto (1951), Symphony (1955), Harpsichord Concerto and 
String Quartet (both 1956). He arranges and permutes divisions of the 
twelve notes of a series (e.g. tetrachords or hexachords) as one does 
when spacing or inverting chords of three, four, five, or six notes in tonal 
harmony. In some ways he was more like Schoenberg than Webern was, 
and could claim just as validly to have developed from a point at which 
Schoenberg left off. 

The third notable immigrant, Mátyás Seiber (1905—60), was a less 
original and audacious composer than Gerhard, but more representative 
of central European advances in composition as observed by a superb 
teacher who assiduously practised composition in all styles. 

Seiber, who settled in England in 1935, was among Kodály's pupils 
and was much influenced by Bartók both before and after his appoint- 
ment at Frankfurt-am-Main in 1928 as director of the newly opened jazz 
department of the Hoch Conservatoire. Such essays in orthodox 
Schoenbergian dodecaphony as the Second String Quartet (1935) and 
the chamber cantata Three Fragments (1957) (the texts taken from 
Joyce’s A Portrait of the Artist as a Young Man) attracted less attention 
than the Third String Quartet (Quartetto lirico, 1951) and the cantata 

1 On Gerhard, see also р. 426. 


560 MUSIC IN BRITAIN 1916-1960 


Ulysses (1949) in which Seiber set passages from Joyce's book for tenor, 
chorus, and orchestra. Here we find, without obvious debt to Berg or 
anyone else, Seiber's easy absorption of dodecaphonic methods along 
with others. Ulysses disproved the charge that serial music is impractical 
or ineffective for choral groups, and Seiber's work in general answered 
critics who supposed that dodecaphonic serialism cannot sustain 
lyrical rapture or that it cannot be integrated with techniques avoided 
by its Viennese pioneers. 

Ex. 245 is from Ulysses. Characteristic of Seiber are both his consi- 
deration for the singers, whose individual lines are easy to pitch, and his 
combination of canonic writing with bouche fermée to create an atmo- 
spheric effect which is nevertheless truly, even traditionally musical in 
the strictest sense. 


Ex. 245 
ri 
а= c Hui jT e 
Шал was. а. М = to. = pisi. — —9" NE ENSE: 
| = Unis. резрг. 
г. Eig а р, pafl 
Ере 
m и was а U - ры -әрі-а me dep (COS ушт 


77 yA 


= [EEE die EEE 


Thatit was a U - to - pi-a E 


Bass 


poco a poco string. 


(4-63) 
espr. cresc 
жы NUIT NE t 
я 5 s В H 21 a 

(бу ee, Не , E НЫ 

т р | Ва в т 7 

(closed) m 

—————— 

/ È » |lL— ПР Я 

ЕЕ" zz 


(closed) m 
Ц р К йй | j ; `~ 


(closed) m 


poco a poco ues M 
e 


2a: RS C M pm 
Ира не S me gru mL 
TEE TE 
Gb EA ЕЕ 


(up to 76—80) Allarg. | 


ан ты 
querat reg rire | 


A tempo (4 = | (d= circa 76) 
Picc. eo dad NM. ---- | 
44121471144 4411 
REI 


da Sm mcd „ер жЕ Де ашый 


HUMPHREY SEARLE 563 


HUMPHREY SEARLE 


The writings of Humphrey Searle (b. 1915), who studied with Webern, 
include Twentieth Century Harmony and translations of Schoenberg's 
Structural Functions of Harmony and Rufer's Composition with Twelve 
Notes. As a composer Searle glanced towards Stravinsky in Gold Coast 
Customs for speakers, male chorus and orchestra (1949), but achieved 
more distinction by his interest in Liszt. Searle hardly did more than 
defy fashion with his first Lisztian piano concerto in 1944 but he proved 
in the next few years that a Webern pupil need not abjure the reflective 
beauty of his native song. His Intermezzo for Eleven Instruments, his 
setting for soprano, flute, oboe, and string quartet of W. R. Rodgers’s 
Put away the flutes, and his best work, the somewhat Bergian Poem for 
Twenty-two Strings (1950) as well as the Variations and Finale for Ten 
Instruments (1958) are stronger works than the more grandiose Piano 
Sonata (1951) with romantic bravura and a Lisztian thematic technique, 
and five symphonies (1953, 1958, 1960, 1962, and 1964). The River-Run 
of 1951 affords more pleasure from the orchestral texture than from the 
speaker of James Joyce’s words with Dublin intonations. 


THE YOUNGER GENERATION 


The new composers who emerged after the Second World War include 
Peter Racine Fricker (b. 1920), Malcolm Arnold (b. 1921), Robert 
Simpson (b. 1921), Iain Hamilton (b. 1922), Anthony Milner (b. 1925), 
and Thea Musgrave (b. 1928). Fricker has given time and labour to a 
relatively small output of high quality. Conscious of developments 
abroad he owes allegiance to no single school or method, and even his 
admitted indebtedness to Bartók is far from obvious. He took lessons 
with Mátyás Seiber, and his 1949 Quartet is clearly influenced by 
Bartók's Third. Before Fricker could have been at all considerably 
influenced by Bartók he had produced harsh but emotional effects by 
his textures; and if he did not command Bartók's rhythmic fertility, 
Fricker has proved himself to be a master of rhythm in the broader sense 
in movement after movement although (or maybe because) his rhythmic 
articula are built into contrapuntal designs derived from those used in 
classical teaching. What gives such distinction to the result is the 
frequent composition of an athletic texture in which vigour, though 
generated by ubiquitous counterpoint, appears to be concentrated in 
an expansive, finely phrased, but urgent melody as in this excerpt from 
the first movement of his Second Symphony: 


564 
Ex. 246 


MUSIC IN BRITAIN 1916-1960 


Allegro 


THE YOUNGER GENERATION 565 


= HU UCET 


Admirable thematic organization makes Fricker's Second Symphony 
(1951) as a whole more rewarding than the First (1949), despite the First 
Symphony's finale where inspired integration transcends its complex 
detail. Perhaps the most assuring aspect of Fricker's composition is its 
consistency and reliability. The symphonies and the Viola Concerto of 
1952 happen to be impressive because they use the full orchestra, but 
the same standard of composition is found in the String Quartet of 1949, 
the Violin Sonata of the following year, and some works for small 
orchestra composed at about the same time—a Violin Concerto and the 
particularly attractive Prelude, Elegy, and Finale for strings, from which 
part of the middle movement may be quoted on account of its lyrical 
appeal: 


566 MUSIC IN BRITAIN 1916-1960 


Fricker does not seem to have introduced any serial features into his 
compositions until 1955 with Litany for Double String Orchestra and 
Sonnets for Piano. Even when his texture is unusually lean Fricker does 
not write music like Webern's, but has brought twelve-tone melodies, 
figurations, and chords into an already assured style. In his Twelve 
Studies for Piano (1960) the contrasts are not between styles learnt from 


THE YOUNGER GENERATION 567 


this or that source—Hindemith, Stravinsky, Bartók, Webern—but of 
keyboard technique in varying moods invented by an integrated 
imagination. 

Fricker successfully cast his First String Quartet (1949) in one move- 
ment; his Violin Sonata (1950) and Second Quartet (1952) finish with 
third movements that are also the slowest; all three movements of his 
Second Symphony he regards as rondos. His command of extended 
form served him well in 1958, when he undertook his only commission 
for an imposing choral work, choosing for his text Cynewulf's Christ 
interwoven with Hallelujahs and extracts from the Mass, and entitling 
the whole work The Vision of Judgement. It moves from a stern peniten- 
tial atmosphere to sustained apocalyptic glory, and is grateful to chorus, 
soloists and the many instrumentalists required. The normal tautness of 
his musical textures is relaxed in the immediately attractive Octet of 
1958. 

Iain Hamilton showed honest romantic rapture in his Op. 1 Varia- 
tions for Strings, but a Symphony heard first in 1951 could not avoid 
comparison with Fricker’s of that year. It took wing powerfully 
in fine melody and reached no satisfactory culmination in its finale. 
ЕпсКегз, on the other hand, showed its first movement to be the 
shortest, remaining menacing and restrained even during passages of 
sultry lyricism; expansive magniloquence crowned the work in its finale. 
Hamilton's Violin Concerto of the following year was well designed and 
intensely emotional in its appeal, but he then turned from this romantic 
mode of expression, with its high violin melodies poised above alluring 
discords. Since then he has experimented with chamber works, song- 
settings, piano pieces, Scottish dances, a concerto for jazz trumpet, and, 
more recently, serial works, following his study of Webern, Stock- 
hausen, and other Europeans. It is only in these later works, which lie 
outside the scope of this chapter, that Hamilton achieves a fully assured 
style. 

Neither Anthony Milner (b. 1925) nor Malcolm Williamson (b. 1931) 
has followed Fricker in concentrating upon concert-hall instrumental 
composition. They have had little difficulty in finding words, liturgical 
and secular, to excite their musical 1magination towards an expression 
as full-blooded as sincerity and modern sophisticated taste will allow. 
Milner is the less romantic of the two and the more conservative 
technically. 

Milner's efficient chamber and orchestral pieces show that he is 
dependent upon words or religious concepts to stimulate his invention. 
In addition to an unaccompanied Mass and setting of vespers-psalms, 
Milner's best works include his /mproperia for double chorus, strings, 


568 MUSIC IN BRITAIN 1916-1960 


and organ (1949) which, from its recourse to instruments, cannot be 
intended for church use in the Good Friday liturgy, The City of Desola- 
tion (1955), a cantata with soloists and orchestra, The Water and the 
Fire (1962), a Three Choirs Festival oratorio on texts chosen by him 
incorporating material from the Easter Liturgy, Our Lady's Hours 
(1957), a song cycle for soprano and piano, and Variations for 
Orchestra (1959) upon a fifteenth-century theme. 

Williamson was born in Australia and heard little twentieth-century 
music until his first visit to Europe in 1950. His deprivation did him 
little harm, for he suffered no pressure from vogue and could form 
his own judgements on Delius, Shostakovich, and the scores of 
‘musicals’. He was ready to receive and examine any music in which he 
recognized vitality. By 1953, when he settled here permanently, he had 
taken lessons with Elisabeth Lutyens and been persuaded to use serial 
techniques. His conversion to Roman Catholicism awakened his interest 
in Dunstable and other medieval musicians (he uses isorhythm in the 
organ variations on the Coventry Carol, Vision of Christ-Phoenix) as 
well as in the music and theories of Messiaen. His one-time employment 
as a night-club pianist and his admiration of tunes in Richard Rodgers's 
musicals led to his own writing of more advanced musicals, including № 
Bed for Bacon. 

Williamson's exuberance finds full rein in his concertos, as it did in 
the opera Our Man in Havana (1963). (One of his best works before 
this was the unaccompanied Symphony for Voices of 1962.) Our Man in 
Havana was an audacious and unequal undertaking, but its best parts 
are not confined to the most serious or the most serial. 

To choose names from among the still younger composers becomes 
increasingly invidious. Of those who had begun to make their mark 
before 1960, which is the chronological limit of this chapter, and have 
gone on to prove their worth, one must mention Alexander Goehr 
(b. 1932), Hugh Wood (b. 1932), Peter Maxwell Davies (b. 1934), 
Harrison Birtwistle (b. 1934), and Nicholas Maw (b. 1935). Goehr, 
Davies, and Birtwistle have become known аз ‘the Manchester school’ 
since they were fellow-students there; Wood studied with Milner, 
Hamilton, and Seiber, Maw with Berkeley and Nadia Boulanger. 


VII 


AMERICAN MUSIC: 1918-1960 


(i) Music In The United States 


By RICHARD FRANKO GOLDMAN 


THE BACKGROUND 


WESTERN musical art is essentially international, but its expressions 
and institutions vary locally. We can often distinguish English, French, 
German, Italian, or Russian music by idiosyncracies of style, and can 
trace these elements of style to national or regional backgrounds of folk- 
music deriving from a remote past and from traditions of composition 
handed down from one generation to another by a succession of 
acknowledged masters. Elaborated or sophisticated art, and the 
institutions for its presentation and dissemination, depend on a sense 
of continuity and on a consciousness of a more or less homogeneous 
culture. It is evident that these conditions for the development of an 
indigenous art did not prevail in America for many generations, and 
that an American music of consequence could not have evolved earlier 
than the twentieth century. 

American music inevitably reflected, as did American culture as a 
whole, the diversity of its European origins. Musicians born or settled 
in America were content to make the best effort within their powers to 
continue, or perhaps to discover, the traditions of a highly developed 
musical language. Musicians arriving from Europe in each generation 
brought with them their ideas, traditions, and technical skills. There 
were, unfortunately, no composers of dominating personality or impor- 
tant attainment among them. American-born composers, until the 
beginning of the twentieth century, necessarily went back to Europe, 
and principally to Germany, to learn their craft. With few exceptions, 
they became very minor German composers; and although almost 
since the settling of the continent there had been a sense of the need for 
discovering an artistic America, this discovery could not be effected by 
exhortation. If certain American materials were already present, the 
conditions for their recognition were not. And while Americanism as 
such remained a preoccupation, there was at the same time an intense 
desire to perpetuate, without the historical and social conditions that 


570 AMERICAN MUSIC: 1918-1960 


had made them possible, many of the institutions and practices of 
European artistic life. One feels that many of these efforts were primarily 
competitive in impulse, and to some extent based on an idea of what 
was proper rather than what was necessary. 

In considering any aspect of American culture, a fundamental condi- 
tion of American society must always be kept in mind: that the country 
was settled by immigrants from every European background, and that 
eventually the consciousness of an African background also became an 
element of primary importance. Thus a part of the problem of the 
American artist has always centred on his search for a real and usable 
past, and one that could be assimilated in a sense as a common past 
for all Americans. The first social and cultural traditions were of 
course English, and English became the language to be learned by 
Germans, Mediterraneans, East Europeans, and all others. But the 
English past, including English folk-song or English hymnody as passed 
on through New England musicians, represented a tradition alien 
to increasing numbers of Americans of non-English background. The 
theoretical *melting-pot' produced its tensions and conflicts, and it 
would be a rash historian who would claim that these have ever been 
completely resolved in American society. 

Thus, while nineteenth-century American composers of serious 
inclination and ambition were endeavouring to establish an American 
art of conventional English and German pattern, they failed to notice 
a lively and diversified popular art which was developing all about them. 
The work of these composers was a provincial echo of the most facile 
styles of Europe. From time to time, gestures were made in the direction 
of the native picturesque, beginning with Hans Gram's Death Song of 
an Indian Chief (1791), said to be the first orchestral score published in 
the United States, and on through a variety of naive and condescending 
evocations of the noble Indian and the happy, carefree Negro. Exception 
must be made for the more interesting efforts of William Henry Fry 
(1813-64), the first musical Americanist of consequence, and of Louis 
Moreau Gottschalk (1829-69), a virtuoso pianist of New Orleans, who 
took a genuine and cultivated interest in the Creole music of the 
Caribbean as heard in Louisiana, and attempted to recapture its 
flavour in many of his once-popular piano pieces. But in general, a 
genteel and conservative manner, typical of provincial urban society, 
set the tone of American concert life, and the representative composers, 
from John Knowles Paine (1839-1906) to Edward MacDowell (1862- 
1908) and Horatio Parker (1863-1919) echoed the styles of Raff or 
Grieg. The so-called Boston school, beginning with Paine, did however 
establish a standard of professional competence and craft, and provided 


THE BACKGROUND 571 


a basis for the schooling of later musicians. Paine became the first 
Professor of Music at Harvard University in 1875; Parker later held a 
similar post at Yale, and MacDowell at Columbia. They helped to 
establish the important precedent of the teaching of composition and 
other branches of musical art in American universities, a development of 
the greatest significance for American music in the twentieth century. 

Popular music in nineteenth-century America was not (nor is it yet) 
a unified folk-music, if indeed it was a folk music at all within the usual 
European definition. Perhaps that is why it was not recognized as such 
even by those composers who most vehemently lamented the absence 
of a folk tradition in America. American popular music was the music 
of a diversity of people remote from the traditions of the concert hall 
or of German symphonic music. It embraced hymns, revival songs, 
temperance tunes, Negro work songs and spirituals, imitation (*black- 
face’) Negro music of the minstrel shows, brass band quicksteps, the 
composed songs of Stephen Collins Foster (1826-64) and others, and, 
in a number of areas, survivals of British, Spanish, French, or German 
secular folk tunes. Eventually ragtime and jazz, a fusion of many 
elements, emerged in and around New Orleans. But all of these vigorous 
manifestations of idiosyncratic music were considered, until the 
twentieth century, as beneath the proper concern of cultivated musicians, 
with the sole exception of the surviving strains of British folk-song. 
These latter were regarded by some few composers as the only proper 
and respectable sources for an American idiom; all else remained either 
vulgar or alien. À new American consciousness, evolving in the twentieth 
century, was required for the recognition and absorption of popular 
diversity—cultural, racial, and even sectional—as the most genuine 
possibility for an indigenous expression. 

American music owes a great debt to Antonin Dvorak. Dvorak was 
among the first composers of universally acknowledged stature to visit 
the United States (1893—5), and the first to advise the American com- 
poser to look about him at home, with particular attention to the music 
of the Negro. Although Dvorák's own knowledge of Negro music was 
superficial, he at least took this music seriously and gave others the 
artistic courage to do so. He contributed greatly towards the creation 
of the new environment for the American musician that began to take 
shape in the early years of the twentieth century. The fact that he 
was a Czech was not without importance. Gilbert Chase! correctly 
points out that MacDowell's contemptuous dismissal of Dvorák's 
advice was probably motivated by the fact that Dvofák was not a 
German composer. But other composers were stimulated by Dvorak’s 

! America's Music (New York, 1955), p. 392. 


502 AMERICAN MUSIC: 1918-1960 


example and advice. Arthur Farwell (1872-1951) is best remembered 
for his serious study of Indian music, undertaken as a direct result of 
Dvofák's influence. But in 1903 Farwell also declared that music 
other than German— French and Russian especially— needed hearing 
in the United States. Farwell, an important figure in American musical 
history (although his compositions have not survived in the repertoire), 
felt strongly the need for an American manner of expression which, in 
his view, would embrace ragtime, as well as the music of the Negro and 
Indian, and which also would demand ‘new and daring expressions’. 
In 1902 Farwell founded the Wa-Wan Press for the publication of new 
American music, with the expressed purpose of launching “а progressive 
movement for American music, including a definite acceptance of 
Dvofrak’s challenge to go after our folk music’. The Wa-Wan Press is 
important in American music as the first concrete and organized 
effort to offer an opportunity of publication to composers representing a 
new orientation and spirit. 

Among the composers published by the Wa-Wan Press was Henry 
Е. B. Gilbert (1868-1928). Although Gilbert was MacDowell’s first 
American pupil, his musical aesthetic was diametrically opposed to that 
of his teacher. Gilbert, like Farwell, saw the importance of the native 
background. He travelled widely throughout the United States, earning 
his living in a variety of ways, and absorbed a great deal of what he 
heard. Like Gottschalk, he was attracted by the Creole music of New 
Orleans, and based his symphonic poem 77e Dance in Place Congo 
(performed as a ballet-pantomime in 1918 at the Metropolitan Opera 
House) on tunes published in 1886 by the New Orleans writer George 
W. Cable. Gilbert's Comedy Overture on Negro Themes (1911) is still 
played occasionally. In other compositions, Gilbert turned his attention 
to ragtime, to the music of Latin America, and to the hitherto neglected 
tunes made popular in the blackface minstrel shows of the 1850s and 
1860s. Gilbert was perhaps the first to recognize that these tunes had 
become a kind of American folk-music. 

Farwell and Gilbert, and other composers of their generation, repre- 
sented what would eventually become one of the main streams of 
American musical development. They were aware of the great variety of 
unexplored and unexploited material available, and their work, though 
not in itself of permanent musical value, prepared the ground for later 
composers by calling attention to that material and to its possibilities. 
Their contribution towards the establishing of an American musical 
consciousness can be seen at a distance as more important than that of 
MacDowell, who was recognized without question in his time (and for 
some time thereafter) as the ‘greatest’ American composer. MacDowell 


THE BACKGROUND 573 


achieved something of an international reputation, the first American 
composer to do so; and this accomplishment was important in that it 
gave Americans of his time a sense that a native composer of acknow- 
ledged stature had finally appeared. MacDowell's music has been over- 
praised and under-estimated; at its best it has charm and individuality, 
but the style he represented was that of a period reaching its end, and its 
connexion with the major movements of musical art was that of a late 
arrival. 

American music may be said to have been, until the end of the First 
World War, in the process of discovering itself. The war of 1914—18 
marked a turning point in the cultural as well as the political history of 
the United States. America's emergence as a world power brought with 
it a heightened national consciousness and a sense of history that could 
be translated into a new and vigorous expression. Earlier consciousness, 
as for example in Whitman, had been largely prophetic. The Transcen- 
dentalists had spoken for the individual. But the new sense of awareness 
was both collective and retrospective; a new identity, formed of 
European elements but in spirit non-European, took shape for the 
first time. It was not accident that the variety of memories and traditions 
existing during the periods of immigration and settlement should at this 
time have begun, at many levels of American culture, to fuse into 
memories and traditions with some meaning for all. The material of 
American culture required synthesis, and in so far as American culture 
can be said to have a character, it is synthetic in the literal meaning of 
the term. In art, as in politics, the emergence of Americans of other than 
English or German origin, dates effectively from 1918, and it is again 
not by accident that in order to be American, American art had first 
to acquiesce in internationalism. The prominence of American com- 
posers of, for example, Italian descent (Walter Piston, Paul Creston, 
Peter Mennin, Norman Dello Joio, Gian Carlo Menotti) and of Jewish 
descent (George Gershwin, Ernest Bloch, Aaron Copland, William 
Schuman, Marc Blitzstein, Hugo Weisgall, Leonard Bernstein, Leon 
Kirchner) is a purely twentieth-century phenomenon. 

From this standpoint, a more immediate interest and awareness of 
contemporary European currents was also of importance. The end of an 
age throughout the world, reflected in art as conscious revolt, gave 
American artists a sense of being able to overcome the lateness of their 
start and to begin a new movement on an almost equal footing. The 
universal awareness of new aesthetic orientations and ideals had a 
liberating influence, both technically and psychologically, on artists 
throughout the world, with whom, for the first time, Americans were 


able to feel a sense of community and of contemporaneity. The 
38 


574 AMERICAN MUSIC: 1918-1960 


American musician felt a welcome sense of release at being able to 
throw off, at the same time as everyone else, the weight of competition 
with Beethoven, Brahms, and Wagner, or even with Raff and Grieg. A 
surge of freedom, reflected in ‘advanced’ music, was necessary not only 
as an assertion of independence, but even as a condition for making 
conservative music once again interesting. 

The influence of European events was felt not only as a result of 
travel and communication, but also through the arrival in the United 
States as permanent residents of musicians of the quality of Edgard 
Varése (1916) and Ernest Bloch (1916). Whether Varése and Bloch 
are to be considered American composers and musicians is open 
to question, but one cannot question their importance and in- 
fluence in the American musical scene. А younger ‘immigrant’ 
whose importance was temporary, but who attracted much attention 
for a brief period, and who did much to make the public conscious of a 
radical new stream of musical thought, was the Russian-born Leo 
Ornstein (b. 1895). In the years before 1920, a small current of French 
music and influence was reflected in the compositions of Charles Martin 
Loeffler (1861-1935), born in Alsace, but long resident in Boston, and 
of Charles Tomlinson Griffes (1884-1920). Griffes was a composer of 
marked gifts, and several of his works entitle him to continued respect 
and attention. Among these are The Pleasure Dome of Kubla Khan, 
written as a piano piece in 1912, and performed in an orchestral version 
in 1919. The White Peacock was similarly composed for piano in 1915, 
and orchestrated in 1919. The Poem for Flute and Orchestra (1918) and 
the Sonata for Piano (1918-19) are also representative. Griffes may be 
described as an eclectic composer, but he was alive to the interesting 
currents of his time, and much of his music can still be heard with 
pleasure. He was aware of the contributions of Debussy, Busoni, and 
Stravinsky, and he was interested in Arabian, Japanese, and Amerindian 
music. It is worth noting that among those who encouraged him most 
strongly was Farwell. 


CHARLES IVES 


Before turning to the main currents of American musical activity 
in the 19206, it is necessary to consider the extraordinary phenomenon 
of Charles Ives, one of the truly great *originals' in the history of music. 
Ives was born in Danbury, Connecticut, on 20 October 1874, and died 
in New York on 19 May 1954. His life thus spans the entire period in 
which American music may be said to have established itself. The 
extraordinary circumstance of Ives's life and work is that none of his 
music was heard or known, even in professional circles, until the 1920s, 


CHARLES IVES 575 


and that important public performances of his major works did not take 
place until 1939 and after. With the exception of the Concord Sonata, 
which Ives had privately printed in 1919, and a collection of 114 Songs, 
similarly printed in 1922, none of Ives's music was available in published 
form until 1929, by which time he had ceased composing. 

The position of Ives in American music, indeed in the history of 
contemporary music, is unique. One need not agree fully with his bio- 
graphers, Sidney and Henry Cowell, that he was one of ‘the four great 
creative figures of the first half of the twentieth century'—the others 
being Schoenberg, Stravinsky, and Bartók— but it is impossible to 
question his originality, his power or his authenticity. He anticipated 
many of the ‘discoveries’ of his contemporaries and juniors, using 
techniques not widely accepted until a generation later, including 
harmony based on fourths, polytonality, tone-clusters and poly- 
rhythms of extreme complexity. Passages in his work have a freedom 
and boldness of imagination that is not only astonishing for its time (all 
of Ives's work was done between 1891 and 1921—the most important 
was completed by 1916) but that still seemed inventive and advanced to 
musicians at the mid-point of the century. Most important, however, he 
was the first to realize completely what it might mean to be an American 
composer, and his career and work illuminate in the most telling 
manner what it is to discover a musical world. 

The world of Charles Ives was a world in which Beethoven and Bach 
met Stephen Foster, New England hymn-singing, ragtime, circus quick- 
steps, country fiddling, and every variety of spontaneous music-making 
that characterizes the activity of a rural environment. This world was 
shaped by an imagination conditioned in childhood to believe substance 
more important than style and to shun dependence on musical customs 
and habits. But it was an imagination coupled with intellect, with deep 
roots in the New England past, and strengthened by a complete 
acceptance of the New England environment as expressed in its most 
ideal terms by Emerson and Thoreau. Ives had the independence that 
can come only from tradition and discipline, the solid base over which 
the most diverse elements can be assimilated and made into a new unity. 

Ives received a conventional training from his father, a bandmaster 
and music teacher who was himself ceaselessly interested in experiment 
and fascinated by the nature of sound itself to the extent of improvising 
devices for the performance of quarter tones. Beside giving his son a 
conventional training, he invented exercises designed 'to stretch our 
ears and strengthen our musical minds'. Later, young Ives studied under 
Horatio Parker at Yale, doing ‘correct’ exercises for the classroom, but 
writing for his own satisfaction music that shocked everyone but 


himself. His Variations on ‘America for organ, composed in 1891, 


AMERICAN MUSIC: 1918-1960 
employed bitonality, and other pieces composed before 1900 utilized 


576 


other techniques and materials that could only have seemed incom- 


prehensible to his contemporaries. Here is an extract: 


Var. III Allegro 


Қ ld. 
ДЇ! ДШ 
GNL GN 


CHARLES IVES 577 


Se rey Der БЕЯ 


oe те 1 
[— JepES.: 7 ccc ecep pn 
Е ае ооо кси сыс ыйы ышк шш 


Characteristically, these early pieces include works based on or para- 
phrasing hymn-tunes, circus marches, or other types of popular music, 
set in textures ranging from extreme simplicity to extreme dissonance 
and rhythmic complexity. | 

Ives evolved his techniques and chose his materials not according to 
апу formulated theory of musical composition, but rather in an attempt 
to solve what to him were important questions about the nature of art 
and expressiveness. To write ‘correct’? music (which he was quite able 
to do) was of no interest to him. Ives sought to express a variety of 
things (this was his view of ‘substance’ as opposed to ‘style’) that to him 
were vital and meaningful: the particular quality of life as perceived 
and enjoyed by a given man in a given time and place, here, in New 
England, and beyond that the transcendental reality of man and the 
universe in terms that had come most immediately from Emerson. Ives 
was not afraid of large thoughts; his own philosophy was forceful and 
articulate, and for him an art divorced from life, whether eternal life or 
daily life, was utterly without point. 

Whatever the merits of Ives's philosophy, the student of American 
music can learn much by reading his Essays before a Sonata,* privately 
printed in 1920 as an adjunct to the Concord Sonata for piano. It 
becomes clear that Ives 15 the first composer in America who did not 
look to European music for models, but to Bach and Beethoven as if he 
felt that he understood what they meant. 

It is not surprising that Ives could find no audience. No public was 
ready for this music, and Ives decided soon after leaving the university 
that he would have to make his living as a business man. Characteristi- 
cally, he did not feel that this involved any conflict. In a statement made 
for the Southern poet and novelist Henry Bellamann, Ives wrote:? *The 
fabric of existence weaves itself whole. You cannot set an art off in the 
corner and hope for it to have vitality, reality and substance. There can 
be nothing **exclusive" about a substantial art. It comes directly out of 


! Reprinted in Three Classics in the Aesthetic of Music (New York, 1962). 
2 *Charles Ives: the Man and his Music’, Musical Quarterly, xix (1933), p. 45. 


AMERICAN MUSIC: 1918-1960 


578 
Ex. 249 
Allegro moderato 
CE Был ылы а 
ЕІ. 
(СР)! 5 Т qe. 3 muss ЕЕ и ииси 
ES ENT мест 5-2: --! уг КӨН RE x 
Ob. = 2 
"Cil 
in Bb 
Bn. 
Hn. |Р 
inF 
"Trpt. 
in Bb 
Tromb. 
Tuba 
Timp. 
Piano 
ЛЛУ М 
Allegro moderato К trem. sempre (ad lib.) a 
em к> ЦЕ he > 
I dw == XL ж 
Violins (ZF) trem. sempre (ad lib.) ге Кен! 
Г: i is 2 < : 
II 
(27) ар. 
> f 
= 
Vioks || (2 4 TN ~y "mds 
- aha == - E 


Cellos (072 ) I A 


CHARLES IVES 579 


в. ees 
мы —— Ó ШЕЛ Ч 
Ob ТЕ И == ы 
сі. 
in Bb 
Bn. 
Hn. 
inF 
Trpt. 
in Bb 
Tromb. 
Tuba 
Timp. 
Piano 
I 
3 [EE OR] a ay 
Violins 
ГИШИ атре абе) 
“еб за ^h ВВ ратман з 4 


tae р Е ЕЕ-Е ЕЕ = 


Violas 


Cellos 


580 AMERICAN MUSIC: 1918-1960 


the heart of experience of life and thinking about life and living life. My 
work in music helped my business and my work in business helped my 
music.’ 

Occasionally Ives was able to hear portions of some of his works in 
readings by friends or occasionally members of local theatre orchestras, 
hired at his own expense. For a time he made efforts to have his work 
performed, but the total lack of understanding, and in many cases the 
derision, with which he met, eventually discouraged him. A reading of 
parts of his First Symphony by Walter Damrosch in about 1910 
produced only annoyance on all sides. Ives resigned himself to the fact 
that his way would have to be made slowly. In 1919 he decided to 
publish himself some of the works he had written, hoping that if enough 
copies were distributed (gratis, of course) a few, through the operation 
of the law of averages, would find sympathetic readers. This proved 
true, and the first sympathetic notices of Ives’s music were written by 
Bellamann. Some time afterwards, the enthusiastic interest of a number 
of musicians, including Nicolas Slonimsky and Henry Cowell, and 
eventually many others, was aroused, and by about 1930 (though not 
earlier) the name of Ives began to be current in avant-garde circles. The 
first major performance of a major work of Ives was that of Three 
Places in New England, performed by an orchestra directed by Slonim- 
sky in 1930 in Boston. Examination of Ex. 249 will give the reader some 
idea of the harsh dissonances, polyrhythms, and performance difficulties 
involved in this work. It was not, however, until the performance of 
the entire Concord Sonata by John Kirkpatrick in New York in 1939 
that Ives achieved anything approaching ‘recognition’ by even a small 
part of the musical public. This example from The Anti-Abolitionist 
Riots will give the reader an idea of Ives’s style of writing for the piano, 
as well as of his improvisatory style without bar-lines: 


Ex. 250 
Adagio Maestoso 


um NP EQ 
6—6] те с 
У ра: 
His —9- 
0—2 = 


М2 
— = quen usos 
Р 7 ^ 3 
be n] А 
ЕЕ 
о а ee ee ee 
| ^ 
"m е 


581 


CHARLES IVES 


— 
=> 
m 


| 
( 


К 


>, 2 
a oe EE occ e 


и даши: 


= В et A 


VY 


582 AMERICAN MUSIC: 1918-1960 


Ives wrote much, including four symphonies, a variety of music for 
various small and large instrumental combinations, some chamber 
music, choral works, music for piano, and songs. The songs are extra- 
ordinary in their variety, and often in their originality, and they con- 
stitute in themselves a chapter in American music. No other American 
composer has approached Ives’s richness or power in this field. The 
Concord Sonata, actually entitled ‘Second Piano Sonata “Сопсога, 
Mass., 1840-60” ’, is almost universally recognized as the most powerful 
and impressive work of its kind in American music. Its technical pro- 
blems are great; Ives was notoriously indifferent to practicability and 
at times wrote music that he himself admitted mighi be impossible to 
perform (see Ex. 250). This constituted, in his view, no reason for not 
writing it, if the expression in itself was justified by what he felt necessary 
to say. It was not until 1964, for instance, that Leopold Stokowski 
was finally able to present a complete performance of the Fourth 
Symphony. 

The Concord Sonata is in four large movements, each an evocation of 
Concord’s past. The first is entitled ‘Emerson’ ; the second, ‘Hawthorne’; 
the third, ‘The Alcotts’; and the final movement, ‘Thoreau’. Ives him- 
self wrote programmes for each of the movements, and described the 
work as ‘an attempt to present one person’s impression of the spirit 
of transcendentalism ...’ He uses the terms ‘impression’ and ‘impres- 
sionistic’ several times; yet the music is not what is generally associated 
with impressionism. It is intensely strong, with a structure obeying its 
own organic laws; there is nothing else quite like it in the literature of the 
piano. Powerful dissonances alternate with quiet quotations of hymns, 
delicate passage work with brash allusions to circus parades. It is an 
impression of all that Ives knew of New England both in thought and in 
sound, an impression of a New England universe. The scope is not 
exaggerated; among the many sketched and unfinished works of Ives 
is a Universe or Universal Symphony, ‘not intended to be completed 
by the composer himself or by any other man, because it represents 
aspects of life about which there is always more to be said’. 

That Ives was ahead of his time in many ways hardly needs emphasis. 
When first written his harmonies could not have failed to seem barbar- 
ous, his melodies and sonorities crude or vulgar, his rhythms unplayable, 
and his musical orthography merely eccentric. By the time his music be- 
came known, however, most of his ‘innovations’ had become familiar in 
the works of other composers, and his use of popular materials had 
become a part of American musical language. But Ives’s importance 
does not reside in his discoveries or anticipations; it resides in the facts 
that his music communicated life and vigour, that he was neither a 


GEORGE GERSHWIN AND AARON COPLAND 583 


miniaturist nor a docile follower, and that what he had to express spoke 
persuasively to a few musicians in the 1920s and 1930s and to a large 
public thereafter. It is perhaps incidental, but by no means unimportant, 
that Ives's music was clearly the first music of large intent and dimension 
that could not have been written by any but an American. He believed 
strongly in himself as a man living in an environment which he con- 
ceived in a way no one before him had done. That is why his place is 
large in American musical history, and why he has come to be an 
image of great importance for so many American musicians. It is the 
substance that in the end justifies Ives, as he himself would have wished. 
There are crudities in the style. Some of these might have been resolved 
had he had the opportunity of hearing more of his music, though this 
cannot be certain. Aaron Copland suggested! that Ives ‘lacked neither 
the talent nor the ability nor the métier nor the integrity of the true 
artist, but what he most shamefully and tragically lacked was an 
audience’. And further, *...the drama of Ives...is that of every 
American composer of serious pretensions. The problem of the 
audience—not a passive audience but an active one—an audience that 
demands and rejects music, that acts as a stimulus and a brake, has never 
been solved.’ 

Obviously, Ives's influence, both technical and spiritual, could not 
have been felt until his music had become known, and in the 1920s it 
was known to very few. But many of the strands of American music in 
that decade were converging towards the kind of synthesis that Ives had 
in a sense achieved, and one can see how justly the discovery of his 
music coincided with the emergence of a new generation of American 
composers. The native materials of ragtime, jazz, and popular hymnody 
were put to use by a number of composers who were at the same time 
learning the devices and techniques of ‘modern’ music from Stravinsky, 
Milhaud, Hindemith, Ravel, Bartók, Prokofyev, and the other new 
masters in Europe, as well as from experimenters at home. It remained 
only for the impact of Schoenberg and his school to be absorbed, some 
years later, for American music to become completely international. 


THE DECADE 1920-30: GERSHWIN AND COPLAND 

The 1920s were eventful years, not only because of the arrival on the 
scene of a number of strikingly gifted composers, but also because for 
the first time concerted efforts were made to give their works a hearing. 
The International Composers' Guild was founded in 1921 by Edgard 
Varése (1885-1965) and Carlos Salzedo (1885-1961) for that purpose, 


! Our New Music (New York, 1941), pp. 160-1. 


584 AMERICAN MUSIC: 1918-1960 


as was The League of Composers two years later. The New Music 
Society of San Francisco was founded by Henry Cowell in 1927, and, 
beginning in 1928, Aaron Copland and Roger Sessions joined forces to 
produce the stimulating Copland-Sessions concerts. These were among 
the most important groups, but there were others; and contemporary 
music, with special emphasis on contemporary American music, began 
to engender an atmosphere of enthusiasm and excitement. These groups 
not only performed, but did active propaganda, and although they could 
not bring new music to the attention of the great public, the stir they 
created eventually spread. A few sympathetic critics, notably Paul 
Rosenfeld, appeared on the musical scene, and many of the composers 
themselves turned their hands to writing. The magazine Modern Music, 
established by The League of Composers in 1925, was, until its dis- 
continuation in 1947, an influential force in American musical life. 

The atmosphere of the early twenties is suggested by some lines of 
Paul Rosenfeld, written in 1922. ‘It 15... no longer true... that there 
is no vitality in the native musical production. What has been charac- 
teristic of the painting and the writing of the land has now become 
characteristic of the music. The country can produce really gifted 
youth. Society has become sufficiently settled to permit the talent to 
assert itself; there is even . . . enough of a real community to permit the 
musical gesture, the gesture of the interpenetrated group, to be made."! 

It may be said that the characteristic of American music during the 
decade was its diversity of style and its unity of spirit. The spirit was one 
of new confidence and assertiveness; composers (and some performers) 
felt united in a common cause, that of new music and the modern 
temper. For the first time, too, American musicians felt that they were 
part of the main current, and not timorous artisans on the periphery. 
The atmosphere was lively and the activity enormously accelerated. 
This was the decade in which the names, and the music, of Aaron 
Copland, Roger Sessions, Roy Harris, Virgil Thomson, Henry Cowell, 
Walter Piston, Randall Thompson, Howard Hanson, George Antheil, 
Carl Ruggles and Wallingford Riegger were becoming known, and 
in which George Gershwin composed his Rhapsody in Blue. It was 
of course also the decade in which the musical style known as jazz was 
for the first time taken seriously, as something vital and new and 
unquestionably American. 

The jazz-influenced idiom was but one of the manifestations of 
American music in the twenties. John Alden Carpenter (1876-1951) 
of Chicago had employed a type of ragtime jazz in his Concertino ` 
for Piano and Orchestra (1915), in his ‘jazz pantomime’ Krazy Kat 

1 Paul Rosenfeld, Musical Chronicle, 1917-1923 (New York, 1923). 


GEORGE GERSHWIN AND AARON COPLAND 585 


(1921), and in Skyscrapers (1926). Carpenter's music enjoyed a vogue 
for a time, for it was basically conservative enough to appeal even to 
the conductors of major orchestras and their audiences. A more 
seriously assimilated jazz appeared in several works by Louis Gruenberg 
(1884-1964), one of the composers active in the early days of the 
International Composers Guild and The League of Composers. 
Gruenberg's The Daniel Jazz (1923) and Тйе Creation (1924), both for 
voice and a combination of eight instruments, had some solidity and 
originality as stylizations of Afro-American material. (It is interesting to 
note that The Daniel Jazz and Milhaud’s La Création du Monde were 
written in the same year.) But it remained for two composers of quite 
different orientation to establish definitely the consciousness of jazz as 
an artistic reality in American music. 

George Gershwin (1898-1937) achieved early fame as a writer of 
popular tunes and musical comedies. But he studied music seriously 
with Rubin Goldmark, then the most esteemed teacher of composition 
in New York, and felt that material derived from jazz and the idiom of 
American popular songs could be used in works on a larger scale, even 
symphonically. The Rhapsody in Blue, designed as a demonstration of 
*symphonic jazz', was commissioned by the band leader Paul Whiteman, 
and was first performed on 12 February 1924, in New York. Its success 
was instantaneous, and the date is a memorable one in American music. 
Probably no work written in America is as well known, nor is there any 
other work that has appealed to such an extent to audiences on every 
level. It may be termed one of the genuine American ‘classics’. This is 
not to make a musical judgement, for the Rhapsody in Blue is a hybrid 
work, which in fact is not jazz, but which combines a real feeling for 
American popular style with a romantically eclectic harmonic and 
orchestral style. Ex. 251, showing one of the Rhapsody’s principal 
themes, clearly illustrates these features: 


Ех. 251 


(poco scherzando) 
а 


Эй 


al 
ИШИ] 
a - 
Ша 


586 


AMERICAN MUSIC: 1918-1960 


Orch. 


GEORGE GERSHWIN AND AARON COPLAND 587 


[21а жа 1 ees poe Що 
Bee EET fel 
==” 


Nevertheless, the Rhapsody in Blue was a genuinely original conception, 
realized with great success; and although the idea of ‘symphonic jazz’ 
could not survive its inherent contradictions, the Rhapsody influenced 
American music both positively and negatively. It demonstrated both 
what could be accomplished on a large scale by a talented composer 
using a popular idiom, and at the same time the serious limitations of 
using a conventional ‘symphonic’ style as a vehicle for this type of 
musical speech. The Rhapsody compelled the attention of both serious 
and popular musicians, however, and the effects of its success cannot be 
under-estimated. Gershwin continued active in both musical comedy 
and serious composition for the remainder of his life. His works in the 
latter category include a Concerto in F for Piano (1925), Three Preludes 
for Piano (1926), the orchestral An American in Paris (1928), a Second 
Rhapsody (1932), the Cuban Overture (1934), and his masterpiece, the 
opera Porgy and Bess (1935). 

Gershwin's natural language was made up of ragtime, the blues and 
jazz, to which in Porgy and Bess he added melodic contours resembling 
spirituals. Out of these elements he made something unmistakably his 
own. Porgy and Bess has not only never been equalled in its genre, it 
has not even been approached. Gershwin called it a ‘folk opera’, and 
whether or not it is opera at all, in the conventional definition, would 
hardly seem to matter. Its tunes are part of the American vocabulary, 
and it has outlasted every other American work for the musical stage. 
Gershwin had the instinct of a natural dramatist, and while Porgy and 
Bess may be defective in terms of formal operatic convention, or even 
material, it is moving and compelling. Its success, not only at home, 
but throughout Europe, indicates that it makes its point as authentic 
American music that could have come from no other place in the 
world. 


588 AMERICAN MUSIC: 1918-1960 


On 11 January 1925, not quite one year after the first performance of 
the Rhapsody in Blue, the first major work of another young American 
composer was performed in New York. This was the Concerto for 
Organ and Orchestra by Aaron Copland. Copland, born in Brooklyn 
in 1900, was, like Gershwin, a pupil of Goldmark’s, but he had gone to 
France where he had become a student of Nadia Boulanger. His 
striking talent was recognized not only by her, but also by the 
American composer Marion Bauer, who had called his name to the 
attention of The League of Composers. Copland came on the scene 
with considerable force; his first works demonstrated unusual power and 
originality, and there was little question from the start that a major 
voice in American music had arrived. The Organ Concerto had many 
derivative elements, notably from Stravinsky, and was not wholly 
successful (Copland later rewrote it as his FirstSymphony, without organ). 
But Copland's next work, commissioned by Sergey Kussevitsky, stands, 
like the Rhapsody in Blue, though on a different plane, as a milestone in 
American music. This was Music for the Theater, a suite in five move- 
ments for small orchestra. Here a ‘jazz’ style, combined with sophisti- 
cated twentieth-century techniques of composition, including poly- 
tonality, resulted in a work of exuberant vitality which for the first time 
indicated how effectively ‘Americanized’ the international idiom could 
become. Although Milhaud had already written his ‘jazz’ ballet, La 
Création du Monde, a comparison of the two works makes clear how 
much more direct and meaningful was Copland's relation to jazz 
material. 

Jazz is the essential American contribution to the world's music, but 
it has given rise to disputes that come near to being as subtle as those of 
the medieval scholastics. It was not, historically, until the 1940s that the 
discussion of jazz assumed the dimensions of a new musicology, with 
learned and pseudo-learned contentiousness devoted to defining what is, 
or was, jazz, and what is not. From the purist point of view, jazz must 
be improvisatory, among other things. Although this is true, it also 
remains true that it can be the informing element in music that is not 
literally jazz, but that is either an evocation of jazzin art music, or else the 
reflection, in formal terms, of some of its characteristics. In the 1920s, 
no such fine lines were drawn, and the distinction between what is 
today properly called jazz and the music that derives from it was 
neither sharp nor consistent. The historian of American music must 
note that ‘true’ jazz had actually to be revived in the 1940s and 1950s, 
and again in the 1970s. 

With these reservations, Copland’s early style must still be described 
as a ‘jazz’ style. All the works of this period, and notably the Piano 


GEORGE GERSHWIN AND AARON COPLAND 589 


Concerto of 1926, are based on characteristically American rhythms 
which Copland himself understood as jazz. He used them consciously 
until he felt that he had exhausted their expressive possibilities. He 
noted also at the time that the use of a jazz style was “ап easy way to be 
American in musical terms';! but later in his career he found other, 
and equally persuasive, ways of writing recognizably American music. 

After 1930 Copland's music alternated between what has been 
called his ‘austere’ style, best represented in the Piano Variations (1930), 
the Short Symphony (1933), and Statements for Orchestra (1934), and a 
style based on folk-music elements, designed to appeal more im- 
mediately to a large public. An important inner evolution seems to have 
taken place in the years immediately preceding 1930, and the Piano 
Variations and Statements for Orchestra remain fundamental for any 
evaluation of Copland's mature achievement. These works, which are 
tight-knit, strongly dissonant, vigorous, and hard in sonority, may 
properly be called severe in style. The Piano Variations has remained, 
thirty years after its composition, a landmark in American composition; 
all of the works of this period, taken together, revealed to musicians 
the emergence of an original personality in full command, intellectually 
and emotionally, of a striking and forceful idiom. 

Copland himself felt that these works might be too advanced for any 
but the most sophisticated audiences, and in the mid-’thirties began to 
make a conscious effort toward achieving what he termed “ап imposed 
simplicity. To accomplish this simplification, and to please a wide 
audience without losing the marks of a personal style is a difficult task, 
yet Copland's success in this respect must be considered remarkable. 
Among the first works in this more accessible vein were the orchestral 
El Salón México, and a small opera, The Second Hurricane, written for 
performance by high-school students, both completed in 1936. Е/ Salón 
México is based on authentic Mexican folk and popular tunes, handled 
with great brilliance and vigour, with characteristically colourful 
orchestration and shifting rhythms. The use of folk-tunes, or of material 
related to folk-music, was carried a step further in scores composed for 
the dance productions Billy the Kid (1938) Rodeo (1942), and 
Appalachian Spring (1944), all of which were recast in concert form. 
Appalachian Spring remains the work by which Copland is probably 
best known to the general public, and through which he most effectively 
realized his hope of communicating with an audience not exclusively 
composed of devotees of avant-garde music. The diatonic and folk-like 
Copland manner may be studied in the excerpt from Appalachian Spring 
(Ех. 252) where a simple Shaker tune (“Тһе gift to be simple") is treated 


1 Our New Music, р. 227. 
39 


AMERICAN MUSIC: 1918-1960 


590 
Ex. 252 


с 1 с | | мо съ | (1 [| ранени ШШ | | 
ги SI села, ЦЕ A ALI: ШАШ 
өш тета? Р ): aum D: |) 
КЮ И М Шо! 
Шү S3 | SN | : 
оо ill ; Dm j 1 т пещ! m ПТ 
CB | ПОЗ. реа | 
Т сее, Ш | 
га : | ша 
ШІ (Ш ІШ! Т | ТІ | 
^ р wu : 
(Ши УШ 
A ES SA moh | 
th, (ЧЕ ГҮ 
0: Wn фт 
MN QU Qu aD Шыр | 
Jess [| ща 
W NIE UE [oth И 
| | КӨШ ШЕШ ШЕ Ur Dr m Ш 
s + | EH A S ШОН A m 
<< МӘ lo. Де 46 26 Ae» qe» RÀ m 
De A = — &,, = oo 48 в = =] c за 
"met 62 4: 974 ШЕ B^ ыт ве 43 E = ы Eos 


591 


GEORGE GERSHWIN AND AARON COPLAND 


D 
Sl ЕР 
FM али жиа аа E 
CS) 1 аара | 
LSY — 
() 
Шр ДЕШЕ, ara и ЕЕ 
sy ы pepe esp pe pon pe Т 
(Ду D 5b л LER 
2 a еее — етене 
0 
ЖЕН а Se eel Fe Se y | 
ee ee "ле 
| (jb —Àu————p———-—] 
eee (ae ae 


e 


: m 


| 


КР Гы 
[TS — -D— == ЫР” ^. 


м | 


(n Bb) I 


| 


{ 


| 
Щи 


ОБЕ 


G 


-( 


Glock 
sounds 8va 


=. 
hind 


ed 
e 
= 
< 


fie ria 
i di 
> ie 2 M 
LM ПЕЙ 
——- 
- = 
> 5 


592 AMERICAN MUSIC: 1918-1960 


canonically in a texture of great transparency. Many of his works in 
this vein have not only remained in the standard American concert 
repertoire, but have also received frequent performances by school 
orchestras and bands, a sure proof that they have deeply penetrated the 
American musical consciousness. Copland's scores for films are 
generally also in his simpler style, and have aided in making his idiom 
familiar. These scores are extraordinarily sensitive, and several must 
rank with the best music of this genre that has yet been produced. 
Copland's film scores include Of Mice and Men (1939), based on the 
Steinbeck novel; Our Town (1940), based on the Thornton Wilder play; 
and Тһе Heiress (1948), after the novel by Henry James. Related in many 
ways to his film scores as well as to his dance scores is the opera The 
Tender Land (1954), one of the few major works of Copland that has 
not enjoyed either popular or critical success. It may be said that 
Copland's music, while full of strength and motor energy, is on the 
whole more lyrical than dramatic in quality. The vocal writing in The 
Tender Land is sensitive, as it is in his settings of Twelve Poems of Emily 
Dickinson (1950), yet it is perhaps too elegant and refined to succeed on 
the large scale of the operatic stage. 

That Copland's work has been frequently imitated, or has served as 
a stylistic point of departure, seems evidence enough that the music 
exhibits a very personal and easily distinguishable style. This is true of 
both the ‘austere’ and the ‘accessible’ styles, which are in fact not as 
different as they may often appear to be on the surface. One may find 
elements of both manners in the Piano Sonata of 1951 or the Sonata 
for Violin and Piano of 1943,! and certainly in the Third Symphony of 
1946. Copland always writes with clarity and brevity, with an honest 
lyricism and lack of grandiloquence, and an elegance of line and phrase. 
The sound is generally open and lean, with a remarkable sense of 
timbre and spacing. These qualities remain constant in his work 
throughout his career and through what appears to be his adoption of 
different manners. 

In later works, such as the Piano Quartet (1950), the Piano Fantasy 
(1957) and the Nonet (1960), Copland adapted serial techniques, and 
although his previous work is essentially diatonic, while freely dissonant, 
the similarity in fundamental musical orientation to his earlier Piano 
Variations is striking. Copland's music is often open in texture, 
occasionally bitonal, and marked by the frequent occurrence of simple 
triads. The language is severely disciplined and the craftsmanship 
impeccable. The first page of the Piano Fantasy, with its wide spacing 
and isolated sonorities, is representative of Copland's more ‘severe’ style: 


1 The slow movement of which is recorded in The History of Music in Sound, x. 


593 


GEORGE GERSHWIN AND AARON COPLAND 


Ex. 253 


Slow d ; (d = circa 76) * in a very bold and declamatory manner 


4 
в 


Ред. оп each note 


9 ee 


же 


ES 
“ы a 


594 AMERICAN MUSIC: 1918-1960 


*Metronome markings throughout are to be understood as approximations only. 


During the second quarter of the century, no composer in America 
exercised greater influence on his contemporaries, both through his 
music (and his leadership in exploiting varieties of material), and 
through his activities as teacher and lecturer, organizer, and propa- 
gandist for his colleagues. Copland was one of the first to make im- 
portant contributions to music for those twentieth-century media, the 
radio and the motion picture. His contribution towards the musical 
re-orientation of the American musical scene cannot be overestimated. 

Apart from the débuts of Copland and Gershwin in the 1920s, there 
was an immense and spontaneous variety of musical activity along 
many lines. In retrospect, it must seem that this decade was in many 


GEORGE GERSHWIN AND AARON COPLAND 595 


ways the most expansive and optimistic of recent times. Financial 
support for ‘modern’ art was forthcoming from enlightened patrons, 
and the artist, as well as the small but growing audience, seemed to be 
looking (despite the often-cited disillusionment of the post-war period) 
confidently at the present and future. After the First World War, 
Western civilization in general began to look at its past in a new light— 
to reinterpret or paraphrase what it did not actively reject. Leadership 
in style and manner still came from the great European figures: Stravin- 
sky, Picasso, Joyce; later, in music, Schoenberg and Webern; and 
the American artist felt closer to these than he ever could have felt to 
Wagner or Brahms or Debussy. 

The decade was one, moreover, in which the new art—whether music 
or painting or literature—was not taken for granted. By mid-century, 
novelty as such was assumed as a matter of course to be a sine qua non in 
art; the composer was expected to produce a new technique or a new 
theory each week. But in 1925, the new could still be surprising and 
often pleasant, and attitudes could be tried out to see if they were 
possible to maintain. Experiment was fresh rather than desperate. 
Critical and public hostility towards new art may have been helpful, 
although not apparently so at the time to the artists concerned; it 
solidified their ranks in a way that has not been known since that period, 
when musical currents had not become as schismatic and mutually 
exclusive as they were to become later. 

All composers of any standing were ‘modern’. Some inclined more 
than others towards experiment for its own sake, or towards the explora- 
tion of ‘new musical resources’, as Henry Cowell phrased it. But even 
some of the most conservative composers felt that they were part of an 
important tide. Thus, for example, Howard Hanson (b. 1896) wrote 
symphonies and other large-scale works in a romantic style perhaps 
continuing the manner of MacDowell with echoes of Sibelius; yet 
Hanson as Director of the new Eastman School of Music organized in 
1925 a series of American Composers' Orchestral Concerts which were 
extremely important in helping to make new American music known, 
and in giving composers a chance (for the most part denied them by the 
established symphony orchestras) of hearing their larger work. Hanson's 
opera Merry Mount (1934) was one of the few works by Americans pre- 
sented by the Metropolitan Opera Association; others heard in the 
1920s and 1930s were two by the even more conservative Deems 
Taylor (1885-1966), The King's Henchman (1927) and Peter Ibbetson 
(1931). The most interesting American opera produced by the Metro- 
politan was Gruenberg's The Emperor Jones (1933), а Ча22” opera based 
on the play by Eugene O'Neill. 


596 AMERICAN MUSIC: 1918-1960 


RADICALS OF THE 1920s 


The avant-garde groups could hardly afford to present large orchestral 
works, much less operas, and their presentations had to be limited to 
chamber music and music in smaller forms. Büt much of this was 
memorable. The stir created by performances of the French-born 
Varése's Octandre апа Нурегризт (1924) still seems vivid. Cowell's 
pounding of tone-clusters on the piano (often with the aid of a board) 
seemed outrageous, and gave Cowell a notoriety that unfortunately 
obscured his achievements in other experimental aspects of com- 
position, and for a long time made people forget the mastery of 
dissonant counterpoint shown in such works as his Sinfonietta of 
1925. Virgil Thomson appeared on the American scene, as a young 
man returned from Paris, with Capital, Capitals, a setting of words 
by Gertrude Stein for four men's voices and piano, performed at 
one of the Copland-Sessions concerts of 1928, and with other works in 
which the inspiration of Satie and ‘Гез Six' made its most effective 
American appearance. The greatest public sensation of the time was 
made by the Ballet mécanique of George Antheil (1900-59), a former 
pupil of Ernest Bloch’s who had spent the years 1922-5 in Europe, where 
he had aroused interest by his compositions utilizing jazz elements. The 
Ballet mécanique, scored for ten pianos and an assortment of noise- 
producers not usually employed in orchestration, proved, when 
revived many years later in New York, a rather dated expression of 
‘futurism’. Nevertheless it was a scandal in its time and so made its 
contribution towards a public awareness of the changing climate. 

Carl Ruggles (1876-1971) is a minor composer of some interest who 
was associated with the ‘radicals’ of the 1920s. With Varése, Riegger, 
Cowell, and others, he was active in the Pan American Association of 
Composers. One can hardly place Ruggles in the mainstream of musical 
development in America, and it is questionable that his tiny output of 
work has exerted much influence. His half-dozen or so completed 
compositions have, however, had ardent admirers. The four major 
works of Ruggles were all written between 1921 and 1933; they include 
Angels (1921), originally for six trumpets, later re-written for four 
violins and three cellos, or four trumpets and three trombones; Men 
and Mountains (1924) for chamber orchestra; Portals (1926) for string 
ensemble or string orchestra; and ТЛе Sun Treader (1933) for large 
orchestra. All these works are characterized by extreme dissonance 
and concentration. The technique is linear and very free, though 
calculated with extreme care. The music is non-tonal, and gives an 
impression of great intensity. Ruggles was an original voice, though a 


HENRY COWELL 597 


small one. For many years his music was rarely heard, though often 
discussed, and it was only in the 1950s that it became generally known 
and admired. 

Works such as these caused the greatest amount of excitement, 
opposition, and noise. Many of them have been forgotten; the works of 
Varése (1883-1965), on the other hand, were understood by very 
few when they were first heard, and the importance and vitality of his 
work came to be appreciated only very much later. In every important 
respect, Varése anticipated the musique concréte of the 1950s, and 
showed а musicality and verve that make many of the newer ‘abstract’ 
composers seem inhibited by their own theories. Varèse was influenced 
by the famous Italian ‘Futurist’ manifesto of 1913, and his training in 
mathematics and engineering makes him seem a destined precursor of 
the mathematically serialized music, and the electronic experiments, of a 
later generation. His works of the 1920s lead directly to his Poéme 
Electronique of 1958, composed directly on electronic tape for the 
World's Fair of Brussels; they can be seen in retrospect as early examples 
of what Varése termed ‘organized sound’. Jonization, composed in 1931, 
is scored entirely for percussion: thirteen players, using thirty-seven 
instruments, including two sirens, with chimes, celesta and piano heard 
briefly towards the end of the work. It differs basically from the Poéme 
Electronique only in the means of sound production. Hyperprism (Ex. 
254), composed in 1924, also uses an extraordinary assortment of 
percussion, and a small orchestra composed entirely of wind instru- 
ments. All these works are original and striking; although stark, after 
many years they no longer impress as being brutal. The music of 
Varése seems to reflect the positive achievement of a mechanical and 
technological civilization, not its doubts, nor does it give that impression 
of retreating which is characteristic of so much twentieth-century art. 
Like Ives, Varése was at least a generation ahead of his audience. 


HENRY COWELL 


The early work and activity of Henry Cowell (1897-1965) placed him 
in the foreground of attention as a radical and controversial figure in 
the 1920s. During his very active career as composer, performer, author, 
and lecturer, Cowell was associated with almost every development 
in American music. In the 1920s he was a radical among radicals, pro- 
ducing works in a bitingly dissonant counterpoint and exploring new 
possibilities in rhythmic organization. At the same time he startled 
audiences with his novel piano pieces in which he used what he termed 
‘tone-clusters’ (massed chords of any size built on major and minor 
seconds) and also struck, plucked or brushed the strings directly, used 


598 AMERICAN MUSIC: 1918-1960 


Ex. 254 
Moderato poco Allegro 
Flûte === ILE 
Petite Flüte gue... 
() = SS eee el 


e 
: Hip тсе irme ERE 
y molto 
E 
2 [а = 
DS pos e га 
ыы e 
ЕЕ a A ш | 
3 ЕЕ 
7 SSS aa 47 
І Hipp E EE LLL 
Тәшен T cep EDEN 
I-A Гиан саас) 
as Aft ж 
E = m—— о 
sourd. V s ите es З = 
те (SS ee 
Bass RES 
Moderato poco Allegro md 


Snare drum A 5 
тр<ұ/- тр 
T sourdement Т) Г) гара 


Indian drum д oe 7 S ws 
mom Букш к= 
Bass drum Z 2 » x 
Tambourine 4 AER 
ж 
Crash Cymbal 4 
sas ш long Laissez vibrer 
hf N et s’éteindre 
2 Cymbals = 2 
ВЕ |е LV. et s'éteindre 
A 


Tamtam A = 


Anvil | 
Slap Stick 2 


А ii 
high ye 989 
cE 


ow 
Lion Roar д =- - 
BENED Lens — | 


Siren 


Flite 
Petite Flite 


Clarinet en тї} 


1 
Cors en fa 
2 
Cors en fa 


3 


1 
Trompettes en ut 
2 


Tenor 
Trombone 
Bass 


Snare drum 


Indian drum 


Bass drum 
Tambourine 


Crash Cymbal 


2 Cymbals 


Tamtam 


Triangle 

Anvil 

Slap Stick 
high 

2 Chinese blocks 
low 

Lion Roar 
Rattle 

Big Rattle 

Sleigh Bells 


Siren 


HENRY COWELL 


T 
м l al 
4 


subito molto crescendo 


prm 
eS 2 phu ae 


[S 


(1) sur rebord D 1 n 


(2) membrane 


frottées l'une 7 == 
ee 
ЯШ 
a Ще mm ет 
mf 


599 


600 AMERICAN MUSIC: 1918-1960 


harmonics, and employed the instrument in other unconventional ways. 
Tiger (Ex. 255) illustrates Cowell's piano technique admirably, and the 
bars of 5 against 4 show another of the composer's musical pre- 
occupations: 

Ex. 255 x 


Allegro feroce 


HENRY COWELL 601 


In 1927, Cowell founded the New Music Society in San Francisco, for 
the promotion of avant-garde music, and edited the New Music 
Quarterly, which published works of Ives and other American and 
European composers at a time when no other outlet was available to 
them. Cowell devoted a great deal of his inexhaustible energy to the 
cause of contemporary music itself, and did much to make American 
music aware of its own strength. 

Cowell had an extraordinary childhood, and was largely self-taught 
until he attracted the attention of Charles Seeger (b. 1886), a remarkable 
composer and musicologist who taught at the University of California. 
Cowell later went to Germany where he studied comparative musicology 
under Hornbostel and attended Schoenberg’s lectures. Webern per- 
formed Cowell’s Sinfonietta in his concerts in Vienna, one of the first 
American works of the period to be heard in Europe. Cowell in turn 
helped introduce music of Schoenberg and Webern, as well as Bartok, 
to avant-garde audiences in America. His interest in oriental music, 
stimulated by his study with Hornbostel, led him to further explora- 
tions, and to an attempt at the reconciliation of music of the East and 
West which became a life-long pre-occupation. Cowell was essentially 
an eclectic composer, taking material where and how he found it, and 
using it without inhibition in compositions of all types. His work 
demonstrates his extraordinary interest in all musical phenomena, of 
any time and place, and the amount of music he left is as huge as it is 
varied. 

As early as 1914, Cowell had hit on the idea of the serialization of 
rhythm, and had experimented with aleatory music, and with novel 
instruments, including an electronic sound-producer invented by Lev 
Theremin (b. 1896). With Joseph Schillinger he had experimented with 
a ‘Rhythmicon’ as an aid in construction with cross-rhythms of great 
complexity. In later years, Cowell explored American popular 
hymnology, folk music of all types, and ‘exotic’ material of widely 
varying kinds. There is, over the years, no single Cowell style, and this 
has puzzled and annoyed critics and historians who feel a compul- 
sion to classify. Copland once said of Cowell that he was more of an 
inventor than a composer, but this is a judgement that time may not 


602 AMERICAN MUSIC: 1918-1960 


bear out. Cowell composed with natural enthusiasm and facility and 
much of his work is both original and well-realized. His later works, 
including most of his nineteen symphonies, are on the whole more 
conservative than the earlier ones, yet the entire body of work has a 
kind of consistency that a highly personal view of phenomena can 
bring to a variety of materials and idioms. Cowell's experiments, and 
his musical interests and aims, have had considerable influence both 
direct and indirect on many of his contemporaries and juniors, notably 
John Cage and Lou Harrison. 


VIRGIL THOMSON AND OTHERS 

Virgil Thomson (b. 1896) went from Harvard to Paris, where he 
studied with Nadia Boulanger and where he remained almost un- 
interruptedly until 1940. Thomson found the attitudes and aesthetic of 
Satie and ‘Les Six’ sympathetic, and became the outstanding American 
exponent of their spirit. But Thomson was an artistic personality of 
great originality and sophistication in his own right, and the deceptive 
simplicity of his usually diatonic and triadic idiom was as representative 
of a new spirit in music as the complex constructions of many of his 
contemporaries. Thomson abandoned the neo-classic style of his 
Sonata da Chiesa (1926), in which he introduced a tongue-in-cheek 
tango, and among his early works of distinct individuality his Symphony 
on a Hymn Tune (1928) stands as one of the first after Ives to utilize 
this type of American material. The work that established Thomson's 
reputation was the opera Four Saints in Three Acts, composed to a 
libretto of Gertrude Stein (1928). Its production in America in 1934 
was a memorable event. Thomson's second opera with Miss Stein, The 
Mother of Us All (1947), remains, with the earlier work, a classic of the 
American theatre. Both works are inimitably original; the music is 
witty, imaginative, subtle, and daring in its innocence (which is only 
apparent) It makes delicate allusions to a variety of sources, from 
modal melodies to folk hymns and ballads and the mock-heroic style, 
but in a way that no one has done before or since. The delicious pseudo- 
simplicity characteristic of Thomson's music is illustrated in this 
quotation from his Four Saints in Three Acts: 


603 


VIRGIL THOMSON AND OTHERS 


di- nar- y 


Ex. 256 


| Е. ІШ 5 ШШШ A 
ERU E FS 
| 2 и 9 “ГЕ QB | 
Е 8 Wy S du 
„с а E А 
4 | 5 | я || " 
5 5 e | Æ 
© lh © № о 
S Des XD YS | 
Р | 44 Па. «ГА -2 
„а HT. ВИЕ а 
Е [|| E | 5 
a a Е иң 
e es © "T 
| gi om 
ла o a Ф 2 
E || ва 
Р ша : 
а d n d | ci 
3 ie “А.а ә а 7 
: а |45 
с. 9 = ә 44 
са S S 
1 | Б | Е Е 
> ы ы ы 
П t Г Е Г il =£ Е „в 
ы ы A P т TF АН мыш ш e 
8 8 а Lg LE LS SI 
П 1 її | at] A Il 
a) ос) о в LE c ч 
© ^s] > z М а 
! П So ee ш = - К. А 
ы = 2 < 5 E < 2 АДА 
E 2 2 е ges иа ( Ф "m | 
E Bie а HST Pat a a 
c ОЕ € E [| 
Во NW IB "et D d E ТУІ TINI 
ld UN | КЫ 
CX 45 AD) AO A 
isa Въ: ГЕЙ [NT a 


1918-1960 


AMERICAN MUSIC: 


604 


or-di-nar-y set-ting which is as 


Sa 
Е-е 
ting which is as 


=: 
ai 
На 

-y Set- 


= 
H = 
1- паг 


= 
ен ай 
ма“ | 
or - di- 


aa 
de—————— — — —31 
Soon 


or-di-nar-y set-ting which is as 


soon 


and noon. 


soon which is as soon 


soon which is as soon 


and noon. 


and noon. 


soon which is assoon 


VIRGIL THOMSON AND OTHERS 605 


These same qualities are apparent in much of Thomson's work in other 
media, and are especially notable in his scores for the motion pictures 
The Plow That Broke the Plains (1936), The River (1937) and Louisiana 
Story (1948). Thomson has occasionally written in a more dissonant 
chromatic style, and has even explored aspects of twelve-note technique; 
but the essential quality of his writing remains a subtle simplicity that 
can take materials ranging from the commonplace to the recondite and 
transform them into a sophisticated personal idiom. Thomson's hand- 
ling of prosody, as demonstrated in his operas, his miscellaneous vocal 
music and his Mass (1959), is masterly. From 1940 to 1954 Thomson 
served as music critic of the New York Herald-Tribune, his work mark- 
ing a high point in American musical journalism. 

Thomson is one of the true ‘originals’ in American music, or, for 
that matter, in international music of the twentieth century. He does 
not appear to be interested in striking out on new paths, or in the 
complications of technique that characterize most contemporary music. 
Yet it is obvious that his path is a new one, and that he is well aware 
of every technical procedure that is current. The music he writes is the 
product of a serious and refined intelligence brought to bear on an 
immediate problem or need for making music. He is not afraid of being 
banal, or of not being 'significant'. One of the mechanisms that makes 
his music effective is his manner of making alarming juxtapositions; 
one could say that he makes Paris and Kansas City realize what they 
have in common, by the fact of his having reconciled them. This is 
beyond mere wit and cleverness. Thomson is one of the few contempor- 
ary artists in any field with a true sense of high comedy. Lesser artists 
use parody, satire, farce, or irony; but the appearance of comedy is 
indeed rare. 

While jazz and the more interesting varieties of novel or experimental 
music were attracting by far the greatest amount of attention, a number 
of American composers were quietly and solidly writing music that 
made its way more slowly. Walter Piston (b. 1894), Quincy Porter 
(1897-1966), Randall Thompson (b. 1899), Bernard Rogers (1893-1968), 
Wallingford Riegger (1885-1961), Douglas Moore (1893-1969), Leo 
Sowerby (1895-1968), Roger Sessions (b. 1896), and Otto Luening (b. 
1900), were names not on everyone's tongues before 1930. No one of 
them burst on the scene with the impact of Copland or Gershwin, or 
aroused the kind of violence that met Varèse or Antheil or Cowell. Nor 
were they acclaimed as extravagantly as Roy Harris was to be only a 
few years later. The music of these composers represents a wide varia- 
tion in style and temper, from the moderately academic conservatism of 


Sowerby, Thompson, and Porter to the more ruggedly dissonant and 
40 


606 AMERICAN MUSIC: 1918-1960 


involved style of Sessions and the more.or less atonal style of Riegger. 
Of the composers born before 1901, these were the ones whose work 
remained of greatest interest at mid-century. There were of course many 
others whose work attracted attention, and the names of Arthur 
Shepherd (1880-1958), John J. Becker (1886-1961) and Frederick Jacobi 
(1891—1953) deserve mention. 

Thompson, Moore, Rogers, Antheil, Porter, and Sessions all were 
at one time pupils of Ernest Bloch (1880-1959), whose importance in 
American music requires no further emphasis. Bloch himself wrote 
some of his most important works in the United States, including the 
Suite for Viola and Piano (1919), the Piano Quintet (1924), Ше “еріс 
rhapsody’ America (1925), the fine neo-classic Concerto Grosso (1925), 
said to have been written as a model for his pupils, the Sacred Service 
(1933), perhaps his most significant work, and the Violin Concerto 
(1937). With Bloch one is faced squarely with the problem of defining 
just who is an American composer, for although Bloch spent more than 
half his life in the United States, it is difficult to think of this remark- 
able man as anything but a musician belonging to a world in which 
nationality is meaningless. Bloch can perhaps best be classed with 
Dvorak, Schoenberg, Milhaud, Hindemith, Křenek, Stravinsky, 
Bartok, Toch, and Wolpe, as visitors from another world who lived in 
America for shorter or longer periods (many of them becoming 
American citizens), and who had great influence as teachers and com- 
posers on several generations of American musicians. . 

Bloch’s influence is not easy to discern in most of the composers who 
studied under him, but there is an almost unanimous testimony as to its 
importance. Bloch should be ranked with Boulanger (whose students 
included Copland, Thomson, Piston, Harris, Carter, and a long list of 
younger composers) as one of the two musicians who did most to shape 
American music in the first half of the twentieth century. And many of 
their pupils, notably Sessions, Porter, Rogers, Piston, Moore, Copland, 
Thomson, became in turn important teachers of a succeeding genera- 
tion. 

Douglas Moore, after his studies with both Bloch and Boulanger, 
joined the faculty of Columbia University in 1926, where he eventually 
succeeded to the Professorship endowed in memory of Edward Mac- 
Dowell. His music is conservative and showed from the start a rejection 
of both contemporary French and German styles. Moore developed a 
simple personal idiom, the originality of which was not perceived at 
once; it was based on a deep understanding of American vernacular 
music and an ability to translate this into an uncomplicated musical 
language that was both sentimental and sophisticated. Moore firmly 


INTERNATIONAL STYLES 607 


believed in musical Americanism, in much the same sense as Ives, but 
he wrote without Ives's complexities. Moore's was a traditional 
Americanism, not a cosmopolitan one. His first orchestral work to 
attract attention was The Pageant of P. T. Barnum (1924). This was 
followed by Moby Dick (1928), Overture on an American Tune (1931) 
and other works clearly reflecting a commitment to American themes. 
With The Devil and Daniel Webster (1938), a one-act opera based on a 
story of Stephen Vincent Benét, Moore showed that vocal writing and 
the musical theatre were his natural media. This unpretentious but 
original work enjoyed great success and has held the stage as a minor 
classic in American music. Moore's greatest accomplishment, and the 
work that finally brought general recognition of his unique talent, was 
the opera The Ballad of Baby Doe (1956), based on a true story of the 
American West, with libretto by John LaTouche. With this work, late 
in life, Moore achieved an American opera with greater success than 
any of his contemporaries or juniors had hitherto succeeded in producing. 


INTERNATIONAL STYLES— WALLINGFORD RIEGGER, 
ROGER SESSIONS, WALTER PISTON 


Piston, Sessions, and Riegger are, each in his way, composers in 
international styles. Each has gained in reputation since first attracting 
notice in the 1920s. By the middle of the century no critic or historian 
could have done other than place them as among the most interesting 
and significant American composers of their time. Riegger, the oldest 
of the three, was the last to gain recognition. His career was curiously 
chequered. He began as a thorough conservative, a student of 
Goetschius in New York and later of Bruch and of the American Edgar 
Stillman Kelley (1857-1944) in Germany. After winning several prizes 
for his early works, Riegger decided to rest and to reconsider for a 
period of three years. At the end of this time, he produced his Study in 
Sonority (1927) for ten violins, a work of extremely dissonant tension, 
using two mixed and extremely acid harmonic complexes as arbitrary 
tonic and dominant, and, as the title suggests, exploring string timbres 
in an unusual way. In his Dichotomy (1932) for chamber orchestra 
Riegger utilized two ‘note-rows’, and the music remains consistently 
within the rather general definition of 'atonality'. This was at a time 
when Riegger was entirely unfamiliar with the work or theories of 
Schoenberg. Riegger's style progressed from the orthodox to the radical 
as he grew older, although in general he adhered to classic forms, which 
he handled with the assurance of an expert craftsman. Yet Riegger 
remained for many years a rather neglected composer, and despite a 
volume of interesting work (much of it composed for dance companies) 


AMERICAN MUSIC: 1918-1960 
Sy Le 88—96) 


608 
Ex. 257 


THAI БҮТҮҮ 
| 


Бозда! 
ННІ 11 
ІШІ 


ИІ 

[M 

" | 

ІҢ || RA BRI 
ШШ ИМ 

i ую ЕНІ? ӨШІН, 


All Horn parts should be doubled throughout 


Е | |. p ШІ! 
sil il г с | i {|| <: B^ jin em ПЫ Ч 
M | Ч A zii : 7 al | ДІ d A 
к AE et С° SNR е к 6 3i пт RNL юн 
| zm ER o зп EE ВЕ 


н HA 
[d 


VII 
VIII 


dg ш isdumi], Я ut ион зподшот 1, 


60 


3 P qx. ND \ 
| 


К & (у 
aid. 


мъ 
4 | | | 


| а 
ІМ | 
= |. 

1 1, | 


ПЫ үй (3 и ( 
SENS 
[| A. Ц 
| ПОШ 
М Ch | а ~. [Ц . a 
ШЕ; Етер Ші An dy Gun A 
А 5 


INTERNATIONAL STYLES 


poco ritenuto 
poco ritenuto 


I 
I 
VII 


— 
= 
= 
> 


Тиба 


j ur ион auoquiory, 


610 AMERICAN MUSIC: 1918-1960 


he was relatively unknown to the American public until his Third 
Symphony (1948). His use of twelve-note techniques is demonstrated іп 
this work, and also in his First String Quartet (1938). Riegger employed 
row construction in a personal and free manner, and combined it with 
traditional structure. He was not a purist, but a composer who 
took materials as he needed them, and employed them with assurance. 
His Second String Quartet (1948) and his Music for Brass Choir (1949) 
show other aspects of his highly personal and sharply dissonant style, 
his strong rhythmic drive, and brilliant sense of instrumental sonority. 
Music for Brass Choir (Ex. 257) has been described as “һе apotheosis of 
the tone cluster'. The effect of these massed sounds on the brass instru- 
ments, and the imaginative handling of line and mass throughout this 
work are indeed original and impressive. 

The career and reputation of Riegger illuminate one aspect of 
American musical life by underlining the importance of prizes, festivals, 
and energetic seeking of public notice. Riegger had been known to 
musicians for many years as a composer of considerable importance, 
and had, in fact, been an active associate of Ives, Varése, Ruggles, and 
Cowell in the Pan American Association of Composers during the 
1920s. But public recognition came to him only after his Third Sym- 
phony, written when he was 62 years of age, had received an award 
from New York's music critics. Fortunately, the impression made by 
this work (which was, incidentally, the first major commission Riegger 
had ever received) resulted in performances of Riegger's other music 
and brought him further commissions for new works. For the last 
decade of his life, Riegger enjoyed some of the recognition and in- 
fluence he clearly deserved, and was at last able to hear his music 
performed by major orchestras and other performing groups. His 
works after 1948 included his Fourth Symphony (1957), much chamber 
and choral music, and two fine sets of variations, for Piano and 
Orchestra (1953) and for Violin and Orchestra (1959). Like Schoenberg, 
Riegger occasionally returned to tonal writing throughout his career, as 
in the Canon and Fugue for Strings (1941) and the Suite for Younger 
Orchestras (1954). 

Basically, Riegger's music, despite its wealth of invention and the 
depth of its technical vocabulary, is uncomplicated. Riegger strove for 
clarity and logic, and felt that the enlargement of the tonal vocabulary 
in the twentieth century was not a licence to greater freedom for the 
composer, but on the contrary imposed on him an ever greater need for 
control and discipline. Riegger's influence on younger composers 
became more evident during the later years of his career, as the original 
nature of his contribution became more generally appreciated. 


INTERNATIONAL STYLES 611 


Roger Sessions, like Riegger, is a composer of marked individuality 
who has also been influenced by twelve-note methods without becoming 
a doctrinaire adherent of the school. Again like Riegger, his musical 
thought gravitated towards serial techniques at a middle stage of his 
career as a composer, although in the case of Sessions this evolution 
was gradual. Sessions's early works, of which the most representative 
is the suite extracted from his music for Andreyev's play The Black 
Maskers (1923), show an inclination towards the richness and com- 
plexity of Central European styles, and reflect the influences of Bloch, 
Mahler, and Strauss. The complexity and richness of texture remain 
through most of Sessions's later work, which from the beginning has 
had a strong appeal to musicians but has never commanded a wide 
public following. The tonal works of Sessions's early period, including 
the First Symphony (1927) and First Piano Sonata (1930), show a 
solidity of technique and sense of musical logic that are very striking, 
and in their straining at the bounds of tonality foreshadow later 
developments in Sessions's music. The key work and turning point in 
Sessions's composition is the Second Symphony (1946), in which the 
impulsion towards a completely chromaticized style reaches the thre- 
shold of atonality. It is after this work that Sessions began to adopt 
twelve-note serial techniques, using them never in the sense of a system, 
but, again like Riegger, deriving thematic material and relationships 
freely and flexibly. Sessions's music is much more inward and intense 
than Riegger's, and is essentially more involved and elusive. At the 
same time, it often has a brooding and disquieting lyricism, perhaps 
at its most evocative in the Second String Quartet (1951). Sessions's 
chamber music is always impressive and satisfying. The Quintet for 
Strings (1958) shows his style at its least opaque and most accessible, 
and also illustrates his very personal use of twelve-note themes. 

In addition to the Third and Fourth Symphonies (1957 and 1958), 
Sessions's major production includes a Violin Concerto (1935), a Piano 
Concerto (1956), a Mass (1958), and a setting of an Idyll of Theocritus 
for soprano and orchestra (1954). Sessions is not a prolific or casual 
composer. Ап intellectual and spiritual affinity to Schoenberg is 
evident in all of his work, although in none of it is there any rigid 
application of Schoenberg's theory or practice; the affinity is entirely 
one of attitude and mentality. The essentially inward temperament 
perhaps is the reason that Sessions seems less happy and successful in 
works for the stage, such as the one-act opera, The Trial of Lucullus 
(1947), after Brecht. 

Sessions was at mid-century a figure of the greatest importance on 
the American musical scene, not only as a composer highly respected 


AMERICAN MUSIC: 1918-1960 


612 


Ex. 258 


ull 
k 
mí 
| 


Violin ПІ 


Ri ; 
Ts 


ТП 
МИ 
є 
Hil 
h 


Viola I 


| 


| т» 
EI 


s м 

Í кезегі 
СЕЕ РЕНН EET 
ІГЕ а 


(uil 


613 


INTERNATIONAL STYLES 


poco cresc. 


Bad 


Ly 
IG 


д 


A 


| 


poco cresc. 


Q 


< 


poco cresc. 


\ 
= 


© 
| 


poco cresc. 


614 AMERICAN MUSIC: 1918-1960 


by all, but also as a teacher and as a direct and indirect influence on an 
entire generation of younger composers. Many of his students, including 
Ross Lee Finney, Hugo Weisgall, Leon Kirchner, Milton Babbitt, and 
indeed a host of others, have achieved prominence not only as com- 
posers, but also, in turn, as teachers. 

Piston, too, has had great influence as a teacher. While Sessionstaught 
at California and at Princeton, Piston was at Harvard from 1926 to 
1961. Both Piston and Sessions have written books on harmony that 
have been widely used in American schools and conservatories. But 
while Sessions's music reminds one in texture and aesthetic of Central 
Europe, Piston's music is that of a neo-classicist. His early works, such 
as his First String Quartet (1933), impressed his contemporaries by their 
elegance, control, and craftsmanship. This impression was confirmed 
by the composer's later work, all of which exhibited a polish and mastery 
of medium that was as refreshing as it was rare. Piston's symphonies 
(his Seventh was completed in 1961) constitute the central portion of 
his work, although he has written in all conventional forms of chamber 
music as well. 

Unlike Sessions or Riegger, Piston has had a fairly wide public 
success. His music is accessible and fluent, in an idiom that is recogniz- 
ably of the twentieth century, but it is not ‘advanced’ or radical or 
apparently complicated by theoretical problems or considerations. It is 
essentially related to early and middle Stravinsky, with occasional over- 
tones of Hindemithian counterpoint. Basically it is conservative and 
even academic, and, in an entirely non-pejorative sense, it may be said 
that Piston represents the legitimate continuation of the ‘Boston School’ 
of Paine, Chadwick, and Parker. 


ROY HARRIS 


If Piston's music is characterized by its sureness and craftsmanship, 
the opposite must be said of the music of Roy Harris (b. 1898). Harris's 
arrival on the American musical scene can only be described as explo- 
sive; as with so many explosions, however, the aftermath is less exciting 
to describe. Harris was for some time a major force in American music, 
and suffered the unfortunate fate of becoming almost a legend at a time 
when American music was searching for something resembling a hero 
sprung directly from the soil. Harris did emerge from the West, appro- 
priately enough from a log cabin; he did not decide to be a composer 
until he was twenty-four, when he studied briefly with Farwell, who 
launched him with considerable effectiveness. An Andante for Orchestra 
was performed by Howard Hanson in 1926, and gained Harris the 
immediate interest of his colleagues. He then went to Paris, on Copland's 


ROY HARRIS 615 


advice, to study with Boulanger; and his Sextet for Clarinet, Piano, and 
Strings (1927) and his first Piano Sonata (1928) are among the best works 
of this period. Indeed they must still be counted among the best Harris 
has written. His First Symphony (1933) aroused the enthusiasm of 
Kussevitsky, the Russian-born conductor of the Boston Symphony 
who almost alone among the conductors of major orchestras en- 
couraged and performed the larger works of American composers a 

this time. Harris's Third Symphony (1939) (see Ex. 259) seems in retro- 
spect to mark the high point of his achievement. Hailed immediately 
as an American masterpiece, it exerted a powerful influence on musical 
thought for many years, and must be counted among the half-dozen 
most important American works of the century. Its interest and appeal 
are difficult to analyse, yet its intensity and novelty made themselves felt 
immediately among musicians and laymen alike. The Symphony 
appeared to be an expression of all that was vast, aspiring and nostalgic 
in American life; it was all things to all people, yet it appealed as being 
ресшапу and unmistakably American in each of its aspects. The 
material ranges from the home-spun (the hymn-tune, as in Ives, Cop- 
land and Thomson) to the epic-heroic: chorale-like brass passages and 
fanfares. Harris's style, as exhibited in this Third Symphony, found 
immediate imitators, and its technical and emotional mannerisms have 
had many echoes in American music. 

Harris's music usually has a long line, more often modal or diatonic 
than chromatic. He is fond of canon and fugue (or of canonic and fugal 
devices); his phrase-structure and rhythm are often irregular and 
shifting. He is fond of triads in non-functional relationship, as sonorities 
in their own right, but often uses a polyharmonic texture, producing an 
overlay of dissonance in an essentially simple tonal context. Cyclical 
structure occurs in many of his larger works. Harris reasons and writes 
about his music, but it remains the music of an instinctive rather than 
of an intellectually disciplined composer, and a lack of a self-critical 
faculty is often evident. In later works Harris proclaimed his mission as 
an American composer through a more obvious and naive use of folk 
and patriotic materials and motives. 

As early as 1941, Aaron Copland (like Cowell and Thomson a most 
perceptive critic of new music) pointed out the qualities and deficiencies 
of Harris's music in a summation that has not since been surpassed for 
acuteness and balance. Copland, after noting the sweep and breadth, 
the power and emotional depth, of Harris's best pages, and noting too 
the indefinably ‘American’ quality of his tunes and rhythms, pointed 
out that Harris often seemed not to know what to make of his materials. 
Formal deficiencies in the music, Copland pointed out, are so obvious 


1960 


1918- 


AMERICAN MUSIC: 


d = 120-128mm (ветрге tempo giusto) 


616 


2 


| І \ lley n | Та i 

| | in i 
ИАН | | 

ІК! ‚ | | ІШ! | ІШ | IC MIRI < Hi 

| е 1$ Т | | | 

| Др || | 


4 


22 
22 
22 


zz 


Б 2 PET) 
irr sellis 
ЕЕЕ е == 


ее] 

я 
: ЕЕ 
ЕЕ ДЕН 


unis. 
=> 


x em КИ б ls 
| үз Us Ie S 
Co cuc съ NE ar м С RG МЈ 
ПЕР” ЧЇЙ ПЕЙО TN ni hid 
ее ро ИД a МАЕ == 
о ЕН > - : ҥн = = 
E c. E E g E ЛЕ, сыра "E Е Е 
B © ds = à B momo b. = ен p > 


Ex. 259 


17 


6 


ROY HARRIS 


Q | | 
| C > 52 
[К H 


me 


| 


ШІ 


| | 

| $ Ad 

T 
ЦИ 


618 AMERICAN MUSIC: 1918-1960 


as to make it evident that Harris’s ‘insecure critical faculty’ handicapped 
him seriously. Many of his pieces lack a sense of direction, and also fail 
to correspond to the copious ‘spiritual’ explanations that Harris so often 
provided. Yet, in the end, Copland concluded that ‘Whatever one may 
think, it is useless to wish Harris otherwise than he is. One may show 
how much better his work might have been . .-. But there is no gain- 
saying that, such as it is, with all its faults and qualities, it is enormously 
important to us in the immediate scene. Plenty of Americans have 
learned how to compose properly, and it has done us little good. Here 
is a man who, perhaps, may not be said to compose properly but who 
will do us lots of good. We can let posterity concern itself with the 
eternal aspect of Harris’s music, if any. The important thing is that it 
has something for us here and now.'! 

Harris's influence on the music of the late 1930s and 1940s has not 
been, on the whole, justly estimated. Its positive qualities are best seen 
in the work of his pupil, William Schuman (b. 1910), and in aspects 
of the work of such composers as Vincent Persichetti (b. 1915) and 
Peter Mennin (b. 1923). But its weaknesses, too, have been reflected 
in a great deal of pseudo-Americana and nationalistic bombast, and 
in a tendency towards the formation of a right-wing school of com- 
position based on nostalgia. With the emergence of Harris as a 
major influence on American music, one is aware that the mood of a 
period has changed, and that ‘modernism’ is no longer a primary 
concern. А music for the people became the preoccupation of many 
composers; Harris himself felt strongly the need to communicate on a 
wide scale, and his Fourth Symphony (1941), based on folk-songs, and 
his Sixth (1944), based on Lincoln's Gettysburg Address, were evidence 
not only of what Harris believed, but also of what the results might be 
in terms of musical content and style. It was at about this time that 
Copland first produced his folk-style ballet scores, in a much more 
transparent and easily assimilable manner than his early works, and 
that Cowell abandoned radical experiment for an innocently folklorish 
manner. The urge towards creating a people's music was felt also in the 
work of a consciously ‘proletarian’ school, of which the most important 
productions were the stage works of Marc Blitzstein (1905-64). Among 
these, The Cradle Will Rock (1936) made a great impression by its 
directness and force, and helped to create a public for a native opera not 
based on the conventions of the Italian or German grand style. 

Many factors, both musical and social, contributed to the establish- 
ment of new tendencies in American music during the 1930s. The 
financial crisis of 1929 and the depression that followed brought a 

1 Ош New Music, p. 175. 


THE 1930s—THE IMPACT OF EUROPEAN IMMIGRANTS 619 


number of changes to American life and thinking, and these affected 
the American artist directly and indirectly. The relief programmes of 
public works instituted during the Roosevelt administration included 
projects designed to provide both opportunity and sustenance to 
writers, painters and composers. The Works Progress Administration 
(WPA), under which musicians were given employment, instituted 
concerts, ‘forums’ and a variety of projects including the collection of 
folk-songs. The WPA orchestras provided more opportunities than had 
previously been available for the performance of new American works, 
and were in general a vitally stimulating influence. Was it a coincidence 
that the number of composers seemed to increase vastly? Certainly the 
WPA provided an encouragement that has since come to be expected 
by American artists, and which has, in fact, been perpetuated through 
the extraordinary abundance since World War II of grants, fellowships, 
awards, and commissions, which appear to fall to the talented and the 
untalented in about equal proportions. During the period of the depres- 
sion, American composers acquired a new sense of their position and 
function in society, and became increasingly aware of the possibility 
of effective collective action not only towards economic improvement, 
but also towards the establishment of an awareness of composers as a 
body politic. 


THE 19305--ТНЕ IMPACT OF EUROPEAN IMMIGRANTS 


A further major influence on the development of music in America 
was the arrival as permanent residents in the United States of some of 
the most eminent composers of Europe. Arnold Schoenberg came to 
the United States in 1933, and after a brief residence in Boston moved 
to California, where he remained until his death in 1951. Ernst Toch 
arrived in 1934, Kurt Weill in 1935, Hindemith in 1937, Kfenek and 
Wolpe in 1938, Stravinsky, Milhaud, and Vittorio Rieti in 1940, 
Martinů and Bartok in 1941. All of these composers, except Stravinsky 
and Weill, made their presences felt in the most direct manner, by 
teaching; but it is evident that the immediate impact of their personalities 
and the performances of their music were also of great importance. 
Nearly all of these distinguished emigrés became members of the music 
faculties of American colleges and universities: Schoenberg at the 
University of Southern California, Krenek at Vassarand Hamline(St. Paul, 
Minnesota), Hindemith at Yale, Milhaud at Mills, Toch also at the 
University of Southern California, Martini at Princeton, Bartok at 
Columbia. The importance and prestige of the teaching of composition 
at American universities, begun by Paine and Parker, was greatly 
emphasized by the presence of these acknowledged masters. Even 


620 AMERICAN MUSIC: 1918-1960 


Stravinsky, who never wished to be officially a teacher of composi- 
tion, gave the series of lectures at Harvard University later published 
as Poétique musicale, a document of major interest in twentieth-century 
aesthetic. 

Of the very great number of composers active in the United States in 
the 1930s, a few began to emerge as well-defined musical personalities 
towards the end of the decade. They represent, for the most part, a 
generation trained in America and directly profiting by the work, both 
musical and propagandistic, of their immediate elders. Among the 
composers achieving prominence around 1940, and maintaining their 
positions during the following twenty years, were Samuel Barber (b. 
1910), Ross Lee Finney (b. 1906), Paul Creston (b. 1906), William 
Schuman, Norman Dello Joio (b. 1913), Morton Gould (b. 1913), 
David Diamond (b. 1915), Vincent Persichetti and Gian Carlo Menotti 
(b. Italy, 1911). Others, equally prominent in the years following World 
War II, had not yet attracted comparable attention. 

The music of these men reflects the customary American diversity of 
style, temperament, and training, but on the whole it is consistently 
conservative and eclectic. It reflects an acceptance and mastery of the 
less adventurous techniques of twentieth-century European masters, but 
with the exception of Finney's later music (after 1950) demonstrates 
no influence of the Schoenbergian disciplines and little desire to enlarge 
musical boundaries. On the other hand, it contains ample evidence of 
the already strong influences of Copland and Harris, added to those of 
the neo-classic Stravinsky, of Hindemith, and to a much smaller extent, 
of Prokofiev. The musical idioms range from the rather sober elegance 
of Barber, through the boisterous vehemence of Schuman, to the 
popular-music style of Gould and the theatricality of Menotti. 

Barber provides an excellent illustration of the internationally- 
oriented traditionalist in American music. His music was readily 
accepted by conservative audiences, and he was the first American 
composer to be performed by Arturo Toscanini (Essay for Orchestra, 
1937, and Adagio for Strings, 1936). Barber's music has often been de- 
scribed as ‘neo-romantic’, a vague word used to suggest the fact that the 
music flows smoothly, is not excessively dissonant, is traditional in form, 
and gives the impression of both warmth and refinement. Barber has, 
however, written in the ‘neo-classic’ style (Capricorn Concerto, 1944) 
and there are elements of greater dissonance and freer concepts of 
tonality in his Second Essay for Orchestra (1942), the Medea Suite 
(1947), and the Piano Sonata (1949). Barber's work after 1950 shows a 
tendency to become more complex, but without losing its mastery of 
line, form, and mood. His opera Vanessa (1958), with libretto by 


THE 1930s— THE IMPACT OF EUROPEAN IMMIGRANTS 621 


Menotti, and his Antony and Cleopatra (1966), are among the few by 
American composers to have been produced by the Metropolitan Opera 
in New York. 

Menotti himself has proved to be much more a composer (and 
author) for the theatre than most of his American-born contemporaries, 
and his success in an idiom remarkably his own has helped to interest 
the American public in new works for the operatic stage. He first 
attracted widespread attention with his one-act opera buffa, Amelia 
Goes to the Ball (1937), produced at the Metropolitan in 1938. Menotti 
came to the United States in 1928 and was trained, like Barber, at the 
Curtis Institute in Philadelphia. Although fundamentally Italianate as 
a composer, Menotti is identified with the American musical scene, and 
has been without question an important factor in the development of 
an American musical theatre. Тле Medium (1946) and The Consul (1950), 
both serious, if rather melodramatic, works, enjoyed unprecedented 
success in the commercial theatre, and have been performed frequently 
throughout the United States and Europe. The chamber opera Amal 
and the Night Visitors (1951) was commissioned by the National 
Broadcasting Company (a private enterprise) especially for television, 
and has been presented annually on Christmas Eve. The unqualified 
success of Menotti's work with even the very broadest public has been 
valuable to other American musicians, creating an awareness of opera 
as a still effective art form and as a potential medium for communicating 
with audiences larger than those available for symphonic music. The 
appeal of Menotti's work, continued with The Saint of Bleecker Street 
(1954), lies in his unerring dramatic flair and directness. The musical 
material is conventional, but treated imaginatively, and the vocal line 
is always handled with mastery. 

William Schuman and Norman dello Joio were among the best known 
and most widely performed American composers during the 1940s and 
1950s. Dello Joio's music poses no problems; it is expertly made and 
quite conservative in style. Dello Joio's principal teacher was Hinde- 
mith, and his music reflects the influence of that master. If dello Joio's 
music is less angular and has more obvious warmth than his master's, that 
is perhaps a reflection of an Italian heritage. Dello Joio has frequently 
used modal melodic material, including Gregorian chant, in a mildly 
dissonant texture. His use of contrapuntal techniques is skilful. Dello 
Joio is an accomplished pianist and organist and is one of the few 
American composers (Persichetti is another) who has written interesting 
music for the keyboard. Among dello Joio's orchestral works, his 
Variations, Chaconne and Finale (1948) has been often performed by 
American orchestras. 

41 


622 AMERICAN MUSIC: 1918-1960 


The chief distinguishing character of Schuman's music is the quality 
of its motor energy. It derives both in spirit and in technique from 
Harris, but goes beyond Harris in achievement. Like Harris, Schuman 
is often concerned with being an American composer; he is, however, 
essentially an urban American, and his music seems a reflection of the 
drive, the buoyancy and the tension of American life. The orchestral 
piece that first brought him notice was an American Festival Overture 
(1939), a brilliantly orchestrated work of great liveliness. His Third 
Symphony (1941) established him securely among American composers 
of his generation, and his work has continued to be performed more and 
more widely. Schuman's work in the 1940s and 1950s includes seven 
symphonies, two concertos, various other orchestral pieces, four string 
quartets, and the music for several dance productions, notably Under- 
tow (1945), Night Journey (1947) and Judith (1949). Schuman has 
writtens ongs, piano music, and choral works as well, and one opera, 
The Mighty Casey (1953), based on a celebrated mock-epic of base- 
ball. With many other American composers (Cowell, Luening and 
Finney among them) Schuman rediscovered the work of one of 
America's authentic ‘primitives’, the New England composer William 
Billings (1746-1800), whose name is associated with sturdy hymns, 
patriotic anthems and what he himself termed 'fuguing tunes'. Schuman 
based his New England Tryptych (1956) on music by Billings, and this 
example of musical Americana has been one of his most popular works. 
Schuman's music is basically tonal, although at times extremely dis- 
sonant. He is fond of major-minor triads and is not afraid of harsh 
conflicts. Like Harris, he often employs a long melodic line and makes 
frequent use of canonic and fugal techniques. The flow of Schuman's 
music is often punctuated by brittle rhythmic passages. It is the drive 
and restlessness of the rhythms, and the brilliance of the orchestration, 
that first attract attention to Schuman's music. A passage from his 
Sixth Symphony (Ex. 260) shows the long string line opposed to the 
noisy rhythmic punctuation of the entire brass section. 


THE SITUATION IN 1950—THE DECADE 1950—60 


At the mid-point of the century the American musical scene was one 
of tremendous activity. Among the ‘established’ composers, the most 
important appeared to be Copland, Barber, Piston, Schuman, Dello 
Joio, Thomson, Sessions and Riegger; and Ives’s reputation was rising. 
But new developments were under way, and other young and middle- 
aged composers were coming to the fore. Almost all the composers 
named above, with the exceptions of Ives and Sessions and possibly of 


THE SITUATION IN 1950—THE DECADE 1950-60 623 


и -НЕ а: 
TTR: 
Ob. 1) | Hy 8 — 2 tr 


Е 
E 
5 
| 
| | 


9 


me 
N 
=a 


p 
5 
ч 
p 
щъ. 
> 


(qi 
ГІШ н 
| 
$ үф | 


о 
I 
| 6 
| у 
у 


= 
— > > 
0 =-= -= у =a = —— =, I, = = 
Е: A = =! (Saas Ses 
(Орк тека а та ppp есе ааа на E 


Me 3 D 4 a 


BS or «таза еее eee ee epe = E prm mc 


0 3 3 


Trpt. in C II 


ш 


Larghissimo J = circa56 molto sostenuto ed intenso Y 
== - 5 5 и ЕЕ 
т! 2 e "= E 


ЕР === 


> 

Fan = | 

Bass E Le TE 2 
= E F 


624 AMERICAN MUSIC: 1918-1960 


() 


IW | ЕЕЕ 


В 
EN eee ст Lo 


Trpt. in СП 


ш 


"Tuba ЕЗ 


THE SITUATION IN 1950—THE DECADE 1950-60 625 


Riegger, were well received in polite musical circles and regarded as 
conservative by a younger generation. Most of the newer composers 
were interested neither in simplicity nor in Americanism. The influence 
of Schoenberg and Webern was beginning for the first time to be 
strongly felt and reflected by American composers, and Stravinsky's 
conversion to serialism prompted a further movement towards the 
international avant-garde style. Prominent among the composers work- 
ing in serial techniques were Milton Babbitt (b. 1916), Ben Weber 
(b. 1916), George Perle (b. 1915), and George Rochberg (b. 1918). 
Many others followed, and the influence of the post-Webernists con- 
tinued to become more marked. Copland and Thomson themselves 
experimented with note-row construction, and a number of more or less 
neo-classical composers, such as Arthur Berger (b. 1912), followed the 
new Stravinsky line of the 1950s. At the same time, new composers also 
emerged on the right wing and in a middle group. Some of the most 
interesting figures, among them Elliott Carter (b. 1908), Hugo Weisgall 
(b. Bohemia, 1912) and Leon Kirchner (b. 1919) could be described 
as left of centre without being committed to serialism or to any other 
easily identifiable school. The original and controversial work of John 
Cage (b. 1912) also attracted attention during the fifties, and the equally 
original work of Harry Partch (b. 1901) began to interest the musically 
curious, without, however, reaching a large public or exerting much 
influence on musical thought or practice. 

The range of interest and of musical style continued to widen during 
the decade 1950-60, so that composers of one extreme tendency seemed 
hardly to exist in the same world as those at the other extreme. This was 
merely a reflection in America of the international situation in music; 
but the important difference for American music lay in the fact that it 
had finally reached a point of development at which it was abreast of 
European currents, instead of being anywhere from ten to fifty years 
behind them. Increased facility of communication was of course a 
primary factor, but it was the achievement of several generations of 
American composers that provided the technical and artistic basis for 
this new relation to the musical art of the world. In the 1950s, the work 
of new European composers such as Boulez and Stockhausen made 
an immediate impact in America, and for the first time a few American 
composers, notably Carter, Cage, and Babbitt, produced music other 
than jazz that was of some influence on musicians of Europe. American 
music had finally become part of the international scene rather than a 
provincial echo. 

It is obviously impossible, in a brief essay, to mention all of the many 
dozens of composers whose works commanded some attention during 


626 AMERICAN MUSIC: 1918-1960 


the decade 1950-1960, nor is it possible to pretend to any degree of 
historical perspective in evaluating an enormous volume and variety of 
production that is still so recent. But among the many composers 
occupying the broad middle ground of musical style at mid-century, one 
should cite some of those who attracted attention and achieved some 
prominence. In such a list one would find the names of Peter Mennin 
and Vincent Persichetti (already mentioned), as well as those of Lukas 
Foss (b. Germany, 1922), Leonard Bernstein (b. 1918), Robert Palmer 
(b. 1915), Ingolf Dahl (b. Germany, 1912-1970), Irving Fine (1914— 
1962), Alexei Haieff (b. Russia, 1914), Elie Siegmeister (b. 1909), Robert 
Ward (b. 1917), William Bergsma (b. 1921), Andrew Imbrie (b. 1921), 
and Alan Hovhaness (b. 1911). To cite these names and to omit others 
is not to make either a critical or an historical judgement. Nor should 
it be assumed that these composers constitute a group in any sense; they 
represent a variety of personalities and musical orientations, and have 
in common only the fact that all were active and respected in their time 
and place, and that their music was performed and known. At so short 
a remove in time, one cannot know if a Charles Ives remains un- 
discovered, or if an Edward MacDowell enjoys a fame beyond his 
deserts. 

Of these composers, brief accounts must suffice. Mennin's style, best 
seen in his symphonies, is broad and singing, rhythmically alive, and 
shows an impressive command of large contrapuntal forms. Ward, 
like Mennin a student of Hanson and Bernard Rogers, has written in 
an easy and recognizably American style in his overture, Jubilation 
(1946), and has progressed to an effective dramatic music in his very 
successful opera Тйе Crucible (1960). Palmer and Dahl write for the 
most part in a vein of Hindemithian neo-classicism, while Fine and 
Haieff represent a neo-classicism deriving from Stravinsky. Bergsma's 
music is generally classical and conservative, but has great sensitivity 
and elegance. Imbrie, a pupil of Sessions, has some of his teacher's 
intensity and drive, and his music represents the Sessions influence 
at its best. Persichetti's style is one in which influences of Copland, 
Harris and Schuman seem to be fused. Foss continued to evolve 
an expression of his own after arriving in the United States at the 
age of fifteen. Influenced strongly by Hindemith, later by Stravinsky and 
eventually by Copland, Foss later turned to serialism in his Time Cycle 
(1960) for soprano and orchestra, and to experiment with improvisatory 
techniques. Hovhaness, of Armenian descent, has attempted in his 
prolific output of composition to reconcile music of the East and West, 
and has succeeded in producing work of quite distinct profile. Bernstein, 
like Foss a gifted pianist and conductor, is a Protean composer, whose 


THE SITUATION IN 1950—THE DECADE 1950-60 627 


serious symphonic works derive from Stravinsky and Copland (with 
occasional overtones of German nineteenth-century composers), but 
whose most convincing accomplishments so far are in the popular 
musical theatre, as West Side Story (1957). 

The techniques and mannerisms of electronic music, aleatory and 
improvisational music, totally serialized music and other manifestations 
of contemporary musical thought or exploration commanded attention 
in the United States as they did in Europe. Among the first to experiment 
with music composed on electronic tape were Otto Luening and his 
colleague at Columbia University, Vladimir Ussachevsky (b. Man- 
churia, 1911). A first concert of their compositions for tape recorder was 
given in New York in 1952. A grant from the Rockefeller Foundation in 
1959 established a laboratory for electronic music at Columbia, where 
further experiment is carried on under the direction of Luening and 
Ussachevsky, with Milton Babbitt and Roger Sessions of Princeton. 
Sessions himself has not composed electronic music; Babbitt, on the 
other hand, has been among the most active in this field, and his work, 
both in theory and in application, has been extremely influential. 
Coincidental with the rise of electronic music, as a phenomenon of 
major interest, was the reappearance, after a silence of many years, of 
Varése. Varése had in many ways anticipated the aesthetic of elec- 
tronic music, and the evolution of its techniques, by a singular poetic 
justice, re-emphasized the daring and originality of the music he 
had written in the 1920s. Deserts (1954), his first new work in almost 
twenty years, was composed for conventional instruments (including 
much percussion) with interludes of ‘organized sound’ recorded on 
tape. With this remarkable piece, Varése once again proved himself to 
be a composer of imagination, power, and vitality. 

Among the independent explorers of new musical styles, none was 
more prominent during the fifties than John Cage. Cage first attracted 
attention with his music for ‘prepared’ piano in the 1940s, taking up a 
line of development first indicated by Cowell. Cage has been influenced 
also by Varése, by oriental music, by Anton von Webern, and by a 
variety of philosophical or speculative ideas. He was perhaps the first 
to use chance as a basic element in musical construction; and his work 
antedates by some years the aleatory music fashionable in Europe late 
in the fifties. His /maginary Landscape, for twelve radios, was performed 
in 1951 at what proved to be the final concert of the New Music Society. 
This work, the principle of which is that it can never repeat itself, 
depends entirely on the chance of what will be on the air at the time of 
performance. ‘Form’ is indicated, but the essence of ‘content’ is absolute 
randomness. The spirit of Cage’s work is reminiscent of Parisian Dada; 


628 AMERICAN MUSIC: 1918-1960 


if its absolute value is difficult to estimate, its influence on many of 
Cage's contemporaries cannot be denied. 

Among other composers of an experimental and non-conformist 
tendency, Lou Harrison (b. 1917), a pupil of Cowell, and Henry Brant 
(b. Canada, 1913) deserve mention. Harrison, in addition to some highly 
sensitive, though eclectic, works for conventional instruments and 
voices, has composed a number of works for percussion orchestra, 
including such unorthodox sound-producers as brake-drums, iron pipes, 
and packing boxes, and these are often of a surprising gentleness and 
astonishing effectiveness. Brant, whose teachers included Antheil, 
Copland, and Riegger, among others, but who was also very strongly 
influenced by Ives and Cowell, was a juvenile prodigy whose early 
works showed a bewildering exuberance and cleverness. He has worked 
as a professional arranger for radio and other commercial media, and 
after repudiating many of his early compositions, has turned his 
attention to experiments in multi-planed music, using separated and 
independent groups of players. The groups are often of unusual con- 
stitution. Typical of Brant's work is his Grand Universal Circus (1956), a 
theatre-piece in three ‘acts’ which embodies Brant's ideas of stereo- 
phonic distribution and polyphonic tempi. Of Brant's music, Henry 
Cowell commented that ‘it constitutes audaciousness іп the grand 
manner . . . and [is] deserving of a place in Ives' idea of a Universal 
Symphony, to which all composers with doughty ears and strong- 
sounding music are invited by Ives to contribute. . . ^ 

Considerably less controversial, and with no elements either of Dada 
or of ‘scientific’ objectivity or abstraction, is the music of Carter, 
Kirchner and Weisgall, who achieved recognition in the fifties as 
composers of impressive stature. Kirchner studied with both Schoen- 
berg and Sessions, and his music has a complexity and intensity 
reminiscent of these composers. But Kirchner's music also shows 
affinities with the music of Berg and Bartók. Kirchner first attracted 
attention with a rhapsodic Duo for Violin and Piano (1947) and fol- 
lowed this with a Piano Sonata (1948) and a String Quartet (1949): 


629 


THE SITUATION IN 1950—THE DECADE 1950-60 


Ex. 261 


Violin I 


Violin II 


ЕЕ 


= 


dinario 


п 


юг 


(они оС 


col legno, 


col legno (battuto) 


lacur d 


p 


ще 


630 AMERICAN MUSIC: 1918-1960 


His Concerto for Piano (1953) is representative of his mature work in 
large forms. There is great force, even violence, in Kirchner's music, 
which at all times remains highly personal and indifferent to theoretical 
positions. Kirchner succeeded Piston as professor of composition at 
Harvard, after many years at Mills College in California. 

Hugo Weisgall also studied with Sessions, and writes in a highly dis- 
sonant atonal style that recalls the expressionist manner of Central 
Europe. Weisgall's forte is vocal music, and his setting of English is 
extraordinarily effective. His principal work is in the field of opera; he is 
almost the only American composer of the fifties to write opera in a 
dissonant, powerfully dramatic style. Ех. 262 from his Athaliah illus- 
trates also the restless semiquaver notation characteristic of Weisgall's 
music. The Tenor (1950) and The Stronger (1952), based on Wedekind 
and Strindberg respectively, made Weisgall known as among the most 
interesting new American composers for the serious musical theatre. 
Six Characters in Search of an Author (1956), after the Pirandello play, 
and Purgatory (1958), after Yeats, established the composer firmly as 
one of the most forceful on the American scene. 

Elliott Carter was among the most respected composers in America 
at the end of the decade. Carter studied with Piston and Boulanger, and 
his early compositions were of a neo-classical trend, with echoes of 
Stravinsky and Copland. Not until 1948 did Carter find an expression 
that was entirely his own, and one that he has developed with increasing 
originality and authority with each succeeding work. In the Sonata for 
Cello and Piano of 1948, he joined the ranks of Ше ‘experimental’ 
composers, but his distinction in this category is that such experiment 


631 


THE SITUATION IN 1950—THE DECADE 1950-60 


Ex. 262 


| 


end of Рат: Опе 


| | 


632 AMERICAN MUSIC: 1918-1960 


as the work represented was subordinate to a remarkable musical 
realization. In this work Carter for Ше first time utilized his invented 
principle of ‘metrical modulation’, a technique for controlling and 
changing the absolute speeds of musical time units. This alone would 
demand attention, but the Sonata also exhibited an expressive power of 
the highest degree and a command of spacing and sonority already fore- 
shadowed in the Piano Sonata of 1946. Carter's work during the follow- 
ing years, the Eight Etudes and a Fantasy for woodwind quartet (1950), 
the String Quartet (1951), the Variations for Orchestra (1955) and the 
Second String Quartet (1960), all impressed as being of major impor- 
tance. Carter's works may be described as intellectual machines in 
which an unusually forceful musical impulse supplies the momentum. 
His ideas in the fields of form and organization, his successful creation 
of a music evolving on several simultaneous planes, had already 
exercised wide influence by 1960. In some respects Carter solved some 
of the problems in polyrhythm and multi-planed music that Ives had 
posed many years before. Ex. 263, from the 1951 String Quartet, is 
typical of his music in this respect: 


Ex. 263 
Уют 42-60 


espr. ——— 
Violin II (d = 135 E 


FAC r exo ЕЕ 


m Й Г 
тг hp rt eee ree По Se 
| E Ду p | Еш Гес 1 I/—& T - «e ТГ rnm 
1 ҮТ Б Дас — — zn Coy SY, ЛИНЕН, 2880] (Фан —- foi 
|]: ———— ЫШЫ E 


633 


THE SITUATION IN 1950—THE DECADE 1950-60 


| 


^ 


| 
| 


Ба 


H р 
pa me ы a рт qma miae 


634 AMERICAN MUSIC: 1918-1960 


Carter's music is always rigidly disciplined, but it is without formula. 
Its texture is generally dense and its colour dark. The calculation of 
sonorities and balances is extremely delicate, and each new work has 
seemed to indicate further progress into still unexplored areas of musical 
thought. 

It is impossible to foresee how the music of the 1950s will appear in 
even the very near future. Judgement becomes increasingly difficult not 
only because of the great diversity of styles and techniques—and the 
consequent possibility that the mid-century on the whole represents a 
point of transition and synthesis—but also because of the enormous 
numbers of composers at work, and the unprecedented rapidity with 
which musical fashions continue to evolve, and in many cases to dis- 
appear. АП these phenomena are characteristic of the twentieth century 
everywhere, and in every field, but they seem to take on peculiarly 
exaggerated forms in America. One is safe only in assuming that 
American music at the beginning of the second half of the century 
possessed both the environment and the tradition to keep abreast of the 
remainder of the world. 


635 


(ii) Music in Latin America 


By GERARD BÉHAGUE 


NATIONALISM 

DURING the first half of the twentieth century Latin-American art 
music was largely dominated by the nationalist trend, with notable 
exceptions. Composers sought in the various folk-music traditions of 
their countries the substance of their works. The large majority of the 
considerable music production for all media at this time reveals varying 
degrees of national concern, from the direct use of folk and popular 
sources to a more subjective assimilation of folk material. Through their 
nationalistic works Latin-American composers were able to win un- 
precedented international recognition, and only in the late 1950s and 
the 1960s did musical nationalism suffer an obvious decline. 

The major exponents of Latin-American art music during the period 
under consideration include Heitor Villa-Lobos, Carlos Chávez, Domin- 
go Santa Cruz, and Alberto Ginastera, respectively from Brazil, Mexico, 
Chile, and Argentina. They and their contemporaries, however, were not 
exclusively nationalist. Frequently they have attempted to incorporate 
twentieth-century European styles and techniques into certain national 
idioms. In many cases they have also followed such trends as impres- 
sionism, neo-classicism or serialism, in which no trace of nationality can 
be detected. 

In Brazil, Heitor Villa-Lobos (1887-1959) dominated the local scene 
throughout his career. Extremely prolific and imaginative, he wrote 
about a thousand works (including various arrangements of many 
pieces) in all possible genres and media. By 1913 he had written some 
fifty-five compositions, among which the Suite dos cántigos sertanejos 
(Folk-song suite) (1910) indicates his first elaboration of thematic 
material derived from folk sources. Further involvement with typical 
national subjects is evidenced in his tone poems and ballets of the late 
1910s, such as (Лғариғи and Amazonas, introducing concurrently 
characteristic harmonic and rhythmic elements of popular music as 
well as children's song-tunes in his piano pieces, such as А prole do bébé 
(Baby's family) по. / (1918) and Lenda do caboclo (Tale of a peasant) 
(1920). During the 1920s Villa-Lobos travelled to Paris where he 
succeeded in establishing himself as a composer and conductor. This 
period had a particular significance for him because he then completed 
the series of the СЙдғов, considered together with the Bachianas 
brasileiras his best contribution to modern music. Inspired by the native 


636 MUSIC IN LATIN AMERICA 


background of the chéros (popular strolling ensembles of serenaders іп 
Rio de Janeiro at the turn of the century), he wrote sixteen compositions 
bearing this title. Not in chronological order, they are intended for the 
most varied media, from solo guitar (Chóros no. 1) to full orchestra with 
mixed chorus (Chóros no. 10). The only common stylistic traits result 
from a highly subjective recreation of various popular and primitive 
musical traditions. Chóros по. 10 quoting the popular song ‘Rasga o 
согасйо” reveals the assimilation of advanced techniques: predominance 
of rhythm and percussion instruments, polytonality, and atonality with 
tone-clusters. The 1920s were also the years of his piano masterpieces 
including the Cirandas (Rounds) (1926), sixteen pieces of high virtuosity 
based on children's songs, and Rudepoema (Rude poem) (1921-26), 
dedicated to Arthur Rubinstein. The Bachianas brasileiras, the last seven 
symphonies, and the last thirteen quartets dominated Villa-Lobos's 
production during the last twenty-seven years of his life. According to 
the composer himself the Bachianas were inspired by the atmosphere 
of J. S. Bach's work, considered by him as a universal source of music. 
Consisting of nine pieces they are written as dance suites preceded 
generally by a prelude and ending with a fugue-like or toccata-like 
movement. The use of baroque composition processes is in fact small, 
although fugue as a formal principle could be construed as a 'neo- 
baroque' device, demanding a clarity of horizontal movement and 
the presence of systematic imitation. Ostinato figures and long pedal 
notes also indicate neo-classic devices. This series was meant as a free 
adaptation of such *baroque' devices applied to Brazilian folk music. 

Among Villa-Lobos's Brazilian contemporaries, Oscar Lorenzo Fer- 
nández, Luciano Gallet, and Francisco Mignone represent the folk- 
music orientation of their time. The best known composers of the next 
generation include Camargo Guarnieri, Luiz Cosme, Radamés Gnatalli, 
and José Siqueira. 

The 1910 revolution had an extraordinary impact on Mexico's artistic 
life. As a result of patriotic fervour musicians adhered to musical 
nationalism whose sources of expression were sought in either Indian or 
mestizo cultures. Manuel M. Ponce (1882-1948), considered the pioneer 
of nationalism in Mexico, drew on all types of mestizo folk music 
incorporated into a neo-Romantic style. The post-revolutionary period 
saw the emergence of the so-called Aztec Renaissance and the con- 
sequent indianista movement in the arts. Carlos Chávez (b. 1899), 
the most influential Mexican composer of the twentieth century has 
been particularly successful in evoking subjectively the remote past, 
character, and cultural setting of the pre-conquest Indian. His works of 
clear Indian inspiration comprise the ballet Los cuatro soles (The Four 


NATIONALISM 637 


Ages), Sinfonia India, Xochipilli-Macuilxéchitl (An Imagined Aztec 
Music’), a Piano Concerto, and the well known Toccata for percussion, 
rearranged as a ballet and titled Тохса!. In his most abstract composi- 
tions such as his Sinfonia de Antígona, La hija de Cólquide (The daughter 
of Colchis), his Violin Concerto, and numerous piano pieces, Chávez's 
highly personal style and Mexican sense appear so intimately connected 
that his music has been characterized as ‘profoundly non-European’. 
As а conductor he has also had a brilliant career. He founded the 
Orquesta Sinfónica de México in 1928 and directed it for over eighteen 
years. Chávez's contemporary, Silvestre Revueltas (1899—1940), became 
internationally known especially through his works Осһо por Radio and 
Sensemayd. 

Outstanding Cuban composers associated with nationalism were 
Amadeo Roldán (1900-39) and Alejandro García Caturla (1906-40) 
who found in Afro-Cuban music the most suitable source of national 
expression. José Ardévol (b. 1911) assumed a position of leadership in 
Cuba as a composer and teacher from the 1930s to the middle 1950s, 
founding in Havana the Grupo Renovación Musical (1943) which pro- 
moted contemporary music and rejected nationalism. 

Musical nationalism in Chile is represented mainly by Pedro Hum- 
berto Allende (1885-1959), who utilized native elements within an 
impressionist style, Carlos Lavín (1883-1961), and Próspero Bisquertt 
(1881-1959). The best known Chilean composers, however, are Domin- 
go Santa Cruz (b. 1899) and Juan Orrego-Salas (b. 1919), both of neo- 
classic rather than nationalist tendencies. Santa Cruz has laid the 
foundation of professional musical life in his country through the 
promotion of music education and the creation of music and concert 
societies. His large production as a composer includes mostly abstract 
works for symphony orchestra, chamber and choral works. Orrego- 
Salas's output comprises a large number of chamber works, such as his 
Sonata a quattro, Op. 55 (1964), three symphonies, ballets, choral pieces, 
and an opera. 

Nationalistic tendencies emerged in Argentina in the works of Alberto 
Williams (1862-1952), the most prolific and influential composer of his 
generation, and of Arturo Berutti (1862-1938), who treated national 
themes in his operas (Pampa, Yupanki). Alberto Ginastera (b. 1916), 
one of the leading creative personalities in contemporary Latin- 
American music, has evolved from a nationalistic orientation in the 
1930s and 1940s (in such works as Impresiones de la Puna, the ballets 
Panambi and Estancia, the series of Pampeanas) to a neo-classical current 
in the 1950s (a Piano Sonata, Variaciones concertantes, and others). In 
the 1960s he turned to an effective manipulation of atonal and serial 

42 


638 MUSIC IN LATIN AMERICA 


techniques, mixed with a meticulous preoccupation with timbres 
(Cantata para América Mágica, a Piano Concerto, a Violin Concerto, the 
operas Don Rodrigo, Bomarzo, and Beatrix Cenci). In some works, such 
as Estudios sinfónicos, Op. 35 (1967), he combined serial and microtonal 
textures with fixed and aleatory structures. 


MODERNISM AND THE AVANT-GARDE 


In opposition to theprevailing nationalist current, a number of Latin- 
American composers active in the 1940s and 1950s practised an abstract 
style through neo-classic, dodecaphonic, and post-Webernian serialist 
idioms. In Argentina Juan Carlos Paz (b. 1897), one of the founders of 
the Grupo Renovación and the Agrupación Nueva Musica, favoured the 
expressionistic aesthetic and became already in 1930 a strong supporter 
and follower of twelve-note techniques. In Chile Carlos Isamitt (b. 
1887), though interested in Indian materials did not neglect serial 
techniques; and Gustavo Becerra (b. 1925) cultivated expressionist 
atonality and serialism before engaging himself actively in experimental 
music. 

In Brazil Schoenberg's theories were first introduced by the German- 
born composer Hans-Joachim Koellreutter (b. 1915) and followed in 
the 1940s by such composers as Claudio Santoro (b. 1919) and César 
Guerra-Peixe (b. 1914). Among the younger generation Edino Krieger 
(b. 1928) has found some interesting compromises within a modernistic 
neo-classic style, as shown by his first string quartet of 1956. The 
Argentineans Roberto García Morillo (b. 1911) and Roberto Саатайо 
(b. 1925) represent the same tendency. The Colombian Luis Antonio 
Escobar (b. 1925) has alternated his style from a neo-classic trend to a 
post-Webern serialism. In Peru an internationalist current appears with 
composers such as Enrique Iturriaga (b. 1918), Celso Garrido Lecca 
(b. 1926), Enrique Pinilla (b. 1927), and others, while in Uruguay the 
most advanced techniques of composition have been used by León 
Biriotti (b. 1929) and Sergio Cervetti (b. 1940). 

Avant-garde tendencies have emerged especially in Argentina, Chile, 
Brazil, Uruguay, Peru, and Mexico. In spite of the limited means of the 
Latin-American scene for experiments in new music, numerous young 
composers (several of them active in Europe or the U.S.A.) have 
advocated and used electronic musical resources, aleatory techniques, 
and indeterminacy, thus breaking definitely with musical nationalism. 


Vill 


МБС АРЕНЕ SOVIEL UNION 


By GERALD ABRAHAM 


THE POLITICAL BACKGROUND 


FEW events in political history have produced such fundamental 
cultural consequences as the October 1917 Revolution in Russia. But 
the consequences followed only by very slow degrees, except in so far 
as individual composers were concerned. Stravinsky was abroad and 
decided to stay there; Rakhmaninov went into self-imposed exile in 
December and was followed the next year by Prokofyev (though not 
before he had conducted the first performance of his Classical Symphony 
in April 1918). These departures, like the deaths during 1914-18 of 
Lyadov, Skryabin, Taneyev and Cui, impoverished Russian musical 
life and seemed to draw a line under the end of an epoch; but a great 
deal in the new Russia was for some time very much like the old so far 
as music was concerned. Glazunov (1865-1936) remained at the head 
of the Petrograd Conservatoire, Mikhail Ippolitov-Ivanov (1859-1935) 
became president of the Society of Writers and Composers in 1922 and 
remained director of the Moscow Conservatoire until 1924, when he 
returned to Tiflis to organize musical life in the Georgian Republic; 
Reinhold Glier (1875-1956) was director of the Kiev Conservatoire 
until his transfer to Moscow in 1920. АП three were Rimsky-Korsakov 
pupils. A fourth, Nikolay Myaskovsky (1881-1950) was still in the 
Army and not demobilized until 1921. Rimsky-Korsakov’s son-in-law, 
Maximilian Steinberg (1883-1946) was the principal composition 
professor at Petrograd and became Director of the (by this time 
Leningrad) Conservatoire in 1934. From the very first the cultural 
policy of the new régime was directed by Lunacharsky, who was no 
enemy of ‘bourgeois culture’, and when the Commissariat of Education 
formed a music section in July 1918 its first head was Arthur Lourié 
(1892-1966), a disciple of Debussy, Skryabin, and Schoenberg. The pre- 
Revolutionary academics and the pre-Revolutionary avant-garde might 
have been supposed to have no enemies but each other. Even the 
earliest experiment in ‘proletarian culture’, the movement known as 
Proletkult (1918—23), largely directed to the training of workers and 


640 MUSIC IN THE SOVIET UNION 


peasants in writing and the practice of the arts in naive ‘leftist’ direc- 
tions, was tempered by the common sense of some of its leaders; these 
included Aleksandr Kastalsky (1856-1926), an authority on choral 
music and folk-song, whose choral folk-song arrangements were 
deservedly popular during the nineteen-twenties. So long as the Civil 
War and Polish War lasted, that is, until 1921; composers were more 
concerned with the production of music of this kind, and *mass-songs' 
for or about the Red Army or the workers, than with symphony or 
opera. The mass-songs of this period not only served as models for the 
later ones of Aleksandr Aleksandrov (1883-1946) (who in 1940 was to 
compose the ‘Hymn of the Soviet Union’) and other composers, but 
often left their broad, undistinguished, diatonic stamp on Soviet sym- 
phony and opera when these appeared. 

The end of the fighting brought a period of relative relaxation, stock- 
taking, and policy-formulating. Lenin had said that 


Art belongs to the people. It must penetrate with its deepest roots into the 
very thick of the broad working masses. It must be understandable by these 
masses and loved by them. It must unite the feeling, thought and will of these 
masses, inspire them. It must awaken in them artists and develop them.! 


These are admirable sentiments, which many eminent Russians, from 
Chernishevsky and Mussorgsky to Stasov and Tolstoy, would have 
endorsed; but, like so many utterances of politicians, they are capable 
of various interpretations. In one sense it was easy to see what should 
be done—and, so far as music was concerned, it was done with a 
thoroughness that put the rest of the world to cultural shame. Musical 
education in the widest sense was put within the reach of all and pro- 
fessional musical education raised to the highest imaginable level. 
“Тһе masses’ were brought into opera-house and concert-hall, and not 
only choirs and orchestras of popular instruments but symphony 
orchestras and string quartets were sent to factories and barracks. But 
the questions *What music is to be understood and loved by the masses? 
What should be the nature of the music of our Brave New World? 
remained unanswered. Lenin himself knew there were no quick 
answers, and in 1921 he gave a public warning that ‘cultural problems 
cannot be decided as quickly as political and military problems"? but 
there were two bodies who were confident they knew the answers: the 
[later: Rossiyskaya] Assotsiatsiya proletarskikh muzikantov (Association 
of Proletarian Musicians: known as APM or RAPM) founded in 1923, 
and the Assotsiatsiya sovremennoy muziki (Association for Con- 
temporary Music: ASM) founded in 1924, the year when Western 


1 О literature i iskusstve (Moscow, 1957), p. 583. 
2 Sochineniya, xxxiii (Moscow, 1950), p. 55. 


THE POLITICAL BACKGROUND 641 


musicians began to visit the Soviet Union and full cultural contacts 
with the outside world were made. APM was the successor to, and in 
some respects the negation of, Proletkult. Proletkult had believed in 
bringing the workman and the peasant to art, including contemporary 
art; RAPM not only detested contemporary music but denounced the 
classics as ‘bourgeois’ and had little use even for national folk-art; to 
them ‘Soviet music’ was music immediately comprehensible to the 
simple workman or peasant. ASM stood for the ‘contemporary’ in the 
international sense and became closely associated with the International 
Society for Contemporary Music. It held that ‘music is not ideology’ 
and defended the autonomy of the creative musician. Naturally each 
body suffered internal dissensions and secessions—ASM, for instance, 
had a more conservative wing in which Myaskovsky, Yury Shaporin 
(1887-1966), and Vissarion Shebalin (1902-63) were prominent—and 
some of the older composers (Glazunov, Glier, Ippolitov-Ivanov) and 
their followers kept aloof from both, preserving 'active neutrality'. But 
broadly speaking these two bodies polarized the main opposing 
tendencies throughout the nineteen-twenties and early nineteen-thirties. 
Thanks to the greater influence of ASM, particularly in Leningrad, 
Soviet Russia was able during 1925-8, the period of maximum artistic 
freedom, to hear such operas as Schreker's Der ferne Klang, Prokofyev's 
Love for Three Oranges, Kienek’s Der Sprung über den Schatten, 
Wozzeck, Le roi David, Mavra and Oedipus Rex, Jonny spielt auf and 
the Dreigroschenoper. ‘Persimfans, the conductorless Moscow 
orchestra which flourished for ten years or so from 1922, played not 
only the classics but Bartók, Skryabin, Stravinsky, Honegger, Ravel, 
Falla, and Prokofyev. 

The ‘modernist’ wing of ASM included Skryabin’s old champion, the 
critic Leonid Sabaneyev, and Lev Knipper (b. 1898), Nikolay Roslavets 
(1881-1944), Alexander Mosolov (b. 1900), Leonid Polovinkin (1894- 
1949), and Gavriil Popov (b. 1904) among the composers. The real 
founder of ASM, Boris Asafyev (‘Igor Glebov’) (1884—1949), the doyen 
of Soviet musical criticism, played an ambiguous róle: conservative and 
uninspired as a composer, he was at first the critical champion of 
advanced Western modernism but as early as 1924! he began to sound 
warnings about the dangers of subjective composition and holding 
aloof from the masses, and by the mid-nineteen-thirties he had become 
one of the severest critics of everything ASM had stood for. RAPM 
was at first insignificant both numerically and in the nature of its 
membership; its only distinguished original member, Kastalsky, died 


1 ‘Krizis lichnovo tvorchestva’, Sovremennaya muzika (1924) no. 4, p. 98, апа 
*Kompozitori, pospeshite', ibid., no. 6, p. 146. 


642 MUSIC IN THE SOVIET UNION 


іп 1926. It was only іп 1929, when it was reinforced by the Prokoll,! a 
group of young graduates from the Moscow Conservatoire—including 
Alexander Davidenko (1899-1934), Boris Shekhter (1900-61), Viktor 
Bely (b. 1904), Dmitry Kabalevsky (b. 1904), and Marian Koval 
(1907-1971)—that it began to carry much professional weight,? though 
its ideals naturally appealed to a much bigger public than those of 
ASM. Тһе battle was already going badly for the ‘contemporaries’; their 
organ Sovremennaya muzika, founded in 1924, ceased publication in 
1929, and in 1931—weakened Бу the secession of Myaskovsky and 
Shebalin—ASM finally collapsed. But the triumph of RAPM was short- 
lived; it was intolerant, dictatorial, and many of its members were 
technically incompetent. Under the ‘proletarian’ régime, standards at 
the Moscow Conservatoire fell so low that composition-students needed 
to offer only two or three mass-songs as their leaving exercise.? 

Parallel struggles had been going on in literature and the other arts 
and on 23 April 1932 the Central Committee of the All-Union Com- 
munist Party stepped in. In 1925, while expressing pious hopes for the 
future ‘hegemony’ of proletarian writers, it had resolved that ‘the Party 
cannot connect itself in any way with any tendency in the domain of 
literary form. ... A style corresponding to the epoch will be created but 
it will be created by other methods; the solution of the problem is not 
yet in sight. .. . The Party must declare for the free rivalry of different 
groups and tendencies. . . . The Party cannot allow a monopoly by any 
group whatever, even to that which is proletarian in ideology: this would 
lead in the first place to the ruin of proletarian literature’.* Now it took 
Soviet culture firmly in hand and put an end to both crude proletarian 
art and sophisticated contemporary ‘formalism’ (art for art’s sake). 
Before long Soviet artists and writers were given a phrase of Gorky’s as 
their slogan: ‘Socialist realism’. Soviet art must be understandable and 
loved by the masses, but it must be worthy of its ancestry in classic 
Russian and world art; and by its strength and optimism it must help to 
build socialism. The artistic debate was henceforth not between creeds 
but about the correct interpretation of the only true creed. RAPM was 
dissolved and the Union of Soviet Composers was established in order to 
safeguard ‘socialist realism’ in Soviet music. 


1 Portmanteau form of ‘Proizvodstvenny kollektiv studentov-kompozitorov Moskovskoy 
konservatorii’ (Production group of the student-composers of the Moscow Conservatoire), 
founded іп 1926: see Sergey Ryauzov, ‘Vospominaniya o “РгоКоПе” °’, Sovetskaya muzika, 
(1949), no. 7, p. 54. 

2 The RAPM programme at this moment of triumph is translated by Nicolas Slonimsky, 
Music since 1900 (3rd edition, New York, 1949), p. 655. 

3 See the collective work, Istoriya russkoy sovetskoy muziki, i (Moscow, 1956), р. 58. 

* Resolution of 18 March 1925. 


THE BEGINNINGS OF SOVIET OPERA AND BALLET 643 


THE BEGINNINGS OF SOVIET OPERA AND BALLET 

During its first decade Soviet opera! had practically only one theme, 
revolution, and its dramaturgy was hardly more sophisticated than that 
of the ‘Western’ film. One of the aberrations of the early nineteen- 
twenties was the performance of operatic classics with new, 'revolu- 
tionary' libretti; thus 7osca was produced at Leningrad in 1924 as 
Borba za Kommunu (The Fight for the Commune). Similarly Les 
Huguenots, which under Nicholas I had had to be disguised as J Guelfi 
e i Ghibellini, now appeared as Dekabristi (The Decembrists). The 
centenary in 1925 of the Decembrist rising, however, also suggested two 
new operas, Vasily Zolotarev's Dekabristi which was produced in that 
year and Shaporin's Pauline Goebel, of which only a few scenes were 
produced at the time. (It had to wait till 1938 before even the first 
version was nearly completed under the title Dekabristi.)? 1925 was 
the true birth-year of Soviet opera, for it saw not only the ‘Decembrist’ 
works but Andrey Pashchenko's Orliny bunt (The Eagles’ Revolt), 
Peter Triodin's Stepan Razin, and Za Krasny Petrograd (For Red 
Petrograd) by Arseny Gladkovsky and E. V. Prussak. These were 
not actually the first new operas after the Revolution; but Triodin's 
Knyaz Serebryany, based on А. К. Tolstoy's historical novel (1923), and 
Yurasovsky's Trilby (1924)? neither belonged to the new age nor 
possessed the breath of life. Pashchenko's opera on the Pugachev 
rising, Triodin's second opera, and the Gladkovsky-Prussak picture 
of the defenders of Petrograd against the White army under Yudenich 
also failed to survive, but the last has at least historic interest as the 
earliest opera on a Soviet theme. The score was mainly the work 
of Gladkovsky (1894—1945), a competent conventional composer, 
while his collaborator Prussak contributed some grotesque, satirical 
music for the Whites; their fighting heroine, Dasha, is characterized 
by music suggesting revolutionary workers songs and her com- 
rades by quotations from the Internationale and ‘Yablochko’ (The 
little apple), the enemy by ‘Vzveytes, sokoli, orlami’ (Soar, hawks, 
like eagles), a song popular in the White armies. However, after 
thirteen performances, the opera succumbed to protests from both 
modernists and Proletkult. Gladkovsky's later version of it, Front i til 
(Front and Rear) (1930) ‘does not contain one single Soviet man who 


is given individual characterization'.* Pashchenko (b. 1885) was more 


1 On early Soviet operas see the chapter by I. Rizhkin and S. Levit in /storiya russkoy 
sovetskoy muziki, i, p. 155, and M. Iordansky, P. Kozlov, and V. Taranushchenko, 
‘K probleme sovetskoy opere’, Sovetskaya muzika (1933), no. 1, p. 19. 

? See p. 688. 

3 In this chapter the dates of operas are those of the first production; for all other works 
the dates are those of composition. 

4 Rizhkin, op. cit., p. 187. 


644 MUSIC IN THE SOVIET UNION 


successful with his Pugachev opera,! not so much with the individual 
characters as with the broad folk-songish crowd-scenes; it was per- 
formed all over Russia and held the stage for ten years or so. His Civil 
War opera, Cherny yar (The Black Crag) (1931) was a failure. 
Zolotarev's Dekabristi also survived at least into the mid-nineteen- 
thirties. A minor pupil of Rimsky-Korsakov, Zolotarev (1873-1934) had 
a real, if unoriginal vein of lyrical-elegiac melody which enabled him to 
make the Decembrist poet Rileyev, the real hero of the opera, a live and 
sympathetic character. Inability to create character in musical terms 
was the general failing of the Soviet opera-composers of this period. 
Even when given strong literary characters, such as Ivan Shishov 
(1888-1947) found for his Tupeyny khudozhnik (The Toupee Artist) 
(1929) in Leskov's terrible story of the love of a serf-hairdresser for a 
serf-actress, they are apt to be musically lost against their social back- 
ground, though Shishov comes near to success in a quintet for the 
fugitive lovers Lyuba and Arkady, the wretched priest and his wife who 
have betrayed their hiding-place, and the enraged Count: 


Ex. 264 
LYUBA 


Akh! ti. . do-lyush-ka pod-ne-vol - па - ya, 
Же = 


ри ы ы ЗУ! О - slush-ni - ki! 
COUNT 
Pro -u - chu, 
ARKADY Go - re gor-ko - e. 


r 220 == 
E = = he =т= 


Priest Оу! О - slush-ni- ki! 


1 See Asafyev's article in Zhizn iskusstva, 1925, no. 46, reprinted in his Izbrannie trudi, v 
(Moscow, 1957), p. 113. 


THE BEGINNINGS OF SOVIET OPERA AND BALLET 645 


Go- - -re gor-ko-e. 


Lyuba: Oh, the lot of the 
un-free, bitter woe. 

Priest and Wife: Disobedi- 
ent! Rebels! 

Count: ГЇЇ teach them. 

Arkady: Bitter woe. Oh, the 
lot of the un-free.) 


- u-chu ya ikh 


do - lyush - ka 
rm ES 


Yet another work in a similar old-fashioned nationalist idiom and with 
a little life, at least in the heroine, was Proriv (The Break-Through, i.e. 
of Mamontov’s cavalry, raiding far in the rear of the Red Army in 1919) 
(1930) by Sergey Pototsky (b. 1883). Tupeyny khudozhnik at least 
brought a refreshing change from the themes of revolt and civil war. 
Aleksandr Krein (1883-1951) and Sergey Vasilenko (1872-1956) sought 
variety by transposing the themes into other lands—in Krein’s opera 
Zagmuk (1929), into a far distant age as well: the revolt of Babylonian 
slaves against their Assyrian oppressors. In Sin solntsa (Son of the Sun, 
also 1929) Vasilenko turned to the China of the Boxer Rising, with a 
tragic love-affair between a Boxer hero and Aurora Walter, daughter of 
the American General Hamilton; ten years later he produced a parallel 
piece, Buran, about the struggle of the Uzbeks with the ‘Tsarist 
colonizers’. 

The settings of Krein’s and Vasilenko’s operas also allowed alter- 
natives to Russian musical idioms; Vasilenko’s Boxers were pre- 
dominantly pentatonic. Throughout the nineteen-twenties the basic 
language of Soviet opera was not unnaturally that of the quarter- 
century before the Revolution, which in turn had been on the whole a 
period of conservatism and epigonism. The only new elements were the 
militant mass-songs, the ‘leftist’? pop-art modernism of Proletkult, and 
‘contemporary Western’ modernism; and these new elements were 
generally associated with Bolsheviks and their supporters. Opera- 
production tended to be ‘modernistic’, sometimes incongruously at 
variance with the music, particularly in Leningrad. The least con- 
ventional works, no doubt inspired conceptually if not musically by 
Stravinsky's Histoire d'un soldat, were Klimenty Korchmarev's Zvan- 
soldat (1927) and Pashchenko's Tsar Maximilian (1929), satirical essays 


646 MUSIC IN THE SOVIET UNION 


in musical folk-theatre, with clowning and popular song. The impact in 
1930 of Dmitry Shostakovich's Nos (The Nose) (after Gogol's story) 
and Knipper's Severny veter (North Wind), the one grotesquely comic 
and with no political content, the other tragic— based on the never 
forgotten or forgiven shooting of the Baku commissars—but both 
genuinely contemporary in musical idiom, must-have been severe. Their 
jagged, nearly atonal melodic lines, motor-rhythms and dissonant 
harmony come from the world of Hindemith and Prokofyev, Кїепек 
and Stravinsky's Mavra; even the diatonic banalities of The Nose are 
more likely to have their roots in Křenek than in Proletkult. Ex. 265 
shows the hotchpotch of styles in which the only common factor is 
anti-romanticism: (i) parody coloratura à [а Mavra and Hindemithian 
counterpoint; (ii) from the galop interlude preceding the solemn sounds 
of the scene in the Kazan Cathedral where the Nose sings (1). 


Ex. 265 (1) 
ба v 


(What do you mean? Explain.) 


THE BEGINNINGS OF SOVIET OPERA AND BALLET 647 
(ii) 


Molto allegro (d = 106) 


It must, however, be remembered that caricature by means of parody 
and the grotesque was by no means always inspired from the West; it 
was constantly employed—for instance, in Za Krasny Petrograd as we 
have seen—against the past and present enemies of the Revolution by 
composers who were anything but modernists. 

Soviet ballet can claim a longer history than Soviet opera, for 
Asafyev, who had composed four ballets in pre-Revolutionary days, 
wrote a Carmagnole during the first winter of the new régime and pro- 
duced it with piano—played by himself—in a Petrograd workers' club 
on the first anniversary of the October Revolution. After that, Asafyev 
contented himself for a long time with the compilation of ballet-scores 
from the music of better composers (Grieg, Tchaikovsky) and returned 
to original composition only in 1932 with Plamya Parizha (The Flames 
of Paris), after which came Bakhchisaraysky fontan (The Fountain of 
Bakhchisaray, after Pushkin) (1934)? and a series of works which all 
failed to achieve lasting success. Vasilenko and Korchmarev also tried 
their hands at ballet. Vasilenko's second effort, Josif prekrasny (based 
on the Old Testament story of Joseph), was the first new ballet to 


1 On the early Soviet ballets, see M. Rittikh in Istoriya, i, p. 202. 


2 Asafyev's own accounts of these works are reprinted in Jzbrannie trudi, v (Moscow, 
1957), pp. 138 and 141. 


648 MUSIC IN THE SOVIET UNION 


be staged at the Bolshoy in Moscow after the Revolution (March 
1925); and Korchmarev's Krepostnaya balerina (Тһе Serf Ballerina) 
(1927), another glorification of the Pugachev rising, won success 
through its beautiful, folk-songish score. But by far the most suc- 
cessful Soviet ballet of this period was Glier's Krasny mak (The Red 
Poppy) (1927; revised version, 1949), set in contemporary China, with 
wicked imperialists and reactionaries and splendid Soviet sailors; part 
of its success was no doubt due to the ballerina Ekaterina Geltser, who 
danced the Chinese heroine; but Glier's score included numbers such 
as the ‘Dance of the Soviet Sailors’ (variations on the popular song 
*Yablochko") which became popular on their own account. 


EARLY SOVIET ORCHESTRAL MUSIC! 


The earliest, and for thirty years the senior, Soviet composer of 
symphonies was Myaskovsky.? He may not cut a very impressive figure 
when seen in world-perspective but, compared with most of the com- 
posers whose operas and ballets have just been enumerated, he was 
outstanding: a master of his craft, fertile in invention, mildly original 
in thought, a genuine supporter of the Revolution, though far from 
being a revolutionary in music. His pre-war compositions had been 
subjective, rather pessimistic in tone, and couched in a late-romantic 
idiom influenced by middle-period Skryabin. Of the two symphonies, 
Nos. 4 and 5, which he wrote at Revel in the first half of 1918, No. 5 
in D—bright and idyllic, with humour and suggestions of folk-music— 
suggests a new departure and has been seen as 'the beginning of Soviet 
symphonism'; but the idea of “а quiet symphony (E, С, D?) in four 
movements; Andante mysterious, with chief theme of lullaby character 
had come to him in April 1914,3 The Sixth Symphony, completed in 
1923, was a different matter. There are still subjective elements; the 
two middle movements were written under the direct impression of the 
deaths of two persons very dear to the composer; the middle section of 
the second suggests a mysticism akin to Holst's: 


!See particularly ‘Glebov’, ‘Russkaya simfonicheskaya muzika za 10 let’, Muzika i 
revolyutsiya (1927), no. 11; Viktor Belyaev, ‘10 let russkoy simfonicheskoy muziki', 
Sovremennaya muzika (1927), no. 24. 

? The chief source of information about Myaskovsky is N. Y. Myaskovsky: stati, pisma, 
vospominaniya, edited by S. Shlifshteyn, two vols. (Moscow, 1959 and 1960). There is a 
good short survey of his work by Kabalevsky in Sovetskaya muzika (1951), no. 4, p. 18, 
reprinted in M. A. Grinberg (ed.), Sovetskaya simfonicheskaya muzika (Moscow, 1955), 
р. 36. On the first twelve symphonies see particularly Tamara Livanova in Istoriya, i. 
p. 233; on the later symphonies see ibid. ii, p. 384, iii, p. 359, and iv (2), p. 235. Studies in 
periodicals are very numerous. 

з Shlifshteyn, op. cit. ii, p. 394. 


EARLY SOVIET ORCHESTRAL MUSIC 649 
Ex. 266 


,Andante moderato (J. — 66) 
Vns. I, div. a3, con sord. (4- Celesta) 


and the main theme of the third: 


Ex. 267 
SE 3. 
() tm т — LR „шм E 
Е ЕЕ 
SS ш —— 
P dolce | 
() 
fet 43 S : 
П ат 


Strings. 
Lm 
м = г! М 
Ht = qe я 
Dit UE Eu Em 
Eee A ass 


(senza D.B.) 


650 MUSIC IN THE SOVIET UNION 


which returns at the end of the finale, is Skryabinesque. The finale, 
written under the influence of Verhaeren's play of the French Revolu- 
tion, Les Aubes, introduces “Са ira’ and ‘La Carmagnole’ (which 
Myaskovsky had once heard sung ‘аз the Paris workers sing them’), 
but the ‘Dies irae’ and a Russian folk-song, taken up near the end by a 
chorus, which sings of the soul being parted from the body and going 
before the judgement of God, seem to triumph over the revolutionary 
songs. On the other hand, the Symphony is undeniably ‘monumental’ 
in scale and conception; it contains a great deal of lyrical diatonic music; 
the end can be construed as an apotheosis of those who gave their lives 
for the Revolution. Modelled on Berlioz rather than Mahler, whose 
music Myaskovsky knew but considered ‘rather banal’,? it is the 
recognized forerunner of the various ‘monumental’ Soviet symphonies 
that have come since. Livanova compares it with Blok’s poem “The 
Twelve’ ‘not only in the analogous treatment of the theme, but for its 
place in the history of Soviet art’ ;? both are documents of a period when 
a portion of the intelligentsiya were still preoccupied with a ‘sacrificial’ 
conception of the Revolution. 

Myaskovsky’s later symphonies were not all so acceptable. No. 8 
(1925), based largely on appropriate folk-songs, reflects—like the con- 
temporary operas of Pashchenko and Triodin—the current interest in 
historical peasant-risings, in this case Stepan Razin’s. But the one- 
movement No. 10 (1927), inspired by the crazy hero of Pushkin’s 
‘Bronze Horseman’, was regarded as a deviation in the direction of 
‘false modernism’, and No. 11 (1932) was (as he admitted) ‘subjective’. 


1 Shlifshteyn, op. cit. і, р. 198, and ii, p. 15. The author of the thematic analysis of the 
Sixth Symphony in the ASM organ, Sovremennaya muzika (1924), no. 3, p. 86, failed to 
recognize either tune. See also M. Tarakanov, ‘Shestaya simfoniya N. Y. Myaskovskovo’, 
Sovetskaya muzika (1956), no. 7, p. 11. 

? Shlifshteyn, op. cit. i, p. 211. 3 Livanova, ор. cit., p. 245. 


EARLY SOVIET ORCHESTRAL MUSIC 651 


He made amends in the lyrical No. 12 (1932), conceived as a ‘Collective 
Farm' Symphony and dedicated “То the Fifteenth Anniversary of the 
October Revolution'. 

Throughout the nineteen-twenties Soviet orchestral music maintained 
a more ‘modernistic’ line than opera. The ‘proletarians’ were not 
interested in the symphony and the symphonists felt no strong com- 
pulsion to indulge in musical ideology. Beside Myaskovsky among the 
more conservative members of ASM stood his talented pupil Shebalin, 
who produced two very Myaskovskian symphonies (1925 and 1929)— 
Myaskovskian in their leaping, energetic allegro themes, their plastic, 
long-drawn lyrical ones, their chromatic harmony, and their slightly 
incongruous folk-elements—and a much older man, Aleksandr Gedike 
(1877-1957), organist and Bach scholar, whose Third Symphony (1922) 
and Concerto for organ and strings (1926) are solid, academic works. 
But there were orchestral essays in more modern idiom, such as 
Knipper's Skazki gipsovovo Buddi (Tales of the Plaster Buddha) (1924) 
and D minor Symphony (1929), Mikhail Gnesin's Simfonichesky 
monument: 1905-1917 (with chorus singing a poem by Esenin) (1925), 
Krein's First Symphony (1925) and Traurnaya oda (Funeral Ode—for 
Lenin) (1926), Polovinkin's Teleskopi I, II, and ИП (1926-8), and 
Shebalin's earlier symphonies (1926 and 1928). Modernistic in a more 
vulgar sense, betraying the influence of Proletkult—which gladly 
accepted *modernistic constructivism’, the ‘neo-classicism’ of the West, 
as an alternative to ‘individualistic subjectivism’ (bourgeous roman- 
ticism)—was Mosolov’s ballet suite Stal (Steel) (1926), perhaps an 
attempt to emulate Prokofyev's Pas d'acier; the first movement of this 
suite the noisy, realistic *Zavod' (Foundry), achieved a notoriety, even 
outside the Soviet Union, far beyond its deserts. 

However, the outstanding orchestral work of the nineteen-twenties, 
the only one to win world recognition and establish a permanent place 
in the repertory, was the First Symphony (1925) of Steinberg's young 
pupil Shostakovich (b. 1906).2 The Е minor Symphony was his ‘diploma 
work' on leaving the Petrograd Conservatoire and some of the material 
is nearly identical with that of still earlier works, such as the Piano Trio, 
Op. 8 (1923). It is eclectic in the sense that the music of any young com- 
poser is eclectic: one easily detects the shades of this or that older 
Russian composer or of Prokofyev. But, whereas in the later Nose the 


1*For me a telescope is a symbol for gazing into the distance or at great manifestations— 
often the same thing’, the composer explained: ‘К moemu avtorskomu kontsertu', 
Sovremennaya muzika (1928), no. 30, p. 140. 

? On Shostakovich's symphonies, see particularly Genrikh Orlov, Simfonii Shostakovicha 
(Leningrad, 1961) which has an excellent bibliography, and (on the first six only) M. D. 
Sabinina, Simfonizm Shostakovicha: put К zrelosti (Moscow, 1965). 


652 MUSIC IN THE SOVIET UNION 


influences are heterogeneous, in the Symphony they are homogeneous 
and have been so far absorbed that a clear and new musical personality 
is revealed. The symphony is ‘pure’—or, as a Soviet critic would say, 
'formalistic'—music, marked as Asafyev pointed out! by ‘the rare 
quality of laconicism’ and the ability to ‘seize upon the characteristic 
quality of a thought and reveal it plastically'; it was only years later that 
Soviet critics began to discover its “һегоіс-ігаріс” nature and the 'festively 
monumental style’ of its conclusion. 

Shostakovich’s next two symphonies were by no means ‘pure’ music. 
The Second (‘October’) Symphony (1927)—the first and only edition 
(Moscow, 1927) is entitled “То October: symphonic dedication’, with 
no claim that it is a “symphony’—and the Third (‘First of May’) (1929) 
are both single-movement works, each with a final chorus. Each is 
marked by a good deal of brass declamation against bustling, washed-in 
backgrounds; each has a curious little quasi-concertino episode for solo 
instruments unaccompanied (in the Second, violin, clarinet, and 
bassoon; in the Third, piccolo, oboe, clarinet, and bassoon). But there 
are also considerable differences. The Second opens Largo, with a 
remarkable built-up sound of muted strings, ppp, against which a solo 
trumpet begins to declaim an angular theme. The main Allegro, indicated 
only by а new metronome mark ( = 152), begins in the manner of the 
First Symphony, but the concertino episode develops in pure counter- 
point: 


xo E pE 


sounds) |^ 


P tutti 


в 
Ist Vis. ГЕ ЕЕ SS = 


pizz. 


1 ‘Glebov’, ‘Russkaya simfonicheskaya muzika’, 


EARLY SOVIET ORCHESTRAL MUSIC 653 


which soon reaches eleven independent parts. A great climax is reached 
and then a transition effected for the entrance of the chorus, which sings 
a typical Proletkult slogan-poem by Bezimensky. The final words, 


There is the banner, the name of the living generations: 
October, 

The Commune 

and Lenin 


are shouted, not sung. The music of this section, of which Ex. 269 (from 
near the end)—'October! Herald of Ше wished-for sun'—gives a fair 
idea: 


MUSIC IN THE SOVIET UNION 


654 


Ex. 269 


Moderato 


Cymbals 


soln - tsa 
3 


E - to 


E-to soln- tsa zhe - 


Chorus 


Ok - tyabr! 
РА 


x 


peat 


ЕЕ х= == 


әй 


# 


Fr 


Я 


vest - nik, ОК ~ tyabr! 


- lan - по-уо 


3p 
- lan - no-vo 


vest 


s 


--------------............-....-....-.-.-..-.-..-...-.-.----..-..-..- 


[d 


d 


аа 


DER 


P » 


EARLY SOVIET ORCHESTRAL MUSIC 655 


is purely diatonic, in the strongest contrast to what has gone before. 
Whatever the composer may have intended, the effect is that of an ASM 
orchestral piece with RAPM finale, and the work was at first actually 
accepted—with reservations—by both parties. The Third Symphony is 
a more successful attempt to integrate Shostakovich's earlier style with 
a proletarian one. The idiom is essentially, if somewhat eccentrically, 
diatonic throughout. Some features, such as the opening clarinet call, 
suggest the impact of Mahler, whose symphonies had been performed 
several times in Leningrad during 1925-7: the already known Fifth 
once (1925), the Fourth once (1927), and the Second three times (1926). 
The purely instrumental part of Shostakovich's Third is merely a string 
of episodes, which forecast—they were not (as some Soviet critics 
have claimed) influenced by—his successful activity as a composer of 
film- and theatre-music. But they are not, at any rate in intention, 
formalistic; one can guess their meaning. The unaccompanied duet for 
two clarinets, following the solo call, obviously evokes the breath of 
spring: 


Ех. 270 


Bruce EI J pi 
ЖЕСІ == 


A trumpet calls, inevitably, and keeps on calling; human excitement 
grows. А march for brass and side-drum suggests the gathering for a 
meeting; the twittering quasi-concertino that interrupts it perhaps 
represents young Pioneers, but it leads not to anything like Ex. 268 but 
to a quiet episode slightly reminiscent of the second subject of the First 
Symphony. The climax of the instrumental part seems intended to 
suggest the excitement of a vast May Day meeting, the entire orchestra 
in octaves against an fff roll on side-drum and kettle-drum (high G) 
declaiming rhetorical, quasi-recitative phrases punctuated by the fff 
cannon-shots of a bass-drum: 


656 MUSIC IN THE SOVIET UNION 


Bass drum 


The chorus then sings Kirsanov's May Day poem in block harmonies 
or unison. 

The Third Symphony was unlucky in that its first performance was 
delayed till November 1931, only five months before the proclamation 
of the ideal of ‘socialist realism’. On 15 February 1932 Shostakovich 
announced in Sovetskoe iskusstvo that he had begun ‘a great sym- 
phonic poem with orchestra, chorus and solo vocal numbers. Its 
theme is “From Karl Marx to our own days” '. The words had been 
written by a Proletkult poet, Nikolay Aseyev, and the first of the five 
parts was already composed; the work was intended to last an entire 
evening. But Proletkult was not socialist realism. How much, if any, of 
the music survives in other compositions, we can only conjecture. 


EARLY SOVIET MUSIC IN OTHER FIELDS 657 


EARLY SOVIET MUSIC IN OTHER FIELDS 


Chamber music, piano music and solo song were cultivated only by 
the conservative academics and the modernist formalists. These kinds 
of music were for the few, the already cultured; they tended to express 
individual ideas and subjective emotion, and the proletarians were 
actively hostile to them. Recitals were given mainly under the aegis of 
ASM, but the predominant idiom was that of fastidious late-romantic 
lyricism. 

One of the earliest post-Revolutionary string quartets must have been 
the veteran Glazunov's Sixth (1921). (His Seventh and last, was written 
in 1930 after his emigration.) Like Glier's Third (1928), it is a return to a 
medium abandoned a quarter of a century earlier. And quite a 
number of ‘Soviet’ chamber-compositions were really refurbishings of 
works written or at least sketched before the Revolution: Aleksandrov's 
Quartet, Op. 7, Lyapunov's Piano Sextet, the third and fourth of the 
quartets Myaskovsky brought out as Op. 33. (And Op. 33, nos. 1 and 2, 
composed about 1930, are inferior inhabitants of the same world.) 
Myaskovskian, too, are the Op. 2 quartet (1923) and the String Trio, 
Op. 4 (1924) of Shebalin, and Kabalevsky's First Quartet (1928): works 
of promise rather than achievement. Only in Shebalin is there a whiff of 
Prokofyev's brand of anti-romanticism, with tougher modernism repre- 
sented by the quartets, trios, and violin sonatas of Roslavets and by 
Popov's Sextet. 

The piano music of the period is similar in spirit but technically more 
accomplished. The sonatas of Myaskovsky, Anatoly Aleksandrov 
(b. 1888), and Samuil Feinberg (1890-1962) belong to the aesthetic 
worlds of Skryabin, Medtner, or Rakhmaninov. It is music for the 
recital-platform; later they had to learn to write in a more intimate 
style and for humbler performers. Nothing could be in sharper contrast 
with this refined if rather etiolated art than the early piano works of 
Shostakovich, the Fantastic Dances, First Sonata and Aforizmi (1922-7), 
which emulate by turns Prokofyev, Stravinsky, and Hindemith.! The 
two wings of ASM piano music may be illustrated by (1) the opening of 
Feinberg's Prelude, op. 8, No. 1, (ii) the opening of the slow move- 
ment of Shostakovich's First Sonata: 


1 For a ‘socialist-realist’ judgement on these works and their immediate successors, see 
Mikhail Druskin, ‘О fortepiannom tvorchestve D. Shostakovicha’, Sovetskaya muzika 
(1935), no. 11, p. 52. 


MUSIC IN THE SOVIET UNION 


658 


Ex. 272 


sempre marcato e cantando 


dolce 


та marcato 


659 


EARLY SOVIET MUSIC IN OTHER FIELDS 


UN 
zi 
| [| 

| 

| 


16 ШЫ" ҮН, 


660 MUSIC IN THE SOVIET UNION Е 

Both were written in 1926; both were objectionable to the proletarians. 
Equally objectionable was another piano-piece of the same year: the 
‘coarsely naturalistic’ railway noises of Re/si (The Rails), by the Prolet- 
kult composer Vladimir Deshevov (1889-1955). Polovinkin, always 
more original in his titles (for instance, his Elektrifikat for piano) than 
in his music, having composed four sonatas during 1924-6, marked the 
‘crisis іп the Soviet piano sonata’ by naming No. 5 (1928) Poslednyaya 
sonata (Last Sonata). 

АП these piano composers, with one exception, composed also for 
voice and piano. (The exception was Shostakovich, whose solo songs 
of the nineteen-twenties—settings of two Krilov fables (1921-2), which 
already show his gift for satire, and of Japanese poems (1928)—have 
orchestral accompaniment.) To them we must add one too-little known 
composer, Vasily Nechaev (1895-1956), a true song-writer who is at his 
best in his Blok and Esenin settings of 1926-8. Their choice of poets was 
symptomatic; instead of the young Soviet poets, they generally preferred 
translations from the Japanese or of Sappho, the Russian classics and 
(above all) the Symbolists and *Perfectionists'—Balmont, Blok, Gippius, 
Akhmatova. The music is essentially a continuation of the polished 
Russian art-song of the pre-Revolution period. There is no social or 
political awareness; even composers, such as Myaskovsky, who had 
shown it in their orchestral works, naturally felt free here to retreat into 
a private world of personal dreams and emotions. Only the egregious 
Mosolov set to music a collection of Gazetnie obyavleniya (Newspaper 
advertisements) (1926). An art like theirs could hardly be expected to 
flourish under the triumphant proletarianism of 1929—31 or even in the 
freer air of early ‘socialist realism’. 


SOCIALIST REALISM IN OPERA 

Given the Marxian postulates, reinforced by the nineteenth-century 
Russian view of the function of art, the Party had to lay down a new 
and firm literary-artistic policy in 1932; and, granted those premises, 
the policy was not so arbitrary as it may seem to the non-Russian. 
ASM had been hopelessly out of touch with ‘the people’; on the other 
hand RAPM had not only lowered standards but had been so intolerant 
that, by comparison, early ‘socialist realism’ was liberal. The Party was 
at this point legislating for a situation, though, as usual with benevolent 
paternalism, it made no allowances for the eccentricities of exceptionally 
gifted children. 

The composers and other creative artists even had a year or two to 
work out in the new Union of Soviet Composers and parallel bodies 
their own interpretation of ‘socialist realism’. This was naturally more 


SOCIALIST REALISM IN OPERA 661 


difficult in music than in literature, though least difficult in opera. As 
always at these policy changes there was trouble with works in progress. 
At the beginning of the nineteen-thirties two composers had been 
working independently on operas dealing with the serf-rising against 
Shuysky led by Ivan Bolotnikov in 1606: Nechaev and Valery Zhelo- 
binsky (1913-46). Nechaev had shown very mild leanings to ‘modernism’ 
and his Ivan Bolotnikov, though accepted for production in 1932 by 
Stanislavsky, never reached the stage even in its second version of 
1936-7. Zhelobinsky's Kamarinsky muzhik' was produced in 1933. 
His model was clearly Boris—there is even a Polish scene, with 
mazurka, and a Polish seductress; the hero was meant to be simply 
a symbol of the oppressed people; the idiom is predominantly vocal, 
folk-songish melody or melodic recitative. But, apart from the fact 
that Zhelobinsky was no Mussorgsky, his characters are lay-figures 
and he was criticized both for his failure to show Bolotnikov’s con- 
nexion with the people and for his ‘grotesque’, ‘ironical’ treatment of 
the boyars. The comic quartet for the bass princes in scene 4, accom- 
panied only by the orchestral basses, certainly recalls Puccini’s Ping, 
Pang, and Pong: 


Ex, 273 . 


DMITRY SHUYSKY 


vi - shli wvpo- le, 


be- = = те ( = 


Руа dnya ne е - li 


1 See М. Glukh, ‘Kamarinsky muzhik—opera У. Zhelobinskovo', Sovetskaya muzika 
(1934), по. 8, р. 3, and S. Levit in Zstoriya russkoy sovetskoy muziki, i (Moscow, 1959), 
pp. 176 ff. 


662 MUSIC IN THE SOVIET UNION 


GOLITSIN 
Ya vi-shel sle- va, 


ее em ee ры 
KURAKIN $ниузку Ya vi -shel sbo-ku, 
Е ee ЕЕ ЕЕ ЕЕ 
Ya vi-shel  sle-va, Ya  vi-shel sbo-ku 
р 2—00 Бр р " Е : i E bg [4 н га eo 
Orch | ESX: Se я 
ве 2. 22 we) 
А EE на. 
Ё ro É * um = 
а АА егт”. 


(We went out in the fields. It was scarcely daylight. For two days we didn't drink. 
For two days we didn't eat. I went left. I went sideways.) 


The chorus of serfs in the last scene is typical of Zhelobinsky's folk- 
songish melodic style and flaccid harmony—and also of a great deal of 
other Soviet opera: 


Ex. 274 


Pti - tsa si-ni-tsa vtem-ni 


(The tomtit languishes in prison; it neither eats nor drinks nor sings.) 


PLATE IX 


(£99 `d 225) YEGI 


‘LOIN 


A^02SO|A ш uonjnpoJd [ешо 3y} шолу 9ua5s v 


LSIG MSNAHSLIN 


AHL AO НИЗЯОРИ AQV'I 


ЯНА SHOIAOV.LSOHS 


SOCIALIST REALISM IN OPERA 663 


Indeed Kamarinsky muzhik is in subject and musical style typical of a 
whole genre of Soviet operas, from Pashchenko's Orliny bunt onward, 
while his Jmenini (The Name-day) (1935), based on a nineteenth- 
century story of a serf-musician's tragedy, is an obvious attempt to 
emulate Shishov's Tupeyny khudozhnik. 

These second-rate works are completely overshadowed by an opera 
which, whatever its faults and however adventitious its fame, is the 
only Soviet opera to make its way all over the world: Shostakovich's 
Lady Macbeth Mtsenskovo uezda (The Lady Macbeth of the Mtsensk 
District) or, to give its alternative title, Katerina Izmaylova. Although 
not produced until 22 January 1934, it had been composed during 
1930-2 and is marked, like the Third Symphony, by a mixture of in- 
compatible styles. Side by side with beautiful or powerful invention, as 
in the passacaglia entr'acte in Act II, and Act IV as a whole—its tragic 
key-note sounded by the old convict's song at the beginning: 

Ex. 215 


- mi - tel- ny | гпоу, 


(Mile after mile the long file trudges, burned by wearying heat.) 
1 See pl. IX. 


664 MUSIC IN THE SOVIET UNION 


there is a great deal of parodic and burlesque music, associated not 
only with the old Izmaylov (where ‘it ‘is defensible and even admir- 
able) but with the anti-hero Sergey and the policemen of Act III. 
And Leskov’s sombre realistic story is falsified not only by this con- 
version into a satire on the nineteenth-century provincial merchant- 
class but by the attempt to make the triple -murderess Katerina a 
sympathetic figure. The dramatic incongruity is as marked as the 
musical. Yet there are master-strokes, such as the subtle recall of 
Katerina’s song of melancholy boredom (a nice parody of the senti- 
mental Russian ‘romance’ of the eighteen-forties) in the third scene of 
the First Act (1) when she thinks of the ‘deep black lake in the forest? 
near the end of the last (ii): 


Ex. 276 
(i) Adagio 


QE EHI EE 


Ya vo - kosh - ko od- nazh-di u- vi - de- 


ЕЕ 


est pod  kri-she-yu  gnez- dish- ko; 


(Once from the window I saw a little nest under the eaves;) 


Gi) Adagio 


Peor 


So-vsem krug-lo- e, о-сһеп gl - bo-ko-e,__ 
-  davnem cher-na - ya, kak mo - 
- ya so-vest, cher- na- ya. 


(quite round, very deep, and its water black—like my conscience, black.) 


SOCIALIST REALISM IN OPERA 665 


and it is not difficult to see why, amid the operatic mediocrity of the 
time, it was hailed as “а most significant landmark on the road of the 
creative development of Soviet musical art’, a genre sui generis— 
*tragic satire'.! Nevertheless when the Party in 1936 decided that com- 
posers were interpreting the directive of 1932 altogether too loosely, it 
was specifically at Тле Lady Macbeth and Shostakovich's recent col- 
lective-farm ballet Svetly ruchey (Clear Stream) that the official 
thunderbolts were directed. On 28 January 1936 Pravda printed its now 
historic article ‘Sumbur vmesto muziki (Confusion instead of music)? 
asserting that ‘from the first minute the listener to The Lady Macbeth is 
dumbfounded by a deliberately discordant, confused stream of sounds'; 
the music was ‘modernist formalism’ of the worst kind. The defining of 
*socialist realism" was too serious a matter to be left to musicians; it was 
taken over by a government Committee for Artistic Affairs. The 
notorious modernist-formalist Mosolov, who made a public protest in a 
Moscow restaurant on 31 January, was unanimously expelled from the 
Union of Soviet Composers for drunken brawling, withdrew to Buryat- 
Mongolia, and reappeared as a composer of simple, conservative 
music only in 1939. 

The opera publicly approved by Stalin and Molotov as a model 
Soviet opera, eleven days before the Pravda condemnation of The Lady 
Macbeth, was Tikhy Don (The Quiet Don) by Ivan Dzerzhinsky 
(b. 1909).3 Based on Sholokhov's novel of Don Cossack life during the 
1914 War and the Revolution, Tikhy Don was first produced in 1935 as 
a result, ironically, of the help and encouragement of Shostakovich to 
whom it is dedicated. Unlike earlier Soviet operas on similar subjects, 
It is concerned with genuine characters who are typical of the masses 
without being mere types; unhappily the simple lyrical score never 
brings them to musical life. A few bars from the episode of the soldier 
half-crazy from shellshock in Act III (Ex. 277) will not at all unfairly 
illustrate the naive technique and dramatic impotence of the entire opera. 

Apart from its wealth of rather undistinguished melody in a diluted 
folk-song idiom, the merits of Tikhy Don are mostly negative; there is 
no trace of any influence more modern than early Debussy; even 


1A. Ostretsov, ‘Lady Macbeth Mtsenskovo uezda: opera Dmitriya Shostakovicha’, 
Sovetskaya muzika (1933), no. 6, p. 9. For a much later Russian study of the opera, see 
Iosif Rizhkin in Istoriya russkoy sovetskoy muziki, ii (Moscow, 1959), рр. 196 ff. 

? Reprinted in Sovetskaya muzika (1936), по. 2, p. 4, the second article, ‘Baletnaya 
falsh’, ibid. p. 6; translated excerpts from both in Gerald Abraham, Eight Soviet 
Composers (London, 1943), pp. 25-26; longer excerpt from the first, Slonimsky, op. cit., 
p. 402. 

3 On Tikhy Don and Dzerzhinsky's views on Soviet opera, see Abraham, op. cit., pp. 
81 f., A. Budyakovsky, ‘Tikhy Don’ I. Dzerzhinskovo', Sovetskaya muzika (1935), no. 
p. 11, 38, and Rizhkin, op. cit., ii, pp. 215 ff. 


666 MUSIC IN THE SOVIET UNION 


Ех. 277 | 
[Slowly, march-like] 


х= сшкш Е ВЕЕ- БН 


slez-mi gor-Ki- mi  ob-mo- yu. 


no-zhen- 


ki, 


(Oh, have pity, I want to live, don’t destroy me, give me a hand, food, stretch out 
your feet and ГЇЇ wash them with bitter tears.) 


the reactionary Listnitskys, father and son, are not caricatured in 
grotesque music. Dzerzhinsky also composed an opera on the sequel to 
Sholokhov's novel, Podnyataya tselina (Virgin soil upturned) (1937)! 
but neither this nor Dzerzhinsky’s later operas achieved success com- 
parable with that of Tikhy Don. 

The way pointed by Tikhy Don was followed by Oles Chishko 


1See Budyakovsky, ‘Ivan Dzerzhinsky i evo opera Podnyataya tselina’, Sovetskaya 
muzika (1937), no. 10-11, p. 44, and Rizhkin, op. cit., ii, pp. 227 ff. 


SOCIALIST REALISM IN OPERA 667 


(b. 1895) in Bronenosets Potemkin (The Battleship Potyomkin) (1937),! 
Zhelobinsky in Mat (The Mother, after Gorky) (1938),? Leon Khodzha- 
Eynatov (1904-54) in Myatezh (The Revolt, on a story by the “рго- 
letarian’ Dmitry Furmanov) (1939) and Semya (The Family, based— 
like Chishko's opera—on a film) (1940), and Tikhon Khrennikov 
(b. 1913) in V buryu (In the Storm, based on Nikolay Virta's notable 
novel Odinochestvo) (1939). Aesthetically similar to these, but re- 
freshingly set in sixteenth-century Burgundy, is Kabalevsky's Master 
iz Klamsi (The Master of Clamecy, based on Romain Rolland's 
Colas Breugnon—by which name the opera has become generally 
known) (1938),* a lyrical, if not very original, score with at least one 
rounded, musically living character, Colas himself. But a greater 
musician than any of these had been waiting for seven or eight years in 
the wings; in 1940 he stepped on to the Soviet stage with a ballet and 
an opera. 

Prokofyev had always been the idol of ASM. In 1927, when he paid 
his first visit to Russia after the Revolution, he was greeted as composer 
and soloist with wild enthusiasm; and late in 1932, when the RAPM 
reaction seemed to have been crushed, he decided to return permanently 
and settle in Moscow. But at first he wrote relatively little:? mostly 
incidental music or film-music—notably for Poruchik Kizhe (Lieutenant 
Kizhe) (1933) and Aleksandr Nevsky (1938)—and, although he had 
finished the ballet Romeo and Juliet in 1936, only concert suites from it 
were performed. Now in January 1940 Romeo and Juliet was at last 
produced and on 23 June his first Soviet opera Semen Kotko. It 
aroused an immediate storm and was contrasted with Khrennikov's 
V buryu very much as The Lady Macbeth had been with Tikhy Don. 

V buryu has a great deal in common with Tikhy Don: the folk-song 
idiom, relieved here and there by watered-down Tchaikovsky (in 
scene 4), and the general weakness of characterization. But it is a better 
Tikhy Don, less helpless technically and with little tonal twists that 
relieve the melodic insipidity, as in Aksinya's song in scene 2: 


1 бее Rizhkin, op. cit, ii, pp. 237 ff. and A. Steinberg, ‘Opera о “Втопеповізе 
Potemkine" °’, Sovetskaya muzika (1937), no. 10-11, p. 55. 

? See M. A. Grinberg, ‘Opera Mat V. Zhelobinskovo,’ Sovetskaya muzika (1939), по. 5, 
p. 9, and Levit, op. cit., ii, pp. 241 ff. 

3See Ivan Martinov, ‘Bratya—opera T. Khrennikova' (Bratya—Brothers—was the 
original title), Sovetskaya muzika (1937), no. 10-11, p. 64, and ‘V buryu—T. Khrennikova’, 
ibid. (1939), no. 11, p. 55, and Rizhkin, op. cit., ii, pp. 249 ff. On Virta's novel, see Gleb 
Struve, 25 Years of Soviet Russian Literature (2nd ed., London, 1944), p. 281. 

4 бее Abraham, op. cit, рр. 73 ff., L. Danilevich, ‘Master iz Klamsi’, Sovetskaya 
muzika (1937), no. 12, p. 35, and Levit, op. cit., pp. 289 ff. 

5 See infra, pp. 672. As early as 1934 he courageously drew attention to the danger of 
Soviet music becoming ‘provincial’, 


668 MUSIC IN THE SOVIET UNION 


Ex. 278 
Andante sostenuto 
(The voice doubles the highest punt и 
Raz-le-ta-lis so- ko - li vraz-ni - е sto-ro-nush- 


(The falcons flew away to different lands, left and forgot the native nest.) 


The action is laid in and around a Tambov village in 1921, with most of 
the peasants at first uncertain whether to side with the Bolsheviks or 
with Antonov's counter-revolutionaries. The drama of decision is 
decided in the minds of the heroine Natasha (a passive character like 
so many heroines in Russian opera), her lover (whose brother is already 
a convinced Bolshevik) and her father, and has an emotional parallel in 
the love-affair; but Khrennikov has no power to convey deep emotion. 
An artificial high-light introduced in the version produced in 1939—it 
does not occur in the unperformed original version of 1937—15 the visit 
which Natasha's father and Listat, the Bolshevik brother, pay to Lenin 
in Moscow (scene 5); this was Lenin's first appearance on the opera- 
stage, but he was not allowed to sing; when he speaks the music stops. 

Semen Kotko is a different matter. As early as 1933 Prokofyev had 
expressed the wish to write an opera on a Soviet theme, ‘heroic and 
constructive',! and he at last found this theme in a story by Valentin 
Kataev. Like V buryu, it is a peasant drama of the Civil War, played out 


1 Prokofyev, ‘Zametki’, Sovetskaya musika (1933), по. 3, p. 99. 


SOCIALIST REALISM IN OPERA 669 


in 1918 when the Communists in the Ukraine still had to contend with 
German troops as well as the counter-revolutionaries. The characters 
are not unlike those of Khrennikov's opera, but Prokofyev was far more 
gifted and much better equipped. Instead of relying almost exclusively 
on a song-based idiom, he commanded one compounded of melodic 
and declamatory elements which enabled him to convey shades of 
feeling and draw character with a subtlety, power and flexibility quite 
beyond the range of his younger colleagues. Consider Remenyuk's 
farewell to the bodies of his murdered comrades: 


-se - Ју mat - ros . s Cherno- mor-sko-vo flo-ta. . 


(My friend, Vasya Tsarev, merry sailor from the Black Sea Fleet—) 


or compare the heart-rending lament of Tsarev's sweetheart, Lyubka, 
out of her mind at his death, with Dzerzhinsky's half-mad soldier 
(Еж, 2/7); 


Ех. 280 


EJS S К 
КҤҮУВҮАГ@ Z-3)—.1—9.—D —42 
[Wap —.9—— — 2, 


44 


670 MUSIC IN THE SOVIET UNION 


Khiv ~ гуа, smi-luy-sya i za-stu - pi - суа! 


(Lyubka: Doll without movement. .. Khivrya: May the holy martyr Khivrya 
pardon and intercede!) 


(The older woman, Khivrya, meanwhile comments in her own accents.) 
There is plenty of strong, long-breathed diatonic melody, especially in 
connexion with the hero Semyon Kotko; it is employed romantically in 
the manner of Rimsky-Korsakov to set the mood of the summer night 
and the three pairs of lovers at the beginning of Act III ;! here much of it 
is related to Ukrainian folk-song, such as the *Rano ranenko' (Early 
in the morning) melody heard as the second subject of the overture, the 
wedding chorus in Act П, and sung by Semyon and his sweetheart, 
Sofya, at the end of the opera. There is humour, as when Semyon 
instructs the young partisans in manning a field-gun in Act IV, and in 
the love affair of his young sister. The Germans are not caricatured but 
characterized, as they had already been in Aleksandr Nevsky, by 
viciously dissonant harmony. АП these qualities were widely recognized 
at the time but they failed to save the opera: it quickly disappeared from 
the repertory and was revived—in concert form—only іп 1958.? 


SOCIALIST REALISM IN THE CONCERT HALL 


The achievement of true socialist realism proved still more difficult 
in symphonic music. Even Myaskovsky, who produced no fewer 
than nine symphonies, Nos. 13 to 21, as well as a Violin Concerto, 
during the period 1933-40, found it difficult to maintain the right 
note. Nos. 14, 16, and 21 were considered the most satisfactory 


! Quoted by Nestyev, Prokofyev (Moscow, 1957): English edition (London, 1960), p. 316. 

2 The prolonged controversy about Kotko filled the Soviet musical press for some time 
and is discussed in all the literature on Prokofyev. On this, and the work generally, see 
particularly M. Sabinina, ‘Semen Kotko’ i problemi opernoy dramaturgii Prokofyeva 
(Moscow, 1963). 


SOCIALIST REALISM IN THE CONCERT HALL 671 


but in his autobiography! he confessed to various backslidings into 
subjectivism. But his ‘monumental’ Sixth was matched in 1932 by a 
work of irreproachable orthodoxy by Shaporin.? As always with 
Shaporin, the musical language is conservative yet he contrives to say 
fresh things in it. The Symphony employs chorus, brass band, and piano 
in addition to a large orchestra and is planned in four movements: 
(1) introducing themes suggesting Russia's remote heroic past, in the 
manner of Borodin, and recalling the heroic days of the more recent 
past (Yablochko?); (2) a dance-scherzo; (3) a dark-coloured lullaby for 
female chorus, which leads into (4) ‘Campaign’, a broad, powerful 
movement that does indeed suggest ‘the movement of colossal human 
masses’.® Six years later, in a much finer 'symphony-cantata' Shaporin 
crystallized similar ideas and emotions around a cycle of poems by 
Blok: Na pole Kulikovom (On the field of Kulikovo).* Here, as in the 
symphonic cantata which at about the same time Prokofyev developed 
from his Aleksandr Nevsky film music? and in the ‘oratorio’ Emelyan 
Pugachev by Marian Koval (1907-71), heroes from Russia's past were 
glorified, in two cases princely and patriotic, not revolutionary, heroes 
who distinguished themselves against foreign invaders. The increasing 
threat of Nazi Germany led to official encouragement of patriotic art. 
Kulikovo and Aleksandr Nevsky are outstanding in a genre hitherto rare 
in Russia: large-scale secular choral composition. In both works, as in 
Koval’s, the choral scoring is very dull by Western standards; the choirs 
sing almost entirely in solid harmony or unison, without contrapuntal 
interest; the half-dozen bars of imitative writing in the ‘Lullaby’ of 
Kulikovo are quite exceptional but the massive effects in the fourth and 
last movements of Nevsky, and the first and third movements and 
Epilogue of Kulikovo are splendid. There is a certain parallel between 
the works as wholes; in each the fifth movement is a battle-scene and is 
followed by a beautiful soprano solo, in one case a lament over the 
Russian dead, in the other the reassuring lullaby of a Russian mother. 

Symphonies and other orchestral works glorifying the Red Army or 
more peaceful activities of the Soviet Union had already appeared: 
Knipper's Symphonies, Nos. 3 (‘Far Eastern’) (1932), 4 (‘Poem of the 


1*Avtobiograficheskie zametki о tvorcheskom рші, Sovetskaya muzika (1936), no. 6, 
р. 3; reprinted in M. A. Grinberg (ed.), Sovetskaya simfonicheskaya muzika (Moscow, 
1955), p. 20. 

2 See Abraham, ор. cit., pp. 90 ff. On Shaporin generally, see Levit, Yury Aleksandrovich 
Shaporin (Moscow, 1964). 

ЗА. Ostretsov, 'Sovetskoe simfonicheskoe tvorchestvo', Sovetskaya muzika (1935), no. 4, 
p. 19. 

а See S. Skrebkov and V. Protopopov in Istoriya, ii, pp. 169 ff., and Abraham, op. cit., 
pp. 94 ff. 

5 Skrebkov and Protopopov, ibid., pp. 163 ff., and Abraham, ibid., pp. 38 ff. 


672 MUSIC IN THE SOVIET UNION 


Fighting Komsomols’) (1934), 6 (dedicated to the Red Cavalry but ‘dry, 
abstract and formalistic’) (1936) and 7 (‘Military’) (1938),! Polovinkin’s 
Fourth Symphony (‘Red Army’) (1933), Steinberg’s Fourth (‘Turksib’) 
(1933), Shebalin’s Fourth (‘Perekop’) (1935), Vasilenko's Red Army 
Rhapsody (1938). Another group of works manifested the interest of 
Russian composers in the native music of the non-Slavonic republics of 
the U.S.S.R.,? which they were officially encouraged to visit—both to 
study folk-lore and to stimulate the development of native art. As early 
as 1927 Glier’s Azerbaydzhanian opera Shakh-Senem had been pro- 
duced at Baku,? and about 1931-3 Knipper composed a series of 
orchestral works on Tadzhik themes,* notably the suite Vanch, 
Shekhter a Turkmenia suite, and the veteran Ippolitov-Ivanov various 
Uzbek and Turkmenian ‘pictures’ and ‘fragments’ for orchestra. And 
before long a native Armenian, Aram Khachaturyan (b. 1903), was to 
appear with a Trio for piano, violin and clarinet (1932), a First 
Symphony (1934), a First Piano Concerto (1936), a Violin Concerto 
(1940),5 essentially Russian but coloured by the folk-music of Armenia 
and Uzbekistan. 

‘Pure’ orchestral music became rare after 1932. Kabalevsky’s First 
Symphony (1932) was inspired by Gusev’s poem ‘The Year 1917’, 
though his attractive Second—in three movements with a scherzo-finale 
(1934)—and Khrennikov’s First (1936) are abstract works and so, 
despite their conservative idiom, aroused little enthusiasm. As for 
Popov’s First Symphony, not performed till 1935, it had been conceived 
as early as 1927—in the high noon of sovremennost (‘contemporariness’) 
—and was now condemned out of hand. Prokofyev, after devastating 
adverse criticism of his Simfonicheskaya pesn (Symphonic Song) 
(1933), his first composition after his return to his homeland, wrote a 
Second Violin Concerto, in G minor (1935), but refrained at this period 
from venturing on a symphony. Shostakovich was bolder but ran into 
trouble. After a deliberately vulgar Piano Concerto (1933), in which a 
solo trumpet plays a scarcely less important part than the piano, he 
had made an almost wholly serious essay in pure instrumental music, a 
Cello Sonata (1934), and followed this in 1936 with a Fourth Sym- 


1 On these symphonies by Knipper, see Abraham, op. cit., pp. 55-60. 

2 See infra, pp. 699—700. 

з Оп Shakh-Senem see Ostretsov, ‘Shakh-Senem’, Sovetskaya muzika (1938), no. 2, 
p. 45, and Rizhkin, Istoriya, i, pp. 165 ff. 

*See Viktor Belyaev, ‘Tadzhikskie narodnie temi v syuite Vanch Г. Knippera’, 
Sovetskaya muzika (1937), no. 4, p. 49. 

5 On these works, see Abraham, op. cit., pp. 45 #., Georgy Khubov, Sovetskaya muzika 
(1939), по. 9-10, p. 18, Martinov, ibid. (1938), no. 5, р. 26, and Tumanina, Istoriya, ii, 
pp. 450 ff. 


SOCIALIST REALISM IN THE CONCERT HALL 673 


phony.! Like so much of Shostakovich's music, it is a mixture of styles— 
to which, as will be apparent from the openings of the second move- 
ment and slow introduction to the third (the finale): 


Ex. 281 
(i) 


Moderato con moto 


that of Mahler was now added. The concertino element of the Second 
and Third Symphonies reappears in the first movement but there is little 
of their revolutionary rhetoric. As with Mahler himself, an un- 
mistakably personal utterance emerges from the heterogeneous musical 
fabric. And the work ends with a faint question mark: against the very 
long-held C minor triad of the strings and celesta, the solo trumpet 
takes 18 bars to resolve its F sharp on G. The Fourth Symphony was on 


1 See Г. Danilevich, Nash sovremennik: Tvorchestvo Shostakovicha (Moscow, 1965), 
pp. 127 ff. 


674 MUSIC IN THE SOVIET UNION 


the eve of performance when the storm over The Lady Macbeth broke. 
Shostakovich decided that it did not exemplify socialist realism in the 
symphony and withdrew it; it was first heard in December 1962. Its 
successor (1938) was described by Shostakovich himself as “а Soviet 
artist's practical creative reply to just criticism’.+ 


The theme of my symphony is the stabilization of a personality. In the centre 
of this composition—conceived lyrically from beginning to end—I saw a 
man with all his experiences. The finale resolves the tragically tense impulses 
of the earlier movements into optimism and joy of living. 


Lyricism and the optimistic D major of the finale cancelled out the 
subjective elements of the earlier movements. But the socialist realism 
of the Sixth Symphony (1939) was more doubtful:? the slow first move- 
ment was found too elegiac, the remaining two movements too noisy, 
empty, and formalistic. 

From this period also date two notable chamber-works by Shosta- 
kovich, the Piano Quintet (1940) and the first (1938) of his series of 
string quartets. Chamber music and solo piano music had flourished 
briefly in the early days of socialist realism, though the piano music— 
for instance, Shostakovich's Twenty-four Preludes (1932-3), Khacha- 
turyan's Тосса а (1932), and Kabalevsky's Sonatinas, op. 13 (1930)-- 
tended to be technically less demanding, as if written for domestic use 
rather than the concert-platform. In the field of solo song composers 
conscious of the dangers of subjective expression turned to arrange- 
ments of folk-song but no longer limited themselves exclusively to 
Russia or the sister-republics: in 1933 both Anatoly Aleksandrov and 
Feinberg published sets of ‘songs of the Western peoples’ (including a 
number of Hebridean, Lowland Scottish, and English examples). Koval, 
on the other hand, searched the Russian classics for socially significant 
poems and published sets of Pushkin and Nekrasov songs: his 
Pushkiniana (1934)? is a cycle of ten rather declamatory songs, inter- 
spersed with readings from Pushkin's diaries and note-books. Koval was 
a former ‘proletarian’ composer and one of his most lyrical Pushkin 
songs may be instructively compared with one from the same period by 
a former ‘contemporary’, Mosolov: 


1 On Ше Fifth Symphony see particularly Khubov, ‘Pyataya simfoniya D. Shostakovicha’, 
Sovetskaya muzika (1938), no. 3, p. 14, and Orlov, op. cit., pp. 62 ff. 

2 Orlov as usual quotes and summarizes the contemporary criticisms, op. cit., рр. 102 ff. 

3 See Shlifsteyn, ‘Pushkiniana M. Kovalya', Sovetskaya muzika (1937), no. 2, p. 42. 


SOCIALIST REALISM IN THE CONCERT HALL 675 


0) KOVAL 


ие | Sh eS SS Е] 


(Useless gift, chance gift; life, why art thou given me?) 


MOSOLOV 
Moderato 
Dp 


Vtvo - уш svet - li- tsu, drug moy nezh 


676 MUSIC IN THE SOVIET UNION 


(Into your little room, my tender friend, I come for the last time.) 


The Pushkin centenary was also marked by the appearance of a 
number of finely wrought settings of his more lyrical verse by Shaporin, 
Aleksandrov and Feynberg. In particular Shaporin's five Pushkin songs, 
op. 10 (1935), with his Dalekaya yunost (Far-off Youth) (1935-40) 
and his Elegii, op. 18 (1945), despite the conservatism of their idiom, 
entitle him by their poetic sensitivity and beautiful craftsmanship to a 
place beside all but the greatest of Russian song-composers. 


PROKOFYEV'S ‘WAR AND PEACE’ 


The outbreak of the ‘Great War for the Fatherland’ evoked a great 
upsurge of patriotic feeling which found the most varied outlets of 
musical expression: in great symphonic works, opera, and even ballet. 
(Khachaturyan, for instance, drastically recast his first ballet, Schastye 
(Happiness) (1939), with a fresh, bellicose scenario as Gayane (1942).) 
Opera-composers drew inspiration first from national heroes of the past 
--УавПепКо in Suvorov, Koval in Emelyan Pugachev (to whom he had 
already dedicated an oratorio), Boris Mokrousov (b. 1909) in Chapaev 
(all three produced in 1942)—later from heroic events in the German 
war itself: Dzerzhinsky’s Krov naroda (The Blood of the People) (1942) 
and Nadezhda Svetlova (1943), Kabalevsky’s Pod Moskvoy (Near 
Moscow: also known as V ogne, In the Fire) (1943), Koval’s Sevas- 
topoltsi (The Defenders of Sevastopol) (1946). Now and again, a com- 
poser would be content to turn to a literary classic, as Viktor Trambitsky 
did to Ostrovsky's Groza (The Storm) (1942) and Anatoly Aleksandrov 
to Lermontov in Bela (produced 1946, but completed several years 
earlier), but such operas were exceptional. Some of the war operas 
proved hardly more than ephemeral; all were overshadowed by 


PROKOFYEV'S 'WAR AND РЕАСЕ” 671 


Prokofyev's Voyna i mir (War and Peace) (original version, 1943; 
produced 1946).1 

Tolstoy's novel seems an impossible opera-subject; its vast field of 
vision, the multitude of fully conceived characters, above all that sense 
of time passing and a whole generation of youth achieving maturity— 
all this is beyond theatre, and music could add nothing to it. What 
Prokofyev did was to extract on the one hand the principal characters, 
and the personal tragedy of Andrey's love for Natasha and her seduction 
by Kuragin, on the other the great set-pieces of 1812: the Battle of 
Borodino, the council of war at Fili (one of the new scenes of the second 
version), the burning of Moscow, and the French retreat. Indeed the 
opera divides naturally into two parts. The first seven scenes are all 
personal drama; at the end of the seventh, when Pierre has put Kuragin 
to shame and flight, Denisov enters with the news that *Napoleon has 
advanced his troops to our border’. After that, only in the wonderful 
twelfth scene where Natasha comes to the dying Prince Andrey (and 
Prokofyev gives a choral embodiment to ‘the soft, whispering voice’ of 
his heart—Andrey did not know whether it was delusion or reality— 
‘incessantly and rhythmically repeating ‘‘i-piti-piti-piti” °): 


Ex. 283 


Lento (J = 60) 


Altos 


1 The original version consisted of 11 scenes; only eight scenes were produced in 1946. 
Prokofyev later made cuts but also substantial additions, including two completely new 
scenes; 11 scenes of this version were crowded into one night in 1955, the complete 13-scene 
form was produced only in 1957. On War and Peace, see particularly A. Khokhlovkina, 
*Voyna i mir Sergeya Prokofyeva', Sovetskaya muzika (1946), no. 8-9, p. 15, Nestyev, 
op. cit. (English ed.), pp. 445 ff. and Rizhkin, Istoriya, iii, pp. 211 ff. 


678 MUSIC IN THE SOVIET UNION 


г 
: 
Ш 
3 
j м 
| 
| 
| 
| 
Li 


(Why is it white by the door?) 


PROKOFYEV’S ‘WAR AND PEACE’ 679 


do we return to the personal drama. We see Pierre at the redoubt at 
Borodino, arrested in captured Moscow, talking to the peasant 
Karataev, rescued (not by Cossacks as in the book but by partizani), 
but we do not apprehend what is going on in his heart and mind. The 
hero is now the Russian people and Kutuzov, the embodiment of its 
immanent will. Like other great Russian operas—Boris itself —Var and 
Peace is a masterpiece not because of its wholeness but because of the 
greatness of its parts. 

The music is worthy of the subject. In this score, better than in any 
other, Prokofyev showed how a great creative artist, denied total free- 
dom of musical speech, may yet achieve unmistakably individual utter- 
ance. Even his earlier works, written in the West, had shown a vein of 
strong diatonic melody; he found this an ever valuable resource after his 
return to Russia; and in War and Peace, as in Romeo and Juliet, he 
achieves a cool classic beauty that is rare in Russian music. It is heard 
straight away in the very first scene, as Andrey sits by the open window 
at Otradnoe in the May moonlight, when the flute plays the melody— 
borrowed from incidental music to a stage version of Eugene Onegin 
(1936)—that is to be associated throughout with Natasha: 


Ex. 284 


Andante assai (4 = 66) 


680 MUSIC IN THE SOVIET UNION 


(Bright spring sky, is not this perhaps an illusion?) 


At times the diatonic idiom is handled artlessly, as in Kutuzov's noble 
but too Borodin-like monologue on Ше abandonment of ‘white-stoned 
mother Moscow' after the council of war. Much more typical are the 
harmonic twists and rhythmic energy that stamp Andrey's assurance to 
Pierre before Borodino: “Вш I tell you: be that as it may, we shall win 
this battle'. Even the big war scenes, spectacular as in Meyerbeer or 
Wagner, remind one more of Mussorgsky than of ‘grand’ opera—in 
many more details than the reading of the proclamation and the 
appearance of the madmen in captured Moscow. Nothing even in 
Mussorgsky is more moving than the scene of Andrey’s dying. 

The first version of War and Peace was produced on 12 June 1946; 
less than five months later (3 November) came another Prokofyev opera, 
one that had been on the point of production five years before, when war 
broke out: Obruchenie v monastire (Betrothal in a Priory: after 
Sheridan’s Duenna).*In spite of his declared intention, Prokofyev allowed 
the comic side, with its brilliantly subtle declamation and characteriza- 
tion, to outweigh the lyrical music of the two pairs of lovers; the boozing 
friars in scene 8 might have come from The Three Oranges. But the 
lyrical vein is as unmistakably Prokofyev's here as in War and Peace; 


good examples are the quartet at the end of scene 5 and Clara's aria in 


1 For Prokofyev's own account of the inception of the work, see S. I. Shlifshteyn, S. S. 
Prokofyev: materiali, dokumenti, vospominaniya (Moscow, 1956), p. 112; English version, 
Moscow, n.d., р. 122). On the opera generally, see Nestyev, ор. cit., pp. 389 ff., and 
Rizhkin, Istoriya, iv (1) (Moscow, 1963), pp. 408 ff. 


INSTRUMENTAL MUSIC, 1941-7 681 


scene 7, and when Louisa is plotting with the Duenna in scene 2 
Prokofyev contrives to write a purely diatonic love-theme in D flat 
(Yes, I love so much that more I cannot love’) which yet avoids the 
almost inevitable lusciousness associated with the key. Here again Ше 
melody is cool and classic. 


INSTRUMENTAL MUSIC, 1941-7 

Orchestral music no less than opera reflected the patriotic mood of the 
war-period, sometimes directly, often by emphasis on the national 
music of the non-Russian peoples of the Soviet Union. (The wholesale 
evacuation of composers first to Nalchik in the northern Caucacus, then 
to Tbilisi (Tiflis), naturally heightened their interest in the national 
musics. Thus Myaskovsky's Twenty-second Symphony (1941) was 
styled 'Symphony-ballad of the Great War for the Fatherland’, the first 
of its three connected movements depicting ‘peaceful life into which 
breaks а menace’, while the other two were conceived as ‘Apprehending 
the horrors of war’ and ‘And the enemy faltered’ (a reference to 
Taneyev’s song, Op. 26, no. 8), while his Twenty-third (also 1941) was, 
like Prokofyev’s Second String Quartet, based on Kabardinian themes. 
Indeed one or two are common to both works and the contrasts in 
treatment are striking: 
Ex, 285 
) _ MyasKovsxy 


Piü appassionato (d = 72) 
W.W. Vns. I, Celli -3 
LS GE 


(ii) PROKOFYEV 
Allegro (J = 152) J 


682 MUSIC IN THE SOVIET UNION 


Prokofyev's other compositions of the war and immediate post-war 
period included the ballet Zolushka (Cinderella) (produced 1945), the 
Fifth and Sixth Symphonies (1944 and 1947), and his last three piano 
sonatas. The symphonies and sonatas all belong more or less to the 
musical world of War and Peace. This is particularly true of the ‘heroic’ 
Fifth Symphony, the opening of which might well be part of the 
characterization of Andrey, while the second movement might have 
served for the caricaturing of Napoleon and the D flat *Russian' theme 
in the finale for the glorification of Kutuzov. Equally, the middle 
movement of the Eighth Sonata (also 1944) would not be out of place in 
the ball-scene. Neither the Fifth nor the Sixth Symphony has a pro- 
gramme, but both were written under the impression of the war, present 
or in retrospect. 

The same is certainly true of the Seventh and Eighth Symphonies of 
Shostakovich (1941 and 1943), while his Ninth (1945) presumably 
reflects the high spirits of victory. Only the first movement of the 
Seventh, the so-called Leningrad, is actually programmatic, the develop- 
ment section being replaced by a long-drawn orchestral crescendo in the 
manner of Ravel's Bolero—the major part of the movement—suggesting 
the inexorable advance of the invading armies, while the recapitulation 
brings back the material of the ‘peaceful’ exposition partly in the minor 
and in mournful distortion; the beginning of the recapitulation is 
marked by fff octave declamation by practically the entire orchestra in 
the oratorical manner of the Third Symphony. But although the com- 
poser had originally intended to give titles to all four movements— 
‘War’, ‘Remembrance’, “The Wide Spaces of ош Land’, ‘Victory’+—he 
suppressed them and later gave no more detailed clues than that the 
second movement, with its echoes of Mahlers Ldndler? in the 
‘humorous’ middle section, is “ап intermezzo, very lyrical, gentle—no 
sort of programme, fewer “concrete facts" than in the first movement’, 
and the third, ‘a pathetic adagio with a dramatic middle section’. In the 
finale his intention was to compose ‘an Ode sounding from all the ends 
of the earth and triumphantly growing’, a statement which has a more 
Mahlerian ring than the music itself. Whatever the extra-musical ideas 
underlying the Eighth Symphony, there can be no doubt of its relation- 
ship to the Seventh, even though the relationship seems to be anti- 
thetical. There are even thematic relationships; the very first entry of the 
first violins broodingly takes up the opening of the ‘invasion’ theme of 
the Seventh; the third movement is again concerned with cruel, in- 


1 Shostakovich, ‘О podlinoy i mnimoy programmnosti’, Sovetskaya muzika (1951), 
no. 5, p. 76. 

2 Orlov, op. cit., p. 168, sees a relationship between the first part of the movement and 
an episode in the finale of Das Lied von der Erde. 


INSTRUMENTAL MUSIC, 1941-7 683 


human automata. But there is now no triumphant conclusion. The 
Eighth Symphony is a tragic monument, one of Shostakovich's most 
powerful and individual scores, but so pessimistic and couched in such 
pungent harmonic terms that in 1948 it was, with Prokofyev's Sixth, 
denounced as 'formalistic and both works disappeared from the 
repertory for nearly ten years. 

By comparison with these symphonic giants, Shostakovich's Ninth, 
thrown off in a few weeks in August 1945, is a lightweight: short, scored 
for a smaller orchestra, humorous but not without serious passages (for 
instance, the obviously programmatic introduction to the finale). It is a 
counterpart to Prokofiev's Classical and, like that work, may be 
expected to outlive the Second Symphonies of Khachaturyan, Khren- 
nikov and Popov (all 1943) or Muradeli's, *dedicated to the victory of 
the Soviet people over Fascism' (1945). 

Another significant stage in Shostakovich's career was now marked by 
increasing preoccupation with chamber music. Before the war, he had— 
after some juvenilia—produced only a decidedly eclectic Cello Sonata 
(1934), a rather naive String Quartet, said to be based on memories of 
childhood and youth (1938), and a fine, if suite-like Piano Quintet 
(1940, but published only in 1956). The year 1944 brought a Piano 
Trio and a Second Quartet, which is also rather naive and suite-like, 
consisting of overture, recitative and romance, valse,! and theme with 
variations. But with the Third Quartet (1946) Shostakovich first showed 
his mastery of the medium and embarked on that series of compositions 
which established him not only as the finest Soviet quartet-composer 
but as also the most prolific.? (The ever-fertile Myaskovsky produced 
six more quartets, nos. 7-12, during the period 1941-7.) Two other 
Soviet string quartets of this period stand out: Kabalevsky's Second 
(1945) and Shebalin's Fifth (1942). Kabalevsky's Quartet, like his 
Second Piano Sonata (written in the same year), is dramatic and 
thematically related to his operas V ogne and Semya Tarasa (Taras's 
family). Shebalin's, the Slavonic, is based on Russian and Ukrainian 
folk-themes in its first and fifth movements, Polish, Slovak and 
Serbian tunes in the middle ones. (The Serbian theme of the fourth 
movement is that on which Tchaikovsky based his S/avonic March.) 
Shebalin further exploited the Russian folk-song vein with fine crafts- 
manship in his Seventh Quartet (1948) and the variation-finale of 
his Piano Trio (1949). Beside these established masters, a newcomer 
appeared in this field: Shostakovich's pupil Yury Sviridov (b. 1915). 


1 Recorded іп The History of Music in Sound, x. 
? See L. Raaben, Sovetskaya kamerno-instrumentalnaya muzika (Moscow, 1963) and 
Colin Mason, ‘Form in Shostakovich's Quartets’, Musical Times, сш (1962), p. 531, 


684 MUSIC IN THE SOVIET UNION 


Sviridov's two quartets, his Piano Quintet, and his Piano Trio all date 
from 1945-6, although the Trio and Quintet were thoroughly revised 
ten years later; as one might expect, they show the influence of his 
teacher's Trio and Third Quartet—in more ways than their suite-like 
structure—but also individual traits, indeed more of the latter than in 
some of Sviridov's later, more highly praised works. 


VOCAL MUSIC, 1941-7 

The solo song was cultivated during these years more than might have 
been anticipated. Naturally heroic, declamatory elements are prominent, 
as in Anatoly Aleksandrov's cycle Tri kubka (Three goblets) (1942) and 
Nechaev's O doblestyakh, o podvigakh, o slave (Of heroism, great deeds, 
and fame) (1943). The nineteenth-century form of the narrative ballad 
was also revived in art-song as well as mass-song; for instance, the 
heroism of the airman Gastello was celebrated by both Viktor Bely 
(‘Ballada o kapitane Gastello’) and Khachaturyan (‘Kapitan Gastello’). 
The wartime alliance also evoked not only a great many arrangements 
of English and Scottish folk-songs but also numerous original settings 
of English and Scottish verse. Shostakovich led the way with his six 
bass songs, op. 62 (1942), to translations of Shakespeare, Walter 
Raleigh, and Burns, originally with piano, later with orchestral accom- 
paniment; his attempt at Sonnet 66 must be one of the worst Shakes- 
peare settings in existence, but he was more successful with Burns. 
Indeed Burns (in Marshak's generally good translations) became a 
favourite with other composers, including Khrennikov. But the Russian 
classic lyric poets—Pushkin, Fet, Tyutchev—were not neglected, and 
Shebalin composed a Heine cycle. 

Yet the most remarkable solo vocal composition of the war years was 
wordless: the Concerto for voice (coloratura soprano) and orchestra 
(1943) by the veteran Glier. It is in two movements, a lyrical Andante and 
a brilliant final valse or jota. The idiom is that of half a century earlier 
but it is a true concerto, highly effective, and it has easily maintained a 
place in the Russian concert repertoire. 

Of the larger works for chorus and orchestra, the patriotic cantatas 
and so-called oratorios, Kabalevsky's Rodina velikaya (The great 
motherland) (1942), Myaskovsky's Kirov s nami (Kirov is with us) 
(1943) and Shaporin's Skazanie o bitve za Russkuyu zemlyu (Story of the 
fight for the Russian land) (1944)! are perhaps the best. They are 
rhythmically square-cut, diatonic, with a preponderance of block 
common chords: official art carried out with the technical competence 


1 On these and other similar works, see Tumanina in Istoriya, iii, pp. 137 ff. 


POLITICAL BACKGROUND AFTER 1947 685 


and occasional flashes of real invention one expects of these composers, 
but even Shaporin's Skazanie is far inferior to The Field of Kulikovo. 


POLITICAL BACKGROUND AFTER 1947 


While the works just mentioned, and many others, conformed 
absolutely to the strictest demands of socialist realism, there were 
some which did not. During the war the Soviet Government and the 
Central Committee of the Communist Party had not been paying close 
attention to literary and artistic orthodoxy and, with control thus 
relaxed, writers and artists had insensibly begun to take liberties; the 
old evils of ‘formalism’ and 'subjectivism' were creeping back. As early 
as August 1946 the Central Committee had begun to worry about 
literature and journalism, about the theatrical repertory, and about 
films; and writers were sharply reminded that ‘there are not and cannot 
be other interests than the interests of the people, of the state... . Hence 
all preaching of that which has no idea-content, of the apolitical, of 
*art for art's sake", is foreign to Soviet literature, harmful to the 
interests of the Soviet people and state, and must have no place in our 
journals’. Writers of the rank of Akhmatova and Zoshchenko were 
coarsely abused in public speeches by the Committee's spokesman, 
Andrey Zhdanov, and the Union of Composers did not fail to notice 
the danger-signals.! But more than a year passed before the Committee 
turned its attention to music. Just as in 1936, the immediate occasion 
was a specific opera: this time Velikaya druzhba (The great friendship) 
by the Georgian-born Vano Muradeli (1908-70), produced in Moscow 
on 7 November 1947, to mark the thirtieth anniversary of the Revolu- 
tion. The highest members of the Government and Party were present 
and reacted precisely as they had done to The Lady Macbeth. Little 
more than a month later (25 December) the performance of Prokofyev's 
Sixth Symphony in Moscow made matters much worse. (It had been 
acclaimed in Leningrad in October.) In January Zhdanov sternly 
addressed a three-day conference of musicians in Moscow, from which 
Myaskovsky absented himself? and on 10 February the Central Com- 
mittee published its ordinance “Оп the opera ТЛе Great Friendship by 
V. Muradeli',? which not only denounced Muradeli's work in terms 
practically identical with those applied (with much less injustice) to The 


1 See the unsigned article, "Problemi sovetskovo muzikalnovo tvorchestva’, Sovetskaya 
muzika (1946), no. 8-9, p. 3, and the account of the week-long plenary meeting of the 
organizing committee of the Union of Soviet Composers, ibid., no. 10, p. 3. 

2 For a copious, though not quite complete, report of the conference, see Alexander 
Werth, Musical Uproar in Moscow (London, 1949), pp. 47 ff. 

3 Full translation of the resolution, Zhdanov's speech, and various statements by com- 
posers in Slonimsky, op. cit., pp. 684 ff.; long excerpts, with commentary, in Werth, 
Op. сії., pp. 28 ff. 

45 


686 MUSIC IN THE SOVIET UNION 


Lady Macbeth and Clear Stream twelve years before but also attacked as 
‘anti-national formalists’ the far more important composers Shostako- 
vich, Prokofyev, Khachaturyan, Shebalin, Popov, and Myaskovsky. 
From the composers' side, the official view was endorsed most en- 
thusiastically by Khrennikov; some—notably Shebalin— defended 
themselves with spirit or remained silent; the majority, including 
Muradeli and Shostakovich, abased themselves and confessed their 
errors. 

The so-called ‘Zhdanov period’ lasted in its full rigour until the death 
of Stalin—on the same day as Prokofyev—in March 1953. A slow ‘thaw’ 
then set in, officially acknowledged in 1956, though it was not until 
18 May 1958 that the Central Committee published another resolution 
‘correcting errors of evaluation’ in the ordinance of 1948, for which 
‘J. V. Stalin’s subjective approach to works of art’ was blamed. Never- 
theless this was no charter of liberty; those who had misread it in that 
sense were put right by Khrushchev in March 1963. 


OPERA: 1948-60 

Muradeli’s The Great Friendship is a weak opera! and the accusations 
of lack of memorable melody or true local colour were justified, but the 
charges of ‘chaos’ and ‘continuous discord’ against so innocuous a 
score were ludicrous. It contains no harmony more pungent than such 
passages as this: 


Ex. 286 
Moderato 


THE 
COMMISSAR [— —Zi 


1 See Rizhkin, Jstoriya, iv (1), pp. 327 ff. 


OPERA: 1948-60 687 


Ко slu-zhitdnyom i no - chyu, 


| 


SSS АЕ таш 2 | Ы, 


^ PLN Г\ ^ 
[fe [uet pem Шш Dp шамы с 1 


(the ploughman who bends over the plough, and the shepherd who works day 
and night, the miner who all day follows his dark road underground,) 


and a very great deal that is less. Perhaps the music was really considered 
less objectionable than the libretto, which purported to show the 
reconciling by a Political Commissar of (allegedly non-existent) enmity 
between the Russians and Lesgians in the Northern Caucasus but con- 
centrated instead on a romantic-melodramatic plot which would serve 
equally well for a ‘Western’ film. It is historically important only for the 
deathly hush that followed its condemnation. Prokofyev completed his 


688 MUSIC IN THE SOVIET UNION 


Povest o nastoyashchem cheloveke (The story of a real man) but it had 
only one private concert performance in December 1948; it was not 
produced till 1960, seven years after his death. Even the Semya Tarasa 
of the more conservative and ‘correct’ Kabalevsky, which had been 
tried out in Moscow in November 1947, was drastically rewritten 
and produced in its revised form in 1950.! (Both these operas are 
on war-subjects; Kabalevsky's depicting the patriotism of a worker's 
family in a part of the Ukraine overrun by the Germans, Prokofyev's 
the heroic will-power of an airman who overcomes the loss of his feet.) 
In the operatic silence were heard only such small, quiet voices as that of 
the Odessa-born Antonio Spadavecchia (Ъ. 1907), а former composition- 
pupil of Shebalin and Prokofyev, with his Khozyayka gostinitsi (The 
Hostess at the Inn) (based on Goldoni’s Locandiera) (1949), and Ше 
official one of Khrennikov with his uninspired comedy Frol Skobeyev 
(1950). 

By far the most notable opera of the nineteen-fifties was Shaporin's 
Dekabristi (The Decembrists), produced on 23 June 1953 after more 
than a quarter of a century of gestation. The original conception had 
been the personal drama of the young nobleman Annenkov, who has 
brought back from the West liberal ideas and a French shopgirl, 
Pauline, whom he is determined to marry—-despite his mother's violent 
opposition. He is drawn into the Decembrist conspiracy and sentenced 
to exile; at a masked ball Pauline was to plead in vain with the Tsar for 
her husband's pardon and resolve to accompany him to Siberia. The 
political events provided no more than a background except in two 
scenes, the equivalents of the definitive fourth and sixth: the meeting on 
the eve of the rising (a scene in which Annenkov's part was insignificant) 
and the great spectacular scene in the Senate Square when the rising 
collapsed. In this form the opera was nearly completed in 1938, with 
only the Senate Square scene and the ball scene unwritten.? In the final 
version Annenkov was supplanted by Shchepin-Rostovsky and Pauline 
by Elena Orlova, the daughter of an impoverished neighbouring land- 
owner. The personal drama remained the same, though the lovers lost 
some beautiful music, but it was now pushed into the background of the 
historical events. (Spadavecchia's Khozhdenie po mukam (1954), based 
on А. М. Tolstoy's well-known novel of the Revolution, translated 
under the title Darkness and Dawn, underwent a precisely similar 
change.) The hero of the final version is no longer Annenkov/Shchepin 
but the Decembrists collectively and, while it would be quite untrue to 


! Tbid., pp. 283 ff. 
2 See the account, with lengthy musical examples, by А. Lepin, ‘Dekabristi—opera Y. 
Shaporina’, Sovetskaya muzika (1938), no. 7, p. 20. 


OPERA: 1948-60 68 


© 


say that the Decembrist leaders are not differentiated, they are certainly 
more important as a group than as individuals. Instead of a lyrical- 
dramatic opera, Тре Decembrists thus became a series of historic 
tableaux in the tradition of Prince Igor. (One would say ‘of Boris 
Godunov’, if only there were one central character, hero or villain, 
studied in depth, but neither Rileyev nor Pestel, nor Nicholas I, is so 
studied.) Elena and Shchepin become hardly more important than 
Musorgsky's Marina and Pretender-as-lover. Yet, granted this 
dramatic weakness and the conservative nature of Shaporin's musical 
idiom, the score contains far too many memorable things to be dis- 
missed as mere epigonism: the beautiful folk-song chorus that opens the 
first and last scenes, the arioso in which Shchepin tells his mother of 
his love (Russian romantic lyricism at its best), Bestuzhev's song of the 
winter road that ends scene 2, the polyphony of folk-songs sung by 
four choral groups in the fair-scene: 


Ex. 287 
Allegro ma non troppo 


ЗБС ат ЕТ ат 
(отпадне ОР ee [eee eee = 


E 


[= зе — — ov mcn] 

2 

rl 

Sra ad 2o aa 
chorus of gypsies 553 а ЕА 


[2 
First chorus Б А6 oep 
[pc Lae) ee), el -Б------і 


Ohpedsants) гс т сс] 


- ra ved - га. Ы 
СШ. =”. пре 
лена ЕЕ 
Da, p - mi = n -~ di 
аи e ту zr Fes 
ЕЕ = 7 ята 


MUSIC IN THE SOVIET UNION 


690 


LT LN m ( йй HN | Ili 


cM | 
сі 
ШЕН 


sho, kho-ro - 


- ka, chtob 
= = а 
== Е-е 0] 
| — = м 
[=== — = NNNM 
I] [Г — T cmn 
EE 
Ee 


% 
ОШ? 


lo - shad - - 
= ara ee] 
ae Sa. ae, БЕЗДЕР 
нардан жән 
Sa a ы Ой Лг 
- ИЕ НЫ] 
kho-ro - 


Гос ae) 
а ғын) 
| ан м 
SS иа) 
Ep ER 
p - vo 

ее 

м 


mo уа. 


ms (Th 1 
Еее ак пат 
plor Io Wa Дена 


DUE ROEONGEGUE ERG қоза” зару 1 
sung ical 
vo tsa = 


йы: Ч | 


OPERA: 1948-60 691 


jeu cH 


Gl ЕШ 


ges SSS 


Oy, do-rozh - ka, ar moy dal-niy, ve 


ЕЕ ЕЕ: (== 


- sho! Vi-no  lush-(che evo!) 


о 3 
б = SS Е-е 
(jt лид. ди БЕЛ s SET С р iz ur elem 
н ) =т= == 
Е Б zz == 
6) 


(Girls: Fly, my horse, so the wind doesn't catch you! Stesha: Hoi, little road, 
my distant мау... Chorus I: Hey, beer is good! Wine's better! Chorus II: Yes, they 
mentioned in the pot-house!) 


the broad, massive handling of the spectacular scene in the Senate 
Square, the waltz (worthy of Tchaikovsky) to which the Emperor 
pursues Elena at the ball. No other Soviet opera comes so near to 
challenging War and Peace on its own ground. 

The diatonic melody and harmony which came naturally to Shaporin 
were cultivated deliberately, in conformity with the Party directive, but 
with much less technical skill, by the younger men: Dzerzhinsky in 
Daleko ot Moskvi (Far from Moscow) (1954) and Groza (Ostrovsky's 
Storm) (1956), Kabalevsky in Nikita Vershinin (1955), Kirill Molchanov 
(b. 1922) in Kamenniy tsvetok (The Stone Flower) (1954, and therefore 
after the composition but before the production of Prokofyev's ballet on 
the same subject), Zarya (The Dawn) (1956), and Ulitsa del Korno 
(Del Corno Street) (1959), and Khrennikov in Mat (The Mother, after 
Gorky) (1956), Even the best Soviet composers failed in the struggle for 
unnatural simplicity. Shostakovich’s ‘musical comedy’ with spoken 
dialogue, Moskva, Cheremushki (1958), is banal beyond belief, and 
Prokofyev’s Story of a Real Man, when it was produced at last post- 
humously in 1960, only showed that his avowed striving for ‘clear 


692 MUSIC IN THE SOVIET UNION 


melodies and the simplest harmonic language possible'! could some- 
times be fatal in his case, too. Only. here and there for a few bars, as in 
no. 14 when old Mikhaylo tells how the maimed Aleksey has dragged 
himself along: 


Ex. 288 
Allegro Е 


ЕЕЕ 
| 4 Жет хасан 


MIKHAYLO 08-456 


B MU MEM 
ka - tit- sya, go- vo-ryat re - bya - ta. 


# SS ee 2 
eo ЕЕ---- 
Sr й 


(And so it goes, so it goes, as children say.) 


is there any hint of the old Prokofyev. Indeed there is more of ‘the old 
Prokofyev’ in Shebalin's often witty Ukroshchenie stroptivoy (The 
Taming of the Shrew) (1957). 


INSTRUMENTAL MUSIC: 1948-60 


The Party resolution of 1948 had similar effects in the field of instru- 
mental music; works that might have incurred the charge of formalism 
were laid aside and composers cultivated simple diatonic melody and 
harmony. Major instrumental composition, so liable to formalistic 
lapses, was indeed largely superseded for a time by the writing of 
cantatas and ‘oratorios’ with ideologically irreproachable texts. 


1 Quoted by Nestyev, op. cit., p. 403. 


INSTRUMENTAL MUSIC: 1948-60 693 


Shostakovich temporarily suppressed his Violin Concerto (com- 
posed in 1948 as Op. 77) and his Fourth Quartet (1949), and com- 
posed a banal ‘oratorio’, Pesn o lesakh (Song of the Forests) (1949) and 
Twenty-Four Preludes and Fugues for piano (1950-1) in emulation of 
Bach. Another field much cultivated during this period of enforced 
innocence was that of music for children—not so much for children to 
perform as for children to listen to: Kabalevsky's concertos for violin, 
cello and piano, Opp. 48-50 (1948-9), Prokofyev's Zimniy koster 
(Winter Bonfire) for readers, chorus of boys, and orchestra (1949), and 
Shostakovich's Concertino for two pianos (1953). Even Prokofyev's 
Seventh Symphony (1952), a lyrical and rather weak work, was originally 
conceived as а ‘symphony for children’ and Shostakovich's Second 
Piano Concerto (1957) were products of the same impulse. Not un- 
naturally, the stronger creative talents show to less advantage within 
such limitations than composers whose more modest powers needed 
less restraint. Kabalevsky's Violin Concerto is a charming little work 
with an andantino melody in the middle movement worthy of 
Tchaikovsky himself: 


Ex. 289 


Andantino cantabile 


whereas the corresponding theme in Shostakovich's Second Concerto 
sounds like Baroque pastiche: 


Ex. 290 
Andante 


1 See S. Skrebkov, ‘Prelyudii i fugi D. Shostakovicha’ Sovetskaya muzika (1953), по. 9, p. 
18, and V. Protopopov, Istoriya, iv (2), (Moscow, 1963), pp. 190 ff. 


694 MUSIC IN THE SOVIET UNION 


The second subject of Kabalevsky's finale has an appropriate affinity 
with the song of Nastya and the Komsomols in Act I of Semya Tarasa: 


Ex. 291 
Moderato 


i 270 
МАЅТҮА и БЕ 


Sdru- zya-mi ras-sta- va - lis mi, bit mo- zhet na-vseg- 


Vivace giocoso 


VIOLIN 
CONCERTO 


- da. Pro-shchay-te, dru - zya, u- kho- dit po - ra! 


-— ет. 


(We part with friends, perhaps for ever. Farewell, friends, it's time to leave!) 


Stalin was no sooner dead than Shostakovich embarked on a Tenth 
Symphony, written during the summer of 1953 and performed in 
December. Judged by purely musical criteria, it is arguably the finest of 
all his symphonies, stamped from beginning to end with his creative 
personality and free both from the cheaper forms of grotesque humour 
which intrude so often in his best scores and from concessions to official 
conservatism. Such a work was bound to attract the charge of ‘formal- 
ism', and few new symphonies can have been debated so fiercely and 
immediately by the composer's colleagues as this.! The Symphony 
manifestly has no quasi-literary programme but it is equally certainly 
not abstract music; it is not so much ‘formalistic’ as ‘subjective’ and 
‘pessimistic’, which, from the Marxist point of view, is just as bad. That 


! See Orlov, op. cit., pp. 249 ff., for particulars and references. 


INSTRUMENTAL MUSIC: 1948-60 695 


much of the Symphony is in some sense tragic is hardly deniable; the 
problems that agitated Shostakovich's colleagues were whether the 
tragedy was personal or national (still the aftermath of the war) and 
whether the tragedy might be regarded as ‘overcome’ in Ше end.! In the 
perspective of ten years, the authors of the official History of Russian 
Soviet Music? were able to sum up in 1963: 


The tragic element in Shostakovich, particularly in his later works, is 
optimistic in tendency. He seeks a way to overcome a tragic beginning. His 
humanistic ideals as a Soviet artist, loving life and freedom, fighting for peace 
and happiness, cannot fail to lead him to life-affirmation. It is possible to 
argue about this or that degree of artistic conviction and about the forms in 
which the tragic is overcome in Shostakovich's works—arguments which 
still continue in musical criticism—but there can be no doubt of the general 
effort to overcome tragedy. 


‘Optimistic tragedy’ is a phrase commonly met with in Soviet criticism; 
both Shaporin's Decembrists and Kabalevsky's Тагазз Family are 
*optimistic tragedies'. But Shostakovich was taking no chances with his 
next two symphonies, No. 11 (1957) and No. 12 (1961), which are both 
programmatic—with programmes beyond political reproach—and, 
particularly No. 12, written in a musical idiom comprehensible to mass- 
audiences. Each is dedicated to a revolution. The four movements of 
No. 11 (entitled The Year 1905) are headed “The Palace Square’ (i.e. in 
Petersburg), ‘9 January’, ‘Eternal Memory’, and ‘Tocsin’, and 
Shostakovich has based most of the score on revolutionary tunes of the 
period,? not simply quoting and varying them, but sometimes even fus- 
ing them into fresh entities. Thus the melodic line of the horns just 
before fig. 107 in the third movement is derived partly from the song 
‘Baykal’, partly from ‘Boldly, comrades, keep in step’. The funeral- 
march third movement is based mainly on the song, ‘Vi zhertvoyu pali 
(You fell as a sacrifice), which was solemnly sung by Lenin and his 
companions in exile when they heard the news of ‘Bloody Sunday’,* and 
the title of the finale almost certainly refers to his article ‘Revolution in 
Russia’ in the émigré journal Vpered on 11 January 1905. Мо. 12 is a 
companionpiece, The Year 1917, with movements entitled ‘Revolution- 
ary Petrograd’, *Razliv (where Lenin lay low before the October 
Revolution), *The Aurora' (the cruiser which dealt the decisive blow to 
the Provisional Government), and “Тһе Dawn of Humanity. The 

1 бее, for instance, Andrey Volkonsky, 'Optimisticheskaya tragediya’, Sovetskaya 
muzika (1954), no. 4, p. 25. 

2 Istoriya, iv (2), p. 145. 

3See Orlov, op. cit, рр. 286 ff, and L. Lebedinsky, ‘Revolyutsionniy folklor v 
Odinnadtsatoy simfonii D. Shostakovicha', Sovetskaya muzika (1958), no. 1, p. 42. 


5 М. К. Krupskaya, Vospominaniya о Lenine (Moscow), 1957, p. 89 (quoted by Orlov, 
ор. cit., р. 303). 


696 MUSIC IN THE SOVIET UNION 


Thirteenth Symphony (1962), on poems by Evtushenko, is less a 
symphony than a cantata for baritone, male chorus and orchestra, 
while the Fourteenth (1969), settings of Lorca, Guillaume Apollinaire, 
Küchelbecker and Rilke for soprano, bass and orchestra, is no more— 
but no less—a symphony than Mahler's Lied von der Erde. 

But the best of Shostakovich at this time was put, not into the 
symphonies (with the exception of the Tenth) but into the Cello 
Concerto (1959)—inspired, like Prokofyev's Sinfonia concertante for 
cello and orchestra (1952, but a reworking of material from the Cello 
Concerto of 1938), by the playing of Mstislav Rostropovich—and the 
string quartets from: No. 4 onward. The Fourth Quartet itself was 
written soon after the outstanding song-cycle From Hebrew Folk Poetry 
(1948) and both are related, either thematically or emotionally, to the 
Piano Trio of 1944, the finale of which is said to have been inspired by 
the discovery of the Jewish death-camp at Majdanek. Rabinovich says! 
that both the Fourth and Fifth Quartets were ‘composed in the same 
atmosphere of thoughts and emotions in which the future Tenth 
Symphony took form’ and ‘perhaps, either directly or indirectly, pre- 
pared the way for it'. They are even related thematically, and stamped 
as essentially ‘personal’ works, by the appearance іп all three, in various 
permutations and transpositions, of the notes О-Ер-С-ВҢ representing 
the composer's initials *D.Sch. (Sch being the German translitera- 
tion of the single Russian letter which begins his surname) a motive 
that was to appear again in the concertos for violin and cello, and in 
other works. One can trace no similar connexion between the later 
quartets and symphonies. The Eleventh and Twelfth Symphonies are 
public utterances; the quartets are intensely personal and subtle in 
workmanship. This is particularly true of No. 8 (1960), with its 
numerous references to the ‘D. Sch.’ theme, prominent from the very 
beginning, and quotations from the composer's earlier works?—the 
opening of the First Symphony near the very beginning, and the 
second subject of the finale of the Piano Trio in the second movement 
(p. 9 of the miniature score)—as well as the revolutionary song 
‘Zamuchen tyazheloy nevoley' (‘Worn out by heavy slavery’) and а 
passage from The Lady Macbeth in the middle of the fourth movement. 

The other older composers continued to produce in their familiar 
veins. Kabalevsky’s Fourth Symphony (1956), Shebalin’s Seventh, 
Eighth and Ninth Quartets (1948, 1961 and 1963) and Fifth Symphony 
(1962), Knipper’s Symphonies 10-14 (1946-54), Aleksandrov’s Piano 
Sonatas 9-11 (1946-55) are all well-written works, with fine passages 


! D, Rabinovich, Dmitry Shostakovich (London, 1959), p. 129. 
? Keldish, “Ап Autobiographical Quartet’, Musical Times, cii (1961), p. 226. 


INSTRUMENTAL MUSIC: 1948-60 697 


(e.g. the Andante of Shebalin's Seventh Quartet), which neverthe- 
less add nothing essential to the composers' images. Khachaturyan's 
colourful ballet Spartak (Spartacus) (1954; produced 1956) is no 
more than a counterpart of Gayane. Nor can it be said that the 
generation born just before or not long after the Revolution brought 
forth anything startlingly new. There have been some very success- 
ful works: Rodion Shchedrin's ballet Konek-Gorbunok (The little 
humpbacked horse) (1955), the ‘oratorios’, Pamyati Sergeya Esenina 
(In memory of Esenin) (1955) and Pateticheskaya oratoriya (Pathetic 
Oratorio) (1959) by Sviridov, and Dvenadtsat (Blok's ‘The Twelve’) 
(1957) by Vadim Salmanov (b. 1912). But neither in these nor in 
the orchestral works of Andrey Eshpay (b. 1925) and Boris Chaykovsky 
(b. 1925)—to name the best composers of their generation—is there 
much, if anything, that could not have been written half-a-century, 
even a century, earlier. Consider, for instance, the theme of the 
storm-wind which not only opens Salmanov's The Twelve but plays an 
important part throughout: 


Chorus 


Orch. 


698 MUSIC IN THE SOVIET UNION 


ре —— — —] 

Ин ПЕН Е 7 ING ПЕСНЕЙ с е2 

D. — — —— Iesum ES es cos 
[ C я 00 Он 


(Black evening, White snow.) 


However, Salmanov has since been influenced by Bartók and his Third 
String Quartet (1960) 15 “ап attempt to fit certain twelve-tone techniques 
into а modal framework.! Another composer who showed dangerously 
modern tendencies was Galina Ustvolskaya (b. 1919), a Shostakovich 
pupil who has written a number of sonatas for piano and various 
instruments. 


1 Boris Schwarz, ‘Soviet Music since the Second World War’, Musical Quarterly, li (1965), 
р. 277. 


THE AVANT-GARDE 699 
THE AVANT GARDE 


А genuine avant-garde, employing twelve-note and aleatory tech- 
niques, emerged when the composers born during the 1930s came to 
maturity thirty years later. They even managed to get a respectful 
hearing for some of their work. When Andrey Volkonsky (b. 1933) 
first performed in public his twelve-note piano suite Musica Stricta 
(1957) it was dismissed as ‘fashionable experiment and nothing тоге”! 
but his later work won toleration though not official approval. Born 
in Geneva of émigré parents, Volkonsky had studied as a child 
with Nadia Boulanger; but he made his home in Russia and com- 
pleted his studies in Moscow under Shaporin. He first attracted 
attention with a setting of Paul Eluard's Poémes pour la paix for mezzo- 
soprano, chorus, organ, and orchestra (1952) and a Piano Quintet 
(1955). By 1962 he was writing a Game in Three for violin, flute, and 
harpsichord, suggested by Stockhausen's Kreuzspiel. Aleatory works 
by other composers followed: for instance, the Dialogi for wind quintet 
(1964) by Sergey Slonimsky (b. 1932), which employ both aleatory and 
serial techniques, and the Crescendo e diminuendo for twelve strings and 
kettledrum (1965) of Edison Denisov (b. 1929). A more important 
work of Denisov's is Soleil des Incas, a cantata for soprano and eleven 
solo instruments on a text by Gabriela Mistral (1964). Other leading 
composers of this avant-garde, whose works are publicly performed and 
seriously discussed, are Boris Tishchenko, a pupil of Ustvolskaya, 
Salmonov, and Shostakovich, Valentin Silvestrov (b. 1937), and two 
Estonians, Arvo Раагї (b. 1935) and Kuldar Sink (b. 1942). Sink has 
experimented with both electronic music and musique concréte. 


MUSIC IN THE NON-RUSSIAN REPUBLICS 


The interest taken by Russian musicians such as Glier and Ippolitov- 
Ivanov (in the nineteen-twenties) and Myaskovsky and Prokofyev 
(during the War) in the music of the Caucasian peoples of the Soviet 
Union has already been mentioned. Such interest was not in itself new, 
of course; but it was of a different nature from the interest of nineteenth- 
century tourists like Balakirev and Taneyev.? The Soviet Government 
from the very early days attached great importance to the stimulation 
of the native arts in the Asiatic republics of the Union, and Russian 


musicians and musicologists were sent to study their folk-musics? and 


1 Vaclav Kucera, Nové proudy v sovetské hudbé (Prague, 1967), p. 19. 

? Taneyev’s study of Kabardinian folk-music, ‘О muzike gorskikh Tatar’, is printed in 
V. Protopopov (ed.), Pamyati 5.1. Taneeva: 1856-1946 (Moscow, 1947), р. 195, with 
Belyaev's commentary, ibid., p. 212. 

3 The results have been published in such monographs as Viktor Uspensky and Viktor 
Belyaev, Turkmenskaya muzika (Moscow, 1928), and Belyaev, Ocherki po istorii muziki 
narodov SSSR, i and ii (Moscow, 1962 and 1963). 


700 MUSIC IN THE SOVIET UNION 


foster the development of more ambitious musical forms on the basis of 
folk-art. Such works as Glier's operas Shakh-Senem (Baku, 1927), 
Leyli i Medzhnun (Tashkent, 1940) and Gyulsara (Tashkent, 1949) —the 
last two in collaboration with the Uzbek composer, Talib Sadikov 
(b. 1907)—and Knipper's orchestral suites Vanch (on Tadzhik themes) 
(1937), Turkmenskie eskizi (Turkmenian sketches) (1940), and Kurumkan 
(on Buryat-Mongolian themes) (1948), are essentially different from 
nineteenth-century Russian essays in orientalism; they are attempts to 
provide models for authentic Caucasian or Central Asian art-music. 
Conversely Caucasian composers—the Armenian Khachaturyan is the 
outstanding example—have settled in Russia and, without sacrificing 
their native accents, made notable contributions to Russian music. 
Others, such as the Georgian, Andrey Balanchivadze (b. 1906), himself 
the son of a composer (Meliton Balanchivadze, 1862-1937), the 
Azerbaidzhanian Kara Karaev (b. 1918), a pupil of Shostakovich, the 
Armenian Eduard Mirzoyan (b. 1921) and the Georgian Sulkhan 
Tsintsadze (b. 1925), have preferred to stay in their native lands. The 
piano concertos of Balanchivadze and the string quartets of Tsintsadze 
have won popularity all over the U.S.S.R. 

Even the Ukraine has produced composers who have followed 
Mikola Lisenko (1842-1912) in refusing to allow themselves to be 
swallowed by Great Russian culture—notably Boris Lyatoshinsky 
(b. 1895), Mikhail Verikovsky (b. 1896), and Gleb Taranov (b. 1905). 
But Ukrainian musical nationalism is perhaps a dying cause. 


BIBLIOGRAPHY 


The bibliography for Chapters III and V is by John Arnn, and for Chapter VI by 
Lewis Foreman 


GENERAL 


ABRAHAM, GERALD: 4 Hundred Years of Music (London, 3rd ed., 1964). 

ADORNO, T. W.: Philosophie der Neuen Musik (Tübingen, 1949). 

AUSTIN, W. W.: Music in the Twentieth Century (New York, 1966). 

BAUER, MARIAN: Twentieth-Century Music (New York, 1933). 

CALVOCOREZSSI, M. D.: Musicians’ Gallery (London, 1933). 

CARNER, Mosco: A Study of Twentieth-Century Harmony (London, 1942). 

COOPER, MARTIN: French Music (From the death of Berlioz to the death of Fauré) 
(London, 1951). 

Совтот, ALFRED: La Musique francaise de piano (Paris, three vols., 1930, 1932 and 
1944). 

DUNWELL, WILFRID: The Evolution of Twentieth-Century Harmony (London, 
1960). 

Dyson, GEORGE: The New Music (London, 1924). 

HARTOG, EDWARD (ed.): European Music in the Twentieth Century (2nd ed., London, 
1961). 

Lissa, ZóriA: ‘Geschichtliche Vorform der Zwólftontechnik', Acta Musicologica, 
vii (1935). 

MELLERS, WILFRID: Caliban Reborn: Renewal in Twentieth-Century Music (London, 
1968). 

— Studies in Contemporary Music (London, 1947). 

MERSMANN, Hans: Die moderne Musik seit der Romantik (Potsdam, 1931). 

Musik der Gegenwart (Berlin, 1923). 

MITCHELL, DONALD: The Language of Modern Music (London, 1963). 

Myers, Кото: Modern French Music (Oxford, 1971). 

—Twentieth-Century Music (2nd ed., London, 1968). 

ROHOZINSKI, L. (ed.): Cinquante ans de musique francaise (Paris, 1925). 

SCHOENBERG, ARNOLD: Style and Idea (New York, 1950). 

SLONIMSKY, NICOLAS: Music since 1900 (New York, 4th ed., 1972). 

SOUVTCHINSKY, PIERRE: Musique russe (Paris, 1953). 

STEIN, ERWIN: Orpheus in New Guises (London, 1953). 

STEPHAN, RUDOLF: Neue Musik (Gottingen, 1958). 

STUCKENSCHMIDT, H. H.: Twentieth-Century Music (London, 1969). 

Twentieth-Century Composers. In 5 vols.: 

Vol. 1 American Composers since 1910. By Virgil Thompson. 

—— Vol. 2 Germany and Central Europe. By Н. Н. Stuckenschmidt (1970). 

WEBERN, ANTON: Wege zur neuen Musik (Vienna, 1960). 

WELLESZ, Есом: Die neue Instrumentation (Berlin, two vols., 1928 and 1929). 

WoOrNER, KARL H.: Musik der Gegenwart (Mainz, 1949). 

— —Neue Musik in der Entscheidung (Mainz, 1954). 
46 


702 BIBLIOGRAPHY 
CHAPTERS I AND II 


THE APOGEE AND DECLINE OF ROMANTICISM and THE REACTION AGAINST 
ROMANTICISM: 1890-1914 


(1) General 

Вовомі, Ferruccio: Von der Einheit der Musik (Berlin, 1922; English translation, 
London, 1957). 

LANDORMY, PAUL: La Musique française de Franck à Debussy (Paris, 1943). 

LENORMAND, RENÉ: Etude sur l'harmonie moderne (Paris, 1913; English translation, 
London, 1915). 

ROLLAND, ROMAIN: Musiciens d'aujourd'hui (Paris, 1908). 


(1) Individual Composers (books and articles dealing with their non-operatic works 
of the period 1890—1914) 


Bartok 

‘Bartok’ numbers of Musikblatter des Anbruch (1921) and Tempo (1949-50). 

Mason, Coun: *Bartók's Rhapsodies’, Music and Letters, xxx (1949). 

—— — Bartók's Early Violin Concerto’, Tempo, 49 (1958). 

—‘Bartok’s Scherzo for Piano and Orchestra’, Tempo, 65 (1963). 

Мшл, EDWIN von DER: Béla Bartók: Ein Beitrag zur Morphologie der neuen 
Musik (Halle, 1930). 

STEVENS, HALSEY: The Life and Music of Béla Bartók (New York, 2nd ed., 1964). 

— ‘Some “Unknown” Works of Bartók’, Musical Quarterly, lii (1966). 

Uirarussv, 7075ЕЕ: Bartók (Budapest, 1965; English translation, Budapest, 1971). 


Berg | 

CHADWICK, NICHOLAS: *Berg's Unpublished Songs in the Ósterreichische National- 
bibliothek’, Music and Letters, lii (1971). 

LEIBOWITZ, RENÉ: ‘Alban Berg's Five Orchestral Songs, Ор. 4, Musical Quarterly, 
xxxiv (1948). 

REDLICH, HANs F.: Alban Berg: Versuch einer Würdigung (Vienna, 1957; English 
condensation, London, 1957). 

Reicu, Милл: Alban Berg (Vienna, 1937). 

——Alban Berg: Leben und Werk (Zürich, 1963; English translation, London, 
1965). 

STUCKENSCHMIDT, Н. H.: ‘Debussy or Berg? The Mystery of a Chord Progression’, 
Musical Quarterly, li (1965). 


Busoni 

‘Busoni’ numbers of Musikblatter des Anbruch (1921) and Rassegna musicale (1940). 

DEBUSMANN, Еми.: Ferruccio Busoni (Wiesbaden, 1949). 

DENT, EDWARD J.: Ferruccio Busoni: a Biography (Oxford, 1933). 

LEICHTENTRITT, HuGo: ‘Ferruccio Busoni’, Music Review, vi (1945). 

Зику, LARRY: ‘The Six Sonatinas for Piano of Ferruccio Busoni’, Studies in 
Music (Perth, Western Australia), ii (1968). 

STUCKENSCHMIDT, Н. H.: Ferruccio Busoni, Zeittafel eines Europäers (Zürich, 1967; 
English translation, London, 1970). 

VLAD, Roman: ‘Busoni’s Destiny’, Score, 7 (1952). 


Chausson 

BARRICELLI, PIERRE, and WEINSTEIN, LEO: Ernest Chausson: the Composer's Life and 
Works (Norman, Oklahoma, 1955). 

*Chausson' number of Revue musicale (1925). 


BIBLIOGRAPHY 703 


Debussy 

D'ALMENDRA, JULIA: Les Modes grégoriens dans l'oeuvre de Claude Debussy (Paris, 
1948). 

BRAILOU, CONSTANTIN: ‘Pentatony in Debussy's Music’, Studia Memoriae Bélae 
Bartók Sacra (3rd ed., London, 1959). 

DAWES, FRANK: Debussy: Piano Music (London, 1969). 

*Debussy' numbers of Revue musicale (1920, 1926, 1964). 

FISCHER, KURT Мом: ‘Bemerkungen zu den zwei Ausgaben von Debussys Ariettes 
oubliées’, Symbolae Historiae Musicae (Federhofer Festschrift), (Mainz, 1971). 

GERVAIS, FRANÇOISE: Debussy et l'évolution de la musique au XXe siècle (Paris, 1965). 

GRUBER, GERNOT: ‘Zur Funktion der “primären Klangformen" in der Musik 
Debussys’, Symbolae Historiae Musicae (Mainz, 1971). 

JAKOBIK, ALBERT: Die assoziative Harmonik in den Klavier-Werken Claude Debussys 
(Würzburg, 1940). 

JANKÉLÉVITCH, VLADIMIR: Debussy et le mystére (Neuchátel, 1949). 

KOECHLIN, CHARLES: Debussy (Paris, 1927). 

KorscH, Hans F.: Der Impressionismus bei Debussy (Düsseldorf, 1937). 

1ле55, ANDREAS: Claude Debussy: das Werk im Zeitbild (Strasbourg, 1936). 

———Claude Debussy und das deutsche Musikschaffen (Würzburg, 1939). 

— —'L'harmonie dans les oeuvres de Claude Debussy’, Revue musicale (1931). 

LOCKSPEISER, EDWARD: Debussy (revised edition, London, 1951). 

— Debussy: his Life and Mind (two volumes, London, 1962 and 1965). 

MELLERS, WILFRID: “Тһе Later Work of Claude Debussy’, in Studies in Contemporary 
Music (London, 1947). 

NicHoLs, ROGER: *Debussy's Two Settings of “Саш de lune 
xl (1967). 

PHILLIPS, C. HENRY: “Тһе Symbolists and Debussy’, Music and Letters, xiii (1932). 

SCHAEFFNER, ANDRE: ‘Debussy et ses rapports avec la musique russe’, in Musique 
russe, i (ed. P. Souvtchinsky) (Paris, 1953). 

SCHMITZ, ROBERT: The Piano Works of Claude Debussy (New York, 1950). 

Ѕтовв, ILSE: Untersuchungen zur Auflösung der funktionaler Harmonik in den 
Klavierwerken von Claude Debussy (Cologne, 1967). 

STUCKENSCHMIDT, H. H.: *Debussy or Berg? The Mystery of a Chord Progression', 
Musical Quarterly, li (1965). 

VALLAS, LEON: Claude Debussy et son temps (Paris, 1932; English translation, 
London, 1933). 

— — Les Idées de Claude Debussy, musicien français (Paris, 1927; English translation, 
1929). 


Delius 

ABRAHAM, GERALD: 'Delius and his Literary Sources', in S/avonic and Romantic 
Music (London, 1968). 

BEECHAM, THOMAS: Frederick Delius (London, 1959). 

Cooke, Оевуск: ‘Delius and Form: a Vindication’, Musical Times, сш (1962). 

‘Delius’ number of Tempo (1952). 

HESELTINE, PHILIP: Frederick Delius (revised ed., London, 1952). 

HOLLAND, А. K.: The Songs of Delius (London, 1951). 

HUTCHINGS, ARTHUR: Delius (London, 1948). 

PALMER, CHRISTOPHER: ‘Delius and Poetic Realism’, Music and Letters, li (1970). 

—— —'Delius, Vaughan Williams and Debussy’, Music and Letters, 1 (1969). 

PAYNE, ANTHONY: ‘Delius’s Stylistic Development’, Tempo, 60 (1961). 


Dukas 
*Dukas' number of Revue musicale (1936). 
FAVRE, GEORGES: Paul Dukas: sa vie — son oeuvre (Paris, 1948). 


LEES 


, Music and Letters, 


704 BIBLIOGRAPHY 


Elgar 

DANN, Mary, G.: ‘Elgar’s Use of the Sequence’, Music and Letters, xix (1938). 
*Elgar' number of Music and Letters (1935). 

KENNEDY, MICHAEL: Elgar: Orchestral Music (London, 1970). 

МСУЕАСН, Diana M.: Edward Elgar: his Life and Music (London, 1955). 
Maine, Basi: Elgar: his Life and Work (London, 1933). 


Falla - 

MAYER-SERRA, OTTO: ‘Falla’s Musical Nationalism’, Musical Quarterly, xxix (1943). 

РАНІЅЅА, JAIME: Vida y obra de Manuel de Falla (Buenos Aires, 1947; English 
translation, 1954). 

TREND, J. B.: Manuel de Falla and Spanish Music (London, 1930). 

—— "Manuel de Falla in Arabia", Music and Letters, iii (1922). 


Fauré 

*Fauré' numbers of Revue musicale (1922) and Monthly Musical Record (1945). 

Favre, Max: Gabriel Faurés Kammermusik (Zürich), 1948. 

JANKÉLÉVITCH, VLADIMIR: Gabriel Fauré: ses mélodies — son esthétique (revised and 
enlarged edition, Paris, 1951). 

KOECHLIN, CHARLES: Gabriel Fauré (Paris, 1927; English translation, London, 
1945). 

MELLERS, WILFRID: “Тһе Later Work of Gabriel Fauré’ in Studies in Contemporary 
Music (London, 1947). 

SUCKLING, NoRMAN: Fauré (London, 1946). 


Granados 

COLLET, HENRI: Albeniz et Granados (Paris, 1926). 

‘Granados’ number of Revista musical catalana (1916). 

LIVERMORE, ANN: ‘Granados and the Nineteenth Century in Spain’, Music Review, 
vii (1946). 


Grieg 

ABRAHAM, GERALD (ed.): Grieg: a Symposium (London, 1948). 

FiscHER, Kurt VON: Griegs Harmonik und Фе nordlandische Folklore (Berne, 1938). 
Horton, JOHN: ‘Grieg’s “ЗааНег” for Pianoforte’, Music and Letters, xxvi (1945). 
SCHJELDERUP-EsBE, Dac: A Study of Grieg's Harmony (Oslo, 1953). 


Holst 
Horsr, IMOGEN: The Music of Gustav Holst (London, 1951). 


d'Indy 

*d'Indy' numbers of Revue musicale (1932, 1937). 

SAINT-SAENS, CAMILLE: Les Idées de M. Vincent d’Indy (Paris, 1919). 
SÉnIEYx, AUGUSTE: Vincent d'Indy (Paris, 1914). 

VALLAS, LEON: Vincent d'Indy (Paris, two vols. 1946 and 1950). 


Koechlin 

CaLvoconsssi M. D.: ‘Charles Koechlin's Instrumental Works’, Music and 
Letters, v (1924). 

MELLERS, WILFRID: ‘Charles Koechlin’ in Studies in Contemporary Music (London, 
1947). 


Magnard 
CARRAUD, GASTON: La Vie, l'oeuvre et la mort d'Albéric Magnard (Paris, 1921). 


Mahler 
ADLER, Сошо: Gustav Mahler (Vienna, 1916). 
BAREORD, PHILIP: Mahler: Symphonies and Songs (London, 1970). 


— — AN 


BIBLIOGRAPHY 705 


BEKKER, PAUL: Gustav Mahlers Sinfonien (Berlin and Stuttgart, 1921). 

LOCKSPEISER, EDWARD: ‘Mahler in France’, Monthly Musical Record, xc (1960). 

MAHLER, ALMA M.: Gustav Mahler: Erinnerungen und Briefe (Amsterdam, 1940; 
enlarged and revised English translation, London, 1968). 

MAHLER, GusrAv: Briefe, 1879-1911 (Berlin, Vienna and Leipzig, 1924). 

‘Mahler’ numbers of Die Musik (1911) and Musikblütter des Anbruch (1920, 1930). 

MITCHELL, DONALD: Gustav Mahler: The Early Years (London, 1958). 

NEWLIN, Пика: Bruckner—Mahler—Schoenberg (New York, 1947). 

PAMER, Fritz EGon: ‘Gustav Mahlers Lieder’, Studien zur Musikwissenschaft, 
xvi (1929) and xvii (1930). 

RATZ, ERWIN: ‘Zum Formproblem bei Gustav Mahler: Eine Analyse des ersten 
Satzes der IX. Symphonie', Musikforschung, viii (1955). 

----<2шп Formproblem bei Gustav Mahler: Eine Analyse des Finales der VI. 
Symphonie', Musikforschung, ix (1956). 

REDLICH, Hans F.: Bruckner and Mahler (London, 1955). 

SCHAEFERS, ANTON: Gustav Mahlers Instrumentation (Düsseldorf, 1935). 

SrEFAN, PAUL: Gustav Mahler (enlarged and revised edition, Munich, 1920). 

TISCHLER, Hans: *Mahler's “Раз Lied von der Erde”’, Music Review, x (1949). 

——‘Mahler’s Impact on the Crisis of Tonality’, Music Review, xii (1951). 

WELLESZ, Есом: “Тһе Symphonies of Gustav Mahler', Music Review, i (1940). 


Nielsen 
NIELSEN, CARL: Levende Musik (Copenhagen, 1925). 
SIMPSON, RoBERT: Carl Nielsen, Symphonist (London, 1952). 


Rakhmaninov 

CULSHAW, JOHN: Sergei Rachmaninov (London, 1949). 

*Rakhmaninov' number of Tempo (1951). 

TsrrovicH, T. E. (ed): 5. V. Rakhmaninov: sbornik statey i materialov (Moscow 
and Leningrad, 1947). 

VASINA-GROSSMAN, V. A.: ‘Romansi Rakhmaninova’, in Russkiy klassicheskiy 
romans XIX veka (Moscow, 1956). 

ZHITOMIRSKY, D.: ‘Fortepiannoe tvorchestvo Rakhmaninova', Sovetskaya muzika: 
sbornik statey, 4 (1945). 


Ravel 

AKERET, KURT: Studien zum Klavierwerk von Maurice Ravel (Zürich, 1941). 

BRUYR, José: Maurice Ravel, ou le lyrisme et les sortilèges (Paris, 1950). 

Carvoconszssi, M. D.: ‘When Ravel composed to order’, Music and Letters, xxii 
(1941). 

JANKÉLÉVITCH, VLADIMIR: Ravel (2nd ed., Paris, 1956). 

JOURDAN-MORHANGE, HÉLENE: Ravel et nous (Geneva, 1945). 

ORENSTEIN, ARBIE: ‘Maurice Ravel’s Creative Process’, Musical Quarterly, liii (1967). 

‘Ravel’ numbers of Revue musicale (1925, 1938). 

ROLAND-MANUEL: A [а gloire de Ravel (Paris, 1938; English translation, London, 
1947). 


Reger 

Васев, Сошо: Max Reger (Stuttgart and Berlin, 1923). 

СОЕМЕМ, PauL: Max Regers Variationsschaffen (Berlin, 1935). 

DENECKE, Н. L.: “Мах Regers Sonatenform in ihrer Entwicklung’, Festschrift Fritz 
Stein (Brunswick, 1939). 

GATSCHER, E.: Die Fugentechnik Max Regers in ihrer Entwicklung (Stuttgart, 1925). 

HUESGEN, R.: Der junge Reger und seine Orgelwerke (Schrammberg, 1935). 

Moser, Hans JOACHIM: “Мах Regers Orchesterwerke’, Festschrift Max Reger 
(Leipzig, 1953). 


706 BIBLIOGRAPHY 


RAHNER, Huco Е.: “Мах Regers Choralfantasien für die Orgel’, Heidelberger 
Studien zur Musikwissenschaft (Kassel, 1936). 

*Reger' number of Die Musik (1921). 

STEIN, Fritz: Max Reger (Potsdam,.1939). 

—Thematisches Verzeichnis der im Druck erschienenen Werke von Max Reger 
(enlarged edition, Leipzig, 1953). 

THERSTAPPEN, Н. J.: ‘Uber die Grundlagen der Form bei Max Reger’, Festschrift 
Fritz Stein (Brunswick, 1939). 

WEHMEYER, СВЕТЕ: ‘Max Reger als Liederkomponist’, Kölner Beiträge zur Musik- 
forschung (Regensburg, 1955). 

WURZ, R. (ed.): Max Reger: eine Sammlung von Studien aus dem Kreise seiner 
persónlichen Schüler (Munich, four Hefte, 1920-3). 


Roslavets 

Gojowv, Detter: “Nikolaj Andreevic Roslavec, ein früher Zw6lftonkomponist’, 
Musikforschung, xxii (1969). 

Козглуетв, NIKOLAY: ‘Nik. А. Roslavets о зере i svoem tvorchestve’, Sovremennaya 
muzika, no. 5 (1924). 


Roussel 

BERNARD, ROBERT: Albert Roussel (Paris, 1948). 

DEANE, BASIL: Albert Roussel (London, 1961). 

Ho£nÉz, A. R.: Albert Roussel (Paris, 1938). 

LOCKSPEISER, EDWARD: ‘Roussel and Ravel’, Music and Letters, xix (1938). 

MELLERS, WILFRID: ‘Albert Roussel and La Musique francaise’, in Studies in 
Contemporary Music (London, 1947). 

PINCHERLE, Макс: Albert Roussel (Geneva, 1957). 

‘Roussel’ numbers of Revue musicale (1929, 1937). 


Satie 

AUSTIN, WILLIAM: ‘Satie before and after Cocteau’, Musical Quarterly, xlviii (1962). 

DANCKERT, WERNER: ‘Der Klassizismus Erik Saties und seine geistesgeschichtliche 
Stellung’, Zeitschrift für Musikwissenschaft, xii (1929-30). 

MELLERS, WILFRID: ‘Erik Satie and the “Problem” of Contemporary Music’, in 
Studies in Contemporary Music (London, 1947). 

Myers, Комо H.: Erik Satie (London, 1948). 

TEMPLIER, PIERRE-DANIEL: Erik Satie (Paris, 1932; English translation, London, 
1969). 


Schmitt 
Hucuer, YVES: Florent Schmitt, l'homme et l'artiste, son époque et son oeuvre 
(Paris, 1953). 


Schoenberg 

ARMITAGE, MERLE (ed.): Arnold Schoenberg (New York, 1939). 

Вевс, ALBAN: Gurrelieder (Führer), (Vienna, 1913). 

Kammersinfonie, Op. 9 (Thematische Analyse), (Vienna, n.d.). 

—— Pelleas und Melisande, Op. 5 (Kurze thematische Analyse) (Vienna, n.d.). 

——‘Warum ist Schönbergs Musik so schwer verstándlich?', reprinted in Willi 
Reich, Alban Berg (Vienna, 1937). 

Воот, PIERRE: ‘L’Oeuvre pour piano de Schönberg’, іп Relevés d'apprenti (Paris, 
1966; English translation, 1968). 

BRINKMANN, REINHOLD: Arnold Schónberg: Drei Klavierstiicke Op. 11 (Studien 
zur frühen Atonalitat bei Schönberg) (Wiesbaden, 1969). 

EHRENFORTH, К. H.: Ausdruck und Form (Schönbergs Durchbruch zur Atonalitát in 
den George-Liedern, Op. 15) (Bonn, 1963). 


BIBLIOGRAPHY 707 


FRIEDBERG, RUTH: “Тһе Solo Keyboard Works of Arnold Schönberg’, Music 
Review, xxiii (1962). 

FRIEDHEIM, PHILIP: Tonality and Structure in the Early Works of Schoenberg (Diss. 
New York Univ., 1963). 

Hopkins, С. W.: ‘Schoenberg and the “Logic” of Atonality’, Tempo, 94 (1970). 

JALOWETZ, HEINRICH: “Оп the Spontaneity of Schoenberg’s Music’, Musical 
Quarterly, xxx (1944). 

NEWLIN, DIKA: Bruckner—Mahler—Schoenberg (New York, 1947). 

PAYNE, ANTHONY: Schoenberg (London, 1968). 

RurER, Joser: Das Werk Arnold Schönbergs (Kassel, 1959; English translation, 
London, 1962). 

SCHOENBERG, ARNOLD: Harmonielehre (Vienna, 1911). 

— Style and Idea (New York, 1951). 

—‘Schoenberg’ Festschriften (Munich, 1912 and Vienna, 1925 and 1934); 
‘Schoenberg’ numbers of Musikblatter des Anbruch (1924), Pult und Taktstock 
(1927), etc. 

WELLESZ, EGON: Arnold Schónberg (Vienna, 1921; English translation, London, 
1925; new edition, 1972). 

WILLE, RUDOLF: ‘Reihentechnik in Schónbergs opus 19, 2’, Musikforschung, 
xix (1966). 


Sibelius 

ABRAHAM, GERALD (ed.): Sibelius: a Symposium (London, 1947). 

Gray, CECIL: Sibelius: the Symphonies (London, 1935). 

HILL, WILLIAM G.: ‘Some Aspects of Form in the Symphonies of Sibelius’, Music 
Review, x (1949). 

LAYTON, ROBERT: Sibelius (London, 1965). 

PARMET, SIMON: Sibelius Symfonier (Helsinki, 1955; English translation, London, 
1959). 

Roma, Емо: Die Symphonien von Jean Sibelius: eine formanalytische Studie 
(Jyvaskyla, 1941). 

TANZBERGER, ERNST: Die symphonischen Dichtungen von Jean Sibelius (Würzburg, 
1943). 

TAWASTSTJERNA, ERIK: The Compositions of Sibelius (Helsinki, 1957-). 


Skyrabin 

Вевкоу, У.: ‘Nekotorie voprosi garmonii Skryabina', Sovetskaya muzika, 
xxiii (1959). 

COOPER, MARTIN: ‘Scriabin’s Mystical Beliefs’, Music and Letters, xvi (1935). 

DICKENMANN, PAUL: Die Entwicklung der Harmonik bei А. Skrjabin (Berne, 1935). 

Сі сн, С. С. J. von: Die sinfonischen Werke von Alexander Skrjabin (Bilthoven, 
1963). 

KELDIsH, YURY: ‘Ideynie protivorechiya v tvorchestve А. М. Skryabina’, Sovetskaya 
muzika, xvi (1950). 

Lissa, Zoria: “О harmonice А. М. Skrjabin’, Kwartalnik muzyczny, viii (1930). 

SABANEYEV, LEONID: А. №. Skryabin (Moscow, 1922). 
Vospominaniya o Skryabina (Moscow, 1925). 

Skryabin, N. A., 1915-1940: Sbornik k 25-letiyu so dnya smerti (Moscow, 1940). 

‘Skryabin’ number of Muzikalny sovremennik (1916). 

WESTPHAL, Кокт: ‘Die Harmonik Scrjabins’, Musikblatter des Anbruch, xi (1929). 


Strauss 

Det Mar, NORMAN: Richard Strauss: a Critical Commentary on his Life and Work 
(three volumes, London, 1963, 1969 and 1972). 

Сүз, Fritz: Richard Strauss (Potsdam, 1934). 


708 BIBLIOGRAPHY 


MUELLER VON Asow, E. H.: Richard Strauss: Thematisches Verzeichnis (two 
volumes, published Vienna, 1955 and 1959). 

LORENZ, ALFRED: ‘Neue Formerkenntnisse angewandt auf Richard Straussens “Поп 
Juan", Archiv für Musikforschung, i (1936). 

ScHUH, WILLI (ed.): Richard Strauss Jahrbuch (Bonn, 1954—00). 

SPECHT, RICHARD: Richard Strauss und sein Werk (Vienna, 1921). 

STEINITZER, Max: Richard Strauss (Berlin, revised and enlarged edition, 1922). 

TENSCHERT, ROLAND: ‘Die Kadenzbehandlung bei Richard Strauss’, Zeitschrift 
für Musikwissenschaft, viii (1925-6). 

——‘Versuch einer Typologie der Richard Strausschen Melodik’, Zeitschrift fiir 
Musikwissenschaft, xvi (1934). 

WACHTEN, EDMUND: ‘Der einheitliche Grundzug der Strausschen Formgestaltung’, 
Zeitschrift für Musikwissenschaft, xvi (1934). 


Stravinsky 

Boys, Henry: ‘Stravinsky: A Propos his Aesthetic’, Score, 2 (1950). 

——‘Stravinsky: The Musical Materials’, Score, 4 (1951). 

Drew, Davin: ‘Stravinsky’s Revisions’, Score, 20 (1957). 

Evans, EDWIN: Stravinsky: ‘The Fire-Bird’ and ‘Petrushka’, (London, 1933). 

‘GLEBOV, IGoR' (Boris Asafyev): Kniga o Stravinskom (Leningrad, 1929). 

Hopkins, С. W.: ‘Stravinsky’s Chords’, Tempo, 76 and 77 (1966). 

SCHAEFFNER, ANDRE: Stravinsky (Paris, 1931). 

ScHUH, WILLI: ‘Zur Harmonik Igor Strawinskys’, Schweizerische Musikzeitung, 
xcii (1952). 

SMALLEY, ROGER: ‘The Sketchbook of The Rite of Spring’, Tempo, 91 (1969). 

STRAVINSKY, IGOR: Poétique musicale (Cambridge, Mass., 1942; English translation, 
Cambridge, Mass., 1947; revised edition, Paris, 1952). 

‘Stravinsky’ numbers of Revue musicale (1923 and 1939), Musical Quarterly (1962) 
and Tempo (1948, 1967 and 1971). 

TANSMAN, ALEXANDRE: Igor Stravinsky (Paris, 1948; English translation, New York, 
1949). 

VLAD, RoMAN: Stravinsky (Milan, 1958; English translation, 2nd edition, London, 
1967). 

Wuite, ERIC WALTER: Stravinsky: The Composer and his Works (London, 1966). 


Szymanowski 

CHOMINSKY, 702ЕЕ: ‘Szymanowski i Skryabin’, in Russko-polskie muzikalnie 
svyazi (ed. Igor Belza), (Moscow, 1963). | 

LoBACZEWSKA, STEFANIA: Karol Szymanowski: Zycie i twórczosc (Cracow, 1950). 

STUCKENSCHMIDT, H. H.: ‘Karol Szymanowski’, Music and Letters, xix (1938). 


Vaughan Williams 

Dickinson, A. E. F.: An Introduction to the Music of R. Vaughan Williams (London, 
1928). 

Foss, HuBERT: Ralph Vaughan Williams (London, 1950). 

Howes, FRANK: The Music of Ralph Vaughan Williams (London, 1954). 

KENNEDY, MICHAEL: The Works of Ralph Vaughan Williams (London, 1964). 

KIMMEL, WILLIAM: ‘Vaughan Williams’s Choice of Words’, Music and Letters, xxi 
(1938). 

——‘Vaughan Williams’s Melodic Style’, Musical Quarterly, xxvii (1941). 

PAYNE, Exsie: ‘Vaughan Williams and Folksong’, Music Review, xv (1954). 

VAUGHAN WILLIAMS, RALPH: National Music (London, 1934). 


Webern 
Соме, EDWARD T.: ‘Webern’s Apprenticeship’, Musical Quarterly, liii (1967). 


BIBLIOGRAPHY 709 


KOLNEDER, WALTER: Anton Webern: Einführung in Werk und Stil (Rodenkirchen, 
1961; English translation, London, 1968). 


CHAPTER III 


STAGE WORKS: 1890-1918 


(1) General 

GILMAN, LAWRENCE: Aspects of Modern Opera (London, 1924). 

GROUT, DONALD Jay: A Short History of Opera (New York, 1947; 2nd ed., 1965). 

IsTEL, EDGAR: Die moderne Oper vom Tode Wagners bis zum Weltkrieg (Leipzig, 
1915; 2nd ed., 1923). 

—‘German Opera since Richard Wagner’, Musical Quarterly, 1 (1915). 

KLEIN, JOHN W.: ‘Verdi’s Italian Contemporaries and Successors’, Music and 
Letters, xv (1934). 

Louis, RUDOLF: Die deutsche Musik der Gegenwart (Munich, 1909; 3rd ed., 1912). 

PROD'HOMME, JACQUES GABRIEL: “Тһе Recent Fiftieth Anniversary of the “New 
Орега”?, Musical Quarterly, xii (1926). 

ScHULLER, KENNETH GUSTAVE: 'Verismo' Opera and the Verists (Diss. Washington 
Univ., 1960). 


(ii) Individual Composers 


d'Albert 

RAUPP, WiLHELM: Eugen d'Albert. Ein Künstler und Menschenschicksal (Leipzig, 
1930). 

5снміт2, EUGEN: ‘Eugen d'Albert als Opernkomponist’, Hochland, vi (1909). 

Товсні, Іллсі: ‘Ghismonda, opera in tre atti di Eugenio d'Albert', Rivista musicale 
italiana, iii (1896). 


Bartok 

LENDVAI, ERNÓ: ‘A kékszakállu herceg vara’ (Duke Bluebeard’s Castle), Magyar 
Zene, i (1961). 

VERESS, SANDOR: ‘Bluebeard’s Castle’, Tempo, xiii (1949); xiv (1949-50); repr. in 
Béla Barték: A Memorial Review (New York, 1950). 


Bloch 

COHEN, ALEX: ‘Ernest Bloch’s Macbeth’, Music and Letters, xix (1938). 

Gatti, Сошо Mania: ‘Two Macbeths: Verdi-Bloch', Musical Quarterly, xii (1926). 

HALL, RAYMOND: ‘The Macbeth of Bloch’, Modern Music, xv (1938). 

HAstTINGs, JOHN: ‘Ernst Bloch and Modern Music’, Music Review, x (1949). 

NEWLIN, Dika: ‘The Later Works of Ernest Bloch’, Musical Quarterly, xxxiii 
(1947). 

TiBALDI-CHIESA, M.: Ernest Bloch (Turin, 1933). 


Busoni 

Dent, EDWARD J.: ‘Busoni’s Doctor Faust’, Music and Letters, vii (1926). 

Gatti, Сошо Maria: ‘The Stage Works of Ferruccio Busoni’, Musical Quarterly, 
xx (1934). 

GOSLICH, SIEGFRIED: ‘Раз Wandbild: Othmar Schoeck und Ferruccio Busoni’, 
Musica, xi (1957). 

GUERRINI, Сошо: Ferruccio Busoni: la уйа, la figura, l'opera (Florence, 1944). 

STUCKENSCHMIDT, HANS HEINZ: ‘Rede über Busonis Doktor Faust’, Schweizerische 
Musikzeitung, xcvi (1956). 


710 BIBLIOGRAPHY 


Charpentier 

DzLMas, Marc: Gustave Charpentier et le lyrisme francais (Paris, 1931). 

HiwoNET, ANDRE: Louise de Gustave Charpentier: Etude historique et critique, 
analyse musicale (Paris, 1922). | 

Hoover, KATHLEEN O’DONNELL: “Gustave Charpentier’, Musical Quarterly, xxv 
(1939). 


Debussy = 

ACKERE, JULES VAN: Pelléas et Mélisande, ou le recontre miraculeuse d'une poésie 
et d'une musique (Brussels, 1952). 

APPLEDORN, MARY JEANNE VAN: 4 Stylistic Study of Claude Debussy's Opera 
‘Pelléas et Mélisande' (Diss. Univ. of Rochester, 1966). 

CHAILLEY, JACQUES: “Ге Symbolisme des thémes dans Pelléas et Mélisande', 
L'information musicale, lxiv (1942). 

DavisoN, ARCHIBALD T.: The Harmonic Contribution of Claude Debussy (Diss. 
Harvard Univ., 1908). 

EMMANUEL, Maurice: ‘Pelléas et Mélisande’: étude historique et critique, analyse 
musicale (Paris, 1925). 

СОВА, ANTOINE: ‘Pelléas et Mélisande’: Analyse poétique et musicale (Paris, 1952). 

LOCKSPEISER, EDWARD: Debussy (London, 1936; 3rd ed., 1951). 

— —Debussy: His Life and Mind (2 vols., London, 1962-5). 

—‘Mussorgsky and Debussy’, Musical Quarterly, xxiii (1937). 

PIZZETTI, ILDEBRANDO: ‘Pelléas et Mélisande ... Debussy’, Rivista musicale 
italiana, xv (1908). 


Delius 

HUTCHINGS, ARTHUR: Delius (London, 1948). 

— —'pelius's Operas’, Tempo, xxvi (1952-3). 

KLEIN, JOHN W.: ‘Delius as a Musical Dramatist, Music Review, xxii (1961). 


Dukas 

Возме, HENRY DE: ‘Ariane et Barbe-bleue de M. Paul Dukas', Mercure musicale, 
iii (1907). 

PIZZETTI, ILDEBRANDO: ‘Ariane et Barbebleue ... de Paul Dukas’, Rivista musicale 
italiana, xv (1908). 


d'Indy 

KUFFERATH, MAURICE: 'Fervaal . . . di V. d’Indy’, Rivista musicale italiana, iv (1897). 

ROLLAND, ROMAIN: ‘L’Etranger de Vincent d'Indy', Rivista musicale italiana, xi 
(1904). 

VALLAS, LEON: Vincent d'Indy (2 vols., Paris, 1946-50). 


Mascagni 

Ромен, EDOARDO: Pietro Mascagni (Rome, 1912). 

Товсні, Luici: ‘Guglielmo Ratclif. .. di Pietro Mascagni’, Rivista musicale italiana, 
ii (1895). 

‘Iris ... di Pietro Mascagni’, Rivista musicale italiana, vi (1899). 


Massenet 

BRUNEAU, ALFRED: Massenet (Paris, 1935). 

MASSENET, JULES-EMILE-FREDERIC: Mes souvenirs (Paris, 1912). 
PouGIN, ARTHUR: ‘Massenet’, Rivista musicale italiana, xix (1912). 


Pfitzner 

BERRSCHE, ALEXANDER: Kurze Einführung in Hans Pfitzners Musikdrama ‘Der arme 
Heinrich? (Leipzig, 1910). 

HALUSA, KARL: Pfitzners musikdramatisches Schaffen (Diss. Vienna, 1929). 


BIBLIOGRAPHY 711 


MANN, THOMAS: Pfitzners ‘Palestrina’ (Berlin, 1919). 
PFITZNER, HANs ERICH: Сезаттейе Schriften (3 vols., Augsburg, 1926). 
Rutz, Hans: Hans Pfitzner: Musik zwischen den Zeiten (Vienna, 1949). 


Puccini 

Витесь A.: ‘La Bohème’ di Giacomo Puccini: Studio critico (Palermo, 1931). 

BONACCORSI, ALFREDO: Giacomo Puccini e i suoi antenati musicali (Milan, 1950). 

CARNER, Mosco: “Тһе Exotic Element in Puccini', Musical Quarterly, xxii (1936). 

— —Puccini: A Critical Biography (London, 1958). 

——‘Puccini’s Early Operas’, Music and Letters, xix (1938). 

Сатті, Сошо Maria: “Тһе Works of Giacomo Puccini’, Musical Quarterly, xiv 
(1928). 

MARIANI, RENATO: ‘L’ultimo Puccini’, Rassegna musicale, ix (1936). 

PARKER, DouGLas C.: ‘A view of Giacomo Puccini’, Musical Quarterly, iii (1917). 

Ricci, Іллсі: Puccini interprete di se stesso (Milan, 1954). 

SARTORI, CLAUDIO (ed.): Giacomo Puccini (Milan, 1959). 

Токсні, Іллсі: ‘Tosca, di С. Puccini’, Rivista musicale italiana, vii (1900). 

Schillings 

LEPEL, FELIX VON: Max von Schillings und seine Oper ‘Mona Lisa’: Ein Ruhmesblatt 
fiir die stadtische Oper in Berlin-Charlottenburg (Berlin-Charlottenburg, 1954). 


Richard Strauss 

DEL Mar, NORMAN: Richard Strauss (3 vols., London, 1962, 1969, 1972). 

ERHARDT, Отто: Richard Strauss, Leben, Wirken, Schaffen (Freiburg/Breisgau, 
1953). 

FAHNRICH, HERMANN: ‘Richard Strauss über das Verhältnis von Dichtung und 
Musik (Wort und Ton) in seinem Opernschaffen', Musikforschung, xiv (1961). 

GREGOR, JosEF: Richard Strauss, der Meister der Oper (Munich, 1939). 

JEFFERSON, ALAN: The Operas of Richard Strauss in Britain, 1910-1963 (London, 
1963). 

KRALIK, HEINRICH: Richard Strauss, Weltbürger der Musik (Vienna, 1963). 

KRÜGER, KARL-JOACHIM: Hugo von Hofmannsthal und Richard Strauss (Berlin, 1935) 

MANN, WiLLIAM: Richard Strauss. A Critical Study of his Operas (London, 1964). 

RorH, Ernst (ed.): Richard Strauss Bühnenwerke (London, 1954). 

ScHMITZ, EUGEN: Richard Strauss als Musikdramatiker (Munich, 1907). 

ScHUH, WILLI: Hugo von Hofmannsthal und Richard Strauss, Legende und Wirklich- 
keit (Munich, 1964). 

—— Über Opern von Richard Strauss (Zurich, 1947). 

STRAUSS, RICHARD and HANS VON BÜLow: Correspondence (English translation, 
London, 1955). 

----апа HuGo von HOFMANNSTHAL: Briefwechsel (Zürich, 3rd ed., 1964). 

— —English translation as А Working Friendship (New York, 1961). 

—— and Комлім ROLLAND: Correspondence, ed. R. Meyers (Berkeley, Calif., 1968). 

----апа FRANZ WULLNER: Briefwechsel, ed. D. Kämper (Cologne, 1963). 

TENSCHERT, ROLAND: ‘Versuch einer Typologie der Richard Strausschen Melodik’, 
Zeitschrift für Musikwissenschaft, xvi (1934—5). 

Dreimal sieben Variationen über das Thema Richard Strauss (Vienna, 1944). 

TRENNER, FRANZ: Richard Strauss: Dokumente seines Lebens und Schaffens (Munich, 
1954). 


Wolf 

BoLLERT, WERNER: ‘Hugo Wolfs Corregidor’, Musica, xiv (1960). 

HELLMER, ELMUND (ed.): 'Der Corregidor' von Hugo Wolf (Berlin, 1900). 

HERNRIED, RoBERT: ‘Hugo Wolf’s *Four Operas"', Musical Quarterly, xxxi 
(1945). 


712 BIBLIOGRAPHY 


REICH, WILLI: ‘Dokument eines Gesprächs (Zur Wiener Erstaufführung von Wolfs 


Corregidor)’, Musica, xiv (1960). "n 
WALKER, FRANK: Hugo Wolf (London, 1951; 2nd ed., New York, 1968). 
Wolf-Ferrari 


GnRISSON, ALEXANDER CAROLA: Ermanno Wolf-Ferrari (Zürich, 2nd ed., 1958). 

PFANNKUCH, WILHELM: Das Operuschaffen Ermanno Wolf-Ferraris (Diss. Kiel 
Univ., 1953). - 

RiNGO, JAMEs: ‘Ermanno Wolf-Ferrari: An Appreciation of his Work’, Rivista 
musicale italiana, xlix (1949). 

Токсні, Luici: ‘La vita nuova di E. Wolf-Ferrari', Rivista musicale italiana, x (1903). 

WOLF-FERRARI, ERMANNO: ‘Meine Beziehung zur komischen Oper’, Zeitschrift 
fiir Musik, cviii (1941). 

ZENTNER, WILHELM: ‘Zum Opernschaffen Ermanno Wolf-Ferraris’, Zeitschrift 
für Musik, cviii (1941). 


CHAPTER IV 


MUSIC IN THE MAINLAND OF EUROPE: 1918-1939 


() General 

AUSTIN, WILLIAM W.: Harmonic Rhythm іп Twentieth-Century Music (Diss. Harvard 
Univ., 1951). 

BABBITT, RICHARD B.: The Harmonic Idioms in the Works of Les Six (Diss. Boston 
Univ., 1963). 

COLLAER, PAUL: La musique moderne, 1905-1955 (Paris, 1955; English translation, 
аз 4 History of Modern Music, Cleveland, 1961). 

CooPER, GROSVENOR W.: Ап introduction to the Analysis of Certain Contemporary 
Harmonic Practices (Diss. Harvard Univ., 1939). 

Davies, LAURENCE: Райз to Modern Music (London, 1971). 

Deri, Отто: Exploring Twentieth Century Music (New York, 1968). 

DuMESNIL, RENE: La Musique en France entre les deux guerres 1919-1939 (Geneva, 
1946). 

FRANCES, Вовевт: La perception de la musique (Paris, 1958). 

Нил, EDWARD BURLINGAME: Modern French Music (Boston, 1924). 

HAUSLER, Joser: Musik in 20 Jahrhundert von Schönberg zu Penderecki (Bremen, 
1969). 

HODIER, ANDRE: Since Debussy: A View of Contemporary Music (English translation, 
New York, 1961). 

KRENEK, Ernst: Music Here and Now (1937; English translation, New York, 1939). 

— —'The New Music and Today's Theatre’, Modern Music, xiv (1937). 

— — Opera between Ше Wars’, Modern Music, хх (1943). 

KoMoROWsKI, HANS-PETER: Die ‘Invention’ in der Musik des 20. Jahrhunderts, 
(Kölner Beiträge zur Musikforschung, lxii) (Regensburg, 1971). 

Квонев, EKKEHART: Impressionismus іп der Musik (Leipzig, 1957). 

LANG, PAUL HENRY and NATHAN BRODER (eds.): Contemporary Music in Europe 
(New York, 1965). 

МАСНІЈ5, ЈОЅЕРН : Introduction to Contemporary Music (New York, 1961). 

MELLERS, WILFRID: Studies in Contemporary Music (London, 1948). 

MERSMANN, Hans: Musik der Gegenwart (Berlin, 1923). 

MEYER, LEONARD B.: Music, the Arts, and Ideas: Patterns and Predictions т 
Twentieth-Century Culture (Chicago, 1967). 

Myers, Коло Н. (ed.): Twentieth-Century Music (London, 1960). 

PERLE, GEORGE: Serial Composition and Atonality: An Introduction to the Music 
of Schoenberg, Berg, and Webern (Berkeley, Calif., 1962). 


BIBLIOGRAPHY 713 


PEYSER, JoAN: The New Music (New York, 1971). 

Rzri, RUDOLPH: The Thematic Process in Music (New York, 1951). 

——Tonality, Atonality, Pantonality: A Study of some Trends in 20th Century 
Music (New York, 1958). 

SALAZAR, ADOLFO: La música moderna: Las corrientes directrices еп el arte musical 
contemporáneo (Buenos Aires, 1944; English translation as Music in Our Time: 
Trends in Music Since the Romantic Era, New York, 1946). 

SALZMAN, Екіс: Twentieth-Century Music: An Introduction (Englewood Cliffs, 
N.J., 1967). 

ScHWARTZ, ELLIOTT and BARNEY CHILDs (eds.): Contemporary Composers оп 
Contemporary Music (New York, 1967). 

SMITHER, HOWARD E.: Theories of Rhythm in the Nineteenth and Twentieth Centuries, 
with a Contribution to the Theory of Rhythm for the Study of Twentieth-Century 
Music (Diss. Cornell Univ., 1960). 

STUCKENSCHMIDT, HANS HEINZ: Neue Musik; Vol. П: Zwischen den beiden Kriegen 
(Berlin, 1951). 

— — "Opera in Germany Today’, Modern Music, xiii (1935). 

— —Twentieth-Century Music (English translation, New York, 1969). 

VETTER, HANS JOACHIM: Die Musik unseres Jahrhunderts (Mainz, 1968). 

VLAD, ROMAN: Storia della dodecafonia (Milan, 1958). 

WEISSMANN, ADOLPH: Germany's Latest Music Dramas’, Modern Music, iv (1927). 

WESTPHAL, Ковт: Die moderne Musik (Leipzig, 1928). 

YATES, PETER: Twentieth Century Music (New York, 1967). 


(1) Individual Composers 


Вагіок 

ABRAHAM, GERALD: “Тһе Bartók of the Quartets', Music and Letters, xxvi (1945). 

ВАввит, MILTON: “Тһе String Quartets of Bartok’, Musical Quarterly, xxxv (1949). 

Bator, Victor: The Béla Bartók Archives: History and Catalogue (New York, 
1963). 

Béla Bartók: A Memorial Review (New York, 1950). 

CITRON, PIERRE: Bartók (Paris, 1963). 

DirrE, DENUS (ed.): Documenta Bartókiana (Budapest, 1964—  ). 

Сомвові, Отто: “Béla Bartok, 1881-1945, Musical Quarterly, xxxii (1946). 

Корду, 7огтАм: ‘Béla Bartók’, Revue musicale, ii (1921). 

Мовевох, SERGE: Béla Bartók (Paris, 1955). 

STEVENS, HALSEY: The Life and Music of Béla Bartók (New York, 1953). 

VINTON, JOHN: ‘Bartók on his own Music’, Journal of the American Musicological 
Society, xix (1966). 

WEISSMANN, JOHN S.: ‘Béla Bartók: An Estimate’, Music Review, vii (1946). 


Berg 

ARCHIBALD, Ковевт B.: Harmony in the Early Works of Alban Berg (Diss. Harvard 
Univ., 1965). 

BERG, ALBAN: “А Word about Wozzeck’, Modern Music, v (1927). 

—— Écrits, ed. and tr. Henri Pousseur (Monaco, 1957). 

FOoRNEBERG, ERICH: ‘Wozzeck’ von Alban Berg (Berlin-Lichterfelde, 1963). 

KLEIN, Тонм W.: ‘Wozzeck—A Summing Up’, Music and Letters, xliv (1963). 

PERLE, GEORGE: ‘Lulu: The Formal Design’, Journal of the American Musicological 
Society, xvii (1964). 

—‘The Music of Lulu: A New Analysis’, Journal of the American Musicological 
Society, xii (1959); corrections: Journal of the American Musicological Society, 
xiv (1961). 


714 BIBLIOGRAPHY 


PLorBscH, GERD: Alban Bergs ‘Wozzeck’ (Strasbourg, 1968). 

REDLICH, HANs FERDINAND: Alban Berg: Versuch einer Wiirdigung (Vienna, 1957; 
English condensation, The Man and his Music, New York, 1957). 

REICH, WILLI: Alban Berg (Vienna, 1937; 2nd ed. Zurich, 1963; English translation, 

The Life and Works of Alban Berg, London, 1965). 

‘Alban Berg’s Lulu’, Musical Quarterly, xxii (1936). 

—‘Alban Berg’s Oper Lulu’, Melos, xix (1952). 

——‘Lulu—the Text and Music’, Modern Music, xii (1935). 

—‘A Guide to Wozzeck’, Musical Quarterly, xxxviii (1952). 

STEIN, ERWIN: ‘Berg and Schoenberg’, Tempo, xliv (1957). 


Casella 

D'AMICO, FEDELE and Соро М. GATTI: Alfredo Casella: con saggi ... appendice 
biobibliografica (Milan, 1958). 

CASELLA, ALFREDO: ‘Matière et timbre’, Revue musicale, ii (1921). 

— —I segreti della Giara (Florence, 1941). 

CASELLA, ALFREDO and ViRGILIO MORTARI: La tecnica dell'orchestra contemporanea 
(Milan, 1950). 

CORTESE, Louis: Alfredo Casella (Genoa, 1935). 


Dallapiccola 
D'AMICO, FEDELE: *Luigi Dallapiccola', Melos, xx (1953). 
VLAD, RoMaN: Luigi Dallapiccola (English translation, Milan, 1957). 


Falla 

ARIZARA, Roporro: Manuel de Falla (Buenos Aires, 1961). 

CAMPODONICO, Luis: Falla (French translation, Paris, 1959). 

JAENISCH, JULIO: Manuel de Falla und die spanische Musik (Zürich, 1952). 

MAYER-SERRA, Отто: ‘Falla’s Musical Nationalism’, Musical Quarterly, xxix 
(1943). 

MiLA, Massimo (ed.): Manuel de Falla (Milan, 1962). 

РАНІЅЅА, JAIME: Vida y obra de Manuel de Falla (Buenos Aires, 1947; English 
translation, Manuel de Falla: His Life and Works, London, 1954). 

WORNER, КАвг Н. ‘Manuel de Falla’, Musica, i (1947). 


Hindemith 

Вілт25ткім, Marc: ‘Hin und Zurück in Philadephia’, Modern Music, v (1928). 

BoATWRIGHT, HowaRp: ‘Paul Hindemith as a Teacher’, Musical Quarterly, i 
(1964). 

Gutman, Hans: ‘Tabloid Hindemith’, Modern Music, vii (1930). 

HrNsEL, HERMAN R.: On Paul Hindemith’s Harmonic Fluctuation Theory (Diss. 
Univ. of Illinois, 1964). 

Німремітн, PAUL: A Composer’s World: Horizons and Limitations (New York, 
1961). 

——The Craft of Musical Composition (New York, 1945). 

Reicu, У/илл: ‘Paul Hindemith’, Musical Quarterly, xvii (1931). 

Rosner, HELMUT: Paul Hindemith: Katalog seiner Werke. Diskographie, Biblio- 
graphie, Einführung in das Schaffen (Frankfurt/Main, 1970). 

SCHILLING, Н. L.: Die Oper ‘Cardillac’ von P. Hindemith (Diss. Freiburg/Breisgau, 
1957). 

STONE, Kurt: Paul Hindemith: Catalogue of Published Works and Recordings 
(New York, 1954). 

STROBEL, HEINRICH: Paul Hindemith (Mainz, 1928; 3rd ed., 1948). 

WiLLMs, FRANZ: Führer zur Oper ‘Cardillac’ von Paul Hindemith (Mainz, 1926). 

——‘Paul Hindemith: Ein Versuch’, Von neuer Musik, i (1925). 


BIBLIOGRAPHY 715 


Honegger 

Ввоув, José: Honegger et son oeuvre (Paris, 1947). 

DELANNOY, Manczr: Arthur Honegger (Paris, 1953). 

GEORGE, ANDRÉ: Arthur Honegger (Paris, 1926). 

GERARD, CLAUDE: Arthur Honegger (Brussels, 1945).) 

Ho£nÉE, ARTHUR: (Honegger) La vie, l'oeuvre, l'homme (Paris, 1942). 
HONEGGER, ARTHUR: Incantation aux fossiles (Lausanne, 1948). 

Je suis compositeur (Paris, 1951; German translation, Zürich, 1952). 
PRUNIERES, HENRY: ‘Honegger’s Judith’, Modern Music, iii (1926). 
TaPPOLET, WILLY: Arthur Honegger (Boudry-Neuchátel, 1957). 


Janácek 

Brop, Max: Leos Janácek: Leben und Werk (Vienna, 1956). 

GERLACH, REINHARD: ‘Leoš Janáček und die Erste und Zweite Wiener Schule’, 
Musikforschung, xxiv (1970). 

HOLLANDER, Hans: ‘Leoš Janáček and his Operas’, Musical Quarterly, xv (1929). 

— Leos Janáček: His Life and Work (English translation, London, 1963). 

—‘Leo§ Janáček in seinen Opern’, Neue Zeitschrift für Musik, cxix (1958). 

RACEK, JAN: ‘Der Dramatiker Janáček’, Deutsches Jahrbuch der Musikwissenschaft, 
v (1961). 
‘Leoš Janáceks und Béla Bartóks Bedeutung in der Weltmusik’, Studia Musi- 
cologica, v (1963). 

STEDRON, Вономік: The Work of Leos Janáček (Prague, 1959). 

VOGEL, JAROSLAV: Leos Janáček: Leben und Werk (Prague, 1958; English transla- 
tion, London, 1963). 


Kodaly 

EOszE, LÁszLÓ: Kodaly Zoltán (Budapest, 1956; English translation, Zoltán Kodály, 
his Life and Work, London, 1962). 

Mason, Сотлм: ‘Kodaly and Chamber Music’, Studia Musicologica, iii (1962). 

STEVENS, НАІЅЕҮ: ‘The Choral Music of Zoltán Kodaly,’ Musical Quarterly, liv 
(1968). 


Martin 

ANSERMET, ERNEST: ‘Der Weg Frank Martins’, Osterreichische Musikzeitschrift, 
xi (1956). 

KLEIN, RUDOLF: Frank Martin: sein Leben und Werk (Vienna, 1960). 

KOoELLIKER, ANDRE: Frank Martin (Lausanne, 1963). 

TUPPER, JANET E.: Stylistic Analysis of Selected Works by Frank Martin (Diss. 
Indiana Univ., 1964). 


Martini 

CLAPHAM, JOHN: ‘Martint’s Instrumental Style’, Music Review, xxiv (1963). 

HALBREICH, Harry: Bohuslav Martini: Werkverzeichnis: Dokumentation: Biographie 
(Zürich, 1968). 

MIHULE, JAROSLAV: Symfonie Bohuslava Martini (Prague, 1959). 

ŠAFRÁNEK, Miros: ‘Bohuslav Martinu’, Musical Quarterly, xxix (1943). 

—— Bohuslav Martini: zivot a dilo (Prague, 1961; English translation, London, 
1964). 


Messiaen 

GoLEÁ, ANTOINE: Rencontres avec Olivier Messiaen (Paris, 1961). 

MESSIAEN, OLIVIER: La Technique de mon langage musicale (Paris, 1944; English 
translation, Chicago, 1957). 

SAMUEL, CLAUDE: Entretiens avec Olivier Messiaen (Paris, 1967). 


716 BIBLIOGRAPHY 


Milhaud 

Beck, GEORGES: Darius Milhaud. Étude suivi du catalogue chronologique complet 
de son ouvrage (Paris, 1949; Supplément (Paris, 1956). 

COLLAER, PAUL: Darius Milhaud (Antwerp, 1947). 

ТОРАТМКОЕЕ, NIKOLAI: “Christophe Colomb’, Modern Music, vii (1930). 

Mason, Coun: ‘The Chamber Music of Milhaud’, Musical Quarterly, xliii (1957). 

MILHAUD, Darius: ‘La mélodie’, Melos, iii (1922). —— 

—‘Polytonalité et atonalité, Revue musicale, iv (1923). 

----Ло1ев sans musique (Paris, 1949; 2nd ed., 1963; English translation, London, 

1952). 

Entretiens avec Claude Rostand (Paris, 1952). 


Petrassi 
WEISSMANN, JOHN: Goffredo Petrassi (Milan, 1957). 
——‘Goffredo Petrassi and his Music’, Music Review, xxii (1961). 


Pizzetti 

GATTI, Сошо Mania: Ildebrando Pizzetti (Turin, 1934; 2nd ed., Milan, 1955; 
English translation, London, 1951). 

GAVAZZENI, GIANANDREA: Altri studi pizzettiani (Bergamo, 1956). 

—Tre studi di Pizzetti (Como, 1937). 


Poulenc 

Durey, Louis: ‘Francis Poulenc’, The Chesterian, xxv (1922). 

HELL, Немкі: Francis Poulenc, musicien francais (Paris, 1958; English translation, 
London, 1959). 

Roy, JEAN: Francis Poulenc: l'homme et son oeuvre (Paris, 1964). 

SCHAEFFNER, ANDRE: ‘Francis Poulenc, musicien frangais’, Contrepoints, i (1946) 


Prokofyev 

BROWN, MALCOLM J.: The Symphonies of Sergei Prokofiev (Diss. Florida State 
Univ., 1967). 

NESTYEY, IZRAEL V.: Prokofyev (Moscow, 1957; English translation, Stanford, 
1960). 

—and С. EDELMAN (eds.): Sergey Prokofyev 1953-1963: stati i materiali 
(Moscow, 1962). 

SHLIFSHTEYN, S. I.: S. S. Prokofyev: materiali, dokumenti, vospominaniya (Moscow, 
2nd ed. 1961; English translation, Moscow, 1960). 

———S. S. Prokofyev: notografichesky spravochik (Moscow, 1962). 


Ravel 

FARGUE, LÉoN-PAUL: Maurice Ravel (Paris, 1949). 

JouRDAN-MORHANGE, HELENE: Ravel et nous: l'homme, l'ami, le musicien (Geneva, 
1945). 

МАСНАВЕУ, ARMAND: Maurice Ravel (Paris, 1947). 

ROLAND-MANUEL, ALEXIS: Maurice Ravel (Paris, 1938; English translation, Lon- 
don, 1947). 

— Maurice Ravel et son oeuvre dramatique (Paris, 1928). 

SEROFF, VICTOR: Maurice Ravel (New York, 1953). 


Respighi 

Мил, Massimo: ‘Probleme di gusto ed arte іп Ottorino Respighi’, Rassegan 
musicale, vi (1933). 

Екөрісні, ErLsA: Ottorino Respighi: dati biografici ordinati (Milan, 1954; English 
translation, London, 1962). 

RiNALDI, Mario: ‘Ottorino Respighi’, Musica d'oggi, iv (1961). 


BIBLIOGRAPHY 717 


Satie 

APOLLINAIRE, GUILLAUME: Selected Writings (English translation, New York, 1948). 

AUSTIN, WILLIAM W.: ‘Satie before and after Cocteau’, Musical Quarterly, xlviii 
(1962). 

COCTEAU, JEAN: Fragments d'un conférence sur Erik Satie (1920; repr. Liège 1957 
and Music Review, у (1944) ). 

EcorCHEVILLE, JULES: ‘Erik Satie’, Bulletin de la Société internationale de musique, 
vii (1911). 

Myers, Кошо H.: Erik Satie (London, 1948). 

TEMPLIER, PIERRE-DANIEL: Erik Satie (Paris, 1932). 


Schoenberg . 

ADORNO, THEODOR WIESENGRUND: ‘Arnold Schönberg’, Die grossen Deutschen, 
iv (1957). 

BonETZ, BENJAMIN and Epwanp T. Соме (eds.): Perspectives on Schoenberg апа 
Stravinsky (Princeton, N.J., 1968). 

BROEKEMA, ANDREW J.: А Stylistic Analysis and Comparison of the Solo Vocal 
Works of Arnold Schoenberg, Alban Berg, and Anton Webern (Diss. Univ. of 
Texas, 1962). 

BUCHANAN, HERBERT B.: “А key to Schoenberg's Erwartung (Op. 17), Journal of 
the American Musicological Society, xx (1967). 

CLIFTON, THOMAS J.: Types of Ambiguity in the Tonal Compositions of Arnold 
Schoenberg (Diss. Stanford Univ., 1966). 

EPsrEIN, Davip M. Schoenberg’s ‘Grundgestalt’ and Total Serialism: Their Relevance 
to Homophonic Analysis (Diss. Princeton Univ., 1968). 

FRIEDHEIM, PHILIP: ‘Rhythmic Structure in Schoenberg's Atonal Compositions’, 
Journal of the American Musicological Society, xix (1966). 

Нил, RICHARD S.: ‘Schoenberg’s Tone-Rows and the Tonal System of the Future’, 
Musical Quarterly, xxii (1936). 

LEIBOWITZ, RENE: Introduction à la musique de douze sons (Paris, 1949). 

—— Qu'est-ce que la musique de douze sons? (Liège, 1948). 

——Schoenberg et son école: l'étape contemporaine du langage musical (Paris, 1947: 
English translation, New York, 1949). 

LESTER, JOEL: A Theory of Atonal Prolongations as used in ап Analysis of the 
Serenade, Op. 24 by Arnold Schoenberg (Diss. Princeton Univ., 1970). 

NAMENWIRTH, SIMON M.: Twenty Years of Schoenberg Criticisms: Changes in the 
evaluation of once unfamiliar music (Diss. Univ. of Minnesota, 1965). 

ODEGARD, PETER S.: The Variation Sets of Arnold Schoenberg (Diss. Univ. of Cali- 
fornia, Berkeley, 1964). 

Pisk, PAUL A.: ‘Schoenberg’s Twelve-Tone Opera’, Modern Music, vii (1930). 

RUFER, JOSEF: Komposition mit zwölf Töne (Berlin, 1952; English translation, 
London, 1954). 

Das Werk Arnold Schónbergs (Kassel, 1959; English translation, New York, 

1963). 

SCHOENBERG, ARNOLD: Style and Idea (New York, 1950). 

‘Structural Functions of Harmony, ed. H. Searle (New York, 1954). 

STEFAN, PAUL: ‘Schoenberg’s Operas’, Modern Music, ii (1925) and vii (1929-30). 

STUCKENSCHMIDT, HANS HEINZ: Arnold Schónberg (Zürich, 1951; 2nd ed. 1957; 
English translation, New York, 1960). 

SUDERBURG, ROBERT C.: Tonal Cohesion in Schoenberg’s Twelve-Tone Music 
(Diss. Univ. of Pennsylvania, 1966). 

WOÓRNER, KARL H.: Gotteswort und Magie. Die Oper ‘Moses und Aaron’ von Arnold 
Schönberg (Heidelberg, 1959; English translation, Schoenberg’s ‘Moses апа 
Aaron’, New York, 1964). 


—‘Arnold Schoenberg and the Theatre’, Musical Quarterly xlviii (1962). 
47 


718 BIBLIOGRAPHY 


Stravinsky 

BLITZSTEIN, Marc: “Тһе Phenomenon of Stravinsky’, Musical Quarterly, xxi (1935). 

BonETZ, BENJAMIN and ЕрууАвр T. Соме (eds.): Perspectives on Schoenberg and 
Stravinsky (Princeton, N.J., 1968). 

CRAFT, ROBERT: Stravinsky: Chronicle of a Friendship 1948-1971 (New York, 1972). 

FLEISCHER, HERBERT: Stravinsky (Berlin, 1931). 

FREDERICKSON, LAWRENCE: Stravinsky’s Instrumentation: A Study of his Orchestral 
Techniques (Diss. Univ. of Illinois, 1960). 

HANDSCHIN, JACQUES: Igor Strawinsky : Versuch einer Einführung (Zurich, 1933). 

KIRCHMEYER, HELMUT: Igor Stravinsky: Zeitgeschichte im Personlichkeitsbild: 
Grundlagen und Voraussetzungen zur modernen Konstructionstechnik (Regens- 
burg, 1958). 

LANG, PAUL HENRY (ed.): Stravinsky: A New Appraisal of His Work (New York, 
1963). 

NABOKOV, NicHOLAs: Igor Strawinsky (Berlin, 1964). 

RAMUZ, CHARLES FERDINAND: Souvenirs sur Igor Strawinsky (Paris, 1929; 2nd ed., 
Lausanne, 1952). 

SCHAEFFNER, ANDRE: Strawinsky (Paris, 1931). 

STRAVINSKY, Тсов: Autobiography (New York, 1962). 

—— Poétiques musicales (Cambridge, Mass., 1942; English translation, Poetics of 
Music, New York, 1956). 

VLAD, RoMAN: Strawinsky (Turin, 1958; English translation, London, 1960), 

WADE, CARROLL D.: “А selected Bibliography of Igor Stravinsky', Musical Quarterly, 
xlviii (1962); repr. in Paul Henry Lang, ed., Stravinsky: A New Appraisal of his 
Work (New York, 1963). 

White, ERIC WALTER: Stravinsky: The Composer and His Works (London, 1966). 


Szymanowski 

GAVEZZENI, GIANANDREA: ‘Karol Szymanowskie il Re Ruggiero’, Rassegna musicale, 
x (1937). 

JACHIMECKI, ZDZISEAW: Karol Szymanowski (Cracow, 1927). 


Varése 

Сноу WEN-CHUNG: ‘Varèse: А Sketch of the Man and his Music’, Musical 
Quarterly, lii (1966). 

CowELL, HENRY: “Тһе Music of Edgard Varèse’, Modern Music, у (1928). 

VARESE, Louise: Varèse: A Looking-Glass Diary. Vol. Т: 1883-1928 (New York, 
1972). 

VIVIER, ODILE: ‘Innovation instrumentale d'Edgar Varése', Revue musicale, ccxxvi 
(1955). 


Webern 

ANTHONY, DoNALD B.: Microrhythm in the Published Works of Anton Webern 
(Diss. Stanford Univ., 1968). 

Anton Webern, special issue of Die Reihe, ii (German ed., 1955; English ed., 1958). 

Brown, Ковевт B.: The Early Atonal Music of Anton Webern: Sound Material 
and Structure (Diss. Brandeis Univ., 1965). 

KaARKOsCHKA, ERHARD: Studien zur Entwicklung der Kompositiontechnik іт 
Friihwerk Anton Weberns (Diss. Tiibingen, 1959). 

KOLNEDER, WALTER: Anton Webern: Einführung in Werk und Stil (Rodenkirchen, 
1961; English translation, London, 1968). 

McKENZIE, WALLACE CHESSLEY: Zhe Music of Anton Webern (Diss. North Texas 
State College, 1960). 

PERLE, GEORGE: ‘Webern’s Twelve-Tone Sketches’, Musical Quarterly, lvii (1971). 

REICH, WILLI (ed.): Anton Webern: Weg und Gestalt. Selbstzeugnisse und Worte der 
Freunde (Zurich, 1961). 


BIBLIOGRAPHY 719 


WEBERN, ANTON: Der Weg zur neuen Musik, ed. by Willi Reich (Vienna, 1960; 
English translation, The Path to the New Music, Bryn Mawr, Pa., 1963). 
Webern Archive of the Moldenhauer Collection at the University of Washington 
(Seattle, 1963). (Now located at Northwestern University, Evanston, Ill.) 


Weill 

BLITZSTEIN, Marc: “Оп Mahagonny’, The Score (1958). 

Drew, Davip: *Topicality and the Universal: The Strange Case of Weill's Die 
Bürgschaft', Music and Letters, xxxix (1958). 

——‘Weill’s School Opera Der Jasager’, Musical Times, cvi (1965). 

GUTMAN, Hans: ‘Mahagonny and Other Novelties’, Modern Music, vii (1930). 

KOTSCHENREUTHER, HELLMUT: Kurt Weill (Berlin, 1962). 

STEFAN-GRUENFELDT, PAUL: ‘Antinomie der neuen Oper: Kurt Weill und Strawin- 
sky’, Musikblatter des Anbruch, x (1928). 

Torksponr, CACILIE: John Gay's 'Beggar's Opera’ und Bert Brechts ‘Dreigroschenoper’ 
(Diss. Bonn Univ., 1934). 


CHAPTER V 


MUSIC OF THE EUROPEAN MAINSTREAM 1940-1960 


(i) General 

ADORNO, T. W.: Philosophie der neuen Musik (Tübingen, 1948). 

— —'Modern Music is growing old’, The Score, xviii (1956). 

BASART, ANN PHILLIPS: Serial Music: a Classified Bibliography on 12-tone and 
Electronic Music (Berkeley, 1961). 

Cross, ANTHONY: “Тһе Significance of Aleatoricism in twentieth-century Music’, 
Music Review, xxix (1968). 

EVANS, PETER: ‘Compromises with Serialism', Proceedings of the Royal Musical 
Association, Іххху (1962). 

FORTE, ALLEN: Contemporary Tone Structures (New York, 1955). 

GERHARD, ROBERTO: ‘Developments in twelve-note technique’, The Score, xvii 
(1956). 

——-'Tonality in twelve-note music’, The Score, vi (1952). 

KRENEK, ERNsT: *New developments of the twelve tone technique', Music Review, 
iv (1943). 

——— Extents and limits of serial techniques’, Musical Quarterly, xlvi (1960). 

LANG, PAUL HENRY and BRODER, NATHAN (eds.): Contemporary Music in Europe 
(New York, 1965). 

КОЕЕВ, ЈОЅЕЕ: Composition with Twelve Notes related only to one another (London, 
1954). 

Ruwert, NicoLas: ‘Contradictions within the serial language’, Die Reihe, vi (1964). 

VLAD, ROMAN: Модегний e tradizione nella musica contemporanea (Turin, 1955). 

Хиллс, WINFRIED: Variationen über neue Musik (Munich, 1959; reissued as Die 
neue Musik: Linien und Portráts, 1963). 


(ii) Individual Composers 


Boulez 

BOULEZ, PIERRE: ‘Propositions’, Polyphonie, ii (1948). 

— АЕ the ends of fruitful land’, Die Reihe, i (1955). 

— ‘Sonate, que me veux-tu?’, Perspectives of New Music, i (1963). 
——‘Alea’, Perspectives of New Music, iii (1964). 

— — Penser la musique aujourd'hui (Paris, 1964; English translation, Boulez on 
Music Today, London, 1971). 

GOLEA, ANTOINE: Rencontres avec Pierre Boulez (Paris, 1958). 


720 BIBLIOGRAPHY 


Dallapiccola 

DALLAPICCOLA, Їллїсї: “Оп Ше twelve-note road’, Music Survey, iv (1951). 

NATHAN, Hans: “Тһе twelve-tone compositions of Luigi Dallapiccola’, Musical 
Quarterly, xliv (1958). 

—‘Luigi Dallapiccola’ (conversations), Music Review, xxvii (1966). 

VLAD, Комам: Luigi Dallapiccola (Milan, 1957). 


Egk 5 
WOÓRNER, К. H.: ‘Egk and Orff’, Music Review, xiv (1953). 


Gerhard 
Drew, Davip: ‘Roberto Gerhard—the musical character’, The Score, xvii (1956). 


Henze 

HENZE, HANS WERNER: Essays (Mainz, 1964). 

Hindemith 

BRINER, ANDREAS: Paul Hindemith (Zürich, 1971). 

HINDEMITH, PAUL: The Craft of Musical Composition (London, 1942). 

—— —A Composer's World (New York, 1952). 

КЕМР, IAN: Hindemith (London, 1970). 

LANDAU, VICTOR: ‘Paul Hindemith: a Case Study in Theory and Practice’, Music 
Review, xxi (1960). 

—— ‘Hindemith the System Builder: a Critique’, Music Review, xxii (1961). 

REDLICH, H. F.: ‘Paul Hindemith: a Reassessment’, Music Review, xxv (1964). 

STEPHAN, RUDOLPH: ‘Hindemith’s Marienleben (1922-48): an assessment of its 
two versions’, Music Review, xv (1954). 

STROBEL, HEINRICH: Paul Hindemith (3rd edition, Mainz, 1948). 

THOMPSON, WILLIAM: ‘Hindemith’s Contribution to Music Theory’, Journal of 
Music Theory’, ix (1965). 


Ligeti 
LIGETI, GYÖRGY: ‘Metamorphoses of Musical Form’, Die Reihe, vii (1965). 


Messiaen 

Drew, рлур: ‘Messiaen—a provisional study’, The Score, x, xii and xiv (1954-5). 
GoLÉA, ANTOINE: Rencontres avec Olivier Messiaen (Paris, 1961). 

MESSIAEN, OLIVIER: Technique de mon langage musicale (Paris, 1944). 


Orff 
WOÓRNER, K. H.: ‘Egk and Orff’, Music Review, xiv (1953). 


Petrassi 
WEISSMANN, JOHN: Goffredo Petrassi (Milan, 1957). 


Pousseur 
PoUssEUR, HENRI: ‘Music, Form and Practice’, Die Reihe, vi (1964). 
—‘Outline of a Method’, Die Reihe, iii (1959). 


Schaeffer E 
SCHAEFFER, PIERRE: А [а recherche d'une musique concréte (Paris, 1952). 


Schoenberg 
See Bibliography to Chapter IV. 


Skalkottas 
PAPAIOANNOU, YIANNIS: ‘Nikos Skalkottas', in Hartog (ed.), European Music in 
the Twentieth Century (2nd ed., London, 1961). 


BIBLIOGRAPHY 721 


Stockhausen 

HARVEY, JONATHAN: ‘Stockhausen: Theory and Music’, Music Review, xxix (1968). 
MARCUS, GENEVIEVE: ‘Stockhausen’s Zeitmasse', Music Review, xxix (1968). 
STOCKHAUSEN, KARLHEINZ: ‘How time passes . . .^, Die Reihe, iii (1959). 

— — Electronic and Instrumental Music’, Die Reihe, v (1961). 

— —'Music in Space’, Die Reihe, v (1961). 

— — Music and Speech’, Die Reihe, vi (1964). 

— "The Concept of Unity in Electronic Music’, Perspectives of New Music, 1 (1962). 
WORNER, К. H.: Karlheinz Stockhausen, Werk und Wollen (Rodenkirchen, 1963). 


Stravinsky 

ВАввит, Милом: ‘Remarks on the recent Stravinsky’, Perspectives of New Music, 
ii (1964). 

Cong, E. T.: “Тһе Uses of Convention: Stravinsky and his Models’, Musical 
Quarterly, xlviii (1962). 

— —'Stravinsky: the Progress of a Method’, Perspectives of New Music, i (1962). 

GERHARD, ROBERTO: ‘Twelve-note technique in Stravinsky’, The Score, xx (1957). 

PAULI, HANsJORG: ‘On Stravinsky's Threni', Tempo, xix (1958). 

STEIN, Erwin: ‘Stravinsky’s Septet—an Analysis’, Tempo, xxxi (1954). 

STRAVINSKY, IGOR and CRAFT, ROBERT: Conversations with Igor Stravinsky (London, 
1959). 

— Memories and Commentaries (London, 1960). 

——Expositions and Developments (London, 1961). 

——Dialogues and a Diary (London, 1963). 

See also Bibliography to Chapter IV. 


Varèse 

Влввітт, MILTON: ‘Edgard Varèse: а few observations of his music’, Perspectives 
of New Music, iv (1966). 

Сноу WEN-CHUNG: ‘Varèse: a sketch of the man and his music’, Musical Quarterly, 
lii (1966). 

WHITTALL, ARNOLD: ‘Varése: and organic athematicism’, Music Review, xxviii, 
(1967). 

WILKINSON, Marc: ‘An Introduction to the Music of Edgard Varése', The Score, 
xix (1957). 


Webern 
See Bibliography to Chapter IV. 


CHAPTER VI 


MUSIC IN BRITAIN: 1918-1960 


(i) General 

BACHARACH, A. L. (ed.): British Music of Our Time (Harmondsworth, 2nd ed., 
1951). 

Вгом, Eric: Music in England (Harmondsworth, rev. ed., 1947). 

Boosey & HAWKES (publishers): ‘England-Heft’ (Bonn, n.d., c. 1953), Musik der 
Zeit, Heft iv. 

BonNorr, JACK: ‘Musikleben in England’, Melos (March 1950). 

Bourr, в ADRIAN: Му Own Trumpet (London, 1973). 

BRADBURY, ERNEST: ‘Modern British Composers’, in Choral Music, ed. Arthur 
Jacobs (Harmondsworth, 1963). 

BRITTEN, BENJAMIN: ‘England and the Folk-Art Problem’, Modern Music (1940). 

CASSINI, LEONARD: ‘Neue Englische Musik’, Musik und Gesellschaft, ix (1956). 

FOREMAN, R. L. E.: The British Musical Renaissance: a Guide to Research, (thesis, 
London, 1972). 


722 BIBLIOGRAPHY 


Foss, HERBERT and GOODWIN, NoEL: London Symphony (London, 1954). 

FRANK, ALAN: Modern British Composers (London, 1953). 

Harris, Rex: “Тһе Influence of Jazz on English Composers’, Penguin Music 
Magazine (1947). 

Howes, FRANK: The English Musical Renaissance (London, 1966). 

Lez, Epwanp: Music of the People: a study of popular music in Great Britain 
(London, 1970). 

LOCKSPEISER, EDWARD: “Trends in Modern English Music’, Musical Quarterly, 
xxviii (1942). 

MANNING, RosEMARY: From Holst to Britten (London, 1949). 

MELLERS, WILFRID: ‘Recent Trends in British Music’, Musical Quarterly, xxxviii 
(1952). 

MILNER, ANTHONY: ‘English Contemporary Music’, in European Music in Ше 
Twentieth Century, ed. Hartog, (London, 1957). 

PALMER, CHRISTOPHER: “Тһе Post Impressionists in England’, Impressionism іп 
Music (London, 1973). 

PORTER, ANDREW: ‘Some New British Composers’, in Contemporary Music іп 
Europe: a comprehensive survey, eds. Lang and Broder, (New York, 1965). 

Raynor, HENRY: ‘Influence and Achievement: some thoughts on Twentieth- 
Century English Song’, Chesterian, xxx (1956). 

RourH, FRANCIS: Contemporary British Music: Britain's Musical Tradition since the 
War (London, 1972). 

SEARLE, HUMPHREY: ‘Growing Pains in England’, Modern Music (1939). 

SEARLE, HUMPHREY and LAYTON, ROBERT: Britain, Scandinavia and the Netherlands 
(Twentieth Century Composers, vol. v) (London, 1972). 

THE Times: Musical Britain in 1951 (London, 1951). 

THOMPSON, KENNETH: A Dictionary of Twentieth-Century Composers (1911-1971) 
(London, 1973). 

Woop, HuGu: ‘English Contemporary Music’, in, European Music in the Twentieth 
Century, ed. Hartog (Harmondsworth, paper-back edition, 1961). 


(ii) Individual Composers 


Alwyn 
Нор, Trevor: ‘The Music of William Alwyn’, Composer (1972). 


Arnold 
MITCHELL, DONALD: ‘Malcolm Arnold’, Musical Times, xcvi (1955). 


Bantock 
ANDERTON, Н. Овзмомр: Granville Bantock (London, 1915). 
BANTOCK, Муввна: Granville Bantock: a personal portrait (London, 1972). 


Bax 

FOREMAN, R. L. E.: ‘Bibliography of Writings on Arnold Bax’, Current Musicology, 
No. 10 (1970). 

Нот, Rosin, M.: A Handbook on Arnold Вах’; Symphonies (London, 1933). 

SCOTT SUTHERLAND, COLIN: Arnold Bax (London, 1973). 


Benjamin 

BOUSTEAD, ALAN: ‘Arthur Benjamin and Opera’, Opera (1964). 

KELLER, HANS: ‘Arthur Benjamin and the problem of popularity: a critical apprecia- 
Поп”, Tempo, xi (1950). 


Bennett, Richard Rodney 
Maw, NicHOLAS: ‘Richard Rodney Bennett’, Musical Times, ciii (1962). 


BIBLIOGRAPHY 729 


Berkeley 

DICKINSON, PETER: “Тһе Music of Lennox Berkeley’, Musical Times, сі (1963). 
—— Lennox Berkeley’, Music and Musicians, xiii (1965). 

—‘Berkeley’s Music Today’, Musical Times, cix (1968). 

HULL, Еовім: “Тһе Style of Lennox Berkeley', Chesterian, xxiv (1950). 
Керисн, Hans F.: ‘Lennox Berkeley’, Music Survey (1951). 


Bliss 

Ві155, SIR ARTHUR: As I Remember (London, 1970). 

HASKELL, ARNOLD L.: Miracle in the Gorbals: a study (Edinburgh, 1946). 

ScHOLES, Percy A.: А Few Notes upon the Work of Arthur Bliss and Especially 
upon his Colour Symphony (London, 1922). 

THOMPSON, KENNETH L.: ТЛе Works of Arthur Bliss (London (rev. ed.) n.d. (1971) ). 
Originally published in Musical Times. 


Boughton 
Ново, MICHAEL: Immortal Hour: the life and period of Rutland Boughton (London, 
1962). 


Brian 

FOREMAN, Lewis: ‘Havergal Brian: a new view’, Composer (1971-2). 

MACDONALD, MALCOLM: Havergal Brian: perspective on the music (London, 1972). 

NETTEL, REGINALD: The Life and Music of Havergal Brian (London, 1937). A revised 
and enlarged version of Ordeal by Music (London, 1945). 


Bridge 
PAYNE, ANTHONY: ‘The Music of Frank Bridge’, Tempo, xxxiv (1973). 
Равте, PETER J.: Frank Bridge (London, 1971). 


Britten 

Brown, Davin: ‘Britten’s Three Canticles’, Music Review, xxi (1960). 

KENDALL, ALAN: Benjamin Britten (London, 1973). 

MITCHELL, DONALD and KELLER, Hans (eds.): Benjamin Britten: a commentary 
on his works from a group of specialists (London, 1952). 

WHITE, Eric WALTER: Benjamin Britten—his life and operas (London, 1970). 
Revised and expanded version of Benjamin Britten—a sketch of his life and 
works (London, 1948; 2nd ed., 1954). 


Bush, Alan 

STEVENSON, RONALD: ‘Alan Bush—committed composer’, Music Review, xxv 
(1964). 

Tribute to Alan Bush on his fiftieth birthday—a symposium. (London, 1950). 


Cooke 
CLAPHAM, JOHN: ‘Arnold Cooke: the achievement of 20 years’, Music Survey 
(1951). 


Elgar 

See Bibliography to Chapters I and IT. 

Finzi 

FERGUSON, HOWARD: ‘Gerald Finzi (1901-1956), Music and Letters, xxxviii (1957). 
Fricker 


Воотн, FRANcIs: ‘Peter Racine Fricker’, in Contemporary British Music (London, 
1972). 


Gerhard 
*Gerhard' issue of 77ie Score, xvii (1956). 


724 BIBLIOGRAPHY 


Goossens 
GOOSSENS, EUGENE: Overture and Beginners: (London, 1951). 


Gurney 
*Gurney' number of Music and Letters, xix (1938). 
Вовтсн, M. A.: ‘Ivor Gurney—a revaluation’, Musical Times, xcvi (1955). 


Hadley - 
PALMER, CHRISTOPHER: ‘The Music of Patrick Hadley’, Musical Times, cx (1969). 


Hamilton 
MILNER, ANTHONY: ‘Some Observations on the Music of Ian Hamilton’, Musical 
Times, xcvii (1956). 


Holst 

Horsr, IMOGEN: Gustav Holst—a biography (London, 2nd ed., 1969). 

The Music of Gustav Holst (London, 3rd ed., 1974). 

ME ters, W. H.: ‘Holst and the English Language’, in Studies in Contemporary 
Music (London, 1947). 

RUBBRA, EDMUND: Gustav Holst (Monaco, 1947). 
‘Gustav Holst as Teacher’, Monthly Musical Record, lx (1930). 

ТІРРЕТТ, MICHAEL: ‘Holst, Figure of Our Time’, Listener (13/11/1958). 

Tovey, DONALD F.: Essays in Essays in Musical Analysis, 11, iv and v (London, 
1936-7). 

WARRACK, JOHN: ‘A New Look at Holst’, Musical Times, civ (1963). 


Howells 

PALMER, CHRISTOPHER: ‘Herbert Howells at 80—a retrospect’, Musical Times, 
cxiii (1972). 

SPEARING, ROBERT: HH—Herbert Howells . . . (London, 1972). 


Ireland 

DICKINSON, А. E. Е.: “Тһе Progress of John Ireland', Music Review, i (1940). 

HOLBROOKE, JoseF: ‘John Ireland’, in Contemporary British Composers (London, 
1931). 

LONGMIRE, JOHN: John Ireland—Portrait of a Friend (London, 1969). 


Lambert 

McGrapy, RICHARD: ‘The Music of Constant Lambert’, Music and Letters, li 
(1970). 

SHEAD, RICHARD: Constant Lambert (London, 1973). 

Leigh 

WIMBUSH, Косек: ‘Walter Leigh’, Monthly Musical Record, Ixviii (1938). 


Leighton 
Соскѕноот, JOHN V.: ‘The Music of Kenneth Leighton’, Musical Times, xcviii 
(1957). 


Lutyens 
LUTYENS, ELIZABETH: A Goldfish Bowl (London, 1972). 


Maconchy 

‘Living British Composers—Elizabeth Maconchy’, Hinrichsen Musical Yearbook 
No. 6 (1949/50). 

MACONCHY, ELIZABETH: ‘A Composer Speaks’, Composer (1971-2). 

Milner 

BRADBURY, ERNEST: ‘The Progress of Anthony Milner’, Musical Times, civ (1963). 


BIBLIOGRAPHY 125 


Моегап 
FLEISCHMANN, ALOYS: “Тһе Music of E. 7. Moeran', Envoy (1951). 


Musgrave 
BRADSHAW, SUSAN: ‘Thea Musgrave’, Musical Times, civ (1963). 


Rawsthorne 

BERKELEY, LENNOX: ‘Alan Rawsthorne—1', Composer (1971-2). 

GREEN, GORDON: ‘Alan Rawsthorne—2’, Composer (1972). 

Howe ts, HERBERT: ‘A Note on Alan Rawsthorne’, Music and Letters, xxxii (1951). 

MELLERS, WILFRID: ‘Alan Rawsthorne and the Baroque’, in Studies in Contemporary 
Music (London, 1947). 


Rubbra 

Dawney, MICHAEL: ‘Edmund Rubbra and the Piano’, Music Review, xxxi (1970). 

HUTCHINGS, ARTHUR: ‘Edmund Rubbra’s Second Symphony’, Music and Letters, 
xx (1939). 

MELLERS, WILFRID: ‘Rubbra and the Dominant Seventh’, in Studies in Contemporary 
Music (London, 1947). 

Orraway, Носн: ‘Rubbra’s Symphonies’, Musical Times, cxii (1971). 

PAYNE, Este: ‘Edmund Rubbra', Music and Letters, xxxvi (1955). 


Scott 

DEMUTH, NORMAN: ‘Cyril Scott’, Musical Opinion, lxxx (1957). 

Ном, A. EAGLEFIELD: Cyril Scott: Composer, Poet and Philosopher (London, 1971). 
Scorr, Суви.: Bone of Contention (London, 1969). 


Searle 
LOCKsPEISER, EDWARD: ‘Humphrey Searle’, Musical Times, cxvi (1955). 
RAYMOND, MALCOLM: ‘Searle—avant garde or romantic’, Musica! Times, cv (1964). 


Seiber 
KELLER, Hans: ‘Matyas Seiber', Musical Times, xcvi (1955). 
—‘Matyas Seiber 1905-1960’, Tempo, xxi (1960). 


Simpson 
JoHNsON, EDWARD (ed.): Robert Simpson—fiftieth birthday essays (London, 1971). 


Tate 
CARNER, Mosco: “Тһе Music of Phyllis Tate', Music and Letters, xxxv (1954). 


Tippett 

ATKINSON, NEVILLE: ‘Michael Tippett's Debt to the Past’, Music Review, xxiii 
(1962). 

Kemp, IAN (ed.): А Symposium for Michael Tippett’s 60th Birthday (London, 1965). 

MILNER, ANTHONY: “Тһе Music of Michael Tippett’, Musical Quarterly, 1 (1964). 


Van Dieren 
APIVOR, DENIS: Bernard van Dieren, Music Survey (June 1951). 


Vaughan Williams 

‘Vaughan Williams Centenary’ number of Opera, (1972). 

‘Ralph Vaughan Williams 1872-1958, memorial number of R.C.M. Magazine, 
(1959). 

Day, JAMES: Vaughan Williams (London, 1961). 

DICKINSON, А. E. F.: Vaughan Williams (London, 1963). 

Douczas, Roy: Working with R.V.W. (London, 1972). 

FOREMAN, Lewis: ‘V.W.—a bibliography of dissertations’, Musical Times, cxiii 
(1972). 


726 BIBLIOGRAPHY 


Foss, HUBERT: Ralph Vaughan Williams—a study (London, 1950). (Includes “А 
Musical Autobiography’ by Vaughan Williams.) 

KENNEDY, MICHAEL: The Works of Ralph Vaughan Williams (London, 1964). 

OrrAWwAY, Носн: Vaughan Williams: Symphonies (London, 1972). 

PAYNE, ELsiE M.: ‘Vaughan Williams’ Orchestral Colourings’, Monthly Musical 
Record (1954). 

——‘Vaughan Williams and Folk-Song', Music Review, xv (1954). 

SCHWARTZ, ELLIOTT S.: The Symphonies of Ralph Vaughan Williams (Amherst, 
Mass., 1964). 

VAUGHAN WILLIAMS, UnsULA: R.V.W.: a biography of Ralph Vaughan Williams 
(London, 1964). 

Young, Percy M.: Vaughan Williams (London, 1953). 


Walton 

*William Walton at 75” number of Musical Times, cxiii (1972). 

CRAGGS, STEWART: Sir William Walton, O.M.: a catalogue, bibliography and 
discography (Thesis, London, 1973). 

Foss, Новект: *William Walton', Тле Chesterian, xi (1930). 

—— —' William Walton’, Musical Quarterly, xxvi (1940). 

Howes, FRANK: The Music of William Walton (London, 1965; 2nd ed. 1973). 

LAMBERT, CONSTANT: ‘Some Recent Works by William Walton’, The Dominant, i 
(1928). 

MITCHELL, DoNALD: ‘Some Observations on William Walton’, The Chesterian, 
xxvi (1952). 

Tovey, DONALD F.: *Walton's Viola Concerto’, in Essays in Musical Analysis, iii 
(London, 1936). 


Warlock 

CoPLEY, IAN A.: ‘Peter Warlock’s Vocal Chamber Music’, Music and Letters, 
xliv (1963). 

—The Published Instrumental Music of Peter Warlock’, Music Review, xxv 
(1964). 


——‘Peter Warlock’s Choral Music’, Music and Letters, xlv (1964). 
GRAY, CECIL: Peter Warlock—a memoir of Philip Heseltine (London, 1934). 
VAN DIEREN, BERNARD: ‘Philip Heseltine’, Musical Times, Ixxii (1931). 


Wellesz 

MELLERS, WILFRID: ‘Egon Wellesz’, in Studies in Contemporary Music (London, 
1947). 

RIDLEY, ANTHONY: ‘The Later Works of Egon Wellesz’, Composer (1966). 

SCHOLLUM, ROBERT: Egon Wellesz—eine Studie (Vienna, 1964). 

Symons, Davin: ‘Egon Wellesz and Early Twentieth Century Tonality’, Studies in 
Music (No. 6, 1972). 


Williamson 
Mason, Coun: ‘The Music of Malcolm Williamson’, Musical Times, су (1962). 


CHAPTER VII 
AMERICAN MUSIC: 1918-1960 


(i) The United States 


(i) General 

BARZUN, JACQUES: Music in American Life (New York, 1956). 
CHASE, GILBERT: America’s Music (2nd ed., New York, 1967). 
COoPLAND, AARON: Our New Music (New York, 1941). 


BIBLIOGRAPHY 727 


COoPLAND, AARON: Copland on Music (New York, 1960). 

COWELL, HENRY: New Musical Resources (New York, 1930). 

———(ed.): American Composers on American Music (Stanford, 1933). 

Goss, MADELEINE: Modern Music-Makers: Contemporary American Composers 
(New York, 1952). 

Нітснсоск, WILEY: Music in the United States: a Historical Introduction (New York, 
1969). 

HOWARD, JOHN TASKER: Our American Music (3rd ed., New York, 1946). 

LAHEE, HENRY C.: Annals of Music in America (Boston, 1922). 

LANG, PAUL HENRY (ed.): 700 Years of Music in America (New York, 1961). 

MELLERS, WILFRID: Music in a New Found Land (London, 1964). 

REIS, CLAIRE: Composers in America (New York, 1947). 

— Composers, Conductors and Critics (New York, 1955). 

ROSENFELD, PAUL: Musical Chronicle 1917-1923 (New York, 1923). 

Discoveries of a Music Critic (New York, 1936). 

An Hour with American Music (Philadelphia, 1929). 

SABLOSKY, IRVING: American Music (Chicago, 1969). 

SAMINSKY, LAZARE: Living Music of the Americas (New York, 1949). 

THOMSON, VIRGIL: The Musical Scene (New York, 1945). 

— Music Right and Left (New York, 1951). 

—— —The State of Music (New York, 1939; 2nd ed., 1963). 


(ii) Individual Composers 


Babbitt 

Bruno, ANTHONY: ‘Two American Twelve-tone Composers’ (Milton Babbitt and 
Ben Weber), Musical America, xxxvii (1951). 

KOSTELANETZ, R.: ‘Two Extremes of Avant Garde Music’ (A Discussion of the 
music of Milton Babbitt), New York Times Magazine, 15 Jan. 1967. 


Barber 
BRODER, NATHAN: Samuel Barber (New York, 1954). 
——'The Music of Samuel Barber’, Music Quarterly, xxxiv (1948). 


Becker 
RIEGGER, WALLINGFORD: ‘John J. Becker’, Bulletin of the American Composers 
Alliance, 1x (1959). 


Bergsma 
SKULSKY, ABRAHAM: ‘William Bergsma’, The Juilliard Review, Spring 1956. 


Bernstein 
BRIGGS, JOHN: Leonard Bernstein: the Man, His Works and His World (Cleveland, 
1961). 


Brant 
SANKEY, STUART: ‘Henry Brant’s Grand Universal Circus’, The Juilliard Review, 
Fall 1956. 


Cage 
Пинк, Ковевт (ed.): John Cage (New York, 1962). 
KOSTELANETZ, RICHARD: John Cage (New York, 1970). 


Carter 

GOLDMAN, RICHARD FRANKO: ‘The Music of Elliott Carter’, Musical Quarterly, 
xliii (1957). 

SKULSKY, ABRAHAM: ‘Elliott Carter’, Bulletin of the American Composers Alliance, 
iii (1953). 


728 BIBLIOGRAPHY 


Copland 

BERGER, ARTHUR: Aaron Copland (New York, 1953). 

Corr, HuGo: ‘Aaron Copland’, Tempo, Spring and Summer, 1966. 

SMITH, JULIA: Aaron Copland: His Work and Contribution to American Music 
(New York, 1955). 


Cowell 
GERSCHEFSKI, EDWIN: ‘Henry Cowell’, Bulletin of the American Composers Alliance, 
iii (1953). 
GOLDMAN, RICHARD FRANKO: ‘Henry Cowell: A Memoir and an Appreciation’, 
Perspectives of New Music, Spring-Summer 1966. 
WEISGALL, Носо: “Тһе Music of Henry Cowell’, Musical Quarterly, xlv (1959). 


Creston 
CoweELL, HENRY: ‘Paul Creston’, Musical Quarterly, xxxiv (1948). 


Dello Joio 
Downes, Epwanp: “Тһе Music of Norman Dello Joio’, Musical Quarterly, xlviii, 
(1962). 


Finney 
Cooper, PAUL: ‘The Music of Ross Lee Finney’, Musical Quarterly, liii (1967). 


Gershwin 

ARMITAGE, MERLE: George Gershwin, Man and Legend (New York, 1958). 

GOLDBERG, IsAAc: George Gershwin, A Study in American Music (New ed. Supple- 
mented by Edith Garson. New York, 1958). 

RUSHMORE, ROBERT: The Life of George Gershwin (New York, 1966). 


Gilbert 
Downes, OLIN: ‘An American Composer’ (Henry F. B. Gilbert), Musical Quarterly, 
iv (1918). 


Griffes 

BAUER, MARION: ‘Charles T. Griffes as I Remember Him’, Musical Quarterly, 
xxix (1943). 

MAISEL, EDWARD M.: Charles T. Griffes: The Life of an American Composer 
(Cleveland, 1961). 


Hanson 
ALTER, MARTHA: ‘Howard Hanson’, Modern Music, xviii (1941). 
TUTHIL, BURNETT C.: ‘Howard Hanson’, Musical Quarterly, xxii (1936). 


Harris 
FARWELL, ARTHUR: ‘Roy Harris’, Musical Quarterly, xviii (1932). 


Harrison 
YATES, PETER: ‘Lou Harrison’. Bulletin of the American Composers Alliance ix (1960). 


Ives 

BELLAMANN, HENRY: ‘Charles Ives: The Man and His Music’, Musical Quarterly, 
xix (1933). 

COPLAND, AARON: ‘(Ives’s) One Hundred and Fourteen Songs’, Modern Music, 
xi (1934). 

CoweELL, HENRY and SIDNEY: Charles Ives and His Music (New York, 1955). 

GRUNFELD, FREDERIC: ‘Charles Ives ... Yankee Rebel’, Bulletin of the American 
Composers Alliance, iv (1955). 

STONE, KURT: ‘Ives’s Fourth Symphony: A Review’, Musical Quarterly, lii (1966). 

YATES, PETER: ‘Charles Ives’, Arts and Architecture, lxvii (1950). 


BIBLIOGRAPHY 729 


Kirchner 
RINGER, ALEXANDER L.: ‘Leon Kirchner’, Musical Quarterly, xliii (1957). 


Loeffler 

ENGEL, Car: ‘Charles Martin Loeffler’, Musical Quarterly, хі (1925). 

——‘News and Reviews’ (A eulogy of Charles Martin Loeffler), Musical Quarterly, 
xxi (1935). 


Mennin 
HENDL, WALTER: ‘Peter Mennin’, The Juilliard Review, Spring 1954. 


Ornstein 
BUCHANAN, C. L.: ‘Ornstein and Modern Music’, Musical Quarterly, iv (1918). 


Palmer 
AUSTIN, WILLIAM: ‘The Music of Robert Palmer’, Musical Quarterly, xlii (1956). 


Partch 
MELLERS, WILFRID: ‘An American Aboriginal’, Tempo, Spring 1963. 


Persichetti 
EvErT, Ковевт: ‘Vincent Persichetti’, The Juilliard Review, Spring 1955. 


Piston 
CARTER, ELLIOTT: ‘Walter Piston’, Musical Quarterly, xxxii (1946). 
Сіткоуу117, ISRAEL: ‘Walter Piston ... Classicist’, Modern Music, xiii (1936). 


Porter 
BoATWRIGHT, Ноуулво: ‘Quincy Porter’, Perspectives of New Music, Spring- 
Summer 1967. 


Riegger 

BECKER, JOHN J., CowELL, HENRY and GOLDMAN, RICHARD FRANKO: ‘Wallingford 
Riegger ... А Tribute’, Bulletin of the American Composers Alliance, ix (1960). 

GOLDMAN, RICHARD FRANKO: ‘The Music of Wallingford Riegger’, Musical 
Quarterly, xxxvi (1950). 


Rogers 
DiAMOND, Davip: ‘Bernard Rogers’, Musical Quarterly, xxxiii (1947). 


Ruggles 
SEEGER, CHARLES: ‘Carl Ruggles’, Musical Quarterly, xviii (1932). 
HARRISON, Lou: About Carl Ruggles (New York, 1946). 


Schuman 

BRODER, NATHAN: “Тһе Music of William Schuman’, Musical Quarterly, xxxi 
(1945). 

SCHREIBER, FLORA RHETA and PERSICHETTI, VINCENT: William Schuman (New 
York, 1954). 


Sessions 

Соме, EDWARD T.: ‘Conversation with Roger Sessions’, Perspectives of New Music, 
Spring-Summer 1966. 

IMBRIE, ANDREW: ‘Roger Sessions: In Honor of his Sixty-Fifth Birthday’, Perspec- 
tives of New Music, Fall 1962. 

SCHUBART, MARK A.: ‘Roger Sessions: Portrait of an American Composer’, 
Musical Quarterly, xxxii (1946). 


Thompson 
FORBES, Етллот: ‘The Music of Randall Thompson’, Musical Quarterly, xxxv (1949). 


730 BIBLIOGRAPHY 


"Thomson 

GLANVILLE-HiICKS, P.: “Virgil Thomson’, Musical Quarterly, xxxv (1949). 
Hoover, KATHLEEN: Virgil Thomson (New York, 1959). 

THOMSON, ViRGIL: Virgil Thomson (New York, 1966). 


Varèse 

BABBITT, MiLTON: ‘Edgard Varèse: A Few Observations of His Music’, Perspectives 
of New Music, Spring-Summer 1966. 

CowELL, HENRY: ‘The Music of Edgard Varèse’, Modern Music, Jan.-Feb. 1928. 

Сноу WrN-CHuNG: ‘Varèse: A Sketch of the Man and His Music’, Musical 
Quarterly, lii (1966). 

DALLAPICCOLA, LUIGI, CARTER, ELLIOTT, NIN, ANAIS, SZATHMARY, ARTHUR, 
FELDMAN, DorTON and WILKINSON, Marc: ‘In Memoriam: Edgard Varèse’, 
Perspectives of New Music, Spring-Summer 1966. 

WALDMAN, FREDERICK: ‘Edgard Varèse’, The Juilliard Review, Fall 1954. 

WILKINSON, Marc: ‘An Introduction to the Music of Edgar Varèse’, The Score, 
No. 19 (1957). 


Ward 
STAMBLER, BERNARD: ‘Robert Ward’, Bulletin of the American Composers Alliance, 
iv (1955). 


Weber 
BRUNO, ANTHONY: ‘Two American Twelve-tone Composers’ (Milton Babbitt and 
Ben Weber), Musical America, Feb. 1951. 


Weisgall 
ROCHBERG, GEORGE: ‘Hugo Weisgall’, Bulletin of the American Composers Alliance, 
vii (1958). 


(ii) Latin America 


(1) General 

ALMEIDA, RENATO: História da musica brasileira (Rio de Janeiro, 1942). 

AUZA, ATILIANO: Dinámica musical en Bolivia (La Paz, 1967). 

AZEVEDO, Luiz HEITOR CORREA DE: 150 Anos de música no Brasil (1800-1950) 

(Rio de Janeiro, 1956). 

Música е músicos do Brasil (Rio de Janeiro, 1950). 

CALCARO, José ANTONIO: La Ciudad y su musica, Crónica musical de Caracas 
(Caracas, 1958). 

CARPENTIER, ALEJO: La música en Cuba (Mexico, 1946). 

CHASE, GILBERT: А Guide to the Music of Latin America (Washington, D.C., 1962). 

ESCOBAR, ROBERTO: Músicos sin Pasado: Composición y compositores de Chile 
(Barcelona, 1971). 

Lisr, GEORGE & ORREGO-SALAS, JUAN (eds.): Music in the Americas (Bloomington, 
1967). 

MaRIZ, Vasco: Figuras da música brasileira contemporanea (Brasilia, 1970). 

MAYER-SERRA, Отто: Música y músicos de latinoamérica (Mexico, 1947). 

— Panorama de la música mexicana (Mexico, 1941). 

PERDOMO Е$СОВАВ, José I.: Historia de la musica en Colombia (Bogota, 1963). 

RAYGADA, CARLOS: ‘Guia musical del Pert’, Fénix, xii, xiii, xiv, (1956-64). 

SALAS VIU, VICENTE: La creación musical en Chile (Santiago, 1952). 

SALGADO, SUSANA: Breve historia de la música culta en el Uruguay (Montevideo, 
1971). 

SLONIMSKY, NicoLAs: Music of Latin America (New York, 1945). 

STEVENSON, RoBERT: Music in Mexico (New York, 1952). 


BIBLIOGRAPHY 731 


(ii) Individual composers 


Ardévol 

ARDEVOL, José: ‘El Grupo Renovación de La Habana’, Revista musical chilena, 
xxvii (1947). 

CARPENTIER, ALEJO: ‘La musica contemporánea de Cuba’, Revista musical chilena, 
xxvii (1947). 


Becerra 
MERINO, Luis: ‘Los cuartetos de Gustavo Becerra’, Revista musical chilena, xcii 
(1965). 


Chávez 

CHAVEZ, CARLOS: Musical Thought (Cambridge, Mass., 1961). 

GARCIA Моввпо, Ковевто: Carlos Chávez: vida y obra (Mexico, 1960). 

(HALFFTER, Roporro): Carlos Chavez, catálogo completo de sus obras (Mexico, 
1971). 


Cosme 
BÉHAGUE, GERARD: ‘Luiz Cosme (1908-1965): Impulso creador versus conciencia 
formal’, Yearbook (Inter-American Musical Research Institute), v (1969). 


Ginastera 

CHASE, GILBERT: ‘Alberto Ginastera: Argentine composer’, Musical Quarterly, 
xliii (1957). 

SUAREZ URTUBEY, РОГА: Alberto Ginastera (Buenos Aires, 1967). 


Mignone 
VERHAALEN, MARION: ‘Francisco Mignone: His music for piano’, Inter-American 
Music Bulletin, \xxix (1970-1). 


Orrego-Salas 
OnREGO-SALAS, JUAN: Pasado y presente de la música chilena (Santiago, 1960). 


Ponce 
Ponce, MANUEL M.: ‘Apuntes sobre música mexicana’, Boletin latino-americano 
de música, iii (1937). 


Paz 
PAZ, JUAN CARLOS: Introducción a la música de nuestro tiempo (Buenos Aires, 1955). 


Revueltas 
MAYER-SERRA, Orro: 'Silvestre Revueltas and Musical Nationalism in Mexico', 
Musical Quarterly, xxvii (1941). 


Santa Cruz 
SALAS VIU, VICENTE: ‘Las obras para orquesta de Domingo Santa Cruz’, Revista 
musical chilena, xlii (1951). 


Villa-Lobos 

Maniz, Vasco: Heitor Villa-Lobos (Rio de Janeiro, 1949). 

МоввЕСА, ADHEMAR: As Bachianas Brasileiras de Heitor Villa-Lobos (Rio de 
Janeiro, 1971). 

ORREGO-SALAS, JUAN: ‘Villa-Lobos: Man, Work and Style’, Inter-American 
Music Bulletin, lii (1966). 

PEPPERCORN, Lisa M.: Heitor Villa-Lobos. Leben und Werk des brasilianischen 
Komponisten (Ziirich, 1972). 


732 BIBLIOGRAPHY 
CHAPTER VIII 


MUSIC IN THE SOVIET UNION 


(1) General 

ASAFYEV, B. V.: Izbrannie trudi, v (Moscow, 1957). 

BERGER, KARLHANNS: Die Funktionsbestimmung der Musik in der Sowjetideologie 
(Wiesbaden, 1963). _ 

BERNANDT, С. and DoLzHANsKY, A.: Sovetskie kompozitori: kratkiy bibliogra- 
ficheskiy spravochnik (Moscow, 1957). 

GRINBERG, M. and PoLvAKOVA, М. (ed.): Sovetskaya opera: sbornik kriticheskikh 
statey (Moscow, 1953). 

GRINBERG, M. A. (ed.): Sovetskaya simfonicheskaya muzika: sbornik statey (Mos- 
cow, 1955). 

KABALEVSKY, D. B., et al (ed.): Istoriya russkoy sovetskoy muziki (four vols., 
Moscow, 1956—63). 

Kress, S. D.: Soviet Composers and the Development of Soviet Music (London, 
1970). 

Kucera, VACLAV: Nové proudy ve sovetské hudbé (Prague, 1967). 

OLKHOVSKY, ANDREY: Music under the Soviets (London, 1955). 

Овгоу, G.: Russkiy sovetskiy simfonizm (Moscow and Leningrad, 1966). 

Sovetskiy fortepianniy kontsert (Leningrad, 1954). 

PRIEBERG, Е. K.: Musik in der Sowjetunion (Cologne, 1965). 

SCHWARZ, Bonis: Music and Musical Life in Soviet Russia 1917-1970 (London, 
1962). 


(ii) Individual Composers 


Myaskovsky 

IKONNIKOV, А.: Myaskovsky, his Life and Work (New York, 1946). 

SHLIFSTEYN, S. (ed.): N. Y. Myaskovsky: Stat'i, рв та, vospominaniya (two vols., 
Moscow, 1959 and 1960). 


Prokofyev 

ARANOVSKY, M.: Melodika S. Prokof'eva (Leningrad, 1969). 

Вгок, V.: Kontserti dlya violoncheli s orkestrom S. Prokof'eva (Moscow, 1959). 

—‘Osnovnie osobennosti neimitatsionnoy polifonii Prokof’eva’ іп С. A. Orlov 
et al (ed.), Problemi muzikal'noy nauki, i (Moscow, 1972). 

BoGANOVA, T.: Natsional'no-russkie traditsii v muzike S. S. Prokof'eva (Moscow, 
1961). 

DELSON, V.: Fortep'yannie kontserti S. Prokof'eva (Moscow, 1961). 

GAKKEL, L.: Fortep' yannoe tvorchestvo S. S. Prokof'eva (Moscow, 1960). 

Кногороу, Y.: Sovremennie cherti garmonii Prokof'eva (Moscow, 1967). 

MNATSAKANOVA, E. A.: Opera S. S. Prokof'eva ‘Voyna i mir’ (Moscow, 1959). 

Opera S. S. Prokof'eva ‘Obruchenie v топаз е” (Moscow, 1962). 

МЕЅТҮЕУ, I. V.: Prokof'ev (Moscow, 1957; English translation, London, 1961; 
revised and enlarged edition, Zhizm Sergeya Prokof’eva, Moscow, 1973). 
—— and EDELMAN, С. Y. (ed.): Sergey Prokof'ev, 1953-1963. Stat'i i materiali 

(Moscow, 1962; rev. and enlarged ed., 1965). 
OLIVKOVA, V. B.: ‘Romeo i Dzhuletta’ S. Prokof'eva (Moscow and Leningrad, 1952). 
ORDZHONIKIDZE, G.: Fortep'yannie sonati Prokof'eva (Moscow, 1962). 
PoLvAKovA, L. V.: ‘Voyna i mir’ S. S. Prokof’eva (Moscow, 1960). 
Rocozuina, N.: Vokal’no-simfonicheskie proizvedeniya S. Prokof'eva (Moscow and 
Leningrad, 1964). 
SABININA, M.: ‘Semen Kotko’ i problemi opernoy dramaturgii Prokof’eva (Moscow, 
1963). 


BIBLIOGRAPHY 733 


SHLIFSTEYN, S. (ed.): 5. S. Prokof'ev: materiali, dokumenti, vospominaniya (Moscow, 
1956; English translation, Moscow, n.d.). 

SLONIMSKY, S.: Simfonii Prokof'eva (Moscow and Leningrad, 1964). 

бококев, Y.: Skripichnoe tvorchestvo S. Prokof'eva (Moscow, 1965). 

ТАВАКАМОУ, M.: Stil’ simfoniy Prokof'eva (Moscow, 1967). 

Уогкоу, A.: ‘Ob odnom printsipe formoobrazovaniya и Prokof’eva’, in Orlov 
(ed.), Problemi muzikal'noy nauki, i (Moscow, 1972). 

——Skripichnie kontserti Prokof'eva (Moscow, 1961). 

Cherti stilya S. Prokof'eva: Sbornik teoretciheskikh statey (Moscow, 1962). 


Shaporin 
Levit, S.: Yuriy Aleksandrovich Shaporin: ocherk zhizni i tvorchestva (Moscow, 
1964). 


Shebalin 
BELZA, І. Е. and Protopopov, У. V. (ed.): Vissarion Yakovlevich Shebalin: stat'i, 
vospominaniya, materiali (Moscow, 1970). 


Shostakovich 

BERGER, L. (ed.): Cherti stilya Shostakovicha (Moscow, 1962). 

Вовкоузку, V.: Kamernie instrumental'nie ansambli Shostakovicha (Moscow, 1961). 

Bocpanova, А. : ‘Sochineniya D. Shostakovicha konservatorskikh let (1919-1925), 
in А. Kandinsky (ed.), 12 Istorii russkoy i sovetskoy muziki (Moscow, 1971). 

DANILEVICH, L. L.: D. D. Shostakovich (Moscow, 1958). 

—— Маз зоуғетеппік: Tvorchestvo Shostakovicha (Moscow, 1965). 

MARTINOV, I.: D. D. Shostakovich (Moscow and Leningrad, 1946). 

Овгоу, G.: Simfonii Shostakovicha (Leningrad, 1961). 

RaBINovicH, D.: Dmitry Shostakovich (English translation only, Moscow and 
London, 1959). 

SABININA, М. D.: Simfonizm Shostakovicha: put К zrelosti (Moscow, 1965). 


48 


a) en 
MIC. 
7 Go қты A 
г INI ако wi LIN : "t 
| (0008 «уну умен нік йй 
«nto ПИ же bg 7 inv Ач м недай d мее i 5 
A сома) i-ar wae Mun "A 
Чїй дот Му ns Vides 


et MTS Nota 


ЖОСА)». 
«лаш 
Аш 
i worse 
y? пай. 
CORITAABNE 
nO | VO TATE 
cata Я 
i 
А 
& 


LIST OF CONTENTS OF 
THE HISTORY OF MUSIC IN SOUND 
VOLUME X 


The History of Music in Sound is a series of volumes of gramophone records, 
with explanatory booklets, designed as a companion series to the New Oxford 
History of Music. Each volume covers the same ground as the corresponding 
volume in the New Oxford History of Music and is designed as far as possible 
to illustrate the music discussed therein. The records are issued in England by 
E.M.I. Records Ltd. (H.M.V.) and in the United States by R.C.A. Victor, 
and the booklets are published by the Oxford University Press. The editor of 
Volume X of The History of Music in Sound is Gerald Abraham. 

The History of Music in Sound is available on LP records, and the side 
numbers are given below. 


IMPRESSIONISM 
Side1 Band 1 Cloches à travers les feuilles (piano) (Debussy) 
Band 2 Ballade que feit Villon à la requeste de sa mére pour prier 


Nostre Dame (voice and piano) (Debussy) 
Band 3 Мотайеза (for piano) (Falla) 


LATE ROMANTICISM 


Band 4 Prelude, Op. 48, No. 2 (piano) (Skryabin) 
Band 5 Etrangeté, Op. 63, No. 2 (piano) (Skryabin) 
Band 6 Blindenklage, Op. 56, No. 2 (voice and piano) (Richard 
Strauss) 
Band 7 Larghetto from String Trio in A minor, Op. 77b (Reger) 
Sideu Band 1 Mafig from String Quartet in Е sharp minor, Op. 10 
(Schónberg) 
Bands 2-3 Ап Leukon (Berg) 
Schlafend trägt man, Op. 2, No. 2 (Berg) 
Nun ich der Riesen Stürksten, Op. 2, No. 3 (voice and 
piano) (Berg) 


ANTI-ROMANTIC REACTION 


Bands 4-6 Trois petites piéces montées (Satie) 
(1) De l'enfance de Pantagruel (Réverie) 
(ii) Marche de Cocagne (Démarche) 
(iii) Jeux de Gargantua (Coin de Polka) 
Band 1 Excerpts from Conversations: No. 1, 
The Committee Meeting, and No. 5, 
In the Tube at Oxford Circus (Bliss) 
Band 8 Trés lent from String Quartet No. 6, in G (Milhaud) 
Side ш Bands 1-2 Ко. 1 of Trois pièces pour quatuor à cordes (Stravinsky) 
Marche royale from L'Histoire du soldat (Stravinsky) 
Band 3 Improvisations on Hungarian Peasant Songs (piano), 
Op. 20, Nos. 3, 4, 5, and 6 (Bartók) 


736 THE HISTORY OF MUSIC IN SOUND, VOL. X 


Band 4 
Band 5 
Band 6 


12-NOTE MUSIC 


Band 7 
Band 8 


Zápisník zmizelého (Diary of one who vanished), Nos. 
15, 16, 17, and 18 (tenor and piano) (Janáček) 

Stillung Мапа ти dem Auferstandenen (from Das 
Marienleben for voice and piano, 1923) (Hindemith) 

Fuga Octava in D (from Ludus Tonalis for piano solo) 
(Hindemith) 


March from Serenade, Op. 24 (Schónberg) 
Goethe-Lieder, Nos. 2, 3, 5, and 6 (for mezzo-soprano 
and three clarinets) (Dallapiccola) 


MODERN ECLECTICISM 


Side iv Band 1 
Band 2 
Band 3 
Band 4 


Band 5 


Allegro moderato from Trio for violin, viola, and cello, 
Op. 58 (Roussel) 

Valse from String Quartet, No. 2, Op. 69 (Shostakovich) 

Lento from Sonata for violin and piano (Copland) 

Sanctus, Benedictus, and Agnus Dei, from Missa 
Cantuariensis, Op. 59 (Rubbra) 

Poco lento from Quartet for clarinet, viola, and cello 
(Rawsthorne) 


INDEX 


Compiled by G. W. Hopkins 


Abraham, Gerald 
Eight Soviet Composers, 665 nn? ?, 667 
n$, 671 nn? $5, 672 nn* 5, 
Grieg: a Symposium, 92 п?. 
*Ach, du lieber Augustin', 16. 
Acts of St. John, 516. 
Adam, Adolphe, 196. 
Adorno, Theodor Wiesengrund, 206. 
Philosophie der neuen Musik, 206 ппЪ?, 
381 n!, 394 n?. 
Afanasyev, Aleksandr Nikolaevich, 213. 
Agrupación Nueva Мизка, 638. 
Akeret, Kurt, Studien zum Klavierwerk von 
Maurice Ravel, 101 n?. 
Akhmatova, Anna, 660, 685. 
Albéniz, Isaac, 30, 315. 
Iberia, 30. 
d'Albert, Eugen, 3, 183-4. 
Tiefland, 184. 
Aleksandrov, Aleksandr, 640. 
*Hymn of the Soviet Union', 640. 
Aleksandrov, Anatoly, 657, 674, 676. 
Bela, 676. 
Sonatas for Piano 9-11, 696. 
Quartet for Strings Op. 7, 657. 
Tri kubka (Three goblets), 684. 
Alfano, Franco, 258. 
Allende, Pedro Humberto, 637. 
Amy, Gilbert, Mouvements, 401. 
Andersen, Hans, The Ice Maiden, 222. 
Andreyev, Leonid, 136. 
The Black Maskers, 611. 
d'Annunzio, Gabriele, 100, 187—8, 200, 259, 
268. 
Ansermet, Ernest, 213. 
Antheil, George, 584, 596, 605, 606, 627. 
Ballet mécanique, 596. 
Apollinaire, Guillaume, 251, 696. 
Apostel, Hans Erich, 418. 
Arbeau, Thoinot, Orchésographie, 520. 
Arcueil, École d', 239. 
Ardévol, José, 637. 
Arndt, Ernst Moritz, 35 n?. 
Arnold, Malcolm, 563. 
Asafyev, Boris (‘Igor Glebov’), 641, 644 n}, 
652. 
Izbrannie trudi,644 n* , 647 п?. 
works, 647. 


Ascot, Rosa Maria, 319. 

Aseyev, Nikolay, 656. 

Assotsiatsiya proletarskikh muzikantov, see 
Rossiyskaya Assotsiatsiya proletarskikh 
muzikantov. 

Assotsiatsiya sovremennoy muziki, 640, 641, 
642, 650 п!, 651, 655, 657, 660. 

Auden, W. H., 391, 478, 550. 

Auric, Georges, 239, 240, 245, 250-1, 254. 

works, 254. 


Babbitt, Milton, 614, 625, 627. 

Bacarisse, Salvador, 319. 

Bach, C. P. E., Sonaten für Kenner und Lieb- 
haber, 330. 

Bach, J. S., 9, 14, 31, 80—5, 192, 213, 217, 
219, 221, 225, 228, 249, 285, 286, 298, 
321, 327, 329, 350, 370—2, 381, 405, 427, 
537, 542, 515, 577, 636, 651, 693. 

Brandenburg Concertos, 228. 

Goldberg Variations, 537. 

Die Kunst der Fuge, 327, 350, 444. 

O Ewigkeit, du Donnerwort, 371 (Ex. 
172 (1)). 

Passions, 542. 

Toccata in C major (BWV 564), arr. 
Busoni, 82 (Ex. 28). 

Das wohltemperierte Clavier, 405. 

Bacharach, Alfred (ed.), British Music of 
Our Time, 526 тї. 

Báck, Sven Erik, 495. 

Badings, Henk, 321, 492. 

Bagier, Guido, 14, 15. 

Мах Reger, 14 n}, 15 п!. 

Baird, Tadeusz, 496. 

Bakala, Břetislav, 304. 

Balakirev, Mily, 33, 92, 94, 102, 699. 

Tamara, 94. 

Balanchivadze, Andrey, 700. 

Balanchivadze, Meliton, 700. 

Balázs, Béla, 202. 

Ballets russes, 172, 174, 177, 197, 199, 200, 
207, 211. 

Balmont, Konstantin, 230, 660. 

Bantock, Granville, 506. 

Banville, Théodore de, Че Thé’, 113. 

Barber, Samuel, 620-1, 622. 

Antony and Cleopatra, 621. 


738 


Barber—(cont.) 
other works, 620. 

Bartok, Béla, 10, 18-19, 22, 63-4, 68, 114- 
115, 117, 127, 200, 202, 209, 228-9, 230, 
274-99, 300, 301, 315, 388-9, 394, 416, 
417, 479, 487, 540, 548, 556, 559, 563, 
567, 575, 583, 601, 606, 619, 628, 641, 
698. 

Literary works 

The Influence of Peasant Music on 
Modern Music, 285 n!. 

(with A. Lord), Serbo-Croatian Folk 
Songs, 283 n! (Ex. 129). 

Das ungarische Volkslied, 277 ппу °, 
278 пп1>2 279 nnb? (Exx. 116 (i, ii), 
117-20). 

Válogatott zenei irdsai, 115 п*. 

Die Volksmusik der Araber von Biskra 
und Umgebung, 281 n? (Ex. 126). 

Die Volksmusik der Rumänen von 
Maramures, 280 n? (Ex. 122), 281 n! 
(Ex. 124). 

Chamber music 

Contrasts, 283. 

Quartet for Strings No. 1, 133, 278, 
285. 

Quartet for Strings No. 2, 133, 285. 

Quartet for Strings No. 3, 287, 288-9 
(Ex. 131), 290, 298, 563. 

Quartet for Strings No. 4, 278, 289-92 
(Exx. 132, 133), 293, 298. 

Quartet for Strings Мо. 5, 278, 283, 
284 (Ex. 130 (ii)), 292-3, 294. 

Quartet for Strings No. 6, 280 (Ex. 
123 (ii)), 293—5 (Ex. 134). 

Sonata for Solo Violin, 297, 298 (Ex. 
IBS) 299, 

Sonata for Two Pianos and Percussion, 
282, 293, 295, 296, 299. 

Sonata for Violin No. 1, 285-6, 287. 

Sonata for Violin No. 2, 280 (Ex. 
123 (1)), 285-6, 287. 

Keyboard works 

Allegro barbaro, 19, 286. 

Bagatelles Op. 6, 19, 127, 128, 133, 283. 

Dirges, 10. 

Elegies, 10. 

Improvisations, 279. 

Microcosmos, 284 (Ex. 130), 296-7. 
Six Pieces in Bulgarian Rhythm, 283, 

297. 

Out of Doors, 286. 

Rhapsody Op. 1, 19, 276 (Ex. 115). 

Suite No. 1 Op. 3, 19. 

Sonata for Piano, 286. 

Orchestral works 

Concerto for Orchestra, 279 (Ex. 121), 
283, 297-8, 299. 

Concerto for Piano No. 1, 286-7, 295. 

Concerto for Piano No. 2, 295. 


INDEX 


Concerto for Piano No. 3, 278, 297, 
298-9, 388. 

Concerto for Two Pianos and Orches- 
tra, 296. 

Concerto for Viola, 297. 

Concerto for Violin No. 1, 19. 

Concerto for Violin No. 2, 293, 294, 
295, 297. 

Dance Suite, 282 (Ex. 127), 286. 

Két arckép (Two portraits) Op. 5, 10, 19. 

Két kép (Two pictures) Op. 10, 19, 115, 
117 (Ex. 52 (iv)). 

Kossuth Symphony, 10, Ex. 21 (iii), 63. 

Music for strings, percussion and celesta, 
278, 293, 295—6, 299. 

Stage works 

A kékszakdlli herceg уаға (Duke 
Bluebeard's Castle), 10, 19, 202 (Ex. 
95), 300. 

The Miraculous Mandarin, 285. 

Baudelaire, Charles, 202. 

Baudrier, Yves, 255, 429. 

Bauer, Marion, 588. 

Bax, Arnold, 504—6, 507. 

Symphony No. 2, 504 (Ex. 221). 
Symphony No. 3, 504, 505 (Ex. 222). 
other works, 504. 

*Baykal (Russian song), 695. 

Becerra, Gustavo, 638. 

Beck, Conrad, 320. 

Becker, John J., 606. 

Bedier, Joseph, Roman de Tristan et Yseult, 
320. 

Beethoven, Ludwig van, 31, 193, 219, 231, 
287—8, 289, 324, 325, 327, 354, 381, 507, 
517555278512 5] SH 

Concerto for Violin, 354. 
Diabelli Variations, 327. 
Fidelio, 193, 324. 
Symphony No. 3, 325. 
Symphony No. 5, 193, 533. 
Symphony No. 9, 6, 78. 

Bekker, Paul, 32 n*. 

Bellamann, Henry, 577, 580. 

Bellini, Vincenzo, 391. 

Belsky, V. I., 175. 

Bely, Viktor, 642. 

*Ballada o kapitane Gastello', 684. 

Belyaev, Viktor, 699 n?. 

Ocherki po istorii muziki narodov SSSR, 
699 n?. 
see also under Viktor Uspensky. 

Belza, Igor (ed.), Rusko-polskie muzikalnie 
svyazi, 36 n°, 

Benda, Jiri, 177. 

Benét, Stephen Vincent, 607. 

Benjamin, Arthur, 547. 

works, 547. 
Benn, Gottfried, 334. 
Berg, Alban, 10, 18, 65, 68, 72, 73-4, 185, 


INDEX 


206, 210, 273, 291, 340-1, 345, 362-72, 
375—6, 377,379, 385, 395, 416, 417, 421, 
423, 448, 494, 497, 500, 540, 559, 560, 
563, 628. 
Altenberg-Lieder, 72. 
Chamber Concerto, 367-8. 
Concerto for Violin, 18, 369-72 (Exx. 
170, 171, 172 (ii)), 417. 
Four Pieces for clarinet and piano Op. 5, 
18. 
Four songs Op. 2, 18. 
Lulu, 187, 273, 363, 364—7 (Exx. 164, 165, 
166, 167), 370, 371, 376, 379, 419. 
Lulu Symphony, 367. 
Lyric Suite, 291, 368-9 (Exx. 168, 169), 
373538762371: 
Piano Sonata Op. 1, 18, 73-4 (Exx. 25, 
26), 104 n?, 133. 
Quartet for Strings Op. 3, 18. 
*Schliesse mir die Augen beide', 368. 
Three Pieces for large orchestra Ор. 6, 18. 
*Traumgekrónt', 18. 
Der Wein, 379. 
Wozzeck, 2773, 363-4, 365, 368, 370, 371, 
448, 497, 641. 
Berger, Arthur, 625. 
Bergsma, William, 626. 
Berio, Luciano, 410, 435, 469, 477-81, 486, 
492. 
Allelujah II, 479. 
Circles, 419-81 (Ex. 214). 
Momenti, 492. 
Nones, 478-9 (Ex. 212). 
Omaggio a Joyce, 478, 486, 488—9, 492. 
Quaderni I, 479. 
Quartet for Strings, 479 (Ex. 213). 
Berkeley, Lennox, 523-4, 536-7, 539, 568. 
works, 536-7. 
Berlioz, Hector, 9, 19, 32, 72, 191, 195, 
650. 
Roméo et Juliette, 191. 
Les Troyens, 191. 
Berners, Lord, 522. 
works, 522. 
Bernhardt, Sarah, 148. 
Bernstein, Leonard, 573, 626-7. 
West Side Story, 627. 
Berutti, Arturo, 637. 
works, 637. 
Bezimensky, Aleksandr Пуісһ, 653. 
Billings, William, 622. 
Biriotti, León, 638. 
Birse, Edward, 38 n?. 
Birtwistle, Harrison, 568. 
Bisquertt, Próspero, 637. 
Bittner, Julius 
Das hóllisch Gold, 148. 
Der Musikant, 148. 
Bizet, Georges, 154. 
Carmen, 154. 


739 


Blacher, Boris, 410—13, 415, 417, 494, 496, 
497, 498, 523. 
Abstrakte Oper, 410-11. 
Orchesterfantasie, 413. 
Orchester-Ornament, 411 (Ex. 190). 
Paganini Variations, 411-13. 
Requiem, 413. 
Rosamunde Floris, 413. 
Blake, William, 551. 
Blei, Franz, 331. 
Bliss, Arthur, 521-2. 
works, 521-2. 

Blitzstein, Marc, 573, 618. 
The Cradle Will Rock, 618. 

Bloch, Ernest, 573, 574, 595, 606, 611. 
Macbeth, 169 (Ex. 79). 
other works, 606. 

Blok, Aleksandr, 152, 660, 671. 
‘The Twelve’, 650, 697. 

Blomdahl, Karl-Birger, 494, 495. 
Aniara, 494 (Ex. 217). 

“Вопда, Мах” (Gertrud Schoenberg), 
B52 

Blunt, Bruce, 520. 

Bócklin, Arnold, 34. 

Boethius, 273. 

Boetticher, Wolfgang, Robert Schumann: 
Einführung in Persónlichkeit und Werk, 
81 nt, 

Boito, Arrigo, 159. 

*Boldly, comrades, keep in step', 691. 

Bolotnikov, Ivan, 661. 

Borodin, Aleksandr, 37, 87, 90, 671, 680, 
689. 

Knyaz Igor (Prince Igor), 174, 689. 
*Spyashchaya knyazhna', 87-8 (Exx. 32, 
33). 

Bossi, Marco Enrico, 256. 

Boston Symphony Orchestra, 223, 297, 
615. 

Boucourechliev, 
453 n*. 

Boughton, Rutland, 506. 

The Immortal Hour, 187. 

Boulanger, Nadia, 536, 547, 568, 588, 602, 
606, 615, 630, 699, 

Boulez, Pierre, 373, 428, 430 n!, 433, 435, 
437, 441—54, 455, 458—9, 464, 466, 467, 
468, 470, 482, 488 n°, 625. 

Penser la musique aujourd'hui, 442 n!, 
445 n?, 453 п1. 

Doubles, 464. 

Etudes, 488 

Improvisations sur Mallarmé, see under 
Pli selon pli. 

Livre pour quatuor, 444. 

Le Marteau sans тайге, 444, 448-53 
(Ex. 203), 470, 484. 

Pli selon pli, 466-7. 
Improvisation sur Mallarmé I, 433 п!. 


André, “РІ selon pili’, 


740 


Boulez—(cont.) 
Improvisation sur Mallarmé И, 453, 
467 (Ex. 208). 

Polyphonie X, 445-6, 448. 

Psalmodies, 442. 

Le soleil des eaux, 444, 445. 

Sonata for Piano No. 2, 442-4 (Ex. 201). 

Sonata for Piano No. 3, 453, 465-6. 

Sonatina for Flute and Piano, 442. 

Structures I, 441, 445, 446—8 (Ex. 202). 

Le Visage Nuptial, 444, 445. 

Brahms, Johannes, 2, 3, 9, 10, 14, 18, 26, 31, 
33, 38, 39, 67, 68, 81, 87, 276, 279, 328, 
343, 354, 381, 574, 595. 

chorale preludes, 2. 

Clarinet Quintet, 2, 78. 

Concerto for Violin, 354. 

Hungarian dances, 276. 

piano pieces Opp. 116-19, 2, 67. 

Quartet for Strings in A minor, 343 (Ex. 

152). 
Romanzen aus Tiecks ‘Magelone’ 
*Ruhe, Süssliebchen', 87. 
Brand, Max, Maschinist Hopkins, 339. 
Brant, Henry, 628. 

Grand Universal Circus, 628. 

Brecht, Bert, 339, 611. 
Brelet, Giséle, 426. 

Chances de la musique atonale, 426 n°. 
Bretón, Tomás, Verbena de la paloma, 30. 
Bridge, Frank, 506, 547. 

Bridges, Robert, 518. 
Brinkmann, Reinhold 
Arnold Schénberg: Drei Klavierstücke 
ОР 
Вгіќеп, Вепјатіп, 494, 502, 506, 507, 520, 
538, 540, 542, 543-56, 559. 
Chamber music 
Quartet for Strings No. 2, 547. 
Sonata for Cello, 555. 
Choral works 
Cantata Academica, 556. 
War Requiem, 542. 
Orchestral works 
Sinfonia da requiem, 544 (Ex. 239), 
547-8. 
Variations on a Theme of Frank 
Bridge, 547. 
The Young Person's Guide to the 
Orchestra, 547. 
Music with solo voice 
Canticles, 551. 
The Holy Sonnets of John Donne, 549- 
550 (Ex. 241). 
Les Illuminations, 550. 
Nocturne, 551. 
Our Hunting Fathers, 550. 
Serenade, 551, 555-6 (Ex. 243 (1)). 
Seven Sonnets of Michelangelo, 550. 
Winter Words, 551-2 (Ex. 242). 


INDEX 


Stage works 
» * Albert Herring, 540, 554. 
The Beggar's Opera, 554. 
Billy Budd, 554. 
Gloriana, 554—5. 
A Midsummer Night's Dream, 555-6 
(Ex. 243 (ii)). 
Paul Bunyan, 554 
Peter Grimes, 540, 544, 548 (Ex. 240), 
554, 555. 
The Rape of Lucretia, 540, 548 (Ex. 
239 (ш)), 554, 555. 
The Turn of the Screw, 554, 556. 
Bruch, Max, 3, 607. 
Bruckner, Anton, 3, 7, 8, 37, 223, 416, 417. 
Bruneau, Alfred, 22, 154, 155 n?, 164—6, 169, 
179. 
L'Attaque du moulin, 164. 
Messidor, 164—6 (Ex. 76). 
L'Ouragan, 164—5. 
Le Réve, 93. 
Bryusov, Valery, 230. 
Büchner, Georg, 363, 497. 
Woyzeck, 363. 
Bülow, Hans von, 7. 
Burkhardt, Willy, 320. 
Das Gesicht Jesajas, 320. 
Burns, Robert, 684. 
Bush, Alan, 523. 
works, 523. 
Busoni, Ferruccio, 31-2, 39, 78, 81-3, 192-5, 
273, 310, 320, 327, 339, 420, 574. 
Entwurf einer Neuen Aesthetik der Ton- 
kunst, see under Von der Einheit der 
Musik. 
Von der Einheit der Musik (The Essence 
of Music), 32 n?. 
Entwurf einer Neuen Aesthetik der Ton- 
kunst, 193 nt, 326. 
Arlecchino, 32, 192-4 (Ex. 91). 
Die Brautwahl, 32, 192. 
Doktor Faustus, 32, 194—5, 
Chamber Fantasy on Bizet's Carmen, 327. 
Concerto for Piano, 31, 32, 78. 
Concerto for Violin, 31, 32. 
Fantasia contrappuntistica, 83, 192, 327. 
Lustspielouvertüre, 31. 
Sonatinas, 327. 
Symphonisches Tongedicht, 32. 
Toccata in C major (Bach BWV 564), 
piano arrangement, 82 (Ex. 28). 
Turandot, 192, 194, 
Turandot Suite, 31, 192. 
Bussotti, Sylvano, 465. 
Piano Pieces for David Tudor, 465. 
Byrd, William, 519, 531, 551. 
Byron, Lord, 357. 


Саатайо, Roberto, 638. 
Cable, George W., 572. 


INDEX 741 


Caccini, Giulio, 267. 

Cage, John, 465, 486, 602, 625, 627-8. 
Imaginary Landscape, 627. 

*Ca ira', 650. 

Calder, Alexander Stirling, 464. 
Calvocoressi, M. D., 30, 101, 277 n!. 
Musicians Gallery, 30 n+, 101 n}. 

Campo, Conrado del, 30. 
works, 30. 
Canaletto, Antonio, 152. 
Capek, Karel, The Makropulos Affair, 303. 
Cardew, Cornelius, 465 n!. 
Carissimi, Giacomo, 542. 
‘La Carmagnole’, 650. 
Carner, Mosco, Puccini, 156 n!. 
Carpenter, John Alden, 584—5. 
works, 584—5. 
Carter, Elliott, 606, 625, 628, 630-4. 


Quartet for Strings No. 1, 632 (Ex. 263). 


Sonata for Cello and Piano, 630-2. 
other works, 632. 

Casanova, André, 435 n°, 

Casella, Alfredo, 256, 267-8, 274, 420. 

La Favola d'Orfeo, 267-8. 

other works, 267. 
Castelnuovo-Tedesco, Mario, 268, 269. 

works, 269. 

Casti, Abbate, 324. 
Catalani, Alfredo, 163. 

La Wally, 153, 163 (Ex. 75). 
Caturla, Alejandro García, 637. 
Cech, Svatopluk, 302. 

Cervantes, Miguel de, 316, 500. 

Don Quixote, 315. 

Cervetti, Sergio, 638. 
Chabrier, Emmanuel, 89, 101, 251. 

Bourrée fantasque, 93. 

Gwendoline, 20. 

Habanera, 101. 

Le roi malgré lui, 93. 

Chadwick, George Whitefield, 614. 

Chambonniéres, 
251. 

Chapi, Ruperto, Revoltosa, 30. 

Char, René, 444, 448. 

Charles IV (Emperor), 305. 

Charpentier, Gustave, 22, 169. 

Louise, 154, 166, 183. 

Chase, Gilbert, 571. 

America’s Music, 571 п!. 

The Music of Spain, 319 її, 
Chausson, Ernest, 2 n?, 22, 25, 26. 

Ровте, 22. 

Quelques danses, 93. 

Le roi Arthus, 2, 20, 22, 169. 

Symphony in B flat, 22. 
Chávez, Carlos, 635, 636-7. 

Los Cuatro Soles, 636. 

other works, 637. 
Chaykovsky, Boris, 697. 


Jacques Champion de, 


Chaykovsky, P. I., see Tchaikovsky. 
Chekhov, Anton, 33. 
Chernishevsky, Nikolay Gavrilovich, 640. 
Chishko, Oles, 666—7. 
Bronenosets Potemkin (The Battleship 
Potemkin), 667. 
Chlubna, Osvald, 304. 
Chopin, Frédéric, 19, 25, 34, 136, 311. 
Clair, René, A nous la liberté, 254. 
Claudel, Paul, 243, 250. 
Clough, Wilson O., 86 n!. 
Cochran, C. B., 522. 
Cocteau, Jean, 167, 220, 236, 239-40, 241, 
243, 250, 251, 254, 428, 543. 
Le coq et l'arlequin, 199 n?, 239. 
Oedipus Rex, 220. 
Colette, 174, 235. 
Collet, Henri, 239. 
Constant, Marius, 495-6. 
Cooper, Martin 
French Music. From the death of Berlioz 
to the death of Fauré, 101 пз, 234 
па 
Copland, Aaron, 536, 573, 583, 584, 588-94, 
601, 605, 606, 614, 615-18, 620, 622, 
625, 626, 627, 628, 630. 
Literary work 
Our New Music, 583 n!, 589 n!, 618 m 
Chamber music 
Nonet, 592. 
Quartet for Piano and Strings, 592. 
Sonata for Violin and Piano, 592. 
Keyboard works 
Fantasy for Piano, 592 (Ex. 253). 
Sonata for Piano, 592. 
Variations for Piano, 589, 592. 
Orchestral works 
Concerto for Organ and Orchestra, 
588. 
Concerto for Piano, 588-9. 
El Salón México, 589. 
Music for the Theater, 588. 
Short Symphony, 589. 
Statements for Orchestra, 589. 
Symphony No. 1, 588. 
Symphony No. 3, 593. 
Solo song 
Twelve Poems of Emily Dickinson, 592. 
Stage and film music 
Appalachian Spring, 589 (Ex. 252). 
Billy the Kid, 589. 
Of Mice and Men, 592. 
Our Town, 592. 
Rodeo, 589. 
The Heiress, 592. 
The Second Hurricane, 589. 
The Tender Land, 592. 
Le Corbusier, 493. 
Corelli, Arcangelo, 256. 
Cosme, Luiz, 636. 


742 INDEX 


Couperin, Louis, 235, 251, 286. 

Cousin, Victor, 238. 

Cowell, Henry, 575, 580, 584, 595, 596, 
597-602, 605, 610, 615, 618, 622, 627, 
628. 

Sinfonietta, 596, 601. 
Tiger, 600 (Ex. 255). 

Cowell, Sidney, 575. 

Crabbe, George, 555. 

Craft, Robert 

Conversations with Stravinsky, 117 n?, 
401 n?. 

Memories and Commentaries, 389 її, 
391 n!, 395 п?, 400 n!, 401 n?. 

Creighton, Basil, 8 n!. 

Creston, Paul, 573, 620. 

Cui, César, 33, 639. 

Cummings, E. E., 480. 

Cynewulf, Christ, 567. 


Dahl, Ingolf, 626. 
Dallapiccola, Luigi, 270-3, 346, 395, 413, 
420-4, 426, 428, 431 nt, 435, 469. 
Canti di liberazione, 423 (Ex. 193), 431 n!. 
Canti di prigione, 273, 421. 
Cinque canti, 423. 
Concerto per la notte di natale, 423. 
Cori di Michelangelo Buonarotti il 
Giovane, 270-3 (Ex. 113). 
Divertimento on Themes from Tartini, 
423. 
Goethe-Lieder, 423. 
Job, 421. 
Liriche greche, 421. 
Il Prigioniero, 273 (Ex. 114), 421. 
Quaderno musicale di Annalibera, 423. 
Variazioni, 423 n!. 
Volo di notte, 273, 420. 
Damrosch, Walter, 580. 
Daniélou, Jean, 220. 
Danilevich, L. 
Nash sovremennik: Tvorchestvo Shostako- 
vicha, 673 nt. 
Dante, 159, 193. 
Inferno, 258. 
Dargomïzhsky, Aleksandr Sergeyevich, 174, 
1, 218: 
Rusalka, 177. 
Stone Guest, 88, 193, 201. 
Daudet, Alphonse, 164. 
David, Johann Nepomuk, 413. 
Davidenko, Aleksandr, 642. 
Davies, Peter Maxwell, 568. 
Dawes, Frank, Debussy: Piano Music, 
101 n?. 
Deane, Basil, Albert Roussel, 25 n!. 
Debussy, Claude, 2, 19-29 passim, 67, 72, 
73, 78, 89-105, 105-27 passim, 140, 144, 
148, 159, 169, 171, 172, 185, 196, 199, 
200, 202, 208, 228, 229, 233, 234, 240, 


259, 262, 266, 267, 268, 274—5, 286, 299, 
7 312, 315, 321, 326, 373, 421, 430, 441, 
445, 466—7, 488, 503, 506, 507, 516, 531, 
544, 574, 595, 639, 665. 
Literary work 
Monsieur Croche, anti-dilettante, 171 пл, 
Chamber music 
Quartet for Strings, 29, 68, 92, 94. 
Sonatas, 29, 100, 104, 144. 
violin and piano, 100 (Ex. 41). 
cello and piano, 144. 
Keyboard work 
Children's Corner, *Golliwogg's cake 
walk’, 99. 
En blanc et noir, 29. 
Estampes, 92. 
‘Jardins sous la pluie’, 92, 94. 
*Soirée dans Grenade', 101. 
Études, 29, 100, 127, 144. 
Images I, 95. 
‘Mouvement’, 99. 
Images II, 95. 
*Cloches à travers les feuilles’, 99, 
L'Isle joyeuse, 128. 
Lindaraja, 101. 
Pour le piano, 92, 93, 94, 101. 
Préludes, 95. 
“Га cathédrale engloutie’, 114. 
*Voiles', 99 (Ex. 40). 
Orchestral music 
Danses for harp and orchestra, 92-3. 
Gymnopédies (Satie, orchestrated 
Debussy), 92. 
Images, 95. 
‘Ibéria’, 95-7 (Exx. 38, 39). 
La Mer, 29, 67, 95. 
Nocturnes, 73, 92, 94, 117. 
‘Fêtes’, 544. 
‘Sirènes’, 78, 94-5 (Ex. 37), 155 п!. 
Prélude à l'aprés-midi d'un faune, 92, 
101, 117, 200, 488. 
Printemps, 91 (Ex. 35), 92. 
Solo songs 
*Les Angélus', 90. 
‘Beau soir’, 89. 
Chansons de Bilitis, 92, 94. 
‘En sourdine’, 27-9 (Ex. 11 (ii)). 
Fétes galantes I, 94. 
‘Fleur des blés’, 91 n°. 
‘Green’, 27 (Ex. 10). 
*Mandoline', 89-90 (Ex. 34). 
Proses lyriques, 92, 94. 
Trois Ballades de Villon, 100. 
Trois Chansons de Charles d'Orléans, 
100. 
Trois Chansons de France, 100. 
Trois mélodies (Verlaine), 94. 
Stage and choral works 
La Demoiselle Élue, 23, 117, 166. 
Jeux, 199. 


INDEX 


Le martyre de Saint-Sébastien, 100, 117, 
200 (Ex. 94). 
Pelléas et Mélisande, 23, 29, 92, 94, 101, 
105, 108, 114, 117, 166-9 (Exx. 77, 
78), 171, 183, 196, 516. 
Trois Poémes de Stéphane Mallarmé, 
100, 466. 
Dehmel, Richard, 15. 
*Waldseligkeit', 12. 
Delibes, Leo, 196, 223. 

Lakmé, 153, 154. 

Delius, Frederick, 39, 43-4, 78, 117, 155 n!, 
184, 503, 520, 568. 

Appalachia, 39, 78. 

Arabesk, 39. 

Brigg Fair, 39. 

Fennimore and Gerda, 39. 

In a Summer Garden, 117 (Ex. 52 (1)). 

Koanga, 184. 

A Mass of Life, 39, 75-8. 

On hearing the first cuckoo in spring, 
39. 

Romeo und Julia auf dem Dorfe (A Village 
Romeo and Juliet), 147 n!, 184 (Ex. 
86). 

Sea Drift, 39, 75. 

Song of the High Hills, 78. 

Songs of Sunset, 44. 

Dello Joio, Norman, 573, 620, 621, 622. 

Variations, Chaconne and Finale, 621. 

Del Mar, Norman 
Richard Strauss, 148 n?. 
Denisov, Edison, 699. 
works, 699. 
Deshevov, Vladimir, 660. 
Relsi (The Rails), 660. 
Diamond, David, 620. 
Dickenmann, Paul, Die Entwicklung der 
Harmonik bei A. Skrjabin, 133 n?. 
Diderot, Denis, 161. 
*Dies irae', 650. 
Dohnányi, Ernó, 9. 
Dolmetsch, Carl, 535. 
Dostoevsky, Fédor, 175, 230, 304. 
Dowland, John, 518, 519. 

Lachrimae, 520. 

Downey, John W., La Musique populaire 
dans l'euvre de Béla Bartók, 282 п\. 

Drew, David, 431. 

Dukas, Paul, 22, 24—5, 43, 105, 108-10, 114, 
115, 169, 202, 255. 

L'Apprenti sorcier, 20, 24, 115. 

Ariane et Barbe-bleu, 22, 23-4 (Ex. 7), 
169, 183. 

La Péri, 25, 105 (Ex. 45), 200. 

Sonata for Piano in E flat minor, 24 
(Ex. 8), 109 (Ex. 47). 

Symphony in C major, 24, 25. 

Variations, interlude et finale sur ип thème 
de Rameau, 24. 


743 


Dunstable, John, 568. 
Duparc, Henri, 22, 26, 169. 
‘Soupir’, 169. 
Dupré, Marcel, 255 
Durey, Louis, 239, 240. 
Dushkin, Samuel, 225. 
Dvorák, Antonín, 3, 6, 9, 10, 31, 37, 68, 301, 
„571-2, 606. 
Cert a Kdca (The Devil and Kate), 177. 
Dimitrij, 177. 
Rusalka, 177. 
Symphony No. 9 (‘From the New 
World’), 78. 
Dyagilev, Sergey, 172, 177, 197, 199, 200, 
207, 230, 522, 524. 
Dzerzhinsky, Ivan, 665—6, 669. 
Daleko ot Moskvi (Far from Moscow), 
691. 
Groza (The Storm), 691. 
Krov naroda (The Blood of the People), 
676. 
Podnyataya tselina (Virgin soil upturned), 
666. 
Tikhy Don (The Quiet Don), 665-6 (Ex. 
277), 669. 


Egk, Werner, 410. 
works, 410. 
Eichendorff, Joseph Freiherr von, 319, 
326 
Eimert, Herbert, 489, 490. 
Etüde über Tongemische, 490. 
Fünf Stücke, 490. 
Glockenspiel, 489 n°. 
*Ein feste Burg', 194. 
Einem, Gottfried von, 411 n!, 497. 
Dantons Tod, 497. 
Ekman, Karl, Jean Sibelius, 38 n?. 
Elgar, Edward, 39, 43-4, 65, 67-8, 74, 75, 
78, 105, 503. 
The Apostles, 75. 
Concerto for Cello, 39. 
Concerto for Violin, 39. 
The Dream of Gerontius, 2, 39. 
Enigma Variations, 39. 
Falstaff, 39, 67-8. 
The Kingdom, 75. 
Symphony No. 1, 39, 75. 
Symphony No. 2, 39, 44, 65, 75 (Ex. 27), 
156. 
ШӨ №. 5., 507. 
Eluard, Paul, Poémes pour la paix, 699. 
Emerson, Ralph Waldo, 575, 577. 
Erbse, Heimo, 496. 
Escobar, Luis Antonio, 638. 
Esenin, Sergey, 651, 660. 
Eshpay, Andrey, 697. 
Esipova, Anna, 229. 
Espla, Oscar, 317, 318, 319. 
works, 317. 


744 


*Es sungen drei Engel', 335. 

Euripides, 264. 

Evans, Edwin, ‘The Firebird’ and ‘Petrushka’, 
[TEE 

Evtushenko, Evgeny Aleksandrovich, 696. 


Falla, Manuel de, 31, 114, 200-1, 274—5, 
315-17, 318, 319, 641. 
El amor brujo, 31, 200-1. 
Atlántida, 317. 
Concerto for Harpsichord, 316-17 (Ex. 
142). 
Fantasia bética, 31, 316. 
Noches en los jardines de Езрайа, 31. 
Quatres piéces espagnoles, 31. 
El retablo de Maese Pedro, 201, 315. 
Siete canciones populares espafiolas, 315. 
El sombrero de tres picos, 31, 201, 315. 
La vida breve, 31, 201. 
Farwell, Arthur, 572, 574, 614. 
Fauré, Gabriel, 25-9, 89, 101, 108, 113, 144, 
231-2, 234, 235, 320, 327, 503. 
La Bonne Chanson, 26, 29. 
La Chanson d'Éve, 29. 
Fantaisie, 232. 
L'Horizon chimérique, 232. 
Le Jardin clos, 144. 
Melodies Op. 58, 26. 
*En sourdine', 27-8 (Ex. 11 (i)). 
‘Green’, 26 (Ex. 9). 
Mirages, 232. 
Pénélope, 172. 
‘Les Présents’, 26. 
Quartet for Strings, 232. 
Quintet for Piano and Strings No. 2, 232. 
Trio, 232. 
Sonata for Cello No. 1, 232. 
Sonata for Cello No. 2, 232. 
Feinberg, Samuil, 657, 674, 676. 
Prelude Op. 8 No. 1, 657-60 (Ex. 272 
(1)). 
Fernández, Oscar Lorenzo, 636. 
Fet, Afanasy, 684. 
Fibich, Zdenék, 177. 
Námluvy Pelopovy, 177 (Ex. 83). 
other works, 177. 
Fine, Irving, 626. 
Finney, Ross Lee, 614, 620, 622. 
Finzi, Gerald, 539. 
works, 539. 
Fitelberg, Grzgegorz, 36. 
Flaubert, Gustave, 148. 
Madame Bovary, 74, 154. 
Fletcher, Giles, 522. 
Fortner, Wolfgang, 413-16, 417, 428, 435, 
497. 
Mouvements, 401, 413-15 (Exx. 191 (i, ii)). 
Symphony, 413. 
other works, 415. 
Foss, Lukas, 626. 


INDEX 


Time Cycle, 626. 


-Foster, Stephen Collins, 571, 575. 


Frangaix, Jean, 256. 
Franck, César, 19, 20, 21, 22, 24, 25, 29, 68, 
75, 91, 170, 320. 
Quartet for Strings, 68. 
Franc-Nohain, 172. 
Frankel, Benjamin, 557. 
Franz, Robert, 10. 
Frescobaldi, Girolamo, 256, 286. 
Freud, Sigmund, 326. 
Fricker, Peter Racine, 563-7. 
Concerto for Viola, 565. 
Concerto for Violin, 565. 
Litany, 566. 
Prelude, Elegy, and Finale, 565 (Ex. 247). 
Octet, 567. 
Quartet for Strings No. 1, 563, 565, 567. 
Quartet for Strings No. 2, 567. 
Sonata for Violin, 565, 567. 
Sonnets for Piano, 566. 
Symphony No. 1, 565, 567. 
Symphony No. 2, 565 (Ex. 246), 567. 
Twelve Studies for Piano, 566—7. 
The Vision of Judgement, 561. 
Fry, William Henry, 570. 
Furmanov, Dmitry, 667. 
Füssl, Karl-Heinz, 410. 


Gabrieli, Giovanni, 218, 398. 

Gallet, Luciano, 636. 

Gastello, Captain, 684. 

Gatscher, Emanuel, Die Fugentechnik Max 

Regers in ihrer Entwicklung, 83 її. 

Gatti, Guido M., 264 n!, 424. 

Gay, John, 339, 554. 

The Beggar's Opera, 339, 554. 

Gedike, Aleksandr, 651. 
works, 651. 

George, Stefan, 16. 

*Entrückung', 16. 
*Litanei', 16. 

Gerhard, Roberto, 318, 425, 426—7, 559. 
Albada, interludi y аапса, 318. 
Alegrías, 318. 

L'alta паіхепса del Rei En Jaume, 318. 
Concerto for Harpsichord, 559. 
Concerto for Piano, 559. 

Concerto for Violin, 426, 559. 

Quartet for Strings No. 1, 427, 559. 
Quintet for Wind, 318. 

The Duenna, 318, 426, 559. 

Symphony No. 1, 427, 559. 
Symphony No. 2, 427. 

Symphony No. 3 (Collages), 427. 

Gershwin, George, 527, 573, 584, 585-7, 

588, 594, 605. 

Rhapsody in Blue, 527, 584, 585 (Ex. 251), 
587, 588. 

other works, 587. 


INDEX 


Gesualdo, Carlo, 520. 

Geyer, Stefi, 19. 

Ghedini, Giorgio, 268, 269, 424. 

works, 269. 
Gibbons, Orlando, 519. 
Gide, André, 220, 222. 
*The gift to be simple', 589. 
Gilbert, Henry F. B., 572. 
works, 572. 

Ginastera, Alberto, 635, 637-8. 
works, 637-8. 

Giordano, Umberto, 31, 161-2. 
Andrea Chénier, 161-2 (Ex. 72). 

Gippius, Zinaida Nikolayevna, 660. 

Giraud, Albert, 16, 144. 

Gladkovsky, Arseny, 643. 

Front i til (Front and Rear), 643. 
(with E. V. Prussak), Za Krasny Petro- 
grad (For Red Petrograd), 643, 647. 
Glazunov, Aleksandr, 33, 639, 641, 657. 
works, 657. 

*Glebov, Igor', see Asafyev. 

Gleich, Clemens-Christoph Johannes von, 
Die sinfonische Werken von Alexander 
Skriabin, 35 n?, 133 п?. 

Glier, Reinhold, 639, 641, 672, 684, 699— 
700. 

Concerto for voice and orchestra, 684. 
Krasny mak (The Red Poppy), 648. 
Quartet for Strings No. 3, 657. 
Shakh-Senem, 672, 700. 
(with Talib Sadikov), Суибага, 700. 
(with Talib Sadikov), Геуй i Medzhnun, 
700. 

Glinka, Mikhail Ivanovich, 32, 88, 218. 
Ruslan and Ludmila, 32, 88. 

Gluck, Christoph Willibald von, 512. 

Gnatalli, Radamés, 636. 

Gnesin, Mikhail, Simfonichesky monument: 
1905—1917, 651. 

Goehr, Alexander, 568. 

Goethe, J. W. von, 149, 193, 319, 380. 

Faust, 8. 
Goetschius, Percy, 607. 
Gogol, Nikolai, 307, 646. 
Government Inspector, 410. 

Goldman, В. Е., 481 n!. 

Goldmark, Rubin, 585, 588. 

Goldoni, Carlo, 189, 191, 192, 264, 266. 

Locandiera, 688. 

Goléa, Antoine, Rencontres avec Pierre 
Boulez, 430 n!, 435 n*, 442 n?, 445 n!, 
446 n!, 453 n!, 465 n?, 466 nn4, 
488 n?. 

Goncourt, Jules and Edmond de, L'Art du 
dix-huitiéme siécle, 193. 

Gorky, Maxim, 642, 667, 691. 

Gottschalk, Louis Moreau, 570, 572. 

Gould, Morton, 620. 

Gounod, Charles, 2, 159, 166, 200, 428. 


745 


Gozzi, Carlo, 192, 230, 267. 
Turandot, 31, 192, 258. 
Gram, Hans, Death Song of an Indian Chief, 
570. 
Granados, Enrique, 30-1, 68, 315. 
Goyescas, 30. 
*Coloquio en la reja’, 68 (Ex. 24 (iii)). 
Gredinger, Paul, 490. 
Formanten, 490. 
‘Greensleeves’, 194. 
Gregor, Joseph, 324. 
Greuze, Jean Baptiste, 161. 
Grieg, Edvard Hagerup, 37, 39, 68, 92, 570, 
574, 647. 
Haugtussa Sang-Cyclus, 37. 
Norske Folkeviser, 37. 
Quartet for Strings in G minor, 92. 
Slátter, 37. 
Griffes, Charles Tomlinson, 574. 
works, 574. 
Grillparzer, Frans, 521. 
Grinberg, M. A. 
(ed.), Sovetskaya simfonicheskaya muzika, 
648 n?, 671 пі. 
Gropius, Manon, 369-70, 371. 
Grove's Dictionary of Music and Musicians, 
300 пт, 424 n*, 556 пі. 
Gruenberg, Louis, 585, 595. 
The Emperor Jones, 595. 
other works, 585. 
Grünewald, Matthias, 334, 407. 
El grupo de los ocho (el grupo de Madrid), 
319. 
Grupo Renovación Musical, 637, 638. 
Guarnieri, Carmago, 636. 
Guerra-Peixe, César, 638. 
Gurney, Ivor, 520-1. 
works, 521. 
Gusev, Sergey Ivanovich, “Тһе Year 1917”, 
672. 
Guthrie, Tyrone, 554. 


Hába, Alois, 309, 310-11. 
Quartet for Strings Op. 12, 311 (Ex. 139). 
other works, 311. 
Haieff, Alexei, 626. 
Halffter, Ernesto, 317, 319. 
Halffter, Rodolfo, 319. 
Hamilton, George H., Painting and Sculp- 
ture in Europe: 1880-1940, 91 п’. 
Hamilton, Iain, 563, 567, 568. 
works, 567. 
Hampshire, Stuart, 404 n?. 
Handel, George Frideric, 221, 249, 334, 339. 
Hanslick, Eduard, Vom Musikalisch- 
Schönen, 80. 
Hanson, Howard, 584, 595, 614, 626. 
Merrymount, 595. 
Hardy, Thomas, 517, 539, 551. 
The Return of the Native, 517. 


746 INDEX 


Harris, Roy, 536, 584, 605, 606, 614-18, 620, 
622, 626. 
Andante for Orchestra, 614. 
Sextet for Clarinet, Piano and Strings, 
615. i 
Sonata for Piano, 615. 
Symphony No. 1, 615. 
Symphony No. 3, 615 (Ex. 259). 
Symphony No. 4, 618. 
Symphony No. 6, 618. 

Harrison, Lou, 602, 628. 

Hartleben, Otto Erich, 16, 144. 

Hartmann, Karl Amadeus, 409, 411, 416- 
417, 498. 

Viola Concerto, 417 (Ex. 192). 
other works, 416-17. 

Hartog, Howard (ed.), European Music in 
the Twentieth Century, 421 n!, 425 
nn? 3, 428 nt, 444 пі, 

Twentieth-Century Music, 410 n!. 

Haubenstock-Ramati, Roman, 487. 

works, 487. 

Hauer, Josef Mathias, 342-3. 

Vom Wesen des Musikalischen: Ein Lehr- 
buch der Zwélftonmusik, 342 n?. 

Zwülftontechnik: Die Lehre von den 
Tropen, 342 п?. 

Haydn, Joseph, 223, 295, 527, 537. 

Hebbel, Friedrich, 319. 

Heine, Heinrich, 684. 

Hell, Henri, Francis Poulenc, 428 n°. 

Henry, Pierre (with Pierre Schaeffer), 
Symphonie pour un homme seul, 488. 

Henze, Hans Werner, 411, 413, 496, 497- 
502. 

Apollo et Hyacinthus, 500. 
Boulevard Solitude, 500. 
Drei Dithyramben, 500, 501 (Exx. 219, 
220). 

Elegy for young lovers, 502. 
The Emperor's Nightingale, 501. 
Five Neapolitan Songs, 501. 
Jack Pudding, 500. 
Kammermusik 1958, 502. 
Kónig Hirsch, 501. 
Ode to the West Wind, 500. 
Der Prinz von Homburg, 502. 
Quartet for Strings, 500. 
Sonata for Piano, 501. 
Symphony No. 2, 498-500 (Ех. 218), 501. 
Variations for Piano, 497-8 (Ex. 218). 
Violin Concerto, 497 n!. 

Heraclitus, 208. 

Heseltine, Philip, 519-20. 
Carlo Gesualdo, 520 n?. 
The English Ayre, 520. 
see also Warlock. 

Hindemith, Paul, 195, 210, 230, 274, 319, 
320, 321, 322, 327-38, 339, 340, 352, 
387, 388, 389, 394, 402-8, 409, 410, 413, 


415, 424, 425, 427, 430, 435—6, 494, 497, 
498—500, 523, 537, 540, 547, 567, 583, 


4 606, 614, 619, 620, 621, 626, 646, 657. 
Literary works 


А Composer's World, 404 n?. 

Elementary Training, 404 n!. 

Traditional Harmony Y and П, 404 пі. 

Unterweisung im Tonsatz, 336-8 (Exx. 
147-50), 404, 413. 


Chamber music 


Kammermusik Op. 24 no. 1, 328 (Ex. 
143), 329. 
Kammermusik Op. 36 no. 4, 330. 
Kammermusik Op. 36 no. 5, 330. 
Octet, 407 n!. 
Quartet for Strings No. 2, 328. 
Quartet for Strings No. 3, 329. 
Quartet for Strings No. 4, 330. 
Quartet for Strings No. 5, 405. 
Quartet for Strings No. 6, 405. 
Schulwerk für Instrumental-Zusammen- 
spiel, 330. 
Choral works 
Sing-und-Spielmusik für Liebhaber und 
Musikfreunde, 330. 
Das Unaufhórliche, 333—4 (Ех. 145). 
When lilacs last in the door-yard 
bloomed, 404. 
Keyboard works 
Ludus Tonalis, 405-6 (Ex. 188). 
Sonata for Piano Duet, 335. 
Sonata for Two Pianos, 406. 
Sonatas for Piano, 335. 
Suite for Piano Op. 26, 329. 
Orchestral works 
Concerto for Horn, 406. 
Concerto for Orchestra Op. 38, 330, 
409 п. 
Concerto for Piano Op. 29. 330. 
Die Harmonie der Welt Symphony, 407. 
Mathis der Maler Symphony, 335, 407. 
Metamorphoses on themes of Weber,405. 
Philharmonic Concerto, 333 (Ex. 144). 
Der Schwanendreher, 335. 
Sinfonia serena, 405. 
Solo song 
Das Marienleben, 329, 332, 407-8 (Ex. 
189 (i, ii)). 
Stage works 
Cardillac, 195, 331-2, 407, 408. 
The Four Temperaments, 406—7. 
Die Harmonie der Welt, 407. 
Hérodiade, 406. 
Hin und Zurück, 332. 
Mathis der Maler, 195, 334-5 (Ex. 
146), 404, 407, 419, 500. 
Mörder, Hoffnung der Frauen, 331. 
Neues vom Tage, 332, 352, 407. 
Nobilissima Visione, 335, 539. 
Das Nusch-Nuschi, 331. 


INDEX 


Hindemith, Paul—Stage works (cont.): 
Sancta Susanna, 331. 
Wir bauen eine Stadt, 330. 
Hitler, Adolf, 409 n°, 
Hodeir, André, Since Debussy, 453 n*. 
Hoérée, Arthur 
Albert Roussel, 25 n?, 112 n!, 232 n!. 
Hoffmann, E. T. A., 32, 202, 264. 
Das Früulein von Scuderi, 331. 
Hoffmannsthal, Hugo von, 149, 152, 153, 
322, 323; 324: 
Jedermann, 321. 
Prologue to Schnitzler’s Anatol, 152. 
Holberg, Ludvig, 191. 
Holbrooke, Joseph, 187, 506. 
works, 187. 
Hölderlin, Johann Christian Friedrich, 410. 
Hollander, Hans, 306. 
Leos Janáček, 306 n!. 
Holmboe, Vagn, 427. 
Holst, Gustav, 39—40, 78, 117, 187, 503, 
508, 513-19, 535, 648. 
Beni Mora, 518. 
Choral Symphony, 517. 
Egdon Heath, 516, 517. 
A Fugal Concerto, 517. 
The Hymn of Jesus, 513, 514-17 (Exx. 
226-8). 
Hymns from the Rig Veda, 517. 
Lyric Movement, 517. 
Ode to Death, 513. 
The Perfect Fool, 517. 
The Planets, 40, 78, 117, 503, 516, 517. 
Savitri, 40 (Ex. 14), 187, 503, 517, 519. 
Scherzo, 517. 
Two Eastern Pictures, 518. 
Holst, Imogen 
Gustav Holst, 117 nt, 517 пп! ?. 
The Music of Gustav Holst, 517 n?, 519 п!. 
Homer, 77e Iliad, 522. 
Honegger, Arthur, 86, 239, 240, 245-50 
309, 419, 488, 524, 641. 
Je suis compositeur, 86 n!. 
Antigone, 250. 
Jeanne d'Arc au bücher, 250. 
Judith, 250. 
Pacific 231, 248, 488. 
Pastorale d'été, 248. 
Quartet for Strings No. 1, 248. 
Quartet for Strings No. 2, 248. 
Quartet for Strings No. 3, 248 (Ex. 106). 
Le Roi David, 249-50 (Ex. 107), 641. 
Rugby, 248, 309. 
Sonata for Violin No. 1, 248. 
Sonata for Violin No. 2, 248. 
Symphony in Three Movements, 248. 
Symphony No. 2, 419. 
Hornbostel, E. M. von, 601. 
Housman, A. E., 521. 
A Shropshire Lad, 506. 


741 


Hovhaness, Alan, 626. 
Howells, Herbert, 521. 
works, 521. 
Huber, Klaus, 494. 
Auf die ruhige Nachtzeit, 494. 
Humperdinck,- Engelbert, 3, 146-7, 148, 
152. 
Hänsel und Gretel, 3, 78, 146. 
Konigskinder, 146-7 (Ex. 64). 


Ibert, Jacques, 256. 
Illica, Giovanni, 154. 
Illica, Luigi, 156. 
Imbrie, Andrew, 626. 
d'Indy, Vincent, 20-2, 25, 29, 49, 63, 64, 90, 
108, 110, 112, 170-1, 232. 
Cours de composition musicale, 20, 21 
пп1,3, 5,6,7, 90, 
Richard Wagner et son influence sur l'art 
musical français, 20 пі. 
Divertissement, 25. 
L'Étranger, 21, 147 n?, 169-71 (Ex. 80). 
Fervaal, 2, 20, 21 (Ex. 21 (i)), 63, 90, 169. 
Jour d'été à la montagne, 21, 25, 110. 
Istar, 21. 
Légende de Saint Christophe, 21. 
Piano Sonata in E, 21. 
Poéme des montagnes, 21. 
Quartet for Strings No. 1, 21. 
Quartet for Strings No. 2, 21. 
Sonata for Violin, 110. 
Symphony No. 2, 21, 108 (Ex. 46). 
Tableaux de voyage, 21. 
‘Internationale’, 639. 
Ippolitov-Ivanov, Mikhail, 639, 641, 672, 
699. 
Ireland, John, 506—7, 547. 
works, 506—7. 
Isaac, Heinrich, 376, 385. 
Isamitt, Carlos, 638. 
Istoria russkoy sovetskoy muziki (ed. D. B. 
Kabalevsky et al.) 642 n°, 643 пп! *, 
647 n!, 648 п?, 650 n?, 661 n!, 665 
пп 3, 666 n!, 667 nn»? 54, 671 пп%5, 
672 nn? 5, 677 пі, 680 n!, 684 n!, 686 
n!, 688 n!, 693 n!, 695. 
Iturriaga, Enrique, 638. 
Ives, Charles, 40, 574—83, 597, 601, 602, 607, 
610, 615, 622, 626, 628, 632. 
Essays Before a Sonata, 577. 
The Anti-Abolitionist Riots, 580 (Ex. 250). 
Concord Sonata, 575, 571, 580, 582. 
114 Songs, 575. 
Symphony No. 1, 580. 
Symphony No. 4, 582. 
Three Places in New England, 580 (Ex. 
249). 
Universe or Universal Symphony, 582, 628. 
Variations on “Атегіса”, 576 (Ex. 248). 
Iwaszkiewicz, Jaroslaw, 314. 


748 


Jacob, Max, 237. 
Jacobi, Frederick, 606. 
Jacobsen, Jens Peter, 15, 39. 
James, Henry, The Heiress, 592. 
Janáček, Leoš, 9, 36, 179, 184, 250, 275, 
301-9, 315. 
Blanik, 307. 
Glagolska mše (Glagolitic Mass), 304-7 
(Ex. 137), 308. 
ей pastorkyňa (Jenůfa), 9, 179-81 (Exx. 
84, 85), 184, 302. 
Katya Kabanova, 302-3. 
Prihody Lišky Bystroušky (The Cunning 
Little Vixen), 301 (Ex. 136), 303. 
Quartet for Strings No. 1, 307-8. 
Quartet for Strings No. 2, 308. 
Sinfonietta, 307. 
Taras Bulba, 307. 
Рес Makropulos 
Affair), 303-4. 
Vylety Pané Brouckovy (The Excursions 
of Mr. Brouček), 302. 
Zapisnik zmizelého (Diary of One Who 
Vanished), 36, 307-8 (Ex. 138). 
Z mrtvého domu (From the House of the 
Dead), 304. 
Jankélévitch, Vladimir, Gabriel Fauré: 
Ses mélodies—Son esthétique, 26 n!. 
Jarnach, Philipp, 194 n!. 
Jelinek, Hanns, 418. 
Anleitung zur Zwólftonkomposition, 418 n!. 
Zwólftonwerk, 418. 
Jenssen, G. Wiers, 7he Witch, 268. 
*La Jeune France', 231, 254-6, 429. 
Job, 508, 512. 
Joergensen, Johannes, St. Francis of Assisi, 
220. 
Johnson, Harold E., Sibelius, 38 n!. 
Jolivet, André, 255, 429. 
works, 255. 
Jone, Hildegard, 379, 380, 382. 
Jonson, Ben, 522, 551. 
Epicoene, 324. 
Joyce, James, 420, 563, 595. 
Portrait of the Artist as a Young Man, 559. 
Ulysses, 492, 560. 


(The Makropulos 


Kabalevsky, Dmitry, 642, 648 n?, 672, 683, 

684, 688, 693. 

Colas Breugnon, see Master iz Klamsi. 

Concerto for Cello, 693. 

Concerto for Piano, 693. 

Concerto for Violin, 693, 694 (Exx. 289, 
291). 

Master iz Klamsi (The Master of Clamecy: 
Colas Breugnon), 667. 

Nikita Vershinin, 691. 

Pod Moskvoy (Near Moscow: V ogne), 
676, 683. 

Quartet for Strings No. 1, 657. 


INDEX 


Quartet for Strings No. 2, 683. 

ма velikaya (The great motherland), 
684. 

Semya Tarasa (Taras's family), 683, 688, 
694 (Ex. 291), 695. 

Sonata for Piano No. 2, 683 

Sonatinas Op. 13, 674. 

Symphony No. 1, 672. 

Symphony No. 2, 672. 

Symphony No. 4, 696. 

V ogne (In the Fire), see Pod Moskvoy. 

Kafka, Franz, 500. 

Kagel, Mauricio, 410, 486. 
works, 486. 

Kallmann, Chester, 391. 
see also under W. H. Auden. 

Kandinsky, Wassily, 203. 

Kant, Immanuel, 359, 375. 

Karaev, Kara, 700. 

Karlowicz, Mieczyslaw, 36. 

Kastalsky, Aleksandr, 640, 641-2. 

Kataev, Valentin, 668. 

Keats, John, 551, 555. 

Keller, Gottfried, 319. 

Die Leute von Seldwyla, 184. 

Keller, Hans, 395 nt. 
see also under Donald Mitchell. 

Kelley, Edgar Stillman, 607. 

Kepler, Johann, 407. 

Kerridge, W. H., 15 n°. 

Khachaturyan, Aram, 672, 686, 697, 700. 
Concerto for Piano No. 1, 672. 
Concerto for Violin, 672. 

Gayane (Schastye), 676, 697. 

‘Kapitan Gastello’, 684. 

Schastye (Happiness), see Gayane. 

Spartak (Spartacus), 697. 
Khachaturyan: Aram (cont.) 

Symphony No. 1, 672. 

Symphony No. 2, 683. 

Toccata, 674. 

Trio for piano, violin and clarinet, 668. 

Khodzha-Eynatov, Leon, 667. 
works, 667. 

Khrennikov, Tikhon, 667-8, 669, 684, 686, 

688. 

Bratya (Brothers), see V buryu. 

Frol Skobeyev, 688. 

Mat (The Mother), 691. 

Symphony No. 1, 672. 

Symphony No. 2, 683. 

V buryu (In the storm: Bratya), 667-8 
(Ex. 278). 

Khrushchev, Nikita Sergeyevich, 682. 

Khubov, Georgy, 672 n°. 

Kirchner, Leon, 573, 614, 625, 628-30. 
Concerto for Piano, 630. 

Duo for Violin and Piano, 628. 
Quartet for Strings, 628 (Ex. 261). 
Sonata for Piano, 628. 


INDEX 


Kireyevsky, Peter Vasilyevich, 215. 
Kirkpatrick, John, 580. 
Kirsanov, S., 656. 
Klebe, Giselher, 496. 
works, 496. 
Klee, Paul, 203, 373. 
Klein, H. F., 368. 
Klopstock, Friedrich Gottlieb, 7. 
Des Knaben Wunderhorn, 6, 7, 72. 
Knipper, Lev, 641, 646, 672, 700. 
Kurumkan, 700. 
Severny veter (North Wind), 646. 
Skazki gipsovovo Ви (Tales of the 
Plaster Buddha), 651. 
Symphony in D minor, 651. 
Symphony No. 3, 671. 
Symphony No. 4, 671-2. 
Symphony No. 6, 672. 
Symphony No. 7, 672. 
Symphonies 10-14, 696. 
Turkmenskie eskizi, 700. 
Vanch Suite, 672, 700. 
Kodály, Zoltán, 10, 19, 115, 275, 276, 277, 
299—301, 559. 
works, 300-1. 
Koechlin, Charles, 29, 110, 112, 113-14. 
L'Abbaye, 113-14 (Ex. 50). 
other works, 113-14. 
Koellreutter, Hans-Joachim, 638. 
Koenig, Gottfried Michael 
Klangfiguren, 490. 
Kokoschka, Oskar, 331, 340. 
Kolneder, Walter, Anton Webern, Ап 
Introduction to his Works, 376 n+. 
‘Kol Nidre', 357. 
Korchmarev, Klimenty, 647, 648. 
Ivan-soldat, 645. 
Krepostnaya balerina (The Serf Ballerina), 
648. 
Kosuszek, С. F., 411 п!, 
Kotoriski, Wlodzimierz, 487. 
works, 487. 
Koval, Marian, 642, 671, 674. 
Emelyan Pugachev (opera), 676. 
Emelyan Pugachev (oratorio), 671. 
Pushkiniana, 674 (Ех. 282 (i)). 
Sevastopoltsi (Тһе Defenders of Sevasto- 
pol), 676. 
Kozlov, P., see under M. Iordansky. 
Krasner, Louis, 369. 
Krauss, Clemens, 324. 
Krein, Aleksandr, 645. 
Symphony No. 1, 651. 
Traurnaya oda (Funeral ode), 651. 
Zagmuk, 645. 
Křenek, Ernst, 338-40, 352, 409, 418-19, 
427 n+, 606, 619, 646. 
Studies in Counterpoint, 340 n!, 345 пі. 
Uber neue Musik, 340 пі. 
Der Diktator, 340. 
Fiedellieder, 340. 


49 


749 


Das geheime Königreich, 340. 
Jonny spielt auf, 340, 352, 419, 641. 
Karl V, 340. 
Kette, Kreis und Spiegel, 418. 
Lamentations, A18. 
Das Leben des Orest, 340, 419. 
Orpheus und Eurydike, 340. 
Pallas Athene weint, 419. 
Quartets for Strings, 340. 
Reisebuch aus den ósterreichischen Alpen, 
340. 
Schwergewicht oder die Ehre der Nation, 
340. 
Spiritus Intelligentiae Sanctus, 419. 
Der Sprung über den Schatten, 340, 641. 
Symphonies, 340. 
Die Zwingburg, 340. 
Krieger, Edino, 638. 
Krilov, Ivan, 660. 
Krupskaya, N. К, 
Lenine, 695 n*. 
Kučera, Vaclav, Nové proudy v sovetské 
hudbé, 699 пт. 
Küchelbecker, Wilhelm Karlovich, 696. 
Kussevitsky, Sergey, 223, 588, 615. 


Vospominaniya о 


Laloy, Louis, 233. 
Lambert, Constant, 523-7. 
Music Ho! A Study of Music in Decline, 
50352405271 
Aubade héroique, 526. 
Concerto for Piano and Nine Instru- 
ments, 526. 
Dirge from Cymbeline, 526-7. 
Horoscope, 526. 
Music for Orchestra, 524 (Ex. 229). 
Pomona, 524. 
The Rio Grande, 527. 
Romeo and Juliet, 524. 
Sonata for Piano, 526. 
Summer's Last Will and Testament, 524 
(Ех. 230), 527. 
Lamberti, X., Stagione, 263. 
Landormy, Paul, 20. 
La musique française de 
Debussy, 20 n?, 22 n?. 
Lanner, Josef, 198. 
LaTouche, John, 607. 
Lavignac, Albert, 91. 
Lavín, Carlos, 637. 
Layton, Robert, Sibelius, 37 n?. 
Lecca, Celso Garrido, 638. 
Léhar, Franz, 171. 
Die lustige Witwe, 171. 
Leibowitz, René, 381, 435, 436, 442, 497. 
Introduction à la musique de douze sons, 
435 п?. 
Schönberg et son école, 435 п?. 
Lekeu, Guillaume, 22. 
Lenau, Nikolaus, 319. 


Franck à 


750 


Lenin, Vladimir Ilyich, 640, 651, 668, 695. 
O literature i iskusstve, 640 n?. 
Sochineniya, 640 п?. 

Leoncavallo, Ruggiero, 31, 154, 256. 

La Bohéme, 155. 
Pagliacci, 154. 
Zaza, 147 n?, 154, 164. 

Leopardi, Giacomo, 274, 424. 

Lermontov, Mikhail Yurevich, 676. 

Leskov, Nikolay Seménovich, 644, 664. 

Lesur, Daniel, 255, 429. 

Levit, S., 643 nn? $, 661 n!, 667 nn? +. 
Yury Aleksandrovich Shaporin, 671 п?. 

Lewinski, W.-E. von, 436 n!. 

Lewkovitch, Bernhard, 495. 

Ley, Rosamond, 32 n?. 

Lidholm, Ingvar, 495. 

Liebermann, Rolf, 494. 

Concerto for Jazz Band and Orchestra, 
494. 

Liess, Andreas 
Carl Orff, 409 n*. 

Claude Debussy: das Werk im Zeitbild, 
90 në, 

Ligeti, Gyorgy, 445, 446, 491-2. 
Artikulation, 491-2. 

Lincoln, Abraham, 618. 

Lindbergh, Charles, 309. 

Lion, Ferdinand, 331. 

Li-Po, 522. 

Lisenko, Mikola, 700. 

Liszt, Franz, 19-20, 24, 31-2, 35, 39, 65, 68, 
78, 82, 102, 167, 192, 200, 235, 276, 279, 
295, 327, 562, 563. 

Christus, 200. 
Faust Symphony, 78. 
Hungarian Rhapsodies, 276. 

Livanova, Tamara, 648 n?, 650. 

Locke, Matthew, 520. 

Lockspeiser, Edward, Debussy: his Life and 
Mind, 90 n?, 920пЪ 3, 101 n°, 105 п?. 

Loeffler, Charles Martin, 39, 574. 

‘Lonely Waters’, 519. 

Lorca, Federico Garcia, 415, 696. 

Lord, A., see under Béla Bartok. 

Lourié, Arthur, 639. 

Louys, Pierre, 94. 

Luening, Otto, 605, 622, 627. 

Lully, Jean-Baptiste, 219. 

Lunacharsky, Anatoly, 635. 

Lutostawski, Witold, 487. 

Muzyka zalobna (Funeral Music), 487. 

Lutyens, Elisabeth, 556-7, 568. 
works, 556-7. 

Lyadov, Anatoly, 33, 229, 639. 

Lyapunov, Sergey, 33. 

Piano Sextet, 657. 

Lyatoshinsky, Boris, 700. 


MacDowell, Edward, 38, 570, 571, 572-3, 
595, 606, 626. 


INDEX 


Machiavelli, Niccolo, 269. 
Mackenzie, Alexander, 38. 


“McNaught, William, 518. 


Modern Music and Musicians, 518 n!, 
520 n!. 
Maconchy, Elizabeth, 557. 
Maderna, Bruno, 469, 479, 492. 
Musica su due dimensioni, 493. 
Notturno, 492. 
Quartet, 479. 

Maeterlinck, Maurice, 16, 183, 202, 203. 
Ariane et Barbe-Bleue, 169, 183, 202. 
Pelléas et Mélisande, 15, 166, 168, 183, 

196, 202. 

Magnard, Albéric, 22-4. 
Symphony No. 3, 22-4 (Ex. 6). 

Mahler, Alma Maria, 8, 369-70. 

Gustav Mahler: Erinnerungen und Briefe, 
Sint, 72n* 

Mahler, Gustav, 1, 2, 5-9, 10, 15, 16, 18, 30, 
33, 34, 37, 39, 43-7 passim, 49, 63-7 
passim, 72, 73, 74, 78, 81, 86, 92, 105, 
133, 140, 147-8, 152, 189, 193, 267, 293, 
302, 364, 372, 507, 547, 548, 611, 650, 
655, 673, 682, 696. 

Ernst von Schwaben, 6. 
Fünf Lieder nach Rückert, 72. 
Kindertotenlieder, 9, 72. 
Das klagende Lied, 6. 
Des Knaben Wunderhorn, 72, 147. 
*Ablósung im Sommer’, 7. 
‘Des Antonius von Padua Fisch- 
predigt’, 6. 
‘Es sungen drei Engel’, 7. 
‘Urlicht’, 7. 
Lieder eines fahrenden Gesellen, 6, 72. 
Das Lied von der Erde, 9, 44, 72, 78, 372, 
682 п?, 696. 
Symphony No. 1, 2, 6. 
Symphony No. 2, 5, 6-7, 8, 78, 92, 655. 
Symphony No. 3, 5, 7, 47, 78, 148. 
Symphony No. 4, 5, 7, 47 (Ex. 21 (ii)), 63, 
655. 
Symphony No. 5, 8, 9, 65-7 (Ex. 23), 81, 
655. 
Symphony No. 6, 8, 45-6 (Ex. 17). 
Symphony No. 7, 8, 9, 128 (Ex. 55 (1)). 
Symphony No. 8, 8, 9, 15, 78, 148. 
Symphony No. 9, 8, 9. 
Symphony No. 10, 8. 
Malipiero, Gian Francesco, 256-8, 262-7, 
268, 424. 
Cantari alla madrigalesca, 262. 
La favola del figlio cambiato, 266-7. 
L'Orfeide, 264—6. 
Pause del silenzio, 262. 
Rispetti e strambotti, 262-3 (Ex. 111). 
Prima sinfonia, in quattro tempi come le 
quattro stagioni, 263. 
Seconda sinfonia (Elegiaca), 263. 
Stornelli e ballate, 262-3. 


INDEX 751 


Torneo notturno, 264, 265-6 (Ex. 112). 
Tre commedie goldoniane, 266. 
Mallarmé, Stéphane, 148, 406, 466. 
Livre, 453, 466. 
Mantecón, Juan José, 319. 
Manuel, Roland, 104, 235. 
Maurice Ravel, 235 п!. 
Marc, Franz, 203. 
Marshak, Samuil Yakovlevich, 684. 
Martin, Frank, 320-1, 419-20, 435, 454, 
494, 537. 
Ballades, 320. 
Concerto for Piano, 320. 
Concerto for Wind, 420 
Der Cornet, 321. 
Golgotha, 420. 
Le Mystére de la Nativité, 420. 
Petite Symphonie Concertante, 419-20, 
Rhythmes, 320. 
Six Monologues, 321. 
Symphony, 320. 
The Tempest, 420. 
Trio sur des chants populaires irlandais, 
320. 
Le Vin herbé, 320. 
Martinů, Bohuslav, 309-10, 389, 428, 619. 
Half-Time, 248, 309. 
other works, 309-10. 
Martinov, Ivan, 672 në. 
Martucci, Giuseppe, 31, 256. 
Marx, Joseph, 9. 
Marx, Karl, 656, 660, 694. 
Marx, Roger, 91. 
Mary Queen of Scots, 273. 
Mascagni, Pietro, 31, 154—5, 177, 183, 193, 
256. 
L’Amico Fritz, 147 nt, 183. 
Cavalleria rusticana, 154, 179. 
Iris, 154—5 (Ex. 68), 162 (Ex. 73), 183. 
Le Maschere, 193. 
Mason, Colin, 453 n?, 526 п\, 
Massenet, Jules, 2, 20, 22, 92, 155, 159, 164, 
165-6, 171-2. 
Grisélidis, 164. 
Jongleur de Notre-Dame, 147 n!, 164. 
Manon, 155, 164, 187. 
La Navarraise, 164. 
Sapho, 164. 
Thais, 164, 187. 
Thérése, 172. 
Werther, 147 n?, 164. 
Massine, Leonid, 335. 
Matisse, Henri, 237. 
Maupassant, Guy de, 148. 
Maw, Nicholas, 568. 
Medici, Lorenzo de', 264. 
Mellers, Wilfrid, 518. 
Studies in Contemporary Music, 518 
rego ЗИ se, 
Mendelssohn, Felix, 219. 
Mennin, Peter, 573, 618, 626. 


Menotti, Gian Carlo, 573, 620, 621. 
works, 621. 

Mérimeé, Prosper, 235. 

Messager, André, 171. 
works, 171. 

Messiaen, Olivier, 255-6, 388, 410, 423, 
428-34, 435, 437, 438-41, 442, 445, 
454, 460, 473, 489, 568. 

Technique de mon langage musicale, 
430 n?. 

Cantéyodjayá, 439 n!, 445. 

Cinq rechants, 410. 

Les corps glorieux, 256. 

Ile de Feu 2, see under Quatre Études de 
rhythme. 

Livre d'orgue, 441 n!. 

Mode de valeurs et d'intensités, see under 
Quatre Études de rhythme. 

La Nativité du Seigneur, 255, 256. 

Neumes rhythmiques, see under Quatre 
Études de rhythme. 

Oiseaux exotiques, 430 (Ex. 196 (1)), 433- 
434 (Ex. 198). 

Quatre Études de rhythme, 442. 
Ile de Feu 2, 432 n?, 439-41 (Ex. 200). 
Mode de valeurs et d'intensités, 434, 

438-9 (Ех. 199), 441, 445, 446. 

Neumes rhythmiques, 431 n?. 

Quatuor pour la Fin du Temps, 431 (Ex. 
196). 

Le Réveil des Oiseaux, 433. 

Timbrés-Durées, 488. 

Trois Petites Liturgies de la Présence 
Divine, 433. 

ТигапгаШа-бутрйоше, 433. 

Vingt Regards sur l'Enfant Jésus, 431. 

Visions de l'Amen, 432-3 (Ex. 197). 

Metner, Nikolay, 33, 657. 

Metzger, Heinz-Klaus, 436 n!, 437 n!. 

Meyer-Eppler, Werner, 454. 

Meyerbeer, Giacomo, 152, 154, 159, 164, 
680. 

Les Huguenots, 194, 643. 
Robert le Diable, 164. 

Michelangelo, 359. 

Mignone, Francisco, 636. 

Milhaud, Darius, 230, 239, 240—5, 262, 309, 
315, 339, 340, 454, 526, 583, 588, 606, 
619. 

Notes sans Musique, 240 пі. 

L'Abandon Ф Ariane, 243. 

Le Beuf sur le toit, 241. 

Catalogue des fleurs 242, 243 (Ex. 104). 

Christophe Colomb, 243—5 (Ex. 105), 340. 

Cinq Symphonies, 242 (Ex. 103). 

La Création du Monde, 241 (Ex. 102), 585, 
588. 

La Déliverance de Thésée, 243. 

L'Enlévement d'Europe, 243. 

Machines agricoles, 242, 243. 

Les Malheurs d'Orphée, 243. 


752 


Milhaud—(cont.) 
Maximilien, 245. 
Мейее, 245. 
Le Pauvre Matelot, 243. 
Les quatre éléments, 245. 
Quartets for Strings 5—9, 245. 
Trois chansons de troubadour, 245. 
Trois élégies, 245. 
Milner, Anthony, 563, 567-8. 
Improperia, 567—8. 
other works, 568. 
Mirmont, Jean de la Ville de, 232. 
Mirzoyan, Eduard, 700. 
Mistral, Gabriela, 699. 
Mitchell, Donald, 72 n!. 
Gustav Mahler: The Early Years, 6 n*. 
(with Hans Keller, ed.) Benjamin Britten: 
a Commentary on his works, 550 n!. 
Moeran, E. J., 519. 
works, 519. 
Moeschinger, Albert, 320. 
Mokrousov, Boris, 676. 
Chapaev, 676. 
Molchanov, Kirill, 691. 
works, 691. 
Moliére, 254. 
Le Misanthrope, 547. 
Molotov, Vyacheslav Mikhailovich, 665. 
Mompou, Federico, 318. 
works, 318. 
Monet, Claude, 91, 94. 
Montemezzi, Italo, 31, 188. 
L'amore dei tre re, 31, 188 (Ex. 88). 
La Nave, 188. 
Monteverdi, Claudio, 262 n!, 268, 536. 
Moore, Douglas, 605, 606-7. 
works, 607. 
Morax, René, 249, 250. 
Morillo, Roberto García, 638. 
Morris, William, 183. 
Morselli, E. L., 268. 
Mosolov, Aleksandr, 641, 660, 665, 674 
(Ex. 282 (ii)). 
Gazetnie obyavleniya (Newspaper adver- 
tisements), 660. 
Stal (Steel), 651. 
Motte, Diether dela, Hans Werner Henze— 
Der Prinz von Homburg, 497 n?, 502 n!. 
Mozart, Wolfgang Amadeus, 193, 225, 226, 
235, 310, 322, 325 391, 536.587. 
Cosí fan tutte, 391. 
Don Giovanni, 192, 391. 
Sonata for Two Pianos K.448, 226. 
Die Zauberflöte, 322. 
Muradeli, Vano, 685, 686-7. 
Symphony No. 2, 683. 
Velikaya druzhba (The great friendship), 
685, 686—7 (Ex. 286). 
Murger, Henri, 155, 156, 164. 
Vie de Bohéme, 155. 
Musgrave, Thea, 563. 


INDEX 


Mussorgsky, Modest, 92, 140, 166, 167,179, 
- 230, 301, 540, 640, 661, 680, 689. 
Boris Godunov, 166, 167, 177, 661, 679, 
689. 

Myaskovsky, Nikolay, 36, 639, 641, 642, 
648-51, 657, 660, 681, 683, 684, 685, 
686, 699. $ 

Concerto for Violin, 670. 

Kirov s nami (Kirov is with us), 684. 

Quartets for Strings Op. 33, 657. 

Quartets for Strings 7-12, 683. 

Symphonies nos. 4 and 5, 648. 

Symphony No. 6, 648-50 (Ехх. 266, 267), 
671. 

Symphonies nos. 8, 10 and 11, 650. 

Symphony No. 12, 651. 

Symphonies nos. 14, 16, and 21, 670. 

Symphony No. 22, 681. 

Symphony No. 23, 681 (Ex. 285 (1)). 

Myers, Rollo, 239 n!. 

Erik Satie, 167 п}. 


Napoleon I, 357. 
Nechaev, Vasily, 660, 661. 
Ivan Bolotnikov, 661. 
О doblestyakh, o podvigakh, o slave (ОҒ 
heroism, great deeds, and fame), 684. 
Nekrasov, Nikolai Alexeevich, 674. 
Nestyev, I. W., Prokofyev, 231 n!, 670 n!, 
677 n!, 680 n!, 692 п!. 
Neumann, Angelo, 32. 
Neveux, Georges, 310. 
Newlin, Dika, 16. 
Bruckner — Mahler — Schoenberg, 16 
nn? 3,4, 68 n?, 85 п!. 
Newman, Ernest, 81 n?. 
Nicholas I, Tsar, 643. 
Nichols, Robert, 522. 
Nielsen, Carl, 37, 38, 73, 191-2. 
Levende Musik, 38 n°. 
Maskarade, 191—2 (Ex. 90). 
Saul og David, 191. 
Symphonies, 38. 
Nietzsche, Friedrich, 7, 35, 149, 166, 217. 
Also sprach Zarathustra, 7, 39. 
Nilsson, Bo, 484-5. 
Zwanzig Gruppen, 484—5 (Ex. 216). 
other works, 484. 
Nizhinska, B. F., 524. 
Nono, Luigi, 373, 413, 434, 437, 441 n’, 
464, 469-77. 
Canonic Variations, 469. 
Il Canto sospeso, 470-3 (Ex. 210). 
Composizione 2 (Diario polacco 1958), 
471 (Ex. 211). 
Cori di Didone, 473. 
Due Espressioni, 469. 
Epitaph for Frederico Garcia Lorca, 469. 
Incontri, 413, 470 (Ex. 209), 473. 
Intolleranza 1960, 469, 473. 
Polifonica- Monodia- Ritmica, 469. 


INDEX 158 


Novák, Vitézslav, 9. 
Noverre, Jean Georges, 196. 
Nüll, Edwin van der, 115. 
Béla Bartók: Ein Beitrag zur Morphologie 
der neuen Musik, 19 n*, 115 n? 
Nystroem, Gósta, 427. 
Sinfonia del Mare, 427. 


Odak, Krsto, 304 n!. 
Offenbach, Jacques, 172. 
La Belle Héléne, 172. 
O'Neill, Eugene, The Emperor Jones, 595. 
Orff, Carl, 409-10, 411. 
works, 410. 
d'Orléans, Charles, 100. 
Orlov, Genrikh, Simfonii Shostakovicha, 
651 п?, 674 ппЪ?, 682 n?, 694 n!, 
695 nn? 4. 
Ornstein, Leo, 574. 
Orrego-Salas, Juan, 637. 
works, 637. 
Ostrovsky, A. N., Groza (The Storm), 302, 
676, 691. 
Overbeck, Johann Friedrich, 183. 
Owen, Wilfred, 522, 542. 


Paart, Arvo, 699. 

Paine, John Knowles, 570-1, 614, 619. 

Palestrina, G. P. da, 20, 195, 200, 300, 312. 
Missa Papae Marcellae, 195. 

Palmer, Robert, 626. 

*Pange lingua', 514. 

Pappenheim, Marie, 204. 

Parker, Horatio, 38, 40, 570-1, 575, 614, 619. 

Parry, Hubert, 38, 40, 506, 508. 

Partch, Harry, 625. 

Pashchenko, Andrey, 643-4, 650. 
Cherny yar (The Black Crag), 644. 
Orliny bunt (The Eagles' Rising), 643, 644, 

663. 

Tsar Maximilian, 645. 

Pauli, Hansjórg, 398 n!. 

Payne, Anthony, Schoenberg, 17 n?. 

Paz, Juan Carlos, 638. 

Pears, Peter, 550, 555. 

Pedrell, Felipe, 30, 201, 275 n!, 315, 318, 

319. 
Cancionero musical popular español, 30. 
Els Pireneus, 30, 201. 
Péladan, Sar, Le fils des étoiles, 93. 
Penderecki, Krzystof, 487. 
Tren—Ofiarum Hiroszimy (Threnody to 
the Victims of Hiroshima), 487. 
Pepping, Ernst, 413. 
Pergolesi, Giovanni Battista, 217, 267. 
Perle, George, 367, 625. 

Serial Composition and Atonality, 137 n?. 
Persichetti, Vincent, 618, 620, 621, 626. 
Peter the Great, 218. 

Petrarch, 346. 
Petrassi, Goffredo, 270, 274, 424—5, 469. 


Concerto for Orchestra, 274. 
Concerto for Piano, 274. 
Third Orchestral Concerto (Récréation 
Concertante), 424. 
Fifth Orchestral Concerto, 424. 
Il Cordovano, 424. 
Cori di Morti, 274, 424. 
Magnificat, 274. 
Morte dell Aria, 424. 
Noche oscura, 424. 
Partita for orchestra, 274. 
Psalm IX, 274. 
Quartet for Strings, 425. 
Serenata, 425. 
Pfitzner, Hans, 3, 49, 147, 195—6, 322, 326, 
334. 
Die neue Aesthetik der musikalischen 
Impotenz, 326. 
Der arme Heinrich, 147. 
Palestrina, 147 п!, 195, 326, 334. 
Die Rose vom Liebesgarten, 147. 
Von deutscher Seele, 326. 
Piani, Casti, 365. 
Picabia, Francis, 237. 
Picasso, Pablo, 229, 236, 237, 595. 
Pijper, Willem, 321. 
works, 321. 
Pinilla, Enrique, 638. 
Pirandello, Luigi, 264, 266, 267, 630. 
Piston, Walter, 573, 584, 605, 606, 607, 614, 
622, 630. 
works, 614. 
Pittaluga, Gustavo, 319. 
Pizzetti, Ildebrando, 250, 256-62, 263, 267, 
268, 269. 
Musica e dramma, 258 n!. 
Musicisti contemporanei: Saggi critici, 
AIST) ja 
Canti della stagione alta, 262. 
Concerto dell'estate, 261—2. 
Debora e Jaele, 259-60 (Ex. 109). 
Fedra, 188, 259. 
Fra Gherardo, 259-60. 
Lo Straniero, 259-60. 
Messa da Requiem, 260-1 (Ех. 110). 
L'Oro, 260. 
Orséolo, 260. 
Rondo veneziano, 262. 
Plato, 175, 215, 238. 
Poliziano, Angelo, 264, 267. 
Polovinkin, Leonid, 641, 660. 
Elektrifikat, 660. 
Poslednyaya sonata (Last sonata), 660. 
Symphony No. 4, 672. 
Teleskopi, 651. 
Ponce, Manuel M., 636. 
Popov, Gavriil, 641, 686. 
Sextet, 657. 
Symphony No. 1, 672. 
Symphony No. 2, 683. 
Porter, Quincy, 605, 606. 


754 


Pototsky, Sergey, 
Through), 645. 
Poulenc, Francis, 230, 239, 240, 245, 250-4, 
309, 310, 428. 
Le Bestiaire, 251 (Ex. 108). 
Les Biches, 251. 
Cocarde, 251. 
Concert champétre, 251. 
Dialogue des Carmélites, 164, 428. 
Litanies à la Vierge Noire de Rocamadour, 
254. 
Les Mamelles de Tirésias, 428. 
Mass 251-4. 
Mouvements perpétuels, 251. 
Quatre Motets pour un temps de pénitence, 
254. 
Rhapsodie négre, 251. 
Sept Nocturnes, 251. 
Pousseur, Henri, 436 п“, 437 n*, 458 п!, 464, 
482-4, 492. 
Impromptu, 483—4 (Ex. 215). 
Mobile, 484, 492. 
Quintet, 483. 
Rimes pour différentes sources sonores,493. 
Scambi, 484, 492. 

Praga, Emilio, 155. 

Preissová, Gabriela, 179. 

Prévost, Abbé, 155, 164. 

Manon Lescaut, 500. 

Prochazka, Frantisek, 302. 

Prokofyev, Sergey, 14, 36, 229-31, 526, 
583, 620, 639, 641, 646, 651, 652, 657, 
667-70, 671, 672, 676-82, 686, 687-8, 
691-2, 693, 699. 

Chamber music 
Quartet for Strings No. 2, 677 (Ex. 
285 (ii)). 
Choral works 
Aleksandr Nevsky cantata, 671. 
Semero ikh (Seven, they are seven), 
230. 
Zimniy koster (Winter Bonfire), 693. 
Keyboard works 
Sarcasms, 230. 
Sonatas for Piano 7-9 682. 
Tales of the Old Grandmother, 230. 
Ten Pieces for Piano, 229. 
Toccata, 230. 
Visions fugitives, 230. 
Orchestral works 
Concerto for Cello, 696. 
Concerto for Piano No. 1, 229, 230. 
Concerto for Piano No. 2, 230. 
Concerto for Piano No. 3, 230, 231. 
Concerto for Violin No. 1, 231. 
Concerto for Violin No. 2, 672. 
Scythian Suite (Ala and Lolly), 230. 
Simfonicheskaya pesn (Symphonic 
song), 672. 
Sinfonia concertante for cello and 
orchestra, 696. 


Proriv (The Break- 


INDEX 


Symphony No. 1, 229, 639, 683. 

Symphony No. 2, 231. 

Symphony No. 3, 231. 

Symphony No. 5, 682. 

Symphony No. 6, 682, 683, 685. 

Symphony No. 7, 693. 

Stage and film music 
Aleksandr Nevsky, 667, 670. 
Eugene Onegin (incidental music), 679. 
Igrok (The Gambler), 230. 
Kamenniy tsvetok (The Stone Flower), 
691. 
Lyubov k trem apelsinam (Love for 
Three Oranges), 230, 641, 680. 
Obruchenie v monastire (Betrothal in a 
Priory), 680-1. 
Ognenniy Angel (The Flaming Angel), 
230, 231. 
Le pas d'acier, 231, 651. 
Poruchik Кігһе (Lieutenant Kizhe), 
667. 
Povest o nastoyashchem cheloveke (Тһе 
story of a real man), 688, 691-2 
(Ex. 288). 
Romeo and Juliet, 667, 679. 
Semen Kotko, 667, 668—70 (Exx. 279, 
280). 
Shut (Yhe Buffoon), 230. 
Voyna i mir (War and Peace), 676-80 
(Exx. 283, 284), 682, 691. 
Zolushka (Cinderella), 682. 
Protopopov, V., 671 nn* 5, 693 ni. 

(ed.), Pamyati S. I. Тапееуа, 699 п?. 
Proust, Marcel, 420. 

À la recherche du temps perdu, 74. 
Prussak, E. V., 643. 

see also under Axseny Gladkovsky. 
Puccini, Giacomo, 31, 78, 140, 155-9, 184, 

189, 195—6, 256-8, 265, 268, 326, 661. 

La Bohéme, 155-8 passim (Ех. 70), 162-3. 

La Fanciulla del West, 158, 257. 

Gianni Schicchi, 159, 258. 

Madama Butterfly, 156, 158, 183, 257. 

Manon Lescaut, 155. 

Suor Angelica, 159, 164, 257-8. 

Il Tabarro, 159, 164, 189, 257. 

Tosca, 147 nn^?, 156-9 passim (Exx. 69, 
71), 162, 257, 643. 

Turandot, 159, 187, 194, 195-6, 258, 661. 

Pugachev, Emelian, 643, 648. 

Purcell, Henry, 518, 519, 520, 532, 540, 548— 
552 passim, 556. 

Pushkin, Aleksandr Sergeyevich, 174, 218, 
647, 674, 676, 684. 

Bronze Horseman, 650. 

The Golden Cockerel, 176. 

The Little House in Kolomna, 218. 

Skupoy Ritsar (The Miserly Knight), 177. 


Raaben, L., Sovetskaya kamerno-instru- 
mentalnaya muzika, 683. 


INDEX 735 


Rabinovich, D., 696. 

Dmitry Shostakovich, 696 n+. 
Radiguet, Raymond, 254. 
Raff, Joachim, 38, 570, 574. 


Rakhmaninov, Sergey, 33, 34, 43, 78, 152, 


177, 639, 657. 
Francesca da Rimini, 155 n!, 177. 
Kolokola (The Bells), 78. 


Skupoy Ritsar (The Miserly Knight), 177. 


Die Toteninsel, 34. 
Raleigh, Walter, 684. 
Ramuz, C. F., 213. 


Ravel, Maurice, 29, 30, 67, 68, 72, 78, 86, 
101-5, 108, 110, 112, 114, 115, 127, 
144, 172-4, 191, 199, 200, 223, 231, 
233, 234-6, 240, 251, 254, 267, 312, 


320, 321, 466, 503, 506, 583, 641, 682. 
Chamber music 
Introduction et Allegro, 104. 
Quartet for Strings, 30, 68, 104. 
Sonata for violin and cello, 30, 235. 
Sonata for violin and piano, 235. 
Trio, 30, 144. 
Keyboard works 
Gaspard de la Nuit, 30, 101, 104. 
*Scarbo', 102 (Ex. 42). 
Jeux d'eau, 101, 104, 199. 
Ma Mère l'Oye, 93, 200. 
Miroirs, 30, 101, 104. 


‘Alborada del gracioso’, 103 (Ех. 43). 


*La Vallée des cloches', 128. 


Prélude à laprès-midi d'un faune 


(Debussy, arranged Ravel), 101. 
Les Sites auriculaires, 101, 103. 
Sonatina, 30, 101, 104. 

Le Tombeau de Couperin, 30. 


Valses nobles et sentimentales, 30, 103, 


127, 200. 
Orchestral works 
Bolero, 235, 682. 


Concerto for Piano for the Left Hand, 


235 
Concerto for Piano in G major, 235. 
Menuet antique, 93. 
Pavane pour une Infante défunte, 93. 
Rapsodie espagnole, 101. 


Sarabande (Debussy, orchestrated 


Ravel), 101. 
La Valse, 235. 
Solo songs 
Histoires naturelles, 104. 
Shéhérazade, 101. 


Trois Poémes de Stéphane Mallarmé, 


30, 115, 466. 
Stage works 
Adelaide ou le langage des fleurs, 200. 


Daphnis et Chloé, 30, 78, 103, 105, 199- 


200. 
L’Enfant et les sortiléges, 174, 235-6. 


L'Heure espagnole. 172 (Ex. 81), 191, 


285; 


Rawsthorne, Alan, 523, 528, 537-9. 
Concerto for Piano No. 1, Ex. 237 (ii). 
Concerto for Violin No. 1, 537. 

The Creel, 537 (Ex. 237 (1)). 

Quintet for wind instruments and piano 
538. 

Sonata for Violin, 537. 

Street Corner Overture, 539. 

Symphonic Sketches, 537. 

Symphonic Studies, Ex. 237 (iii). 

Razin, Stepan, 650. 

Rebikov, Vladimir, 36, 133. 

Alpha i Omega, 207. 

Beliya pesni (Chansons blanches), 133. 
Bezdna (The Abyss), 136 (Ex. 59). 
Dances Op. 51, 133. 

Redlich, Hans F. 

Alban Berg—Versuch einer Würdigung, 
18 n+, 367 пі, 

Reger, Max, 9, 10, 11-15, 16, 18, 20, 
49, 36, 68, 78, 81-6, 108, 140, 144, 
19] 992} 19572295812731978257329: 
427. 

*Maienblüten' Op. 66 no. 5, 11. 

*Morgen' Op. 66 no. 10, 11 (Ex. 4). 

Phantasien Op. 52, 14. 

Quintet for Piano and Strings in C minor 
Op. 64, 15, 68. 

Schlichte Weisen, 11. 

Serenade in G major, 14, 229. 

Sinfonietta, 15. 

Sinfonischer Prolog zu einer Tragódie, 14, 
15: 

Sonata for Cello Op. 5, 14. 

Suite in E minor, 14. 

Trio Op. 2, 14. 

Trio for Strings Op. 77b, 15 n?. 

Variationen und Fuge über ein Thema von 
J. S. Bach, 83-5 (Ехх. 29, 30). 

Vier Tondichtungen nach Arnold Bócklin, 
14. 

"Waldseligkeit! Op. 62 по. 2, 12-14 
(Ex. 5 (ii)). 

other works, 85. 

Régnier, Henri de, 110. 

Reich, Willi, see under Anton Webern. 

Respighi, Ottorino, 268-9. 

Lucrezia, 268—9. 
other works, 268. 

Revueltas, Silvestre, 637. 
works, 637. 

Reznicek, Emil Nikolaus von, 3. 

Rheinberger, Josef, 3, 38, 189. 

Riegger, Wallingford, 584, 596, 605, 606, 
607-10, 611, 614, 622, 625, 628. 

Dichotomy, 607. 
Music for Brass Choir, 610 (Ex. 257). 
Study in Sonority, 607. 
other works, 610. 
Riemann, Hugo, 14. 
Rieti, Vittorio, 619. 


756 


Rilke, Rainer Maria, 329, 696. 

Die Weise von Liebe und Tod des Cornets 
Christoph Rilke, 321. 

Rimbaud, Arthur, 90, 94, 550, 556. 

Rimsky-Korsakov, Nikolay, 32, 33, 44, 68, 
92, 103, 115, 117, 137, 174—7, 196, 197, 
200, 211, 229, 268, 639, 644, 670. 

Kashchey bessmertny (Immortal Kash- 
chey), 33, 174—5, 177. 

Mlada, 116, 174. 

Mozart i Salieri, 174. 

Noch pered Rozhdestvom (Christmas 
Eve), 174. 

Sadko, 147 n!, 174. 

Servilia, 174. 

Skazanie o nevidimom grade Kitezhe i deve 
Fevronii (The Legend of the Invisible 
City of Kitezh and the Maiden 
Fevronia), 33, 175 (Ex. 82). 

Skazka o Tsare Saltane (The Tale of Tsar 
Saltan), 174—5. 

Snegurochka (Snowmaiden), 174, 177. 

Tsarskaya Nevesta (The Tsar’s Bride), 
174. 

Zolotoy Petushok (The Golden Cockerel), 
33, 44, 116-17 (Ex. 51), 137, 175, 176—7, 
197. 

Ripellino, Angelo Maria, 473. 

Rittikh, M., 647 n!. 

Rizhkin, Iosif, 643 пп1>4 665 пп! ?, 666 пі, 
667 nn^?, 672 n°, 677 n!, 680 n!, 
686 n!, 688 n!. 

Roberts, E. T., 17 п8. 

Rochberg, George, 625. 

Rodgers, Richard, 568. 

Rodgers, W. R., 563. 

Rodrigo, Joaquin, 318. 

Concierto de Aranjuez, 318. 
Roger-Ducasse, Jean Jules Aimable, 29. 
Rogers, Bernard, 605, 606, 626. 
Roland-Manuel, see Roland Manuel. 
Roldán, Amadeo, 637. 

Rolland, Romain, 2. 

Colas Breugnon, 667. 

Musiciens d'aujourd'hui, 2 nt. 

Roosevelt, Franklin, 619. 

Ropartz, Guy, 22. 

Rosenberg, Hilding, 427, 494. 

Rosenfeld, Paul, 584. 

Musical Chronicle 1917-1923, 584 n!. 
Roslavets, Nikolay, 137-40, 641, 657. 

Deux Compositions pour piano, 137. 
Rossini, Gioacchino, 193. 

Rossiyskaya Assotsiatsiya  proletarskikh 
muzikantov, 640, 641, 642, 655, 660, 
667. 

Rostropovich, Mstislav, 696. 

Roussel, Albert, 22, 25, 64, 110-13, 231, 
232-4, 309, 518. 

Chamber music 
Divertissement for piano and wind 


INDEX 


quintet, 25, 110 (Ex. 48 (i)). 
. .Quartet for Strings, 234. 
Sonata for Violin in D minor, 25, 63 
(Ex. 22), 110 (Ex. 48 (ii)). 
Trio for Strings, 234. 
Choral works 
Évocations, 25, 78, 110. 
Keyboard works 
Rustiques, 110. 
Suite for piano Op. 14, 112 (Ex. 49). 
Orchestral works 
Concerto for Cello, 234. 
Concerto for Small Orchestra, 233. 
Faunes et dryades, 25. 
Forét d'hiver, 25. 
Le Роёте de а forét, see Symphony 
No. 1. 
Pour une Féte de Printemps, 233. 
Renouveau, 25. 
Soir d'été, 25. 
Suite in F, 233. 
Symphony No. 1 (Le Poéme de la 
forét), 25, 110. 
Symphony No. 2, 233. 
Symphony No. 3, 233. 
Symphony No. 4, 233. 
Solo songs 
Deux poémes chinois, 112. 
Quatre Ровтез (Régnier), 110. 
Stage works 
Aeneas, 234. 
Bacchus et Ariane, 234. 
Le Festin de l'araignée, 25. 
La Naissance de la lyre, 233-4. 
Райтдуа!, 25, 233, 518. 
Le Testament de la Tante-Caroline, 234. 
Rózycki, Ludomir, 36. 
Rubbra, Edmund, 518, 523, 528, 531-6, 539. 
Symphony No. 1, 532, 533. 
Symphony No. 3, 533 (Ex. 236). 
Symphony No. 4, 531. 
Symphony No. 5, 532-3 (Ex. 235). 
other works, 535. 
Rubinstein, Artur, 218. 
Rubinstein, Ida, 235. 
Rufer, Josef, 411 n!. 
Die Komposition mit zwölf Tönen, 411 n?, 
415 n!, 418 n?, 420 n°, 427 пі, 494 п?, 
500 n!, 563. 
Ruggles, Carl, 584, 596—7, 610. 
works, 596. 
Ruwet, Nicolas, 458 n!. 


Sabaneyev, Leonid, 133, 137, 641. 
Vospominaya о Skryabine, 36 п\, 67 n!, 
Ей mS ПӘЛЕ 
Sabinina, M. D. 
*Semen Kotko' i problemi opernoy drama- 
turgii Prokofyeva, 670 п?. 
Simfonizm Shostakovicha: put k zrelosti, 
651 п?. 


INDEX 


Sadikov, Talib, see under Reinhold Glier. 
Šafránek, Miloš, Bohuslav Martini, 428 пі. 
Saint-Exupéry, Antoine de, 273. 

Saint Francis of Assisi, 220, 335. 

Saint John of the Cross, 229. 

Saint-Saéns, Camille, 19, 25, 93, 101, 108. 
Danse macabre, 237. 

Scherzo Op. 87, 93. 

Salazar, Adolfo, 318-19. 

Salieri, Antonio, Prima la musica e poi le 

parole, 32A. 

Salmanov, Vadim, 697-8, 699. 

Dvenadtsat (The Twelve), 697 (Ex. 292). 
Quartet for Strings No. 3, 698. 

Salzedo, Carlos, 583. 

Santa Cruz, Domingo, 635, 637. 

Santoro, Claudio, 638. 

Sappho, 660. 

Sarasate, Pablo, 30. 

Sardou, Victorien, 155, 156, 164. 

Satie, Erik, 92-3, 114, 167, 231, 236-9, 250, 

251, 254, 596, 602. 

Gnossiennes, 93. 

Gymnopédies, 92. 

Mercure, 237. 

Parade, 236-7, 239. 

Prélude de la Porte Héroique du Ciel, 93. 

Preludes for Le fils des étoiles, 93 (Ex. 
36 (1)), 128. 

Reláche, 237. 

Reláche: entracte | cinématographique, 
238. 

Sarabandes, 93. 

Socrate, 238 (Ex. 101). 

Sonneries de la Rose Croix, 93 (Ex. 36 
(ii)). 

Sauguet, Henri, 239. 

Savonarola, Girolamo, 273. 

Scarlatti, Domenico, 30, 235, 286, 316-17. 

Schaeffer, Pierre, 488. 

A la recherche d'une musique concréte, 
488 ппу 2, 

(with Pierre Henry) Symphonie pour ип 
homme seul, 488. 

Schaeffner, André 
Stravinsky, 117 n!. 

Scherchen, Hermann, 469. 

Schiffer, Marcellus, 332. 

Schilling, H. L., Paul Hindemith's 'Cardillac'. 
Beitrüge zu emem Vergleich der beiden 
Opernfassungen, 332 n1. 

Schillinger, Joseph, 601. 

Schillings, Max von, 3. 

Schlózer, Tatyana, 35. 

Schmidt, Franz, 9, 417. 

Schmitt, Florent, 29—30, 68, 110, 112, 199. 
Quintet for Piano and Strings, 29—30, 68. 
Psaume XLVI, 30. 

La Tragédie de Salomé, 105 (Ex. 44), 199. 

Schmitz, Robert, The Piano Works of 
Claude Debussy, 101 n?. 


757 


Schnebel, Dieter, 435. 

Schnitzler, Artur, Anatol, 152. 

Schoeck, Othmar, 319-20. 

works, 319-20. 

Schoenberg, Arnold, 9, 10, 15—18, 19, 21, 
39, 46-7, 48-9, 63, 65, 68, 72, 74-5, 78, 
85, 105, 115, 127-8, 133, 137-44 
passim, 200, 202, 203—6, 208, 209, 210, 
217350200 8518283201321 322, 926, 327, 
329, 336, 337, 338, 340-62, 363, 364, 
367, 368, 373, 375, 377, 380, 385-9 
passim, 394—5, 417, 419, 420, 423 n?, 
425-8 passim, 436, 437 n3, 438, 442, 
455, 469, 470, 479, 482 n!, 490, 494, 
497, 523, 556, 557, 559, 562, 515, 583, 
595, 601, 606, 607, 610, 611, 619, 620, 
625, 628, 638, 639. 

Literary works 
Der Biblische Weg, 359. 
Harmonielehre, 16, 17, 19, 21 n!, 105, 
115, 128, 140 п?, 142, 342 пі, 
Structural Functions of Harmony, 133 
її? 563. 
Style and Idea, 17 n>, 74 nn» ?, 140 n?, 
354 п!. 
Chamber music 
Quintet for Wind Op. 26, 318, 347-8, 
349-50. 
Quartet for Strings in D major, 9, 
141-2 (Ex. 61). 
Quartet for Strings No. 1, 15, 18, 21, 
68, 74-5. 
Quartet for Strings No. 2, 16, 18, 74-5. 
Quartet for Strings No. 3, 349-50, 355, 
356. 
Quartet for Strings No. 4, 355-6 (Ex. 
161). 
Trio for Strings Op. 45, 359, 395 п!, 
Suite Op. 29, 348, 349 (Ex. 156), 353. 
Verklarte Nacht, 2, 15, 16, 68. 
arrangement, 68. 
Choral works 
Four Pieces for chorus Op. 27, 348, 354. 
Gurrelieder, 15, 17, 75-8, 142. 
Die Jakobsleiter, 341 (Ex. 151), 358, 359. 
Kol Nidre, 356—7, 358. 
Six Unaccompanied Pieces Op. 35, 354. 
Three Satires Op. 28, 348, 354. 
Keyboard works 
Five Piano Pieces Op. 23, 144, 346, 
347 (Ех. 155). 
Piano Piece Op. 33a, 338, 353 (Exx. 
159 (i, ii)). 
Piano Piece Op. 33b, 353. 
Six Little Piano Pieces Op. 19, 17, 18, 
142, 341. 
Suite for piano Op. 25, 344 (Ex. 153), 
347. 
Three Piano Pieces Op. 11, 17, 68 (Ex. 
24 (ii)), 75, 142 (Ех. 62), 341, 455 
ici 


758 


Schoenberg—(cont.) 
Variations on a Recitative for organ, 
ЗАД 
Orchestral music 
Begleitungsmusik zu einer Lichtspiel- 
szene, 353-4. 
Chamber Symphony No. 1, 15-16, 18, 
48-9 (Ex. 20), Ex. 21 (iv), 63, 65, 74- 
75, 128-33 (Ех. 56), 357. 
revised editions, 48 пт, 68. 

Chamber Symphony No. 2, 357. 

Concerto for Piano, 357-8. 

Concerto for Violin, 354—5 (Ex. 160), 

357. 
Five Pieces for Orchestra Op. 16, 17, 18, 
127-8 (Ex. 53), 144, 341, 377, 438. 
Pelleas und Melisande, 15, 16, 117, 128 
(Ех. 54), 133, 203. 

Suite in С for strings, 354. 

Theme and Variations for 
Orchestra, 357. 

Variations for Orchestra Op. 31, 350-2 
(Exx. 157, 158), 380. 

Music with solo voice or speaker 

Das Buch der hängenden Gärten, 16, 
17 n*, 341. 
Herzgewáüchse, 16, 203. 
Ode for Napoleon, 357, 358. 
Orchestral Songs Op. 22, 16, 341. 
Serenade Op. 24, 144, 346-7 (Ex. 154), 
348. 

Pierrot Lunaire, 16, 17, 19, 115, 142-4 
(Ex. 63), 210, 341, 357, 420. 

Sechs Orchester-Lieder Op. 8, 46 (Ex. 
18), 72. 

A Survivor from Warsaw, 358-9. 

Stage works 

Erwartung, 16, 18, 204 (Ex. 96), 206, 
341. 

Die glückliche Hand, 10, 16, 17, 18, 36, 
206, 341. 

Moses und Aron, 354, 358, 359-62 
(Exx. 162, 163). 

Von Heute auf Morgen, 352-3. 
Schoenberg, Gertrud, see Max Blonda. 
Schopenhauer, Artur, 326. 

Schreker, Franz, 9, 184, 185-7, 310, 326, 

340. 

Der ferne Klang, 19, 185—7 (Ex. 87), 326, 

641. 

Die Gezeichneten, 187, 326. 
Der Schatzgrüber, 326. 
Schroeder, Hermann, 454. 
Schubert, Franz, 6, 340, 507, 519, 520, 521, 
55158552! 
Quintet in C major, 519. 
Die Winterreise, 340. 
Schuman, William, 573, 618, 620, 621, 622, 

626. 

Symphony No. 6, 622 (Ex. 260). 
other works, 622. 


Wind 


INDEX 


Schumann, Georg, 3. 
Schumann, Robert, 18, 19, 21, 25, 34, 39, 
80-1, 177, 326. 
Fugues Opp. 60 & 72, 80-1. 
Schweitzer, Albert, 81. 
J. S. Bach, le musicien-poéte, 81. 
Scontrino, Antonio, 31. 
Scott, Cyril, 506. 
Scott, Robert Falcon, 508. 
Scott-Gatty, Alfred, 184. 
Searle, Humphrey, 17 n5, 563. 
Twentieth Century Harmony, 563. 
works, 563. 
Seeger, Charles, 601. 
Segovia, Andrés, 269. 
Seiber, Mátyás, 424, 559-60, 563, 568. 
Ulysses, 560 (Ex. 245). 
other works, 559. 
Seidl, Arthur, 6. 
Séré, Octave, Cinquante ans de musique 
française, 25 п?. 

Serov, Aleksandr Nikolaevich, The Power 
of Evil, 177. 

Sessions, Roger, 584, 605-6, 607, 611-14, 
622, 626, 627, 628, 630. 

Quintet for Strings, 611 (Ex. 258). 

other works, 611. 

Seurat, Georges, 199. 

Séverac, Déodat de, 101. 
works, 101. 

Sgambati, Giovanni, 31, 256. 

Shakespeare, William, 264, 269, 507, 520, 

539, 555, 684. 
Shaporin, Yury, 641, 671, 676, 684—5, 688- 
691, 699. 

Dalekaya yunost (Far-off Youth), 676. 

Dekabristi (The Decembrists), 643, 688- 
691 (Ex. 287), 695. 

Elegii Op. 18, 676. 

Na pole Kulikovom (On the field of 
Kulikovo), 671, 685. 

Pushkin songs Op. 10, 676. 

Skazanie o bitve za Russkuyu zemlyu 
(Story of the fight for the Russian 
land), 684, 685. 

Symphony, 671. 

Shchedrin, Rodion, Konek-Gorbunok (The 

little humpbacked horse), 697. 
Shebalin, Vissarion, 641, 642, 651, 657, 683, 
684, 686, 688. 

Heine songs, 684. 

Quartet for Strings Op. 2, 657. 

Quartet for Strings No. 5, 683. 

Quartet for Strings No. 7, 683, 696, 697. 

Quartet for Strings No. 8, 696. 

Quartet for Strings No. 9, 696. 

Symphonies, 651. 

Symphony No. 4, 672. 

Symphony No. 5, 696. 

Trio for Piano and Strings, 683. 

Trio for Strings Op. 4, 657. 


INDEX 


Ukroshchenie stroptivoy (The Taming of 
the Shrew), 692. 
Shekhter, Boris, 642, 672. 
Turkmenia suite, 672. 
Shelley, P. B., 500. 
‘Shema Ysrael’, 358. 
Shepherd, Arthur, 606. 
Sheridan, R. B., 318, 559. 
The Duenna, 559, 680. 
Shishov, Ivan, 644, 663. 
Tupeyny khudozhnik (The Toupee Artist), 
644 (Ex. 264), 645, 663. 
Shlifshteyn, S. 
(ed.), N. Y. Myaskovsky: stati, pisma, 
vospominaniya, 648 nn?:?, 650 nn’ ?, 
S. S. Prokofyev: materiali, dokumenti, 
vospominaniya, 680 nt. 
Sholokhov, Mikhail, 665, 666. 
Shostakovich, Dmitry, 568, 646, 651—6, 657, 
663—5, 672-4, 682—3, 684, 686, 691, 693, 
694—6, 698, 699, 700. 
Chamber music 
Quartet for Strings No. 1, 674, 683. 
Quartet for Strings No. 2, 683. 
Quartet for Strings No. 3, 683, 684. 
Quartet for Strings No. 4, 693, 696. 
Quartet for Strings No. 5, 696. 
Quartet for Strings No. 8, 696. 
Quintet for Piano and Strings, 674, 683. 
Sonata for Cello, 672, 683. 
Тпо for Piano and Strings Op. 8, 651. 
Тпо for Piano and Strings Op. 67, 683, 
684, 696. 
Choral work 
Pesn o lesakh (Song of the Forests), 
693. 
Keyboard works 
Aforizmi, 657. 
Fantastic Dances, 657. 
Sonata for Piano No. 1, 657-70 (Ex. 
272 (ii)). 
Twenty-four Preludes, 674. 
Twenty-Four Preludes and Fugues, 693. 
Orchestral works 
Concertino for two pianos, 693. 
Concerto for Cello, 696. 
Concerto for Piano No. 1, 672. 
Concerto for Piano No. 2, 693 (Ex. 
290). 
Concerto for Violin, 693, 696. 
Symphony No. 1, 651-2, 655, 696. 
Symphony No. 2, 652-5 (Exx. 268, 
269), 673. 
Symphony No. 3, 652, 655-6 (Exx. 
270, 271), 663, 673, 682. 
Symphony No. 4, 672-4 (Ex. 281). 
Symphony No. 5, 674. 
Symphony No. 6, 674. 
Symphony No. 7, 682. 
Symphony No. 8, 682-3. 
Symphony No. 9, 682, 683. 


759 


Symphony No. 10, 694—5, 696. 
Symphony No. 11, 695, 696. 
Symphony No. 12, 695, 696. 
Symphony No. 13, 696. 
Symphony No. 14, 696. 
Solo songs 
From Hebrew Folk Poetry, 696. 
Krilov fables, 660. 
Six bass songs Op. 62, 684. 
Stage works 
Lady Macbeth Mtsenskovo uezda (Lady 
Macbeth of the Mtsensk District: 
Katerina Izmaylova), 663-5 (Exx. 
275, 276), 667, 674, 685, 686, 696. 
Moskva, Cheremushki, 691. 
Nos (The Nose), 646 (Ex. 265), 652. 
Svetly ruchey (Clear Stream), 665, 686. 
Shuysky, Vasily, 661. 
Sibelius, Jean, 37-8, 67, 74-5, 503, 519, 595. 
En Saga, 38. 
Kullervo, 37, 78. 
Symphony No. 1, 37. 
Symphony No. 3, 74-5. 
Symphony No. 4, 38, 74-5. 
Symphony No. 5, 75. 
Symphony No. 7, 38. 
Tapiola, 38. 
Voces intimae, 74-5. 

Siegmeister, Elie, 626. 

Silcher, Friedrich, 348. 

Silvestrov, Valentin, 699, 

Simpson, Robert, 563. 

Carl Nielsen, Symphonist, 38 п“. 

Sinding, Christian, 37. 

Sinigaglia, Leone, 31, 256. 

Sink, Kuldar, 699. 

Siqueira, José, 636. 

Sirola Božidar, 304 n!, 

Sitwell, Edith, 522, 526, 551. 

Sitwell, Osbert, 528. 

Sitwell, Sacheverell, 522, 527. 

‘Les Six’, 210, 231, 234, 236-54, 255, 256, 
318, 319, 428, 524, 526, 596, 602. 

Skalkottas, Nikos, 387, 425-6. 

Fourth Suite for Piano, 425 (Ex. 194). 

Skrebkov, S., 671 пп“»5. 

Skryabin, Aleksandr, 33, 34-7, 63-5, 67, 68, 
72-3, 78, 102-3, 127, 133-7, 140, 141, 
199, 202, 229, 255 пт, 342, 639, 641, 
648, 650, 657. 

Poem ekstaza (Poem of Ecstasy), 35 п?. 

Le divin poéme, see Symphony No. 3. 

‘Etrangeté’ Op. 63 no. 2, 137. 

Étude Op. 65 no. 1, 136. 

Poema ekstaza (Poem of Ecstasy), 35 (Ex. 
21 (v)), 63, 64, 65, 67, 73. 

Preludes Opp. 11, 13, 15, 16-17, 34. 

Prelude Op. 37 no. 3, 136 (Ex. 58). 

Preludes Op. 74, 137. 

Prometheus, 35-6, 67, 78, 133, 136, 137 
(Ex. 60). 


760 


Skryabin—(cont.) 

Satanic Poem Op. 36, 68 (Ex. 24 (1)). 

Sonatas for Piano 1-3, 34. 

Sonata for Piano No. 4, 35. 

Sonata for Piano No. 5, 35. 

Sonata for Piano No. 6, 36. 

Sonata for Piano No. 7, 36, 137. 

Sonatas for Piano 8-10, 36. 

Symphony No. 1, 34-5 (Ex. 12), 78. 

Symphony No. 2, 35. 

Symphony No. 3 (Le divin poéme), 35, 
229. 

Works for Piano Opp. 71-4, 36. 

Slonimsky, Nicolas, 580. 

Music since 1900, 642 n?, 665 n?, 685 n°. 

Slonimsky, Sergey, 699. 

Dialogi, 699. 
Smetana, Bedfich, 177, 301, 309. 

From My Life, 309. 
Smith-Brindle, Reginald, 473 n!, 479 n!. 
Smyth, Ethel, 187. 

works, 187. 

Solovyev, Vladimir, 175. 

Song of Solomon, 507. 

Sophocles, 220, 233, 250, 410. 

Souvtchinsky, Pierre (ed.), Musique russe, 
92 n*. 

Sowerby, Leo, 605. 

Spadavecchia, Antonio, 687. 

works, 687. 

Spontini, Gasparo, 267. 

Stalin, Joseph, 665, 686, 694. 

Stanford, Charles Villiers, 38, 39, 40, 187, 
506, 516, 521. 

Shamus O'Brien, 187. 

Stanislavsky, Konstantin, 661. 

Stasov, Vladimir Vasilievich, 640. 

Stefan, Paul, Gustav Mahler, 6 n°. 

Stein, Fritz, Max Reger, 82 n?. 

Stein, Gertrude, 522, 596, 602. 

Steinbeck, John, Of Mice and Men, 592. 

Steinberg, Maximilian, 639, 651. 

Symphony No. 4, 672. 

Steinecke, Wolfgang, 435. 

Steiner, Rudolf, 175. 

Stenhammar, Wilhelm, 37. 

Stevens, Halsey, The Life and Music of Béla 
Bartók, 291 n!. 

Stockhausen, Karlheinz, 373, 401, 410, 
427 n?, 434—5, 437, 454-65, 467, 468, 
470, 473, 477, 479, 480, 482, 484, 485, 
489 n!, 490—3 passim, 497, 500, 567, 
625, 699. 

Carré, 464, 465 n!. 

Gesang der Jünglinge, 454, 464, 488-9, 
490-1, 492. 

Gruppen, 401, 459, 460—4 (Exx. 206, 207), 
479, 484. 

Klavierstücke I-IV, 455-9 (Ex. 205). 

Klavierstück XI, 464—5. 

Kontakte, 493. 


INDEX 


Kontra-Punkte, 454—5 (Ex. 204). 
. Kreuzspiel, 454, 699. 
Refrain, 465. 
Studie I, 490. 
Studie II, 490. 
Zeitmasse, 401, 459-60, 464, 467, 485. 
Zyklus, 465, 480. 

Stoin, Vasil, Narodni pesni ot Timok do Vit, 
282 n? (Ex. 128). 

Stokowski, Leopold, 582. 

Storb, Ilse, Untersuchungen zur Auflösung 
der funktionaler Harmonik іп den 
Klavierwerken von Claude Debussy, 
90 n?. 

Storm, Theodor, 368. 

Stósslová, Kamila, 309. 

Stramm, August, 33]. 

Strauss, Johann, 223, 323. 

Strauss, Richard, 2-8 passim, 10-16 passim, 
19, 24, 30, 36, 39, 43-7 passim, 49, 63, 
64-5, 67-74 passim, 78, 81, 86, 105, 
117, 133, 140, 144, 146-53 passim, 157- 
158, 164, 169, 176, 183-4, 185, 190, 191, 
193, 195, 196, 202, 204, 229, 267, 268, 
302, 312, 321-6 passim, 332, 363, 503, 
505-6, 507, 540, 611. 

Chamber music 
Duet Concertino, 325. 
Sonatina No. 1, 325. 
Sonatina No. 2, 325. 
Orchestral works 
Alpensymphonie, 6. 
Also sprach Zarathustra, 3, 4, 5, 7, 19, 
81, 149. 
Aus Italien, 3, 117. 
Concerto for Horn No. 2, 325. 
Concerto for Oboe, 325. 
Don Juan, 3. 
Don Quixote, 3, 4, 5, 8, 67-8. 
Ein Heldenleben, 3, 4—5 (Exx. 2, 3), 8, 
14, 15, 19, 44, 45-6, 67. 
Macbeth, 3. 
Metamorphosen, 325. 
Symphonia domestica, 4, 5, 73, 74. 
Till Eulenspiegels lustige Streiche, 3, 5. 
Tod und Verklärung, 3. 
Solo songs 
*Blindenklage' Op. 56 no. 2, 10. 
“СасШе”, 10, 72. 
*Heimliche Aufforderung’, 10. 
*Morgen', 10, 12, 72. 
‘Traum durch die Dämmerung’, 10. 
Vier Gesdnge Op. 33, 72. 
Vier letzte Lieder, 72, 325-6. 
*Waldseligkeit', 12-13 (Ex. 5 (1)). 
Stage works 
Arabella, 196, 323-4. 
Ariadne auf Naxos, 144, 153, 176, 322, 
323, 324. 
Die aegyptische Helena, 196, 323. 
Capriccio, 324, 325. 


INDEX 761 


Daphne, 324, 325. 

Elektra, 4 (Ех. 1), 17, 64, 148, 149-52 
(Exx. 66, 67), 153, 164, 193, 204, 322, 
505. 

Feuersnot, 6, 146, 148. 

Die Frau ohne Schatten, 196, 322. 

Der Friedenstag, 324. 

Guntram, 2, 6, 46. 

Intermezzo, 190, 322, 324. 

Die Liebe der Danae, 196, 324. 

Der Rosenkavalier, 44 (Ex. 16), 47, 
Ex. 21 (vii), 63-4, 78, 147 nt, 152-3, 
191, 196, 322, 323. 

Salome, 17, 73, 128 (Ex. 55), 148-50 
(Ex. 65), 153, 158, 187, 202, 204, 322. 

Die schweigsame Frau, 324. 

Stravinsky, Igor, 33, 34, 36, 115, 117, 127 n!, 
137, 144, 177, 193, 196—202 passim, 
207, 210-29, 230-4 passim, 238, 243, 
250, 251, 254, 266, 267, 268, 273, 274, 
276, 299, 306, 309, 310, 315, 319, 321, 
330, 335, 380, 387-8, 389-402, 409, 
410, 411, 413, 416, 419, 420, 424, 430, 
435, 444, 479, 495-8 passim, 500, 512, 
513, 516, 523, 524, 530, 536, 547, 548, 
559, 563, 567, 574, 575, 583, 588, 595, 
606, 614, 619, 620, 625, 626, 627, 630, 
639, 641, 645, 646, 657. 

Literary works 
Chroniques de ma vie, 211 n!, 214, 

218 n!, 220 n!, 222. 
Conversations, 117 n?, 401 n?. 
Memories and Commentaries, 389 n!, 
391 п!, 395 n?, 400 nt, 401 п’. 
Poétique musicale, 34 n!, 223, 620. 

Chamber music 

Duo Concertante, 225-6. 

Octet, 217-18. 

Ragtime, 213. 

Septet, 394, 396. 

Three Pieces for String Quartet, 210. 

Choral works 
Cantata, 394, 396 nt. 

Canticum Sacrum, 397—8, 413. 

Mass, 393-4. 

Symphony of Psalms, 217, 223-5 (Ex. 
99), 228, 513. 

Threni, 398-401 (Ехх. 185, 186), 513. 

Keyboard works 
Concerto for two pianos, 226-7 (Ex. 

100), 380. 
Five Easy Pieces, 218. 
The Five Fingers, 218. 
Piano Rag-Music, 213, 218. 
Serenade in A, 219. 
Sonata for Piano, 219. 
Sonata for Two Pianos, 389, 393. 

Orchestral works 
Agon, 397, 398. 

Capriccio for piano and orchestra, 219, 
225. 


Concerto for piano and wind instru- 
ments, 219. 
Concerto in D for string orchestra, 389. 
Concerto for Violin, 225, 226. 
Dumbarton Oaks Concerto, 228. 
Ebony Concerto, 389. 
Fireworks, 115. 
Movements, 395 n?, 396 n!, 401-2 (Ex. 
187). 
Scherzo fantastique, 115,117 (Ex. 52 (ii) ). 
Symphonies of Wind Instruments, 217, 
218. 
Symphony in C, 389-90 (Ex. 182). 
Symphony in Three Movements, 389, 
391. 
Solo songs 
Faun and Shepherdess, 220. 
In Memoriam Dylan Thomas, 397. 
Pastoral, 220. 
Pribaoutki, 210. 
Shakespeare Songs, 396 (Exx. 184 (i, 11)). 
Three Japanese Lyrics, 115, 210. 
Stage works 
Apollo Musagetes, 220, 221-2, 398, 
513, 524. 
Le Baiser de la Fée, 222-3. 
L'Histoire du soldat, 207, 211, 212-15 
(Ex. 97), 220, 223, 225, 243, 645. 
Jeu de Cartes, 223. 
Мауга, 218, 222-3, 641, 646. 
Les Noces, 211, 215-16 (Ex. 98), 218, 
224, 273, 306, 410. 
Oedipus Rex, 213, 220-1, 222, 228, 243, 
250, 393, 513, 516, 641. 
Orpheus, 220, 393. 
Perséphone, 217, 220, 222, 228. 
Petrushka, 33, 78, 127, 197, 198—9, 201, 
2, 218, 219, 220, 512: 
Three Movements, 218. 
Pulcinella, 217-18, 223. 
The Rake's Progress, 193, 389, 391-3 
(Ex. 183), 394, 559. 
Renard, 207, 211-12, 213. 
Solovey (The Nightingale), 115, 197 
(Ex. 92), 212. 
Vesna svyashchennaya (The Rite of 
Spring), 33, 127, 133 (Ex. 57), 144, 
197, 199, 200, 210, 217, 219, 220, 230, 
395953? 
Zhar-ptitsa (Yhe Firebird), 33, 36, 117, 
137, 177, 197 (Ex. 93), 199, 513. 
Strindberg, August, 630. 
Strobel, Heinrich, Paul Hindemith, 405 п!. 
Struve, Gleb, 25 Years of Soviet Russian 
Literature, 667 n?. 
Stuckenschmidt, Hans Heinz, 17. 
Arnold Schönberg, 17 n?, 204 пі, 
Suckling, Norman, Fauré, 231 n?. 
Suk, Josef, 9. 
Sullivan, Arthur, 38. 
Sutermeister, Heinrich, 494. 


762 


Sviridov, Yury, 683-4. 
Pamyati Sergeya Esenina (In Memory of 
Sergey Esenin), 697. 
Pateticheskaya oratoriya (Pathetic ora- 
torio), 697. 
other works, 684. 
Szymanowski, Karol, 36-7, 40, 275, 311-15. 
Harnasie, 312 (Ex. 140). 
Król Roger (King Roger), 37, 314 (Ex. 
141). 
Quartets for Strings, 315. 
‘Smutna wiosna’ Op. 24 no. 5, 40 (Exx. 
15915): 
Stabat Mater, 312-14. 
Symphony concertante, 314-15. 
other works, 36. 


Tailleferre, Germaine, 239, 240. 

Tallis, Thomas, 511, 519. 

Taneyev, Sergey, 33, 639, 681, 699. 

‘And theenemy faltered’ Op. 26 no. 8, 681. 
Taranov, Gleb, 700. 
Taranushchenko, V., see 

Iordansky. 

Tate, Phyllis, 557. 

Taylor, Deems, 595. 

works, 595. 

Tchaikovsky, P. I., 1, 9, 32, 33, 34, 37, 174, 
196, 218, 222, 223, 647, 667, 683, 691, 
693. 

Iolanta, 174. 

Pikovaya Dama (The Queen of Spades), 

174. 

The Nutcracker, 196. 

Slavonic March, 683. 

The Sleeping Beauty, 196. 

Swan Lake, 196. 

Symphony No. 6 (Pathétique), 78. 
Tennyson, Alfred, 551. 

Theocritus, 522. 

Theremin, Leon, 601. 

Thomas, Ambroise, Mignon, 237. 

Thomas, Dylan, 397. 

Thompson, Randall, 584, 605, 606. 

Thomson, Virgil, 584, 596, 602-5, 606, 615, 
622, 625. 

Capital, Capitals, 596. 

Four Saints in Three Acts, 602 (Ex. 256). 

The Mother of Us АП, 602. 

Sonata da Chiesa, 602. 

Symphony on a Hymn Tune, 602. 

other works, 605. 

Thoreau, Henry David, 575. 

Tippett, Michael, 540-3. 

Quartet for Strings No. 2, 540 (Ex. 238). 

other works, 542-3. 

Tishchenko, Boris, 699. 

Toch, Ernst, 606, 619. 

Todi, Jacopone da, 264. 

Tolstoy, Aleksey Konstantinovich, 643. 

Tolstoy, Aleksey Nikolaevich, 688. 


under М. 


INDEX 


Tolstoy, Lev, 175, 640. 
- The Kreutzer Sonata, 308—9. 
War and Peace, 677. 
Torrefranca, Fausto, 256. 
Giacomo Puccini e l'opera internazionale, 
256 n!. 
Toscanini, Arturo, 620. 
Ттаһегпе, Thamas, 539. 
Trambitsky, Viktor, 676. 
Groza (The Storm), 676. 
Trend, J. B., Manuel de Falla and Spanish 
Music, 275 пі, 
Triodin, Peter, 643, 650. 
works, 643. 
Tsintsadze, Sulkhan, 700. 
Tumanina, Nadezhda, 672 n5, 684 пі, 
Turina, Joaquín, 31, 317-18, 319. 
Cuentos de Езрайа, 318. 
other works, 317. 
Turner, Joseph Mallord William, 547. 
Tyrwhitt, Gerald, see Lord Berners. 
Tyutchev, Fédor Ivanovich, 684. 


Uhland, Johann Ludwig, 319. 

Ungaretti, Giuseppe, 473. 

Uspensky, Viktor (with Viktor Belyaev), 
Turkmenskaya muzika, 699 n?. 

Ussachevsky, Vladimir, 627. 

Ustvolskaya, Galina, 698, 699. 


Valen, Fartein, 321, 387, 427. 
Piano Variations Op. 23, 321. 
Vallas, Léon 
Les Idées de Claude Debussy, 91 пт, 144 п. 
Vincent d'Indy, 20 п“, 108 п?, 
Van Dieren, Bernard, 520. 
Varése, Edgard, 255, 389, 489, 492-3, 574, 
583, 596, 597, 605, 610, 627. 
Déserts, 492-3. 
Hyperprism, 596, 597 (Ex. 254). 
Ionization, 597. 
Octandre, 596. 
Poéme électronique, 493, 597. 
Vasilenko, Sergey, 645, 647. 
Buran, 645. 
losif prekrasny, 647-8. 
Red Army Rhapsody, 672. 
Sin solntsa (Son of the Sun), 645. 
Suvorov, 676. 
Vaughan Williams, Ralph, 39-40, 114, 117, 
503, 504, 507-13, 517, 531. 
Benedicite, 511. 
Fantasia on a Theme by Tallis, 504, 511. 
Flos Campi, 507, 512. 
Job: A Masque for Dancing, 512. 
The Lark Ascending, 512. 
London Symphony, 504, 511. 
Mass, 531. 
On Wenlock Edge, 504, 511. 
Pastoral Symphony, 504, 509 (Ex. 223). 
Riders to the Sea, 507, 508. 


INDEX 


Sea Symphony, 40, 78, 504, 511. 

Sinfonia antartica, 512, 513. 

Symphony No. 4, 510 (Ex. 224), 513. 

Symphony No. 5, 510 (Ex. 225). 

Symphony No. 6, 508. 

Symphony No. 8, 512. 

Toward the Unknown Region, 504. 

The Wasps, 507. 

Veracini, Francesco Maria, 256. 
Verdi, Giuseppe, 32, 155, 157, 159, 191, 
221, 258, 420, 421, 469, 501, 516, 
554. 

Aida, 159. 

Falstaff, 32, 78, 191, 258. 

Otello, 169, 554. 

La Traviata, 154, 157, 187. 

Verga, Giovanni, 154. 
Verhaeren, Emile, Les Aubes, 650. 
Verikovsky, Mikhail, 700. 
Verlaine, Paul, 20, 26, 94, 193. 
Jadis et naguère, 20 n°. 
‘Vexilla regis’, 514. 
*Vi zhertvoyu pali' (You fell as a sacrifice), 
695. 
Vidakovié, Albe, 304 n!. 
Villa-Lobos, Heitor, 635-6. 
works, 635-6. 
Villon, Frangois, 100. 
Virgil, Georgics, 243. 
Virta, Nikolay, Odinochestvo, 667. 
Vivaldi, Antonio, 219, 256, 262 n!. 
Vlad, Roman, 225, 420 n?, 421 п°, 424, 
427 п?, 495-6. 

Dallapiccola, 420 п?. 

Modernità e tradizione nella musica con- 
temporanea, 419 n!, 420 nn**, 424 
nn? 5, 7. 

Stravinsky, 115 n*, 225 n!, 397 n?, 400 пі, 

Volkonsky, Andrey, 699, 695 n!. 

works, 699. 

*Vsveytes, sokoli, orlami’ (Soar, hawks, 
like eagles), 643. 


Wagner, Richard, 1-2, 3, 8, 9, 10, 14, 15, 17- 
18920521520 429831832233 435330439; 
39, 43, 44, 49, 63, 69—70, 81, 86-7, 91, 
102-3, 140, 145-8, 152, 153, 154, 157, 
159, 163-7 passim, 169—72 passim, 
175, 177, 182, 183-4, 185, 187, 188, 191, 
193, 195, 196, 200, 201, 202, 209, 240, 
259, 285, 300, 324, 326, 332, 334, 341, 
363, 370, 423 n°, 505, 506, 527, 531, 
574, 595, 680. 

Der fliegende Hollünder, 169—70. 

Gütterdüámmerung, see under Der Ring des 
Nibelungen. 

‘Im Treibhaus’, 86-7 (Ex. 31). 

Lohengrin, 145, 146. 

Die Melstersinger, 81, 128, 145, 146, 147, 
152, 191, 195, 326, 334. 


763 


Parsifal, 1, 2, 17, 43, 44, 146, 167, 169, 
175, 183, 185, 187, 200. 
Der Ring des Nibelungen, 2, 32, 145, 183, 
324. 
Gótterdümmerung, 1, 81. 
Siegfried, 175. 
Die Walküre, 145, 165, 324. 
Tristan und Isolde, 1, 2, 43, 145, 146, 169, 
184, 188, 369, 505. 
Wagner, Siegfried, 148. 
Der Barenhduter, 148. 
Walter, Bruno, 557. 
Walton, Isaak, 537. 
Walton, William, 523-31, 537, 538. 
Belshazzar's Feast, 528. 
Concerto for Cello, 530. 
Concerto for Viola, 524, 530, 531. 
Concerto for Violin, 531. 
Еасайе, 526. 
Portsmouth Point, 527, 531. 
Scapino, 531. 
Sinfonia concertante, 524, 526. 
Symphony No. 1, 528-30 (Exx. 231-4). 
Symphony No. 2, 530, 531. 
Troilus and Cressida, 530—1. 


Мага, Robert, 626. 


works, 626. 
Warlock, Peter (Philip Heseltine), 519—20, 
521 
works, 520. 
Watteau, Jean Antoine, 193. 
Weber, Ben, 625. 
Weber, Carl Maria von, 32, 147, 219, 223. 
Webern, Anton, 10, 18, 19, 63-5 passim, 67, 
68, 72-3, 75, 117, 140, 210, 215, 216, 
340-1, 346, 362-3, 372-86, 387-8, 393, 
396, 397, 398, 409, 411, 423, 431, 436-7, 
438, 442, 444, 453, 454, 455 n!, 460, 
461, 469, 482, 483, 489, 496, 497, 502, 
523, 556, 559, 563, 566, 567, 595, 601, 
625, 627, 638. 
(ed. Josef Polnauer), Briefe an Hildegard 
Jone und Josef Humplik, 380 n+. 
(ed. Willi Reich), Der Weg zur neuen 
Musik, 380 n?, 387 n!, 397 пі, 
Chamber music 
Fünf Sätze for string quartet Op. 5, 18. 
orchestral version, 68. 
Quartet Op. 22, 378—9, 483. 
Quartet for Strings Op. 28, 381-2 (Ex. 
179), 384. 
Trio for Strings Op. 20, 375. 
Choral works 
Das Augenlicht, 382-4 (Ех. 180). 
Cantata No. 1, 379, 382, 384-5 (Ex. 
181). 
Cantata No. 2, 376, 379, 382, 385, 
397 пт, 
Fünf geistliche Lieder Op. 15, 376. 
Zwei Lieder Op. 19, 373 (Ex. 173 
(ii)). 


764 


Webern—(cont.) 
Keyboard works 
Variations for piano, 380-1 (Ex. 178), 
382. 
Orchestral works 
Concerto Op. 24, 379-80 (Ex. 177). 
Fiinf Stiicke for orchestra Op. 10, 18. 
Im Sommerwind, 18. 
Passacaglia Op. 1, 18 (Ex. 21 (vi)), 
63-4, 67. 
Sechs Stücke for orchestra Op. 6, 18, 
ӨЛШ (Ех Эа 
Vier Lieder Op. 13, 18. 
Symphony Op. 21, 375-8 (Exx. 174, 
175), 379-82 passim. 
Variations for orchestra, 382, 419. 
Solo songs 
Drei geistliche Volkslieder Op. 17, 373. 
Drei Gesünge Op. 23, 379 (Ex. 176), 
382. 
Drei Lieder Op. 18, 373 (Ex. 173 (1)). 
Drei Lieder Op. 25, 379, 382. 
Wedekind, Frank, 326, 364, 365, 630. 
Die Büchse der Pandora, 364. 
Der Erdgeist, 364. 
Weelkes, Thomas, 518, 519. 
Weill, Kurt, 330, 338-9, 619. 
Die Dreigroschenoper, 339, 641. 
other works, 339. 
Weisgall, Hugo, 573, 614, 625, 628, 630. 
works, 630. 
Weissman, John S., 398 n!. 
Goffredo Petrassi, 424 n?. 
Wellesz, Egon, 65, 557. 
Arnold Schönberg, 15 n?, 127 n?. 
Die neue Instrumentation, 65 n!. 
Quartet for Strings No. 8, Ex. 244. 
Werth, Alexander, Musical Uproar іп 
Moscow, 685 пп?» 3, 
White, Eric Walter, Stravinsky, the Com- 
poser and his Works, 397 n?. 
Whiteman, Paul, 585. 
Whitman, Walt, 39, 404, 508, 522, 573. 
Whythorne, Thomas, 519. 
Widor, Charles Marie, 81. 
Wilbye, John, 519. 
Wilde, Oscar, 148. 
De Profundis, 148 n?. 
Salome, 148. 
Wilder, Thornton, Our Town, 592. 
Williams, Alberto, 637. 


INDEX 


Williamson, Malcolm, 567, 568. 
-Works, 568. 
Wittgenstein, Ludwig, 557. 
Wittgenstein, Paul, 235. 
Wolf, Hugo, 2-3, 10-11, 26, 47, 68, 147, 319, 
539. 
Der Corregidor, 147. 
Mórike-Lieder, 68. 
“Schon streckt’ ich aus’, 47 (Ех. 19), 48. 
Wolf-Ferrari, Ermanno, 189-91, 192. 
Die vier Grobiane, 189 (Ex. 89), 191. 
other works, 189-91. 
Wolff, Christian, 437 п?. 
Wolpe, Stefan, 606, 619. 
Wood, Hugh, 568. 
Wörner, Karl H., 465 nt. 
Gottesdienst und Magie, 362 n!. 
Neue Musik in der Entscheidung, 17 n?. 
“Wynter wakeneth al my care’, 406. 


*Yablochko' (The little apple), 643, 648, 671. 
Yeats, W. B., 520, 630. 

Yudenich, Nikolay Nikolaevich, 643. 
Yurasovsky, A. L, Trilby, 643. 


‘Zamuchen tyazheloy nevoley’ (Worn out 
by heavy slavery), 696. 
Zandonai, Riccardo, 188. 
Francesca da Rimini, 188. 
Zemlinsky, Alexander von, 9, 369. 
Lyric Symphony, 369. 
Zhdanov, Andrey, 685-6. 
Zhelobinsky, Valery, 661-3. 
Imenini (The Name-day), 663. 
Kamarinsky muzhik, 661-3 (Exx. 273, 
274). 
Mat (The Mother), 667. 
Zillig, Winfried, 409. 
Variationen über neue Musik, 409 nn? 3, 
436 n!, 494 п1, 
Zimmermann, Bernd Alois, 496. 
Symphony, 496. 
Zola, Emile, 148, 153, 154, 164, 166. 
Soirées de Médan, 164. 
Zolotarev, Vasily, 643, 644. 
Dekabristi (The Decembrists), 643, 644. 
Zoshchenko, Mikhail, 685. 
Zuckerman, Eliot, Тйе first 100 years of 
Wagner’s ‘Tristan’, 148 n+, 
Zweig, Stefan, 324.