THE OXFORD HISTORY OF MUSIC

by W, •M.

VOL. VI

THE ROMANTIC PERIOD

BY

EDWARD DANNREUTHER

OXFORD AT THE CLARENDON PRESS

1905

HENRY FEOWDE, M.A.

PUBLISHER TO THE UNIVERSITY OF OXFORD

LONDON, EDINBURGH NBW YORK AND TORONTO

\c\o\ V . 6

.

NOTE

IT is not fitting to speak here of the heavy loss which English Music has sustained in the death of Mr. Edward Dannreuther. To his long career of unselfish and devoted labour there have been paid elsewhere tributes which no words of mine could en- hance. But, from respect to his memory, I would ask leave to offer a brief explanation of the circumstances under which this his last work is presented to the public.

The manuscript was finished and partly revised by the Autumn of 1904. All that remained was to complete the re- vision and to make a selection of the musical examples. During the winter Mr. Dannreuther was prevented by illness from con- tinuing the work ; and at his request, and under his instructions, I carried it on to the best of my ability. The volume as it stands embodies the results of his research and the verdicts of his critical judgement : but it did not receive the final touch of his hand.

There is one more point to which the attention of the reader may be directed. When the Oxford History was first planned it appeared advisable to end with Schumann, and to leave to some future historian the more controversial topics of our own time. This view it has been found necessary to modify, and the present volume contains reference to the principal works, of whatever date, which in origin or character can be directly attributed to the Romantic movement.

My cordial thanks are due to Mr. J. A. Fuller Maitland for his assistance in the correction of the proofs.

W. H. HADOW.

PREFACE

Ax attempt is here made to show, with the aid of copious examples, analyses, and comments, how the course of Music has gradually changed since Beethoven's day. Not to disturb the impression of historical sequence each department is treated chronologically, but as the different classes of music are discussed separately the dates necessarily overlap. Musical quotations are in some instances rather long, because the reader, to be in a position to judge of the calibre and style of a passage, ought to have at least one complete sentence or period before him. They are compressed in so far as compression is consistent with perspicuity, and the original is always faithfully reproduced. In connexion with the description of the works that stand for the various phases of the Romantic movement, certain questions arising out of the attitude towards artistic problems taken by leading masters are discussed as they come into view ; and it is the ever-varying aspect of such questions that forms both the link and the contrast between chapter and chapter, sometimes even between one paragraph and another.

A book so closely in touch with the actualities of present-day musical life must needs contain some controversial matter. It ought to be trustworthy as to facts, but it can hardly avoid the expression of disputable criticisms. These must be understood to represent merely the personal opinions of the writer.

EDWARD DANNREUTHER.

CONTENTS

PAGE

CHAPTER I. INTRODUCTORY SKETCH. Romanticism in literature and music. German, French, Italian Romantic opera. Overtures and Symphonies. Pro- gramme Music. Oratorios and Cantatas. Piano- forte pieces and Solo Songs. Virtuosi. Musicians as writers on music . . . . . i

CHAPTER II. GERMAN ROMANTIC OPERA FROM WEBER TO SCHUMANN. Weber's Preciosa, Per Freischutz, Euryanthe, Oberon Spohr E. T. A. Hoffmann Marschner Schumann's Genoveva . . .16

CHAPTER III. ROMANTIC OPERA IN PARIS. Auber's La Muette Rossini's Guillaume Tell Scribe and Meyerbeer's Robert le Diable, Les Huguenots, &c. Halevy Herold Adam Berlioz' Benvenuto Cellini ......... 36

CHAPTER IV. ITALIAN OPERA. Bellini Donizetti

Verdi 58

CHAPTER V. THE FURTHER DEVELOPMENT OF RO- MANTIC OPERA. Comic operas : Lortzing Flotow Nicolai. Operas and Operettas in other European countries : Berlioz' Les Troy ens and Beatrice et Benedict. Works by Gounod Ambroise Thomas Bizet Offenbach Peter Cornelius - Goetz Glinka Moniuszko Smetana Sullivan 67

CONTENTS vii

PAGE

CHAPTER VI. OVERTURES AND SYMPHONIES. Mendels- sohn's Overtures Schumann's Overtures Schumann's and Mendelssohn's Symphonies - Wagner's Eine Faust- Ouvertiire .... 80

CHAPTER VII. PROGRAMME Music. Berlioz* Sympho- nies and Overtures Felicien David's Le Desert Liszt's Symphonies and Poemes symphoniques . IIT

CHAPTER VIII. ORATORIOS AND CANTATAS. Mendels- sohn's St. Paul, Elijah, Lobgesang, Die erste Walpur- gisnacht Schumann's Paradise and the Peri, Der Rose Pilgerfahrt, Choral-Balladen. Religious pieces. The music to Goethe's Faust. Berlioz' La Dam- nation de Faust, and L'Enfance du Christ

CHAPTER IX. RELIGIOUS Music. Berlioz Liszt Rossini Verdi Wagner's Biblical Scena, Das Liebesmahl der Apostel . . . . . . j 73

CHAPTER X. CONCERTOS AND CHAMBER Music . . 225 N

CHAPTER XI. PIANOFORTE Music. Weber Field

Mendelssohn Schumann Chopin Liszt . . 237

CHAPTERXH. SOLO SONGS. Schumann Mendelssohn Robert Franz Berlioz Liszt Wagner. Balladen. Melodrama 271

CHAPTER XIII. VARIOUS WORKS BY OTHER MASTERS. Anthems by Samuel and Samuel Sebastian Wesley. Glees and Catches. Miscellaneous pieces by Niels Gade Sterndale Bennett Onslow Rubinstein Stephen Heller Sullivan Volkmann Lachner - Kiel - - Goetz Kirchner Jensen Peter Cornelius Hans v. Billow Raff. ^The Neo- Russians : Balakirev Rimsky-Korsakow Cui

viii CONTENTS

PAGE

Musorgsky Alexander Borodine and his opera Prince Igor Dargomijsky Tchaikovsky. Grieg

Max Bruch Felix Draeseke C. V. Alkan. Organ pieces by Mendelssohn Liszt Rheinberger

The Wesleys 289

CHAPTER XIV. THE ROMANTIC OPERAS OF WAGNER

AND THE INCIPIENCY OF THE MUSIC-DRAMA . . 333

CHAPTER XV. MUSICIANS AS WRITERS ON Music. Historical studies Antiquarian research Editing of classics ........ 350

CHAPTER XVI. SUMMARY AND CONCLUSION . . 361 INDEX .......... 367

THE ROMANTIC PERIOD

CHAPTER I

INTRODUCTION

Music, polyphonic and harmonic, considered in relation to other departments of artistic endeavour, is somewhat late in its manifestations. It was not until the fifteenth and sixteenth centuries, when Counterpoint flourished in the Netherlands, England, and Italy, that the affinity between polyphony and Gothic architecture became apparent. The religious painting of the early Italian Renaissance hardly found a counterpart in music before Palestrina's time. Some aspects of the secular poetry and painting of the later Renaissance are reproduced in the Italian and English Madrigals and the Spanish, French, and English lute music of the seventeenth century. The spirit of Protestantism acquired its musical voice very gradually first in Schiitz, then in Bach and Handel. The elaborate courtesy and urbanity of the first half of the eighteenth century are, to a certain degree, reflected in the operas of Gluck and Mozart. Echoes of the worship of nature, the humanitarian enthusiasm, and the social upheaval of the second half of that century can be traced in the symphonic work of Beethoven, which properly belongs to the first quarter of the nineteenth. Even in the revival of Teutonic myth, and in the dreamland of Norse, Celtic, or Slavonic poetry, the relation of things poetic to things musical exhibits the same order : the songs of Tieck^s Schone Magelone had to wait for Brahms, the revival of old German stories of gods and heroes for Wagner. But as time goes on and new processes are discovered the intervals become shorter. The

DAXSKEUTilEli \\

a THE ROMANTIC PERIOD

rate of growth is accelerated, and musical art comes nearer to the sister arts, to thought, and to life. Examined from this point of view, the period of musical development here to be discussed the Romantic period, from Weber to Wagner, i. e. from Der Freischutz to Lohengrin and Tristan, or from Weber's overtures to the symphonic pieces of Berlioz and Liszt is, on the emotional side, seen to be imbued with the spirit of romantic poetry and literature, whilst, on the technical side, it is a time of transition from the formal to the characteristic, from the f Singspiel 3 or the f Opera seria ' to the ' Opera caratteristica,' and the Wagnerian Music-drama, from the Sonata to the 6 Characterstiick/ from the Symphony to the fPoeme sym- phonique/

Romantic music is, in some sense, an offshoot of literature ; a reflex of poetry expressed in musical terms; a kind of impressionism which tends to reject formality, and aims at a direct rendering of its object ; a desire to produce musical effects suggested by natural phenomena ; an art eager, sensitive, impulsive, which seeks its ideal of beauty through emotional expression. With Wagner it is ancilla dramatis a powerful rhetoric which, like scenery and action, is made subservient to the purposes of the Theatre.

Literary Romanticism, about 1800, found a voice for the thoughts and feelings which by natural reaction had begun to invade the rationalistic world of the eighteenth century. It was not so much a protest against classical work as against some aspects of the reasoned taste in art that had sprung from the spirit of rationalism. It gave voice to a keen love of the past, especially of the religious aspect of past ages, and to a keen passion for nature. The Christian ideals and the ideals of Rousseau met in the sentimentality of Bernardin de St. Pierre, Chateaubriand, de Senancourt, and many others who mistook the furore espressivo for a symptom of strength. ( L'aspiration du sublime/ says George Sand, cetait meme une maladie du temps c'etait quelque chose de fievreux qui s'emparait de la

INTRODUCTION 3

j eunesse.5 In Germany as in France, a little later on, the growth of music proceeded on similar lines. German music followed the traces of German literature at an interval of a generation or so whilst in France, more directly in accordance with the ficole romantique, music and literature came to be very nearly contemporaneous. On every side it is evident that music received, and in its peculiar way brought to efflorescence, modes of feeling which had their roots in literature.

Goethe claimed to have been the first to use the elusive term Romantic in contrast to Classical. 'The familiar conception of the classical and the romantic arose in my mind and in Schiller's. My maxims were in favour of the objective method of treatment. But Schiller preferred his own subjective method, and defended it in his essay on "Naive and Sentimental Poetry." He showed that I was romantically inclined in spite of my desire to be otherwise, and that my Iphigenie, because of the preponderance of sentiment, was by no means so antique in feeling as I thought. Later on the brothers Schlegel took up the matter; and people now talk glibly of classicism and romanticism, to which fifty years ago no one gave a thought V Romanticism in Germany reached its acme during the period of reaction and restoration following the Napoleonic wars, when the Germans, in their ardour for historical studies, developed the historical sense a high degree of sympathy with modes of feeling long lost, a taste for conditions and characteristics of past ages and climes remote. The historical studies at the Universities induced a leaning towards criticism. A school of literary critics arose ; innumerable translations were attempted (Homer, Shakespeare, Calderon, Cervantes), philologists worked at Sanscrit, Persian, Old Norse, and mediaeval German. Teutonic mythology was reconstructed ; legends, fairy tales, remnants of folk-lore, popular ballads, and songs, were collected and compared.

1 J. P. Eckennann, II. 137, Oesprache mil Goethe: three parts published in 1837. The preface to Part I is dated October 31, 1835.

B a

4 THE ROMANTIC PERIOD

Following in the wake of Thomson, J. J. Rousseau, and their own Goethe, German romanticists in prose and verse such as Novalis, Tieck, Fr. and A. W. Schlegel, Eichendorf, Brentano, and later on Chamisso, Hoffmann, Fouquet, Jean Paul Richter, Uhland, Ruckert, Platen, Lenau, Heine attempted an emo- tional interpretation of nature. The prevailing sentimentalism, together with their special historical predilections, prompted the earlier of these writers to deal mainly with mediaeval legends, with magic superstition, with knight-errantry, and the worship of woman. Mythology and poetry, symbolical legend and art, they asserted, must be considered as one and indivisible; 'the deepest want and deficiency of modern art lies in the fact that artists have no mythology to fall back upon V And they insisted that only in the service of religion had art ever flourished only in that service could it flourish. Some of them, like Tieck and A. W. Schlegel, inclined towards mediaeval Catholicism the revival of which they greeted with the sympathy of poetical enthusiasts ; others, like Fr. Schlegel and Novalis, accepted its doctrines in full ; whilst contemporary German painters, such as Overbeck and Cornelius, under the same influence, busied themselves with an attempt to revive the forms of early Christian art. Goethe, with some disdain, spoke of the painters as ( Nazarenes/ and of the litterati as 'forced talents/ What he particularly disliked in the pro- ductions of the eccentric younger men was not so much the prevalent musical note in their verse, for in their schemes of verbal melody they often followed his example, as the quasi-musical mood underlying the things set forth vague visions and states of feeling, dissolving views bathed in moon- light and made faintly vocal with the notes of the Aeolian harp.

The German writers of fantastic prose or verse (E. T. A. Hoffmann excepted) were but dilettanti in music lovers of fine sound with little knowledge of the art. Yet it is of real interest

1 Fr. Schlegel, Gesprache fiber Poesie, pp. 263, 374.

INTRODUCTION 5

to note how persistently their instinct turned towards music as a possible fulfilment of their aims, and how much that is sane and good in their efforts has found expression in artistic music.

Premising that music, be it vocal or instrumental, is incapable of describing matters of fact, we may speak of the romantic element in music as poetical suggestion by musical means. The suggestiveness and witchery of certain fragments of impressionist music is best illustrated by citation from Weber and Wagner. Take a few typical examples : the last eighteen bars of the second act of Die Meistersinger, when the moonlight peers through the streets of old Nurnberg after the riot, the forest voices in Siegfried, the strange chords when Briinnhilde sinks to sleep. Weber, Wagner's great prototype in this respect, has such things in abundance e.g. the introductions of the overtures to Der Freischiitz and to Oberon, the intermezzo in the overture to Euryanthe. Effects of mystery dreams of the past, with all the illusions of a dream * The horns of Elfland faintly blowing ' a rustle of fairies, the chant of some nameless knight dim memories and suggestions, Mondbeglanzte Zauber- nacht, Liebe denkt in siissen Tonen, Alles singt zu dieser Stunde. Nor is the suggestiveness confined to snatches of Weber or Wagner. Entire pieces or strings of pieces by com- posers contemporary with both of them are imbued with it. Take, for further instance, Schubert's C major symphony ; com- pare the tones of the horn in the introductory 'Andante' and the ' Andante con moto ' ; take the sound of the strings and oboe in Mendelssohn's overture The Hebrides, and of the horns in his Scotch symphony, or Schumann's trumpet effects in the overture to Manfred, his Song-cycle Liederkreis von Eichendorf, or certain of his pianoforte pieces such as theKreisleriana. Every- where there are musical echoes of nature, recollections of ' old unhappy far-off things ' ; appeals to sentiment and emotion which would lose half their force had they not already been anticipated by literature.

6 THE ROMANTIC PERIOD

Of course musical romanticism in the nineteenth century, be it German, French, or Italian, was not a new departure. Pre- monitions of it are already discernible in Schiitz, d'Anglebert, Froberger, Buxtehude, J. S. Bach, Beethoven, Dussek, Schubert, Spohr ; also in the work of certain men who made their mark in Italy or in Paris Gluck (parts of Alceste, particularly the superb recitative chorus and aria, e Grands Dieux,' Act III, No. 3) ; Pergolesi (Stabat Mater] ; Spontini (La Vestale, notably the chorus of vestals in E f) ; Mehul (Uthal and Joseph en figypte) ; Boieldieu (Jean de Paris and La Dame blanche}. No one can mistake the romantic sensibility expressed in Froh- berger's Tombeaux (' Laments ') ; in J. S. Bach's Prelude in E !? minor1, the Chromatic Fantasia, the recitative arioso Ach Golgotha, unseliges Golgotha, and Am Abend da es kuhle war; in that Passion according to St. Matthew, in Dussek's Elegie harmonique ; or in some of Beethoven's smaller pieces, such as Adelaide, An die Hoffnung, Op. 94, In questa tomba oscura ; Liederkreis, An die feme Geliebte, the first movement of the Sonate pathetique, the second movement of the Trio in D, Op. 705 and in many a song and instrumental piece of Schubert.

It is worthy of note that the touches of romantic sentiment in Schiitz or in the works of some of Bach's precursors like Buxtehude and Bohm, as in J. S. Bach's own compositions, occur but sporadically, whilst the technical means employed to express them are in no way exceptional. The case is very different with the composers of the nineteenth century. Romantic sentiment with them is always present, and at the same time they exhibit a continuous striving to keep pace with the spiritual transmutation that^ is going on around them. Under some personal impulse, some suggestion without, always with a view to musical characterization, they are seen to be taking pains to elaborate this or that point, or trying to discover

1 Compare it with Chopin's Etude in C g minor, or with the Preludes Nos. a and 4 in E minor.

INTRODUCTION 7

more suitable technical means to attain this or that particular effect.

From Weber's time, about 1820, a new spirit was in the air and an increasingly rapid process of change and expansion resulted from its appearance. It can be traced from Spohr and Weber to Mendelssohn, Schumann, Gade, Sterndale-Bennett, Rubinstein, and Tchaikovsky ; from Berlioz to Liszt ; and from Schumann, Liszt, and Berlioz to the ingenious Neo-Russians such as Balakirev, Borodine, Cui, and Rimsky-Korsakov ; and again from Wagner, Berlioz, and Liszt to Anton Bruckner, and, mutatis mutandis, to Richard Strauss. Taken altogether, the romantic movement in so far at least as instrumental music and the orchestra is concerned appears as an unconscious tendency towards the relaxation of the laws of structure in favour of characteristic details, an almost total rejection of organic design on self-contained lines, and, step by step, an approach to a sketchy sort of impressionism and a kind of scene-painting a huge piling up of means for purposes of illustration. No doubt it was guilty of many excesses. It was often crude, often extravagant ; sometimes apparently inspired by mere defiance and bravado. But, when all this has been said, it remains true that the net gain, the widening both of the range of knowledge and of the scope of emotion, which has resulted from the move- ment, is a possession the value of which cannot be overrated *.

After Weber the change from the formal to the characteristic and the tendency towards programme music went on apace. The attitude of Mendelssohn and Schumann, regarding titles and descriptive indications generally, like that of Beethoven and the earlier masters, was the attitude of the specific musician. Up till about 1850, music on self-sufficing lines was by far the main concern from the professional point of view. With

1 The reaction against vague impressionism or pictorial illustration came in Germany under Brahms, and in England under Parry, Stanford, and the other masters of the English Eenaissance ; and it seems to be coming in Eussia. The stragglers, adherents of Berlioz and Liszt (their name is legion), hardly count.

8 THE ROMANTIC PERIOD

Mendelssohn and Schumann, conscious poetical intentions, admittedly present in many instances, appear on the second or third plane as it were by implication only and do not directly touch the musical design. Every good overture or introduction to an opera has in the nature of things a sort of programme implied or avowed. The programme consists of the opera to follow witness Beethoven's overture to Fidelio 1. Mendelssohn held that any attempt to reduce definite combinations of tone and rhythm that is to say, musical expression generally to verbal expression or vice versa, must result in failure. Musical utterance, he maintained, is positive, and remains so whether or not a more general significance be attributed to it ; whilst any verbal or pictorial allusion to the effects of music will be less definite than the music itself. To this it may be added that music contrived with intent to illustrate a ready-made programme may by lucky chance, and in a way of its own prove to be a satisfactory statement of essentials. But it is by no means sure to be so, and this was the point of view Wagner took regarding Liszt's symphonic programme music.

Not only in Germany, but all round, the spirit of Weber's opera led the romantic development. * Questo e inventore/ as Jommelli said of Piccinni. Through the whole field of musical art Weber's temper prevailed. From Der Freischutz onwards, opera (apart from the bel canto of the Italians) is in a large measure derived from Weber, and many of the novel procedures in instrumental music rest on his method. Thus, on the one hand, the innovations and experiments in the treatment of operatic forms first suggested to Weber by theatrical considera- tions, constitute the starting-point of the changes that led to Meyerbeer's Robert le Liable, Marschner's Vampyr and Hans Heiling, and the earlier operas of Wagner2. And, on the other

^ Spohr in his Faust has gone the length of explicitly acquainting the audience with what he wishes them to imagine.

2 Lohengrin was the last of Wagner's works which he called a ' romantic opera.'

later application of Beethoven's symphonic music to his own ends as a musical

dramatist-in Der Ring des Nibelungen, Tristan und Isolde, Die Meistersinger

INTRODUCTION 9

hand, the orchestral pieces the overtures and symphonies of Mendelssohn and his followers, as well as the orchestral pieces of Schumann, bear traces of Weber's spirit and practice. Besides, in several minor departments, such as music for the pianoforte, and part-songs for male voices, Weber stands in the position of a pioneer x.

In France the immediate musical sequel to the romantic movement in literature was restricted to about a dozen works. Several grand operas belong to it: Auber's La Muette de Portici, Rossini's Guillaume Tell, Meyerbeer's Robert le Diable, Halevy's La Juive; as also the symphonies and overtures of Berlioz, his early opera semi-seria Benvenuto Cellini, and an 'Ode-symphonic,' Le Desert, by Berlioz' disciple Felicien David.

The representatives of the opera comique Boieldieu, Auber, Adam, Herold moved on the ordinary light French lines of Gretry, Monsigny, Dalayrac, Isouard, and kept far away from the glimmer of romance ; whilst in the operas of Italian com- posers the influence of French literary and pictorial romanticism (Victor Hugo, Delacroix, Ary Scheffer) can be traced in a few cases only in the sentimentality, for instance, of Bellini's Norma, La Sonnambula, and / Puritani, which had its souree in Rousseau, through Chateaubriand and Madame de Stael, and in the violence of some of Verdi's early operas, Ernani, Rigoletto, both of which are founded on plays by Victor Hugo. The power of the romantic literature of the day is also manifest in certain early Russian, Polish, and Bohemian operas, such as Glinka's La Vie pour le Tzar, Russian et Ludmilla, Serof's Judith, Moniusko's Halka, and Smetana's Prodana nevhta (< The Bartered Bride').

von Number g, and Parsifal is best considered apart; though in many ways, even in the latest phase of Wagner's art, Parsifal, the romantic impulse is felt to be present, and the master appears intimately connected with romanticism both on the literary and the musical side.

1 The entire German 'Mannergesang' of the present day, with all its political significance, springs from Weber's initiative in such spirited little masterpieces as Lutsow's wilde Jagd and Gebet vor der Schlacht.

io THE ROMANTIC PERIOD

Many an important development in orchestral music, apart from the stage, starts indirectly from Weber's overtures to Freischiitz, Euryanthe, Oberon. Weber's vivid imagination and fiery impulse found complete expression in these typically romantic pieces. Less compact and strictly consistent on musical lines than Beethoven's overtures, they are, to a large extent, made up of extracts from or allusions to scenes of the operas to which they serve as introductions. But they are so well put together, the contrasting modes are so well arranged, that there is no suspicion of any shortcoming in design. Mendelssohn in his concert overtures improved on Weber's type, inasmuch as the thematic materials he worked with, figures, and melodies, are more homogeneous and suit- able for development. The plaintive tone pervading Fingal, the translucent atmosphere of Meeresstille, the tender and delicately passionate romance of Melusina, have little in common with Weber's dramatic power; but Weber's instru- mental technique is present, notably in the overture to A Mid- summer Night's Dreamy and many a subtle device of Weber's orchestration is adopted and put to ingenious use. In his sym- phonies, just as much as in his overtures, Mendelssohn aimed at the expression of distinct moods and definite ideas, but always on lines of purely musical design. In no instance did he ven- ture beyond distinctive titles or inscriptions, and it is significant that he has hardly ever furnished sub-titles, or any other hints of his meaning, though he might just as well, for instance, have called the slow movement of the Italian symphony a ( Pilgrim's Procession' as the finale a ' Saltarello.' Schumann, the Romantic par excellence after Weber, was more inclined to give special titles to his musical poems. One of his symphonies, the first in Bt>, was known to his friends as the Spring Symphony, and another, Et>, Op. 97, as the Rhenish. His overtures, Manfred and Genoveva, were intended to be close reproductions of the moods which underlie the subjects indicated or suggested by the titles. But with Schumann, as with Mendelssohn, the

INTRODUCTION n

titles are never meant to furnish programmatic details which can in any way control the course of the music.

Nothing could better prove the strength of the impulse towards characterization than the fact that Spohr, who in most things was a follower of Mozart, should in his latter days have com- posed symphonies illustrating ideas more or less alien to music. The true initiator, however, the path-finder in the direction of musical illustration, was Berlioz. He and Liszt are the most conspicuous and thoroughgoing representatives of programme music, i.e. instrumental music expressly devised to illustrate in detail some play or poem or some succession of ideas or pictures.

In pieces such as the first and last movements of Berlioz* Symphonic fantastique, the first and last movements of his symphony Harold en Italic, Liszt's Poemes symphoniques Ce qu'on entend sur la montaane, after a poem by Victor Hugo, and Die Ideale, after a poem by Schiller, the hearer is bewildered by a series of startling orchestral effects which are not explicable on any principle of musical design.

The use of the oratorio and the cantata for concert rather than church purposes which in Germany began with Haydn's Creation and Seasons, and with Beethoven's setting of Goethe's Meeresstille reached a climax when Mendelssohn produced his cantata Die erste Walpurgisnacht and the oratorios St. Paul and Elijah. Before that time Spohr, notably in The Last Judgement, had found the oratorio better suited to his powers than the opera or the symphony. Contemporaneously with Mendelssohn, Schumann made a new start with a secular oratorio (or rather a set of three cantatas) after Moore's rimed story { Paradise and the Peri' in Lalla Rookh. Berlioz, too, attempted something resembling oratorio in the ' Trilogie sacree ' L'Enfance du Christ, and something like a dramatic cantata, or an e opera de concert,' in La Damnation de Faust. With his Messe des Marts, and the Te Deum, he aimed at the seculari- zation of church music, purposely setting the words of the Catholic service on a grandiose scale and somewhat histrionically..

12 THE ROMANTIC PERIOD

Liszt, in a more religious spirit, followed with the Graner Fest- Messe and Ungarische Kronungs-Messe, the Thirteenth Psalm, the oratorios St. Elizabeth and Christus.

A tendency to deviate from the lines of the sonata, akin to the tendency which brought about changes in the concert-over- ture and the symphony, can be traced in the concerto, the quartet, and other forms of concerted chamber music. To Schumann and his contemporaries it seemed difficult to produce string quartets true in spirit and technique, yet distinct from the classical models, unless they could venture upon some modi- fications of design in the direction of the ( Characterstiick/ Schumann appears to have felt that the chances of making the most of his peculiar gifts were not entirely favourable on such lines. His three string quartets, written so rapidly that they may almost be described as improvisations, are of more account for beauty of detail than for any general mastery of design, and it is significant that after the single outburst which created them he turned, as though with a sense of relief, to the combination of pianoforte with stringed instruments by which the modern taste for warm colour and volume of sound can be gratified without fear of conflict between technical means and the end in view.

The tendency towards concise expression of emotion reached its zenith in some of the short lyrical pieces, both vocal and instrumental, by Schumann and in a number of the solo piano- forte pieces by Chopin. Chopin and Schumann appear as the greatest lyrists among romantic composers of the century. In their best pieces for the pianoforte, both forsook the old ordin- ance of the sonata, and treated the pianoforte as the confidante of their personal feelings. Thus they found new ways and new patterns of expression, discovered abundance of novel and striking effects of sonorousness, and brought about a notable change in the spirit as well as in the technique of pianoforte playing. Some detailed account of these representative pieces, as well as of certain concertos for the violin, the pianoforte, and

INTRODUCTION 13

solo pieces by Weber, Mendelssohn, Schumann, Chopin, Liszt, and others, will be attempted later on when it will also be con- venient to touch upon the Lieder of Schumann, Mendelssohn, Robert Franz, Liszt, Wagner, the melodies of Berlioz, the Balladen of Loewe, and the church music of Samuel and Sebastian Wesley.

This sketch of the most important work produced during the period would be incomplete without a reference to the influence of the instrumental virtuosi who from Paganini to Liszt, Thalberg, Ernst, Vieuxtemps, Wieniawski, Joachim aroused the enthusiasm of the public, and to the good work done by some of the leading composers as critics and writers on music. About the beginning of the nineteenth century a change in the status and habits of certain classes of musicians took place. The 'Musicien de chambre' (' Kammermusicus ') became the wandering virtuoso who appealed to the miscellaneous public with a show of manipulative skill. Composers of concerted and solo music had in consequence to reckon with an element of display l which soon came to be regarded as an important ingredient in the composition of instrumental pieces. The inclination to emphasize virtuosity in lieu of musical quality came to a head with the appearance of Paganini, who exhausted the technical capabilities of the violin, and of Liszt and Thalberg, who did the same with the pianoforte. Inevitably one amongst the results was an undue preponderance of glitter and show and an inordinate display of gymnastics in finger, wrist, and arm. But at the same time there was brought about a result of which we are now reaping the benefit that is to say, technical mastery, absolute command of all the capabilities of an instrument. To exhaust the technical possibilities in every direction was the task that executants of genius like Paganini and Liszt set themselves to accomplish ; and the command they attained over every kind of executive difficulty was the most salient, if not the most valuable

1 It was always there —witness Each and the concertos of Moiartaud Beethoven but not prominently and pointedly so.

i4 THE ROMANTIC PERIOD

among the many factors which contributed to the style of their playing.

Dilettanti and literary men fond of music and eager to discuss it had not been wanting in the eighteenth century, from Mattheson, Marpurg, Rousseau, and Grimm onwards. But professional musicians with a taste for literature, themselves competent to act as writers and critics, were hardly known to exist in Germany before the time of Weber and in France before Berlioz. E. T. A. Hoffmann lawyer and litterateur by profession, composer and conductor by choice had written a number of fantastic articles about Mozart, Gluck, and Beethoven, showing a rare combination of poetical with practical insight, and gaining for himself the reputation of a prophet of romanticism. But Weber was really the first among professional musicians to put forth his opinions with the distinct object of instructing the public. He helped to start Meyerbeer, Marschner, Hoffmann, Fescaj and many others. In early days it appeared as though Schumann would be able to make better use of his rare gift of verbal expression than of his specifically musical gifts. Generously putting aside his own claims to recognition as a composer, Schumann, with comprehensive sympathy, acted for a number of years as the advocate of Chopin, Mendelssohn, Gade, Sterndale-Bennett, Heller, Liszt, Berlioz and, finally, of Brahms. Thus he materially assisted in bringing about a change for the better in the relation between aspiring composers and the public. Liszt's enthusiasm was most helpful in the cause of Wagner. Berlioz furnished reports ' a travers chants ' for some twenty- eight years. His peculiar view of music and its effects on the mind and body can in a manner be taken as embodying a romantic leader's profession of faith, and may fitly close this chapter of outlines. It is an exaggerated view, perhaps, but it fairly represents the spirit of the period l.

' La musique, s'associant a des idees qu'elle a mille moyens

1 A travers chants, p. I .

INTRODUCTION 15

de faire naitre, augmente 1'intensite" de son action de toute la puissance de ce qu'on appelle la poesie . . . reunissant a la fois toutes ses forces sur Poreille qu'elle charme, et qu'elle offense habilementj sur le systeme nerveux qu'elle surexcite, sur la circulation du sang qu'elle accelere, sur le cerveau qu^elle embrase, sur le coeur qu'elle gonfle et fait battre a coups redoubles, sur la pensee qu'elle agrandit demesurement et lance dans les regions de Pinfini ; elle agit dans la sphere qui lui est propre, c'est-a-dire sur des etres chez lesquels le sens musical existe reellement/

CHAPTER II

GERMAN EOMANTIC OPEKA FROM WEBER TO SCHUMANN

A GERMAN 'Singspiel' is a play of .light texture in prose or verse copiously supplied with incidental music. A German f Oper ' is a musical play in which a minimum of dialogue is employed for the exposition of the situations, while music serves as the exponent of emotion. With the exception of Weber's Euryanthe and Schumann's Genoveva, German operas up to Wagner are merely enlarged ( Singspiele.'

The increase in the function and the efficacy of music in connexion with the actor's art ; the widening of its scope, as it gradually rises from the position of a merely incidental embel- lishment in a play, to melodrama, where it accompanies, illustrates, and enforces the action; and again, the rise from melodrama to the aria and operatic scena where it serves as the principal means of expression are well seen by the com- parison of certain German dramatic pieces wherein incidental music plays a part or contributes to the denouement for example, Schiller's Turandot and A. P. Wolff's Preciosa with Weber's music. In Turandot, a play in Chinese garb, adapted from Gozzi's // Re Turandotte, Weber's music is incidental only, consisting of orchestral pieces an overture, marches, &c. In Preciosa, a romantic play in four acts, the music answers the purpose of embellishment as well as direct expression and characterization. Some of the dance tunes, choruses, and portions of the melodrama, might be omitted without loss to the action ; but Preciosa's solo dance in the first ' Ballo,' and particularly her song ' Einsam bin ich nicht alleine,5 are essential to the part. It is worthy of note that this song

OPERA FROM WEBER TO SCHUMANN 17

illustrates the character and contributes to the portrayal of Preciosa as much as * Meine Ruh' ist hin ' illustrates the character of Gretchen in Goethe's Faust, or ( Freudvoll und leidvoll' and 'Die Trommel geriihret' portray Clarchen in Egmont.

The triumph of romanticism in operatic music begins with Weber's success Der Freischiitz (1821) 1. In the story of this opera the motives contrast greatly with the rather stilted pathos of the older opera seria and with the equally conventional comicalities of the opera buffa. Certain romantic elements, such as the myjstery of the forest, the interference of demoniacal powers in the life of men, the redemption of a man's soul through a woman's devotion, appealed powerfully to the instincts of the German people and assisted in establishing Weber's work in the position of a national favourite.

After the failure of Napoleon's invasion of Russia, Weber had come to the fore as the musical exponent of German aspirations for independence with his setjing for male voices of some of the fighting songs from Korner's Leyer und Schwert, such as 'Liitzow's wilde Jagd,' 'Gebet vor der Schlacht,' fDu Schwert an meiner Linken.' His melodies stimulated the national enthusiasm, and Germany hailed Der Freischiitz as the first artistic expression of its patriotic senti- ment 2. It is the most German of operas ; the music, in some respects, is the very quintessence of contemporary German popular melody. The workmanship throughout is conscientious and sincere. Never before, unless it be in the second act of Beethoven's Fidelio, has so intimate a connexion between the orchestra and the stage been attempted. Weber's fine feeling for effect and his extraordinary sense of instrumental colour served to define and contrast the scenes and situations, and to intensify the emotional expression. 'The various characters

1 Weber was born in 1786, he died in 1826.

1 ' Die verschiedensten Richtungen des politischen Lebens trafen hier in einem gemeinsamen Punkt zusammen : von einem Ende Deutschlands zum anderen wurde Der Freischiits gehb'rt, gesungen, getanzt.' Wagner- Schn/ien, i, p. 266.

DANNKBUTBEB

i8

are perfectly identified with the music they have to sing. Caspar, the reckless meddler in dangerous magic, was easily drawn; but the heroine Agathe and the lighter spirited Aennchen both also keep their musical identity quite well, even when they are singing together. The scenes are separate, but the final transition to the continuous music of later times is happily illustrated in such a case as Agathe's famous scena, in which a great variety of moods and changes of rhythm and speed and melody are all closely welded into a complete and well-designed unity V Perhaps the most remarkable section of the whole work is the melodramatic Finale of the second act, the scene of the casting of the magic bullets. Here Weber found splendid opportunity for the suggestive and descriptive power of his music. The orchestral effects are as novel and telling to-day as they were at the first performance : ' For such things must be heard,5 as Beethoven said of them. They told on contemporary musicians, hostile or friendly, just as they tell to-day a.

In his scenas and melodramas Weber represents the environ- ment, as well as the emotions, of the characters. ' His music conveys a pictorial and a psychological impression simultane- ously 3/ as in Max's scena and aria ' Durch die Walder, durch die Auen ' (Freischiitz), in the cavatina f Glocklein im Thale ' (Euryanthe), wherein the musical phraseology is equally well adapted to every phase of emotion that passes through Euryanthe's mind as to her woodland surroundings, and more distinctly still in Agathe's scena in Der Freischiilz, ' Wie nahte

1 C. Hubert H. Parry, The Evolution of the Art of Music, p. 316.

8 A telling effect familiar to every concert-goer, the long-drawn melody for the clarinet in the overture, is thus described by Berlioz : ' Cette longue m&odie ge'missante, jet<5e par la clarinette au travers du tremolo de 1'orchestre, comme nne plainte lointaine dispers^e par les vents dans les profondeurs des bois, cela frappe droit au cceur; et, pour moi du moins, ce chant virginal qui semble exhaler vers le ciel un timide reproche, pendant qu'une sombre hannonie fr^mit et .menace au- dcssous de lui, est une des oppositions les plus neuves, les plus po&iques et les plus belles qu'ait produit en musique 1'art moderne.'

1 W. H. Hadow, Lecture delivered at the Royal Institution, 1902.

OPERA FROM WEBER TO SCHUMANN 19

mir der Schlummer,' in Rezia's scena in Oberon^ Ocean, thou mighty monster/ and in the great melodrama already mentioned which forms the Finale of the second act of Der Freischiitz.

The mere names of Weber's operas and Singspiele suffice to show the wealth of opportunity for the display of colour which the romantic subjects afforded him. Thus Silvana (1812) an improvement on Das Waldmadchen (1800) and Der Freischiitz (1821), serve to exhibit German forest legends. J*re.ciosa_(i$2i) is based on a Spanish novel of Cervantes. Euryanthe^ (i 823), based on an old French story, illustrates feudalism. Turandot (1808), Abu Hassan (1811), and Oberon (1826), represent the East.

The novelty and peculiarity of Weber's method in opera consist in the close and persistent attention to characterization and the use of special devices of orchestration for particular purposes, so that the musical speech of one character shall be palpably distinct from that of another. Weber attains his object by the use of instrumental tone-quality, i. e. ( colour ' ; also by the use of folk-songs and dances, or of melodic and rhythmic traits belonging to them, i. e. local colour. It must not, however, be supposed that devices to obtain particular descriptive effects by suggestion were a new thing in Weber's time. Such devices were by no means rare in the older classical opera, but they were employed there in single and separate pieces only as wtyen Gluck introduced the choruses of Scythians in Iphigenie en Tauride, when Mozart employed the so-called Turkish music in the Entfiihrung aus dem Serail, or when he made use of a Spanish fandango in the second act of Le Nozze di Figaro. But Weber, in Der Freischiitz, and still more in Euryanthe, kept up the colour once adopted throughout an entire scene, or an entire act. And he managed to do this by the use of striking melodic curves, figures, rhythms, and such peculiarities of instrumentation as are suggested by the particu- lars of the action and the environment.

The vigorous rhythms, characteristic figures, and ingenious orchestral contrivances which go to make up Weber's design also

C a

20 THE ROMANTIC PERIOD

assist in the presentation of certain scenes and generally in the portrayal of the characters. ' That fellow stands like a house/ Beethoven said of Caspar in Der Freischutz. Caspar is depicted as a burly peasant, with an uncanny touch of evil ; and indeed Weber's originality in the invention and the use of striking vocal phrases and instrumental effects to depict special characters is surprising1. Weber's strong feeling for ' local colour > had its concomitant in a love of folk-songs and dances. Adopting a method entirely different from that of Beethoven, who occasion- ally takes a hint from a folk-tune or embodies fragments of folk- song or dance as a builder might have used curious bits of stone, Weber introduces popular melodies just as he may have heard them sung or played in the fields or streets, or found them in books, e. g. the Gipsy march in Preciosa, the Peasants' march or the waltz in Der Freischutz, or the Chinese tune in Turan- dot, which he found in J. J. Rousseau's Dictionnaire de la musique. His own peculiar type of melody, closely akin to the contemporary German Volkslied, is based upon the major scale and often consists solely of a statement of the notes of that scale resting on chords, of the tonic or of the dominant or sub- dominant chords. Apart from this native type, which is Weber's personal note and prevails in those of his works that count most, his fondness for exotic tunes— Bohemian, Spanish, Polish, Hungarian, and even Turkish or Chinese is in all cases typically romantic. Thus Weber, together with Schubert, is the initiator of those picturesque touches of exotic tonality or instrumentation, and of that tendency towards nationalism in melody, rhythm, and even harmony, which is now so prominent a feature in music. The German tradition of spoken dialogue is still maintained in Der Freischutz. Once only, in Euryanthe2 (1823), Weber

1 The first part of Caspar's drinking song consists of a succession of three-bar phrases, the second part of a succession of four-bar phrases. In Oberon there are several cases where the melodic outlines show the juxtaposition of four bars or five. Compare the first part of the subject in Brahms' variations Ueber ein eigenes Thema, Op. 21.

2 Euryanthe, grosse heroisch-romantische Oper in drei Auf ziigen ' (literally,

OPERA FROM WEBER TO SCHUMANN 21

discarded it and trusted to music alone, or rather to music in close connexion with poetry, mimetics, and scenic accessories. "The proper effect of my new work/ he wrote in 1824 (letter to the Musik-Verein of Breslau), { can only be expected from the united efforts of the sister arts.' The ideal towards which his instincts led him was that of the musical drama as subsequently realized by Wagner in Lohengrin. He intended to make the design of the several musical movements conform to the course of the action, and the details were to spring directly from the verse with as little repetition as possible of single words or lines. But his intuition of the theatrical concentration necessary to present his effects was not equal to his genius for musical expression. He made a mistake in the choice of subject. His imagination was captivated by the glamour of romantic incidents displayed in an old French story of a lady's constancy, a version of which he had read in Count Tressan's Bibliotheque des romans l. But he disregarded, or at least underrated, the want of true interest in the leading motives. He was not far wrong in his belief that the principal personages concerned in the * Histoire ' were sufficiently distinct to serve as types for musical characterization. His librettist, Frau von Chezy, published a translation of the original text in F. Schlegel's Sammlung romantischer Dichtungen des Mittelalters, 1804, and a revised version 1823, as Geschichte der tugendsamen Euryanthe. But she failed to produce an intelligible play from the materials at her disposal. As the opera now stands 2 the difficulty from the point of view of the stage lies in the fact that the plot rests on the existence of a certain secret, constantly referred to but

Grand heroic-romantic opera in three rises of the curtain). First performed at Vienna in 1823 ; first heard in London in 1833, and again in 1882.

1 ' Histoire de Gerard de Nevers et de la belle et vertueuse Euryante de Savoye, sa mie ' a tale probably known to Boccaccio in its original verse form, ' Roman de la Violette/ by Gilbert de Montreil (Decamerone, second day, ninth tale), and again in Boccaccio's version known to Shakespeare (Cymbdine).

3 According to Frau von Chezy it was rewritten eleven times. It has been revised again, Vienna, 1904, seventeen pages of the pianoforte score cancelled, some cuts sanctioned by Weber restored. It seems a hopeless case.

22 THE ROMANTIC PERIOD

not clearly explained. Weber, when he had already written the greater part of the music, came to realize this source of trouble. Apparently in despair, and at the last moment, he adopted an expedient. In the course of the overture, at the Largo played by the strings pianissimo possibile, the curtain is directed to rise upon a tableau vivant showing Adolar and Euryanthe, the hero and heroine, at the tomb of Emma, Adolar's sister who, at the death of her own betrothed, is supposed to have sucked poison from a ring and who now, to the weird sound of the music, tries to inform Adolar and Euryanthe in dumb show that ' her soul will never find rest until the fatal ring is bathed in tears of innocence/ This absurdity exhibits the crux of Romanticism in music, that is to say the gap between the end in view and the means to attain it.

Is it surprising that a work which commands the admiration of musicians when they read it, leaves them dissatisfied when it is .acted, sung, or played ? The excellence of the musical material is incontestable; the score contains finer individual passages than any other work of Weber's, the exigencies of musical design do not hamper the action, the style is noble, broad, consistently German ; and yet Euryanthe has failed.

Wagner, loth to charge Weber with a lack of discernment in stage effects, attributed the indifferent success to a want of balance between the musical and the dramatic factors. ' Critics/ he said, ' have not thoroughly sifted the heterogeneous elements which meet and contradict one another in this work, nor have they tried to show that the composer's inability to combine them into a harmonious total was the true cause of failure. Never since opera began has there been a work like this, in which the contradictions of the entire operatic genre have been so methodi- cally exposed, by a composer so gifted, genuine, and high-minded. These contradictions are, absolute self-contained melody as an end per se, and dramatic expression which shall be true through- out. Assuredly the one or the other, the melody or the drama, must give way. Rossini sacrificed the drama : Weber tried to

OPERA FROM WEBER TO SCHUMANN 23

restore it by the power of his expressive melody, and had finally to acknowledge the impossibility of the task V

Viewed from the standpoint of Lohengrin, where scenic arrangement, verse, and music are justly combined, Wagner's contention cannot be denied, though it is more a defence of his own position than an appreciation of Weber's work. The words of the historian Ambros 2 are more to the point: 'The libretto of Weber's greatest work Euryanthe is a romantic product ; and it is greatly to Weber's credit that he succeeded in giving flesh and blood to these moonlit phantoms of Prove^al knights and ladies. The role of Eglantine in his hands becomes a demo- niacal figure such as had not yet been depicted in music. The part of Ortrud in Wagner's Lohengrin is modelled upon that of Eglantine, even to the wild burst of triumph at the end of both operas. Similarly the role of Telramund in Lohengrin rests on that of Lysiart. Euryanthe is truly an epoch-making work. The roots of Wagner's art, as we have it in Derfliegende Hollander, Tannh'duser, and Lohengrin) spring from this score of Euryanthe, which is also the source of much of Marschner's operas and of some part of Meyerbeer's operatic writing.' Weber's fairy opera Oberon, on the lines of an enlarged Sing- spiel after the manner of Der Freischutz, was written to order for Covent Garden under Kemble's management in the year of Weber's death, 1836. It proved a popular success in London and abroad, but lailed to hold the stage. Well aware of the puerility of the book, WTeber expressed his intention of having it recast, and of rewriting the music, ' so that it shall deserve to be called an opera.' The fame of Oberon now rests on the delightful fairy choruses, the superb overture, and the great aria ' Ocean, thou mighty monster/

The nature of the stories o^Spohr^s 3 principal operas (Faust, 1815-8, Zelmira und Azor, 1718-22, Jessonda, 1823, and Die

1 Wagner, Operund Drama, iii, p. 361.

J A. W. Ambros, CuUurhistorische Bilder aus dem Musikleben dvr Qegenwart, p. 45.

* Spohr was born in 1784, he died in 1859.

24 THE ROMANTIC PERIOD

Kreuzfahrer, 1 844), together with a certain plaintive sensitiveness, i. e. chromatics in the inner parts, which is the personal note that pervades Spohr's music, has induced German writers to present him as one of the originators of musical romanticism, and by reason of the early date of his Faust (he began the work I in 1813), as the precursor of Weber in romantic opera1. A study of his scores does not, however, bear this out. It is a far cry from romantic elements in a libretto to true romanticism in music. Spohr's predilections and, what is more important, his musical method, are distinctly Mozartian to say classical would perhaps be saying too much. The formal finish of his pieces and the easy mastery of orchestral effect fascinated contemporary musicians. But nothing can be further from the drastic verve and vivid suggestiveness of Weber's musical speech than the languor of Spohr's melody and his fondness for square rhythms and square structure 2.

It would have been nearer the mark to represent E. T. A. Hoffmann3, with his opera Undine (1816), after de la Motte- Fouque's Marchen, as a precursor of Weber. But though Hoffmann acted for a number of years as a professional musician and writer on musical subjects, he never rose above the level of a highly gifted dilettante. His opera did not and could not gain a firm footing on the stage, it was far too tentative. Weber introduced it to the public of Prague. The story of Undine is

1 The overture to Faust has a kind of programme which Spohr caused to be printed at the head of the libretto : ' The tone pictures of the Allegro vivace, Largo grave, and Tempo primo are meant to suggest to the auditor the changing moods and conditions of Faust's inner life.'

2 The chromatic and enharmonic intervals in the inner parts, of which Spohr was so inordinately fond, may once and again have furnished hints to Schumann or Wagner. But it is abundantly evident that the chromatics of romantic music, as we find them in Schumann, Chopin, Wagner, are really, in so far at least as harmony is concerned, derived from J. S. Bach and his sons, and only to a very limited extent from Mozart, or Haydn, or Spohr. Compare introduction to Mozart's quartet m C and Haydn's orchestral pieces, ' Chaos ' in the freation, and Summer ' in the Seasons.

* Hoffmann, 1776-1 82 2, wrote several singspiele, three operas, a mass, a requiem, a symphony, several overtures and sonatas, some chamber music, and a number of smaller vocal and instrumental pieces.

OPERA FROM WEBER TO SCHUMANN 25

that of a water-nymph, who has no soul, but who acquires one through her love for a knight. Together with the birth of the soul, she also receives knowledge of human distress and pain ; her hus- band breaks his faith, and she kills him with a kiss. The music to Undine, though performed twenty-one times in all (1816-7), has not been published, but manuscript copies of the score, preserved at Berlin, bear out Weber's appreciation of it : { The work is one of the most ingenious of recent years . . . and so consistent that details disappear and the interest is absorbed by the composition as a whole . . . the composer avoids empha- sizing any particular piece to the detriment of another, he is careful not to hamper the action and always strives for true dramatic expression. The part of Kiihleborn stands out as that of the most prominent character, by reason of the particular cast of melody and instrumentation which persistently accom- panies his uncanny appearance V

Weber objected to certain weak points in Hoffmann's music, such as his love of short phrases and figures which lack variety, monotonous employment of violoncellos and violas, amateurish use of sequences of diminished sevenths, and of cadences which are abrupt and occur too often in the same shape. On the whole, it may be said that the characters and situations in Hoffmann's opera are well depicted. The declama- tion, in the airs and the comparatively few recitatives, is remarkably direct and spirited. The overture and the short instrumental pieces which serve as introductions to the second and third acts rather poor and somewhat incoherent are made up of scraps and hints of things to come. The orchestra- tion, particularly of the vocal pieces, shows an acute sense of instrumental colouring and considerable knowledge of effect. The following extract from Undine may be taken as fairly representative 2 :

1 Compare ' Caspar * and ' Samiel ' in Der FreiscMtz.

* Compare the excellent article on Hoffmann by Vianna da Motta, the quotations in Bayreuther Blatter, 1898, and Die Musik, p. 1666.

THE ROMANTIC PERIOD

Kuhleborn.

ft r

u I

Siehst Du die' Wel - lea in blan . ker Ver -

fflrffbdj

1

r

r'

r r r

s r

fir-rung? Siehst Du die Wel - len iu blari - ker Ver

flat

-r-^rpg-q

ffi & r

TBbd

Y

TJndine.

= g

wii

-rung?

Ist's

recht so ?

«/

1st reel

it.

-

•*•

r-

^

rK

J

P 1 Horns 1

rf

r.

r

4-

-

-f-h

:-Hr^

1 J «i

H ^ : 1

^ «M

r r

r r

-fc

a

{

i

j

gpjL^I 1 L 1 Dreht Euch zur

hel .

len be -

"J r

tau - ben - den

r J r

Ir - rung.

"

TT-

^

- ^

•«*•

•r-

.

rtt?

^

"-r-f-f^rr

r

p

r-r-C 4=

r— f—

f r r^

^ r

OPERA FROM WEBER TO SCHUMANN 27

Marschner1 had the tact to select subjects fit for theatrical presentation and favourable to the display of his peculiar musical gifts, which lay in the direction of jovial popular humour, com- bined with a striking and, from a theatrical point of view, very effective combination of sentiment with a feeling of awe and horror. Friar Tuck, in Der Tempter und die Judin, is a good example of the former quality; and the latter Marschner's idiosyncrasy is exhibited, in a more or less prominent manner, by each of the heroes of his three best operas Der Vampyr (1828)2, Templer und Judin (1829), and Hans Heiling

As early as 1820, Weber had credited an opera of Marschner's, Heinrich der Vierte und D>Aubigne, with c vivid original invention and careful workmanship.' Schumann, twenty years later, summed up his impression of Der Templer und die Judin thus : ' the music is occasionally restless : the instrumentation not sufficiently discriminate. There is a good deal of clever melody, considerable dramatic talent sundry echoes of Weber. A gem not entirely cleared from its rough covering.'

Wagner used to point to certain portions of the Templar's long scena, No. 1 2, Act II, particularly the passage where the rapid triplets of the wind instruments depict a feverish state of excitement, as remarkably spontaneous and original examples of emotional expression :

1 Heinrich Marschner (1795-1861) became Weber's assistant as conductor of German opera at Dresden in 1824, and fora number of years subsequently acted as conductor at Hanover.

1 The sixtieth performance of Der Vampyr took place in London, 1829. The libretto is based on a little known fragment of a novel by Byron, Augusta Darnel (first published together with Mazeppa), which Byron began in 1816 at Geneva, when Mary Shelley wrote Frankenstein.

THE ROMANTIC PERIOD

.gg-g—f-^r fi **" ^-y^^

r r | <-> z-\

Dooh dass ich's bin, wer an - ders 6va. $^z If.

als ein Weib bracbt'

=1P} \ i JT\

tf tf

-j-—* g z=|

f f

r r -^

Allegretto J . = 96.

dim

OPERA FROM WEBER TO SCHUMANN 29

m

Als ea Bo

Stark In jeg

is Ouil

li - cher

S

dim,

%5i

ff f

man, Klang

8i-a.

be riihtnt der Min

Al lie

len, der

^{ii^^

&&

mei Mnt

Da mir

8i'i.

me

iiu

Na Her

men

zen

^

£=C

i

THE ROMANTIC PERIOD

dim.

dim.

OPERA FROM WEBER TO SCHUMANN 31

Hans Heilinff, an opera based on one of the manyjegends in which a goblin in disguise woos a maiden, appears to be the immediate precursor of Wagner's Hollander as regards both the play and the music. In like manner the opening of its prologue, when the hero departs from the subterraneous abode of the Queen of the Goblins, contains the germ of the scene in the first act of Tannhduser, which culminates in Tannhauser's flight from the Venusberg.

Wagner's admiration for Marschner, though sincere and warmly expressed both in public and private, was yet by no means unqualified. He strongly objected to certain banalities which now and then disfigure Marschner's melody and to his rather slovenly declamation. He also drew the line between ( the mellifluous choral sing-song' of some of the concerted pieces for male voices, which Marschner in his operas and in separate publications addressed to the populace, and the noble and touching choruses which have dramatic significance, such as those in the Finale to the second act of Der Templer.

Marschner's operas, later in date than those mentioned, and sundry miscellaneous pieces of vocal and instrumental chamber music of no particular importance, never gained a hearing out- side Germany ; and even there the repute of his three typically romantic operas is distinctly on the wane. But the fact that they form a link between Weber and Wagner's early operas secures for them a permanent place in the history of the German operatic stage.

Schumann J, with his introspective ways and his devotion to personal ideals, was the least theatrically minded musician that it is possible to conceive. A man totally devoid of mimetic gift and as far removed from the theatre as an educated German can well be, he had but scant acquaintance with the aspects of opera from the standpoint of the audience, and knew next to nothing about its conditions as they appear to actors and singers. Beyond listening to an opera once in a while, he does not seem

1 Schumann was born in 1810, he died in 1856.

32 THE ROMANTIC PERIOD

ever to have come in contact with the operatic stage and its belongings; nor did he trouble to make a special study of the conditions of success, when the desire to produce an opera took hold of him. The ' Theaterbiichlein ' little theatre-book, printed at the end of his collected writings records impressions of some fifteen operas heard at Dresden during the years 1847- 50. His words show him to have been keenly responsive to certain musical points, such as details of instrumentation, the treatment of the voice in connexion with particular instruments, the use of the chorus. But the histrionic side, the peculiar position and function of music in combination with stage action, does not appear to have occupied much of his attention. While seeking a suitable subject for an opera, he examined and rejected the stories of the Nibelungenlied, the contest of Minnesanger at the Wartburg, Die Braut von Messina, Abelard et Helo'ise, Faust, Sakuntala, Byron's Corsair and Sardana- palus, Moore's Veiled Prophet of Khorassan, and many others, amongst which was a sketch of his own, after E. T. A. Hoffmann's Doge und Dogaressa. In the end his choice fell on the legend of St. Genevieve perhaps of all the subjects that came under consideration the least amenable to effective treatment as an opera. Apart from its popularity as a favourite story like that of Patient Grissel, Genoveva seems to have attracted him as bearing a certain affinity to the story of Weber's Euryanthe (both stories tell of maligned innocence, banishment, and ultimate rehabilitation), and perhaps also because his ambition was stimulated by the prospect of producing something like a match for Weber's work. The construction of a libretto seemed to be easy, since the legend of Genoveva had already been treated in dramatic form, for reading purposes by Tieck, and for performance by { Maler Miiller' (the poet of Schubert's 'Die Schone Mullerin'),Raupach,andHebbel. Tieck's tragedy1, written in alternating rimed verse and prose, is a long shape-

1 Leben und Tod der heiligen Genoveva, ein Trauerepiel von Ludwig Tieck, 1799 ; Oenoveva, cine Tragddie in funf Akten von Friedrich Hebbel, 1843.

OPERA FROM WEBER TO SCHUMANN 33

less, pious, and rather insipid expansion of the old Volksbuch, Geschichte von der heiligen Genoveva. HebbePs is an ex- travagant acting play in blank verse, with a touch of brutality in the treatment. Starting with such material, Robert Reinick, painter, poet, and a friend of the composer, sketched a scenario, but failed, after repeated attempts, to satisfy Schumann, who then, in vain, applied to Hebbel for assistance. Finally, following as he thought the example of Lortzing and Wagner *, Schumann himself undertook the task of arranging a libretto. He managed, with very indifferent success, to contrive an amalgamation of the two plays. In accordance with Hebbel rather than with Tieck, he chose to eliminate most of the legendary features, which lend a charm to the mediaeval story. Thus certain telling traits, delicately developed in Tieck's version Genoveva's long sojourn in the wilderness, the friendship of her little son with the doe and the beasts and birds, even that most musical incident, the chance meeting with her husband and the recogni- tion and reconciliation are rejected ; and their place is taken by certain ugly scenes from HebbePs play, which exhibit the insults and brutalities Genoveva suffers at the hands of her domestics. e Do not expect to find the old sentimental Geno- veva 2 ; ' ( I rather believe it is a piece of actual life, as a dramatic poem should be.' Golo, the traitor squire, who, with Tieck, wavers inconstantly between wickedness and contrition, is pre- sented as a cowardly sensualist and scoundrel. Genoveva's husband Siegfried, and Margaretha the witch-wife, are little better than lay figures. The final scene of the second act, where the rabble of servants who a moment before might have sworn to Genoveva's innocence force their way to her apartment, is as repulsive as the murder of the old Seneschal Drago, Genoveva's supposed paramour. Part of the third act is occupied by a series of pictures shown in a magic mirror whilst an invisible

1 Genoveva , 1849-50, was completed two years after the first performance of Tannhauser.

* Schumann : Letter to Dora, 1849.

DAJiKBEUTUEB D

34 THE ROMANTIC PERIOD

chorus comments in the background. The fourth act contains a repetition of the tiresome incidents which make up the close of Weber's Euryanthe a visit to the wild wood, an attempt to murder, a rescue, and the usual jubilant Finale. 'When Schumann consulted me about the libretto of his opera V says Wagner, f I failed to induce him to alter the ineptitudes of the third act, especially the magic pictures : he lost his temper and appeared to believe my warnings were meant to mar his best efforts/

It is evident that Schumann had no just conception of the magnifying or dwarfing effect of stage presentation. He did not perceive that a particular incident, though sufficiently inter- esting in narrative, may yet appear puerile or artistically impos- sible from a histrionic point of view ; and that to read about brutality is one thing, to see it presented on the stage, with the details emphasized by music, is a very different matter altogether*.

Schumann adopted Weber's method of connecting the scenes and fusing the words and the music into one single and coherent act, as in Euryanthe, but he fell far short of Weber's grip and brilliancy. In Genoveva the composer's power of invention appears to be on the wane. The daring originality, the force and passion of the younger Schumann is gone. The opera con- tains no spoken dialogue, and nothing resembling plain recitative, which might perhaps have acted as a foil to the lyrism that pervades the whole. Neither in general design nor in detail does the music spring direct from the dramatic situation. Through- out there is a lack of actuality of vivid contrast and telling colour. The composer occupies the position of an annotator. He stands outside the story, and puts forth his own musical comments on the situation which it portrays. Rarely, if ever, does he rise to the height of his opportunity and succeed in making the

1 Wagner-Schriften, x. pp. 222, 223.

8 Liszt's verdict on Schumann's work was generous. Writing to a friend in 1855, when he was rehearsing Genowwx at Weimar, he says : ' I prefer certain faults to certain virtues— the mistakes of clever people to the effects of mediocrity. In this sense there are failures which are better than many a success.'

OPERA FROM WEBER TO SCHUMANN 35

characters speak their own language. He cannot paint affresco, as in the salient points of an operatic scena he should. The wealth of clever detail in melody, harmony, and at times even the orchestration, is no doubt interesting ; but it must be added that in performance the majority of these subtleties do not make the impression of being in perfect accordance with the action. They are too minute, and therefore do not produce the effect intended *.

1 On its first performance at Leipzig, 1850, Schumann himself conducting, Genoveva was coldly received. Subsequent performances at Dresden, Vienna, Weimar, Leipzig and also a revival in English, carefully prepared by the pupils of the Royal College of Music under Sir Charles Stanford in 1893 one and all resulted in a lukewarm succis cfestime.

D 2

CHAPTER III

ROMANTIC OPERA IN PARIS

IN France, as compared with Germany, the powerful romantic movement in literature was less in accord with the national taste in music. A leaning towards romanticism in music was mainly confined to those members of literary and artistic coteries, amateurs for the most part, who felt the influence of Byron, Scott, Moore, or Goethe, and to some extent of Bee- thoven. Parisian musical romanticism was but a reflex of the ferment in French literature. It came to the fore at the Opera, where everything is sung in French ; whilst the so-called opera comique, where the entertainment consists of light music alter- nating with spoken dialogue, was hardly touched by it. At the Opera, the way was opened, in 1828, by the Muette de Portici (Masaniello) of Scribe and Auber *. This work, romantic, ex- travagant, revolutionary in spirit, is rich in captivating tunes, full of clever instrumental effects, and remarkable for the novel use of massing the chorus, so as to permit them to take a prominent share in the action. ' So lively an opera had not yet been seen V It was the first realistic drama in five acts, with all the attributes of a tragedy, particularly a tragic end. I well remember that the latter circumstance made an especial sensation. Hitherto operatic stories (in Germany at least) had always ended comfort- ably— no German composer could venture to send people home in a sad mood. When Spontini came to Dresden to conduct

1 Scribe, 1 791-1861, produced, or at least lent his name to, 422 pieces 47plays, 28 grand operas, 95 comic operas, 344 vaudevilles, and 8 ballets. Auber was born in 1783, he died in 1871.

* Wagner, Erinnerungen an Auber, ix. 55 (1871).

ROMANTIC OPERA IN PARIS 37

La Vestale, he waxed wroth when he found that, after Julia had been happily saved from death, we intended to let the opera conclude with the scene in the cemetery. He would not permit such a thing. The scenario had to be changed, the Bower of Roses with the Temple of Venus had to appear, the priest and priestesses of Love had to lead the happy pair to the altar: ' Chantez, dansez J . . . impossible that things should be other- wise. . . . But all this conventional business suddenly came to a stop when La Muette appeared. Here was a grand opera, a tragedy in five acts, completely set to music ; without a trace of stiffness, of empty pathos, of so-called classical dignity, warm enough to burn, heady enough to intoxicate. German musicians confessed themselves bewildered by the new prodigy, and, after some acrimonious discussion of its merits and defects, finally cut the knot by referring it to the influence of Rossini. This judgement, like many others in the history of Music, has been reversed on appeal. No doubt Rossini was in a sense the father of modern operatic melody, yet even he was unable to produce or rival the particular quality that gave such dramatic power and effect to this music of Auber's ; moreover, the fates denied, not only to Rossini, but to other Italians and Frenchmen, and even to Auber himself, a chance to continue in the path of La Muette 1.

Before we speak of Rossini's romantic masterpiece Guillaume Tell 2 it may be well to touch on his career in Italy, and to trace through early years of conflict the preparation for his triumphs in London and Paris. In the main a self-made man, he studied Mozart's operas, Haydn's and Mozart's symphonies, and tried to score some of their quartets. He rapidly acquired facility, though not complete mastership. At first he composed for the lesser Italian theatres, rapidly producing serious or comic operas, which lived, at best for a couple of seasons, and at worst for

1 Performances of La Muette have repeatedly furnished an excuse for political demonstrations.

3 Rossini was born in 1792, he died in 1868.

38 THE ROMANTIC PERIOD

a single day. Then, in 1813, he achieved his first great successes with Tancredi, an opera seria, and U Italiana in Algerij an opera buffa. The latter, a forerunner of // Barbiere di Sivifflia, was a surprising display of comical gaiety and verve with a savour of Cimarosa's // Matrimonio segreto, but still with a distinct note of its own J. These were followed by Elisabetta, reginad' Inghilterra (Naples, 1815) and II Barbiere (Rome, 1816) on a subject already treated by Paisiello, after which came La Cenerentola and La Gazza ladra (1817), Mose in Egitto (1818), and, in 1819, La Donna del Lago, after Scott's The Lady of the Lake. The failure of Semiramide one of his most ambitious works at Venice, in 1823, prompted him to go to London, where he laid the foundations of his fortune, and in the following year to Paris, where the authorities appointed him Director of the Opera with a salary of 20,000 francs and a share in whatever tantiemes he might be entitled to in case he chose to write a new work or rewrite an old one. At first he adopted the latter plan and rewrote two of his best Italian scores Maometto II, which at the Opera became Le Siege de Corinthe (1826), and Mose in Egitto, which was called Mo'ise. A third piece, Le Comte d'Ory, a pretty comic opera (1828), was a pasticcio of old and new fragments. Finally, he crowned the edifice with Guillaume Tell.

In the transformation of older work the influence of French theatrical art and French taste in music is felt ; and there is no doubt that Rossini was bent on making the most of his talents. He added and changed a good deal, especially with regard to instrumentation, rendered the declamation more precise, the accents more incisive, and revised the entire workmanship with a fuller comprehension of the requirements of the stage.

A grand opera, Italian in all essentials, yet French in aspect, elaborate in style and rich in melody, Guillaume Tell (1829) came upon the world as a surprise. Certain qualities always

» The once popular sentimental tune from Tancredi, < Di tanti palpiti,' now serves for the professional song of the tailors in the third act of Wagner's Mdstersinger.

ROMANTIC OPERA IN PARIS 39

rare with Italian composers, and particularly rare with Rossini characterization in great things and in small, the orchestra in touch with the action, careful declamation, appropriate local colour1 combined to form a masterpiece forwhich neither school was entirely prepared. Up to the time of Guillaume Tell Rossini had mainly addressed himself to hearers who had a sense of musical verve and movement, as for instance in // Barbiere di Siviglia. In Guillaume Tell he chose to speak seriously to the elite of the public and of professional musicians. He aimed at perfection of musical style under cosmopolitan con- ditions, and attempted a fusion of the good qualities of Italian, French, and German masters in opera. The workmanship in the score of Guillaume Tell is good throughout, the melody spon- taneous, the harmony often refined in a very original manner; the treatment of the solo and chorus voices and the orchestra masterly. The overture ranks high amongst overtures of the potpourri sort, and is only excluded from the first place by the bustling vulgarity of its close. The power and originality of the principal pieces is best shown in the scene of the conspiracy and the taking of the oath at the Riitli ; also in the duet between Tell and Arnold, in Matilda's recitative and romanza at the beginning of the second act ; in the dainty ballet tunes and the Tyrolienne of Act III ; the quartet in Act I ; and the storm in Act IV. The solo parts exhibit a wealth of device for the display of fine emotional singing ; and nowhere does Rossini's affluence of vocal melody fail him. It may be that such tune- ful facility as is his rests on the traditional musical speech of the earlier Italian composers such as Piccinni, Paisiello, Cimarosa, and in some measure on that of Mozart; but the melodies themselves, even if at times they exhibit touches of superficial emotionalism or border on triviality, have a stamp of their own and possess a peculiar sensuous charm. Guillaume Tell exhibits full measure of the scenic display that especially belongs to the

1 Local colour so perfect was not again seen or heard in opera till 1875, when Bizet's Carmen was produced.

40 THE ROMANTIC PERIOD

Parisian opera grand choral masses, a large corps de ballet, sumptuous pageantry, dazzling effects of light and colour, &c. and the skilful rhetoric of the libretto has given rise to many a novel and telling detail in the musical setting.

To illustrate the power of Rossini's temperament when it touches upon a strong theatrical situation no better example exists than the Terzetto in Act II between Arnold, Tell, and Walter tenor, baritone, and bass. The scene shows Arnold (the son of a patriot leader, connected with the governing party by his love for the governor's sister) at the moment when he is informed of the ruthless slaying of his father j and the music reflects the conflict of his emotions.

f Tell.

[ Walter.

*»il

ri - mor so che il cor gli mar

~±- Jl J— •&

ra dell' . . a - mor og - ni no - do spez-

ROMANTIC OPERA IN PARIS

Arnoldo.

z6 !

io mo

ro !

I

fTell. 1 Walter.

quel duo - lo giii ca - de e de

Arnoldo.

JTell.-P6' a 1 Walter. N k

il pa quel duol gia de - Hi

dre, ohi - me ' ! ' mi ma-le.

i f-

gia la ben - da strap -

THE ROMANTIC PERIOD

va,

ed io

1' El - ve - zia allor tra-

ra, a quel duol gia de - li

-^ h. k ^ K

Iff

p6,

la ben - da strap

•q -i : . •! -1 -=%:

=3 9 1

di

va,

T P

o Ciel 1

quel duol gii de -

m

po, a quel duol gia de - li

?A-

m

Ciel! li

mai piii lo ri - ve

ra, gia la ben - da I a

ra, gia la ben - da fa - ta

strap-

m

**

t

ROMANTIC OPERA IN PARIS

43

gE§ES_g*te-g-3

dro, mai piii lo ri -ve-dio, no, no, mai piii non lo ri . ve - dr6 talo, si la ben - - da atrap-po

P

_-j i ^±3 -=r at

There is in this piece direct presentation of feeling very effective from the actor's point of view and a considerable degree of musical originality. It will be found worth while to compare it with certain celebrated operatic ensembles of a later period, such as the quartet in Bellini's / Puritam, the sextet in Donizetti's Lucia, the quartet in Verdi's Rigoletto, or even with the superb quintet in Wagner's Meistersinger, the middle portion of which is musically, if not emotionally, cast in a similar mould.

As time advances it appears evident that Rossini made a mistake in not demanding more than a mere picturesque book of words from his librettist, Etienne de Jouy1. Guillaume Tell contains little that resembles a plot. The interest in the story wanes after the second act, and the succession of pretty scenes does not make amends. In performance, the order of things produces a sense of diminuendo there is a gradual falling off from the scene of the morning sun on Alpine summits, after the nocturnal meeting of the conspirators, to the storm on the lake, the leap from the little boat, and the shooting at the apple. In Germany, where Schiller's Wilhelm Tell had

1 Writer of the libretti to Spontini's La Vestale and Olympic, and mentioned in Byron's Letters, vol. vi. p. 230, as the author of the tragedies Scylla and L'Hermite.

44 THE ROMANTIC PERIOD

familiarized people with the peculiar order of scenes1, this drawback was hardly noticed, and the opera, accepted on its musical merits alone, became a great favourite.

It is characteristic of Rossini that he does not indulge in experiments. There is always a personal note about his work, be it trivial or passionate the intuition of a great personality. Something resembling a distinct personal note is also felt to be present in the work of his principal successors, Bellini, Donizetti, and Verdi, notably Bellini, but in a far less marked degree. The successors too had something of their own to say, and, under prevailing theatrical conditions, did say it, often in a convincing way. No one can justly assert that Bellini, Donizetti, and Verdi derived directly from Rossini. Did they openly imitate his ways or copy his mannerisms? Bellini never attempted such a thing; Donizetti, an Italian eclectic aspiring to cosmopolitan sway, here and there followed him ; Verdi, late in life, worked on totally different lines2. The kinship is the kinship of tradition. The ways of Italian opera persist, though the lines of its development may deviate there is the spirit of continuance in its treatment of the theatre and in its entire absence of introspection. Together with Donizetti (Lucia di Lammermoor), Bellini forms the link in the growth of opera from Rossini's exuberant force and the consummate savoir-faire of his later years to the more earnest and consciously cosmopolitan art of Verdi. In France, however, soon after the appearance of Guillaume Tell, people began to hint at defects which they could not describe they felt that there was something wanting or something amiss. The fact that Rossini's individual art had gained much by its contact with the French stage was gracefully acknowledged. f But ' it was asked in artistic circles fif the maestro's masterpiece does not entirely fulfil the promise of romanticism, cannot

1 The order is not that of a play, but rather of an epic poem, the plan for which Goethe communicated to Schiller. 8 Compare his OteJlo with Rossini's.

ROMANTIC OPERA IN PARIS 45

some other musician be discovered who will reproduce the fantastic or demonic side of it, and create a truly romantic opera ? '

This other musician was soon found, and proved to be Jacob Meyer Beer, better known as Giacomo Meyerbeer l. The wild side of romanticism ran riot in Scribe and Meyerbeer's Robert le Diable, 1831. In this work the most strenuous theatrical and musical means are employed to bring about contrasting effects. All the elements of romantic and operatic excitement are made to serve the same purpose : characters and situations as extravagant as possible, demons and men in conflict, plain-chant and ballet-tunes intermixed, church-pageantry transported to the stage, prayer alternating with bacchanalian song, simple tunes interlarded with gruesome melodramatic chords, and the most ethereal effects of instrumentation in conjunction with vulgar noise. On the dramatist's side there is an exhibition of extremes ; on the musician's an accumulation of Italian, French, and German devices grossly exaggerated ; and the total is contrived with little regard for consistency of style, and with hardly a trace of artistic conscience. 'Meyerbeer's object was to make the mere externals tell. He did not care in the least whether his details were common- place or not. His scores look elaborate and full of work, but the details are the commonest arpeggios, familiar and hackneyed types of accompaniment, scales, and obvious rhythms. Musically it is a huge pile of commonplaces, infinitely ingenious, and barren. There is but little cohesion between the scenes, and no attempt at consistency with the situation in style and expres- sion. No doubt Meyerbeer had a great sense of general effect. The music glitters and roars and warbles in well-disposed contrasts, but the inner life is wanting. It is the same with the treatment of his characters. They metaphorically strut and pose and gesticulate, but express next to nothing; they

x Meyerbeer was born in 1791, six years before Schubert and eighteen years before Mendelssohn ; he died in 1864.

46 THE ROMANTIC PERIOD

get into frenzies, but are for the most part incapable of human passion. The element of wholesome musical sincerity is wanting in him, but the power of astonishing and bewildering is almost unlimited V

When Scribe shifted the slides of the lantern to replace the romantic phantasmagoria of Robert le Diable by the quasi- historical pictures of Les Huguenots (1836) and Le Prophete (1849), his method remained unchanged. He continued to exhibit the same mixture of operatic contrasts of ecclesiastical display and voluptuous ballet, of passion torn to tatters, and violent death. Nor was Meyerbeer's musical procedure modified in any important way. There is in both these later operas some increase of means such as enlarged choral masses, greater swarms of figurantes, a fuller and even noisier orchestra. There is also some gain in the choice and variety of instrumental colour, some advance in the precision and energy of the declamation; but hardly anything deserving the name of musical polyphony either vocal or orchestral, and what little there is of it savours of banality. Of invention, novelty in the contrivance of melody, harmony, rhythm, there is very little that in any way surpasses the average quality of the musical materials in Robert le Diable. It may interest students to see how this condition of things struck a great contemporary between 1836 and 1850. After protesting against certain frivolous tendencies in the book of Les Huguenots, Robert Schumann wrote of the music as follows: *It would take volumes to comment on the music. Each bar has been considered and reconsidered by the composer, and something might be said about it. To astonish and to amuse is Meyerbeer's object, and he succeeds with the vulgar. As for the ubiquitous chorale, " Ein' feste Burg," about which French journals rave, I confess that if a clever pupil were to submit counterpoint of that sort to me I would beg of him not to do worse in future. How studiously shallow, how carefully superficial 1 C. Hubert H. Parry, The Evolution of the Art of Music, p. 315.

ROMANTIC OPERA IN PARIS 47

is all this obtrusive screaming of " Ein* f este Burg " at the groundlings ! A great fuss has also been made about the " Benediction of the Swords " in the fourth act. I admit that the piece has a good deal of dramatic force, that it contains several clever and striking traits, and that the chorus is particularly effective and makes a great show. The situation, the stage accessories, and the instrumentation support one another ; and since the terrible, and the horrible, are Meyerbeer's predilection, he has done his share of the work with enthusiasm. But, if we examine the particular melody closely, is it other than a bedizened marseillaise ? And does it really take much artistic wisdom to produce a strong effect by such means and in such a place? I do not blame the use of every possible means of effect in the right place but people ought not to cry aloud and marvel if a dozen trombones, trumpets, and ophicleides, together with a hundred men singing in unison, are audible at some distance. A special refinement of Meyer- beer's must not be forgotten here. He knows the public too well not to perceive that too much noise might produce apathy. And how cleverly he gets over this. Directly following upon such rattling movements as the "Benediction of the Swords," he inserts entire airs with the accompaniment of a single instrument as if to say, " Behold, ye people, how much ,1 can do with so little." Certainly some degree of esprit cannot in this instance be denied him 1. It would be an easy task to point out traces of the style of Rossini, Mozart, Herold, Weber, Bellini, and even of Spohr. Meyerbeer's

1 The telling effect of contrast Schumann here alludes to really belongs to Weber, who in Euryanthe, Act II, after the violent duet between the evil characters Lysiart and Eglantine, introduces Adolar's aria ' Wehen mir Lufte Huh' ' with a long delicate ritornello. Compare the similar situation in Wagner's Lohengrin, Act II, where, after the duet of rage and hate between Ortrud and Telramund, Elsa's appearance on the balcony is accompanied by a lovely melody played upon the clarinet. With Meyerbeer the trick, for such it becomes in his hands, is first employed in Robert le Liable, Act III, when Bertram rushes into the cave amid a most violent orchestral uproar, which is immediately followed by the soft ritor- nello of Alice's ' Eomanza.'

48 THE ROMANTIC PERIOD

exclusive speciality, however, consists in that famous ambiguous 'rhythm which appears in nearly all the themes of the opera 1.

( I had already begun to note the passages in which it occurs, but I soon grew weary of the task. Ill-will and envy could alone deny the presence of many better things, even of some truly grand and noble points. Thus Marcel's battle-song is telling, the song of the page is lovely, the greater part of the third act, with its scenes of common life, is interesting. The first part of the duet between Marcel and Valentine is both interesting and character- istic, and so is the sextet. The chorus of derision has a comic effect. The "Benediction of the Swords" is comparatively original ; but, above all, the duet between Raoul and Valentine, which follows it, is distinguished by fluency of thought and musicianly treatment. Yet how can all this atone for the vulgarity, grotesqueness, ambiguity, and anti-musical quality of the whole ? Thanks to Heaven, we have reached an end things cannot come to a worse plight/

Schumann did not cry aloud over Le Prophete. He recorded the first performance at Dresden thus :

( Prophet von Giac. Meyerbeer. (Den a. Febr. 1850.)'

If it be permissible to view and estimate the value of Meyer- beer's work from the standpoint of Wagner's achievements, Schumann's strictures, taken together with Sir Hubert Parry's weighty words, may be taken to represent the verdict severe, perhaps, but not unjust 2.

i I n 5 Li/ r qj* r gj

Leggiero

!! I r ' L2J LIT Llf I &c-

* Meyerbeer's early works— after some failures such as the setting of the 98th Psalm, an oratorio Gott und die Natur, i8u, the operas Jepkthas Tociiter and

ROMANTIC OPERA IN PARIS 49

Meyerbeer's manner is more or less apparent in the operas of Halevy and Herold, and even rouses echoes in the works of Gounod, Bi/et, Massenet, Ambroise Thomas, and Saint-Saens. It is felt in Mercadante's // Giuramento, Donizetti's Lucrezia Borgia, Dom Sebastien, La Favorita ; in Verdi's Rigoletto, La Forza del destino, Don Carlos, and A'ida. There are touches of it in Wagner's Rienzi, and still more in Goldmark's Konigin von Saba and Merlin.

Ludovic Halevy1 stands to Meyerbeer not so much in the position of a disciple as in that of a partial imitator, and in some sense a rival. In La Juive (i 835), Scribe, the librettist, produced a very striking lyric tragedy, and one peculiarly fit for Halevy's talent. At its best, Halevy's music is distinguished by a certain gloomy sublimity. It is often full of dramatic animation, and rarely, if ever, sinks to the depths of Meyerbeer's bathos. The musical movements of La Juive, connected by means of recitative, are for the most part of large dimensions, deftly put together, well written for the voices, well scored, and remarkable for the skilful use of the wood-wind instruments such as corni di bassetto, corni inglesi, bassoons, oboes, and clarinets, in com- bination with modern brass. U Eclair ( 1 835), the only other one of Halevy's many operas (upwards of thirty in all) that was altogether a success, forms a strange contrast to this sumptuous theatrical display. There is nothing in the short libretto of U Eclair that can be called a plot; and the music is for two tenors

Abimelek oder die beiden Kalifen, 1813 (Munich and Vienna), all of which were written while he was still in a state of pupilage (Weber and Meyerbeer studied under the Abbe" Vogler, Browning's Vogler, about 1810-12) consist of a number of Italian operas : Romilda e Constama, Padua, 1 8 1 8 j Semiramide riconosciuta, Turin, 1819; Emma di Resburgo, Venice, 1819; Margherita d'Angiit, L'Esule di Granata, Milan, 1820 and 1822 ; and 11 Crociato in Egitto, Venice, 1824, which was repeated in Paris, 1826. The later operas are : Das Feidlager in Schlesien, 1842, given in Vienna as Vielka, 1844, and rewritten for Paris as U&toile du Nord, 1854 ; Le Pardon de Ploermel (Dinorati), a so-called op^ra comique, Paris, 1859 ; and the grand opera UAfricaine, which, though posthumous, really belongs to the time of Le PropJiite. UAfricaine was first performed in Paris, 1865, one year after the composer's death.

1 Born 1799, d'c<l J862.

DANNREUTHER E

50 THE ROMANTIC PERIOD

and two sopranos without chorus1. In connexion with La Muette and Robert le Diable, Adam's Le Postilion deLongjumeau (1836), and Herold's Zampa (1831), together with his Pre aux clercs (1832), may be mentioned 2.

Benvenuto Cellini, Berlioz' 3 first opera, was produced at the Academie Royale in 1 838, as an excuse' for a ballet, and with- drawn after the third performance. Though meant for an ( opera semiseria ' the work was presented as an opera proper. Originally it consisted of two acts only, each divided into two tableaux. The two acts were turned into four, when Liszt in 1852 conducted the work at Weimar. Subsequently, by dint of omissions, and with Berlioz' consent, the four acts were reduced to three, in which latter form Berlioz himself (1853) conducted it in London, and Biilow revived it at Hanover in 1878. It was never a success. Reasons sufficient to account for the persis- tent failure may perhaps be found in the nature of the subject, which, though lively enough and far from commonplace, does not offer many interesting situations. But the peculiar character of Berlioz' music, the rarity of genuine pathos in the melody, and a continuous striving after novelty of rhythmical effect have had quite as much to do with the disappointing general impression as any defects in the subject or faults in the construction of the book. The music throughout is clever but artificial. For the most part it is anything rather than dramatic. In rapid move- ments the variety and bizarre originality of the rhythms together with the dazzling instrumentation produce a sense of haste and restless excitement, and in slow movements the phraseology

1 Probably Le Guitarrero and La Heine de Chypre would now be completely forgotten, were it not for the fact that the partitions de piano of both are among the journeyman tasks Wagner executed for publishers during his first stay in Paris. He speaks in warm terms of the quartet ' En cet instant supreme,' at the end of the fifth act, and remarks that in the first two acts there are instances of miscalculated effects, when the composer expects clarinets and oboes to do the work of horns and valve-trumpets. This early experience of Wagner's led him, later on, to suggest certain emendations in the scoring of the Scherzo of Beethoven's gth symphony, which, it must be said, are not wanted.

* The latter piece, one of the most popular of light operas, was given for the thousandth time in 1871, and is still occasionally to be heard in Paris.

8 Berlioz was born in 1803, he died in 1869.

ROMANTIC OPERA IN PARIS

fails to convince for lack of warmth and fluency. Always ingenious, Berlioz, offers a superabundance of clever devices in rhythm and orchestration. He seems to be addressing himself to an audience of experts, and consequently ' il faut de Pesprit pour lui en trouver ' ; that is to say, just the kind of musical esprit with which even the experts in his day were but scantily furnished. The result is best described in his own words: f On fit a Pouverture un succes exagere, et Pon siffla tout le reste avec un ensemble et une energie admirable.' Considerable wit and finesse are shown in the whispering duet, Act I, and in the bantering aria of Scaramoglio. There is instrumental humour in the carnival scene which forms the finale of Act II. By the side of such pieces are movements, the scene of the oath for instance, that properly belong to the grand opera. Berlioz described his own score as containing ' une variete d'idees, une verve impetueuse, et un eclat de coloris musical que je ne retrouverai peut-etre jamais et qui meritaient un meilleur sort ' (Memoires, p. 214). The following quotations may serve to illustrate the validity of his claim :

Allegro <=! = 108.

i

Feramosca.

r

. «

f Vi - ve 1'es - cri - me, c'est roon fort, oui, c'est mon fort,

*

iff

Allegretto J = 160.

Rhythme a 7 Temps.

A A A'

E 2

THE ROMANTIC PERIOD

t> Temps.

5 Temps.

S3E

accelerando poco a poco une,

i 3^

5 Temps.

o Temps.

une,

' J

une, niort ! san.s pi - titS je per - ce son

ROMANTIC OPERA IN PARIS

53

pi .

sans pi - titf,

* iF rW

_ _ _ f^- H J J J

TF+f— L±-^

s

un jjoeo «'<.

a tempo

^fr=g

je suis vain - queur

ff

Allegro eonfuoco a = 100.

Teresa.

Cellini.

Quand des som

P

m

Quand des som mets, des com

mets de la nion - ta

Quand des

^^

meta de la mon - ta -

Quand des

54

THE ROMANTIC PERIOD

Allegro confuoco e marcato attai . CHORUS. = 184

Tenors.

Basses.

£j. i c 6=gi

Si la terre aux beaux jours se cou - ron - ne

^P

-* e

-r— i— t

. de

P

S

ger - bes, de fruits et de fleurs, . . en ses flancs I'hoin

f=X

^ c^~

^•-f-*- ^2Epi

ROMANTIC OPERA IN PARIS

55

2nd Tenors.

Dang ti ms lea temps, dans toug les temps dee tic -

me mols » son - ne . .

dans tons les temps, dans toug leg tempe des tr6

N

^=r=^^=^^^

_

p e

son . . meil-leun, hon - neur am mal - trea c: - M-leura !

son . meil-leura. Hon - neur aux ma! - tres ci - se-leurs !

=> «'• .

1st and 2nd Tenors.

hon - ne\ir aux mat - tres ci - se - leura ! . Aux flancs

hon - neur aux mai - tree ci - se - leure !

THE ROMANTIC PERIOD

*=?=?

fSE

de laterre en mois-sonne en tout temps destre'sors meil - leurs. 1st and 2nd Busses.

Hon-neur, hon.

enmob-sonneentouttempsdestresorsmeil - leurs. Hon-neur, hon-

- neur aux mal -tres ci - se - leurs ! hon - neur, .

hon . neur au

J1

- neur aux mal - tres ci - se - leurs ! hon - ne

hon - neur aux

•tf ff if .. FJ

j L v_ i^ r P-

j|.

'

F3/5

-'-

ed.

mat - tres ci - se - leurs !

mat tres ci - so, - leurs !

#

ROMANTIC OPERA IN PARIS 57

Auber's principal contribution to the repertoire of grand opera, as has been stated, was La Muette (Masaniello). All his life long he wrote for the Opera-Comique, and produced (mostly in conjunction with Scribe) upwards of forty light operas and operettas ephemera all of them always bright and amusing, frankly written for the market and addressed to the bourgeoisie. ' Que voulez-vous ? C'est le genre, ' answered Auber, when Wagner expressed his astonishment at certain banalities. At the Opera-Comique, Scribe and Auber met on equal terms. Both show esprit, grace, theatrical instinct, at times even passion but the one, in Heine's phrase, lacks poesy, the other lacks music. Apart from the French stage, Fra Diavolo (1830), Le Domino Noir, Le Philtre (1831), and the little masterpiece Le Maqon (1825) are the best known. The latter had a great run in Germany as Maurer und Schlosser. 'C'est de la futilite indestructible. '

CHAPTER IV

ITALIAN OPERA

OF the Italian composers who made their mark in Paris and London after RQSsmi'j GuittaMms^TdLf the most conspicuous are Vicenzo .Bellini1 with his La Sonnambula (1831), Norma (1832), and I Puritani (1834), and Gaetano Donizetti2 with his Lucrezia Borgia (1833), Lucia di Lammermoor (1835), and La Favorita (1840). None of these productions, though they are their composers' best, will stand close scrutiny as a whole, but each contains one or two pieces that, from a vocal and a theatrical point of view, possess high and genuine merit. Thus the quartet and chorus f A te, o cara,' in / Puritani, the sextet in Lucia, and above all, the Finale to the second act of Norma, are in their way masterpieces. They exhibit a minimum of musical elaboration, yet there is much more contained in them than mere sentimental cantilena. In each the vocal expression rises to genuine pathos and passion, j

Bellini, the favourite of the public and of the great vocal virtuosi such as Pasta, Grisi, Rubini, Lablache, met with scant justice at the hands of professional musicians, especially in Germany. ' Bellini's melody is of a monotonous type, it depends on the bel canto for its effect.' f His cadences are weak, the choruses noisy and trivial, the orchestration childish/ Strictures of this sort may be true enough in the main, but the emotional quality of the pieces already mentioned, and of many a single recitative, aria, 01 scena besides, makes up for much that is poor or defective in Bellini's work as a whole 3. In La Sonnambula, the idyllic

1 1802-35. u 1797-1848.

3 Note, however, the wide divergence of Wagner's estimate of Bellini's abilities

ITALIAN OPERA 59

mood of a slight ' Liederspiel/ a song-play, for such the little opera virtually is, does not suffer much from the preponderance of vocal fireworks that form part of the arias, whilst in Norma (the best of Italian tragic operas before Verdi's Otello) the pre- vailing elevation of sentiment is sufficiently well sustained, in spite of occasional banalities in the shape of noisy tunes in choral unison, long drawn-out sequences of thirds, and the like 1. In the middle of his career Bellini2 had the good fortune to meet the man who for his special purposes proved to be an ideal librettist, Felice Romani a person of considerable literary attainment, of sufficient stage experience and of a rare instinct for that peculiar compromise between stage action and song of which in the time of Rossini and Bellini the traditions still survived from the early days of the operatic spectacle in Italy. Romani skilfully contrived the book of Norma (1831) after a little known French play by Soumet, and that of La Sonnam- bula (also 1831) after a now forgotten vaudeville ballet by ^cribe_^. In these model libretti, Romani provided Bellini with the outlines, skilfully drawn and precisely adapted for musical colouring : dramatic situations easily understood, and demand- ing few stage accessories for their proper presentation, but care- fully arranged and graduated for the lyrical utterance of passion : headlong words for rapid recitatives, telling scenas, culminating in some cluster of verses apt for emotional cantilena. All the

from that of other contemporary German critics. Compare Wagner's account of I Capuletti e Montecchi, 1834.

It cannot be overlooked that critics accustomed to the weakness of most contem- porary German translations of operatic libretti, and, what is worse, accustomed to the lax methods of the German operatic singers of the time, were not in a position to appraise the value of lyrical effusions, such as those of Rossini in parts of Otello, or of Bellini in parts of Norma or I Puritani.

1 Earlier operas of Bellini that deserve passing record are H Pirata (1827), La Straniera (1829), I Capuletti e Monlecchi (1830). His first opera, Adelson e Salvina, written and produced at the Naples Conservatoire, was never published. A manu- script copy of it is preserved in the British Museum.

* He died at the age of 33.

3 Count Pepoli, who wrote the book for I Puritani, followed Romani's lead, though with far less success.

60 THE ROMANTIC PERIOD

rest was left to the composer who again, on his part, knew exactly how to adapt his knowledge and sense of vocal effect to the altogether exceptional gifts and attainments of the great singers for whom he wrote. Of the two composers, Bellini was the more delicately gifted and original, whilst Donizetti was the better trained. The dates of Donizetti's best known operas are: Anna Bolena, 1830; Marino Falieri, 1833; Lucrezia Borgia, 1 833 ; Lucia di Lammermoor, 1 835 ; La Figlia del Reggimento, 1 840 ; La Favorita, 1 840 in all, he produced sixty- six operas before his career was cut short by mental disease. Together with Donizetti, Bellini the sentimentalist forms the link between the gaiety and verve of Rossini and the more strenuous art of Verdi. Bellini has well expressed not only the genuine feeling, but also the prevailing sentimentality of his time. As a master of elegiac melody, he indicates his claim by one fact which outweighs a host of disparaging remarks the fact that Chopin, the modern melodist par excellence, paid him the homage of conscious or unconscious imitation. For some of Chopin's most telling cantilena, no matter how subtle and refined it may appear as he presents it, is essentially the can- tilena of Bellini1.

About seven years after Bellini's untimely death, Giuseppe Verdi2 began to attract attention with his Nabucodonosor (1842), / Lombardt alia prima crociata (1843), Ernani (1844), after Victor Hugo's Hemani, and particularly Rigoletto (1851), after Hugo's Le Roi s'amuse3. In his first operas, such as / Lombardi, he exhibits an audacious temperament and power- ful theatrical instincts rather than high musical attainments, but already in Rigoletto, II Trovatore, and La Traviata,

1 See post, p. 254. a Verdi was born in 1814, he died in 1901.

" Macbeth, after Shakegpeare, had been produced in 1847 ; 1 due Foscari in 1844 and II Corsaro in 1848, both after Byron ; Giovanna d'Arco in 1845 ; I Masnadieri (Die Rduber) in 1847; Luisa Miller (Cabale mid Liebe} in 1849 an^ Don Carlos in 1867, after Schiller; Les Vtpres Siciliennes, in imitation of Meyerbeer's Les Huguenots, in 1855 ; Un Batto in maschera in 1859— the libretto is identical with that of Auber's Le Bal masque— and La Forza del destino in 1862, the latter after a Spanish play by the Due de Rivas.

ITALIAN OPERA 61

the accomplished musician is evident, whilst the marked racial and theatrical qualities remain unimpaired. Riffoletto, the chef-d'oeuvre of Verdi's first period his seventeenth opera was at least equalled in popularity by the success of II Trovatore and La Traviata, both of which were brought out in 1855. The subjects as well as the construction of the libretti of these operas are each in its way typical of the curious con- ception of operatic romance that prevailed among fashionable circles in Paris, Venice, Rome, and Milan about the middle of the nineteenth century. The music runs on the cosmopolitan lines which Verdi affected in those emotional days. His musical gifts and predilections were exactly fitted to reflect the ex- travagant tragical situations of Ernani and Rigoletto, the lay- figure and gipsies, monks, knights, and ladies of II Trovatore, the story of a consumptive courtesan and her ultimate purification by love and death in La Traviata1. The great vogue of / Lombardi, Nabucodonosor, Ernani, and especially Rigoletto, with its appeal to the revolutionary spirit, is partly explained by the political circumstances of the time. The romantic movement, which elsewhere on the continent told in favour of reaction, became in Italy the handmaid of revolution. By a curious accident the letters of Verdi's name were adopted by the Italian populace as an emblem connected with the liberation of Italy ' Viva Verdi ! ' really meant £ Viva Vittorio Emanuele Re D5 Italia ! J And it is certain that Verdi himself, when dealing with operatic situations that happen to have a problematic social significance situations such as could be turned to account for political purposes was very much in earnest, and consciously made the most of them. Of his perfect sincerity in such matters there can be no doubt witness the fine scena in the first act of Rigoletto, 'Pari siamo. lo ho la lingua; egli ha il

1 The libretto of La Traviata is based on Dumas fils' La Dame aux Cornelias. 1 Musical art,' it has been well said in this connexion, 'cannot depict the repulsive without some glimmer of beauty. It penetrates and idealizes the elements of corruption, and transmutes the terrible reality of the drama into a melancholy dream.' Hanslick, Die moderne Oper, i. 233.

62

THE ROMANTIC PERIOD

pugnale/ and other such semi-personal utterances. Rigoletto was originally called La Maledizione. The censorship objected to King Francis I being cursed by a court fool, as is the case in Victor Hugo's play. Eventually the monarch was turned into a duke of Mantua, and the opera appeared under the fool's name Rigoletto.

Musically it is interesting to compare the famous quartet in Rigoletto with its model, the quartet in Bellini's I Puritani, and to note the advance the former shows in the direction of la musica caratteristica, towards which Verdi came to lean more and more in course of time. The change for the better with regard to the independence of the vocal part-writing, the individualization of the characters, together with a wider range of harmony, may be slight in this particular case, but it is remarkable all the same. In so far as the outlines of form are concerned the two quartets are closely alike in the cast of the melody, the changes of harmonic centres, the culmination of vocal effects towards the close, and in the contrivance of the coda. The general resemblance may be easily detected by a comparison of the following quotations :

Largo

^y ^ p P -Nl

^ tr-1 S

G£fi j'^g q r c r "~erTfT'-|-f Vr

'r r -.-HK

-_j L- H

^^

-g+^-H

* u

ITALIAN OPERA 63

Verdi at first derived his manner and style from Mercadante and Bellini; then he felt the influence of Donizetti (Lucrezia Borgia, Don Pasquale), later on, and in an increasing degree, that of Meyerbeer. The role of Azucena in II Trovatore, for instance, is but that of Fides from Le Prophete translated into Romany. Then came the influence of Halevy (La Juive). Finally, he was swayed in a curious manner by Wagner (Tannhauser, Lohengrin).

During the first twenty years of his career Verdi produced twenty-three operas. In the following period of twenty-seven years, only three Aida, Otello, Faktaff. Fully up to middle age the trend of his genius was evidently more inclined to theatrical than to musical ideals. Exceptionally gifted for the naturalistic expression of passion it was partly a gift for finding emotional vocal melodies in the manner of Bellini, partly in that of Meyerbeer for combining such melodies with sharply accentuated rhythmical figures he was able, by the aid of very simple choral and orchestral devices, to produce telling theatrical effects. Thus his music, as it were by fits and starts, is now operatically effective, now sentimentally weak or vulgar, now blatant with theatrical pathos, now genuinely original and dramatic.

At the age of 57 Verdi wrote Aida for Cairo (1871-2), and produced it at the European centres as a pendant to Meyer- beer's UAfricaine. Then in his seventy-third year came Otello, first performed at La Scala, Milan, in 1887, the book by Arrigo Bo'ito after Shakespeare ; and, finally, Falstaff, first performed in 1893, the book again by Bo'ito after the Merry Wives of Windsor.

Verdi's efforts to keep pace with the movement towards characterization in opera became more and more apparent from the date of Rigoletto onwards, and gradually brought about changes in his manner and considerable improvement in his technique. Aida contains much that is mere pageant music or picturesque illustration of scenic evolutions, but also several

64 THE ROMANTIC PERIOD

scenes of emotional power and impressiveness. Otello very distinctly and, to a greater extent still, Falstajf shows a change in method. The master seems to have reversed his artistic direction, and to have adopted a more intellectual speculative * Wagnerian ' gait. That Verdi did assimilate some minor points in the method of Wagner is certain. Yet any one familiar with the scores of both masters will readily distinguish between the perhaps equally futile designations ' Maestro ' and * Meister.' In Verdi there is hardly a trace of Wagner's peculiar orchestral polyphony or variety and richness of tone- colour. Verdi, it is true, employs the representative phrase, the f Leitmotiv/ but not in Wagner's many-sided, contrapuntal way, and only in so far as the device is helpful in clearing up special points in the action. Moreover the character of Verdi's representative phrases is as distinctly his own, that is to say Italian, as that of Wagner is Teutonic.

Comparison of Rossini's Tragedia lirica, Otello (1816) with the Otello of Verdi throws a vivid light on the changes brought about by the spirit of romanticism and the example of Wagner. Rossini's Tragedia has many traces of the older opera seria. Verdi's is an opera caratteristica of a very pronounced type. With Rossini there is but a faint shadowof Shakespeare's tragedy, and the music might, for the most part, be sung in solfeggio. With Verdi the librettist adheres closely to Shakespeare's text, whilst the composer strives to develop his powers of dramatic realization, and to find proper accents, passionate or tragic or comical, to tally with the characters and situations. It is a matter of give-and-take between dramatist and musician ; Boito's book inspires the composer to a new mode of utterance, and a good performance of Otello leaves the impression of a tragic drama dissolved in music. In certain particulars, however, the older Otello holds its own, and the points are not altogether in favour of Verdi and la musica caratteristica. In Desdemona's ' willow ' song, for example, Rossini's melody is simple and beautiful; Verdi's slightly bizarre, though most

ITALIAN OPERA 65

effective in its place l. A note of warning, ' Wagner in the air/ was uttered in Italy after the production of Otello ; it became a cry of alarm after Falstaff'^ but Verdi even here remains true to himself and the traditions of his country. In no case has he traversed the Italian doctrine that vocal melody of some sort is the main concern, even when the music takes the place of rapid dialogue or passionate soliloquy. His melody, in Falstaff, is more inclined towards recitative than cantilena ; and its power, though its presence is felt throughout, is but rarely con- densed to actual song. In one instance only Fenton's little arioso ( Bocca bacciata ' there is something like the lilt of the younger Verdi's tunes. For the most part the music has the character of a lively conversation, with here and there some bits of energetic declamation or emotional cantabile. Bo'ito's libretto to the Merry Wives would be quite effective as a play without music. Few lines in it appear to be written with a direct view to formal solo or concerted pieces. But whenever Verdi chooses to make use of an opportunity to write ' in form* as for instance in the ensemble ( fe un ribaldo, un furbo, un ladro/ in the second part of Act I, the duets f Labbra di foco/ and ' Dal labbro il canto estasiato vola/ Acts II and III, and the fugal finale 'Tutto il mondo e burla/— the result proves fine in effect, distinctly artistic and perspicuous, though less striking and impressive than earlier pieces, excepting, of course, the comical vocal fugue at the end, which is perfect in its way.

Falstajf is perhaps less remarkable than Otello in point of musical invention. But the sardonic vivacity of its humour is surprising. There is not a page in Falstajf which does not exhibit touches of musical as well as verbal wit of the most entrancing kind. 'Such scenes as the assignation made by Falstaff with Dame Quickly, with its playful iteration of the notes associated with the words " Dalle due alle tre," the whole

1 There is in Verdi's setting a touch of artificiality, recalling the song ' Le roi de ThultS ' in Berlioz' Damnation de Fausf. The prominent part played by the corno inglese, i. e. bass oboe, also recalls Berlioz.

DANNREUTHER

66 THE ROMANTIC PERIOD

scene of the buck-basket, the fat knight's soliloquy after his immersion in the Thames, and above all the working up of the final scene, are monuments of humorous power V

In Aida, Otelloy and Falstaff the robust naturalistic expres- sion of passion, so characteristic of the early operas, appears less crude, the declamation less violent and more carefully balanced ; the outlines of the melody more sinuous, the harmony and modulation richer and bolder, the instrumentation less coarse and commonplace, whilst the telling quality of the music, the sum total of its effect in combination with proper stage manage- ment (provided always that the later operas are performed at theatres of more reasonable dimensions than La Scala of Milan or Covent Garden), can hardly be said to fall short of what it was at the outset, that is to say, in such popular works as Rigoletto, II Trovatore, and La Traviata. Allowing for the curious cosmopolitanism in his choice of subjects and the eclecticism in their musical treatment, the occasional crudity and frequent vacillation in his style, together with the not un- common cases of perhaps unconscious borrowing of other men's devices, it is none the less evident that the operas of Verdi represent a forward movement in several branches of the musico- histrionic art, and that his music, taken altogether, is the result of self-developed Italianism, and the expression of a strong Italian individuality.

1 Mr. J. A. Fuller Maitland.

CHAPTER V

THE FURTHER DEVELOPMENT OF ROMANTIC OPERA

AFTER Weber and Marschner the progress of operatic art did not move so rapidly in Germany as might have been expected. It is true the spirit of romantic opera had already entered the mind of Wagner, to await there its most complete embodiment in Lohengrin. But it was not till past the middle of the century that any of Wagner's works came to be an appreciable factor in musical life 1.

In the meantime the operas of lesser musicians, such as Konradin Kreutzer, Reissiger, Lortzing, Flotow and Nicolai, . gained some degree of popular favour. It is enough to mention Kreutzer's Nachtlager von Granada (i 834)2, Reissiger's Felsen- miihle, Flotow's Martha (1847), Lortzingjs( 1801-51) Czar und Zimmermann (1837), Der Wildschutz (1842), and Nicolai's Die lustigen Weiber von Windsor (1849). Lortzing, an experienced actor singer and conductor, wrote his own libretti, which, with considerable skill, he adapted from already existing plays. His practical knowledge of stage effect to some extent made up for the rather commonplace character of his music. Compared with the power of Marschner's work, Lortzing's is but that of a gifted dilettante, who was able to make good use of his experience in the theatre. With the aid of his lively tunes and his actor's chic, Lortzing managed somehow to express the provincial humours of the period (about 1840-50) in a manner

1 The first performance of Lohengrin took place under Liszt, at Weimar, in 1850.

2 Konradin Kreutzer, 1780-1849. Some of bis choral songs for male voices, such as ' Die Kapelle,' ' Der Tag des Herrn,' ' Marznacht,' are models in their way, and had an immense vogue.

F 2

68 THE ROMANTIC PERIOD

sufficiently artistic to ensure a widespread popularity in Northern Germany. His successor in popular favour was Flotow, another quasi-amateur, whose Martha made the round of Europe ; but Flotow's melody is at once more commonplace and more sentimental, and the neatness of style, to which his vogue was mainly due, does little more than borrow a few epigrams from the current phraseology of French opera comique.

Nicolai, an excellent all-round musician, conductor and singing-master, who had produced operas in Italy and church music in Germany, put forth the ripe fruits of his experience in Die lustigen Weiber von Windsor, e komisch-phantastische Oper in 3 Akten/ which was brought out shortly before his death, at Berlin in 1849, and was received with acclamation. The bright and spontaneous good humour that pervades the music chimes with the gaiety of Shakespeare's play. The airs and rapid conversational ensemble pieces are connected by short snatches of dialogue, after the manner of Auber's Le Maqon. Nicolai makes no attempt at close characterization, such as Weber, for instance, achieved in the duet between Agathe and Aennchen in Der Freischiitz ; his Merry Wives, indeed, might exchange their tunes throughout, as they actually do, when, in an amusing duet, they compare FalstafPs letters. But the humours of Falstaff, Mrs. Ford, Dr. Caius, are kept sufficiently distinct. The finales of both the first and the second acts are well contrived and effective. The declamation is good and the treatment of the voice admirable. But it is not till the third act, during the fairy scene in Windsor Forest, that the limitations of Nicolai's talent become apparent. Here his routine proves insufficient, and the music, pretty and brisk as it is, lacks the touch of poesy.

It was said that Lesueur, Berlioz' master at the Paris Conservatoire, put so much drainatic life into his church music that there was none left for his operas. In an analogous way it may be said that Berlioz' dramatic vein was nearly exhausted

DEVELOPMENT OF ROMANTIC OPERA 69

by his symphonies when he began to write Les Troyens, which he intended to be his magnum opus for the stage.

The words and music to the ( Poeme lyrique en deux parties ' Les Troyens: I. La Prise de Troie, opera en trois actes, II. Les Troyens a Carthage, opera en cinq actes avec un prologue, were completed in 1858. An opera comique, Beatrice et Benedict, after Shakespeare's Much Ado about Nothing, followed in 1 86 a. As at first intended, Les Troyens, like any other grand opera, was to occupy one evening only; and Beatrice et Benedict was planned as a musical comedy, a 'Lever de rideau,' in one act. Both works were repeatedly revised, unduly expanded, and spoilt. ( On peut dire de lui (Berlioz), comme de son heroique homonyme, qu'il a peri sur les murs de Troie V Les Troyens, although intended as an equivalent to Wagner's Der Ring des Nibelungen, is simply an opera upon an unusually large scale, bearing no sort of resemblance to the Wagnerian music-drama2. From the stage-manager's point of view, the disposition of Les Troyens (Berlioz' own) is unsatisfactory, and there are but few cases where the music supports and furthers the action or makes amends for defects in dramatic construction or scenic arrangements. Even the dance tunes, such as those of the Combat de geste 3, the Pantomime avec choeurs in La Prise de Troie, the Pantomime in the second act, the airs de danse in the third act of Les Troyens a Carthage, have a touch of artificiality. La Prise de Troie is a ponderous prelude to the main work. In the opera proper, Les Troyens a Carthage, the dramatic interest lies solely in the departure of Aeneas, and in the scene of Dido's death. The style of some of the airs and scenes and of the short choruses recalls the manner sometimes of Gluck, sometimes of Spontini. The best pieces in La Prise de Troie are a spirited choral ensemble,

1 Gounod, Preface to Berlioz' Lettres intimes.

2 Part I, La, Prise de Troie, was inadequately performed in Paris at the Theatre Lyrique, and withdrawn after twenty repetitions. A German version of both parts of Les Troyens was produced at Carlsruhe in 1 897.

3 Compare Wagner's Rienzi, Act II.

yo THE ROMANTIC PERIOD

a fine Octet and chorus, 'Chatiment effroyable, mysterieuse horreur,' a touching prayer, 'Puissante Cybele,' for female voices, Cassandra's air (in the manner of Spontini) ( Malheureux roi,' and a 'Marche tyrienne,' which occurs again in Les Troy ens a Carthage, l dans le mode triomphal ' and ' dans le mode triste.' In Les Troyens a Carthage, the more remarkable numbers are the introductory Lamento and the ensembles belonging to Act I, including the fine Chant national ' Gloire a Didon,' the ambitious and curious Chasse royale et Orage, called a symphonic descriptive avec choeurs, which forms the second act, the Quintet ' O pudeur,' and the Septet 'Tout n'est que paix et charme autour de nous,5 the love duet, Dido and Aeneas (a rifacimento of the moonlight scene in The Merchant of Venice), the delicate 1 Chanson du jeune matelot,' the noble fragment of a duet, Act V, ' Va, ma soeur, 1'implorer,' Aeneas' air f Ah, quand viendra 1'instant supreme?' Dido's solo ( Je vais mourir,' and f^nee, Enee, ah, mon ame te suit,' and some portions of the picturesque choral music which illustrates the funeral ceremonies.

Beatrice et Benedict is a light opera, the texture of which, shot with strands of comedy and romance, was woven in the loom of Benvenuto Cellini. The influence of the earlier and more robust work is noticeable in many solo numbers and ensembles ; such for instance as Benedict's rondo ( Ah ! je vais 1'aimer,' the trio ( Me marier, Dieu me pardonne,' and the whimsical duet 'L'amour est un flambeau.' One of the strangest numbers is an '^Ipithalame grotesque' written in Berlioz' rather heavy-handed counterpoint, sung, according to the stage-direction, in tones of extravagant emphasis, and repeated with a farcical accompaniment of oboes and bassoons. Like the 'Amen' chorus in Faust it is an obvious satire on academic methods, the purport of which hardly atones for its ugliness. But the opera contains two numbers which are fresh and spontaneous : the Sicilienne which serves as entr'acte, and the duet ' Vous soupirez, Madame,' which in Act I occupies the place of the finale. The latter in particular is a shapely and

DEVELOPMENT OF ROMANTIC OPERA 71

beautiful composition, which in style and feeling forms a worthy pendant to the shepherd's chorus in the Enfance du Christ.

The music of Charles Gounod1, an eclectic in the good sense of the word, shows traces of Mozart, Weber, Meyerbeer, Halevy, Auber, Schumann, and early Wagner. But Gounod so completely absorbed and assimilated the results of a close study of these masters as to place himself in a position to produce something that is new in effect if not new in substance. The distinctive personal note of his music consists in the expression of tender sentiment and longing as in certain parts of Faust et Marguerite and Romeo et Juliette. It is worthy of remark that his lovers, Faust and Marguerite, Romeo and Juliette, Vincent and Mireille, Philemon and Baucis, all seem to make love to the same tune. Gounod rarely reaches the heights of passion. He contrives, however, to reflect the changes of light emotion in Mireille for instance, or in Philemon et Baucis, andZ/e Medecin malgre lui. Based upon just declamation, as it was practised in the days of Lully and Gluck, the accents of Gounod's melody never contradict those of the words. The refinement of his style is peculiarly French, and he shows a consummate knowledge of orchestration for stage purposes. His first opera, Sapho (1851), was a failure, and so was La Nonne sanglante, in 1 854. Success came with Le Medecin malgre lui, from Moliere's comedy, which appeared in 1858, and particularly in 1859 with Faust et Marguerite, from the first part of Goethe's Faust, which, after a short period of suspense, brought him fame and position. Gounod professed to hold his own religious music the oratorios La Redemption (1882), Mors et Vita (1885), and sundry songs of a pious character in higher esteem than his theatrical works ; yet his talent was essentially imitative, histrionic, and his best work belongs to the operatic stage 2.

> 1818-93.

a After Faust et Marguerite Gounod produced Philemon et Baucis (1860), and La Reine de Saba (1862), both at the Grand-0pe"ra ; Mireilk (1864), and Romeo et Juliette

73 THE ROMANTIC PERIOD

Ambroise Thomas l, an eclectic like Gounod, but of a somewhat weaker type, and more inclined towards the methods of Meyer- beer, made his mark in 1866 with Mignon, after Goethe's Wilhelm Meister\ Hamlet followed in 1868, and Franqoise de Rimini in 1882. But the next French opera of world- wide fame after Gounod's Faust et Marguerite was Bizet's Carmen, produced at Paris in 1 875. Bizet2 had already won some reputa- tion with The Fair Maid of Perth (1867), Djamileh (1872), and with his brilliant incidental music to Alphonse Daudet's Arlesienne (1872), but it is into this, the latest and greatest of his compositions, that he put his best work. Despite a touch of diablerie which Georges Bizet's music shares with Merimee's story, or perhaps because of that very touch, Carmen made a distinct and strong impression. The public was fascinated by the sensuous and picturesque dances and songs, and the murder in the Bull Ring ; musicians were fascinated by the novelty of the Spanish gipsy measures, the subtle cleverness of the melodic, rhythmical, and harmonic devices, and by the strange realistic effects of instrumental local colour. In Carmen (as in Weber's Preciosa) the music is always in accord with the action, yet never crude or vulgar. Every note sung by the chief personages seems to belong to them by natural right. The music is singularly free from reminiscences of the classical composers ; throughout it savours of Spanish and Proven9al folk-songs and dances. Its beauties are too well known to need quotation, yet we may mention the Seguedille et Duo, the Habanera, the Chanson Boheme, and the dance with which Carmen fascinates at once her lover and her audience. Among composers of lesser account Edouard Lalo (1823-92) produced a con- siderable impression with his opera comique Le Roi a" Ys, and Leo Delibes (1836-91) with Le Roi Va dit and Lakme.

(1867), both at the The"atre-Lyrique ; Cinq Mars (1877) ; Polyeucte (1878) ; Le Tribut de Zamora (1881) again at the Grand-Op&a.

1 1811-96.

a 1838-75.

DEVELOPMENT OF ROMANTIC OPERA 73

For a decade or so, during the Second Empire (1860-70), Jacques Offenbach l influenced, and in a manner controlled, public taste in France and elsewhere on the Continent, in a manner disproportionate to the musical value of his productions. Offenbach's opera bouffe, which eclipsed the genuine opera comique, was a perfect echo of the cynical caprice of the third Napoleon's time theatrical extravagance paired with farce—- satire with vulgarity a theatrical journal pour rire, the stress laid upon the comic licence of the stage business and the licence emphasized by the music of a Meyerbeer en miniature.

Yet Offenbach had an individual gift of melody, his harmony at times was refined, his instrumentation often ingenious, though the means were simple. From the comedian's point of view the facility and rapidity of his invention was remarkable. His satirical vein never failed. Perhaps the operettas Orphee aux enfers and La Chanson de Fortunio represented him to best advantage. In the once celebrated La Grande-Duchesse de Gerolstein he appears at his worst. He returned to his first manner before his death in Les Contes dy Hoffmann.

Two German comic operas demand notice here : Peter Cornelius' 2 Der Bar bier von Bagdad, produced at Weimar in 1858, and Hermann Goetz'3 Der Widerspenstigen Zdhmung, after Shakespeare's Taming of the Shrew, produced at Mannheim in 1875. Cornelius' work shows the influence of Berlioz' Benvenuto Cellini; Goetz' of Schumann generally, and, far away, of Wagner's Meister 'singer. Both operas are genre pictures full of delicate details, but at times the music is too frail for stage effect. Goetz was not a man of theatrical instincts and had hardly come in contact with the theatre when he began to compose his opera. He had previously produced chamber music, pianoforte pieces, and songs. His sense of comedy lacked power, and, though he never actually lost sight of the stage action, yet, like his model Schumann, he deliberately chose to lay his chief stress on musical detail. After the performance at 1 1822-80. * 1830-74. 3 1843-76.

74 THE ROMANTIC PERIOD

Mannheim, Der Wider spenstig en Zdhmung was warmly received at Vienna in 1875, and it is still occasionally given.

Wagner's Meistersinger apart, the book of Cornelius' Barbier von Bagdad is, from a literary point of view, far and away the best comic libretto in German. There is nothing to approach it in any other language, unless it be Boito's libretto to Verdi's Falstaff— which has the great advantage of livelier stage action. In Der Barbier von Bagdad, Cornelius sometimes employs parlando recitative, and even the patter song, upon an elaborated orchestral background and with astonishing result as for instance in the fifth scene : f O wiisstest du, Verehrter, was ich fur ein Gelehrter/ The various movements are formally com- plete, as is the case in Berlioz' Benvenuto, and Cornelius tried hard to resist Wagnerian influences to which, however, in his last opera, Der Cid, and the unfinished Gunlb'd he finally succumbed. Compared with Goetz, Cornelius' originality and fertility of invention is very striking.

Opera, more or less on the lines of racial and national characteristics, began in Russia with Glinka's 1 La_ Vie pour le Tsar (1836), Russian et Ludmilla (1842) (the text after^ar romance in verse by Pushkin), and Serov's Judith, produced in 1863; in Poland with Moniuszko's Halka (1858); in Bohemia with Smetana's Prodana nevesta (The Bartered Bride) (1866); in Hungary with Erkel's Hunyady Lazlo (1844).

The operas of Glinka and Smetana are distinguished by their musical value apart from theirposition as national representatives. Glinka, Russian by birth, chose to adopt the characteristics of North-eastern European folk-song, both in the vocal and instru- mental part of his operas. In La Vie pour le Tsar, Russian and Polish elements are combined 2. In his incidental music to a

1 Glinka was born in 1804, he died in 1857. His memoires were published in 1870.

2 In its strongest moments the music to La Vie pour le Tsar appears as a kind of scene painting, very bold and effective. That of Russian et Ludmilla is of a fantastic , semi-oriental character, and differs so greatly from the first opera that one might guess at another composer. La Vie pour le Tsar has been given hundreds of times, and its popularity shows no signs of abating.

DEVELOPMENT OF ROMANTIC OPERA

75

tragedy, Le Prince Kholmsky, by Koukolnik, there are remini- scences of Hebrew melodies. His technique, vocal and instru- mental, is that of a master with a faint touch of dilettantism, trained in Italy and Germany. The melody, apart from Russian influence, is reminiscent of South-western Europe ; the orchestra- tion, too, has a Southern touch French, Italian, or Spanish it is always simple, often very effective, but occasionally thin. A song such as the following, from La Vie pour le Tsar, represents him well.

Adagio

Con molta anima &-

76

THE ROMANTIC PERIOD

Certain orchestral pieces of Glinka's deserve mention : Jota Aragonese, described as a ' Capriccio brillante/ a Fantaisie sur des themes espagnols, ' Souvenir d'une nuit d'ete a Madrid/ and La Kamarinskaja, which last is the true ancestor of Russian instrumental music. It consists of an orchestral fantasia on two Russian folk-tunes, a wedding song and a dancing song, rich in novel contrapuntal devices and orchestral contrivances of con- siderable originality 1.

Alexander Serov 2, critic, librettist, amateur composer, in early life came under the influence of Glinka, and later pn under that of Wagner. His first opera, Judith, hesitating and inadequate in point of style, though written when he was upwards of forty, is

| National elements, Russian, Spanish, Italian, Polish, are always present in

Uinka's songs— some eighty in number. He accepts existing dance-rhythms and

takes no pains to modify or improve upon them, as Chopin did in his Mazurkas and

Polonaises. The list of Glinka's works includes pianoforte pieces, chamber music

vocal quartets, choruses.

DEVELOPMENT OF ROMANTIC OPERA 77

laid out on the lines of Meyerbeer and scored in the manner of Wagner's Rienzi. It was produced in 1863 and met with an extraordinary and lasting popular success in Russia. His second attempt at opera, Rogneda, contains, according to Tchaikovsky, ' certain oases in the desert, of which the music will pass muster.' In 1867 Serov tried something faintly resembling a Wagnerian music drama in Russian, on a Russian subject, and with the aid of Russian folk-tunes, thus following in the wake of Glinka. He founded his libretto on Ostrovsky*s rather sordid play, The Power of Evil, but did not live to finish the music, of which the orchestration was completed by Soloviev an arrangement that still keeps the stage 1.

Smetana's most famous opera, The Bartered Bride, like Nicolai's Merry Wives of Windsor, is an enlarged Singspiel of the family of Mozart. There are traces and touches of Beethoven's Fidelio and Cherubini's Les Deux Journees. The music throughout is fresh and bright, the melody refined, the ensembles masterly, and there is a great deal of amusing inter- play and episode. Six other operas by Smetana have been performed. His cycle of Six Poemes symphoniques, entitled My Country, shows considerable cleverness and some originality.

Moniuszko's Halka is the favourite Polish opera. The original two acts were first performed at Wilna in 1854. With two further acts interpolated, it was heard at Warsaw in 1858, and repeated there for the five hundredth time in 1900. The interpolations weaken the total effect, but the charm of the tunes keeps the work afloat. It is admirably written for the voices and admirably scored. Moniuszko put forth a total of fifteen Polish operas, several Masses, cantatas, and a number of songs.

The achievements of English composers during the first half of the nineteenth century, the instrumental music of Sterndale

1 Liszt {Letters to Madame de Wittgenstein, Hi. p. 38) mentions some candid advice on the subject of Serov's opera Judith : ' . . . que je lui ai conseille de traiter comme Judith avaitfait d'Holopherne ! Imagine: que Serov se figure qu'il est le Wagner

7 8 THE ROMANTIC PERIOD

Bennett and the church music of the Wesleys excepted, were by no means imposing. Judged by quality, the operas of Balfe and Wallace, such as Balfe's still popular Bohemian Girl (1843), or Wallace's Maritana (1845) an(^ Lurline (1860), are not in- ferior to the lighter operatic ware produced in France and Italy for the delectation of middle-class audiences but, then as now, musicians must have found them weak and insufficient. Both Balfe and Wallace had a facile gift of melody. Expert vocalists and instrumentalists (Balfe was a famous singer, and Wallace has the reputation of having been a virtuoso upon the violin), they understood the requirements of popular operatic performers, and were sufficiently experienced as musicians to handle a small orchestra and a small chorus with ease and skill.

The operettas of Burnand and Gilbert, the librettists, and Sullivan the composer from Box and Cox, and Trial by Jury, to The Gondoliers j The Yeomen of the Guard, and The Mikado are the leading English contributions to the devolution of opera comique to the opera bouffe. These amusing pieces were hailed with delight by all English-speaking people, chiefly on account of the fresh air and healthy laughter that pervades their humorous extravagance. When Sullivan appeared with his music to Shakespeare's Tempest, Balfe and Wallace were near the end of their careers. Sullivan, a gifted and accom- plished musician, had acquired the mannerisms of Mendelssohn, and felt the simple charm of Schubert ; he came by degrees to emulate the savoir-faire of Auber, and to approach the satire of Offenbach. His first essays in operetta, Box and Cox (1867), and Trial by Jury (1875) (both libretti by Burnand), were avowedly due to Offenbach's example, but from the outset they were free from the grotesque eccentricities of their French models. There is a distinct personal note about Sullivan's lighter operatic tunes ; amiable, tender/ slightly ironical, always graceful and artistic. His lucid sense of humour stood him in good stead. In agreement with the Merry Andrew in the prologue to Goethe's Faust, he seems to have said to himself :

DEVELOPMENT OF ROMANTIC OPERA .79

* Posterity ! Don't name the word to me ! if / should choose to teach posterity, where would you get contemporary fun ? ' Technically Sullivan was a master all round; a good vocalist, and well acquainted with every instrument used in the orchestra, or in military bands. In the matter of orchestration he was completely at his ease, and at once found the simplest and best means of attaining his end. His melodic vein flowed readily and copiously ; it was never deep nor passionate, yet at its best sufficiently capable of expressing emotion. His declamation was easy and natural : the words and the tune seemed to spring from the same source. His own favourite among the operettas was The Yeomen of the Guard (1888), probably because of the touching dramatic story. Ivanhoe, a serious effort in opera, can hardly be said to have fallen flat since there were one hundred and six consecutive performances in London only, but it failed to compel assent.

CHAPTER VI

OVERTURES AND SYMPHONIES

WEBER'S music, both vocal and instrumental, owes its character indirectly to the romantic nature of the legends and scenes which he employed as a basis of his works. His innate tendency which in later years became a conscious aim was to achieve a complete rendering of the emotional essence of a dramatic situation in terms of music, be it instrumental or vocal. Thus his Concertstiick is essentially a f Dramatic Con- certo/ and the three overtures to his mature operas are the finest Dramatic Fantasias extant, reproducing, in a concentrated form, the sentiment of the scenes and situations which are to ensue.

Apart from the stage, romantic effects in instrumental music arose from a desire to reproduce impressions derived either from imaginative literature, or directly from natural phenomena to express the prevailing emotion, the mood of some particu- lar poem or story, or of some particular aspect of nature.

The latter maybe illustrated by Mendelssohn's1 famous over- ture The Hebrides (Fingal), written in 1830.

The music here conveys a sense of distance, of solitude, and of moving water. Further on, there are suggestions and effects as of storm, or of wind-shaken surges, of shifting gleam and cloud, of the sea-mew's plaintive cry, and the shimmer of north- ern seas. Most hearers will confess to having received some such impressions-— even if they do not happen to possess Heine's or Schumann's gift of evolving pictures from musical sounds, 1 Mendelssohn was born in 1809, he died in 1847.

OVERTURES AND SYMPHONIES

81

and are not aware that the work was conceived amid the rugged scenery of Staffa and the adjacent islands.

Allegro noderato

Flauti.

Oboi.

Clarinetti iu A.

Fagotti.

Corni in D.

Trombe in D.

Timpani.

Violino I. /

Violino II.

Viola. •<

Violoncello.

Basso.

ftp***:

eg u

Allegro moderate

&E

i/ p

-^P 1 _LJ rl -••'Ti ^^ --a-!^

^ jirr^IFh^cj^

DANSUECTHIK

THE ROMANTIC PERIOD

Flanti. f ^*

Oboi.

Clarinetti in A.

Fagotti.

Corn! in D.

Trombe in D.

Timpani.

Violino I.

Violino II.

Viola.

Violoncello.

Basso.

SiJi

PP

ii

1

I

&c.

perdendosi

F fCTl? *f i^?P?* 1*1"% !.••-' -• ^~^

^^|LB T B L&Trt:^^^

OVERTURES AND SYMPHONIES

' Mendelssohn,' Wagner said l, ' was a landscape painter of the first order/ and this overture is his masterpiece. Note the extraordinary beauty of the passage where the oboe rises above the other instruments with a wail as of sea winds over the sea :

Oboi.

Viol in 1 1 1.

Violino II.

Viola.

Violoncello.

Basso.

m

P=

»/

^f:

dim.

g^-^-J^-g-r--^

* /

Solo.

dim.

iE£

^

dim.

w

^

--£.-=&-

=p=i=

dim,

pp

ff

Grove's Dictionary of Music, article Wagner. G 1

84 THE ROMANTIC PERIOD

It must not, however, be supposed that this piece is merely made up of a series of more or less picturesque devices of orchestration, or that it is in any way meant to be an example of programme music. It is pure instrumental music on musical lines *.

Its originality consists in the nature of the fresh and charac- teristic subjects, especially the first subject and in the masterly treatment of the orchestra. Mendelssohn's other overtures A Midsummer Night* s Dream (1826), Meeresstille und gliickliche Fahrt (1828), and Zum Marchen von der schonen Melusine (1833) are to a certain extent reflections of literature. In Meeresstille the composer is guided by the poet's order of ideas Goethe's pictures of a calm at sea and a prosperous voyage. In Melusinaj Tieck's version of the old French story prompts the music, which seems to depict, alternately, the beauteous water-nymph turned human, her pathetic distress on discovery, and the return to her former condition. Here again, the music is meant to tell its own tale in purely musical terms and on purely musical lines. The title contains all that the composer deemed needful to guide the audience 2.

Mendelssohn's nearest approach to the role of musical illus- trator is contained in the Scherzo of an early work, the octet for strings (1825), and the instrumental introduction to the cantata, Die erste Walpurgisnacht. Originally the Scherzo in the octet was headed by a stanzg from the f Intermezzo ' in Goethe's

Faust :

Orchester pianissimo.

Wolkenzug und Nebelflor Floating cloud and trailing mist

Erhellen sich von oben ; Bright'ning o'er us hover ;

Luft im Laub, und Wind im Kohr, Airs stir the brake, the rushes shake,

Und alles ist zerstoben. And all their pomp is over.

The materials are arranged in accordance with the usual scheme of harmonic distribution— the outlines of the sonata form : I. Exposition, first subject in B minor, second subject in the relative major D. II. The working-out section, wherein fresh harmonic centres are touched upon. III. Recapitulation, with both subjects in the principal key.

2 Mendelssohn's aversion to anything resembling a detailed programme came out very clearly in his sarcastic answer to a question as to the « meaning' of Metusina : ' Hm does my music hint at a mesalliance ? '

OVERTURES AND SYMPHONIES 85

The Introduction to Die erste Walpurgisnacht describes : i. eDas schlechte Wetter/ and 2. ' Der Uebergang zum Friihling.'

The germs of the overture to A Midsummer Night's Dream can be traced to a reading of Shakespeare's play in Schlegel and Tieck's translation, when Mendelssohn was still in his teens. The score was finished in 1826, before he had completed his eighteenth year. It is easy to see how impressions of certain scenes took shape in the young man's mind, and how, with Weber's overtures for a model, he fused and welded them together so as to form a consistent whole. The music seems to convey sugges- tions of Titania asleep, revels of fairies, a dance of clowns with the bray of ( Bottom translated,' the lovers' hide and seek, and the nuptial festivities. Technically the overture is a carefully planned and carefully finished piece of work. The orchestra- tion is remarkable for its clearness and practical efficiency. Every fantastic effect is produced with perfect ease. The spirit, however, is Weber's from the first note to the last, and in this important respect this overture is inferior to The Hebrides, and perhaps even to Melusina. The rest of the Midsummer Night's Dream music is of later date. In 1843 Mendelssohn was asked to write incidental pieces for a performance of the comedy at Berlin. The orchestral numbers Scherzo, Elfinmarsch, Inter- mezzo (Hermione seeking Lysander), Notturno, Wedding march, a comic Funeral march and a Clowns' dance show the master at his very best.

There is no need to dwell on the other overtures Ruy Bias, Athalie, the overture for a Military band, published as Op. 24, and the so-called Trumpet overture in C for, apart from the technical merits which they possess in common with all Mendels- sohn's orchestral pieces, they have not sufficient spontaneity and weight to make their gradual disappearance a matter for much regret.

Schumann wrote a number of overtures, of which Manfred (1848) and Genoveva (1847-8) are the only two that really represent his powers. The rest, such as Faust, Julius Caesar,

86

THE ROMANTIC PERIOD

Die Braut von Messina, Hermann und Dorothea, belong to that unfortunate period of feverish productiveness which preceded his final collapse (1850-4) ; all more or less ineffectual, they leave an impression of weariness and vain effort.

The overture to Manfred stands forth among Schumann's works even more conspicuously than The Hebrides among Mendelssohn's. In some degree Schumann's tone-poem sug- gests the atmosphere of Byron's Manfred, though it can hardly be called Byronic in spirit ; a generic name, such as Brahms chose for his * Tragic ' overture, would suit it better. With the Manfred overture, a piece sombre in tone, deeply felt and very personal, Schumann came nearer to the inner fane of music than Mendelssohn with The Hebrides,orwith Meeresstille or Melusina. Though perfectly original in matter and sentiment, the method of construction employed in Manfred recalls that of Beethoven's third period. The subjects are strong and novel in style. The passionate melody, the vehement rhythm, the keen chromatic progressions and poignant dissonances, combine to produce an effect of restless longing and fierce excitement, to which, in its particular kind, the history of music can afford no parallel.

Langtamer J = 63. |x x ,

a. .

ptd.

OVERTURES AND SYMPHONIES

i2£

m

7 r

/t\b-

. DP* i.

tT

~lp J

1 l>j bf"^r -1 —^- H */

u.

yp-

r^f 1 ^

s& ' i .... *r *

\ * p -

t= -r-3—

ped.

2 r & r «•

T^r^f-^

ped. 4& ped. ped. ped. ped. ped.

Tn leidenichaSllichem Tempo. J = 144.

A U _ •^'^1* .^I*'~V'»

ft^^^^E

b..

pet?. ^

88

THE ROMANTIC PERIOD

JtB

c. .

cor

3=

JTJ1.

OVERTURES AND SYMPHONIES

Flutes and_O^222L_

. j ^,, H n^ZlXJ j J.

f

Violins.

'==~=:~ ^

Three Trumpets.

r-^r¥

"T

Tnimjiets and Trombones.

:W

:*•

9o THE ROMANTIC PERIOD

Next to Manfred the overture to Genoveva, a shapely and spirited piece, full of refined melody, still holds its own in the concert-room.

Composers, after Beethoven, who desired to attempt the grand form of orchestral music, found the breaking of new ground a matter of considerable difficulty. Schumann shared Mendelssohn's reluctance to experiment in symphonic design. Both masters, however, introduced some novel features with a view to unity, such, for instance, as the use of the same subject in several movements and the more or less close con- nexion of one movement with another, as in Mendelssohn's symphony in A minor and in Schumann's symphonies in D minor and B flat.

Mendelssohn, in the so-called Scotch and Italian symphonies, depicts moods reflecting impressions which he received in Scotland and in Italy (1830-1). He also makes occasional use of certain characteristic traits of Scotch and Italian folk-music, apparently with a view to a musical picture of manners and local colour. In the symphony known as the ( Reformation,' in the symphony to the Lobgesang, and in the overture to St. Paul he appeals to religious sentiment and the associations of worship, by means of leading phrases, such as ' Alles was Odem hat, lobe den Herrn,' in the Lobgesang, or the ( Dresden Amen ' and the chorales f Ein' feste Burg ist unser Gott,' in the ( Reformation ' symphony, and f Sleepers wake,' in St. Paul.

Schumann, more introspective than Mendelssohn, more of a mystic and an intellectualist, and less open to external impressions, sought to express his personal desires the glow of enthusiasm, the ardour of love, or joy, or sorrow J. Thus his four sym- phonies make a stronger appeal to the emotions, and stimulate the hearer in a more direct manner than those of Mendelssohn. They evince a greater power of invention in harmony, melody, and the rhythms and figures which constitute the thematic

1 In every case the man and the musician always strove to speak simultaneously ' Mensch und Musiker suchten sich immer gleichzeitig bei mir auszusprechen.'

OVERTURES AND SYMPHONIES

material. There is more energy in the Allegros, deeper pathos in the slow movements.

The freshness and originality of Schumann's talent is fully apparent, even in his first symphony in B flat, Op. 38 l. Among his friends this work was known as the Spring Symphony ; and it is indeed full of the glory of spring and joyous youth. The Introduction starts with a stately phrase for horns and trumpets :

Andante vn poeo mautoso. Trumi>ets,

T C r r f r

which is really the first subject of the Allegro :

Allegro motto virace.

-•£ .f-'f r- r, *

it occurs again in a larger form at the close of the working-out section. After the recapitulation of the subjects the music breaks away into a new theme, a spring song which carries to a full climax the emotional fervour of the movement.

dot. String?. ^" J~~^

^^-*i-

N

/ P

i-t: L-i

r|» * ,i I

1 First performed undtr Mendelssohn at the bewandhaus, Leipzig, in 1845. At the same concert Mendelssohn's Violin Concerto was first heard.

THE ROMANTIC PERIOD

Towards the end of the Larghetto, a gentle sensuous melody with variants recalling the early manner of Beethoven, the hearer is again surprised by the appearance of new melody. This time, however, it is not climax but anticipation. The new subject arises, ghostlike, to the sound of softly swelling trom- bones, and then, with a sudden forte, bursts forth as the Scherzo :

«T= 66.

Trombones.

Strings.

"

m

m

^

fel

Schtrzo. Molto vii-aee

m

r-

W

The Scherzo has two trios, the first of which consists of a curious device in groups of chords rather weak in effect, alter- nating between strings and winds in f time. The same groups of chords are again introduced at the end of the movement, so as to serve as a kind of bridge to the Finale. Probably the construction of a suitable sequel to three movements of such exceptional calibre presented unforeseen difficulties. Schumann, serious by nature, when he wants to be jocose, is apt to become trivial. That he disliked triviality, so far as he saw it, seems proved by the fact of his employing the full orchestra, trombones

OVERTURES AND SYMPHONIES

93

and all, to emphasize a concatenation of violent phrases, which are meant to serve as a contrast to the somewhat flimsy leading tunes of the Finale.

The result as a whole is hardly satisfactory. Incongruous effects of this kind may be * humorous,' or ' romantic/ or what not they are certainly eccentric. And the first leading tune when it returns after a polite little cadenza for the horns and the flute, comes dangerously near to a Pas seul.

Flute. Solo, cadema

1 $ J7 **'~ =

p— *c« -i

$f

wnpoco ritard. a Tempo

fa 1 rz^d

Strings. ',_; '

p w n p p

P p

94 THE ROMANTIC PERIOD

With the symphony in D minor, No. 4 *, Schumann continued his efforts to attain unity. To this end, the leading phrase of the introduction to the first movement is repeated in the second a ' Romanze ' ; the principal figure of the first Allegro is again employed to form a link between the close of the Scherzo and the beginning of the final Allegro ; and all the movements are joined together so as to avoid a break. In spirit as well as in technical execution both the weak and the strong side of Schumann's talent are fully in evidence. The work contains bold, tender, and fantastic ideas presented in a very free and original manner, and, as in the second part of the first Allegro, often treated with remarkable skill. But both the first and last movements seem, at times, to suffer from a want of air and perspective.

The high-water mark of Schumann's symphonic music is reached in the symphony in C, Opus 61 (1846). Laid out on consistent lines, this work shows perfect unity of spirit, although there are fewer special devices to unify the movements than in the symphony in D minor. Some points in the Finale excepted, the entire work is strikingly original in topic, and the first three movements are admirably concise in form. As compared with the symphonies in B flat and D minor the efficiency of the orchestration is worthy of note. The sounds emitted by the various groups of instruments are characteristic and spontaneous in effect ; and most of the themes give the impression that they were directly conceived in the actual form in which they reach the ear. The work opens with a solemn contrapuntal introduc- tion, a broad theme for trumpets, horns and trombone, accom- panied by a flowing counter-subject on the strings.

1 This work is really the second symphony, having been sketched in 1841, soon after the first, though it was not completed till 1851. It was at first entitled ' Symphonische Phantasie,' and it appears, indeed, as a landmark on the border between the older symphony and the ' Poeme symphonique.'

OVERTURES AND SYMPHONIES

95

Trumpets.

Horns.

Flutes.

Oboes.

Clarinets

Bassoons.

ombone Alto.

et olin.

Second Violin.

Viola. J

jllo.

^'•.-ittinito iiKfiii J = 76.

&• « i 1,.^, n r-^

i

m

r

pp

g U r

_____!

P

96

THE ROMANTIC PERIOD

Truinpeta.

Horna.

Flutes.

Oboes.

Clarinets inBb.

Bassoons.

Trombone Alto.

First / Violin.

Second Violin.

'Cello.

Bass.

I

J

m

Sr^=

=3

- i-

i F

\S*=p

3Z=±

&

=3$A

OVERTURES AND SYMPHONIES

97

P

-^ =4-

i

sa=

-j r Irrrarrr

tP

m m *

DANNREUTHER

98

THE ROMANTIC PERIOD

Trumpets.

Horns.

Flutea.

Oboes.

Clarinets inBb.

Bassoons.

Trombone Alto.

First / Violin.

Second Violin.

Viola. •<

'Cello.

Bass.

V

'J. »

J. r r _

" "

OVERTURES AND SYMPHONIES

99

i

r*

fejUE

±±

SI

P

f

r r r-

3^

=P=^

^^

I- C 1=

:p --— •-H-L-ti-

J.

H 2

ioo THE ROMANTIC PERIOD

This in due course leads to an Allegro full of virile and impulsive energy. The breadth and conciseness of this move- ment are more noticeable than in any other of Schumann's orchestral compositions; the working-out section in particular has something of the grand style and the classic dignity. Next follows a vehement Scherzo with two well-contrasted trios, and then a slow movement of poignant pathos, remarkable alike for its closeness of texture and for several fine effects of instru- mentation. Schumann's special power of vivid expression is well illustrated by his treatment here of the oboes and bassoons, and by the superb chain of trills for the violins which holds in unbroken unity the flying scales and arpeggios of the wood wind.

There is less of concentrated power in the Finale. The subjects are telling enough, but the persistent reiteration of a jerky rhythm bars 13-46 and elsewhere produces an im- pression as of a crowd of superlatives jostling one with another. The development section, in which part of the subject of the Adagio reappears both in an inverted form and recte, is ingenious, and the Coda, in which the phrase from the Introduction, quoted above, is interwoven with the other subjects, is as irresistible in movement as it is rich in texture. It cannot be denied, however, that the full effect is marred by want of balance and economy in the distribution of strongly emphasized points.

Taken as a whole this symphony, together with Schubert's symphony in the same key, ranks as the greatest achievement in symphonic form after Beethoven and before Brahms. It stamps Schumann as the most original of contemporary instru- mental composers and, together with the Manfred overture, places him as the leading representative of the romantic spirit of his day.

Opus 97, in E flat, known as the Rhenish, written in 1850 and published as No. 3, in 1851, is the last and longest of Schumann's symphonies. It is also the least spontaneous and the most laboured. There are five movements. The personal note, that important element in Schumann's work, is not particularly

OVERTURES AND SYMPHONIES 101

prominent, and the technique of the experienced artist can hardly make amends for its absence. In the first movement the process of elaboration is felt to be artificial, and the middle portion of the last movement, for the same reason, appears dry and dull. A so-called Scherzo in C major, £, molto moderate, which in Schumann's words 'probably reflects a bit of life/ has the lilt of a folk-song, and was at first called 'Morning on the Rhine/ The most subtle and carefully written section is an impressive contrapuntal movement in E flat minor, placed between the Andante and the Finale. It was conceived at a solemn ecclesiastical function held in Cologne Cathedral and was originally headed 'Im Charakter der Begleitung einer feierlichen Ceremonie ' ' in the manner of music at a religious ceremony/ The music of Schumann's later life is in more than one respect touched with religious influence. In the days of the Myrthen, of the Pianoforte quintet, of Paradise and the Peri, he took the Romantic movement on its human side and expressed in a heightened and intensified form the joys and sorrows of customary human experience. But the second part of Goethe's Faust opened to him a new world of thought and feeling, and from thenceforward we find him striving towards a deeper and more mystic utterance. It was not always with success, for, with all its reticence, his natural temper was human and sympathetic : when he attempted to scale the remoter and more solitary heights he climbed with a vacillating and uncertain tread, due partly to the unfamiliarity of his surroundings, partly, no doubt, to the gradual encroachment of disease. The craftsmanship would not always answer the requirements of the thought : there is a want of that supreme mastery which, in Bach or Beethoven, can embody the highest truth without appearance of effort. Hence he shows at his best when, as in the movement under discussion, he represents the external pageantry of spiritual fervour: the medium through which it appeals to that romantic element in human nature which even religion itself has not disdained to employ.

THE ROMANTIC PERIOD

FEIERLICH. J = 54.

TIMPANI in

Eb and lib TRUMPETS

E? LJ _

in Eb

§* P " |J

1

^ 'L^/LA ^--4—^-- z^A^. uj i

1

£M9 _—

in Eb

Mr «• -v.

j? PflJ J I J J

OBOES.

e i i =t

CLARINETS

P

57T J— ] 1 J bj bj ^J 1 j ^bjLj J-T

^ wf f ' T f f r T ^ r

tt..fe n i.. -i.iJ j •*--^r^l -J- J uJ ,

TROMBONES.

' PP'T* r f- 'T E-

-s * T^

r.bb (; L.J *i r - P =1 p^ tr* P

tf ft pizz.

'CELLOS. I

if f3 pizz.

BASSES. /

if fl)3 pizz.

f

OVERTURES AND SYMPHONIES

103

r

r^^r

i i i - ~ *=^ ^=

fe^ J. 3 N . J J *t^

1EE

^=g=fc

a j

=L-* .cat'T

1 1 J3 uJ"

r i i -) —i i r

f arco

'J3J qJ3l »

farm

3=g=

^=^c

5&F

H

g

f arco

JJ^^UPH-J-

r

/ orco

104 THE ROMANTIC PERIOD

One of Schumann's lesser orchestral productions may fitly be mentioned here. It is a short Suite or Sinfonietta entitled 'Ouvertiire, Scherzo und Finale/ Op. 52, in E (1841-5),— a bright, slender, somewhat sketchy piece, with a touch of Cherubini in the Introduction and first Allegro.

The most notable weakness in Schumann's Symphonic work is his deficiency in orchestration. Now and then he succeeded in producing novel effects of great beauty, such as the weird sound of the chords for trumpets and trombones in the passage quoted above, from the overture to Manfred, or the thrilling tones of strings ending in the long chain of shakes in the Adagio of the second symphony. New and successful touches of in- strumental colour are, however, rare with him. He shows but small sense of the possibilities of instrumental tone, and for want of intimate acquaintance with the peculiar qualities and capabilities of orchestral instruments either single or grouped, his musical ideas do not seem in the first instance to have presented them- selves to his mental ear through the medium of the orchestra. His instrumentation, in consequence, falls short of Mendels- sohn's habitual clearness and brilliancy, and even as compared with the instrumentation of lesser contemporary composers, German or French, it often appears inept, turgid, or dry. Occasionally, indeed, a page of his scoring looks and sounds like so many bars of rather clumsy pianoforte music writ large.

The finest of Mendelssohn's symphonies, the Scotch (1821-31 -42) and the Italian (1831), form a remarkable complement to Schumann. They certainly have more of the practical efficiency that comes of all-round ability, training and experience especially in respect of copiousness and fluency of diction, and variety of instrumental colour. They are most carefully designed and finished down to the smallest details, and they have a picturesqueness and a poetical atmosphere of their own which leaves nothing to be desired in point of originality ; yet they do not approach the emotional elevation of Schumann's work. The

OVERTURES AND SYMPHONIES 105

conrincing personal touches are absent ; especially in the case of the slow movements wherein Mendelssohn's habitual attitude of reserve in the expression of deep emotion is most apparent. Schumann's disposition always prompted him to deal directly with passion, and strongly to emphasize the human element; whereas Mendelssohn preferred to depict moods which are, more or less directly, the results of external impressions. In other words, Mendelssohn in his leading symphonies, and almost as much in his best overtures, reveals himself as one who chooses to express, in musical terms, the moods of a 'landscape* or ' genre ' painter. Thus the first sixteen bars of the Scotch sym- phony, which now act both as an introduction and as close to the first Allegro, were written down as an impression de voyage (1829) in the chapel of Holyrood Palace, 'open to the sky and surrounded with grass and ivy, and everything ruined and decayed/ The subjects of the first Allegro (A minor, f ), the Scherzo (F major, f), and the Allegro maestoso which follows the Allegro guerriero (A major, f), recall Highland scenery, fighting legends, and the lilt of Scottish tunes. The first subject of the Scherzo is a happy transformation from minor to major of ' Charlie is my darling.' In the case of the Italian symphony it is equally obvious that the most characteristic movements the Allegro, the Andante, and the Finale are records of musical moods suggested by things actually seen and heard. The Andante con moto (D minor £) recalls the music of a religious procession, a Pilgrim's March; and the final Saltarello, named after two strains of a popular dance of the Romagna, contains also a bit of a Neapolitan Tarantella. Heine (1842), in one of his half-serious moods, compared Felix Mendelssohn's talent with the talent of Mademoiselle Rachel Felix, the actress : ' Peculiar to both,' he says, ' is the fact that they are seriously in earnest, they have a decided, almost an aggressive predilection for classical models, they delight in the most in- genious calculations of delicate effects, they show singular clever- ness and, finally, a total lack of na'ivete. But,' Heine adds, ( is

io6

THE ROMANTIC PERIOD

there such a thing in art as genuine originality without na'ivete"} hitherto a case has not occurred. '

It seems a far cry from Mendelssohn to Wagner. But Wagner's overture Faust distinctly belongs to the Central phase of the Romantic development and is best discussed here.

Der Gott, der mir im Busen wohnt, Kami tief mein Innerstes erregen ;

Der tiber alien meinen Kraf ten thront, Er kann nach aussen nichts bewegen ;

Und so ist mir das Dasein cine Last, Der Tod erwtinscht, das Leben mir verhasst.

The God that dwells within my breast, Can stir the inmost of my being,

Holds all my power at His behest, Yet naught without marks His decree- ing ;

And so my whole existence is awry, Life hateful, and my one desire to die.'

These are the lines from Goethe's Faust that serve for a motto to Wagner's f Eine Faust-Ouvertiire,' an orchestral piece written at Paris, in 1840, and originally intended to be the first movement of a Faw*/-symphony ; it was entirely rewritten in 1 854-5, and is now complete as it stands. The second part of the projected (symphony was to have depicted Gretchen. The overture, as we have it, is concerned with Faust's moods alone. It depicts Faust in solitude, with his day-dreams, his sadness, and his despair. In spite of the motto, it would be wrong to interpret this eminently independent and original work as a piece of programme music. The designation ( Eine Ouvertiire ' is equivalent to Characterstiick or Stimmungsbild a picture of a particular mental state or expression of feeling, as is the case with certain overtures of Beethoven such as Eamont, Coriolan9 Leonora, No. i. There is in Wagner's Faust no conflict or discrepancy between externals and the innermost being of man ; the state of feeling is conveyed to the mind by purely musical means, and the design depends upon purely musical devices. There is nothing left to be gathered from a printed programme, and there are no traces of a symbolical or histrionic kind. ' Eine Faust-Ouvertiire ' invites comparison with the Coriolan over- ture of Beethoven and the Manfred overture of Schumann. Faust's moods are as vividly expressed by Wagner, as the essentials of Coriolanus's character, and the emotional influences

OVERTURES AND SYMPHONIES 107

that touch upon it, are depicted by Beethoven ; and Wagner's music reflects Faust's passion more closely perhaps than Schumann's music reflects the passion of Manfred. In the striking originality of the Coda and in the conciseness and perspicuity throughout, Wagner's Faust forms a parallel to Beethoven's Coriolan. The instrumentation is as efficient and telling as Beethoven's, and already foreshadows Wagner's later practice of grouping the instruments in accordance with deli- cate affinities of timbre.

The slow introduction is a model of condensation, and as it were a presage or anticipation of the Allegro. It contains some of the leading subjects and figures in nuce :

Viola. *• Sthr gehalttn

Nothing could be more characteristic of Wagner than this bleak and rugged theme. Its incisive rhythm, its tragic in- tensity, its challenging defiance of conventional melodic beauty, all may find a hundred parallels in his later compositions.

io8 THE ROMANTIC PERIOD

The opening themes of the Allegro are as follows :

a. J

Strings. ^

Sehr lewegt

ff

I Hi

OVERTURES AND SYMPHONIES

109

The second subject, presumably suggested by the lines :

Ein unbegreiflich holdes Sehnen

Trieb mich durch Wald und Wiesen hinzugehn,

Und unter tausend heissen Thranen Ftihlt' ich mir eine Welt entstehn.

A sweet uncomprehended yearning Drove forth my feet through woods and meadows free,

And while a thousand tears were burning I felt a world arise for me.

consists of a broad melody in F major, typically instrumental by reason of its great width of range. A marvel in the beauty of its instrumental colour, it is at first intoned by a group of wood wind and horns, and then repeated in A major by the strings.

^ K-Bl

D r

fi 1 r-V-

^, 3

-j

u^

>~

_ x

Through a series of transitions of a type fully developed in Tristan und Isolde it approaches the third subject, also in F, which during some sixty bars of ingenious modulation gradually comes to a crisis, and prepares for the development section commencing thus :

Wood wind.

Bassoon and LSaas.

no

THE ROMANTIC PERIOD

, _fl M- 14 M-y-r-*- *-/-.-••*

BEr?irirM i^^-i^irpf-fr^r

£g±=iti_jLr rr r=«^! ' i i ' ! ! ! ! "^

and finally leading to a climax of force and power. The conclusion consists of a condensed recapitulation, and a long- drawn-out Coda, pianissimo.

f In instrumental music I am a reactionnaire, a conservative/ said Wagner in 1877. fl dislike everything that requires a verbal explanation beyond the actual sounds/ Together with the Huldigungsmarschj the Kaisermarsch, and that miracle of rare device, the Siegfried-Idyll, this overture, so consistent in design and execution, is a practical illustration of his views.

CHAPTER

PROGRAMME MUSIC

BERLIOZ and Liszt were not content with the novel possibilities opened up by the infusion of the romantic spirit into the established forms. They desired, rather, to bring about some sort of direct alliance of instrumental music with poetry. To effect this they chose to make use of the means of musical expression for purposes of illustration. Thus they came to lay stress upon the conjunction of music with some particular poem or some special order of ideas, and, in the construction of their pieces, to rely directly on suggestions received from literary or other non- musical sources. The result was the production of a curious hybrid which has been called Programme music, i.e. music posing as an unsatisfactory kind of poetry l.

In the case of Berlioz' symphonies, the programme forms an important part of the artistic scheme and the music is hardly intelligible without it. His symphonies comprise the Symphonie fantastique (1828-30), and its sequel, Lelio, ou le retour a la vie monodrame Iyrique(i8^2) ; Harold en Italic (1834); and the Symphonic dramatique, Romeo et Juliette (1839).

In Berlioz' overtures, such as Waverley, Les Francs-Juges, Rob Roy, Roi Lear, Le Corsaire, Le Cameval romain, the mere names ought to be sufficient to define the nature of the appeal to the hearer's imagination. But in a large part of the Symphonie

1 Lesueur (1763-1837), Berlioz' composition master at the Conservatoire of Paris, in his pamphlet Expose d'une musique imitative et particuliere, so early as 1786 pointed at imitation as the true aim of music, and his compositions were written in accordance with a ' plan raisonn^,' which plan or programme was printed for the benefit of the audience. In this respect there appears to be a ' filiation artistique ' between Lesueur and his pupil.

\

ii2 THE ROMANTIC PERIOD

fantastique, notably the Finale, the music is sheer nonsense unless the hearer has knowledge of the programme ; yet even if he has full knowledge, the heterogeneous factors interfere with one another, and leave an annoying sense of incoherence and incongruity. The same may be said of parts of such movements as the ( Orgie des Brigands,' which forms the finale of Harold en Italic, or the f Invocation et Reveil de Juliette,' and the middle portion of the ( scene d'amour,' which forms the central move- ment of Rom6o et Juliette.

Berlioz, who claimed to have { taken up music where Beethoven laid it down/ deserves respect for the power and persistence of his efforts. About his originality there can be no contention. His aim was always high, his ambition unbounded. Whether he succeeded in expressing what he meant to express, as Beethoven almost invariably did, is another matter. After more than half a century of controversy, it is by no means easy to estimate the value or to gauge the genuineness of his claims. None, how- ever, need take him at his own valuation or at that of his partisan Liszt. The problems involved are: Were his con- ceptions formed in harmony with the conditions of musical art ? Did his methods of composition do justice to his conception? Is his style convincing ? And the answers, even if they are mainly in the negative, may now be given with some hope of general acceptance. The whole question hinges on the special nature of Berlioz' gifts and attainments, and on their particular limitations. He was a man of excitable temperament and vivid imagination, a great master of instrumental effect, an adept in the use of colour and rhythm, a melodist of limited scope and power, a poor harmonist, and an equally poor contrapuntist. His instinct for form and just proportion was defective, his judgement wanting in balance.

His disposition was in reality more poetically imaginative than musical. He was inclined to emphasize the histrionic side of art and to indulge in extravagance. His ideals of beauty and style in music appear to have been narrowly personal, and he chose at

PROGRAMME MUSIC 113

times to embody them in a bizarre and eccentric manner. He was familiar with the principal operas of Gluck and Spontini ; and with those of Weber, in so far at least as they can be ap- preciated without a knowledge of German. On the other hand, the assistance he derived from the example of the masters of self-dependent instrumental music was comparatively small. His acquaintance with the bulk of their work, Beethoven excepted, was little more than casual, his knowledge of Bach practically nil. The view that he took of Haydn and Mozart is best stated in his own words : ' When I hear Mozart I am troubled with a slight nightmare, when I hear Haydn I am troubled with a big one.' It is undeniable that in spite of his sympathy with Beethoven's symphonies, quartets and sonatas, he was unable fully to grasp and value the principles of design which underlie them and to appreciate the many subtle devices by which Beethoven contrives to give them life and unity.

Berlioz' failure to combine his many excitable impulses into an organic whole, caused him to feel the need of a * programme ' which should explain the intention of his music. The programme, as employed by him, serves on the one hand to disguise a lack of constructive power, and on the other hand assists in the attempt to express literary and theatrical ideas in terms of orchestral colour and rhythm. It follows that those sections of his in- strumental works in which design is of the greatest importance, that is to say, the principal Allegros and the Finales of the symphonies, exhibit his peculiar powers in the least favourable light.

The programme of his first important instrumental produc- tion, the Symphonic fantastique (1830), informs us that the work is intended to depict an f episode de la vie d'un artiste,' and goes on to state how a young musician is affected by f le vague des passions,' which causes the image of a beloved maiden always to appear to him accompanied by a musical phrase, a Leitmotif, £ comme une double idee fixe.' At a ball, the young man in an exalted mood stands and gazes upon the

DANNRKUTHER

THE ROMANTIC PERIOD

scene, but the ' idee fixe/ both maiden and tune, continues to trouble him. Of an evening, in the country, he hears two shepherds piping to one another from afar; he reflects that the maiden might be false to him : ( Bruit eloigne de tonnerre . . . solitude . . . silence . . . ' Having acquired the certitude that his sentiments are misunderstood, he takes a dose of opium. He dreams he has murdered the beloved and has been condemned to death, but is yet in a position to witness the execution: cMarche au supplice^; at the end of which ' Fidee fixe ' appears again, but is cut short by the fall of the axe. Finale : Visions of a witches' sabbath f howls, laughter, cries of pain, waitings/ the melody appears yet again, now turned into a vulgar dance tune f Demoniacal orgies, death bells, the Dies irae and Ronde du Sabbat combined amid wild yells/ and so forth.

The outlines of the customary movements of a symphony are shadowed forth in this curious scheme. I. ' Reveries passions/ Largo and Allegro. II. ( Un Bal/ a dance movement. III. ' Scene aux champs/ Adagio. IV. £ Vision a. Marche du supplice ; b. Songe d^une nuit du Sabbat/ Allegro strepitoso. A few quotations will show the character of the themes employed :

Largo J = 56.

a.

PROGRAMME MUSIC

J* j

£

80-

L'idfe fixe. Allegro agitato es = 132.

b.

EN

mf

X-—

r

=7-1, ^>

P^iF=-^*-H

r^rn

v

Ff2^"!

^V

Sspretiivo eon passione

fr'

jjoco ^/brzato

" 1

w>

J

-»-^P^4 •— P-

TO ur

&C.

I 2

THE ROMANTIC PERIOD

Oboe solo.

C.

'"IT

-

3 «^ 1

ly ~M~

Stprettivo e

*

I

1

?

:^:

-

^--

"

E

P^ f

* i Strings

•L

1

P*

r

*

t

r

'Cello

6

\L_

8

J J J J J J 1

i

i

P

&C.

:fr f

u

The c idee fixe ' (Example a) may be taken as a fair sample of Berlioz* melody artificial rather than warm or spontaneous, and not susceptible of much variety of harmonic or contrapuntal treatment. The combination of tunes extending to some forty- eight bars (Example b) is an instance of that peculiar process of rabbeting which with Berlioz often takes the place of counterpoint, and may be compared with the subsequent treatment of the Ronde du Sabbat and Dies irae. The Valse tune approaches the commonplace :

PROGRAMME MUSIC

Strings.

Valtt. j = 60.

The f Scene aux champs,3 the most consistent and most music;il section of the symphony, shows the influence of Beethoven's Sinfonia pastorale, notably of the s Scene am Bach/ Though there are no direct reminiscences of that movement, its mood is reproduced tant bien que mal', and an equivalent for the quaint imitation of bird notes at the end of it is sought in a theatrical effect of distant thunder, which is produced, very realistically,

u8

THE ROMANTIC PERIOD

by four drummers- on four drums struck with s baguettes a tete d'eponge/ Berlioz ends his scene with a variant of the duo between oboe and corno inglese which forms the opening a subtle and telling device that makes for symmetry :

Adagio - 84.

a.

y#-*-J -r q 1

Corno Inglese.

* * J J '

F===

Oboe. jjl

h r S q f\

gO

dolce

ffi*:

^

^=^

^^

Viola.

PTft

f=pFTi

&c.

PROGRAMME MUSIC

119

Corno Ingleae.

b.

/rfci

T": r. fl

[.. J 3 , -=]

Timpani.

^ ' 4 ^J

P^' /"^/^ *P

1 Timpani.

w f

/tab - h

f 1 | J a J l-UErgr-: d 1

&?- m

% P% *

••I .

^±J&-L- If 1 ^=^= 1^-,-^^=^=

h •'- h-~^"^=1

The drummers are busy, with and without sourdines, ' et avec des baguettes de bois,' during the fortissimos in the Marche au supplies, as well as in the Ronde du Sabbat, and the travesty of the Dies irae which constitutes the Finale. In the latter movement they are assisted by two further drummers who beat a ( grosse caisse roulante/ and by a host of brass instru- ments, a pair of cymbals, and two bells behind the scenes.

120

THE ROMANTIC PERIOD

Strings.

Drums with mutes.

J J J

^f

dim.

i

jGTV— t

r '^.'r^:^-! ai

* ^

Dies irae et Ronde du Sabbat ensemble. Strings.

b.

^

ff

Brass and Wood wind.

~t^ -&•

PROGRAMME MUSIC

121

The limitations of Berlioz' mental horizon, his histrionic bias and tendency to attitudinize, are nowhere so much in evidence as in that collection of romantic platitudes, the monologues of '1'acteur qui parle et agit seul sur 1'avant-scene ' in Lelio, ou le retour a la vie: monodrame lyrique, which forms the sequel to the Symphonic fantastique. It is a mere pasticcio made up of heterogeneous fragments and pieces loosely strung together by means of rhapsodical declamation.

Apart from his love for a loud noise, every page of his scores manifests Berlioz' instinct for sonorousness and tone quality, and the extraordinary care with which he worked out his combinations of instrumental colour indeed the lucidity and beauty of his instrumentation throughout is remarkable. A number of his novel effects, however, be they beautiful, violent, or grotesque, are produced by the mere sound of the instruments rather than evolved from the musical conception; they seem to have had some separate existence, perhaps in the composer's notebook, before the notion of either programme or symphony occurred to him. ( C'est fort beau, quoique ce ne soit pas de la musique.' From a technical point of view, Berlioz' orchestration is chiefly remarkable for its distinctness. It is,

122 THE ROMANTIC PERIOD

as he calls it, eclatant in every sense of the word, varied, daring, original. The stress is laid more often on the novelty than on the beauty, but things sound as they are intended. The various procedures have nothing about them that might be called meretricious no particular devices predominate. His powers of invention, in this respect, are almost as great as Weber's and far greater than those of his disciples in orchestra- tion such as Liszt, Felicien David, Saint-Saens.

The second symphony, Harold en Italic, contains also a persistent melody, which, like that of the Symphonic fantas- tique, is somewhat jejune in character, and somewhat incapable of harmonic variety.

Adagio = 76. Esprestivo e largamente

Viola

:if* 3 J rf

^-^ p= ~ |***j J-pJ. j~T~p~h» "J 1 * ~i

HIP

: J| 4 ±.

7)^

iEs^sh

"j^ i h-H

M+-W

*• * ^ ^^ _

Originally designed as a concerto for Paganini, this work grew under Berlioz5 hand into a l Symphonie en quatre parties avec UH alto principal 3 ; and it is interesting to observe that two of its chief themes are taken almost exactly from the discarded overture to Rob Roy 1. One, ' the persistent melody,5 stands for ' une sorte de reveur melancolique dans le genre du Childe Harold de Byron V The other tune found a place in the first Allegro. The entire symphony forms a record of the composer's musical recollections of Italy: I. f Harold aux montagnes, scenes de melancolie, de bonheur et de joie,' Adagio and Allegro. II. * Marche de pelerins chantant la priere du soir,5 Allegretto. III. ' Serenade d'un montagnard

1 Written at Rome, 1832, and said to have been destroyed after a fiasco in 1833. The full score was, however, found and published in 1900. When Lully, obeying his confessor, put the score of his Armida in the fire : ' Eh quoi, Baptiste, lui dit- on, tu as pn bruler une si belle chose ! Paix, paix, j'en ai garde copie.'

2 Eecte, Dans le genre du R^ne", de Chateaubriand.

PROGRAMME MUSIC

123

des Abruzzi a sa maitresse,' a sort of Scherzetto. IV. ' Orgie de brigands, souvenirs des scenes precedentes,' Allegro frene- tico. The scheme is closely akin to that of the Symphonic fantasttque, embracing a similar order of movements, with the histrionic Leitjnotif in each, and Berlioz' favourite ' Orgie,' by way of Finale. The fragmentary restatement of themes, described as ' Souvenirs des scenes precedentes' is a device derived from the Finale of Beethoven's Ninth Symphony (the score of which appeared in 1826). Advocates of Berlioz5 method can point to the ' Marche de pelerins' with its mysterious charm, and the * Serenade,' as genre pictures of rare originality, the undeniable success of which goes far towards justifying realistic effects, such as the imitation of bells and the responses of the Litany which occur at intervals in the march, or the pseudo-contrapuntal ingenuity with which the Harold motive is joined to the main theme :

Allegretto j = 96.

„- Oboe, i' ^J. r . -— ; •--. . ' v

Marche de pelerins.

*

Strings.

124

THE ROMANTIC PERIOD

dim.

Wind.

&c.

poco f

Viola.

b. \

Strings.

/ J f. 1 1

1

pizz.

/^"i~T'Vi in

E= 4 ^J

y^-p-"'*- *

mp. 'Cellos and Bassoons Cantando __

^*i*P::r:=='T i "^~=F3-

i 5^= g i-

i^w-rL-^

\ ^-Jt-l-J- yj [

i* =r

ytzz.

Solo.

m/ Theme de I'adagio

g

«*N-g

^ Canto.

-

TUT

n s

5

11

fe?

^4^==^

1;

PROGRAMME MUSIC

125

P Canto.

eret, Ac.

3=

•=*»?

In the c Serenade ' the interweaving of dissimilar melodies and rhythms is carried to an extreme with surprising cleverness and picturesque effect :

Allegro aiaai J = 138.

p

Wi

*=j~t

*

f

fc*^

•1 8va. . .

~J.

^"

F=f=

J4.

r r JF : JF :

mf

r r

in ? f i

N 8 »-f T

r 'j r r j r

h h r h

Allegretto J = 1S9. Moitit mains vite. Viola solo.

-I- " n T I

Flute, Harp (harmonics).

Let Altai contervent le meme mouvement (Allegro assai)

^ffmfff

i:

* J

s2-

P

126 THE ROMANTIC PERIOD

| ; j

j£pr- 1

^-|»- '

If ' =1

& j : j : i rr»-fpt-

1

j M h rn m

~JS» = 1

&c.

With f Romeo et Juliette symphonic dramatique, avec choeurs, solos de chant et prologue en recitatif choral, composee d'apres la tragedie de Shakespeare ' (1838) Berlioz professed to believe he had reached cl'ile merveilleuse ou s'eleve le temple de Part pur/ What he did in fact reach, or at least approach, was a kingdom of Cockayne. And he seems to have been dimly conscious of this fact when, in the preface to the edition of 1857, he tried to explain that his efforts were not meant to be an ( opera de concert ' nor a cantata, but a l symphonic avec choeurs/

In the first instance the work was- undoubtedly intended to be a symphony illustrating certain incidents of Shakespeare's play. The conception comprised the following movements : Andante malinconico and Allegro (f Romeo seul/ ( Grande fete chez Capulet '), Adagio (' Scene d'amour '), Scherzo (' La reine Mab '), Finale, instrumental and choral (c Rixe des Capulets et des Montaigus Serment de reconciliation ') ; evidently a scheme resembling that of the Symphonic fantastique and Harold, influenced by the vocal Finale of JJeethoven's Ninth. If a programme thus laid out had been furnished separately, or if only the titles of the sections had been given, the composer might have succeeded in justifying his scheme ; but he chose, instead, to preface the principal instrumental movements by a lengthy vocal introduction : which consists of a Prologue set as a choral recitative, followed by two Couplets for contralto solo containing sentimental reflections. Later on, the symphonic movements are interrupted by a chorus of male voices singing

PROGRAMME MUSIC

127

1 Reminiscences de la musique du bal,' and again by a choral and instrumental piece, entitled : ' Convoi funebre de Juliette.' This is followed by a melodramatic orchestral piece, ( Romeo au tombeau des Capulets' and the close consists of a kind of operatic scena which includes a lengthy 'Recitatif et air du pere Laurence ' and a choral ' Serment de reconciliation/

Thus the complete work has seven divisions: i. Orchestral introduction, 'Combats, tumulte, intervention du Prince'; Prologue, ' Recitatif choral et Scherzetto, La Reine Mab,' tenor

solo and chorus. 3. Romeo seul ( Tristesse,' orchestra alone ; c bruit lointain de bal et de concert ; grande fete chez Capulet.' 3. Ojrchestra and chorus 'Nuit sereine; le jardin de Capulet silencieux et desert; les jeunes Capulets, sortant de la fete, passent en chantant; scene d'amour' orchestra alone. 4. Scherzo { La reine Mab, ou la fee des songes ' orchestra alone. 5. Orchestra and chorus f Convoi funebre de Juliette.' 6. Orchestra alone * Romeo au tombeau des Capulets ; invoca- tion, reveil de Juliette ; elan de joie delirante, brise par les premieres atteintes du poison ; dernieres angoisses et mort des deux amants.' 7._Finjle-— Prchestraj solo voice, and double chorus fLa foule accourt au cimetiere; rixe des Capulets et des Montaigus ; recitatif et air du pere Laurence ; serment de reconciliation.'

Is it in any way surprising that such a conglomerate should prove a failure in spite of a host of novel and beautiful details ? Undoubtedly practical skill and insight is shown in the selection of incidents which lend themselves more especially to musical illustration, and produce once and again effects in a high degree striking and original. But the plan of the whole, if plan it can be called, is thoroughly unsatisfactory the design of some of the separate movements equally poor. The constant reference to incidents in the play is the cause of much that seems musically inconsistent ; it accounts also for certain lapses into mediocrity and for an abundance of histrionic effects which tend to lower the general quality of the music. Effort after effort is made to

ia8 THE ROMANTIC PERIOD

attain dramatic expression. The style recalls the rhetoric of the grand opera. It is the music of attitudes. It constantly seems to demand stage action, and to be striving to illustrate or emphasize things more or less alien to musical art.

The vocal pieces, both choral and solo, are distinctly inferior to the instrumental ; the artificiality of Berlioz' melody reduces their value. With the exception of the Convoi funebre, the touching effect of which is more truly orchestral than vocal, they are not particularly interesting, either from the singer's point of view or from that of the hearer. The opening of the movement entitled f Romeo seul Tristesse ' is on a par with the fine introduction to the Symphonic fantastique. Here is genuine music; in both pieces the chromatic wail of strings produces a striking effect of melancholy and longing. It is interesting to compare such poignant chromatic passages in Berlioz' symphonies with the richer and more musical employ- ment of similar devices in the introduction to Wagner's Tristan und Isolde. Compare also the melody and harmony of the Andante malinconico with the chant of syrens in the first scene of Tannhauser, or the following bars from the finale (Pere Laurence) with the sixteenth to the twenty-fourth bar of the overture to Tannhauser.

Au li . vre du par - don in - scri . ra ce ser-

Si«U_

'Celli divisi.

PROGRAMME MUSIC

129

In both cases Wagner's superiority is not less unmistakable than his indebtedness to Berlioz.

Then follows a Larghetto espressivo, and an Allegro; the themes of which are finally combined and produce an effect of extraordinary brillance. But in the next movement Berlioz has something better than brilliance to offer. The ( Scene d'amour J contains some 200 bars of the richest and most delicately passionate music in existence. There is nothing in the whole range of French music to approach it, and nothing but the 'O sink hernieder, Nacht der Liebe' in the second act of Wagner's Tristan, to surpass it. Unfortunately the instru- mental recitatives, which are meant to represent Shakespeare's lines about the lark and the nightingale, produce an incongruous effect, and spoil the middle of the movement. The instrumenta- tion is remarkable and beautiful from beginning to end :

2nd Violins.

Adagio

J*= 88.

Flute and Clarinets.

tip con sordini Basses.

^

HP

DANNItEUTHER

THE ROMANTIC PERIOD

Clarinets and Corno Inglese.

J-

mm

1

-l -i

^7

-J7

m

It.

sss

PROGRAMME MUSIC

fcfc

t--b5==

IstViolinO

.^ 3 tt:

ms

¥

1 *- H

ill

2nd Violins.

Bassoons.

rr

=

IT

K 2

132

THE ROMANTIC PERIOD

Clarinet and Corno inglese.

^

1st Violins, fcfcfc

--— =a- tr

' -^ .d— <•-

&c.

r ^ n

i»Pr a-

fit

The Queen Mab Scherzo is an orchestral tour de force, dealing largely in the sound of violin and harp harmonics. In form it is akin to the Scherzos of Beethoven's Third and Seventh Symphonies. The fantastic element predominates to such an extent that the effect is more curious than beautiful, indeed, borders upon the ludicrous f un petit bruit semblable a celui des seringues mal graissees.' The ' Convoi funebre' which follows is described in the score as a *marche fuguee instrumentale d'abord> avec la psalmodie sur une seule note dans la voix : vocale ensuite, avec la psalmodie dans Porchestre.' The following extract will convey some idea of its effect :

And'

'ante J

Trebles

and Tenors divided.

"Cellos, etpressivo

m

PROGRAMME MUSIC

pp

133

Je - tez des fleura pour la

Oboe.

S

vierge ex - pi - r^ - - e !

'Cellos :m

P

Violaa.

I v

Flutes and Clarineta.

134

THE ROMANTIC PERIOD

f* | h >

- J 4 + i n=

-i i -

Je-tez des fieurs !

Je - tez des fleurs pour la vierge ex-pi - r6 -

- e ! 1st Violins.

tr t "

poco a poco cres.

poco f

Of the reception of the sixth division, a melodramatic instru- mental piece illustrating the scene in the tomb, Berlioz seems to be in doubt, and the score contains a note advising its omission on the ground that f le public n'a point d'imagination ; les morceaux qui s'adressent seulement a ^imagination n'ont done point de public. La scene instrumental suivant est dans ce cas, et je pense qu'il faut la supprimer toutes les fois que cette symphonic ne sera pas executee devant un auditoire d'elite auquel le cinquieme acte de la tragedie de Shakespeare avec le denoument

PROGRAMME MUSIC 135

de Garrick est extremement familier, et dont le sentiment poetique est tres eleve. C'est dire assez qu'elle doit etre retranchee quatre-vingt-dix fois sur cent.' If imagination on the part of an intelligent audience can make up for incoherence on the part of the composer, this somewhat arrogant manifesto may perhaps be justified. Questions as to the appropriateness of his method seem never to have troubled Berlioz. But judged on its musical merits alone, the movement is a failure. There are genuine accents of passion here and there, and towards the middle the pathos of the ' Invocation 3 is well sustained for some four-and-twenty bars, then the strutting recitative of the operatic actor again makes its appearance, and the tragedy collapses. The incongruous and lengthy vocal scena, which forms the seventh and concluding division, shows traces of Meyerbeer's Huguenots. It needs no comment beyond the statement that it is as dull as it is elaborate. A few words of Wagner's l sum up the situation perfectly : ' Berlioz added to, altered, and spoilt his work. This so-called Symphonic dramatique, as it now stands, is neither fish nor flesh strictly speaking it is no symphony at all. There is no unity of matter, no unity of style. The choral recitatives, the songs, and vocal pieces have little to do with the instrumental movements. The operatic Finale, Pere Laurence especially, is a failure. Yet there are many beautiful passages in the work. The Convoi funebre is a very touching and masterly piece. The opening theme of the Scene d'amour is heavenly ; the garden scene and fete at Capulet's wonderfully clever; indeed Berlioz was diabolically clever (verflucht pfiffig).'

Brief mention of the remaining orchestral pieces of Berlioz suitable for concert performance may conveniently be made here. They are : a piece ^occasion, entitled : f Symphonie funebre et triomphale pour grande harmonic militaire, avec un orchestre d'instruments a cordes et un choeur ad libitum' (1834-40), and eight overtures, viz. Waverley (1827-8), Les Francs-Juges

1 See Grove's Dictionary of Music, article Wagner.

136 THE ROMANTIC PERIOD

(1827-8), Roi Lear (1831), Rob Roy (1832), Benvenuto Cellini (1837), Le Carnaval romain (1837-43), Le Corsaire (1831-44— 55), Beatrice et Benedict (1862). The Symphonic funebre et triomphale,for military band originally written for the ceremony of transferring to their monument in the Place de la Bastille the remains of the victims who fell in the revolution of July 1 830 consists of a sombre and rather heavy march-like movement in F minor ; and an { Oraison funebre 9 for trombone solo, followed by a cleverly contrived flourish of trumpets, cornets, trombones, and drums, leading to an 'Apotheose' for the combined orchestras, in which the chorus joins 1. The overtures are satisfactory or the reverse according as the composer indulges or restrains his e poetical intentions/ Thus Waverley, Rob Roy, Les Francs- Juges, are all open to objection on the ground of some deviation from musical common sense. Le Roi Lear and Le Corsaire, the latter remarkable for consistency of form on novel lines, i.e. lines analogous to those of the first movement of Beethoven's Sonata, Op. 109, in E major2, though far behind Beethoven's ideal, follow in the track of Egmont and Coriolan; whilst Benvenuto Cellini, together with Le Carnaval romain and the overture to Beatrice et Benedict, are frankly operatic and free from the burden of a programme. Technically there is little to choose between them.

Berlioz' eminent gift of calculation and combination, his peculiar method of interweaving heterogeneous phrases, arrests attention already in his early overture, Les Francs-Juges (F minor). This work, which belongs to the period of Berlioz' pupilage, shows as much reverence for orthodoxy of form as its mercurial composer ever allowed himself to exhibit. Yet even here the histrionic method appears. By way of working

1 Here, again, it is instructive as well as curious to compare Wagner's two occasional pieces, the Huldigungsmarsch, originally for a military band, and the Kaisermarsch, with chorus. The spontaneity and technical superiority of Wagner's work are very striking.

" As has been pointed out by Mr. W. H. Hadow in his Studies in Modern Music, vol. i. p. 141.

PROGRAMME MUSIC 137

out, an attempt is made to depict the gradual approach of a band of warriors. Scraps of melody, scraps of accompani- ment are heard, the rhythmical beats as from afar, then the fragments coalesce, and the arriving hosts are greeted with a shout of welcome. Still more evident is the dramatic intention in the music of the overture Le Roi Lear. In this piece the form of expression, especially in the introductory part, is vivid enough for a tragic opera with Leitmotive which might be labelled Lear, Goneril, Regan, Cordelia ; so vivid, indeed, that, given the general designation, even an unimaginative hearer is likely to take the composer's meaning, and to find the proper names for the themes.

We may agree with Felix Draeseke * in assigning to the over- ture Benvenuto Cellini (G major) the foremost place among Berlioz' shorter instrumental pieces. The companion piece, Le Carnaval romain (A major), invites comparison with the Saltarello of Mendelssohn's Italian Symphony, though not to its advantage in point of freshness or finish. It is made up of the principal tunes from the second finale of Benvenuto Cellini.

Berlioz, his partisans assert, was grand, sublime and powerful, terrible or comic as he chose ; to which detractors not unfre- quently retort that he was grotesque, inflated and pretentious. Perhaps there is a modicum of truth on both sides. He wor- shipped the grandiose. It is vain to look for traces of mental growth in his work; a comparison of dates reveals next to nothing in regard to the development of his genius, unless, indeed, the dates which belong to the last twenty years of his life be taken as marking retrogression.

One of his most ambitious efforts as an illustrator, the Marche funebre pour la derniere scene a" Hamlet (1848), a piece little known and very rarely performed, deserves at least as much attention as any other bit of descriptive music by his Russian or German disciples. It is written for grand orchestra and

1 Die Musik, p. 1257 (1902).

138 THE ROMANTIC PERIOD

voices in chorus, plus '6 Tambours voiles ou sans timbre, Grosse caisse, Cymbales et Tambour.'

A curious fact in connexion with his programme music is the evolution of what is known as the Leitmotif from the ' Double idee fixe ' of the Symphonic fantastique and the 'melodic caracteristique' of Harold en Italic. The use of a musical phrase as a subordinate element of coherence, a ready-made appeal to the imagination and memory, was an innovation in Berlioz5 time. It survives to this day as the technical process which has proved most serviceable among the many experiments that Berlioz made in search of novelty. Partially adopted by Meyerbeer in Les Huguenots and Le Prophete, and systematically developed by Wagner, it is now familiar to musicians as the method of employing representative themes, and serves as an efficient device in the technique of dramatic music or oratorio. At first it was an ( Erinnerungs- motif,' i.e. a motive appealing to the memory, which was gradually developed into the f Leitmotif ' as we now have it.

Among the few works directly inspired by the example of Berlioz, perhaps the most noticeable is the Ode-symphonie, Le Desert, by Felicien David. Le Desert was performed at the Conservatoire of Paris in 1844 and speedily made the round of Europe1. This work, inasmuch as it consists of orchestral movements, descriptive or melodramatic fragments, marches, dance tunes, vocal solos, and choral pieces for male voices, each preceded by some lines of explanatory verse or prose, resembles Berlioz' Melologue Lelio, mentioned above. There is a cleverly contrived sequence of events far less of a makeshift than the order of scenes in Lelio showing the progress of a caravan through the Sahara and including a hymn to Allah, a struggle with the Simoon, and a rest at the well of an oasis. Effective use is made of typical Arabic phrases and tunes -as in the 'Priere a 1' Allah,' the 'Chant du Muezzin,' and the 'Danse des

1 Chiefly in consequence of a very appreciative article by Berlioz in the Journal des Debats, December 15, 1844.

PROGRAMME MUSIC

139

Almees.5 There is nothing really symphonic about David's ode, but^it pleased those who had a liking for descriptive experiments : and Le Desert, for a number of years, furnished hints to seekers after novel modes of picturesque expression 1. The Muezzin's chant at sunrise, as David puts it, is well worth quoting :

Tenor.

Uii poco motto e con elevaiione

^-.^ -eg

If* r e e p ' ff i

BE=EE

i

£1 . . .Sii-liiin a

- lek, a - leUkoum el 8a>

V w_jJ^_J_

IJCT— i— i rpr r , \fp \ ^_j* {* ' f^p— {-P-FiH

i f . . r -=pq

A 1 - lali

a^, j ,— : ,

/r\

1 h rf ^

dim.

/7\

boa ak - bur,

ia lea Sa - lah . .

m

^

1 Compare Borodino's Dans les steppes de I'Asie Centrale, and the ' Au convent * and ' Serenade ' in his Petite Suite ; some of Rubinstein's Persische Lieder ; portions of Saint-Saens' Africa, and his Fifth Concerto; the 'Danse Indienne' in Rimsky- Korsakov's Milada ; the Reverie orientale and Bhapsodie orientate by Glazounov.

THE ROMANTIC PERIOD

la Al - lah . . . il Al - lah ou Mu- ha- rued ras-soul' Al-

dim.

f r

:^pA«-j

=

t

p

*=

-f-f

?±:

^p

ffi i— l-

- lah

===!

c=

Al lah

p

-

.

/T\

- -C_t

=

r f - 1

dim.

3=

ad lib.

hou ak - bar, ia leg Sa-lah

dim.

fp

f ere*

Al - lah ... il Al - lah ! .

PROGRAMME MUSIC 141

David's second Symphonic avec choaurs, Christophe Colomb (1847), failed to attract attention1.

Liszt2, with his largest productions Sine Symphonic zu Dantes Divina Commedia (i 847-55), and Sine Faust-Symphonic (i 854-7) shows less directly than David the trace of Berlioz' influence. Like Berlioz he aims at effects derived from a desire to illustrate, and he relies on points of support outside the pale of music proper. Saturated with contemporary literature, imbued with the Parisian romanticism and the spirit of the Catholic revival of 1 830-40, and stimulated, musically, by the example of Berlioz, he set himself in his own way to work out, on poetic lines, the problem of musical illustration. Composed of imagination and impulse, his musical genius was one that could hardly express itself save through some other imaginative medium. He devoted his extraordinary mastery of instrumental technique to the purposes of illustrative expression; and he was now and then inclined to do so in a manner that tends to reduce his music to the level of decorative scene painting or affresco work. His orchestral compositions consist of the two symphonies already mentioned, two illustrations to Lenau's Faust, twelve lesser pieces with the happily chosen title ( Poemes symphoniques/ and a few miscellaneous works of comparatively slight importance.

Of all the subjects Liszt chose for symphonic treatment the Divina Commedia was the best suited to his peculiar tempera- ment, as it offered superb opportunities for the display of his powers as a master of instrumental effect. Already, in 1847, he had planned musical illustrations of certain scenes from Dante's poem with the aid of the newly-invented Diorama. The original

1 Such was also the fate of his oratorio Moise au Sinai, the Jfystere Eden (1848), the comic operas La Perle de Bresil (1851), LaUa Rookh (1862), Le Saphir (1865), and the grand opera Herndaneum (1859). David with advancing years gave way to an inartistic desire to illustrate all manner of non-musical ideas. He reached the height of absurdity in the score of L'Avant-homme, where by means of marginal notes he endeavoured to combine certain curious musical effects with a statement of the principles of antediluvian geology.

* Liszt was born at Raiding, Hungary, in 1811 ; he died at Bayreuth in 1886.

142

THE ROMANTIC PERIOD

idea was never carried out, nor does there seem to be much con- nexion between certain sketches he made with a view to it and the symphony as it now stands. An eccentric e Fantasia quasi Sonata' for the pianoforte (Annees de pelerinage ii, 7), suggested by a poem of Victor Hugo, c Apres une lecture de Dante,5 seems also to have been merely tentative. A number of lines from the Inferno, as at first selected for illustration, have, however, been retained. They now appear in the score of the Dante Symphony, as inscriptions elucidating a number of musical phrases which act as Leitmotive. To give an idea of the character of the whole composition would require more exten- sive quotation than is possible here, but the Leitmotive will at least exhibit the general conception of the music. The entire work is laid out so as to form three divisions, entitled Inferno, Purgatorio, and Magnificat the last in place of Dante's Paradiso.

Inferno.

Trombones, Ti Strings in u

iba, and aison.

i? cJT^'-1^

r=*£-^~

Lento ff Per me

]<&•—{* - -=( - -

Sa

si

ra nel-la cit - ta do

h .! 'r^rTJ Ei-l

len - te: Timpani.

A $

^

z5 ~3

Tam-tam.

Per me si va nell' e - ter - no do - lo - re :

Timpani.

Tam-tam.

Per me si va tra la per - du - ta gen - te.

*/:

PROGRAMME MUSIC

<f\ "i ^i ^

143

T T - sciji . te o - gni spa ran Tremolando

99 t

I

TO!

ch'en tra

to!

-I "K

^

2t f=»-

The furious chromatics of the first Allegro, entitled Inferno (D minor mainly), do full justice to the poet's words :

Diverse lingue, orribili favelle, Parole di dolore, accent! d' ira, Voci alte e fioche, e suon di man con

elle

Facevan un tumulto, il qual »' aggira Sempre in quell' aria senza tempo tinta, Come la rena, quando al turbo spira.

Languages diverse, horrible dialects, Accents of anger, words of agony, And voices high and hoarse, with sound

of hands,

Made up a tumult that goes whirling on For ever in that air for ever black, Even as the sand doth, when the whirl- wind breathes.

For musical reasons, much is made of the Paolo and Francesca episode with its characteristic theme in £ time :—

^<nrfan<« amoroto (Tempo ruftato)

r^

144

THE ROMANTIC PERIOD

'|

tK

^

« J-

=*» =

The theme ( Lasciate ogni speranza ' is repeated with tremen- dous vehemence at the end of the movement. The strains of the Purgatorio, after a tranquil introduction reflecting Dante's ' tremolar della marina,' are meant to suggest the soul's passage through expiatory pain and a state of contrition, towards a sense of beatitude :

Metto

Purgatorio.

a.

g^g

P Woodwind.

rJ-

Flibile

b.

?

' Trombone

Strings.

tip Si

La-neutrino

PROGRAMME MUSIC

1ft

The fugato, 62 bars in B minor (e), which forms the centre of the movement, its theme derived from a recitative for the bass Clarinet in the preceding movement, may be compared with the ' Convoi funebre ' of Berlioz5 Romeo Symphony l. For the Magnificat, Liszt employs the intonation of the third tone, supporting it by long-drawn consonant harmonies consisting of major chords exclusively, which towards the end, where the bass descends by slow steps of whole tones (£), produces a novel effect of great beauty :

a.

Chorus of c Women's i Voices.

Magnificat. Dolce

9-

Mag

ni.fi. cat

Do

mi . num .

1 Compare also with fngato in the third movement of Liszt's Faust Symphony.

THE ROMANTIC PERIOD

a ]

i a 1 a i s=^pgteP=q

ffl) * "S_!teg-=^

=g -i _^

l=i

^

c- 1 ) ca 1 btp 1

The ideal aimed at in <Eine Faust-Symphonic indrei Charakter- bildern l ' is akin to that of the Dante Symphony. The moods of Goethe's characters Faust, Gretchen, and Mephistopheles are depicted in three instrumental movements, and a chorus of male voices supplies a sort of comment by way of a close. The method of presentation is again by means of Leitmotive and by their reaction upon one another, as well as by allusions to those incidents of the play which are susceptible of musical illustration. The portraiture of Faust is the object of the first and weightiest movement : the Andantino which follows represents Gretchen : the spirit of Mephistopheles animates the Scherzo (Allegro ironico), in which the themes of the earlier movements are per- verted and caricatured, much as the e Idee fixe ' and the Dies irae are turned into burlesque in the Finale of Berlioz' Symphonic fantastique. A setting of Goethe's chorus mysticus 'Alles Vergangliche ist nur ein Gleichniss/ for tenor solo and male

1 Ary Scheffer's Faust pictures had some influence on the conception of Liszt's Faust Symphony.

PROGRAMME MUSIC

147

chorus, concludes the work. The changing moods of doubt, despair, desire, enthusiasm, are shadowed forth in the Faust section. Like the Inferno of the Dante Symphony, this move- ment exhibits the outlines of symphonic structure ; exposition of themes, development, recapitulation ; but the details for the most part have reference to the exigencies of the 'poetical idea,' and such exigencies are permitted to overrule considerations of musical consistency and beauty.

The rather mechanical sequence of chords with the aug- mented fifth, which forms Liszt's principal theme :

lento attai Strings. ^ Con lonlino

f

Hfcu==d

V Sit

/—

[j£ r*1 -IT

-^^^

^^-^

if

E

r^^

Oboe. .. ^

=^

Violins.

^i

^ 1 Joiente

<=* ^

rz^-^7 p=

Fii^

1^-

-D^—

-TO -3-f*r- «/ -*-'

rf^

r'i..,.;,,af.

P

4< i-

rsc

pt3& ^

ptnlendo ,

may profitably be compared with the theme of Mozart's Fantasia in C minor (a), Rossini's Guillaume Tell, Act II (d), and finally, Wagner's Die Walkiire, Act II (c).

Adagio

L 2

i48

THE ROMANTIC PERIOD

c.

The Andantino, a portion of which refers to the garden scene in Goethe's poem, invites comparison with the * Scene d'amour* in Berlioz' Romeo et Juliette. Here, as there, and for similar reasons, the thematic materials fail to produce their full effect. The Gretchen themes are delicate and tender, the dainty illus- tration of ' Er liebt mich liebt mich nicht,' is of charming naivete, and the modifications of the Faust subjects, quoted or adopted from the first movement, are sufficiently eloquent. Yet the section, taken as a whole, produces an incongruous effect J. Liszt's endeavour to illustrate certain incidents directly leads to combinations which are rather detrimental from the point of view of consistent musical design. On the other hand, it may be asserted that the recurrence of Leitmotive furnishes a clue to the spiritual connexion of one movement with another, a connexion that certainly exists in this instance, and which probably could not be so clearly expressed by other means. In the Scherzo the travesty of the Faust subjects (those referring to Gretchen are barely touched upon) is piquant in the extreme. Compared with the violence of Berlioz' burlesque treatment of the themes in the ( Songe d'une nuit de Sabbat,' Liszt's method is delicate and light-handed, equally clever in instrumental device, and far more successful in its appeal to the imagination 2.

The chorus mysticus ('Alles Vergangliche ist nur ein Gleichniss') indicates one clear limitation of the descriptive method : its inability to express those solemn emotions to which devices and tricks of phrase are inappropriate. It endeavours to complete the story by combining themes assigned to the char-

1 Just as in Berlioz' 'Scene d'amour,' where the composer tries and fails to reproduce the sense of Shakespeare's words about the lark and the nightingale. * Saint-Saens in his Danse macabre follows Liszt in pursuit of the grotesque.

PROGRAMME MUSIC 149

acters of Faust and Gretchen, but it is perforce confined within restrictions where pure music alone can move freely, and in its attempts to be reverent it approaches perilously near to dullness.

Two other orchestral Faust illustrations, ' Zwei Episoden aus Lenau's Faust : (i) Der nachtliche Zug, (ii) Der Tanz in der Dorfschenke* (Mephisto-Walzer, No. I), are best mentioned here. They belong to the year 1859, and rank as rather eccentric * paralipomena ' to the Faust Symphony.

Liszt's Poemes symphoniques (1850-60), companion pieces to the Faust and Dante symphonies, may be described as the out- pourings of a musician's fancy as stimulated by certain pieces of poetry and painting. They are as follows :—

1 . Ce qu'on entend sur la montagne (after a poem by Victor IJugo).

2. Tasso : Lament o e Trionfo (partly after Byron's The Lament of Tasso).

3. Les^JPreludes (after one of Lamartine's Meditations poetiques).

4. Orjyhfa.

5. Promethee (an introduction to Herder's dramatic scenes, ' Der entfesselte Prometheus ').

6. Mazeppa (after Byron and Victor Hugo.).

7. Fest-Klange (Prelude to Schiller's ' Huldigung der Kiinste ').

8. Hero'idefunebre (Part I of a symphony composed in 1 830).

9. Hungaria.

10. Hamlet.

11. Hunnen-Schlacht (after a painting by Kaulbach).

12. Die Ideale (after a poem by Schiller).

13. Von der Wiege bis zum Grab (after a sketch by Michael Zichy).

With the exception of the little masterpiece Orphee, which is pure self-contained music, all are impromptu illustrations, corresponding to some poem, or picture or group of concepts expressed in words. They are mere sketches arranged in

150 THE ROMANTIC PERIOD

accordance with some poetical plan, extraneous, and more or less alien, to music. The words ( Lamento e Trionfo,' which mark the two sections of Tasso, show the scheme of the majority ; thus the essence of Promethee is ( Malheur et gloire Une desolation triomphante par la perseverance de la hautaine energie '; Mazeppa's ride closes with a glorification, ' II tombe . . . et se releve roi ' ; Les Preludes ends with an Allegro marziale ; Hungaria with an Allegro trionfante ; Hunnen-Schlacht with f The triumph of Christianity ' ; Die Ideale with an ' Apotheosis 3 of certain symbolical themes. Orphee, as has already been said, is a gem in its way, and is exquisitely scored. Les Preludes, Tasso, and Fest-Kldnge bid for popularity ; they are melodious, effective, readily intelligible, but have a touch of the common- place. The two former are in reality Variations, arranged in accordance with the contrasted images contained in one of Lamartine's Meditations poetiques ({ Notre vie est-elle autre chose qu'une serie de preludes a ce chant inconnu dont la mort entonne la premiere et solennelle note ? '), and with the com- poser's own meditations on the fate of Tasso. By means of rhythmical changes, contrasting subjects are evolved from the same order of notes an ingenious device that seems to be Liszt's peculiarity, but which he derived from Berlioz, who again was by no means its originator *.

Mazeppa and Hungaria are expansions of early pianoforte pieces : the first of an Etude, the germs of which are contained in Op. i (1826-7, aet. 16), the second of a { Heroischer Marsch im ungarischen Styl' (1840-4). There is quite a family of Mazeppas, though only the two latest take the name: (i) A little c Exercise in the manner of Czerny,' Op. i, No. 4; (a) Grandes Etudes, No. 4 (1839); (3) Etudes d'execution transcendante, No. 4, Mazeppa ; (4) the Poeme symphonique ;

1 The device in question is in fact at least as old as the Partite (variations) and Suites of Froberger and other seventeenth-century precursors of J. S. Bach, who frequently constructed successive movements upon subjects consisting of the same notes rhythmically changed. In Handel's Suites, too, movements occur which are variants of one another.

PROGRAMME MUSIC

and finally (5) a transcription for two pianofortes of the latter piece. The title Mazeppa, and the reference to Byron's ' Away, Away,' and Victor Hugo's f II tombe . . . et se releve roi,' appear in the third version. The early Herdide funebre (1830-56), Hamlet (1858-61), and a late piece Von der Wiege bis zum Grab (1881-3) exhibit his method in a less favourable light. Ce qu'on entend sur la montagne (1848-56) and Die Ideale (1853-7) are the longest examples of tborough-going programme music among Liszt's works of that description. On various pages of the score of Die Ideale he has inscribed extracts from Schiller's poem, much as mediaeval painters used to add legends to parts of their pictures. Unity is sought by the use of representative themes, of which the principal one undergoes various rhythmical changes. In its first and complete shape it recalls the effect of measured prose :

Alltgro tpiritoto

« -^""V f- f* f-

py,. ,u i r~ ^:i I

ff

THE ROMANTIC PERIOD

The sections are connected, and the main lines of symphony- form can be traced, though the contours are somewhat blurred and indefinite. Auf schwung, aspiration . . . Allegro spiritoso ; Enttauschung, disillusion . . . Andante maestoso; Beschafti- gung, occupation . . . Allegretto mosso ; Apotheose . . . maes- toso, Allegretto fuocoso assai. The Leitmotive are summed up, as it were, in the final section. An episode in E major, symbolical of friendship it would seem, merits quotation for its tenderness and beauty. It is but a fragment, however, and remains undeveloped :

Andante

r— r

pp

s

5fe*

i

J.

w

m

F

r

PROGRAMME MUSIC

153

The singularities and oddities of Lis/t's style roused the ire of contemporary critics and conservative musicians such as Ferdinand David, Rietz, Hauptmann and other professors of the Leipzig Conservatorium. 'The bulk of Liszt's orchestral pieces/ it was said, 'is needlessly and gratuitously ugly he delights in eccentric combinations solely because of their eccentricity.' Censure and hostility were directed, not so much against his 'poetic' tendencies and the nature of the subjects he chose for illustration (which latter to some extent appear to justify Liszt's system of procedure), as against ill- contrived outlines, forced connexion of chords, strained melody, and bluster. Uncouth, ' impure ' harmony was the principal subject of complaint 1. Liszt's friends and pupils, when they failed to discover satisfactory reasons for this or that ugly distortion, fell back on 'poetical intentions' but an abundance of cases remained for which they found it hard to discover a plausible excuse. It is on record that certain extravagances in Liszt's early pianoforte pieces annoyed Chopin, and that sundry perversities in the Poemes symphoniques puzzled Wagner.

What, then, can have induced a born ffrand seigneur in the realms of music, a man so highly gifted as Liszt, to put his trust in harmonic crudities and chromatic horrors, from which sensitive musicians shrink with an instinctive dislike ? One fact which throws a strange light upon the matter, is that Liszt in early days (about 1 832) earnestly worked at a notion vaguely expressed 1 This was Hauptoiann's objection.

154 THE ROMANTIC PERIOD

by Fetis, the lexicographer, as a possible ordre omnitonique : which ordre, according to Fetis, might be destined at some distant date to supersede our present tonality. But it is a fact stranger still that Liszt, all his life long, should have retained such a notion, and that he desired to make, and was ever ready to encourage, experiments in tonality which led to effects of interesting ugliness l.

Another fact to be taken count of is the pianoforte virtuoso's prodigious mastery in chromatic ' double stops,' major and minor thirds, augmented fourths, diminished fifths and sevenths, and the like. This feature, conspicuous among the many specialities of Liszt's pianoforte technique, was the result of his strenuous practice of chromatic passages in every imaginable form. And it is indisputable that mechanical practice of chromatics, if carried to an excess, particularly on the piano- forte, will prove detrimental to the inner ear, and induce decadence of that native instinct for tonality and harmonic relation generally, without which a modern musician can hardly hope to keep steady and hold his own as a harmonist 2. Thus it seems sufficiently clear that Liszt's habit and his taste were subject to many an out-of-the-way influence, and, without press- ing the point further, it may be surmised that certain cases of harsh dissonance, juxtaposition of chords having little or no harmonic affinity, perverse and ugly modulation, may be the result of poetical or rhetorical guesswork, materialized through

1 Witness certain curious productions of his old age, such as the close of the second Mephisto-Walzer score published 1881 'La lugubre Gondola,' and por- tions of the last number of Annees de pelerinage, iii, 'Sursuni corda,' 1882. Not only in his original compositions but also in his numerous transcriptions and arrangements Liszt shows a predilection for harmonic piquancies and haute gouts. ' Gieb ihm Weissbrod, so streut er Paprika darauf,' said Wagner in 1877 : ' Give him white bread, he will sprinkle red pepper upon it.'

2 Berlioz, when composing, professed to prefer the guitar to the pianoforte. Wagner always made use of the latter instrument for testing purposes ; but his independence of any sort of digital mnemonics was remarkable. With him the inner ear was always paramount. For instance, the full orchestral score of Tristan was engraved long before the composer had heard a note of it, and not a note was ever altered.

PROGRAMME MUSIC 155

the pianist's fingers. Other traits, such for instance as the tempestuous chromatic crescendos in Les Preludes, the harp glissandos representing gusts of wind in the Dante Symphony, and kindred effects of crude realism got by means of rapid chromatic scales or arpeggios arranged in chords of the diminished seventh are but a reflex of certain technicalities of pianoforte playing and a reproduction of some of Liszt's peculiar ways at the pianoforte *.

Though the widest difference of merit exists between the good and bad, it must be admitted that, from the point of view of musical design, a lax and loose conception of art prevails more or less through all the Poemes symphoniques. In those pieces, in place of melody, Liszt offers mere fragments of melody touching, it may be, and beautiful, passionate or tinged with triviality; in lieu of a rational distribution of centres of harmony in accordance with some definite plan, he presents clever com- binations of chords, and ingenious modulations from point to point ; in lieu of musical logic and consistency of design, he is content with rhapsodical improvisation. The power of persistence seems wanting, orchestral polyphony is not attempted. The musical growth is spoilt, the development of the themes is stopped or perverted by some reference to extraneous ideas. Everywhere the programme stands in the way and the materials refuse to coalesce.

No doubt, Liszt as sincerely believed in the symbolical efficiency of his representative themes as Berlioz believed in the dramatic significance of Uldee fixe and La Mtlodie caracttristique. Both the great illustrators were convinced that a close union of instrumental music and poesy is possible and desirable. Both masters may have erred in their method ; and programme music, as they conceived it, may in the end prove to have been a dubious hybrid of insufficient vitality.

1 Wagner to Liszt, 1856. Letters, vol. ii. pp. 129-30: 'So gelten mir deine Orchesterwerke jetzt gleichsam als eine Monumentalisinmg deiner personlichen Kunst, und unverglelchbar, dass die Kritik lange Zeit brauchen wird, um nur irgendwie zu wissen, wohin damit'

CHAPTER VIII

OBATORIOS AND CANTATAS

THE steadily increasing tendency towards closeness of charac- terization, which forms the distinguishing feature of the romantic period, is apparent even in the work of Mendelssohn, who was by nature and training averse to innovation or experiment. Mendelssohn's attitude in regard to Protestant church music was peculiar. Looking at music mainly from the artistic point of view he thought it ill-placed in the Lutheran church service. He confessed his inability to see how artistic music could be made to form an integral part of such a service so that it would appear other than ' a mere concert with more or less of devo- tional effect* as, for instance, is the case when J. S. Bach's Cantatas are performed in church l. Accordingly, he wrote his Psalms, the Lobgesang, and the oratorios, St. Paul and Elijah, for purposes not directly connected with worship, that is to say, for the use of choral societies and with a view to performance at festivals. Thus his choral work of this description must be considered as concert music of a serious kind touched, it is true, with the spirit of devotion and with Christian symbolism, but by no means ecclesiastical.

Mendelssohn's strength in oratorio and cantata lies in the mastery of polyphonic choral technique which he had acquired by the study of Bach and Handel, in his facile gift of melody, and in his command of instrumentation. To this may be added a marked inclination towards the formal side of musical art; an instinctive love of form for its own sake; and also, perhaps, the influence of individual temperament, of hereditary

1 Mendelssohn's Briefe, vol. ii. p. 75.

ORATORIOS AND CANTATAS 157

bias, and the love of religious emotion. In the choice of the scriptural account of St. Paul, and of Elijah as subjects for oratorio, as well as in the arrangement of the ground-plan and general disposition of the materials with a view to the best musical effect, Mendelssohn relied upon his own judgement, and friends merely supplied a choice of texts and occasionally the wording of a recitative or an aria.

The book of St. Paul, which comprises a cento of texts interspersed with Lutheran chorals, is mainly concerned with the death of Stephen, the conversion of Saul, and the firmness of the Apostle's faith. In the first part of the book of Elijah, the symbolical element which a Christian mind may find in the Old Testament appears to be latent only; but it occupies a more prominent position towards the close of the second part, where allusion is made to the coming of Christ. The Biblical account of Elijah forms the groundwork, while a superstructure of texts is arranged so as to serve musical purposes. The framework consists of a series of scenes and situations laid out dramatic- ally, and held together by the figure of Elijah. The narrative in the form of recitative which in St. Paul connects the vocal numbers is eliminated, so that in this oratorio the principal character, the subordinate characters, and, in some instances, even the chorus, are made to speak dramatically. The book embraces just enough action and contrast to sustain the interest, and allows enough of expansion to give a chance to the full musical expression of the inner aspect and the inner ground upon which the events are set forth.

In St. Paul the narrative consists of recitative delivered by soprano and tenor voices in turn, with simple accompani- ments for strings. The declamation, in German, is precise and accurate, though without undue insistence. When the narrative demands the expression of emotion, the recitative rises to cantilena as at the death of Stephen and such touches are among the best and most sincerely felt details resulting from the composer's mode of treatment. Side by side with such

158 THE ROMANTIC PERIOD

touches, rank the poignant accents of remorse and contrition in pieces such as the air, ' Lord, a broken heart, and a contrite heart' ; the burst of rage and hate, ' Confound them all, Lord Sabaoth'; and the expression of strong indignation, as in the aria, ' Is not His word like a fire ?'

From a musician's point of view, the weight and power of Mendelssohn's oratorio music is best seen in the choruses. They are masterly in the full sense of the word, characteristic and varied, and evince a command of contrapuntal resource beyond the reach of any of his contemporaries. The following are conspicuous examples from St. Paul : the powerful opening chorus in A, ( Lord, Thou alone art God ' ; the chorus, f Rise up, arise and shine ' ; the choruses of Hebrews, ( Now this man ceaseth not to utter blasphemous words,' and ( Stone him to death ' ; the chorus of Gentiles, ' O be gracious, ye immortals ' ; the earnest and serious chorus of Christians, in D minor, ( But our God abideth in heaven,' which contains the melody of a chorale, cHis will directeth all the world,' inserted after the manner of J, S. Bach. As in Bach's Passion-music, harmon- ized chorales are introduced at intervals between the airs and choruses to act like ideal comments upon events, and so as to enforce a mood or close a section.

The secret of the greater success especially in England of the later oratorio, Elijah, lies in its dramatic scheme, and in the enhanced opportunities for solo and concerted music which it afforded the composer. In Elijah, Mendelssohn exhibits his talent at its full maturity. Taken as a whole and compared with St. Paul, Elijah stands on a higher plane. It is stronger in spirit, freer, broader, more direct, and less tinged with Lutheran influences. Whether the composer was conscious of the fact or not, it is evident that every conspicuous feature in the method and technique makes for characterization. The dramatic force of the scenes, their conciseness and vivid colouring, the pointed rhythms of the choruses, the highly wrought orchestration, the energy expressed in the choruses of Baal's priests, the fierce

ORATORIOS AND CANTATAS 159

savagery of ' Woe to him, he shall perish,' and certain almost histrionic effects in the rain scene, all point in the same direction. The influence of the Lutheran service, which in St. Paul is apparent in the symbolism of the chorals and in the pietistic tone of certain airs and choruses, is less conspicuous in move- ments of the same order in Elijah. The dramatic sections of the latter oratorio are as free, as vivid, and as frankly pictorial, as anything in Handel. As an example, note the impressive ' prophecy of the drought ' which, preceded by four solemn chords of brass and placed before the overture, produces a very remarkable effect. Further examples are contained in the scene of the widow, the priest's scene with the flesh, of fire from heaven, the scene of Elijah standing on the mount, and above all the great scene which describes the giving of the rain ; all striking and expressive in idea, all remarkably original in form and workmanship. The choruses at the beginning of Part I describing the drought, and also the duet and chorus following 'Lord, bow Thine ear to our prayer,' are both touching and impressive. In Part II the splendid chorus, ' Be not afraid,' and the chorus, f He that shall endure to the end,' are typical examples of the master's sentiment in music and of his great knowledge of effect. The chorus last named Andante sos- tenuto, F major is especially attractive ; it contains nothing that resembles or even aims at passion, nothing that is even telling, it is merely the record of a tranquil and peaceful mood, truly and beautifully expressed. Wonderfully characteristic in their way are the choruses of the priests of Baal. In the scene of Elijah standing on the mount, the repetitions of ' Behold, God the Lord passed by,' produce a kind of rhythmical effect, framing, as it were, the description of storm, earthquake and fire. An impression of resistance and striving against natural phenomena, and at the same time an irresistible onward impulse, is conveyed to the mind by the use of choral imitations leading to the crisis, c And after the fire there came a still small voice : and in the still small voice onward came the Lord.' After the

i6o THE ROMANTIC PERIOD

Holy, Holy/ of the angels, and the arioso, f For the mountains shall depart and the hills be removed,5 Elijah is silent ; his ascent to heaven is depicted in the chorus, ' Then did Elijah the prophet break forth like a fire.' The five numbers following, which point to the coming of Christ, form rather an anticlimax and are not particularly interesting as music.

Mendelssohn intended to compose another oratorio, Christus, as a sort of sequel to Elijah. He began to work at it in 1847, the year of his death.

Apart from the two oratorios Mendelssohn's other choral works demand little comment. Planned as a kind of choral symphony, the disposition of the Lobgesang (1840) is open to a charge of tautology, inasmuch as the choral movements follow- ing upon the instrumental, cover the same ground, and act as a restatement by the aid of words of that which has been already expressed by the orchestra without them. Connexion and unification is attained by the use of a leading phrase or motto, like Berlioz' ( Idee fixe/ which consists of the intonation of the eighth tone as sung to the Magnificat, and forms the subject of the first movement of the ' symphonia,' the first chorus, and the close of the cantata.

TONE VIII.

(\/L >sr-_

r=«r-— H-=ej

69 <=> =-

Mag -

»«« 1 j 3

ni.fi . cat.

WM^ 1""

_JL_-T JLJ-n

Al . les

-j: H

was O > dem bat.

Like the overtures to Mendelssohn's oratorios, the three orches- tral movements of this work are fluent and finished but not very fresh or spontaneous. The choral movements, six in number, have full measure of the technical merits that always characterize the master's choral pieces ; yet they do not come near to the direct impulse and power, or to the warmth and tenderness of feeling he managed to express in certain choruses belonging to

ORATORIOS AND CANTATAS 161

Elijah and St. Paul. The entire work consists of serious music, individual and independent in style, but neither inspired nor inspiriting.

In the three motets, Op. 39, for female voices and organ (1830), and in the Lauda Sion, for chorus and orchestra (1846), Mendelssohn tried to compose in a style suitable for the Roman Catholic service. The Psalms for chorus and orchestra, e As the hart pants/ f Come, let us sing/ ' When Israel out of Egypt came/ ( Not unto us, O Lord/ though more or less important in the list of his works, belong to the Protestant concert-room.

The case is very different with Goethe's wild scene, ' Die erste Walpurgisnacht J (1831-32-43), which forms the subject of a cantata expressly intended by Goethe for music, and laid out with a view to that end :

Miige dies den Sanger loben, Ihm zu Ehren war's gewoben.

This scene is one among three cantatas contained in Goethe's collection of f Books of words for music J a companion piece to the RinaldOy so beautifully set by Brahms. The poem is triumphantly successful a masterpiece in a field where master- pieces are extremely rare ; so simple and consistent in motive that it seems like one spontaneous growth, so direct in its appeal to the imagination that there is no need of scenic accessories, yet so precise and vivid in its fantastical way, that it could be set, sung, and acted on the stage, without the change of a scene or a line. Whether it was expressly written for Mendelssohn does not appear, but it exactly met his views and afforded fine oppor- tunities of which he thoroughly availed himself. On the approach of spring, in face of warning that heathen practices are punished with death, Druids assemble on the hill-tops at dusk to kindle flames in honour of All-father. Below, on the borders of the wild wood, they post sentinels who all night long, with diabolical din, feign a witches' orgy to scare the profane. Christian watchmen hold aloof for fear of devils, whilst the fires burn above, and the worship proceeds. As has already been

DARRKXUTBXK

i6a THE ROMANTIC PERIOD

pointed out, Mendelssohn in the f Overture ' to this cantata is a musical illustrator with a definite idea before him. In this Introduction for such it is and not a true overture he depicts in a realistic manner the rough winter weather and the gradual approach of spring ; then, by a series of vocal solos and choral songs, he leads up to one of the most vivid and striking of his pieces the wild and fantastic chorus, l Kommt mit Zacken und mit Gabeln,' Allegro molto, A minor, f , which recalls certain weird instrumental effects first heard in the incantation scene of Weber's Freischutz. A kind of hymn for baritone solo and chorus Andante Maestoso, C major, £, rounds off and closes the work.

Passing mention may here be made of the works which Mendelssohn wrote to order for the Court Theatre at Berlin : overture, solos, and choruses, to Racine's Athalie (1845), and the music to Donner's translation of Sophocles' Antigone (1841), and of Oedipus in Colonos (1845), for male voices and orchestra. The fragments of an opera, Lorelei, the text by Geibel, are essentially concert music, and one fails to see how it could gain by stage action.

Much more than in Mendelssohn's oratorios, the spirit of romanticism with its innocent striving after emotional expression pervades Schumann's Das Paradies und die Peri, Der Rose Pilgerfahrtj and other choral works for the concert-room. The presence of romantic emotion is felt throughout even in bits of choral description, and in solo narrative, which in place of the ordinary recitative often rises into definite arioso.

After having produced a large amount of original and very interesting music for pianoforte solo, Schumann, about 1840, began in splendour the career as a composer of songs and choral pieces which he ended in darkness. In the zenith of his brilliant genius he put forth a treasury of song ; from about 1850 onwards, in failing health, he flooded the market with vocal mediocrity. His secular oratorio, Das Paradies und die Peri (1843-4), consists of the translation in rather perfunctory

ORATORIOS AND CANTATAS 163

verse of Moore's story of 'Paradise and the Peri,' slightly expanded, and set as a series of three cantatas for solo voices, chorus and orchestra. It made a great impression and was immediately popular in Germany. Novel in style, romantic and sentimental in spirit, it won the sympathy of the upper bourgeoisie, and pleased the members of Singing Societies. To the luxurious oriental pictures of Moore's poem, the music adds a warmth of feeling peculiarly German and peculiarly Schumann's own. The airs contain an abundance of sweet, perhaps over-sweet, melody ; and several of the lesser choruses or choral songs are pleasant to sing, as, for instance, the delicate f Sleep on, in visions of odour rest,' which closes the second cantata, the canon for female voices called the ' Song of the Houris/ and the choruses of the Genii of the Nile the two latter forming part of the few slight additions to Moore's poem. The larger choruses, however, fall short of the mark. Schumann's choral technique is almost always defective*,. It will not bear comparison with that of Handel, Bach, Haydn, or Mendelssohn. It is often inept, inefficient, and trying to the voices. The lack of practical experience is seen to be a real drawback. Despite strenuous efforts and a not inconsiderable display of contrapuntal skill in choruses such as ' Denn heilig ist das Blut,' and ' O heil'ge Thranen inn'ger Reue ' the result is disappointing. Yet such occasional shortcoming in choral and orchestral effect does not detract much from the originality of the work taken as a whole. The true source of its weakness lies rather in the disposition of the poem and the decrease in interest as it proceeds. Moore's story of the three gifts that should open the gate of Eden to the disconsolate Peri exhibits a sequence of events diminishing in excitement and picturesqueness. From f Mahmudh of Gagna's wrath ' and ' The warfield's bloody haze,' to ' Blest tears of soul-felt penitence,' there is a diminuendo very detrimental to the total effect of the music.

The historical interest which belongs to Paradise and the Peri will probably be found to exceed the intrinsic value of the

M z

164 THE ROMANTIC PERIOD

work. It is the first instance of a long poem not especially written for music, being set from beginning to end without interference with the verse or the disposition, and with minute attention to the poet's meaning. Schumann's music covers Moore's poem like a close-fitting garment. Formerly, poems intended for music were expressly contrived to meet the real or supposed exigencies of the art ; or else composers appropriated some poet's work, and cut, compressed, or by means of repeti- tion, expanded it to suit their particular purposes as, for instance, Handel expanded the f Alexander's Feast ' of Dryden. It was not until after Beethoven's time that musical resources were sufficiently developed to allow of a composer attempting so difficult a task as the complete absorption and adequate reproduction of a poem of some length. Nowadays, and since Schumann set the example, any poem that is musical in its nature may be aptly treated, and presented in conformity with the poet's intention. It has been proved that the sequence of moods which makes a good design in poetry will also make a good design in music *.

With the exception of Paradise and the Peri, a portion of the music to the second part of Goethe's Faust, i. e. the ( Transfigu- ration,' the 'Requiem fiir Mignon' (1849) one of the poems contained in Goethe's Wilhelm Meister, Op. 98 b and the short Nachtlied, Op. 108 (poem by Hebbel), which also belongs to 1 849 and is both poetically and musically suggestive and original, none of Schumann's choral works have stood the test of time. Der Rose Pilgerfahrt, a companion piece to Paradise and the Peri, was a production for the market. The music to the Adventlied a kind of choral ode (poem byRuckert), 1848 is reminiscent

1 Beethoven's cantata Meeresstille and Mendelssohn's Walpurgisnacht, together with Schumann's successful initiative in Paradise and the Peri, and in part of his music to Goethe's Faust, may be regarded as the starting-points of that extensive development of the Ode and Cantata which starts from Bach, and which Brahms furthered in his setting of Holderlin's Schicksalslied and Goethe's Rinaldo. The development culminates in some recent versions of great English poems by English composers.

ORATORIOS AND CANTATAS 165

of Mendelssohn's Psalms, but does not reach their level of excellence. The Balladen for solo voices, chorus and orchestra (1851), Der Konigssohn, Des Stingers Fluch, Das Glilck von Edenhall (after poems by Uhland), the four Balladen by Geibel (1852-3), entitled <Vom Pagen und der Konigstochter/ are failures one and all. The motet, l Verzweifle nicht im Schmer- zensthal,' for a double choir of male voices, with organ ad libitum, 1849 (poem by Riickert), has several fine moments, but it is monotonous, long, and heavy; and, owing to certain chromatic progressions, rather difficult to sing. There is hardly any relief to the forte of the chorus and organ, and the effect of the intricate harmonies at the words ' Harre aus im Leid ' is distressing. For the sake of completeness certain lame and tame attempts at church music a Mass and a Requiem, both composed in 1852 and published after the composers death may be mentioned here. The Requiem has a few points of interest. The Mass was stillborn. In the Requiem the opening and closing movements, ' Requiem aeternam dona eis Domine/ which are written in a tender and delicate strain the fugal * Pleni sunt coeli/ which has a certain degree of life and power portions of the closing section of the Benedictus and also the Sanctus and Agnus Dei, are still worth the attention of the student.

The ecstatic verses of the closing scene of Goethe's Faust, Part II, the Transfiguration, exercised a peculiar fascination over Schumann's mind; he set them line by line, for solo voices, chorus and orchestra. The music pertaining to this section belongs to his best period, about 1 848 ; it is complete in itself and important. In later years (1853-6) he tried his hand at certain other scenes from both the first and second part of Faust. Ultimately the detached pieces were gathered together and prefaced by an overture, so as to form a kind of concert oratorio. Thus the huge posthumous publication entitled ( Scenen aus Goethe's Faust ' is in fact a mere conglo- meration, having no more real coherence than can be claimed

1 66 THE ROMANTIC PERIOD

for its rival, Berlioz' Damnation de Faust. The first division consists of portions of Goethe's garden scene, Gretchen before the shrine of the Mater dolorosa, and the scene in the cathedral. The second contains the scene of the dawn and Ariel's hail to sunrise, the scene of the e four grey sisters ' Want, Guilt, Misery, and Care striking Faust with blindness, and the scene of Faust's death. The third, as has been stated, is concerned with the mystical Transfiguration, the translation of Faust's soul to heaven.

The spirit of Goethe's great scene in heaven exactly suited Schumann's temperament and stimulated his genius to its best. He seems to have had no difficulty in dealing with the verse. Except in the chorus * Gerettet ist ' and the chorus mysticus at the end, hardly a line is repeated. There is little flagging or shortcoming ; things seem to take form and fall into their place quite naturally. The rhythmical variety of the verse is reflected in the movement and speed of the music ; and the gradual increase of animation and ecstasy expressed in the words of the Hermit-fathers and the Doctor Marianus, in the chorus of blessed youths and the younger and elder angels, the chorus of female penitents, and the voice of ( una poenitentium sonst Gretchen genannt ' is rendered to perfection.

The delicate swing and balance of rhythm in the choral song, * Dir, der Unberiihrbaren,' is a good specimen of the ease and spontaneity that distinguishes Schumann's declamation. The Doctor Marianus and chorus address the Mater Gloriosa thus :

ORATORIOS AND CANTATAS JTV

167

r

die leiclit Ver . fUhr - ba - ren Truu-lich zn dir

nom men, Dass

CHORI-S.

*-J 'fl: J. <=3»t

der Un - be rUKr - ba - ren, 1st ea nlcht be

kom-men, Dir,

4

die leicht Ver - fUhr - ba - ren Trau-lich zu dir kom-men. I

j±±4^

&c.

From the beginning to the end of the Transfiguration, Goethe's verse is reproduced in melody of equal beauty and subtlety. Of the opening and concluding choruses two versions have been published. The principal choral numbers are difficult to sing ; but the effect of a correct rendering is distinctly fine almost throughout. The music emphasizes the points of the poem, and explains its meaning more effectually than any commentary. This is particularly the case with the words of the Doctor Marianus, and the final chorus mysticus :

Alles Vergiingliche

1st nur ein Gleichniss ; Das Unzulangliche,

Hier wird's Ereigniss. Das Unbeschreibliche,

Hier 1st es gethan ; Das Ewigweibliche

Zieht uns hi nan.

All things transitory

But as symbols are sent ; Earth's insufficiency

Here grows to event ; The Indescribable,

Here it is done ; The Woman-Soul leadoth us

Upward and on.

168 THE ROMANTIC PERIOD

If Schumann's Faust music is but a conglomerate of scenes from Goethe's poem arranged in a sequence for concert purposes, the so-called L&gende dramatique, La Damnation de Faust, by Berlioz (1846), may be described as an accumulation of stray pieces some taken from the composer's first publication, f Huit scenes de Faust, oeuvre Ier,' which was cancelled, some totally alien, like the arrangement of the Rakoczy March, and some few specially written to fill up gaps in the curious scheme of a grand opera or rather of an opera de concert, in accordance with which the materials were finally put together. The designation 'Legende dramatique' (Legende fantastique would have been more appropriate) merely serves to conceal the absence of any definite plan. In his preface, Berlioz is quite frank on this subject. ' Pourquoi 1'auteur a-t-il fait aller son personnage en Hongrie ? Parce qu'il av; it envie de faire entendre un morceau de musique instrumentale dont le theme est hongrois. II 1'avoue sincerement. II 1'eut mene partout ailleurs, s'il cut trouve la moindre raison musicale de le faire.' As to the violence done to Goethe's poem, the author of La Damnation asserts that ' Le titre seul de cet ouvrage indique qu'il n'est pas base sur 1'idee principale du Faust de Goethe, puisque dans 1'illustre poeme Faust est sauve.' Portions of the libretto are taken from Gerard de Nerval's translation of Faust; the first, fourth, sixth, and seventh scenes were written for Berlioz by an acquaintance, the remainder by Berlioz himself.

The separate airs, choral songs, and orchestral pieces, which go to make up the four main divisions are picturesque in effect, highly coloured, and superbly scored. As regards intrinsic musical value they differ greatly. Occasionally, in scenes such as Faust in the fields, or Marguerite in her room alone, when the situation appeals to the sober and more serious side of his mind, the composer rises to heights of noble and original music. In the Chansons called ' Histoire d'un Rat,' ( Histoire d'une Puce,' and in the ( Serenade du Diable/ the comical and ironical sense of the words is well brought out. In this last piece (which in the

ORATORIOS AND CANTATAS 169

suppressed Opus I appears as a song for a tenor voice with guitar accompaniment), in Marguerite's beautiful romance just men- tioned, ' IVamour 1'ardente flamme/ in her Chanson gothique, <Le Roi deThule/ in the gracefuland delicate 'Danse desSylphes,' the ' Menuet des Follets,' and the famous * Marche Hongroise,' the cleverness of the instrumentation is remarkable even for Berlioz, with whom the treatment of the orchestral instruments is always a strong point. The choral numbers are comparatively weak. The Easter hymn, ( Christ vient de ressusciter,' is not particularly impressive. The burlesque fugato on the subject of the song of the Rat, ' Requiescat in pace, Amen,' is in fact rather a display of Berlioz' shortcomings as a contrapuntist, than a telling skit on fugal composition as he meant it to be. Compare the queerly contrapuntal double chorus for male voices (soldiers and students) in La Damnation de Faust.

The * Course a 1'abime ' (Faust's and Mephistopheles' ride to hell), and the scene in ' Pandaemonium/ with its 'Choaur de demons et damnes' screaming in Swedenborgian 'langue infernale,' attempt to inspire terror, but achieve no more than distaste, whilst the 'Epilogue sur la terre,' and the chorus of blessed spirits singing the f Apotheose de Marguerite dans le ciel,' are trite and rather insignificant. They would appear wholly commonplace were it not for a new and beautiful effect at the end which is obtained by an ingenious combination of harps and boys' voices.

All the numbers airs, vocal ensembles, and orchestral pieces are more or less redolent of the opera. There is a sense of theatrical rhetoric, of musical pose and tirade, quite correct in its way and perhaps effective in an opera de concert, but not always good music. At times the instrumental noise as in the 'Marche Hongroise' and the scene in 'Pandaemonium' is almost deafening. By devices of orchestration the colourist tries to reach that which the melodist fails to attain for want of warmth, the harmonist for want of power, and the designer for want of skill. Everywhere the composer is prone to speak in tones

170 THE ROMANTIC PERIOD

acceptable to that kind of public taste which is caught by peculiar tricks of sound or glaring colours. For the most part the music is histrionic in style, it stimulates the nerves and is frequently constrained to attract attention by strange or barbaric sounds, or to excite interest by appeals to ideas and associations outside the pale of musical art. Purely musical expression, as we find it in Bach, or Beethoven, or Brahms, may not have been intended ; but even where it does in fact exist, it is not well sustained, and its quality is hardly pure enough to tell as absolute music. From a rhetorical point of view, Berlioz in this his fo3uvre le plus accompli ' is successful enough ; not so, tested by the standard of the greatest masters.

Berlioz5 Salon Oratorio, the Trilogie sacree, L'Enfance du Christ (1854) a series of three cantatas consists of nine short scenes of studied conciseness, irregularly grouped in three divisions: I. Le Songe d'Herode; 2. La Fuite en Egypte; 3. L'Arrivee a Sa'is. The music is daintily written for a small chorus, small orchestra, solo voices and solo instruments. By the side of much that is, essentially dull like the 'Marche nocturne/ the overture to fLa Fuite en Egypte,' and ' L'Arrivee a Sa'is ' or both dull and grotesque, like the conjuration of the soothsayers (said to be based on Hebrew tunes) this triptych of cantatas contains several movements distinctly interesting and charming from a musical point of view as, for example, the choral song, ' II s'en va loin de la terre,* the Choeur mystique, *6 mon ame* (which Berlioz palmed off on a select Parisian audience as the work of a for- gotten mediaeval composer), the duo fLJ6table de Bethleem,' and the interlude fLe Repos de la Sainte Famille/ The latter movement is most delicately scored for wood wind and strings, to which at the end are added a tenor voice and eight female solo voices representing angels at a distance. The following extract will give an example of its melody :

ORATORIOS AND CANTATAS

171

S^T1

"il- <• i

r^-r

^ J^^

~f-^ ^=— H*

fejTfff

&c.

r r

Two pieces by Liszt may fitly be mentioned here. One is a Chorus of Reapers, A major, f No. IV of an ambitious set of choruses for Herder's ( Entfesseltem Prometheus' (1856), to which the Poeme symphonique * Prometheus' forms the over- ture. An oasis in a desert, this f Chor der Schnitter ' is a good specimen of a Pastorale, closely akin to certain pieces belonging to the cancelled Album Suisse and now contained in Les Annees de pelerinage, i, such as 'Eglogue' and 'Au bord d'une source.' The other is a setting of the opening scene of Longfellow's Golden Legend (1875), * Die Glocken des Strass-

172

THE ROMANTIC PERIOD

burger Miinsters,' for baritone solo, chorus, full orchestra, and organ. This approaches the category of Schumann's Balladen, being rendered in the manner of a dramatic dialogue between baritone and chorus. The liturgical motif of the Preludio recurs in the Vorspiel to Wagner's Parsifal, and elsewhere in that work.

CHAPTER IX

RELIGIOUS MUSIC

THE Romantic movement in France was a revolution which invaded not only the theatre and the concert-room but the Church also. Not that it set up any worship of the ' Goddess of Reason ' from this, indeed, it was in many ways removed but it made a definite and conscious effort to break through all ecclesiastical restrictions, and to treat religion as one department of human feeling, different in object but not different in kind from the sentiments evoked by natural beauty or by the tragedy of history and romance. Liszt, in a letter to the Gazette Musicale (1834), described his own and Berlioz5 ideal of romantic religious music thus : ' For want of a better term we may well call the new music Humanitarian. It must be devotional, strong, and drastic, uniting on a colossal scale the theatre and the Church, dramatic and sacred, superb and simple, fiery and free, stormy and calm, translucent and emotional/ And Berlioz furnished a practical example of what such music would be like, when in 1 836-7 he composed the Requiem Mass which was performed in the church of the Invalides at a memorial service for General Damremont and the French soldiers who fell at the taking of Constantine.

The words ' on a colossal scale, theatre and Church, dramatic and sacred' suffice to indicate Berlioz' views when he wrote

174

THE ROMANTIC PERIOD

this wilfully eccentric score. He particularly emulated, and hoped to eclipse, the fame which Cherubim, the then Director of the Paris Conservatoire, had gained with a grand Requiem mass in C minor1. When, in the most striking movement of Cherubim's Requiem, to the trumpets and trombones of the * Dies irae,' a gong is struck, fortissimo, we may readily under- stand that the impressiveness of the subject has passed beyond the ordinary resources of the composer. But what is the beating of a gong or a sudden clash of cymbals to Berlioz' Tuba mirum, with its ( Quatre orchestres d'instruments de cuivre places aux quatre angles de la grande masse chorale et instrumentale,' together with sixteen kettledrums and a grosse caisse roulante, a gong, three pairs of cymbals, and another grosse caisse f avec deux tampons' ? No such volumes of sound had been heard in Paris since the taking of the Bastille. No such orchestra had ever been collected Gossec's Te Deum is pale and ineffectual in comparison.

There is not space to quote the host of instruments in detail,

Tuba minim.

Andante maestoso m = 72.

RELIGIOUS MUSIC

175

Laiucz le mouvement t'animer Mi peu.

tt^SiE

but some of the drums, the cymbals, and the gong, figure in the score as follows :

THE ROMANTIC PERIOD

Wood wind.

Brass.

Eight pairs (alternating)

of Kettle-drums.

Two big

drums.

Gong and

three pairs of

Cymbals.

Sopranos and Tenors.

CHORUS.

Basses.

Strings.

*

_

.g^-i-Sa- f jg-. .. _ | -SB ^r=q

S-l be? \k<a> I bag - ^H

jf Tamtam.

*

==| ^=j

[11 faut f rapper sur six cymbales avec des baguettes a tete d'eponge.]

_IN ^

-j ib

In - dex,

1

iu-dex er

8P

Iu-dex, iu-dex er - - - go cum se-

ff

RELIGIOUS MUSIC

177

-£EEl

$

,

ff &

ff ^

Sat

ff &

-t - r- " - 1 -J- -*-••*- 3 i

go cum oe » <!«• - bit,

cum Be - de - bit,

- cic - bit,

cum se - de - bit,

DAKNHEUTIIER

i78

THE ROMANTIC PERIOD

After this the 'cellos and basses resume the tune of the ( Dies irae,' piano, accompanied by little phrases played upon the corno inglese and bassoons, while the tenor voices, also piano and { avec un sentiment d'humilite et de crainte/ sing 'Quid sum miser tune dicturus/ This short movement makes a subtle appeal to the imagination. It acts, of course, as a contrast and relief, and is, in its way, both novel and beautiful, though the musical substance does not rise much above mediocrity. A similar effect of contrast is expressed in the next movement between the fortissimo, { Rex tremendae maiestatis/ and the supplication, piano, f Salve me/

The monotonous psalmody of the Offertoire produces a cal- culated effect on the nerves *. As in the Convoi funebre from the symphony Romeo et Juliette, the device here consists of a long-drawn instrumental fugato, 154 bars (A), interrupted at intervals by a plaintive wail of voices on two notes only which are repeated again and again (B).

Moderate J Sopranos.

84.

w=±= =J

Tenors and Basses.

A,

Violins. _____ =^^;

^.•T—7—=Z J

H— *H

g . , %i

i d JH

D.

Do

m

1 Wagner admired the skilful contrivance of this Offertoire, and repeatedly spoke of the singular effect it produces : ' If a man cares to be hypnotised musically, here is his chance.'

RELIGIOUS MUSIC

W

179

Do

ne,

Wood wind. «/

Wood wind.

^^

5

The 'Hostias' which follows depends for its effect upon a surprising trick of instrumentation, i. e. chords played piano, crescendo, and again piano by three flutes and eight trombones :

3 flutes.

8 trombones.

Tenors.

Basses.

I

1

±1:

mf Ho» - ti - as et pre - ces ti - bi laud - is of . f«r - i -

»

r r *

x

P^

sfe

This contrivance, though but little suited to the dignity of the occasion, is a capital example of Berlioz' ingenuity and

N 2

i8o

keenness of perception in matters of instrumental effect, and of the marked attention he paid to those remoter capabilities of the orchestral instruments with which the majority of professional musicians in his day were not familiar. The score contains a note drawing attention to a particularly clever con- trivance in the employment of certain bass notes of the tenor trombone at the end of the movement *.

3 flutes.

8 trombones. ,

!SF •" -^

p-=

ev - ,

p^=

§

»—

•M U

^. jg~~i

The most musicianly number is the opening ( Requiem aeternam,5 which is repeated in a shortened form at the close of the work. It is original, sincere and consistent; marked by a touch of austerity which is rare in Berlioz, and approach- ing more nearly than was his wont to a contrapuntal style. Its various themes are effectively contrasted : the sombre melody with which it begins, the whispered monotone of its fKyrie eleison/ the transition to the major mode at the words 'Te decet hymnus/

' ' Ces notes graves de trombone te*nor sont peu connues meme des exe'cutants ; elles existent cependant et sortent meme assez aisement lorsqu'elles sont ainsi amene'es.'

RELIGIOUS MUSIC

181

A ndante, un poco lento, •' = 69.

Soprani 1 and 2.

Tenori 1 and 2.

Bassi land 2.

£T ' Imi Soli.

Be - qui . Imi Soli, mezzo, voce g . ^.

Re - - qai -em.. ae . ter

lempre ft =••

^ ^ J ' J J J 1 J. ^

Imi Soli, metza voce

Re

qui - em

em ae » ter

do - na

i

-I c-l

=3=

r

182

THE ROMANTIC PERIOD

i

ter

•un poco piii

Do

mi ne,

qui-

re qui » em ae ter - naua do - na

r

do na e is, do

ae - ter

is, Do

mi . ne,

r r f

r

RELIGIOUS MUSIC

183

i

to*-J^H£

Oft e - is,

Do

mi - ue,

& -tt

SE

Imi & 2di

qui > em . ae ter

-J^— .j-

i*Hm

Sans pretser tt largcmtnt

Tenori

Soli.

Te de . cet hym nua,

De-ua, in

Si ' on

Ky.ri - e e.le.i-son, Ky-ri . e e-le-i-son, Ky-ri

i - son.

1 84

THE ROMANTIC PERIOD

And as this is the most ecclesiastical number so the fLacry- mosa ' is the most operatic. Written with every consideration for the voices, and admirably scored, it is wholly theatrical in character ; it is no more suited to the Church than Rossini's Stabat Mater, and in the midst of all its artificial passion and histrionic effect it even stops to borrow a moment's in- spiration from Bellini :

Unison. Dolce assai

Sop. 2di

K£^—

Ten. Imi

La .

P cry

w

mo ' j

^^

aa 0 ^

•fl | ! I

1 1 il

dl - es

2E

' 1 J J^

J * J~

La - cry - mo

-1 l—T-1— -N

il . la

qua re -

jTT"j ""•< a=

mr - get, qua re -

V »*• C-ff ^ •LJ-J—

1 __j » J_* L

1

di.es il

la,

/ 3: I 1 r* V

gi-3 -<J-- Jj J:

- aur - get ex fa

SB j-, =

. vil - la.

v ^— -^ •'-J J^ J .

di

es il - - la .

In other movements, however, notably in the unaccompanied { Gluaerens me ' this charming trait appears as a mere trick that comes in conveniently when contrapuntal invention fails. Some- times, too, as in the ( Dies irae,' the choral writing has an instru- mental air. The words are put under the notes, in accordance with the popular French formula of progress, l Marche, ou je t'assomme V

1 Cited by George Sand in Lettres d'un voyageur.

Sopranos.

Tenor*.

Basses.

RELIGIOUS MUSIC

Modtrato f

jj'" "V pTfg= =5 ^

gu ^

Di - . - - es il - - la

f

\ SLtt-b 1

i i ' 1 ' ' "*"" '

Quau - tus tre mor

st fu - tu rus,

=rd^:_^ n

H

aol -

if

P

Di - es i . rae, di - ea

i) - l.i, di et i- rac,

quan

i

<p

&**=?

clum.

m

il - la,

do

dez,

di - es il - la, sol vet aae-clum.

*

I

do iu - dex.

Curiously enough, the instrumental phrases thus sung by the tenors recur in the Offertoire (bars 96-105), where, more appro- priately, they are allotted to the strings.

Taken as a whole, the Requiem shows little of religious resignation, quiescence, or repose of the soul. Grandiloquent effects predominate. It is musical scene-painting ; illustration on a large canvas and with glaring colours.

The Te Deum (1849-54) was performed once only during Berlioz* lifetime, and under his personal direction, in the church of St. Eustache, Paris, April 30, 1855 to celebrate the opening

1 86 THE ROMANTIC PERIOD

of the Palais de 1' Industrie on the day following. For its due performance the composer demands an orchestra of 134 executants, a double choir of 200 voices each, and a choir of 600 children's voices1. In a note prefixed to the score, he directs that the orchestra and two principal choirs, with the third choir on a separate platform and at some distance from them, are to be placed at one end of the church opposite to the organ.

It is difficult to speak plainly and without exaggeration of this very ambitious but very disappointing work. The rather limited range of Berlio/' melody has already been criticised, and there is little evidence of increase or development in the measure of his genius. Indeed, the later date of the Te Deum (1854) as compared with that of the Requiem Mass (1837), and of the operas Les Troyens and Beatrice et Benedict as compared with Benvenuto Cellini,, would seem to indicate a decadence of inventive power. Musicians are agreed that strange instances of inept harmony or melody occur in the early works, and it cannot be said that they are less frequent in the later ; indeed, inepitude gradually becomes more obvious. In a general sense the Te Deum is distinctly inferior to the Requiem. The composer alters the arrangement of the ritual text, and repeats or omits words for the sake of dramatic effect. Side by side with passages undeniably clever though rarely beautiful, it contains much that is inchoate, dull or perverse, and much that is marred by abstruse and laboured details, instances of vain effort and errors of tact and taste which 4a man of forty ought not to have committed2/

1 The use of such choral Masses was not an entire novelty in the churches of Paris at that time. Mehul, in the Chant du n juillet 1800, had already employed three independent choirs and full orchestra.

2 Schumann : private letter to Amhros. Franchomme, the violoncellist, Chopin's friend, asserted that as early as 1833 Chopin declared that he had expected better things from Berlioz, and that Berlioz' music was of ' such a quality as to justify any man who chose to break with him ' a singular expression of opinion on the part of one so reticent as Chopin, if it is strictly true as reported.

RELIGIOUS MUSIC

187

A lltgro moderate

Sopranos.

Tc De - um lau da » - mua, te Do - iui-nuiu con. ft

Alltgro non troppo Tenon and Basses.

. Oi fS . -j I

b- I g) * <h ^-\-r p-

Tu, Chris te,

tu

Rex glo - ri - ae,

*

Fa tris seiu - pi - ter nus Pi - li - us.

These quotations, which show the theme of the first number, the 'Te Deum laudamus' and the scale-like theme of the fTu Christe/ cannot be said to possess much distinctive character. Neither of them is particularly suitable for contrapuntal treat- ment, and the manner in which they are developed is never really effective. Although the orchestral and choral means employed are abundant and the cumulative mass of sound is great, both movements are tedious.

The sixteen bars of an Andantino which form the opening and the end of the second number Hymne, e Tibi omnes angeli incessabili voce proclamant* are perhaps the most fluent in the entire work.

i88

THE ROMANTIC PERIOD

.$>— ;

'•=3=

r

^

Pff

J J J J .fe

The close of the third number < Priere : Dignare Domine '- on the whole a weak number, approaches the same level :

Moderate quasi andantino, = 69.

~J »

Mi . se - re

re nos

tri, mi - se - re -

3

£6=^

r

RELIGIOUS MUSIC

189

ppp

re, dig

na - re ia

- to

1 ^ J J1— >,«! H di - e sine pec -

I

=

S- 'J 1

1

r

rallent. « jxrdfndo

But the one movement that really shows the composer's power and originality is the ' ludex crederis/ No. 6. This is a piece of large dimensions, grandiose in style, and in spirit akin to the 'Rex tremendae' and the 'Lacryraosa' of the Requiem. It resembles the latter in the cast of the themes and mode of treatment. A very powerful and original effect is attained by the restatement a semitone higher, step by step, of its striking and vigorous melody :

I9o THE ROMANTIC PERIOD

Alleyretto un poco maettoso. j . = 69.

Base.

lu - dex ere do ris es - se ren tu - - rua. In

t~~ I 1 * \

te, Do - mi - ne, ape - ra - vi, non con - fun - dar in ae

^=,r i bf g ii?r-Jg-jQ» g r- ipj in

- ter . num, non con fun - dar in ae - ter - num.

The final movement, a ( Marche pour la presentation des drapeaux,' in B flat, Allegro non troppo, leaves the impression of empty rhythmical noise rather pretentiously put forth.

It has been pointed out that Liszt's orchestral Programme- music derives in a large measure from Berlioz' Symphonic fantastique, Harold en Italic, and Romeo et Juliette. In like manner it may be asserted that much of Liszt's miscellaneous church music and a large portion of the oratorios St. Elisabeth and Christus emanate from Berlioz' Requiem and L'Enfance du Christ. Wagner's method of employing representative themes is adopted by Liszt in the Graner Festmesse (written for the consecration of the Basilica at Gran, 1855-6), as well as in Die Legende von der heiligen Elisabeth (1861-5) and Christus (1866-73).

In St. Elisabeth, which is published as a concert-oratorio, Liszt virtually produced something like an opera sacra in which guise the work has been performed at Munich, Weimar, Hanover, Leipzig, &c. According to the authorized biography1 the success of St. Elisabeth, when performed with scenic accessories, came as a surprise to the composer. But, however that may be, the nature and structure of the work seem to call for theatrical pomp and circumstance. The conception of St. Elisabeth is rooted in the enthusiastic admiration felt by both Wagner and

1 Lina Ramann, Franz Liszt als KunsUer und Mensch, iii. p. 444.

RELIGIOUS MUSIC

191

Liszt for certain of Calderon's Autos sacrament ales l. A libretto was wanted which should give a composer of sacred music an opportunity to illustrate the legend of a saint, as Wagner had illustrated the stories of Tannhduser, Lohengrin, and Siegfried. Taking his cue from the order of Moritz v. Schwind's frescoes, which illustrate the history of St. Elisabeth of Hungary in the restored hall of the Wartburg at Eisenach, Liszt planned six scenes for which Otto Roquette 2 furnished the verses. The scenes are : i. The arrival of the child from Hungary a bright sunny picture ; 2. The rose miracle a forest and garden scene ; 3. The crusaders a picture of mediaeval pageantry; 4. Elisabeth's expulsion from the Wartburg a stormy nocturne ; 5. Elisabeth's death ; 6. Solemn burial and canonization. Five sections belong to the dramatic presentation of the story. The sixth and last, the burial and canonization, is an instrumental and vocal epilogue balancing the long instrumental movement which acts as a prologue. The ( Leitmotive,' five in number, consist of tunes of a popular type : I. ( In festo Sanctae Elisabeth 8 ' :

Andante moderato

2. A Hungarian folk-tune

Andante moderate

3. A Pilgrims' Song

1 See the Wagner-Liszt correspondence, anno 1858.

* Author of the well-known poem Waldmeisters Brautfahrt.

3 Compare Brahms' Qeisttiches Wiegenlied, Op. 91, No. a

192 THE ROMANTIC PERIOD

4. A hymn-tune assigned in the oratorio to Saint Elisabeth

-^-r-i , :|| v

/!\

J J J j-Hhr*-

/tN « [— fl

bgL-jLj

5. The intonation of the Magnificat frequently employed by Liszt as symbolical of the Cross 1 :

Mag - ni - fi - cat. Crux fi - de - lis.

Of these tunes the first proved the most pliable and the most fruitful of results. Its gentle contours supply the main lines and suggest the development of the two best sections of the oratorio the instrumental Introduction and the scene of the Rose-miracle. The latter is a little masterpiece : the most touching and most artistic scene in the whole work, and altogether one of the best things Liszt ever produced. The story is one of charming naivete, and seems to call for music : Landgrave Ludwig, Elisabeth's husband, is surprised to meet her in the wild wood alone, carrying a covered basket. In reply to his questions, she timidly pretends to have wandered while gathering roses. Questioned further, she confesses to be on her way to the sick poor with bread and wine. At the Landgrave's command, she uncovers the basket, and it is found to be full of roses. c A wonder, a marvellous wonder hath the Lord done unto us.'

1 For example, in the Dante Symphony, Granermesse : nachtlicher Zug No. I of the illustrations to Lenau's Faust, and in the Poeme symphonique Die Hunnen- scUacht.

RELIGIOUS MUSIC

193

un poco rait.

Liindgrave,

Sf*&^

1 IT : p-Qfr i*-; r r^ : r ^ _ /H

^fr*^™

Ge .

lieb - te, kannst du mir ver - zeih . n?

/TV

y

Elisabeth.

m It

Er-Bchiit-tert steh' ich

und er -

ho . -

•~N.

( %*fy r, . h M j [

F^Fh i J Ji i

1 [>-».

X

. ben.

jM/l n

/TV

^.-1 <~ z^=:

P

r

-

1

J:

=

-

^E

•-

Ul ' '

ttprtx*.

^=p=

r | I-

DANNRKUTHKR

194

THE ROMANTIC PERIOD

Andante relifiio

p

Ihm, der uns die - sen Se - gen gab, ihm . . lasst uns dan . ken !

Landgrave. ~a=~ .r-^ CZ,

f.-h

Ihm, der uns die - sen Se - - - gen gab, ihm lasst uns dan - ken !

##=

1 " fTTT r 1

[ _ j""^

4^

g : I g

^L . J-

i

I- I- -— K-"-- W=3P^

Er sei uns Leuch - te,

^

Er sei uns Leuch - te, er sei uns

~rT^rr~ ~r~- ~r

->=- \"j ^

=P=E

Er sei uns Stab,

'

er sei uns Leuch - te,

"V s~

-(22- -m- -(&- -9- -t&- -(2-

r-

h,

Stab,

er sei uns Leuch - te, er sei uns

rf^Tr

RELIGIOUS MUSIC

195

E£j&-^ r^=F

er Mi una

Stab, wenn

wir im . . .

8tab,

wenn

wir im

[--jrA 1 A A -JH

Pj£-* I* ^-2 g, 1

^=s

F r-*

--e=f£

Dun -

keln wan ken,

Dun

keln wan » ken,

f

T P"

f-

r r

*t±i=

wenn wir im Dun ... keln wan - ken.

wenn wir im Dun ... keln wan - ken.

t- h

r

O 2

196 THE ROMANTIC PERIOD

The chorus enters and continues in the same strain with subtle changes leading to a climax of praise; then the two solo voices, alternating with the chorus, resume their song, and the scene ends as quietly as it began.

Everything following this scene is comparatively weak. The chorus and march of the Crusaders are poor and trivial ; the scene of the expulsion culminates in a theatrical thunderstorm : that of Elisabeth's death recalls the last act of Tannhduser; indeed, a steady diminution of power and effect is manifest from the close of the miracle scene to the close of the work. There is as complete a miscalculation in this case of St. Elisabeth as in that of Schumann's Paradise and the Peri the subject of which offered material enough to make one fine cantata, whereas Schumann wrote three weak ones and Liszt chose to write a series of six.

Next to St. Elisabeth, the setting of the thirteenth Psalm for tenor solo, chorus and orchestra (1855-65)— occupies a prominent place among Liszt's contributions to sacred music. In Luther's version of the Psalmist's words the force of the personal pronoun comes out strongly: fWie lange soil ich sorgen ? Ich aber hoffe Ich will dem Herrn singen.' This is Liszt's cue. He exhibits the Psalmist's passionate appeals, his trust and hope, and his final conviction that he has been heard and will find help. In his own words, fthe tenor part (that of the Psalmist) is very important.' fl have permitted myself to sing personally, and I have tried to convert the ways of my own flesh and blood to the ways of King David 1.'

The following recitative-like setting of the words ' How long wilt Thou forget me, O Lord ? for ever ? how long wilt Thou hide Thy face from me ? ' shows the character of the main themes :

1 Liszt, Letter to Brendel.

RELIGIOUS MUSIC

197

A adante maeitoio

Tenor

-= & iP^

Solo.

Heir

0 """"

=^

i

vie

]

an - ge

^ a;

willst da

*s ^s

-

^^-H

B-^ £j

5S

•».

£

'

;

- -^

.^ 1

i , :

II

u'i . ner

«-*

80 g

IT rer

, gea

sen?

a

3=

^=

i

& ,

T6S dim.

r r

hO * re inidi, Hi IT, niein Gott !

198

THE ROMANTIC PERIOD

The chorus repeats these supplications and the orchestra strengthens and enforces them. The first appeal is followed by a f ugato l : ' How long shall I take council in my soul, having sorrow in my heart daily ? how long shall mine enemy be exalted over me ? '

Andante eon moto

mf esprets.

crescendo

ten. ten.

. ten. . ten.

Then comes another supplicating strain : ' Consider and hear me, O Lord my God ; lighten mine eyes, lest I sleep the sleep of death'; after which the supplications are again resumed, and again the pain seems assuaged: fBut I have trusted in Thy mercy ; my heart shall rejoice in Thy salvation.5

1 Compare the contrapuntal fragments quoted above, pp. 120 and 133, from Berlioz' Symphonic fantastiqw andf rom Romeo et Juliette. Liszt's essays in counter- point are, perhaps, more successful than those of Berlioz, though his fugue subjects are equally artificial and he fails to make the most of them. Both masters seem to have concocted rather than composed their fugues.

RELIGIOUS MUSIC 199

Etpreuivo

Ich a-ber hof - te dar - auf , daas du so gntt - dig, so gnfc - dig bist

The close consists of an Allegro impetuoso, ( I will sing unto the Lord V

Allegro impetunto

L«_S

Herrn sin - - - gen, will dem

E=3^

Among Lis/t's many contributions to the repertoire of Catholic church music2 the Missa solemnis, known as the f Graner Festmesse/ is the most conspicuous. Written to order in 1855, performed at the Consecration of the Basilica at Gran, in Hungary, in 1856, it was Liszt's first serious effort in the way of church music proper, and shows him at his best in so far as personal energy and high aim are concerned. 'More prayed than composed/ he said, in 1856, when he wanted to smooth the way for it in Wagner's estimation ( more criticised

1 In connexion with the setting of the thirteenth Psalm it is curious to note the contrast between Liszt's ultra-romantic pose of passion, and Brahms' studied reticence and purity of diction when dealing with the same words. Liszt's setting was published in 1865 ; Brahms' Op. 27, ' Der 13. Psalm for dreistimmigen Frauen- Chor mit Begleitung der Orgel,1 appeared at about the same time. Party strife was then at its height ; and the two versions may be taken to represent as nearly as possible the conflicting ideals of style, under the influence of which German musicians ranged themselves in hostile camps ' Zukunftsinusik ' on the one side, uncompromising classicism on the other.

J I. Missa quatuor vocum ad aequales (two tenors and two basses), concinentc organo. II. ' Graner Festmesse.' III. Missa choralis quatuor vocum, concinente organo. IV. ' Ungarische Kronungsmesse.' V. Requiem, for male voices and organ. Psalms 13, 137, 23, 18, 116, 129. VI. Twelve Kirchenchorgesange.

300 THE ROMANTIC PERIOD

than heard/ when it failed to please at the Church of St. Eustache, in Paris, in 1866. It certainly is an interesting and, in many ways, a remarkable work.

Liszt's instincts led him to perceive that the Catholic service, which makes a strong appeal to the senses, as well as to the emotions, was eminently suited to musical illustration. He thought his chance lay in the fact that the function assigned to music in the ceremonial is mainly decorative, and that it would be possible to develop still further its emotional side. The Church employs music to enforce and embellish the Word. But the expansion of music is always controlled and in some sense limited by the Word for the prescribed words are not subject to change. Liszt, however, came to interpret the Catholic ritual in a histrionic spirit, and tried to make his music reproduce the words not only as ancilla theologica et ecclesiastica, but also as ancilla dramaturgica. The influence of Wagner's operatic method, as it appears in Tannhduser, Lohengrin, and Das Rheingold *, is abundantly evident ; but the result of this influence is more curious than convincing. By the application of Wagner's system of Leitmotive to the text of the Mass, Liszt succeeded in establishing some similarity between different movements, and so approached uniformity of diction. It will be seen, for example, that his way of identifying the motive of the Gloria with that of the Resurrexit and that of the Hosanna, or the motive of the Sanctus and the Christie eleison with that of the Benedictus, and also his way of repeating the principal preceding motives in the ( Dona nobis pacem,' especially the restatement, at its close, of the powerful motive of the Credo, has given to the work a musical unity which is not always in very clear accordance with the text.

Liszt's illustrative, decorative, and dramatically expressive style is seen in the Kyrie and Christe, as well as in the Gloria, of which movements the following quotations will give some notion :—

1 The score of Das Rheingold was finished in 1854.

A.

Soprano*. Altos.

Tenors.

RELIGIOUS MUSIC

Andante toltnne

201

SBE

,- es

4-—^

^- *

r r

^

Ky

£*..

> i>=

/3 ere«.

202

THE ROMANTIC PERIOD

m

J-r£

^

poco riten.

i > son

i

Solo.

i - son

if f HT||

e - le - i -

rinforzando

Tutti. /,

JU

e - le . i -

Solo, i

Ky

1

Solo, espress. mf poco riien.

e - le . i - son

* t=-

e - le - i - son

Efc

Tempo

-$£-

RELIGIOUS MUSIC

203

B.

aprtu.

le . i - son

fT-fr—^: ^

s

Chri - «te, Chri dolce J I i

ste,

e > lei son,

Chri

ste

c.

non troppo Strings. '

£ *

-is*-.— r^'pr "•"."_' p* . " T p^T^ *~*?

gz3f^? ~~l? ?

Flutea.

s

->•• ••..-•

Oboes.

i i i

s £

> fe i

ri a in ex eel

sis De o,

D.

Flute.

JTT:! fTfttfixu2&

Soli, p et

in ter

ra pax ho

T

ni . bus

bo

204

THE ROMANTIC PERIOD

f ^

. nae vo - lun

^, -p- - a -*

- ta

1 tis

^I^^J_±I-— - 1 _^-

^==-

The sentences of the Credo, so difficult to weld into a con- sistent musical whole, are treated separately, and a semblance of unity is attained by means of repetition and expansion of the principal subject.

Altogether it may be said that Liszt's treatment of the text of the Mass his treatment of the Gloria, for instance, as a chorus of angels with the accompaniment of whirring violins is here and there strikingly picturesque and effective. Com- pare the example C, above. But it must be added that with the exception of the Kyrie and its little offshoot, the Benedictus, no complete movement is sufficiently well knit together to bear severe scrutiny ; that the music is made up of scraps of melody, of fragmentary counterpoint, and sudden changes of key ; and that the prevailing restlessness and the theatrical character of some of its instrumental effects are not in just accord with the spirit of a religious ceremonial.

In the Hungarian Coronation Mass (' Ungarische Kronungs- messe/ 1866-7) Liszt aimed at characteristic national colour, and tried to attain it by persistently putting forward some of the melodic formulae common to music of the Hungarian type, such as :

A.

' 1ft

B.

m

which occurs in the national f Rakoczy March ' and in number- less popular tunes or an emphatic melisma, such as :

RELIGIOUS MUSIC

205

C.

D.

known to everybody through the famous Rhapsodies. From beginning to end the popular Hungarian element is represented by devices of this kind in a manner which is always ingenious and well suited to the requirements of a national audience. The following bars from the Benedictus and the Offertorium will serve for examples :

Adagio molto

\-

•9?

*

/v * p—ir i r P '

t/ *•

*f»)ij)re dolcittimo

Vffe g

j^I 5 r

^ i

ao6

THE ROMANTIC PERIOD

pfc-T-T-ff-

dim. pentoido

EEE3

.fEf-

£FTT- efrrriT-

rfim. perdendo PPP

fe

-r I "

RELIGIOUS MUSIC

207

i

.. e £ g iei££ £»£;££ g °e

••g " i r

i

/,to ,:

i

Ac.

-

B.

Lento

<7\ tostenuto ed trpreutirn

ao8

THE ROMANTIC PERIOD

r

v*

+=£ i ,J~jS

j. i

J t— s -T-

TT

*u

I1"*-

i t-

*> <

fed^Szft:

"1 f

1

lA

?^^i^y

-4 j -i

^

« i $

:^=s=i

^

rilenulo

^I\ «V

•i K •• r-4-

misterioio

RELIGIOUS MUSIC

309

piu ere*.

to,

But the style of the entire Mass is as incongruous as a gipsy musician in a church vestment doubly strange to students of the present day, who in Liszt's Rhapsodies and Brahms' Ungarische Tdnze have become familiar with the rhythmical and melodic phrases of the Hungarian gipsy idiom, and who all along have known them in their most mundane aspect. Apart, however, from its incongruities of style, the Offertorium is a shapely composition with a distinct stamp of its own 1.

Liszt's manner of writing for solo and choral voices is generally practical and effective. The voice-parts are carefully written so as to lessen the difficulties of intonation which the many far-fetched modulations involve, and are skilfully disposed in point of sonority. The orchestration, always efficient, is fre- quently rich and beautiful.

The oratorio, Christus(i 863-73), thelargestand most sustained of Liszt's efforts, and the magnum opus of his later years, is remarkable for the fact that its conception is essentially Roman Catholic, devotional, and contemplative in a Roman Catholic

1 Allowing for *ce double caractfere national et religieux,' Liszt asserts that ' dans ces ^troites limites, La Hesse du Couronnement est plus concentrique qu'e- courte'e, et que les deux tons principaux du sentiment national hongrois et de la foi catholique s'y maintiennent et s'harmouient d'un bout a 1'autre.' Letters to Madame de Wittgenstein, iii. 181.

DANNKEUTIIEK P

aio THE ROMANTIC PERIOD

sense both in style and intended effect. It contains nothing that is not in some way connected with the Catholic ritual or the Catholic spirit; and, more than any other work of its composer, recognizes and obeys the restrictions imposed by the surroundings of the Church service. The Latin book of words (Liszt's own selection) consists of Biblical and liturgical texts, with the Beatitudes (Matt. v. 3-10) for the centre. The person of Christ is treated with great reticence, though His words are used in the Beatitudes, in the illustration of the storm, ' Quid timidi estis, modicae fidei ? ' and in the scene of the Passion, 'Tristis est anima mea usque ad mortem.' There are three main sections: I. Christmas oratorio; z. After Epiphany; 3. Passion and Resurrection. These sections or parts are subdivided thus : Part I. Christmas Oratorio I. Instrumental introduction with the prophet Isaiah's words for a motto : 'Rorate coeli desuper et nubes pluant iustum, aperiatur terra et germinet Salvatorem.' a. Instrumental Pastorale, and the Annunciation : c Angelus Domini ad pastores ait V 3. { Stabat Mater speciosa.' 4. Instrumental Pastorale, the music of the shepherds. 5. Instrumental March, the Three Kings, ( Et ecce stella quam viderant in Oriente antecedebat eos.' Part II. After Epiphany: T. The Beatitudes. 2. ( Pater noster.' 3. The foundation of the Church : e Tu es Petrus et super hanc petram aedificabo Ecclesiam meam.' 4. The storm-miracle. 5. The entry into Jerusalem : ' Hosanna, Benedictus qui venit in no- mine Domini, Rex Israel.' Part III. Passion and Resurrection : i. ' Tristis est anima mea.' 2. ' Stabat mater dolorosa.' 3. Easter hymn, ' O filii et filiae.' 4. Resurrexit.

A mere cento, it may be said. But it would be difficult to make a better selection for the special purposes which Liszt had in view.

Of the Leitmotive, several of which consist of liturgical

From c Angelus Domini ' to the words ' Coelestes exercitus ' Liszt quotes the Gregorian chant note for note (soprano solo) ; even the chorus ' Laudantium Deuia et dicentium,' which follows, is almost entirely taken from the same source.

RELIGIOUS MUSIC

211

phrases and are therefore meant to be taken for symbolical as well as representative themes, two may be quoted1: I. The intonation of the ' Rorate coeli ' (Introitus ; Advent IV)

Ko ra - - - te coe

which Liszt has as follows :

P = =

de - eu

per

=3?

Bo - ra - » - te coe . . . li de - su . - -

2. The intonation of the ' Pater noster ' :

per

Pa u-r no - ster, qui ea in coe 111,

which Liszt has thus :

1ft

1

4^ r r

Pa - ter no - ster,

qui ea in

The themes of the introductory fugue, the first Pastorale (a, below), and the March of the Three Kings (by below), are evolved from No. i, above :

Allegretto moderate, pattora.lt,

, Allegro non troppo.

. ten.

1 F

The phrase from the first bar of No. I appears in the introduction to the Beatitudes, Part II, in the Easter hymn, Part III, and

1 Compare Biiumker, ' Das katholische deutsche Kirchenlied in seinen Sing- woisen,' where ' Gtesangbucher ' of Mayence, 1661 and 1665, are quoted, i. p. 476.

P 2

212

THE ROMANTIC PERIOD

elsewhere. Both the Pastorale movements and the March, though they are somewhat long-drawn-out, considering the quality of the musical texture, are full of simple, popular melody and effective instrumentation. Instances may be found in the pifferare tunes of the second Pastorale, the tender melody in F major from the same number, the picturesque effect of the high notes of the violins and flutes in the trio of the March, meant to depict the star of Bethlehem, or the rich sound of the following passage, also from the March, illustrating the adoration of the Magi, c Apertis thesauris suis, obtulerunt Magi Domino aurum, thus et myrrham ' :

Adagio sostenuto ed esprettivo a*sai Strings. | | .

m

Cl. and Bassoons.

-

^ 9

^ j

^E£

^

r

RELIGIOUS MUSIC

213

The ' March of the Three Kings,' it is said, was inspired by a picture in Cologne Cathedral. * Stabat mater speciosa,' No. 3, Part I, in spite of its cloying sweetness proves singularly impressive in performance. On the whole, the entire Christmas section the first part of the oratorio is distinctly good, and in its naive way deserves to rank with the charming Children's chorus of welcome and the fine scene of the Rose-miracle in St. Elisabeth. With one exception, namely, the beautiful illustrations to the Sermon on the Mount (the Beatitudes), it cannot be said that the second and third parts take rank with the first. There is an increasing preponderance of effort and dubious experiment as the work proceeds.

The exception, however, is very noticeable. The Beatitudes contain more refined music, convincing in itself by reason of its beauty, than all the later movements put together. The antiphonal disposition of this piece is simple and very effective. The protagonist, baritone solo, begins : ' Beati pauperes spiritu ' (E major), and the chorus repeats and responds :

Baritone solo. Lento dolce ,

Be . a

pau - pe res spi - ri - tu, poco rit. ^.

quo . m -

Oi£

-jf.

- am . . ip - so - rum eat reg - num ooe » lo

Sopranos and Altos.

P

Be - a Tenors and Bass**.

pau pe - res spi

ri-

-P—

i -

214

THE ROMANTIC PERIOD

quo - ui - am . . ip so - rum oat reg - num coe -

I* ft J

L_ I"

Advantage is taken of the opportunity for expressive modulation offered by the third, fourth, and fifth versicles :

(a)

ttprct*.

"-CTT

Be - a . ti qni lu - gent, quo - ni am ip - si dolee

con - so la - bun tur.

Eifefe

Be - R ti qni e su - ri - xint et si - ti - unt ins . t i

ti - am.

(r)

Be - a

ti mi - ee

li . cor

dee.

RELIGIOUS MUSIC

215

and of the chance to introduce an effect of energy at the eighth response : ( Beati, qui persecutionem patiuntur propter iustitiam.'

In tempo ed tnergico

Solo.

Chorus. J

Organ. •(

??=£

"1= -F-:—3=*

Be - a . ti, Be a

/ :=-

Be - a . ti,

mf

r ..-fr-

- ti,

Be - a

n'n/orj. , A ?»o»o B^. *• .

Be - a ti,

qoi per-ae - cu . ti o-nem pa - ti > UD - tur

ti,

S^

It 1 1

Be- a - ti, ^ J J

B«- -=• C*-

> hM

^

ai6 THE ROMANTIC PERIOD

f Beati, Beati, Amen ' forms the close, pianissimo. Then comes the ' Pater noster,' a choral movement in a similar mood, equally well written for the voices, though far less original. The section entitled f The foundation of the Church ' (' Tu es Petrus) 3 consists mainly of a choral song, ( Simon loannes, diliges me ? ' written in 1863, under the title of cLJHymne du Pape,5 and published with French and Italian words, in 1865, at a time when there was talk of Liszt being appointed Maestro della Cappella Pontifica. The storm-miracle in Christus, like the wild storm-scene in St, Elisabeth, is a piece of decorative affresco work partly instrumental with descriptive indications such as f Ipse vero dormiebat/ printed in the score, partly choral with short exclamations chanted by a chorus of male voices : ( Domine, salva nos, perimus ' ; it culminates in a phrase sung by a baritone voice which represents the voice of Christ : ' Quid timidi estis, modicae fidei ? ' This curious specimen of hybrid composition comes as near to failure as any similar deviation in Berlioz5 work.

In the scene of the entry into Jerusalem the music, admirably adapted to the pageant, culminates in an animated fugato, 'Hosanna Filio David/ with a coda, in graver mood, to the words ' Benedictum quod venit regnum patris nostri David : hosanna in altissimis.'

The f Tristis est anima mea usque ad mortem } of Part III illustrates the sufferings of Christ as realistically as it is possible to present them in terms of music. The persistent chromatics affect the ear as an excess of sombre chromes and browns may affect the eye in a picture. In the somewhat prolix ' Stabat mater dolorosa,' the process of emphasizing the expression of suffering is continued with similar insistence, by an ingenious and at times really fine accumulation of tones. The poignant expression is carried to an extreme of pathos, strikingly dramatic in nature and effect : such, for instance, as the following passage from the ' Inflammatus ' :

RELIGIOUS MUSIC

217

con lomma pauione

Soprano

r r

per te, vir

go,

aim fl A

de

fen A t

;M^-£

^^=fe

^

iF=*F= *m

Chorus. -AT""""^ A A -g-' ^^^

^:f-F i f ^=£q^=:^

In flam - ma - tus et ac - cen - BUS

But the main impression left by this part of the oratorio, i. e. the Tristis and the Stabat, is as wearisome and oppressive as that of a collection of realistic pictures of martyrdom l.

1 We have got accustomed to all manner of chromatics and the most poignant dissonances in Tristan, Gotterdammerung, and Parsifal ; but we have also learnt to appreciate the closeness of Wagner's musical diction, which responds at once and precisely to the stimulus of emotion. Moreover, the extremes of emotion, as

2l8

THE ROMANTIC PERIOD

A nai've little hymn, a favourite old tune tastefully set for female voices with harmonium (or flutes, oboes, clarinets and corno inglese), is introduced by way of relief and contrast after the Stabat. It is intended to be sung and played by invisible choristers and instrumentalists :

tin poco aniinato

p Al - le - lu

Al - le lu - in, Al - le - lu

mor - te sur - re

The oratorio closes with an ecstatic, and indeed somewhat militant, 'Resurrexit/ in which the voices are almost overpowered by the tumult of brass instruments. The following excerpt may serve to illustrate the manner in which it is conceived :

Allegro mosso (Alia breve)

2-2E

pT r , i

r !-i

Chris - tus

via - cit,

Chris . tus

leg - nat,

=F=S

Chris -

tus iin

pe - rat

in

seni

1 - pi

Chris -

tus via

cit, Chris

. tus

reg

nat,

1 P

A ± r J-

'i r

na sae

Wagner sometimes deals with them, are explained by the dramatic action. Thus the utmost violence in musical combinations may be intelligible and justifiable from the dramatist's point of view, even when, taken as mere music, combinations of the kind seem to pass the limits of intelligibility, and appear simply ugly and repelling. Compare Wagner's own view as expressed in Ueler die Anwendung der Musik aufdas Drama, Schriften x. 231-50.

RELIGIOUS MUfc

Chris - tns im - pn - rat

>IC 219

in gem pi

A J ^ i r

J A.' .J. j =g_l » -^=^

L * E

^=^— 1 J

: _L J

pi - ter

Was Liszt's bias essentially histrionic, as the use of repre- sentative themes and operatic effects of illustration in his Pro- gramme-music seems to suggest ; or was it ecclesiastical, as the use of Intonations serving as symbolical Leitmotive in his Masses, Psalms, oratorios, and even in a number of instrumental pieces, would imply ? Perhaps both. All his life long Liszt was a faithful, if somewhat wayward, son of the Catholic Church ; and from the first, some of his music echoes the tone of the cloister, and bears traces of the faicts et gestes of the priest- hood. It is difficult to deal justly with a certain class of Liszt's productions pieces imbued with religious sentiment and pro- duced at all stages of his career from the morbid ' Pensee des morts ' of 1 834, the rather austere ' Pater Noster ' (in C) or the devout ( Ave Maria' (in B b) of 1847 13 and the painfully chro- matic 'Stabat mater dolorosa' (Fminor) of 1886 (Christus), to the various azure and ultramarine Pieties, such as the * Ave Maria ' (in E) of 1863, the 'Pater Noster' (in A t?) of 1866 (Christus), or the Angelas ' Priere aux anges gardiens ' (in E) of 1 883 2. Several of these effusions are very ambitious : the scene of the canoniza- tion of St. Elisabeth, for instance, or the distressing ( Stabat mater dolorosa ' just mentioned, or the beautiful Beatitudes ; others are so slight as to be hardly appreciable, like the ' Stabat mater speciosa,' and the naive Easter hymn, f O filii et filiae' (Christus}, the Sposalizio of the Annees de pelerinage, or the delicate little

1 Harmonies poetiques et religieuses, Nos. 2, 4, 5.

2 Annees de pelerinage, iii. No. i.

220 THE ROMANTIC PERIOD

pianoforte pieces called ' Consolations.' Among such widely divergent compositions, religious in tone and character, those of later days than the simple ' Pater Noster 3 of 1 847 are by no means extravagant or problematic, or in any sense outside the pale of music proper ; yet the majority of them can hardly be accounted good music in the full sense of the word. Taken simply as music, and without regard to any symbolism or casual associa- tion with the ritual, they convey an undefinable sense of effort and weakness. Nevertheless, some of these very pieces, notably the f Stabat mater speciosa ' and the Beatitudes, have been known to make a deep impression on persons not particularly suscep- tible to influences of an ecclesiastical description or subject to sudden devotional impulses.

Looked at from a musician's point of view apart from the glamour of an ancient ceremonial, apart also from the fascination of Liszt's unique personality a large proportion of these compositions appear wanting in that specific musical character and in those distinctive features which make for consistency and coherence of musical interest. Pieces, to a certain degree well put together, are found to contain bare and arid stretches, full of intention perhaps, and full of feeling, but full, also, of wearisome and pointless particulars. The means of effect employed by Liszt are neither commonplace, nor especially eccentric, extravagant, or in any technical sense deficient. The devotional feeling that prompts their use is evidently sincere, amounting now and again to true fervour and passion ; yet, in the end, the entire endeavour fails to convince the mind's ear, and leads to little that is complete or even likely to prove of enduring value as artistic work. If a man chooses to employ the pianoforte or the chorus and orchestra for devotional purposes he is bound to be watchful of his mode of musical expression ; mere emotional improvisation will not suffice ; for his experienced hearers are always inclined to resent any shortcomings in the musical substance or workmanship, and to assert, with increasing emphasis, that the cause of piety

RELIGIOUS MUSIC 221

is but ill served by deficiency in the essential elements of composition J.

The pieces sacrees of Rossini and Verdi may fitly be men- tioned here. Rossini's consist of the well-known Stabat mater, begun in 1831, finished and performed 1841-2; three choruses for female voices: La Foi, L'Esperance, and La Charite (1844), of which the first two belong to a forgotten opera, Oedipus ; a ' Tantum ergo 3 for two tenors and bass with orchestra, a ' Quo- niam' for bass solo and orchestra, and SO Salutaris' for four solo voices; and the so-called 'Petite Messe Solennelle' (1864), the scoring of which was completed in 1867, shortly before his death. Writing about the first Parisian performance of Rossini's Stabat Mater, in 1842, Heine2 slyly stated that the impression he had received reminded him of a curious representation of the Passion by little children, which he professed to have witnessed at Cette. ' The ineffable martyrdom was presented and repro- duced, but in the most naively juvenile way the terrible plaint of the Mater dolorosa was intoned by little maiden voices,' &c. It is sufficient to state broadly that Rossini's Stabat is fine music from the professional singer's point of view, and not always devoid of devotional sentiment. The Introduction and first Chorus, the duet, ' Quis est homo/ the ' Inflammatus,' and the unaccompanied quartet, s Quando corpus morietur,' rank with the most effective of Rossini's work.

Verdi's contributions to sacred music are a Requiem Mass for solo voices, chorus, and orchestra, in memory of Alessandro Manzoni ; a Pater Noster for two soprani, contralto, tenor, and bass ; an Ave Maria for soprano and strings, and ' Quattro pezzi sacri.5 In 1 875, his sixty-second year, Verdi, amid rare enthusiasm, made the tour of the principal European concert- rooms with the Requiem. It is in the nature of things that the religious music of a man who from youth to old age devoted

1 Witness in recent years the failure of Gounod's oratorios The Redemption and Nors tt Vita, Tinel's Frandscus, Perosi's Transfguration, &c. 3 Werke, x. 331.

222 THE ROMANTIC PERIOD

himself exclusively to the composition of operas should have something dramatic about it. The marvel, under the circum- stances, is rather that this religious music of his (it is not precisely church music) should have so little of the histrionic or the theatrical about it. The expression of sorrow, terror, despair, supplication, and hopeful expectancy in the Requiem is perhaps too personal and passionate, but it is sincere. There is no trace of frivolity, and, what to a musician is more important, it will bear inspection from a strictly musical point of view. Much of it is admirable in its way, and convincing ; some portions, where the contrapuntal treatment of choral parts comes in sporadically, are now and then weak, though by no means inept. The best numbers owe their success to the composer's native gift for vocal melody, to his able treat- ment of the voice-parts in ensemble pieces, also in some measure to certain realistic effects produced by choral noise and orchestral blare, as in the ' Tuba mirum,' which shows that Verdi had studied Berlioz' Requiem to some purpose. The elegiac reposeful numbers have the charm of sincerity and tender feeling ; such is particularly the case with the ' Requiem aeternam,' in which a subdued expression of sorrow alternates with a tender ray of hope, and the Kyrie and Christe. The mezzo soprano solo, ' Liber scriptus,' and the bass solo, ' Confutatis,' which form part of the 'Dies irae,' are remarkably effective. The extremely simple 'Agnus Dei' is as original as it is masterly, a melody of thirteen bars six times repeated with ingenious changes in the voicing, and a few bars of coda. The closing ' Requiem aeternam ' is as touching as the opening of which it is an expansion. Of the two fugues, £ Sanctus' and ( Libera me/ the second is the best as regards the invention of the subject and also in point of workmanship ; its climax acts as a telling foil to the pianissimo of the final ' Libera me.'

In the course of 1 898, his eighty-fifth year, Verdi published the latest of his religious pieces : Quattro pezzi sacri, Ave Maria (scala enigmatica armonizzata a quattro voci), Stabat mater

RELIGIOUS MUSIC 223

(per coro a quattro parti ed orchestra), Laudi alia Vergine Maria (per quattro voci bianche), and Te Deum (per doppio coro a quattro parti ed orchestra). The first of these, the Are Maria, a harmonic puzzle, looks as if it were a jeu d" 'esprit, meant to be a skit upon certain curious experiments of Liszt's in chromatics. It is based upon an ( enigmatic 3 scale ascending C, D b, E, Ffc G ff, Afc B, C ; descending C, B, A J, Gfl, F fl, E, D[7, C. This 'scala' in minims is employed as a cantus firmus, first in the bass, then successively in the tenor and alto, finally in the soprano. The queer counterpoint which Verdi applies to it is far-fetched and difficult of intonation ; the total effect is almost, if not quite, as musical as it is curious. From a vocalist's point of view, the ( Laudi per voci bianche,' female voices, is an exquisite piece sweet and tender, showing Verdi as a singer of genius. It produces a wonderful effect of purity and happiness. The words are from the last canto of Dante's Paradiso :

Vergine Madre, figlia del tuo Figlio, Thou Virgin Mother, daughter of thy Umile e alta piu che creatura, H-ji ftnd

Termine fisso d'eterno consiglio, creature,

The limit fixed of the eternal counsel, Thou art the one who such nobility Nobilitasti si, che il suo Fattorc Tohuman naturegave, that its Creator

Did not disdain to make Himself its Non disdegnfc di farsi sua fattura. creature.

Nel ventre tuo si raccese 1' amore, Within thy womb rekindled was the love,

By heat of which in the eternal peace Per lo cui caldo nell' eterna pace After guch wige this flower ^ ^

Cosl e germinato questo fiore. minated.

The Stabat and Te Deum, in the violent contrasts of colour employed, the noisy instrumentation, and the crude use of chromatics, show traces of the operatic stage of Berlioz' Requiem, and even of Liszt's Christus; but they also exhibit high imagination and some genuine musical quality.

Wagner's Das Liebesmahl der Apostel : eine biblische Scene, for male voices and orchestra, belongs to this part of the subject, though it can hardly be ranked under the head either

224 THE ROMANTIC PERIOD

of orthodox or of secular church music. It is religious in spirit and tinged with the partly religious, partly humanitarian and political, enthusiasm of the years which preceded the revolution of 1848. It was composed as apiece d' occasion, for a festive gathering of men's choral societies in Dresden. Like some scene from an opera seria of Spontini, it is planned to form one continuous movement. Remarkable rather for the spirit in which it is conceived than for its actual musical value, Das Liebesmahl recalls the style of Wagner's first successful opera, Rienzi (1840-2). As in Rienzi, the melodic and harmonic effects are simple, direct, and telling, though occasion- ally somewhat blatant and commonplace. Certain passages fSende uns deinen heiligen Geist,' f Machet euch auf, redet f reudig das Wort ' show traces of that mystical fervour which many years afterwards found fuller expression in the choruses of the Knights of the Grail in the first act of Parsifal, Three separate choirs a cappella (each consisting of tenors and basses divided), who represent the Disciples, are supplemented by twelve solo bass voices, who represent the Apostles, and by a choir of ' voices from on high,' who represent the Trinity ; ultimately the mass of voices is joined by a very full orchestra, which, as in Rienzi, includes certain extra wind and percussion instruments such as additional bassoons, a ' Serpent,' valve horns, valve trumpets, a bass-tuba and two kettledrums, besides the usual trumpets, horns, trombones, and timpani. The sound of * voices from on high ' as in Parsifal •' Seid getrost, Ich bin euch nah, und mem Geist ist mit euch' was heard in its fullest beauty when the words were sung from the cupola of the Frauenkirche, at Dresden, on the sixth of July, 1843; and the fine effect thus produced was made to serve again, thirty-nine years later, in the first and third acts of Parsifal (1882)— f Der Glaube lebt, die Taube schwebt.'

CHAPTER X

CONCERTOS AND CHAMBER MUSIC

and shorter pieces chiefly for pianoforte or violin with orchestra are conspicuous among the instrumental music of this period. They noav be^ 4'^fififl \yijgr, foy°- ^ea:48- v J&^figJ3 rem^arka.ble^focxu'iginitlityjof^oncgpldon and a high quality of material and ^workmanship such as WefrprN ^^"^rtitMrV fcy jpranoiorte, Spohr's eighth Concerto, fln modo d'una scena T caniante' (1816), MepH^^Qhai'8 Concerto for violin, and Schu-

mann's for pianoforte (both 1 84.^) ; secondly, pieces wherein stress is laid on a display oT the solo player's attainments as a virtuoso

_Q»oT) nfi PagtHinavioln oncertoin D (1820), Ernst's in F g

^^* '^WM^JWnHMMMM>MHn*MHPIiMMM^

minor (1836), Vieuxtemps' in E (1835) and D minor (1853), Moscheles' third pianoforte Concerto in G minor1 (1822), Henselt's in F minor, jistj^ Fi I? ""^ il his concert

variations on the Dies irae known as ' Todtentanz ' (Danse macabre), 1 850, and furthermore, three cha^^jfp«^'^aUy XTQ f innai concertos, distinrtly

% though comparatively recent in date, viz. Joachim's Hungarian Concerto in D minor for violin, Grieg's .Norwegian Concerto

•ffl A "minor for pianoforte, and Tchaikovsky^s Russian Concerto. also |pr jjianofortej in B b minor.

*^rebe?TCSllL'WlHlUl!k (iSiij),""1 Larghetto affettuoso, Allegro passionate, Marcia e Rondo giojoso,' was first designed on the scheme of a romantic story, which Weber finally did not publish, since he deemed the sequence of movements sufficiently intel-

1 The first movement of which contains music of a high order, undeservedly neglected, and represents the technique of Hummel, grafted on that of dementi and Beethoven.

DANNREUTHEU

236 THE ROMANTIC PERIOD

ligiblfc without verbal explanation. His own views on the subject of this piece and its relation to ' programme music ' are expressed with great frankness : f I am planning/ he wrote to Rochlitz, the critic, ' a pianoforte concerto in F minor. The choice of key seems curious seeing that concertos in a minor key so rarely please unless there be some rousing idea connected with them. A sort of story has somehow taken hold of me ; it will serve to link the movements together and determine their character in detail, as it were, dramatically. I intend to entitle them : Allegro, " Separation " ; Adagio, " Lamento " ; Finale, " Deepest grief, comfort, return and jubilation." I find it very difficult to accustom myself to this conception, as I particularly dislike all musical pictures with specific titles yet it irresistibly forces itself upon me and promises to prove efficacious. In any case I do not care to put it forward for the first time at any place where I am not already well known, for fear of being misunderstood and counted as a charlatan.'

As the Concertstiick now stands, the sequence of movements reflects the changes of mood in some operatic scena, thus : A lady sitting in her bower and thinking of her knight, \vho has gone to the Crusade (Larghetto affettuoso). She fancies him in battle (Allegro passionato) and longing for one more sight of her before death. She is near to fainting away, when suddenly from the woods without comes the sound of men approaching (Tempo di mar da). She looks out anxiously. There is her lover and with a wild cry she rushes into his arms (Presto giojoso)1. Technically, the Concertstiick marks a new departure in the treatment of t

C, amHs full of ingc'nious devices both in the solo part and the orchestration.

* ; __ , mm -iimn i

1 An ingenious commentator, C. F. Weitzmann (the first good historian of clavier music), inspired hy Liszt's grand rendering, regarded the piece from a political point of view, as an echo of the glorious uprising of Germany towards the close of the Napoleonic wars, and not without some reason, in so far as the spirit of the music is concerned.

CONCERTOS AND CHAMBER MUSIC 227

=-*"^E^s£=F=a

w^ e i- u •* i : 3 -*—sr<r-f-a i i-

V 6 ^ b B

Con duolo e ben marcato la me^odia

__ ft- t f T ,

lH aT" T* P B

1

H

-^J j

@a,»> L^*-

-W a =1 P 1

L^ ^

=H^

^

=*4=t-

^ i

=fe=^

F^F= ^ i

3z±rz£r^

The novel effect of a sustained melody with arpeggiando and quasi-pizzicato accompaniments, senza pedale (Introduction), has been reproduced by Mendelssohn in his Capriccio brillante, mentioned above ; and the staccato octaves, rippling semi- quavers, and certain details of instrumentation (Finale) reappear in Mendelssohn's Concerto in G minor. The plaintive wail produced by the high notes of the bassoon against the throbbing chords of the strings an operatic effect just before the Tempo di Marcia, is singularly telling in its place :

Bassoon solo. Adagio

f> dolce *2^ JL . . ^t

I^rinrn1~'~ ' 1

.

^ r" £,

A L Adagio

ten.

njf-^

Strings, pp

ten.

SSl jy^i-t 1

= J~* * * '

1

Q 2

228

THE ROMANTIC PERIOD

^g^ .

g" a^ere W^£

i \ : 1 n-rrr-

^±E

iT'^a <en'

"=— Iht5-

B===p '?• ^j.,,.

§£

(&rr

ten.

"^ n l!ij?^-i r *g'tt~

ten. ^

> : -A

- *jTT2 J 3QE

After {hfo the hearers are astonished by rapid octave jj]U over the white keys of 'the^TflgMb'fu'f Ui,' a virtuoso effect com-

paratively^ easy ^on the oTcT Viennese instruments, but nc impossible on.,niodoM» jgroiadfr^

ere is yet another even more surprising effect, and more distinctly Weber's own, but too long for quotation the sustained crescendo from pianissimo to fortissimo for sixteen bars before the first Allegro on the same harmony, where the sound of the instrument is allowed to accumulate with the dampers raised 2.

Before the Concertstiick Weber wrote two concertos, one in C and another in Eb. The second of these appeared in 1812, a year after Beethoven's great Concerto in E 7. Thus Weber's surprising choice of the key of B major for the Adagio 3 is accounted for ; it appears as a reflex of Beethoven ; there are also other points of resemblance slight but unmistakable.

Though not strong enough for performance in public at the present day, Weber's Concerto for bassoon in F, his concertino in E b, Op. 26, and the Concertos in F minor and E b for clarinet, and especially the interesting Concertino in E minor for horn (1806-18) occasional pieces rapidly written for virtuoso friends deserve notice as containing the best solo music extant for

1 It is apparently due to Beethoven, who made use of it in his Trio in C minor, Op. i, in the first Concerto in C, and in the Finale of the Waldstein Sonata.

a This is the celebrated crescendo that Goethe asked for when Weber visited him in 1812. Here again Beethoven was the initiator, and Weber, with his theatrical ' flair,' the first to make the most effective use of it.

3 Virtually C J? major the key-note E I? becomes D ;$ and forms the third of the new key.

CONCERTOS AND CHAMBER MUSIC 229

the particular instruments. The latter piece, for the first time in recorded music, exhibits a curious trick of obtaining the effect of three- or even four-part harmony from a solo horn. (The player ( blows ' a note and at the same time sings or rather s hums ' another if this is done perfectly in tune the acoustical result is of a chord in three or four parts.) Taken together Weber's show-pieces for wind instruments form a compendium of the good effects the instruments can produce individually, and here again, as with some of the solo pianoforte pieces to be mentioned later on, the details appear like studies for something to come. Compared with Spohr's early violin concertos, the freshness of Weber's pieces is remarkable, in spite of their rather flimsy texture. The best piece in which the clarinet takes part is the duo concertante Op. 48, in Eb, a spirited and showy sonata for that instrument and pianoforte.

Musicians and virtuosi of the present day are agreed that after Beethoven's concertos in C minor, G major, and E b, and before Brahms' in Bb major and Tchaikovsky's in Bb minor, Schumann's is the great pianoforte concerto; and similarly, that after Spohr's ( Scena cantante,' and Beethoven's concerto in D, but before Joachim's f Hungarian' in D minor and Brahms' in D major, Mendelssohn's is the great concerto for the violin. So grateful to the violinist and so much in vogue with the public is the latter, that it bids fair to outlast the interest in the rest of Mendelssohn's solo and ensemble pieces.

The device of joining the movements of a concerto so as to form a continuous whole has frequently been ascribed to Spohr, who, in his c Scena cantante' (Gesangsscene) just mentioned, imitates an operatic scena of tragic import. Moscheles in his ( Concerto fantastique ' worked on similar lines. Mendelssohn made use of it with very good effect in both of his pianoforte concertos and in the concerto for violin. But the credit of having originated the happy innovation belongs to Beethoven, whom Weber followed in the Concertstiick (compare the beautiful transition from the Adagio to the Finale in Beethoven's Eb

230 THE ROMANTIC PERIOD

concerto with Weber's bassoon passage quoted above). Mendels- Rf>nn^r»2^t^2il ftf . flO,ll)biining * J^^tB«vetaenU..derive& dicejclly |rom Weber's ConcertstiickA as Spohr's design derives from Mozart's "fantasias in D minor and C minor. Mendelssohn's concertos for pianoforte, the Capriccio brillante, the Serenade and Allegro giojoso, and the Rondo brillante, are not included in the first rank of concert pieces for the reason that even the best of them, the G minor concerto and the Capriccio, bear palpable traces of borrowing from Weber, and the remainder, though clever in detail and eminently practical, exhibit conspicuous mannerisms. Schumann's pianoforte concerto stands with Weber*? finTrMi:-

as a typical representation offlie RomantitTperjo^ ^ The_ opening Allegro was written first as a Ivric fantasia : ,tne tHteMfie'Z20""fffitl' CiiaF Allegro were adaed alter an interval of

^l»»i«.y<''»W*«M»ma»«n»g>iB<.»Mj»i>iiaMlt«»«»i»ii'»iii "" ...... '•••••••.naiiiiiiiliim*"""

some years. Traces of Schumann's aphoristic manner are * present throughout, particularly in the first movement, but the design is firmly and consistently maintained. Among his larger instrumental pieces there is none that offers a more complete and well-balanced expression of his individuality, and in none of his works (the Manfred overture and the Adagio of the C major Symphony excepted) has he so perfectly succeeded in setting forth the delicately passionate sentiment and the fiery exaltation that represents the normal state of his musical mind. The technique of the solo part is original and sufficiently effective, though there isnblb a

_ sake. The instrumentation, without being striking or clever,

muim Mtnm>**>'

yet leaves nothing to be desired. Less satisfactory is the Introduction and Allegro appassionato, Op. 93, an ambitious work, but rather monotonous and ineffective, and still less the Concert Allegro with Introduction, dedicated to Brahms, Op. 134. Chopin's two concertos were early works (1833-6). In the general outlines they recall the style of Hummel, as is shown in the arrangement of tuttis and solos, the distribution of cantilena and passage work, and in certain technicalities be- longing to the treatment of the solo instrument. The cantilena

CONCERTOS AND CHAMBER MUSIC 231

in the E minor concerto is Italian in spirit, and, like much of

melody, shows traces of the influence of Bellini l. Agdinst both concertos and indeed against all Chopin's pieces with ^orchestra the Krakoviak, which ranks with the Rondos of thfe concertos, the Andante spianato and Polonaise in E b, Op. 32,Ahe Fantasia on Polish airs, Op. 13, the variations on ' La ci/darem/ all of them fascinating from a virtuoso's point of vfew and very clever as compositions there is serious objection : icy apjeajito .better advantage without orchestra, and with the accompaniment J^JgdlW ? s^^p^d pianoforte. Chopin did not know enough about orchestral instruments, either singly or m combination, to employ them with proper effect., ffis tuttis Jack sonority, aflfl Wfl£fa the pianoforte "enters*. Jhe _would-be~" accompaniment fails' Jbo blend with_.the solo instrument. One or two fine and original effects, however, must not be overlooked, viz. the alternation of strings, pianissimo and unisono, with soft chords of wood winds, in the beginning of the Larghetto in the F minor concerto ; and the long tremolo of strings inter- spersed with the solemn pi/zicato of the double basses which supports the recitative of the pianoforte in the same movement -. Regarded from the pianoforte player's point of view, Henselt's3 concerto in F minor (1838), the most ambitious among that pianist's pieces, has very considerable merits. As a record of Henselt's personal achievements at the keyboard, the work is remarkable for the rich effects of sound attained by the use of widespread chords in the most complex form of arpeggio, the intricate filigree work of passages, the rapid fioriture, the broken

1 The Italian eighteenth-century vocal cantilena as transferred to the violin by Legrenzi, Tartini, Viotti, and afterwards from the violin to the pianoforte, constitutes the cantilena of Mozart's and Hummel's concertos, and from Mozart and Hummel, Field and Chopin in the main derive theirs.

3 Attempts at re-instrumentation, such as those of the F minor concerto by Klindworth and of the E minor concerto by Tausig, or of instrumentation direct such as that of the Allegro de Concert in A by Nicode, have not justified their existence. Almost throughout, the solo part, as Chopin has it, is complete in itself. The accompaniments, whether the composer's own or added by commentators, act as drags and obscure rather than enhance the effect.

s 1814-89.

232 THE ROMANTIC PERIOD

octaves, and the other devices which afford such ample oppor- tunity to the skill and endurance of the virtuoso. But in spite of the brilliancy of the protagonist's part, and the fairly good orchestration., there is a noticeable lack of convincing effect. One misses the chief of Henselt's peculiarities that smooth, sentimental ( Lieder ohne Worte ' tunefulness, which distin- guishes many of his Etudes and Impromptus and this perhaps more than anything else has stood in the way of a complete success. Though the concerto is well planned and carefully written, it breathes an air of pedantry, chiefly owing to the rather trite character of the themes which have all the mediocrite distinguee of Henselt's master, Hummel. Certain details in the two principal movements, Allegro patetico and Allegro agitato, are obviously appropriated from Chopin's F minor concerto, and thus derive from Hummel at second hand. The middle movement Larghetto in D b major and C Jf minor, f, a piece more distinctly Henselt's own in point of melodic invention and sentiment, starts and terminates as a kind of Nocturne after the manner of Field or Chopin. The centre portion, CJf minor, very remarkable from a pianist's point of view, contains a bold effect of sonority l : a broad cantilena for both hands in double octaves sustained by the pedal, in imita- tion of heavy bass and tenor instruments, at first piano, later on forte, and thenj^X", played simultaneously with the accompany- ing chords, which, also laid out for both hands, follow each main note of the melody in semiquavers higher up on the key-board.

Schumann expressed a wish to write a piece which should consist of an extended movement, the opening section to stand for the first Allegro, a cantabile section for the Adagio, and a bright close for the customary Rondo. Liszt tried to carry out some such idea. Taking a h^tf from Berlin?, Wflgper's system of Leitmotive, as employedj L strove^foy unity hjr

1 Recently copied key, effect, notation on four staves by S. Rachmaninoff, in the celebrated C minor Prelude.

CONCERTOS AND CHAMBER MUSIC 233

serve both in quick a^d glow, .*ti?lga J?yT-roafePJih-a ^QfifolHZ e.pfe P'prpf and, by stringing together the it& variants ami all accessory melodies in tfyr . This novelty in form, designed as self-depen- dent music without regard to a programme, proved to be a success. Indeed Liszt's two concertos in E b and A major, to which may " Be adSed'tfie so-called Concerto' for two pianos without orchestra in E minor, and the ' Todtentanz ' (Danse macabre), would rank among the best of concert pieces, were it not for the lack of weight and beauty in their main themes. As virtuoso pianoforte music these efforts are magnificent, the orchestration superb particularly in the Concerto in A and the wildly fantastic l Dance of Death V The Danse macabre, sketched in 1839 (written and rewritten in 1 849 and 1 859) a piece that belongs to the category of Berlioz' Nuit d'un sabbat, the Orgie des brigands, and Liszt's own Scherzo ironico, Mephistopheles, has for an avowed programme Orcagna's frescoes representing the Dance of Death, at Pisa, together with a reminiscence of Holbein's Dance of Death, at Basle. The piece consists of a series of grotesque variations on the old intonation of the ' Dies irae ' used by Berlioz in the Nuit d'un sabbat. Remarkably clever as an example of the extremes of pianoforte technique, and equally clever as an example of grotesque instrumentation, it is very effective when properly played to an audience in the mood for such things.

Bare enumeration must suffice for John Field's pianoforte Concerto in Ab (the 7th), Ferdinand Ries' in C$ minor (the 3rd), Sterndale-Bennett's in F minor (the 4th), Ferdinand Killer's in F$ minor (1863), Rubinstein's in G major and in D minor, Joachim Raff's in C minor (1870), as well as for Schumann's Concerto for violoncello, and his Concerto for four horns. All these works stand, more or less, apart from the main line of

1 The Concerto in E [? and a good number of Liszt's earlier works owe much of their telling effect to Joachim Raff, who for several years acted as Liszt's secretary.

234 THE ROMANTIC PERIOD

gradual change which marks the Romantic period, and none of them have left an appreciable impression upon professional executants, to whom they were, in the first instance, addressed and whom they mainly concern.

Both Mendelssohn and Schumann made strenuous efforts in concerted chamber music for pianoforte and strings. Mendels- sohn's two trios, D minor and C minor, and his two sonatas for 'cello and pianoforte, fine and finished as they are, particularly in the first movements and the Scherzos, have, for the present at any rate^ lost their vogue. His three early pianoforte quartets are of small moment.

Written in 1842, Schumann's famous Quintet in Eb soon became, and (apart from Brahms's in F minor) still remains, the favourite of concert audiences, despite the lugubrious In modo d'una marcia, which has been maliciously described as an ( Elegy on the death of a Philistine/ Next to this ranks the pianoforte quartet also in Eb. Schumann in his later years wrote three trios in D minor, F major, and G minor of which the first is the strongest and the last a failure. With the exception of the first Allegro of the Trio in D minor and perhaps the Adagio of this and the Larghetto of the Trio in F, none of Schumann's trio movements reach the level of Mendelssohn's. Two sonatas for violin and pianoforte, in A minor, Op. 105, and D minor, Op. 121, not so well considered, in point of form, as Mendelssohn's 'cello sonatas, belong to the period of Schumann's decline, when he wrote in feverish haste. The themes of the first movements of both sonatas are passionate, but the treat- ment produces an effect of effort and forced agitation. The slow movements, in F and G respectively, especially the first one, are in their way good, the finales mediocre. It is enough to mention the three Phantasiestiicke, Op. 73, for clarinet (or violin) and piano, and the three Roman/en, Op. 94, for oboe (or violin) and piano, as pieces perhaps better suited for the violin than either the oboe or clarinet. Miirchenbilder, for pianoforte and viola, Op. 113, Marchenerzahlungen, for pianoforte, clarinet,

CONCERTOS AND CHAMBER MUSIC 235

and violin, and Stiicke im Volkston, for violoncello and piano- forte, Op. 1 02, are in the main dull, though not without an occasional glimpse of beauty.

In his three string quartets, Op. 41, Schumann tried to make each movement exhibit some definite mood, in a manner which should depend for its effect upon a concise and direct expression of the idea rather than upon a complex scheme of contrasting subjects and balanced developments. And he thus managed to say things aphoristically which had never been so expressed before. He was well aware that in the most intellectual de- partment of instrumental music, the string quartet, any effect produced by mere mass or colour, anything which gave the impression of trickery, would instantly be revealed as an error in style, and in this respect his three quartets are more satisfactory than Mendelssohn's seven. Not that Mendels- sohn's music for stringed instruments can be called other than masterly; but the fact remains that his Quartets and Quintets contain, here and there, certain effects which suggest the orchestra or the pianoforte such, for instance, as the use of the tremolo in the first Allegro of the D major Quartet, Op. 44, No. i, and the Quintet in F minor, Op. 80, or the use of syncopated accompaniments in the first Allegro of the Quartet in E minor, and other devices of the kind. Mendels- sohn's fine Octet for strings, published as Op. 20, with its very clever and poetically suggestive Scherzo, is too well known to need any detailed description. It was a truly astonishing feat for a boy half-way through his seventeenth year l. Unification is attempted by a repetition in the Finale of the principal subject of the Scherzo with the same end in view, subjects from the first movement of the Quartet, Op. 12, in Eb, are also reproduced in the Finale.

Verdi's one contribution to chamber music, a string quartet in E minor, is interesting and original throughout. It consists of the usual four movements : (i) a rather lengthy, yet effective,

1 Compare Sir George Grove in the article on Mendelssohn, Diet. ii. p. 258.

236 THE ROMANTIC PERIOD

Allegro, £ , consistent in form, masterly in treatment, and full of novel effects; (z) an Andantino, f, in C, of piquant inven- tion both as regards melody and modulation; (3) a short Scherzo Prestissimo ; and (4) a Scherzo Fuga, ^, allegro assai mosso, very cleverly elaborated and brought to a bright close in E major. The work, like Borodine's second quartet, is worthy of more serious attention than it has received.

CHAPTER XI

PIANOFORTE MUSIC

AN anthology of Weber's compositions for the pianoforte would exclude all the variations on popular tunes and include the Concertstiick, the sonatas in Ab major and D minor, the so-called Perpetuum mobile (Finale, Sonata in C), the Mo- mento capriccioso, the Rondeau brillant in Eb, the Polacca in E, the Aufforderung zum Tanz (L'Invitation a la valse) and a few of the four-hand trifles, Op. 60. It is a small list, but a weighty one, for it consists of the most original and technically the most advanced pieces after Beethoven and Schubert and before Schumann and Chopin. In most of these pieces Weber has broken new ground and has proved to be the pioneer of later developments : in almost all of them he added a good deal to the keyboard technique of Dussek, Clementi, and even of Beethoven. Thus, for instance, taking a hint from Beethoven, he produced special and very distinct effects of sonority without the aid of the pedals, or by some particular use of them ; he developed Dussek's showy passage work of scales and broken chords as we know it in that master's sonatas in F minor, Op. 77, called L' Invocation, and Op. 70 in A 9* called Le Retour a Paris still further in the direction of pliant grace and glitter. Unfortunately Dussek's rather lax and patchy construction also reappears in the Allegros of Weber's sonatas, which, like those of Dussek, are concert pieces intended for the use of professional players. The predominance of sentiment over closeness and concentration of design is fully apparent as for instance in the fine first movement of the Sonata in Ab. A number of Weber's early pianoforte works look

238 THE ROMANTIC PERIOD

like stepping-stones to his operas ; even the Polacca in E, the Invitation, and the Concertstiick seem to belong to the same group and to point in the same direction. It is worthy of note that up to the present day Weber's influence is felt in the ball-room. The chevaleresque spirit and subtle grace of his Aufforderung zum Tanz (IS Invitation a la valse) l has changed the character of the German Walzer, which it made the richer by a note of brilliant gaiety, of dignified ease, and gentle manners, unknown before. Up to Weber's time the Walzer resembled a rustic dance known as the Landler (compare the Waltz in Der FreischiitZ) or the middle part of the Presto alia tedesca of Beethoven's Sonatina, Op. 79), or else it was like a fluent Minuetto with a touch of sentimentality like Schubert's so- called Sehnsuchtswalzer in A b, Op. 9, No. 2.

Then came Weber with his dashing Allegro con fuoco, and his frank enjoyment of life and movement. Thus, together with Schubert, he appears as the originator of the modern Valse and the father of the music of the Strausses and other masters of the dance. His influence is perceptible even in Chopin, whose lighter valses owe as much of their freshness and charm to the Invitation as some of his Polonaises owe their fire to the Polacca in E 2.

Before passing on to the works of Mendelssohn, Schumann, and Chopin, John Field must be mentioned3. To find any- thing so dainty in sentiment, so novel and perfect in diction, as Field's Nocturnes, No. 4 in A, £ and No. 7 also in A, f , one would have to go back to Mozart's Rondo in A minor, or forward to certain Nocturnes of Chopin. The designation 1 Nocturne ' is Field's own, and only nine or ten of the pieces

1 The graceful pantomimic music of the Introduction and the Epilogue accounts for that title.

2 Viennese dance music from 1820 to 1850, with Labitzky, Strauss the elder, and Lanner, reflects the spirit of the South German bourgeoisie of that time. With the younger Strauss and Gungl the valse becomes Pan-Germanic and cosmopolitan. With Chopin and Brahms it leaves the confines of the public ball-room and returns to the domain of graceful fancy.

3 Field, a pupil of dementi, was born in Dublin 1782, he died in 1837.

PIANOFORTE MUSIC 239

so called are genuine1. Field's frail little pieces are remark- able for originality of spirit and novel technique. Each bar shimmers with the gleam of romance. To realize their merit it suffices to remember their date, and to compare them with some of their offspring, such as Nos. I, 18, 19, 37, of Men- delssohn's Lieder ohne Worte, or the first set of Chopin's Nocturnes, Op. 9. High and varied as is the artistic quality of these particular pieces of Mendelssohn's and Chopin's, the mysterious voice of poesy does not so unmistakably resound in them as in Field's. Besides the two Nocturnes just men- tioned, those in Bt?, Ab, and Eb merit attention2. It is evident that Chopin in his elegiac mood is much indebted to Field. The kind of emotion expressed in Chopin's Noc- turnes, the type of melody with its graceful embellishments, the waving accompaniments in widespread chords with their vaguely prolonged sound supported and coloured by the pedals, all this and more Chopin derived from Field. Even from the executant's point of view, there is as much trace of the study of Field's pieces in Chopin's case, as there is of Clementi's or Berger's 3 in Mendelssohn's, or of Cramer and Hummel in Schumann's.

Mendelssohn intended his Lieder ohne Worte, eight books in all, to be straightforward, simple, and naive, that is to say, Mozartian in the expression of emotion. In their effects the majority of the Lieder are graceful and pleasing. There is refined musical sentiment, perfect savoir-faire, balance, com- plete finish, but not music in the fullest and warmest sense, as we get it so often in Schumann and Brahms. In spite of occasional titles Gondellied, Volkslied, Jagdlied, Friihlings- lied, Spinnerlied (the three latter, though generally adopted, are not furnished by the composer), each Lied rests solely upon its musical merits. Among the finest of them we may point

1 Publishers, by including arrangements, have increased the number to 18.

2 The latter is the prototype of Chopin's Op. 9, No. a, also in E t?.

3 Berger was Clementi's pupil and Mendelssohn's teacher.

240 THE ROMANTIC PERIOD

to the Gondellied in A minor, the Volkslied in the same key, the so-called Spring Song in A major, the first Lied, in E major, of Book I x, and the first, E b major, of Book II 2.

Besides the Songs without words, Mendelssohn's most im- portant contributions to the solo literature of the pianoforte consist of six Preludes and Fugues, a Scherzo a capriccio in F # minor, Op. 35, and the Variations serieuses, Op. 54. The first of the six fugues, in E minor, is a masterpiece on the lines of J. S. Bach, whose manner it often recalls both in texture and in movement 3 ; but with the remaining numbers the interest lies more in the Prelude, e.g. the one in A b, F minor, and Bb. The * Serious' Variations on a beautiful theme in D minor, Andante f , are remarkable for ingenuity of treatment and grouping, and for skilful handling of the instrument. The scheme is akin to that of Bach's Chaconne for violin and Beethoven's Variations in C minor with the very effective turn to the major key towards the end common to both, and a showy close in the minor. These Variations in the matter of invention break no new ground as, on the contrary, is the case in all of Beethoven's, the majority of Brahms', and in some of Schumann's Variations but the unity of style, the balance of effects, and the mature craftsmanship shown throughout are qualities beyond praise.

Other sets of variations, the posthumous Op. 8 a in E b and those in Bb, look like preparatory studies for the Variations serieuses. The two sets last mentioned can hardly count as representative pieces, though they are quite up to the average level of the master's work.

When Schumann took to composing, at first he devoted

1 Chopin's favourite.

3 It may be added that several of the Lieder make good studies for specialities of touch like Nos. 10, u, 15, 18, 24, 30 and 32. Fanciful sub-titles, such as have been furnished for all the Lieder by Stephen Heller, or poems cited by way of illustration, are entirely superfluous and often misleading.

3 Compare the wonderful fugues in Beethoven's Op. 101, 106, and no, which are also written on the lines of Bach, but in every bar bear the stamp of Beethoven's impetuous individuality.

PIANOFORTE MUSIC 241

himself exclusively to the pianoforte— he relied more upon his temperament and his gift of improvisation, than upon any system or tradition of style ; and from the earliest sketches for pianoforte to the apotheosis of Faust, he was influenced by the romantic spirit of the time as it reached him through litera- ture. With regard to the pianoforte pieces, it is wonder- ful that so early in his career he should have been able to condense and express so many heterogeneous suggestions emanating from non-musical sources. When he began to publish, his actual professional attainment was insufficient, distinctly less than that of Mendelssohn, Chopin, or Berlioz. His early manner, from Opus i to about Opus 9, was modelled on the style of his favourite author Jean Paul. ' I have learnt,' he said, 'more counterpoint from Jean Paul Richter than from any music-master/ The counterpoint is not particularly in evidence; but Heine's humorous account of Jean Paul's manner will throw some light on the matter. 'Jean Paul's periods,' wrote Heine, ' are constructed like a series of diminu- tive chambers, which are often so narrow that if one idea happens to meet another there is sure to be a collision ; the ceiling above is provided with hooks to hold up all manner of ideas, and the walls are furnished with secret drawers to con- ceal emotion V And Schumann's style, like Jean Paul's, was the result of impulsive improvisation and a constant desire to symbolize, with apparently no knowledge of the art of selection. He seems to be trying to reproduce Jean Paul's figurative and metaphorical mode of expression in terms of music, and appears to be playing with poetical metaphors, unable to find full expression for his meaning. In the early sets of solo pieces with suggestive titles, Schumann deals in terse epigrammatic phrases, which he joins one with another, but with little or no attempt at evolving anything further. So long as such phrases are sufficiently novel and the pieces concise, the result is both striking and fascinating, as is the case in the Intermezzi, Op. 4,

1 Heine-Schriften, vii. p. 268.

PAHSRKUTHKB R

34* THE ROMANTIC PERIOD

the Davidsbundlertdnze, Op. 63 the Scenes mignonnes entitled Carneval, Op. 9, Kreisleriana, Op. 16 (1837) (the title is meant to recall the fantastic figure of E. T. A. Hoffmann's Capellmeister Kreissler), particularly the highly original num- bers i, a, 4, and 6. The fantastic miniatures 1 that go to make up the majority of Schumann's publications from Op. 9 to 23, some numbers of the Phantasie in C and Kreisleriana, the Allegro Op. 8 and the Sonatas excepted, are each the brief expression of a single mood, each remarkable for concentration and power of suggestion. But this method of stringing together a number of independent paragraphs, as in the lengthy Hu- moreske, Op. 20, or in the last of the Novelletten, F $ minor, Op. 2,1, does not commend itself. The result is ill-balanced, incongruous, and, at times, even wearisome. In several sets of pieces earlier than the Carneval, such as Papillons, Op. 2, Intermezzi, Op. 4, Impromptus, Op. 5, Davidsbundlertdnze, Op. 6, there is much that is inchoate, though not exactly indefinite. Even in later and more mature pieces surprises and contradictions abound, as for instance in No. 7, Traumes- wirren, where a series of abstruse chords interrupts a lively piece of salon music. Sometimes Schumann obtrudes his particular whims and even his personality, for instance when he introduces mysterious quotations from his own Papillons in the Carneval or from the 'Abegg' variations in the last number of the Intermezzi, Op. 4.

Schumann in his early days reproduced what he had been taught or what he had studied. It is not surprising, therefore, that his early pieces contain traces of Hummel and of Schubert. His indebtedness to Schubert has been frequently pointed out, but that to Hummel and even to Moscheles has been overlooked. The Allegro in B minor, for instance, might well be called ( Reminiscences d'Hummel' ; compare also the Finale of the Sonata in F $ minor, Op. 1 1, and the Toccata, Op. 7, with Hummel's Sonata in F $ minor, Fantasia in E b, and 1 Hauptmann called them Sachekhen, Tit-bits.

PIANOFORTE MUSIC

243

the Scherzo of the Sonata, Op. 106, in D. The fantastic titles, some of them interchangeable ad libitum such as Arabesque, Blumenstiick, Novelletten, Humoreske, Nachtstiicke, Kreis- leriana derive directly from Jean Paul and E. T. A. Hoffmann. It has already been pointed out with regard to Schumann's orchestral works that he stopped short of actual programme music, although here and there he makes use of an inscription or a motto in verse or a musical quotation from some work of his own or of his bride, Clara Wieck. The same practice appears in his pianoforte music. For example, the great Phantasie in C major, Op. 1 7, exhibits the following lines of F. Schlegel's by way of a clue :

Durch alle Tone tonet Im bunten Erdentraum Ein leiser Ton gezogen Fur den der heimlich lauscht.

Midst all tones that vibrate Through earth's mingled dreams One whispered note resounds For ears attent to hear.

The ( ear attent to hear ' will readily perceive the uniting tones that run through all the pictures which the imagination of the composer unrolls.

a.-

244

THE ROMANTIC PERIOD

-£- '

<^^E3^EE£i EEES

M.< ~^jfL*^. m^ .

T-i= tr £ fL «: ^ ^ ^tt, •*-•!»•

^±=- *^H-^i ^T i r *r i £_*£=£=.*--

Again, the Intermezzo No. 2, Op. 4, is inscribed l Meine Ruh ist bin/ i. e. Gretchen's song from Goetbe's Faust ; Novelletten No. 3, as at first published, had a motto taken from Macbeth, and ' Verrufene Stelle ' in Waldscenen has a motto from Hebbel. Such things are meant to be merely accessories, indications of the mood in which the piece is to be heard or interpreted. Out of a total of seventy-four instrumental pieces thirty-four have characteristic titles and musical quotations. In the majority of instances the titles and inscriptions help to explain the mood of the music. It may be that the hearer misses his way if he goes from the piece to the inscription or the quotation, but from the inscription to the piece the path is straight and

PIANOFORTE MUSIC 245

a knowledge of what is intended adds to the pleasure of listening. The delicate touch of romanticism shows to per- fection in pieces like Des Abends. This little masterpiece conveys just that which music can convey, and words or colours cannot. There is something new here, both in spirit and in technique. In the latter respect the novelty lies in the continuous contradiction of the prevailing f time of the harmonic accompaniment and the quaver triplets of the melody, to which the constant use of the pedal adds a vague atmospheric effect ; a mere trifle, it may be, but complete and perfect in itself. The companion piece entitled Warum ? is equally good. Here the charm lies in the syncopation of the accompaniment against two responding parts which overlap, one phrase beginning before the other has come to an end. There is no need to dwell upon other such buds and flowers of poesy ; but attention must be called to Nos. 4 and 6 from Kreisleriana, ' Arlequin ' and ' Eusebius ' from Carneval, ' In der Nacht ' from Phan- tasiestucke, Op. 12, the Aria from the Sonata in F$ minor, the last of the Etudes symphoniques in G $ minor, and, above all, to those inimitable examples of musical miniature, the Kinderscenerij Op. 15.

Apart from his Concerto in A minor, none of Schumann's larger pianoforte works, i.e. the three Sonatas, the Etudes symphoniques, the Phantasie in C, Op. 17, the Faschings- schwank, the Humoreske, Op. 20, the Novelletten, are entirely without flaw or shortcoming. The power of invention and the emotion displayed are astonishing; so is the wealth of detail in rhythm, harmony, melody, and the persistency in the attempts to produce new effects of sonority. But the formative power is defective or imperfectly developed ; the materials are not completely welded together, the profusion of detail tends to obscure or upset the balance, the structure shows a lack of unity, the music is not so much an organic whole as it is a fusion of parts, and, at times, the treatment of the instrument leaves much to be desired. Indeed the key-board technique is

246 THE ROMANTIC PERIOD

here and there so clumsy that the novel effects fail to be effective.

To make Schumann's pianoforte music sound right a far greater and more persistent use of the pedal is required than in the music of any earlier composer. He is generally content with the indication f con pedale/ and leaves the application in detail to the executant. If the use of the pedal, no matter how frequent, were restricted to the sustaining of particular notes or harmonies there could be no objection; but if a composer chooses to sustain certain important notes or chords in defiance of the context and without regard to the ' muddy ' confusion and contradiction of harmony which results, he wilfully does an injury to his cause. Compare for instance, Schumann's Faschingsschwank, the Intermezzo in E b minor, bars fourteen and fifteen and bars twenty-nine and thirty, or Etudes sympho- niques, No. VIII, the last four bars where the blur of conflicting harmonies is the unavoidable result of the prescribed c pedal obbligato,' and these bars cannot be played without, the pedal. As instances of impracticable technical experiments see the Intermezzo entitled 'Paganini5 (No. 16 of the Carneval), and note the chords ff and pp before the return ; other examples occur in the last twenty bars of Kreisleriana, No. 3, in the Humoreske (middle of the Intermezzo), and in the third section of the Blumenstuck, where during some sixteen bars the thumb of the right hand is expected to hold down certain keys whilst the left hand is to touch the same keys staccato.

Of the three Sonatas (in F $ minor, Op. 1 1 ; F minor, Op. 14; and G minor, Op. 22), the first two are the strongest and warmest ; the third is formally finished, but not very significant. The Adagio Introduction to the Sonata in F$ minor, 'the most romantic of sonatas,' as Liszt was wont to call it, is full of passionate melody ; the Allegro which follows is forcible and vigorous, though, by a curious error of judgement, it reaches its emotional climax before the close required by the sonata form, and thus the interest declines at the very point where,

PIANOFORTE MUSIC 247

with Beethoven, it would have been chiefly concentrated. This means that the design of the entire movement is feeble with regard to the distribution of key centres. The boisterous Scherzo has an intermezzo 'Alia BurlaJ and a burlesque instrumental recitative to lead back to the theme. The Finale contains fine points, notably the first subject and a coda of entrancing warmth ; but it is long and patchy, there is evidence of its having been still longer in the first instance, and the shears seem to have been ruthlessly applied, so that the relative positions of key centres is even more anomalous than it is in the first Allegro. Compare the section in E b and in C with the context. For the absurd title of the Sonata in F minor, Op. 14, * Concert sans orchestre/ Schumann disclaimed responsibility. The second edition presents a revised and partially rewritten text. In turn fiery, passionate, tender, humorous, the work covers a wide range of feeling, though it suffers from uniformity of key, all the movements being in F minor. As is the case with the first of the solo sonatas, the materials are not completely unified, and there is now and again a sense of incoherence.

The phrase that acts as a musical motto is part of the subject for the Variations that form the third movement. Under various disguises it appears throughout the work. Among these disguises the last of the variations is particularly interesting as a very characteristic and personal effusion striking in its emotional sincerity.

248

THE ROMANTIC PERIOD

b.-

.JW-j-3

^•V^ A 4 •r^-jt

3*===:*:

\- I "^^ fl

IT J~ J.. ^ 1....J— ^— g|

J J^ J -I

^^-^^--^=2==-^^

•&- j J"

^=4=

**

-i-l

The final Presto, a sort of Toccata akin to the Finales of Beethoven^s Op. 37 and 54, with its incessant, almost delirious whirl of rapid semiquavers and the persistent rhythmical anticipation with every change of chord, produces a disquieting effect on the hearer l. The Andantino belonging to the Sonata Op. 22, f, a moonlight scene ' mit allem romantischen Zubehor 2,' is the best of the four movements, the final Rondo the weakest. The Novelletten, a total of eight numbers, are pieces of an illustrative kind in some sense programme music. To the

1 This movement is an extreme instance of the mechanically contrived effects of anticipation and syncopation, of which Schumann was so fond. Compare the long series of hardly interrupted syncopations in the middle of the first movement of Faschingsschwank, Op. 26.

8 Schumann's letter to Miss Laidlaw anent the PJumtasiestucke, Op. 12.

PIANOFORTE MUSIC 349

exquisite little miniatures called Kinderscenen already mentioned may be added the second of the Romanzen (F$ major), the charming trifles contained in the Album fur die Jugend, the Albumbldtter, Op. 124, and the four-hand pieces, Op. 85.

The Carneval, so called by Schumann himself, consists for the most part of very clever variations on a rather unmusical theme of four notes A, E \) (German Es), C, and B fl (German H) ; thus A-S-C-H, which notes, besides being the musical letters in Schumann's own name, also happen to spell the name of the birthplace of his friend Ernestine von Fricken l. These notes will be found embedded in most of the little pieces. The arrangement with a view to contrast, and the notion of a musical carnival, were an afterthought. The enigmatic presentation of the four notes as ' Sphinxes ' is evidently intended as a joke in the manner of Jean Paul a riddle without an answer2. J. S. Bach in his younger days set the example of using the letters of his name in this way, and Schumann took the hint in his six Fugues on the name of B-A-C-H, as did Liszt afterwards. The theme of Schumann's 'Abegg' variations, Op. I, is a sort of musical acrostic belonging to the same style of experimental composition ; so is the little piece d' occasion, 'Greeting to Gade,' G-A-D-E, in the Album fur die Jugend, and several others. The signatures Florestan and Eusebius (in imitation of Jean Paul's Walt und Wult) which Schumann appended to his critical articles, and which appear as noms de guerre in the Davids- biindlertanze, and the F $ minor Sonata, are meant to represent Schumann himself in his humorous and sentimental moods. The series of sketches entitled Waldscenen, already mentioned, contains one number, ( Einsame Blumen,' that in its delicate loveliness ranks high among the lesser pieces 3. ( Canonische

1 The ' Estrella' of the Carneval.

* Of course the breves are not meant to be played, though Anton Kubinstein used to bang them, slowly, fortissimo, and look solemn.

3 ' Gesange der Friihe,' the latest publications for the pianoforte, belong to the last sad years that were darkened by Schumann's genius ater.

250 THE ROMANTIC PERIOD

Studien* and 'Skizzen* for Pedalfliigel, Op. 56 and 58, together with the six ' Bach 3 Fugues, Op. 60, for the organ, seven ' Characterstiicke in Fughetten form ' for pianoforte, Op. 126, are fruits of the special studies in Counterpoint that Schumann began to make about the middle of his career. They are all interesting ; some, like the first of the Canonische Studien (a minor *•/•) and Nos. i and 2 of the ' Bach ' Fugues, are ingenious and beautiful.

As a composer of pianoforte music Schumann had but one superior among his contemporaries— Chopin, pre-eminently the poet of the piano, the genius of the instrument, who by divine instinct realized the impossible and hardly seemed conscious of the fact. There is in his best work a breath and glow as of the south wind. His fervour of spirit, the fire and force of his fancy, his pathos, and, in his lighter moods, his ease, grace, and consummate taste, are unique. Some part of his work, not a large part, appears over-refined, hectic, and morbid ; a small part belongs to the Parisian salon ; most is poetical work of a high order, perfect, not only in fragments and sporadically, but in entire pieces and entire groups of pieces. The music rings true. Chopin does not pose for pathos and emphasis. The sensitive delicacy of his nature kept him within the limits of courtesy and prompted him to shun the more violent accents of passion ; his canon of taste was the result of his temperament. He shrank from the robust, open-air power of Beethoven and was now and then inclined to emphasize those elements that make for sensuousness. The most artistic of romanticists, he never forgot or over- stepped the limits of the art. He avoided everything that might seem pedantic, dogmatic, or theoretical. He had nothing to preach or teach, unless it be his own incommunicable gift of beauty. The fire of his genius increased in intensity as time went on. His skill ( in the use of the sieve for noble words ' enriched his work and saved it from extravagance. To a student, the perfect finish of Chopin's pieces affords

PIANOFORTE MUSIC 251

evidence of the care and labour that he expended upon them. A comparison of the rather flimsy early pieces which were published as oeuvres posthumes with those that he published himself, say from Op. 9 to Op. 65, inclusive, will suffice to show that he rejected music enough to fill scores of pages. As he was fond of types such as the Mazurka, the Polonaise, the Nocturne, in which some sort of rhythmic and melodic scheme is prescribed at the outset, he virtually set himself the task of saying the same thing over and over again. Yet he appears truly inexhaustible ; each Impromptu, Prelude, Etude, Nocturne, Scherzo, Ballade, Polonaise, Mazurka presents an aspect of the subject not pointed out before; each has a birthright of its own. Chopin indeed is one of the rarest inventors, not only as regards the technicalities of pianoforte playing, but as regards composition. Besides being a master of his particular instrument, he is a singer in that high sense in which Keats, and Coleridge, and Tennyson are singers. He tells of new things well worth hearing, and finds new ways of saying them. He is a master of style a master of flexible and delicate rhythm, a fascinating melodist, a subtle harmonist. The emotions that he expresses are not of the highest : his bias is always romantic and sentimental. In his earliest productions his matter and manner are alike frequently weak ; in his latest now and then turgid. But in the bulk of his work, be the sentiment what it may, he makes amends for any apparent want of weight by the utmost refinement of diction. With him the manner of doing a thing is the essence of the thing done. He is ever careful to avoid melodic, rhythmic, or harmonic commonplace ; and he strove so hard to attain refinement of harmony that in a few of his latest pieces, such, for instance, as the Polonaise-Fantaisie, the Violoncello Sonata, and the last set of Mazurkas, he appears to have spun his progressions into useless niceties. The impressions Chopin received in Poland during boyhood and youth remained the principal sources of his inspiration. Personal impulses, later

252 THE ROMANTIC PERIOD

on, added radiance and intensity to his expression of passion, but the influence of the Parisian environment is felt only in such pieces as the Valses, the Bolero, the Tarantella, and a few of the Nocturnes.

Seen on paper, much of Chopin's work appears to be unduly ornate. Frequently, indeed generally except in the cantabile of the Nocturnes, Scherzos, and Sonatas the thought is stated in terms of ornament. In consequence there is a softening and clouding of outlines and things look very complex on paper, but in performance the main lines stand forth clearly enough.

A considerable variety and novelty of form may be found in the collection of pieces such as Preludes, Etudes, Impromptus, Mazurkas, and Ballades ; and, with the exception of the first movements of the Concertos and of the three Sonatas, Op. 35, 58, and 65 (the last of which is the Sonata with violoncello men- tioned above), there is no shortcoming. A very delicate feeling for balance and proportion is generally present ; no matter how novel the scheme or how complex the details, the outlines are simple, telling, and self-contained, requiring no title or explana- tion. Certain exceptional works, such as the Preludes in E minor and D minor, the Prelude, Op. 45, in C$ minor, the Etudes in A minor and C minor, Op. 25, the Berceuse, the Barcarolle, the Nocturne in G major, Op. 37, No. 2, the Finale of the Sonata in B b minor, Op. 35, seem to mark a new de- parture, as of poems upon new lines. The art is here so com- plete that it disappears.

Beethoven excepted, Chopin invented more that is valuable in the way of pianoforte effect and the technical treatment of the instrument, than any of his predecessors or contemporaries. His pupils and other witnesses agree in using the same words to convey a notion of his mode of playing his own pieces : * veiled, graduated, accentuated, evanescent/ { the harmonic notes vaguely blending, yet the transitions from chord to chord and phrase to phrase clearly indicated/ 'ever-changing and undulating rhythms/ ' indescribable effects of chiaroscuro/ i.e.

PIANOFORTE MUSIC 253

effects of sustained tone produced with the aid of the pedals. Heine, who was intimate by instinct with the nuances of Chopin's musical expression and style of playing, speaks of him with becoming warmth: 'Not only does he shine as a virtuoso with a perfect technique, but he accomplishes things of the highest value as a composer. He belongs to the sphere of Mozart or of Raphael. His true home is the dreamland of poesy. When I hear him, I entirely forget the mastery of his pianoforte playing and sink into the sweet abysses of his music.' (Heine-Schriften, x. pp. 287 and 342.)

Of his solo sonatas two alone count : Op. 35 in B b minor and Op. 58 in B minor; the third in C minor (Op. 4) is an early and immature work which was published posthumously.

Op. 35, the sonata that contains the funeral march (pub- lished 1840), is a great composition, Chopin's own from the first note to the last. There is no hint as to the composer's meaning in the title of any of the movements; all that we know is that the extremely emotional music was called forth by the struggle for independence in Poland, and that the spiritual connexion of one movement with another is to be sought in this direction. The first movement conveys a sense of strife, of a resolve to conquer or to die. It is a true sonata movement, with the usual two contrasting subjects, a working-out section and a recapitulation. Then follows a fervent Scherzo, having something of the same fierce impulse in its leading part, with a piu lento exquisitely tender and graceful ; then the Marcia f unebre, with the cantilena which we all know by heart ; finally there is a wail, like the night wind's cry over the graves of vanquished men *. If this Finale is played exactly as Chopin directs pianissimo and with hardly any gradation of tone the effect produced is weird in the extreme. This is the movement of which Mendelssohn is reported to have said : ' Oh, I abhor it. There is no music no art,' and of which Schumann asserted that it contained 'more mockery than music.' But

1 Compare the Prelude in E j? minor, Op. 28, No. 14.

254 THE ROMANTIC PERIOD

supposing it to be on the verge of, or even outside the pale of music proper, what is it to be called ? It is a piece unique in its way, and of a genius not less than that of the three movements preceding it. Both Mendelssohn and Schumann seem to have overlooked the fact that the little toccata is perfectly orthodox in form J. The Sonata in B minor, Op. 58, published some five years later (1845), is less concise and less well planned ; this is particularly the case in the first movement, of which the work- ing-out section is as lax in design as overwrought in style, and consequently somewhat chaotic in effect no matter how well played. But the long-drawn-out melodies of the Allegro maestoso and the Largo are remarkable even for Chopin, the supreme master of elegiac cantilena. In such melodies of Chopin there are frequent touches of Bellini no note-for-note resemblance, but obvious spiritual connexion. Such, for ex- ample, are the second subject of the first Allegro in Chopin's E minor concerto ; the corresponding passage in his B minor Sonata, Op. 58 ; the long melody in D b of the Scherzo, Op. 31 ; the melody that forms the trio of the Marche funebre ; the principal melodies in the Nocturnes in F$ and Ab and C$ minor ; the second part of the Nocturne in B, Op. 9, No. 3 ; the posthumous Impromptu in C $ minor ; the Prelude in D b, and many others.

The majority of Chopin's Etudes, unlike those of dementi, Cramer, and Moscheles, have no didactic purpose ; the best are characteristic pieces, studies for masters, not for pupils. The f Etiiden,' Op. 2 and 5 of Henselt, the ( Etudes d'execution transcendante ' and ' Etudes de concert ' of Liszt, may be said to vie with them. But if we look for originality, beauty, and variety of effect, Henselt's Studies are left far behind, and Liszt's, though remarkable from a virtuoso's point of view, lack

1 After four introductory bars it starts in the key of B j? minor ; with the 24th har it moves on to something like a second subject in the relative major, D |? ; with the 39th bar the four introductory bars recur, and the return from the 43rd bar to the end, in B j? minor, forms a complete recapitulation of the first section. The movement is in fact unified on the lines of certain Preludes of J. S. Bach.

PIANOFORTE MUSIC 255

the musical inspiration of Chopin's l. In a number of cases Chopin contrives to exhibit the theme of an Etude in different aspects and under different lights. The Etude in Ab, for instance (Op. 10, No. 10), weaves into a single texture the diverse aspects of the leading figure. Other such Etudes are Op. 25, Nos. 3 and 5. But, technicalities apart, the most glorious of the Etudes are the two in C minor, Op. i o, No. 1 2, and Op. 25, No. 12, and No. n in A minor, Op. 25 pas- sionate lyrics in the form of studies. Some of the Preludes, Op. 28, many of them little tone-poems that convey something of a passing perfume, correspond in style with particular Etudes contained in Op. 10 and Op. 25, though with the exception of Nos. 8, 16, 19, and 24 (F$ minor, Eb, Bb minor, and D minor), which are Etudes in the full sense, they are but sketches 2.

The Impromptus, so called by Chopin, have some affinity to the Impromptus and Moments musicaux of Schubert; the exquisite melody and style are of course Chopin's own 3. Two or three of the early Nocturnes, Op. 9, No. 2, parts of Op. 32, Nos. i and 2, show traces of Field 4, but in all the rest Chopin speaks in his own magical way. The Nocturne in G major, Op. 37, No. 2, is one of the most original and subtly beautiful pianoforte pieces extant. Other fine pieces are the Nocturnes in C $ minor and D b, Op. 27, the tragic Nocturne in C minor, Op. 48, No. I, the dreamy and perhaps a little over-elaborated Nocturne in E major, Op. 62, No. i, and the Duet-Nocturne in E b, Op. 55, No. 2, to which professed students of Chopin

1 In two instances, at any rate, Liszt's fitudes are Chopin at second hand. Compare Liszt's very clever ' fitude de concert* in F minor (No. 2) and the ' fitude d'ex^cution transcendante ' (No. 10) with Chopin's two Etudes in the same key. Henselt too imitates and dilutes Chopin witness fitudes, Op. 5, Nos. 2, 9, and 10.

3 Taken as sketches they may be compared with the second and third set of Beethoven's ' Bagatelles ' merely experiments, it may be, but of high value.

s Compare Schubert's Moment musical, No. 4 in C 4 minor, with Chopin's fourth Impromptu, the posthumous one, also in C ft minor.

* As Leopardi's darker mood reflects the pessimism of Byron, so Chopin's elegiac melancholy is closely in touch with that of Field and Bellini.

256 THE ROMANTIC PERIOD

have not yet given the attention that it deserves. e I do not care for the Ladies'-Chopin,' Wagner remarked in 1877, * there is too much of the Parisian salon in that ' ; but, whether one cares for the salon or not, the wit and finesse of French society seem to be more accurately caught and reflected in some of Chopin's lighter pieces than anywhere else in art. Un- doubtedly, within these confines of elegance and pleasant trifling, Chopin is unrivalled. But let no one suppose that the true weight and significance of his music is to be found there.

It is difficult to say anything adequate of that glorification of Polish national music which Chopin has accomplished in his Polonaises and Mazurkas. The latter range from mere jeux d? esprit to highly elaborated pieces. Some of the Polonaises, such as those in A, C minor, A b and F minor, are grandiose pictures of pomp and pageantry. In both Mazurkas and Polonaises melodic and rhythmical idioms belonging to Eastern Europe abound. To a western ear some of the exotic melodies based on unfamiliar scales and the resulting harmonies sound strangely impressive. The impulsive rhythm, the delirious swirl, or the languor of certain Mazurkas (Nos. 39, 3, 10-13, 40, 22, 23, 29-32), the dithyrambic enthusiasm, the barbaric din and clang, of certain Polonaises (Ab, F minor, Op. 53 and 44), convey impressions as of oriental exaltation, languid sensuousness, mili- tant enthusiasm, or dithyrambic excess. Compared with the fresh open-air spirit of Beethoven's Scherzos, the burly humour of Schumann's, the bustle of Mendelssohn's, Chopin seems to have struck a new vein ; he invented the sardonic Scherzo. In the four pieces which he calls by this title, the music has an air of impatience and questioning irony that alternates with mo- ments of dreamy pathos. A fifth Scherzo, the best of all, that in E b minor, contained in the Sonata, Op. 35, already men- tioned, belongs to this group. It is remarkable for conciseness, for concentrated energy, and for the strange grace of the trio four-bar and five-bar rhythms overlapping. In the Barca-

PIANOFORTE MUSIC 257

rolle, the Berceuse, and the Ballades pieces of an illustrative cast, experiments with an unwritten but implied programme Chopin discovered a form of expression peculiar to himself; the music, especially in the Ballades, appeals to the imagina- tion, like a narrative poem. The third Ballade, in A b, is the most perfect as a well-balanced and carefully-designed piece ; the second, in F, is the most fascinating and fantastic one longs for a clue to the mysterious tale which the music unfolds ; the first is perhaps the most impassioned ; the fourth is the most elaborate, as it is the richest. Certain harmonies that look unfamiliar in Chopin's text because of the complex notation with all manner of accidentals, sound like pure chords J. The practice of employing transient chromatic harmonies in the same manner as transient chromatic single notes or groups of notes began with Chopin, as, for instance, in the trio of the Polonaise, Op. 40. It has been much extended by later com- posers, notably by Wagner in Tristan und Isolde and in Parsifal.

Chopin always kept a metronome on his teaching piano. His tempo rubato was not an eccentric swaying to and fro in point of speed. 'The singing hand/ he taught, cmay deviate the accompanying hand must keep time.' l Fancy a tree with all its branches swayed by the wind the stem is the steady time, the moving leaves are the melodic inflexions V It follows that certain readings of Chopin, which are dear to the heart of many a virtuoso, must be discarded as caricatures. He disliked exaggerated accentuation : * It produces an effect of didactic pedantry.' 'You must sing if you wish to play hear good singers and learn to sing yourself.'

Since the expiration of the copyrights, Chopin's text has suffered much at the hands of editors. It is true that Chopin's method of notation does not always express the full musical sense, inasmuch as the effects of sustained sound, which are

1 Compare the Barcarolle, the Sonata in B minor, first movement, the Polonaise- Fantaisie.

8 This saying is vouched for by Liszt.

DANWRKUTHEE S

358 THE ROMANTIC PERIOD

best gained by the use of the pedal, are written out only in so far as the player's finger can hold down the key. In like manner his notation of fioriture, consisting of an irregular number of delicate grace-notes, expresses exactly what the fingers are to do, but no more. Nor is anything more desirable. Attempts to write out note-values in full, or to group the little ornamental notes so as to fit them into the time of the bar, are thoroughly misleading, for they tend to destroy the graceful ease of the music and to foster pedantry on the part of the executants. Tellefsen's presentation of the text, based upon Parisian proof-copies (the only proofs read by Chopin himself), is perfunctory and insufficiently revised for the press. Klindworth's valuable edition, a marvel of careful musical philology, contains too much of the editor's own views as to details of notation, fingering, &c. ; it is Chopin seen through the temperament of a very masterful editor. The only recent edition free from undue interference with the notation, valuable also as a partial1 record of Chopin's pecu- liarities in the matter of fingering, pedalling, and the like, is that of Mikuli, the last of the master's professional pupils.

Chopin and Liszt, as pianists and composers for the piano- forte, have often been compared. Facing the audience Liszt was triumphant. But when it is asserted that Liszt has out- stripped Chopin as a composer for the instrument we must protest the fact being that Liszt, in many instances, is but the imitator or the exaggerator of Chopin. Liszt's publications for the pianoforte solo may be ranged thus: I. Fantaisies dramatiques. II. Annees de pelerinage. III. Harmonies poetiques et religieuses. IV. Sonata, Concertos, Etudes, and mis- cellaneous works. V. Rhapsodies hongroises. VI. Partitions de piano of Berlioz and Beethoven's Symphonies, Beethoven's and Weber's Overtures, &c. ; Transcriptions of Paganini's Caprices, Rossini's Soirees musicales, Schubert's Soirees de

1 Frequently when Mikuli has no authentic materials as to fingering, he falls hack upon Kliudworth ; hence certain inconsistencies.

PIANOFORTE MUSIC 259

Vienna (Valses), Songs by Schubert, Schumann, Robert Franz, arrangements for two pianos of Beethoven's ninth symphony, Beethoven's Concertos in C minor, G, and E b ; and also the majority of Liszt's own orchestral pieces1.

Liszt was always ready to make speculative experiments in form. The earliest of these were the ' Fantaisies dramatiques/ which belong to the period of his early manhood, 1830—49-50, when he led the life of a travelling virtuoso. These pieces derive from the ' Variations brillantes ' and ' Variations de concert,' mainly on operatic tunes, which were equally beloved by the virtuosi and the public of those days. Starting from such facile types Liszt added an Intrada, certain connecting links to make the design continuous, and a Finale. The idea was to combine the tunes and variations in such wise that the entire piece, from the introduction to the final climax, should consist of a crescendo of effects reproducing the mood of some dramatic situation or condensing an entire act. By means of dazzling execution and the personal magic of Liszt himself, some of these fantasias, such as Norma, Sonnambula, Robert le Diable, and Don Juan, took the musical world by storm. As the vogue of the theme wanes the chances of arresting the attention of an average audience diminish. Still the beauty of certain melodies such as Mozart's ' La ci darem la mano ' and the Finale of Bellini's Norma will support Liszt's pianoforte effects for a long while to come. From the virtuoso's point of view, the technical difficulties contained in the fantasias remain as a supreme test of the executant's efficiency. Liszt exhibits true fancy in the general arrangement, and remarkable cleverness in the treatment of finger and wrist. The most celebrated of the fantasias, Don Juan, apart from its merits as a piece of display, is really interesting as a composition. Liszt takes Mozart's entire duet

1 At Borne in 1839 Liszt, finding no proper partner for ensemble music, gave the first of those pianoforte recitals of which we have since felt the benefit and the boredom. In his case the ' ennuyenx soliloques musicaux,' as he frankly called them, were entirely a one-man show, executant and composer combined.

S 2

a6o

THE ROMANTIC PERIOD

1 La ci darem la mano ' for the subject of his variations, and frames them by means of an Intrada, derived from the overture to Don Giovanni and a final presto based on the Brindisi from the third act. The whole piece is well designed and leaves nothing to be desired in consistency and unity of effect. The fantasias, Robert le Diable, Norma, La Sonnambula, and others, contain clever combinations of two or three different tunes played simultaneously a virtuoso trick contrived in imitation of Berlioz' practice of dovetailing diverse melodies and rhythms l. Taking into account the restricted possibilities of the key- board and the difficulties which arise from it, the following bars from La Sonnambula show the highest degree of manipulative ingenuity. They contain a combination of two tunes and a staccato bass, with a persistent trill at the top.

1 For examples of Berlioz' methods see pp. 120, 125.

PIANOFORTE MUSIC

261

A curiosity called Hexameron, 1837, which bears Liszt's name and is included in the thematic catalogue of his works, claims a few words. It consists of a series of variations on a melody known as ' La Marche des Puritains ' from Bellini's / Puritani ; the variations were composed and played upon six pianofortes by six pianists of repute : five played on full grands, Chopin sat at a two-stringed semi-grand. Liszt contributed the Intrada, the connecting links, and the Finale ; the others one variation each. Chopin's variation, a little Larghetto in E major, 1 7 bars of square time, shines like a gem set in pinchbeck.

Les Annees de pelerinage, published in three divisions, bear dates ranging from 1835 to 1883. For the most part the pieces are but slight sketches several among them, belonging to the first division, are strikingly true to nature and suggestive. They appear to be records of impressions directly derived from natural sights and sounds the beginnings of what may be called V impressionisms musical. The charm of such trifles as <Au lac de Wallenstadt,' ' Pastorale,' ' Au bord d'une source ' is indefinable. For adroitness and elegance in the treatment of the pianoforte it would be difficult to match the latter piece. Personal effusions, like cTre Sonetti di Petrarca,' which belong to the second division and are transcriptions for the pianoforte of certain melodies set to Petrarca's sonnets, have the peculiarly Italian note of ecstasy that distinguishes the melodies of Bellini. Other pieces, like * Sposalizio,' after Raphael's picture in the Brera Gallery at Milan, ' II Penseroso,' after Michael Angelo,

262 THE ROMANTIC PERIOD

( D'apres une lecture de Dante,' after a poem by Victor Hugo so entitled, show the first attempts at illustration made by the great musical illustrator. The third division of Les Annies is a collection of lesser value belonging to Liszt's old age.

Harmonies poetiques et religieuses, of which No. i, 'Bene- diction de Dieu dans la solitude,' and No. 10, ' Cantique d'amour,5 are the best, is made up partly of transcriptions of vocal pieces and partly of attempts at the illustration of poetry. All have descriptive titles or mottoes in verse. A set of six little pieces called 'Consolations/ belonging to Liszt's prime, about 1850, may be taken as corollaries to the Harmonies. Distinguished by a dreamy personal note, the meditative and pious ' Consola- tions ' take as high a rank among Liszt's pieces as the naive Kinderscenen among Schumann's. Nothing better than these little sets of miniatures could be found to exhibit the two composers' widely divergent temper and mode of work.

The Etudes, which head the thematic catalogue of Liszt's works, show, better than anything else, the transformation his style has undergone ; and for this reason it may be well to trace the growth of some of them 1. ( Etudes en douze exercices, par Fra^ois Liszt, Op. i,' were published at Marseilles in 1827. They were written during the previous year, Liszt being then under sixteen. The second set of ^Etudes, ' dediees a Monsieur Charles Czerny,' appeared in 1839, but were cancelled ; and the ( Etudes d'execution transcendante,' again dedicated to Czerny, f en temoignage de reconnaissance et de respectueuse amitie de son eleve,' appeared in 1852. The now cancelled copy of the Etudes which Schumann had before him in 1 839, when he wrote his brilliant article 2, shows these studies to be more extravagant and, in some instances, technically more difficult than even the final version. The germs of both the new versions are to be seen in the Op. i of 1827. Schumann transcribed a couple of

1 With the permission of Messrs. Augener & Co. the above paragraph regarding the ' Etudes d'exe"cution transcendante ' is quoted from the preface to the complete edition of Liszt's Etudes which the present writer prepared for them in 1899.

2 Gesammdte Schriften, iii, pp. 166-8.

PIANOFORTE MUSIC 263

bars from the beginning of Nos. J, 5, 9, and u, from both the new and the old copies, and offered a few of his swift and apt comments. The various changes in these Etudes may be taken to represent the history of the pianoforte during the last half of the nineteenth century, from the 'Viennese Square5 to the concert grand, from Czerny's Schule der Gel'dufigkeit to Liszt's Danse macabre. Czerny might have written the original exercise No. I, but it would not have been so shapely a thing as Liszt's final version. The difference between the two versions of No. I is, however, considerably less than that which separates Nos. a, 3, and 4 from their predecessors. If the earlier and the later versions of No. 3 in F and No. 4 in D minor were signed by different composers, the resemblance between them would hardly attract notice. Of No. a little remains as it stood at first. Instead of a reduction there is an increase (38 to 103) in the number of bars. Some harmonic commonplaces which disfigure the original, as, for instance, the detour to C (bars 9-16), have been removed. The remainder is enlarged, so as to allow of more extensive modulation, and thus to avoid redundancy. A short introduction and a coda are added, and the diction throughout is thrown into high relief. * Paysage,' No. 3 in F, has been subjected to further alteration since Schumann wrote about it. In his article he commends the second version as being more interesting than the first, and points to a change of movement from square to triple time, and to the melody which is superadded, as improvements. On the other hand he calls an episode in A major ' comparatively trivial,' and this, it may be noticed, is omitted in the final version. As it now stands, the piece is a test study for pianists who aim at refinement of style, tone, and touch. The Etude entitled 'Mazeppa' is particularly characteristic of Liszt's power of endurance at the instrument, and it exhibits the gradual growth of his manner, from pianoforte exercises to symphonic poems in the manner of Berlioz. It was this fitude, together perhaps with Nos. 7 ('Vision'), 8 ('Wilde Jagd'), and la ('Chasse-neige'),

364

THE ROMANTIC PERIOD

that induced Schumann to speak of the entire set as ' Wahre Sturm- und Graus-Etiiden ' (Studies of storm and dread), studies for, at the most, ten or twelve players in the world 1. The original of No. 5, in B b, is a mere trifle, in the manner of J. B. Cramer the final version entitled l Feux follets ' is one of the most remarkable transformations extant, and perhaps the best study of the entire series, consistent in point of musical design and full of delicate technical contrivances. ' Ricordanza/ No. 9, and e Harmonies du soir/ No. n, may be grouped together as showing how a musical e Stimmungsbild ' (a picture of a mood or an expression of sentiment) can be evoked from rather trite beginnings. Schumann speaks of the melody in E major, which occurs in the middle of the latter piece, as ' the most sincerely felt'; and in the last version it is much improved. Both pieces, 6 Ricordanza' and f Harmonies du soir/ show to perfection the sonority of the instrument in its various aspects. The latter piece, ( Harmonies du soir 9 in the first, as well as in the final version, appears as a kind of Nocturne. No. 10, again, begins as though it were Czerny's (a), and in the cancelled edition is developed into an Etude of almost insuperable difficulty (b). As finally rewritten, this study is possible to play and well worth playing (c).

Moderate

a. j

1 This is no longer the case ; we might multiply the twelve by teu and still be below the number, so much has the mastery over the mechanical difficulties of pianoforte-playing increased of late.

PIANOFORTE MUSIC

265

Prtito, moltoagitato

No. 12 also has been recast and much manipulated, but there is no mending of weak timber. We must also mention 'Ab Irato/ an fitude in E minor cancelled and entirely rewritten ; three fitudes de concert (the second of which has already been mentioned as Chopinesque) ; and two fine fitudes, much later in date and of moderate difficulty, 'Waldesrau- schen ' and ( Gnomentanz.' The Paganini Studies, i. e. tran- scriptions in rivalry with Schumann of certain Caprices for the violin by Paganini, and far superior to Schumann's, do not call for detailed comment. They were several times rewritten (final edition, 1852) as Liszt, the virtuoso, came to distinguish between proper pianoforte effects and mere haphazard bravura,

266 THE ROMANTIC PERIOD

and also, as the pianoforte makers afforded him better oppor- tunities in point of touch and carrying power.

About Liszt's technique as pianist and composer of piano- forte music, it may be said that it rests on the teaching of Czerny, who brought up his pupil on Mozart, a little Bach, more of Hummel and still more of Czerny himself. Hummel, the solid respectable classic, on the one hand, and Carl Czerny a trifle flippant perhaps, and inclined to appeal to the gallery on the other: these are the musical ancestors of the young Liszt. Then appears the Parisian incroyable and grand seigneur ; then the imitator of Paganini and Chopin ; and last the passionate and devoted student of Beethoven, Weber, Schubert, and Berlioz. Thus gradually there develops the mature master who, both as player and composer, bore to the end of his days the double marks of his origin *.

Taken together with the Concertos, Liszt's ambitious Sonata in one movement, B minor (( Senate in einem Satz, an Robert Schumann '), completely represents him in his more serious and manly mood. Etudes and Rhapsodies hongroises apart, it shows the ripest phase of his technique both as pianist and composer. The scheme consists of a novel and rather specula- tive device, akin to that of the Concertos ; that is to say, the composer strives for unity by employing single phrases in quick as well as in slow time, and by arranging the materials so as to make a continuous movement of the entire piece. And in the

1 Prom about 1863 onwards the writer has at times had the good fortune to hear Liszt play, in private, pieces of such various descriptions as the following : a number of Bach's Preludes and Fugues and single movements from the five later and several of the earlier sonatas of Beethoven, bits of Chopin, some of his own Ehapsodies, transcriptions from Schubert's ' Divertissement a la Hongroise? sundry valses by Schubert, fragments from his own operatic fantasias, &c. There was an air of improvisation about his playing the expression of a fine and grand personality perfect self-possession, grace, dignity, and never-failing fire. His tone was large and penetrating, but not hard; and every effect was produced naturally and easily. Performances, it may be of the same pieces, by younger men, such as Rubinstein or Tausig, left an impression as of Liszt at second hand, or of Liszt past his prime. None of Liszt's contemporaries or pupils were so spontaneous, individual and convincing in their playing ; and none, except Tausig, so infallible with their fingers and wrists.

PIANOFORTE MUSIC 267

case of this sonata as in the Concertos already discussed he does so on consistent musical lines without reference to a programme. The work is a curious compound of true genius and empty rhetoric, which contains enough of genuine impulse and originality in the themes of the opening section, and of suave charm in the melody of the section that stands for the slow movement, to secure the hearer's attention. Signs of weakness occur only in the centre, where, according to his wont, Liszt seems unable to resist the temptation to tear passion to tatters and strain oratory to bombast. None the less the Sonata is an interesting study, eminently successful in parts, and well worthy the attention of pianists.

Two Ballades, a Berceuse, a Valse-impromptu, a Mazurka, and two Polonaises sink irretrievably if compared with Chopin's pieces similarly entitled. The 'Scherzo und Marsch,' in D minor, an inordinately difficult and somewhat dry piece, falls short of its aim. Two legends, ( St. Francis of Assisi preaching to the birds,' a clever and delicate piece, and ' St. Francis of Paula stepping on the waves,' a kind of Etude, are examples of picturesque and decorous programme-music.

At the present day Liszt's reputation as a composer of piano- forte music rests largely upon the success of his Rhapsodies hon- groises. These transcriptions of Hungarian songs and dances, ostentatiously rhythmical, and by no means discreet in character, are the most dazzling of show pieces in the hands of virtuosi. The arrangement of some of them for full orchestra has doubled their brilliancy and increased their intoxicating effect.

Liszt adopts the incisive Hungarian tunes as the itinerant gipsy bands are wont to play them ; he finds many ingenious modes of imitating the orgiastic sounds of the cymbalon and ably develops the luxurious semi-oriental ornamentation and the crude harmony 1. The Rhapsodies, starting from short transcriptions of Hungarian tunes, were elaborated at intervals,

1 It seems worth while to add here that Brahms, in his Ungarische Tanze and all other movements that show the Hungarian influence, was careful to preserve

368 THE ROMANTIC PERIOD

published, cancelled, rewritten, and republished in some cases three times over. Schubert's Divertissement a la Hongroise was the prototype of Liszt's Ungarische Melodien, which began to appear in 1838; Melodies hongroises followed in 1846; the final version, entitled Rhapsodies hongroises, in 1854; this consists of a total of fifteen pieces, ending with the f Rakoczy March V and was accompanied by a curious attempt to prove the existence of a Gipsy epic.

In the so-called ' Partitions de piano,' transcriptions reproduc- ing orchestral effects as closely as the pianoforte permits and without regard to difficulties of execution, Liszt has accomplished some of his best work. The task he set himself was akin to that of an engraver, who must have knowledge of the painter's and designer's art ere he can hope to apply his own technique to advantage. It is astonishing to find how well Liszt succeeded in the apparently impossible cases of Berlioz' Symphonic fantastique and Harold en Italie. The transcriptions of Beethoven's first eight symphonies, for pianoforte solo, and especially that of the ninth (choral), as well as those of the three Pianoforte Concertos in C minor, G, and E b, for two pianofortes, are marvels of skill. So are the transcriptions of Beethoven's and Weber's overtures, the overtures to Berlioz' Francs Juges and to Wagner's Tannhduser, and of sundry other pieces culled from Lohengrin, Tannhauser, Derfiiegende Hollander, Tristan, and Die Meister singer. Together with the ( Partitions de piano ' certain arrangements for orchestra deserve to be mentioned, such as the arrangement of Schubert's Fantasia in C, set out so as to serve as a pianoforte concerto ; Weber's Polacca in E, to serve as a concert-piece with orchestra, the instrumentation of Schu- bert's ( Marches a quatre mains,' and the accompaniments to some of Schubert's songs.

the principal rhythmical and melodic characteristics of Hungarian music ; that he generally reproduced them in his own firm idiom, and very rarely touched upon the Gipsy vernacular.

1 Later additions to the number, all feeble, are of no account, be they authentic or not.

PIANOFORTE MUSIC 269

Liszt was also a master in the notation of pianoforte music a very difficult matter indeed, and one in which even Chopin frequently erred1. His method of notation coincides in the main with that of Beethoven, Berlioz, Wagner, and Brahms. Let the player accurately play what is set down and the result will be satisfactory. The perspicuity of certain pages of Liszt's mature pianoforte pieces, such as the first two sets of Annees de pelerinage, Consolations, Sonata in B minor, the Concertos, the Danse macabre, and the Rhapsodies hongroises, cannot be surpassed. His notation often represents a condensed score, and every rest not absolutely necessary is avoided ; again, no attempt is made to get a semblance of an agreement between the rhythmic division of the bar and the freedom of certain rapid ornamental passages, but, on the other hand, everything essential to the rendering of accent or melody, to the position of the hands on the key-board, to the details of special fingering and special pedalling, is faithfully recorded. Thus the most complex difficulties, as in the Fantaisies dramatiques, and even apparently uncontrollable effects of tempo rubato, as in the first fifteen Rhapsodies or the fitude ' Ricordanza,' or the ( Tre Sonetti di Petrarca/ are so closely indicated that the particular effect intended cannot be mistaken. One simple example of the notation of tempo rubato will suffice to show the method. In the vocal version of the first Sonnet (recently republished with the composer's latest emendation) the effects are obtained by the contrasting rhythms of the voice-part and the accompaniment (a), and a corresponding effect in the pianoforte transcription by means of slightly delaying the main notes of the melody (b).

1 As he did in the second part of the Nocturne in F A, Op. 15, No. a ; and in which Schumann frequently showed himself regardless of practical expediency as, for instance, in the first Intermezzo belonging to Kreisleriana, No. II, where what is intended for both hands is crowded into the lower stave, whilst the upper stave is left empty, or in Novelletten, V. (Vol. iv. p. 55 of Madame Schumann's edition), where, besides the perversely crabbed diction, a most awkward task is assigned to the left hand.

THE ROMANTIC PERIOD

Andante, un poco mono

dolce

3 " 3 3 > ,3

1 i M

Be - ne det to aia il giorno, e '1 mese e

iS^fi

a..

fl dolciitimo tempre

±=t* h frijji^zrirr— E -1 ' K

Fed. * -+ •*-* 9^

W^W j-.-.jij_3^^

S

1'an - no, e la sta - gio ne, e '1 tempo, e 1' ora, e '1 pun - to,

_J

^=P^

*qc

Sempre motto con intimo tentimtnto

R canto mezzo forte etpressivo un poco marcato^

Pho corda (Z'ac gnamento sem)

<L~^f S*

:ompa. rre dolce) I*

">-«•

r v ?r n5 S-

^.Zffi-

J j

-, J

1 J J - J J 1

\ ^s.t)

rr$-Tt\4J- M

J

pg2^3l3^^= Y1 r 'Ji a-g-^^T^

^

CHAPTER XII

SOLO SONGS

THE romantic problem, the many-sided question of an equi- poise between poesy and music, has presented itself under various aspects again and again since Beethoven's time. One impor- tant side of it, the relation of verse to music and of music to verse, is best studied in connexion with the German Lied, in which direct appeal to the heart of man is made by the fusion of the two. In countless instances attempts at such an alliance or interchange of forces have failed, chiefly for the reason that lyric verse possesses greater rapidity of movement than music. Even with the Germans, whose lyric poetry is closely akin to the folksong and therefore best fitted to associate with music, instances of complete success, such as Schubert's Gretchen am Spinnrade, Schumann's Friihlingsnacht, Mendelssohn's Fruh- lingsliedj Durch den Wald, Robert Franz' Zu Strassburg auf der Schanz and Stille Sicherheit, Wagner's Traume, Brahms' Wann der silbeme Mond, Feldeinsamkeit, and Wie raff? ich mich auf in der Nacht, are by no means common. Another reason is that poets often aim at effects resembling actual vocalization as Tennyson does in that despair of composers, the f Bugle Song ' or cast their stanzas in epigrammatic form with the point at the end, in which case the composer is at a loss, and must pass on to something else ere he can bring his melody to a satisfactory conclusion. On the musician's side it may be contended that the musical exposition, inasmuch as it is more protracted, makes a stronger appeal to the senses and therefore acts more powerfully on the emotions than verse alone.

ay* THE ROMANTIC PERIOD

Thus, for instance, the music in Brahms' song Wie raff't' ich mich aufin der Nacht, just mentioned, enforces the passion and melody of Platen's verse in a truly wonderful manner. But in every case the balance between the poetical and the musical factors is a matter of considerable difficulty, and a completely successful fusion is a rare achievement.

Apart from Brahms, who belongs to a later period, Schumann taken at his best is the greatest composer of songs after Schubert. When composing a song he was always instinctively guided by the idea rather than by any traditional conventions. Whether he utters a poet's passion or his own personal cry, Schumann is true and strong. In a supreme degree his best Lieder, such as Widmung,Ujbe^m^^arten durch die Lufte, Mondnackty JJie Jboioshlume, Schone Fremde, Er der herxtichste von alien, Waldesgesprach, possess the rare quality which Wordsworth failed to discover in certain metrical works of \Croethe's/old age absolute spontaneity. 'The verses are not inevitable enough ' was Wordsworth's way of putting it. But the lyrical pathos of Schumann's songs is indeed inevitable, original, spontaneous. Schumann produced the bulk of his Lieder in 1840, the year of his marriage. Saturated as he was with German romantic literature Jean Paul Richter, E. T. A. Hoffmann, translations of Byron and'Moore, Eichendorff, Heine, Riickert saturated still more with the emotional music of his predilection, Schubert, Beethoven, Chopin, he was often able to produce two or three songs in a single day. With him each song is the full musical utterance of the poem, without sacrifice of meaning and without repetition of words. The principal inflexions of the voice-part spring directly from the words, and every subtlety is emphasized by characteristic harmony or reiterated figures of accompaniment, or by some significant prelude, interlude, or coda. The balance between the voice and the instrument is well maintained, each factor makes for definite articulation and contributes towards a consistent and homo- geneous whole. Thus the majority of Schumann's Lieder are

SOLO SONGS 273

convincing as the direct utterance of his personal feeling *, Among the less well-known songs there is occasionally a touch of weak ecstasy, as in Du Rose meines Herzens ; or of pedantry, as in Rathsel, where the singer is made to sing of * Gottes- gelehrtheit und Philosophic' (Myrthen, No. 16), or Zahnweh, Op. 55, No. 2 ; or of sentimentality, as in Frauenliebe und Leben, No. 6, where there is something artistically wrong that invites parody. In certain other songs of a Spanish type, e. g. Der Hidalgo (Op. 30, No. 3), Der Contrabandista (sequel to Op. 74), as well as in the sets of songs, duets, and quartets, called Spanische Liebeslieder, Op. 138, it is to be feared that Schumann produced de VAndalou de Leipzig.

Schumann, in his Lieder and choral pieces, was the first of the Germans who troubled about correct declamation. Before him, neither in opera nor in simple songs did any one take offence at prosodical absurdities ; and it is significant that Weber, Marschner, and Mendelssohn educated men, and not devoid of humour should have allowed so many anomalies to pass. The source of many a curious instance of obtuseness in this respect composers, singers, and the public are alike implicated may be sought in the fact that the tunes of German popular songs and chorales, from Luther's time downwards, were generally older, often much older, than the words. Throughout the history of music, and not in Germany alone, it has been a common practice to fit new words to old tunes as, for instance, Moore did in his Irish Melodies and nobody seems to have cared whether or not the words and the tune meet on equal terms. To this, again, must be added the universal habit of singing successive stanzas to the same tune, as in the German

1 Ami, as he was the most German of contemporary musicians, and the most intimately connected with his own language, it would seem to follow that they had better not be sung in translation. Translations of German romantic songs, be they ever so faithful, are heavily handicapped because the musical and verbal effects belong together. Any translation of Schubert's Du bist die ftufi, Sei mir gegntsst, Der Wanderer, Der Erlkonig, Schumann's Mondnacht, Brahms' Feldeinsamkeit, or Wagner's Traume, is foredoomed to failure.

DANNRECTHER f

274 THE ROMANTIC PERIOD

Balladen of Zumsteeg and Zelter, or in many of the sea songs of Dibdin.

Compared with Schumann's method, the connexion of the melody with the verse in Mendelssohn's songs is rather lax. The melody reflects the mood of the poem well enough, but it rarely starts directly from the sound and sense of the words. Always facile, graceful, delicately refined, the music seems to stand aloof from the verse, and in many cases it appears as though either the words or the tune might be other than they are. This severance of verse and music marks Mendelssohn's songs as distinctly inferior to Schumann's ; though the best of them show a mastery of their own which, from a vocalist's point of view, is supreme as is the case, for instance, in the setting of Lenau's Fruhlingslied, Gpethe's Suleika I : Ach, um deine feuchten Schwingen, West, wie sehr ich dick beneide, and Suleika II : Was bedeutet die Bewegurig ? In like manner the setting of Heine's Auf Flilgeln des Gesanges is above praise as a vocal expression of the mood of the poem. Yet nevertheless, cases where Mendelssohn's melody chimes perfectly with the words are all too infrequent. Perhaps Frage, Op. 9, No. I, shows him at his best in this respect and a prosodical blunder such as Friihlingsmachtig in Op. 47, No. 3, at his worst.

There is the same difference between the vocal duets of Schumann and Mendelssohn as between their songs. Schumann's melody, in the treatment of the vocal parts, is more emotional and more closely in connexion with the words, whilst Mendels- sohn's is more effective from the singer's point of view. In the department of vocal quartets and part-songs without accompaniment, Mendelssohn's choral mastery shows to advantage; he proves himself superior in point of choral technique quite as clearly as in the choruses of the oratorios and cantatas J. Mendelssohn's productions in song form embrace eighty-three solo songs, thirteen duets, twenty-eight quartets for

1 Some of the weaker songs of Op. 8 and 9 (Nos. 2, 3, 4, and 7, 10, 12, respectively) are by his elder sister, Fanny Hensel.

SOLO SONGS 275

mixed voices, and seventeen quartets for male voices. Wer hat dich, du schoner Wald, one of his four-part songs for male voices, has become a folk-song in Germany.

Robert Franz's * Lieder und Gesange, forty-four books in all, containing some 340 numbers, have compelled the admiration but failed to elicit the full sympathy of later musicians. Finished in structure and technique, they lack the human sympathy of Schumann, the fluency of Mendelssohn, the weight and power of Brahms. Apart from three or four spontaneous outbursts, such as the celebrated Er ist gekommen and the less well-known Volker, spielt auf, Op. 27, No. i, Franz is subtle, delicate, contemplative ; he often exhibits a note of resignation or of quietism derived, it would seem, from the Freylinffhauser Gesang-Buch, the hymn-book of the later Protestant Pietists, so well known to J. S. Bach. Complete unity of expression was Franz's aim. A song, with him, is intended to be the reproduction of a single mood, simple or complex, and all the factors, voice and verse, melody, harmony, figures of accompaniment, are co-ordinated and made to con- tribute their share towards the end in view. Most of Franz's songs come and go like a gleam they are nearly all too short and frail for performance in public. The matter is ingenious in weaving up the voice parts and the accompaniment without interfering with the flow of the words. f I merely illustrate the words, and my music does not pretend to be much by itself. In this respect my Opus I is no better and no worse than my last ' (Opus 52). Franz is in fact generally content with the articulation of the poem, supplemented by a rather complex pianoforte part. Technically, he is a master in the fullest sense. But impulse is impulse, and he had little of it. Com- pared with Schubert, and still more with Schumann, he is impersonal and, in so far, weaker than they. He fails in personal charm, and his pathos at its most touching (e. g. Verfehltes Lieben, verfehltes Leben, Op. 20, No. 3) strikes the

1 Recte Knauth, 1815-92. T 2

276

THE ROMANTIC PERIOD

hearer as somewhat factitious. A sense of weakness is almost always present when Franz unites his music to verse by his friend Osterwald. He is at his best when he sets Lenau's, Heine's^ or Goethe's verse ; as in the following example :

Andante Innig.

Aua mei-nen groa . sen Schmer - zan ' canto .. N

macli' ich die klei - nen

-

dolce

S^

fed. *

**

Lie - der, die he ben ihr klin gend Ge - fie - der und

« •Mr1

-Jw. * Ped-

-&-.

Ped.

flat tern nacb ih - rent Her

ritard.

tiprcuivo

^i> JM-

*3^

Sie

SOLO SONGS

277

•j j,j^==£gg!E=^

fan . den den Wef zur Trau - ten, doch kom-men ale wie der Und

» i

m

PP

»=3iC

kla - gen, und kla gen und wol » len nicbt sa

gen, was

.

Pal.

278 THE ROMANTIC PERIOD

To understand Franz's ideal, his setting of Tieck's Ruhe, Stiss- liebchen, im Schatten der Matten, Op. 10, may be compared with Brahms' treatment of the same words (Romanzen aus Tiecks Magelone, Op. 33, No. 9), or his setting of Lenau's Durch den Wald, den dunklen, geht, Op. 52, with Mendelssohn's in the Frilhlingslied already mentioned. Franz usually expresses his sentiment in a rather complex and somewhat laboured manner, apparently natural to himself. His style is an amalgam of Bach, Handel, Schubert, Schumann, the German folk-song, and the Lutheran chorale, all of which Franz, from time to time, absorbed and reproduced in his own way. Hence, the quasi- abstract, scholastic sound of so much of his work. He was, in fact, essentially a scholar, and much of his most congenial work is to be found in his textual commentary on the scores of Bach and Handel.

Of a very different character are the pieces written by Berlioz, Liszt, and Wagner for solo voice with the accompaniment of pianoforte or small orchestra. The songs of Schumann and Franz are effusions belonging more or less to the intimacy of private life, whereas the aim of Berlioz and Liszt was to meet the requirements of singers or actors in the concert-room. Berlioz offers the Chanson and the pathetic Arioso ; Liszt both the Chanson and the German Lied, or the Lied expanded to a short scena ; Wagner, by the side of his three Chansons, has a Ballade in French, a German Lied, and five pieces which he calls e Poems set to music,' Gedichte in Musik gesetzt, and which, as reminiscences of the lyric stage, appear to form a link between the two categories. It is pleasant to find that the tendency to eccentricity rarely appears in Berlioz's Chansons. Ditties, such as Chanson de Paysan, Petit oiseau, La Belle Voyageuse, Elle s'en va seulette1, or the Villanelle No. i of Les Nuits d'ete (verses by Theophile Gautier), are gems with a real charm of their own. No. 2 of the Gautier set, Le Spectre de la rose, is a broad Arioso. Connaissez-vous la tombe blanche ? No. 5, has a touch of the

1 Paroles imitees de I'anglais de Thomas Moore, par Thomas Gounet.

SOLO SONGS

279

same melancholy which is expressed in the introduction to the Symphonie fantastique. No. 3, the so-called Lamento, Sur les Lagunes, and No. 4, Absence, rank among the finest histrionic examples of forlorn passion. For their due effect Les Nuits d'ete should be sung, not at the pianoforte, but with a small orchestra as originally written. Berlioz himself pointed to La Captive, Op. 12 the poem taken from Victor Huj^Ps Les Orientales as his supreme achievement in the way of solo song ; and, from the musical impressionist's point of view, this Reverie, as he calls it, is truly a fascinating piece. The exotic mood of the poem is well expressed in a compact and beautiful tune, and the scoring for a small orchestra is a marvel of skill a treasury of novel and picturesque effects, all directly illustrating the verse.

AiiJantino non troppo lento (/"= 132).

Hi jo n't) -

Fed.

S^

tais cap - ti

J'ai - me-rais ce p* .

I i I

5 J. J- ' d "F

E^

Perf.

a8o

THE ROMANTIC PERIOD

et cet - te mer plain-ti

I

f

Ptd.

# ' Ped.

Et ces champs do ma - is .*.

Et

+.

Jlf.0.

y/r/i-o rforzato

fP

'g— r =f.

ces ns - tres sans nom bre. Si

le lon

du luur

«/ P

4=-

*/ 1'ed.

som - bre NM - tin - ce - lait dans 1'om - bre Le

LLJ

# Fed.

~t *

SOLO SONGS

281

•a bre da Spa his,

" w r

^nHEJ.! JI

-Ped.

cJ

/LL1

* Fed.

Fed.

Fed.

Liszt, when composing for solo voice (a total of about sixty songs, published in nine books), took less account of the poem as a whole than of its successive details. He tried to intensify the effect of his lyrics by emphasi/ing the more important single words or clauses, or by strongly contrasting one word or clause with another, and was thus led to develop an exaggerated style that tends towards the incoherence of melodrama. Brief and unified expression is rare with him, though we meet with it in his setting of Victor Hugo'flj ^Q^^ntt dis^nt-ils ? Oh^guand je dors, and S*ll esl 'urT charm an t (jazoiif as well as in Es muss ein wunderbares sein, and Utjoeihs Kennst du das Land ?

again 'Tie approaches the Ballade, as n he most picturesque of his songs, Die drei Zigeuner, or the Scena, as in Lorelei. Frequently he appeals to that public taste which is caught by over-emphasis, as in Enfant, si fetais roi and Vergiftet sind meine Lieder, or in Es war ein Konig in Thule and Wer nie sein Brod mit Thrdnen ass, in which latter song

282

THE ROMANTIC PERIOD

the words f Der kennt euch nicht, ihr himmlischen Machte/ are at first directed to be ( spoken softly,' and, when they recur, ' to be sung with full force/ Again, he is disposed to indulge in over- sentimentality, as in Ich mochte hingehn and Tre Sonetti di Petrarca, or in decorous platitudes, as in the two songs called Marienstrausslein, or in languorous sensuousness^ as in Liebes- traume, Nos. i and 3. In point of accent and declamation Liszt's French songs are admirable.

Trit ani»if

S 5 * 14

^r CJ

staccato (quasi Chitarra)

~h -(• I r

see

parU

Com . ment,

di-saient ilg,

*A

[iVirj^r -*-j|

1 j =3=1

vec nos na -

eel . les

^ J* -T^i i**i

fuir les

^^-J J PJ ,___ _(. _* ,

1 i |. b 1

SOLO SONGS

283

al - gua-zila?

m

ry r ^X u^

itn pen retenu

MH p«u reftnu

/ifV i

(fo "g j«^: £/ ~~V~

•9-S^i

»»<<><-«••"">

284

THE ROMANTIC PERIOD

a tempo

The German ones leave much to be desired ; an annoying discrepancy is felt to exist between the true sounds of the words and the musical accent for example, the stress on du and die in Goethe's Kennst du das Land ? is detestable ; and there are other errors of declamation in the rest of the song.

Sehr langtam, tehniuchttvoll

Kennst du Uas Land, wo

una corda

*

^m

_ J_ nrj

die Ci - tro - nen blUh'n? Im dun - klen Laub die

Ped.

:i='- P«^.

SOLO SONGS 285

fe=£gssEE Eg^=,F^=j=E£EEgEl EEi

Gold .o ran - - gen gllili'n.

Musically considered, Kennst du das Land? is a song remarkable for its romantic colour and the exquisite touches of longing expressed in the refrain. The atmosphere of Die drei Zigeuner, too, is wonderfully characteristic.

Wagner's French Chansons, Dor*, mon enfant ; Mignonne ; Attente ; and the Ballade Les Deux Grenadiers a translation of Heine's Die beiden Grenadiere belong to the period of his first sojourn in Paris (1841-2), when he finished Rienzi and wrote Der fliegende Hollander. Another Ballade, Der Tannenbaum, words by Scheuerlin, was written at Riga in 1839, when only two acts of Rienzi were completed. Yet this piece has some connexion with the style of Tannhauser, as the Fiinf Gedichte, the verses of which were written by Frau Mathilde Wesendonck (1855-7), are closely connected with Die Walkiire and Tristan. The earlier of Wagner's songs and vocal pieces differ as much in style from the later as his operas differ from the tone-dramas. The third and fifth of the Fiinf Gedichte, < Im Treibhaus' and ' Traume/ are offshoots or forerunners of Tristan und Isolde. 'Traume' prefigures the love-scene in the second act, flm Treibhaus' recalls the instrumental introduction to Act HI. Again, ' Stehe still,' No. 2, is connected with the third act of Tristan, ( Schmerzen,' No. 4, and ' Der Engel,' No. I, with Die Walkiire. It is curious to watch Wagner listening to him- self, as it were, in his own workshop. His personality is as perfectly revealed in these five songs as in the later Siegfried- Idyll t which belongs to Siegfried, Act III.

a86 THE ROMANTIC PERIOD

Wagner laid great stress on prosody. Compared with Schumann, Robert Franz, or Brahms, who each are occasionally lax in their ways, and more or less consciously take up a recal- citrant position in regard to declamation, Wagner is an uncom- promising purist. In every case, he insisted, the spirit and sense of the language must be respected and the laws of prosody and of musical rhythm should form an equilibrium.

Lapses such as Robert Franz's Ich stand gelehnet an den Mast, or Ihr Thrdnen, bleibt mir aus den Augen (Wasserfahrt, Op. 48, No. 3), roused his ire. f If Franz,' he said, ' did not care to sacrifice the characteristic figure or the tune, he might have found a better figure or a better tune.'

In connexion with the partially dramatic lyrics of Liszt and Berlioz mentioned above, it seems convenient here to touch upon certain earlier forms of Ballade and of Melodrama proper. Carl Loewe 1, who is practically the originator of the German Ballade as it now exists, relies on various means of artistic effect, justifiable it may be, but not entirely musical. Declamation, histrionic changes of voice, and even mimicry are called upon to bear their part. With the aid of the musical actor's art, some of Loewe's Balladen, such as Edward, are effective enough, but they need such external assistance to cover defects in the music. The poetry usually is allowed to tell its own story, but the music is subordinate. There is a sense of insecurity. Rarely does the musical mood embrace the entire poem, and almost invariably the stress is laid upon the externals of the story rather than upon the lyrical emotion which underlies it. The impression left is that of a partially musical recitation by an actor, not the consistent outpouring of a musician 2.

Excepting the pieces mentioned, to which may be added Herr Oluf, Prinz Eugen, Der Pilger von St. Just, and Der Wirthin Tochterlein, the musical ideas are neither new nor deep, and they

1 1796-1869.

8 This may help to explain Wagner's inordinate fondness for Loewe's Edward and Erlkimig, as well as the fascination which these pieces had, and still have, with operatic singers.

SOLO SONGS 287

occasionally approach the confines of bathos. Loewe's manner of writing for the pianoforte is a little in advance of the point reached by his predecessors Zumsteeg and Zelter, yet, like theirs, it is somewhat commonplace. There is abundant evidence of a gift for rapid improvisation, but little restraint or self-criticism. Hence the superiority of certain Balladen by later masters, such as Schumann's Belsatzar, Die beiden Grenadiere, and Die Lowenbraut. The early dates of Loewe's best work must not be overlooked; Edivard and the Erlkonig belong to 1818, Der Wirthin Tochterlein to 1 824; while, on the other hand, Archibald Douglas appeared in 1857, which was Loewe's sixty-second year a date which may account for certain instances of appa- rently direct indebtedness to Wagner's Derfiiegende Hollander. Melodramatic music, such as that contained in Mendelssohn's Midsummer Night's Dream and Schumann's Manfred, is an offshoot of the great melodramas in Beethoven's Fidelio, or in Weber's Freischiitz and Preciosa, or a combination of both styles, as in Marschner's Vampyr. Schumann, besides the three melodramatic fragments in the music to ^Byron's Manfred the calling of the witch of the Alps, the invocation to Astarte, and Manfred's address to Astarte published three pieces for declamation with pianoforte accompaniment, Schb'n Hedwig and Vom Haidenknaben, both by Hebbel, and Die Fluchtlinge, Op. 122, taken from Shelley's Fugitives. These pieces, again, prompted Liszt to write melodramatic music to illustrate a recitation of Burger's Lenore, and a Ballade by Lenau called Der traurige Monch1. As early as 1773 J. J. Rousseau, with his Pygmalion, set the example, and the amateurish character of this early instance appears to have tainted all its successors. There was the alternative of recitation and music, musical illustration or mere support of the reciter's voice then again music, and again recitation, and so on. The unity of effect was difficult, if not

1 Liszt's music to the latter poem is a curious experiment in ugliness an entire piece built on a whole tone scale. Compare Verdi's Scala enigmatica quoted above, p. 223.

288 THE ROMANTIC PERIOD

impossible to attain, because as the reciter's topic or his mood changes, the music must change and, as music must recur to its beginning or remain inchoate, the two aspects rarely fit together.

In connexion with the stage, melodramatic music is accept- able in so far as it underlines the words ; it supplies emphasis and descriptive touches and, in the intervals of speech, completes the expression of the emotion. How well it can serve in the latter capacity may be felt in the great prison scene, Act II of Beethoven's Fidelia, and the invocation and address to Astarte in Schumann's Manfred. Apart from the stage, however, a true fusion of the poetical text with the musical accompaniment seems impossible. The speaking voice and the music fail to blend. If the reciter is competent he will absorb the interest, or else the accompanist will disturb the reciter. Musically considered, melodramatic effects appear tolerable only in con- nexion with a poem that contains a certain proportion of definitely musical elements, such as Tennyson's Enoch Arden, for instance, to which Richard Strauss has recently added illus- trative music ; but even in this case the result is of doubtful value.

CHAPTER XIII

VARIOUS WORKS BY OTHER MASTERS

REFERENCE to the followers of those Romantic masters who founded a school, as the phrase goes, has hitherto been avoided. We may now turn to the adherents of Mendelssohn such as Niels Gade, Sterndale Bennett, Rubinstein, Stephen Heller, Sullivan ; of Schumann such as Volkmann, Kiel, Goetz, Theodor Kirchner, Jensen ; of Liszt and Berlioz such as Peter Cornelius, Hans v. Billow, and the eclectic Joachim Raff. It is pleasant to record that most of these men were independent enough to indulge their personal note and guard their spon- taneity, though they frankly adopted the methods and even some of the mannerisms of their leaders.

But before going into details we may consider the case of two composers who are not followers at all, and whose claim to recognition rests entirely on its own merits: the twin masters of English church and organ music (they happen to be father and son), Samuel Wesley (1766-1837), and Samuel Sebastian Wesley (1810-76). And in connexion with their productions we may also touch upon certain specifically English forms of vocal music the glee, the round, and the catch.

In the musical history of the nineteenth century the work of the two Wesleys is of real importance. They are by far the weightiest composers who wrote for the Anglican Church Service at a period when English music in general was at a low ebb. They tower above their English contemporaries, laymen or churchmen ; and in their particular department in which they are by no means imitators need not shun com- parison with continental celebrities such as Spohr, or even with a master such as Mendelssohn. The fine eight-part antiphon for double chorus and organ in In exitu Israel, the motets Dixit Dominus and Exultate Deo, the bold motet for two altos, tenor, and bass in B[7, Levate capita vestra, and the noble Ecce Pants in D minor for soprano, alto, tenor,

DANNREUTUER U

290 THE ROMANTIC PERIOD

and bass, the ' Carmen funebre ' for five voices, Omnia vanitas, by Samuel Wesley1, together with S. Sebastian Wesley's eight- part anthems, Let us lift up our hearts, and 0 Lord, thou art my God ; his Morning and Evening Service in E ; his masterly five-part anthem The Wilderness; his four-part anthem Man that is born of a woman, with its direct expression and fine pathos ; his beautiful setting of the Nicene Creed, which forms a part of the Morning Service just mentioned ; and the poignantly expressive Wash me throughly from my wickedness are the most valuable of their pieces. There is nothing in the range of modern religious music more sincerely felt and ex- pressed than, for example, the anthem last mentioned, Wash me throughly neither in Spohr, with whose practice certain chromatic progressions seem to coincide, nor in Mendelssohn, with whose oratorio style there is a certain resemblance in phraseology. S. S. Wesley's way of expressing religious emotion appears more individual than either Spohr's or Mendelssohn's, and it is for that very reason better worth hearing. Always in close connexion with the traditions of English vocal music, the choral technique in the work of both masters is of a high order witness the elder Wesley's five-part madrigal 0 sing unto my roundelay, and Samuel Sebastian's five-part glee J wish to tune my quivering Lyre. The sheer musical invention in S. S. Wesley's 0 Lord, thou art my God, in the Credo belonging to the Morning Service, and in Wash me throughly, is that of a virile genius, who knows his J. S. Bach not only contrapuntally but emotionally, and loves him2. The quotation subjoined may appear inordinately long, but it is not possible to convey in a few bars an adequate idea of the persistent strength of this contrapuntal music, that ought to be studied and recog- nized as masterly wherever the English language is spoken.

1 A complete list of S. Wesley's pieces will be found at p. 446 of Grove's Dictionary, vol. iv.

3 The elder Wesley's enthusiasm for Bach's organ works and the 48 preludes and fugues, which the son fully shared (S. Wesley, together with C. E. Horn, brought out the first English edition of Das icoftltemperirte Clavier in 1110), is well shown in the familiar epistles known as the 'Bach letters ' written to Benjamin Jacob in 1808, which were not published till 1875.

VARIOUS WORKS BY OTHER MASTERS 291

Andante e marcato

/ubh (S

i j

L__ _

ORGAN. m

5> lf(f -. , =

7»tf Sempre Ugatt

jS>

i

1 r '

96.

,^.

F"^5^ H ^

i r r !»-^

=4b£=J

PeO. 16/«.

-r

^-^-r^

r r

£=&

J j i

r r r r r r ^ A- J-

^n-

f

^^ i . JJ J

Treble

p

Alto.

1

rt L Tenor (8ve lower).

mf Decani.

1- 1

Bass, mf Decani.

,

Q

Lord,

O

Lord,

The

U

art my

m

-j,n

a

J ,

^ J.

, ***lr-r- qp-r— p~P

Mf f

rp: . , -r-j- _-_^_^_ -j,-—-^

u r Ui i f i^

U 2

393

THE ROMANTIC PERIOD

g====j

~\ r =|

Thou, Thou . . . artmyGod;! will . . ex-alt. . . Thee, I . . . willpraiseThy

Thou . . . art my God;. . I will ex- alt .... Thee, I. . . willpraiseThy

-

sit

j J^H J-?J-Jh4

£^3: -^ +~H*'/'-

.J.

rsCT

-•i JbJ.J h-1

Cantoris.

cBm.

y coun-sels of old arefaith-ful-

Thy Cantoris.

th-ful-nessand truth. Thou hast, Thou

dim.

"fny coun-sels of old arefaith-ful-nessand truth. Thou hast, Thou Cantoris.

Name; Thy coun-sels of old arefaith-ful-nessand truth. Thou hast, Thou Cantoris.

Name ; Thy coun-sels of old arefaith-ful-nessand truth. Thou hast

""Ml '—

£**&*

~P

±=Fi

VARIOUS WORKS BY OTHER MASTERS 293

p Decani.

O^i ^~|~b^ . * y -

bant. . done won-der-ful things; Thy couu-sela of old . . are faith-ful-nessand era. p Decani. >.

I I =T

liast done won-der-ful things; Thy coun-selsof old are faith-ful-ness and

-_ era. P Decani. =*

hast done won-der-ful things; Thy coun -sels of old are faith-ful-ness and

era. p Decani. >

frr-er r.

done . . won-der-ful things; Thy coun-sels of old are faith-ful-ness and

•*^ 1, 25 » Ofc-' K k ^S

i i -'i r

r- r

r

truth. O Lord, Thou art, . .Thou art my Qod; Thy coun - sels of =- / '»/ > P

•i -J U "

truth. O Lord, =- f

Thou art my God ; Thy coun - sels of mf > P

tnith. O Lord,

Thou art my God ; Thy conn - sels of mf =- p

tnith. O Lord,

Thou art my God ; Thy coun - sels of

J

294

Decani.

THE ROMANTIC PERIOD

mf

old are truth, are truth,

Thy coun - sels of

mf =~

old are truth, are truth,

are truth,

mf >

old are truth, are truth,

are truth,

mf a.

old are truth, are truth,

Cantoris. J>

are truth,

Thy coun - sels of old . . are

3^F

coun - sels of old . . are

Thy coun - sels of old . . are

Thy coun - sels of

VARIOUS WORKS BY OTHER MASTERS 295

old are faith - ful -ness and truth. . . Thou

Thou art,

O Lord, /

Thou

m

Thou art, O Lord, Thou

crei. f

Thou art,

O Lord, Thou

j=r=f=fr-

r *

faith - ful - ness and truth Thou, Thou

faith - fill-ness and truth

Thou, Thou

I

faith > fill-ness and truth.

Thou, Thou

old are truth.

1 - J - ,

Thou, Thou

m

m

^

296

THE ROMANTIC PERIOD

(\$~?~3~ 1 1 ^ 1

, . 0p-^ - ^_ """•

1— ' i - F —&

1 j ^- 1

art my God, Thou, Thou . . art ere*. ff dim.

-?- 1" -H my God ;

art my God, Thou, Thou . . art ere*. ff dim.

j£6k-_fSr= i -— \'=-<s> \ h— ^ »* m 1 '-(»— -

my God;

-r~ \ (*-^=\

art my God, Thou, Thou . . art era. jC2. jf x- "•^ dim.

-J P r— J

my God ;

f \ I f— j

art my God, Thou, Thou . . art

n u _ cres. ff ^-— ^ rfi»i.

my God;

art my God, Thou, Thou . . art . .

n __ ere*. ff dim.

my God ;

art my God, Thou, Thou . . art

n L ere*. ff dim.

my God ;

--? \r^-*A

art my God, Thou, Thou . . art

cres. ff dim.

__t -i i

my God ;

art my God, Thou, Thou . . art

-fr- ^ ' ^ my God;

py_b L^_ J D Us; off . 1»_^ ,_ _^L £T p .5- 53 '—^y—. :t-L~L-p-1

...J | rt? HI=? QjS . __

( vg$L-jp=^33. ^—^-^ F^ -^F <*

: i—

Lf ' .

VARIOUS WORKS BY OTHER MASTERS 297

Not less magnificent is the closing chorus of the same anthem. Indeed, the best examples of S. S. Wesley contain an expression of the highest point up to that time reached by the combination of Hebrew and Christian sentiment in music. They are well worthy of comparison with Mendelssohn's psalms, with the best things in Spohr, and with the Beatitudes of Liszt and of Cesar Franck, to which their relation may be illustrated by the following excerpt from the anthem Wash me throughly.

TREBLE.

ORGAN.

IT-™

Waah Larrihetto

tl iroi

igh - ly

s=»=

from in

v

"

wick - ed - ness,

pj .J |

ir

^

g:

^ r-

\

^

-* 1

~

Sir. /<«<< * 1

E

cT a

A-

id

fo

r

giv

-T 1-

1 J 1

e me

1

r *•

M ni)

^ J

E **

sin, . .

pi

»-:

f

>«_

S

-

r r

Lr— r^

^

£

la

1^

m b ^ 1 '

__! L_

wiish me

/f^ =^

through . ly

^ . J |

from my

wick - ed - nesd,

^ J 1— i

P^=

^p~^-=~z> *^~

P ^

1

29« THE ROMANTIC PERIOD

TREBLE.

And ALTO.

fur - give me all

For ... TENOR (8ve lower).

43- .

give

=p

me all

my sin,

dim.

And . . for give BASS.

me all

my sin,

and

and

for - give

m

m

and for . give me all

my sin,

and for give, . . for give . . me all . . my

i

-aJ I dr f frjr

m

£=*=

for » give . . me all dim.

S&7—&

-r r

and

for give me

all . . my sin, all

^g~f^

1 J-JY

-W 1

=T=°g:

*=E

VARIOUS WORKS BY OTHER MASTERS 299

5fc=p===fs=:

|X-^ —-

wash me

L_ through - ly

from my

wick > ed - ness,

Bin, . .

and P

for ...

give, for- 1 J =P|

sin, . .

\\.-L-II roe

through . ly

| - ^=j

and f or-

mv

^-.

,

al id fur.

~^J J

"^J 1

I

^ f ^

"^i r

tS— * y J

y

i^1^ -^

F?— - J

i ^ i

and fur

give

me all my sin,

. give, for - give

me all my sin, all ...

^

i-P -rj

^BE

^^=

=j

- give, for - give . me all my sin, and for-

J I -J =*fe

^m

give roe

all

mr ain, and for-

^J"" ^

300

THE ROMANTIC PERIOD

&r^

a.l,

all

f

my sin,

all .

my sin.

- give

me all,

all, all

my sin.

In the wake of these masterpieces (longo intervallo) certain contemporary compositions written for the Anglican Service, and for Societies of Glee-singers, merit a passing notice. Thomas Attwood Walmisley's (1814-56) Magnificat and Nunc Dimittis in D minor are sincerely felt; his Thanksgiving Anthem in G minor, If the Lord Himself had not been on our side, shows individuality, is strong at the beginning and the end, but has an unfortunate touch of sentimentality in the middle (G major |). A dainty five-part madrigal, Sweete Floweres, ye were toofaire, also deserves mention. S. Webbe senior's (1740-1816) Discord, dire sister of the slaughtering power, a glee for alto, two tenors, and bass, Y minor |, is concise and powerful with a suave close in F major 1. Sir John

1 Dr. Callcott's (1766-1806) rather saccharine With sighs, sweet rose, I mark the faded form is a homophonous four-part glee in E |?, for alto, two tenors, and bass, E. Spofforth's (1768-1827) five-part glee for two altos, tenor, two basses, Come, bounteous May, and William Horsley's (1774-1858) By Celia's Arbour, a four-part glee, are both well contrived for the voices. Thomas Moore's verse is reproduced

VARIOUS WORKS BY OTHER MASTERS 301

Goss's anthems, If we believe that Jesus died and rose again, and O Saviour of the World, are sincere and beautiful, and the part-writing is masterly. His Ossian's Hymn to the Sun, as well as the anthem Praise the Lord, lacks the personal note. Finally, we may mention J. L. Hatton's lively four-part song, King Witlaf's Drinking-horn, and R. L. de Pearsall's Sir Patrick Spens a ballad-dialogue in ten real parts, as he is careful to note. Pearsall's glee for four voices, When Allan- a -dale went a hunting, is spirited and deservedly popular *.

This would seem to be the right place to call attention to a species of concerted music for solo male voices unaccompanied, which is worthy of note as peculiarly English, intrinsically genuine, and in its peculiar way good. Setting aside the madrigal, it may be said that from about the fourth quarter of the eighteenth century to the end of the first quarter of the nineteenth, the round, the catch, and the glee held the field now occupied by the part-song for mixed voices. From Cromwell's time onwards, and especially before and during the Napoleonic wars, a truculent male element was conspicuous in English society, and was reflected in a manner which needs no further particularization, in the ' tavern-catches ' on which a good deal of current musical invention was employed. The glees are mostly sober, gently bucolic, or sentimental; but with a large proportion of the rounds, and with most catches, there is only one sentiment ergo bibamus. Yet, whether emotional or lively or boisterous, many rounds and catches are distinctly effective both from the musical and from the histrionic point of view 2. The quotation of a catch by Purcell,

in the music, and, as in Moore's, there is a curious air of sensuousness obviously at second hand.

1 Other well-known productions of Pearsall's, such as There is a Paradise on earth, from the German of Hiilty, Lay a garland on my hearse, also an eight-part madrigal, Great God of Love, though nobly sonorous, are rather dry, and the music hardly chimes with the spirit of the words.

a It appears that certain catches were not only sung but acted.

302

THE ROMANTIC PERIOD

though it belongs to the seventeenth century, may perhaps be condoned, since it serves to exhibit the social function of such jeux d'esprit l.

1.

2.

3.

Fie, lf. ,. |?-g-

nay pri thee, John,

=====£==£==

do nut quar-rel, man,

i h r r n N * -v 1^1

You're a rogue, you've cheat-ed me, I'll prove be -fore the

-0 ^ « •£• (• = P r m % •-

com -pa - ny, I

lijgF*: g p ^-fc=5 rr^-r

^=S=^^~

T~rT^\

Sir, you lie, I scorn your word, or a - ny man that wears a sword. For

•H s '

i 1

$-*-* ' " J J.=-

let's be met - ry and

g drink

3^^

a - bout.

caren't a far - thing six for all you

are

_H i J

so stout.

u ' e r*=n

G "* M >— r..-J— J=— ^~

9 i n

all your huff, who cares a fig, or

who

cares for you?

The first voice starts, the second enters after the fourth bar, the third after the eighth bar. Thereafter the first starts again at the point where the second entered, and finally at the point where the third entered. The process consists in giving out, first a solo, then a duet, finally a complete harmonic trio which, with successive changes and histrionic exaggeration, may go on for ever. 'The catch in music/ says Dr. Hayes, the editor of several sets of catches, f answers to the epigram in poetry, where much is expressed within a very small compass, and unless the turn is neat and well pointed it is of little value.' Like the round, the catch is a short Canone infinite in the unison or the octave, and the e epigrammatical touch ' is sought in the connexion of pointed words with the musical sounds.

1 Compare also Purcell's catch in G minor, If all be true that I do think, reprinted in Samuel Webbe junior's Convito Armonico, p. 423.

VARIOUS WORKS BY OTHER MASTERS 303

A glee may be described as a sort of harmonic madrigal for three, four, five, or more male voices i. e. combinations of alto, tenor, bass (with the occasional addition of a boy's treble), depending on the deft interweaving of vocal parts in which the contrapuntal element is usually more or less present1. Points are started and taken up, much as in the madrigal proper; indeed, though less persistently developed, they are more frequent than in the madrigal, and they are generally poetical points. The musical structure is always strictly har- monic (there is no trace of the modes), and the total appeal is rather to the poetical perception of the auditor that is to say, attention is drawn to the drift of the verse as emphasized by means of harmonic music. Of course the specifically musical charm is never entirely absent. Frequently a glee consists of various movements in succession, suggested by the course of a poem; and this again differentiates the glee from the madrigal, in which the same words are repeated to support contrapuntal points. Many a glee, round, or catch is remark- able for good craftsmanship and delicate knowledge of vocal effect. In a number of rounds, catches, glees, something like a democratic note may be felt: each voice seems to represent an individual holding his own, yet harmonically co-operating with the others. This applies, for instance, to a fine work by a recent master of contrapuntal and vocal effect, Sir John Goss's2 five-part glee, Hark! heard ye not that strange tumultuous sound?

1 It is entirely absent from the fine homophonic glees of R. G. S. Stevens. a 1800-80.

3°4

THE ROMANTIC PERIOD

Con moto, ed animato. e3 = 84.

ALTO.

1st TENOR / (Sve lower). |

2nd TENOR i (Sve lower). \

1st BASS.

2nd BASS.

PIANO

(ad lib.)

dfcdtirJS;

G&alS

^zgzzo^as^^

-S-. rw— *• * •*•

Hark! heard ye not that strange, tu-mul-tnous sound?

mf t »

=£=p:

a^E===±=H EL-HE=&±£Eg=p

Hark !

mf

Hark ! hark ! heard ye not that

S

Hark! mf

Hark! hark!

Hark!

mf

Hark ! hark !

Hark!

Con moto, ed animato

Hark! hark!

mf

\

=i

1

Hark ! heard ye not that strange tu - mul - tuous

e^g^^^B -=^^^^^g-^^^

strange tu - mul - tuous sound ?

Hark ! hark ! heard ye not that

Hark ! heard ye not that strange sound ? heard ye not that

Hark ! heard ye not that strange tu - mul tuous

Hark ! hark !

(

I

hark ! hark ! heard ye not that em.

^ ft

VARIOUS WORKS BY OTHER MASTERS 305 #-

sound, that grates, that grates dia . cor - dant on the star - tied

sound, that grates, that grates dia cor - dant on the star - tied

sound, that grates, that grates dis - cor dant on the star tied

f^i I,

sound, that grates, that grates dis - cor - dant on the star tied

'-C~- *-

i - 1

sound, that grates, that grates dia cor - dant on the star tied

i i j 1 * 1 1— i

i

1

J J J J.

'*-*• *— - ^-i-r5

**:-w ^*

g?tvl 1-

I* I*

^f-^-Tf^-^f^-

ear? Too . . faith - ful e cho, too . . faith ful e - cho of a

l5g " - |

^ l^ag_ [• rI9f r «i-H^I

ear?

feg^— -I— ±=£=£

Too faithful e - cho of a

W

* -J -J— jl-

ear?

Too faithful e - cho, too faithful e cho of a

ear?

Too faithful e cho, too faithful e - cho of a

g=g==]

ear?

Too faithful e - cho of a

r

DANNREUTHZR

306

THE ROMANTIC PERIOD

I

ngj 'CS^

jar . ring world, Too . . faith - f ul e - cho, too . . faith ful 43

jar ring world,

Too faithful

jar ring world, Too faithful e - cho, too faithful

jar - ring world, Too faithful e * cho, too faithful

,gp , ftp

Sr^ i -j- i i

e cho of a jar »

ring

world,

e > cho of a jar

ring

..^

world,

e . cho of a jar

^SS> GiJ

ring

^zrr==^^

world,

e - cho of a jar

ring, jar-ring world, Where all the i 1 1

e . cho of a jar

ring

world,

" "P" "D" "**''*'" '^ 5s>" ' ^rJ^2__j..j-.| ,, ^^__J__^^_

| :

l^L-<9 rrj-- ^^^— f=

\=z^>-?-* r f-\

-t ' [T I tJ

VARIOUS WORKS BY OTHER MASTERS 307

/

J- -v

Where all the passions most de » gra

£

ding

Wheie all the pas-aions most de-gra-ding

Where all the

E

j wis-fcions most

de gra - ding, de > gra - ding, -f

^=3=2

.

Where all the pas-sions most de- gra- ding

f f r r f ~- f

i

jLJN^J.

man, Where all the passions most de - gra

ding

man, Where all the pas sions most de - gra - ding

pas-sions most de-gra-ding man,

de - gra

ding

^

J *

Where all the pa**ions niott de- gra. ding

man, Where all . . the . . pas . sions most de - gra

ding

*&=

i±tz:

^FW t i f

f

X 2

jo8 THE ROMANTIC PERIOD

man are

gen der'd, and break forth in

-1-— 1—— 1 \—\ "-E L ' "|

man . , . . are . .

gen - der'd, and break forth in

-f*-- T~ * * ~\ ^ F ~==\

man .... are . .

gen - der'd, and break forth in

r ' f- r r i- n ]

man are

fr*-fr

gen - der'd, and break forth

\ ^ ~—

man,

are gen-der'd and break

u i i i r j _- H j =TJ

r ' t r * r i

T r r P1^ F

in suits,

" 0 J ^ \ broils, loud up - roar,

- r r ^-p^3--^ ==g=^

in - suits,

are gender'd, and break forth in

in ' suits, in

/ TFi r r r r f

suits, broils, . .

are gender'd, and brea

k forth in up-roar, rage, in

forth in 0 h 1

in - suits, in broils, in-

j ~d3~~]-d 1

Qi » ^ j "T "T "T Y" f^

-* r r r ^r r r r f r

VARIOUS WORKS BY OTHER MASTERS 309

rage .... in - f u riate, dire re - venge, loud up - roar, rage, .

up ronr,

dire re - venge, loud up - roar, rage in -

•±±=+—J—l

rage

in - fu riate, dire re venge, loud up - roar, rage,

m

ftf-3

»*=£

*^-

fu

re i" ~i

riate, dire re - venge, loud up roar, rage in

^g

- fu

riate, dire re venge, loud up roar, rage in .

m

3= r4 \ rf.

^m + !*=.

I

•T I? i J I _i r~ -J-4—

~ i -of p-J . -m i «* i - ffl \ "^

dire

re - venge, dire re - venge, dire re venge. &c.

fu riate, dire re - venge, dire re - venge, dire re venge. &c.

-p-r^t

dire

re - venge, dire re - venge, dire re - venge. &c.

f u-riate, dire re - venge, dire re - venge

Ac.

=3fe£

m

fu-riate, dire re - venge,

dire le - venge. Ac.

3 . .1

r-

^a

310 THE ROMANTIC PERIOD

The best glees deal with poetry in a sympathetic, often very human and manly way. The composers are generally mindful of their poets, and their music keeps closely in touch with the language. No doubt, in the case of many a favourite glee e.g. Stevens' The cloud-capped towers or Ye spotted snakes with double tongue the words far outshine the musical expression ; yet it must be conceded that correct declamation, as here represented, is a feature worth having and accentuating, inasmuch as it makes for conciseness and sincerity in musical diction, and leads to effects which chime with the modern feeling for elasticity in point of Tempo. Too frequently in this curious by-way of vocal art an ill-tilled wheat-field, where tares predominate the musical expression is cheap, dull, narrowly provincial, yet it is sometimes novel, ingenious, beautifully contrived for the voices and delightful to sing. The admixtures of the so-called alto voice, i.e. a bass or baritone in falsetto, brings about variety, and if the singing is well done there is sure to be beauty of tone-colour in the ensemble. And though the total effect may be a trifle sensuous, it need not be sickly. Incidentally too, the use of the alto voice which, as an artificial product, is useless unless it is well trained and well managed makes for artistic vocal style 1.

Among the followers of Mendelssohn, Niels Gade and Sterndale Bennett were the most conspicuous. Gade (1817-90), Mendelssohn's successor in the conductorship at Leipzig, subse- quently held a leading position in Copenhagen akin to that of Sterndale Bennett (1816-75) in London. Both men were friends and disciples of Mendelssohn, and in some sense, particularly in their later days, disciples of Schumann as well, yet their style, far from appearing as a mere reflex of the greater masters, exhibits a distinct physiognomy of its own. Gade, who was more of an expert in instrumental

1 In the matter of vocal ensemble, just intonation, and correct phrasing, lovers of English music have always had a good example in the performances of the trained vocalists belonging to the cathedral choirs.

VARIOUS WORKS BY OTHER MASTERS 311

colour than a master of design, in such pieces as his overtures, Reminiscences of Ossian (1841), and In the Highlands, shows a fine vein of imagination, recalling the spirit of Scandinavian folk-song and Northern scenery. His themes, though rarely passionate, are spontaneous, and never without some special grace of colour or sentiment, or a tinge of Norse melancholy. The fascination usually lies in the prevailing sense of beauty and poetical suggestion conveyed by the entire piece, rather than in any prominent feature of tune or harmony or rhythm, and the details are always apt to the particular instrument concerned. His cantatas, Comala and Erlkonigs Tochter, met with well-merited success. He published eight symphonies, two violin concertos, Novelletten for orchestra, and a mass of chamber music, amongst which an octet for strings and a sonata in D minor for piano and pianoforte are the most conspicuous.

With less pathos than the best effusions of Gade, Sterndale Bennett's dainty pieces leave a less definite impression. Yet Bennett (1816-75) accomplished the work of a high, though but a secondary master. His style was distinguished by fluency, as well as fineness and delicacy of fancy. There is grace and natural beauty in much of his work. His attitude towards Schumann's musical poems with suggestive titles was at first one of hesitating assent. Later in life he made some com- promise with programme music as in his last publication, the sonata called The Maid of Orleans, Op. 46 (the themes of which are labelled as they occur, like the themes in Lizst's Ideale], and the ' Phantasie-Overture ' to Moore's Paradise and the Peri, Op. 42. His cantata The May Queen (1858), which suffers from a weak libretto, and his oratorio The Woman of Samaria (1867), fell flat, as did the Symphony in G minor, Op. 43, produced three years earlier the only one of his symphonies that was published. His best overtures are Parisina (1835), Op. 31, The Naiads, Op. 15, and The Wood Nymph ( \ 836), Op. 20. The passage work of Bennett's piano-

THE ROMANTIC PERIOD

forte pieces, like that of Mendelssohn's, is derived to a large extent from the methods of the older harpsichord players, and from Clementi ; but his treatment of the instrument is neither vigorous nor various. The Toccata in C minor, the exquisite sketches entitled The Lake, The Millstream, The Fountain, the Barcarolle belonging to the Concerto in F minor, and the Rondo piacevole are good examples of his individuality and style1. His Chamber Trio for pianoforte, violin, and violoncello, in A, is really good chamber music of a delicately reticent sort.

George Onslow (1784-1852), a composer of English origin, though born and chiefly resident in France, gained an ephemeral reputation with a comic opera, Le Colporteur, and enjoyed a wide popularity in amateur circles, both in France and in England, chiefly by reason of his string quintets with double bass, which have no inconsiderable merit.

Like Liszt, Anton Rubinstein 2, in some respects incompar- able as a pianist, was given to rapid improvisation as a com- poser. But, unlike Liszt, he chose to ignore the labor limae with the unfortunate result that many an ambitious piece, fine in impulse, remained sketchy, flimsy, and diffused. He wrote many operas, of which the least unsuccessful is The Demon, after a Caucasian legend versified by Lermontov; { sacred operas/ i.e. oratorios contrived with a view to stage per- formance; symphonies, concertos, quintets, quartets, trios, sonatas, preludes, etudes, smaller pieces for the pianoforte, and something like two hundred miscellaneous songs for one or two voices. The larger works operas, both secular and sacred, symphonies, &c. were for the most part stillborn, though it is true that some were received with acclamation when the composer personally introduced them. But in a few instances has there been a success other than ephemeral. The Symphony in C entitled Ocean, six movements, the Symphonic dramati-

1 Felix Moscheles' Autobiography : ' It sounds like Mendelssohn, it must be Sterndale Bennett.' 8 1830-94.

VARIOUS WORKS BY OTHER MASTERS 313

que in D minor, the Pianoforte concertos in G major and D minor, the Trio in B b, the 'Cello sonata in D, are the most likely to survive. Certain trifling improvisations, little piano- forte pieces, like the Barcarolles in G major and F minor, and especially a few little songs, have a singularly oriental charm. The best of the songs, Gelb rollt mir zu Fiissen der rauschende Kur, belongs to a set of so-called Persian Lieder1. With a minimum of elaboration, and in spite of two glaring faults in declamation, this Lied is both novel and charming; and, somehow, with its Eastern melismata, haunts the memory. Other such Lieder, though on a lower level, are No. i of the Persian songs, Nicht mit Engeln im blauen Himmelszelt mein Mddchen vergleich ich, and a fine setting of one of Heine's most original poems, Und der Slave sprach : Ich heisse Mahomet, ich bin aus Jemen, und mein Stamm sind jene Asra, welche sterben wenn sie lieben. As in the poem there is considerable originality in the music.

In contrast to the diffuseness of Rubinstein's pianoforte concertos, etudes, &c., we may note the graceful futility of Stephen Heller's pieces (1815-88). Heller, in spite of his conspicuous mannerisms and constant production for the market, was a conscientious worker in the field of solo piano- forte music and free from affectation. His etudes, Op. 16, 47, 46, 45, his preludes, Op. 81, Promenades d'un Solitaire, Op. 78, may be taken to represent his delicate talent. His pianoforte technique is sufficiently effective and refined, though he rarely gets full value out of the instrument.

Arthur Sullivan, apart from his true domain, the operetta, comes into view under other aspects. There was an English note already in his early music to Shakespeare's Tempest (1863), which ranks with his best productions ; also in the duet from The Merchant of Venice, ( How sweet the moonlight sleeps upon this bank,' which forms part of the cantata Kenilworth, 1864; and particularly in the set of six Shakespearean songs

1 Lieder des Mirza Schafiy, edited by Friedrich Bodenstedt.

314 THE ROMANTIC PERIOD

written and published in 1 865, one of which, Orpheus with his lute, with its genuine tunefulness, gained great popularity and deservedly retains it. There is something English too, though not so pronounced, in the effect of the overtures In Memoriam (1868), Di Ballo, Macbeth (1888), and, we may even add, in the so-called Irish Symphony (1866). Of Sullivan's quasi- oratorioSj The Martyr of Antioch, a sacred music drama, has actually been performed on the stage. Other oratorios, The Light of the World (1873), Mendelssohnian in style and arrangement, and The Prodigal Son (1887), which shows a little more of his individuality, do not at the present day count for much. But the sincerely expressed Golden Legend, half cantata, half oratorio, proves as attractive as ever. A collec- tion of the best modern English songs would as certainly ex- clude Sullivan's hypersentimental The Lost Chord, as it would include Hatton's To Anthea, and Sullivan's Orpheus with his Lute.

As equivalents to certain simple songs by Rubinstein and Sullivan, just mentioned, we may point to Luigi Gordigiani's Canti populari Toscani, which, with their artless charm, have found their way to the hearts of cultivated amateurs, as have Rossini's Soirees musicales, Gounod's Quand tu chantes bercee le soir, and certain German sentimentalities by Abt, Kiicken, Lassen, and other Capellmeister. Taubert's na'ive Kinderlieder deserve mention.

Robert Volkmann (1815-83), a German who lived in Hungary, and whose music contains many Hungarian traits, merits con- sideration as the composer of a good pianoforte trio, B b. minor, Op. 5 (1852), and a fine set of solo pianoforte pieces, with characteristic titles after the manner of Schumann, called Visegrad. The trio, Op. 5, attracted the attention of Liszt and was frequently performed by him at Weimar. It is of elegiac import, a sort of forerunner to Tchaikovsky's trio A la memoire d'un grand Artiste, and, technically, fully up to the mark of that interesting work. In Visegrad Volkmann

VARIOUS WORKS BY OTHER MASTERS 315

appears as an illustrator, of power and originality and on con- sistently musical lines of his own. For the most part his pieces are like German translations of the Hungarian idiom pomp and pride, a ponderous sort of grace, and some pathos altogether well set for the instrument. Five string quartets, two symphonies, a violoncello concerto, and two serenades for strings, still belong to the staple concert repertoire in Germany. Volkmann also published two Masses for male voices and other choral music.

Schubert's friend Franz Lachner (1804-90), who made his name with an opera, Catarina Comoro, and a number of orchestral suites, may be bracketed with Friedrich Kiel (1821- 85), a North German composer of considerable attainment who is best remembered as a master of counterpoint. A Requiem Mass, Op. 20, appeared in 1862, a Missa Solennis in 1867; and these, with the two oratorios, Der Stern von Bethlehem (1866) and Christus (1872), and a second Requiem (Op. 80), produced shortly before his death, are the most conspicuous of his works for chorus and orchestra. Good in their way, there is nothing of real importance about them. Kiel also composed a large amount of chamber music three pianoforte quintets, two pianoforte quartets, seven pianoforte trios, four violin sonatas, a viola and a violoncello sonata, two string quartets, and a set of waltzes for strings, of which the same estimate may be offered.

Goetz, as a composer of instrumental music, must be included on account of an overture entitled Fruhlingsouvertiire, Op. 15, and a Symphony in F, Op. 9 (1876), both of which pieces, like the late and latest efforts of Gade and Bennett, were at first well received, but failed to make a definite mark. The one concerted piece by Goetz likely to survive is his Nanie, Op. 10, a poem by Schiller set for chorus, solo voices, and orchestra. Not par- ticularly strong, but emotionally genuine and technically finished, it forms a sort of pendant to Brahms' Gesang der Parzen iiber dem Wasser and the Schicksalslied. Goetz' church music is insignificant.

316 THE ROMANTIC PERIOD

About 1 860 there was a ready sale for Theodor Kirchner's l smaller pianoforte pieces, and for Jensen's 2 pieces and songs with the usual result that both men continued to compose for the market, with increased facility, but, as time went on, with decreasing freshness. This may not be entirely true of Jensen, whose songs show progress, but it is near the mark in both cases.

It was not until after his death that Peter Cornelius' verse and music began to attract attention. He wrote much and published little. Each carefully considered publication, however, represents something in the growth of his talent. His three operas, written to his own librettos, have already been mentioned. It remains to touch upon his remarkable choral music a capetta, and his songs and duets with pianoforte accompaniment, which are also for the most part attempts to set his own verses to music. The songs, some fifty in all, are little more than an array of trifles occasional pieces, akin to certain little lyrics of Goethe each the expression of some particular emotional experience, many among them perfect in their delicately reticent way. Neither as a writer of lyrical verse or of lyrical music does Cornelius aspire very high. But what he has put forth in each department shows perfect sincerity, a rare sense of fitness, and considerable technical attainment. Everywhere one hears the voice of a man who is somebody, a man with a delicate ear for balanced beauty of verbal and musical expression. In a number of instances the verbal expression appears richer than the musical ; but this applies to the solo songs, duets, and the operas more than to the choral pieces, a capella, which latter belong to the best modern work in that department. Here, especially in the unaccompanied choral pieces, Cornelius combines contrapuntal mastery on the traditional lines of canon and fugue, with the chromatic harmony of the later romantic development, in the manner of Berlioz, Liszt, and Wagner. The majority of these secular anthems are therefore difficult to 1 1823-93. a 1837-79.

VARIOUS WORKS BY OTHER MASTERS 317

intone, but when the choristers have mastered the strange intervals the result is good, and at times very impressive, as for instance in the setting of Abbot Notker's (Balbulus) Sequence Media vita in morte sumus (Mitten wir im Leben sind von dem Tod umfangen), which perfectly represents the Neo-German ideal of declamation, or the remarkable setting of Uhland's Die Vatergruft ('The ancestors' tomb') for solo baritone and four- part chorus of mixed voices, which is the most original piece of vocal programme music in existence. Of Cornelius' songs, one at least deserves special mention Ein Ton, Op. 3, No. 3, a curiously speculative and original piece, in which the voice reiterates one note (B natural) whilst the elaborated piano part develops the poetic idea l. Among the duets the most remark- able is No. 3, Op. VI., a strict canon for baritone and soprano, the voice parts a crotchet asunder. This is a setting of the old hymn attributed to Abbot Wernher of Tegernsee, Ich bin dein, du bist mein, dess sollst du gewiss sein.

Icb bin dein,

du but mein,

solUt

eaft^-g1

Ich bin dein,

du bist mein,

dcsa aollst du ge-

1 Compare Chopin's Preludes, Op. 28, No. 15, in D b and Xo. 6, in B minor; Op. 4, No. 2, ' Kouim, wir wandeln zusammen im Mondenschein, So zau- berisch glanzt jedes Blatt, Vielleicht steht auf einem geschrieben, Wie lieb mein Herz dich hat ' the music to these lines, in D t? and in f and f time, is as simple and direct as the words. Masterly too, are Die Hirten, and Die Kdnigin, Nos. 2 and 3 of Weinachtslieder, Op. 8, and Op. 5, No. 5, Der Pelion sprach zum Ossa.

3i8

THE ROMANTIC PERIOD

du ge - wiss sein.

Dich hab' ich ge-sclilos-sen ein . .

i

wiss sein.

Dich hab' ich ge - schlos-gen ein

-~

. tief in meines Her-zens Schrein, ver lo - ren ist das

tief in meines Her-zens Schrein, ver » lo - ren ist das Schlusse

m

--*-

SchlUsse-lein,

nun musst du e - wig drin-nen sein, nun

lein, nun musst du e - wig drin nen sein, nun musst du

VARIOUS WORKS BY OTHER MASTERS 319

wig, nun auf e

wig drin-nen sein.

/ P

•=>• -^ >•;«.

3EEEE£

Hans v. Bulow (1830-94) in his younger days was ambitious to excel as a composer. He wrote pianoforte pieces, songs and duets, a Poeme symphonique, Nirwana, also an overture and incidental music to Shakespeare's Julius Caesar, an Orchestral Ballade, Des Stingers Fluch, after a poem by Uhland, £c. But the vein of his invention proved essentially jejune, and in spite of technical attainment and undeniable musical ability he failed to make a mark as a composer. The cleverest and least dry of his pianoforte pieces are a set of modern dances, entitled // Carnevale di Milano, Ballabili ed Intermezzi, dedicated to an Italian opera dancer, as Op. 21. His editions of classical pianoforte music, particularly those of Beethoven's later Sonatas, beginning with

320 THE ROMANTIC PERIOD

the Waldstein Sonata, Op. 53, and of certain pieces which formed part of his concert repertoire, are very instructive, in spite of the fact that he is prone to indulge his personal whims.

Joachim Raff (1822-82) was the most curiously eclectic among the modern Germans, and, in so far as technical mastery and versatility of production are concerned, the most accomplished. His works exhibit traits belonging to a variety of contempora- ries, from Mendelssohn to Liszt. He composed operas, oratorios, eleven symphonies, many concertos, quintets, quartets, trios, suites, sonatas, a host of solo pianoforte pieces, songs, &c., of very unequal value. Some of them are merely productions for the market, whilst others show artistic aim. Two symphonies still evince signs of vitality, Leonore (1869), and Im Walde (1872). Both have a full programme; thus, Im Walde: (i) Daytime, impressions and emotions ; (2) Twilight, dreams, dance of Dryads ; (3) Night-time, silence and darkness, coming and going of the ' Wild-chase ' with Odin and Venus ; (4) Break of day. In Leonore (after Burger's well-known Ballade) Raff tries to depict the lovers' farewell, the war, the return of the dead lover and the spectre's ride. But in both cases an annoying discrep- ancy between the programme and the exigencies of musical logic impairs the value of the work. The composer, to satisfy his instinct for musical symmetry, finds himself compelled to violate the continuity and progress of his story ; he falls between two incompatible ideals, and his music, in spite of its skilful instrumentation, strikes us as essentially dull and artificial. Next to these symphonies may be ranked his pianoforte quintet in A minor, the trio No. i in E minor, a suite in E flat, Op. aoo, for pianoforte with orchestra, a solo suite for pianoforte in E minor, Op. 72, and a very effective set of variations on a Gigue belonging to the suite in D minor, Op. 91, also for pianoforte.

In this place mention may fitly perhaps be made of a remark- able set of quasi-amateurs who called themselves the ( Five Neo-

VARIOUS WORKS BY OTHER MASTERS 321

Russian innovators/ a coterie held together by friendly rivalry and patriotic ambition. It consisted of four men of uncommon talent, Balakirev the leader, Cui, Musorgsky, Rimsky-Korsakov, and a man of genius, Alexander Borodine, who was a son of a Prince of Imeretia in the Caucasus. Their cry was ( Russian music for the Russians.' Following the example of Glinka and Dargomijsky, they studied ecclesiastical melodies, folk-songs and dances, investigated the various oriental elements which are intermingled with Russian art, and strove for novelty in melody and harmonization, and for piquancy of orchestral effect. In the Liturgy of the Greek Church, together with the semi-oriental songs and dances of the peasantry, they found a vast amount of material that strikes Western ears with a strange sense of power and spontaneity. Balakirev (1836), a man of keen intelligence and an accomplished musician, collected Russian folk-songs, composed orchestral pieces in the manner of Berlioz and Liszt, and pianoforte pieces in a manner of his own of which the oriental fantasia Islamey (variations on an Eastern tune) is the most ingenious. Rimsky-Korsakov, born in 1844, has written some twelve operas, many songs, a pianoforte concerto in Eb, two Poemes symphoniques, Antar and Sheherazade, and published two valuable collections of folk-songs taken down from oral tradition *. Cui, born in 1835, a truculent critic, produced eight operas, over 160 songs, and a large number of small pianoforte pieces many of which are but Schumann at second hand. Musorgsky (1839-81), the 'most Russian of the Russians,' in his vocal efforts appears wilfully eccentric. His style impresses the Western ear as barbarously ugly. Alexander Borodine (1834-87) composed two symphonies (two movements of a posthumous third have been published), an orchestral sketch Eine Steppenskizze aus Mittel-Asien, two string quartets, a Petite Suite in D minor for pianoforte, twelve songs, some to

1 Korsakov is fond of reproducing the peculiar metrical structure of certain Russian folk tunes, f, J and the like, and so gets surprising effects of rhythm and colour. He is a master of orchestration.

DASXBEUTHEK

THE ROMANTIC PERIOD

his own words ; and left unfinished one of the most original productions of modern times, Prince Igor, an opera on a Russian subject, which, after his death, was completed by Rimsky- Korsakov and Glasounov. The picturesque oriental elements upon which Balakirev laid so much stress, the use of chromatics, augmented seconds, and sequences of whole tones, frequent changes of rhythm and surprising modulations, abound in Borodine's works. His instrumental pieces, almost throughout, are programme music sincerely felt and expressed, and without a trace of affectation. The thematic material is novel, the workmanship careful, the sense of variety and beauty in instru- mental colour striking. Dans les steppes de I'Asie centrale forms a good pendant to Berlioz/ Marche des Pelerins 1. The second symphony in B minor, the two movements of a third, the Asiatic sketch just mentioned, the second quartet in D major, and the Petite Suite, have permanent value. For completeness' sake the name of Dargomijsky (1813-69) must be included. A disciple of Glinka, he wrote several operas Roussalka (f Water nymph') has been most frequently given about 100 songs, a number of which are noteworthy by reason of a curious alternation of passion with oriental languor, and a grotesque ( Cossack-dance ' (Kosatshok) for orchestra, that made the round of European concert-rooms, and is indeed very clever and characteristic.

Borodine is, however, the national genius after Glinka. So far as concerns sheer novelty in the scenes depicted and originality in the musical material and treatment, his Prince Igor, an opera in four acts with an overture and a prologue, published in 1889,

1 The programme is as follows : ' Dans le silence des steppes sablonneuses de I'Asie centrale retentit le premier refrain d'une chanson paisible russe. On entend aussi les sons me"lancoliques des chants de 1' Orient ; on entend le pas des chevaux et dee chameaux qui s'approchent. Une caravane, escortee par des soldats russes, traverse 1'immense desert, continue son long voyage sans crainte, s'abandonnant avec confidence h, la garde de la force guerriere russe. La caravane s'avance toujours. Les chants des Russes et ceux des indigenes se confondent dans la mSme harmonic, leurs refrains se font entendre longtemps dans le desert et finissent par se perdre dans le lointain.'

VARIOUS WORKS BY OTHER MASTERS 323

equals, perhaps surpasses, Bi/et's Carmen. It is the strangest and the strongest production of the Neo-Russian school. Borodine, better than any other composer, represents the poetry of the manners and the sights and the sounds of the north and south-east. He does not consciously strive to produce charac- teristic Eastern music ; with him it is true, spontaneous, and irresistible. Throughout his work Russian local colour is supreme so much so that Rubinstein's and even Tchaikovsky's oriental tints fade before its vivid rays. Borodine's invention never flags he is ever ready with something new, strange, and appropriate. In Prince Igor there are popular Russo-Asiatic motives of surprising delicacy and charm, as for instance in the dances, the songs, and the choruses for female voices. Certain virile movements again, such as the Introduction and the close of Act I, the Warriors' dance at the end of Act II, or the march at the beginning of Act III, are pieces of barbaric splen- dour, which, for all their colours, are as compact and logical in construction as they are subtle and penetrating in style. Better materials for a Russian opera could hardly have been found J, yet Prince Igor is perhaps more of an epic pantomime than an opera proper. It consists of a series of scenes, choruses, ensembles, dances, songs for the most part of startling origin- ality, so vivid indeed that once heard they persistently linger in the memory. A professional librettist might denounce the book as inorganic it is not weak or inept, but sketchy rather, and unfinished. It has only a semblance of a plot and is pervaded by a curious duality : two khans, two Vladimirs, two ladies in love, two fools, two outrages on the princely dignity, two captive princes, two victorious armies. It should be added that the composer's command of musical form and diction, of the treatment of solo and choral voices and the orchestra, is that of a bold, highly accomplished master one who never rouses the suspicion that he may have intended one thing and by lucky

1 Of course the work must be heard in Russian ; for reading purposes the French translation may be called fairly good, the German is clumsy.

Y 2

3*4

THE ROMANTIC PERIOD

chance achieved another 1. For delicacy and strange charm it would be hard to surpass the following twenty-eight bars of Kontchakovna's Cavatine (Prince Igor, No. 9) :

Largo = 48.

Kontchakovna.

Choeur des - jeunes fillea.

Soprani (14-16).

Alti (12-14).

=i—

4

Ah !

ces sons lea chants,

fflg^jy-f^s

m

-&?•

ar - re - tons noa jeux.

Le jour

1 And this astonishing man was an army surgeon who for a quarter of a century acted as a professor of chemistry. His scientific publications aroused the attention of his colleagues.

VARIOUS WORKS BY OTHER MASTERS 325

f

nieurt !

Kt sur

—a i

- •&•

£=£:

de la mi it sont le« voi-lea.

a jnaetrr

^S

fff at

zr*—:-

N > N * J <

^o. tempo

m

Oh 1

des

cenda des

cienx blcus,

f>p con Fed. una. conia,

^

326

THE ROMANTIC PERIOD

Dans

r

man-teau

r

d'6 - toi

les,

•-=)-

6 ... nuit, 6

r

bel

nuit : Le

doux

re . ve

1 j b[jj_ -

>• I l :

1 Le

doux

r6 - ve

Doux

!*• "S

-**i -. i i ... j

[J. J. J23JJ.

Z=^i|==

VARIOUS WORKS BY OTHER MASTERS 327

r9ltT." f-

clian

suit, Bel

. tent

unit, bel

merit, bel

unit, bel

Poro piu ani

suit.

nuit !

nuit !

Pocopiu animato

&C.

&c.

^^5=g

So exquisite a sample of the exotic element in artistic music has not been seen since Chopin's Mazurka, Op. 1 7, No. 4, and the Trio of his C minor Polonaise *. The idiom is not altogether

1 One of the most extraordinary attempts at musical humour (musicians take their humour seriously at St. Petersburg) is a pianoforte composition for four hands entitled, ' Paraphrases sur le theme favori et oblige", d&liees aux petits

338 THE ROMANTIC PERIOD

new : indeed it has been in some measure anticipated both by Dargomijsky and by Glinka, but no master has ever employed it to such admirable effect.

The total work of the Russian innovators is a distinct, though very late outcome of that development of the historical and critical sense which has been already discussed 1. On the basis of the Russian language, coloured by ecclesiastical chants, most of which are modal, by peasants' songs and dances, by funeral laments and festive tunes, together with the quaint wail of Hebrew and other Semitic melismata, something has of late years been achieved in Russia that exhales 'le rude et viril parfum de la terre slave V and offers one of the most important contributions to the history of romantic music. The composers, like their precursors and masters, Schumann, Berlioz, Liszt, looked at music through the lenses of literature. Hence their pronounced tendencies in favour of a programme, whether it be avowed or not. Hence also their leaning towards the further or nearer East which, apart from oriental barbarisms deliber- ately chosen for barbaric ends, as in Musorgsky, has hitherto made for good. Kept within proper artistic bounds, the Russian movement now in full course may ultimately lead to illustrative instrumental music of the highest beauty arid value. There is no need that the young Russian composers should hark back to partially exhausted formulas. The laws of musical design, the principles of good sense and proportion will make themselves

pianistescapables d'ex&uter le theme avec un doigt de chaque main ! ' The theme is that known in England as the ' Chopsticks Waltz,' and is played continuously by one of the two performers, while the other accompanies it with galops, mazurkas, requiems, fugues on the name of Bach, and similar incongruities. Started by Borodine this curious jeu d'esprit was a joint production by himself, Cui, Liadow, Rimsky-Korsakow and, later on, by Liszt and N. Stcherbacheff (who well replaces Musorgsky). Some of these very speculative variations rival those of Schumann's Carneval in point of beauty, and for variety and subtlety of invention may be said here and there to surpass them. A similar piece of still more serious fooling is the Quatuor sur le nom de B-la-f (Belaieff the generous publisher's name), but this concoction, though enormously clever, is really too serious.

1 See ch. i. pp. 3-5.

3 Alfred Bruneau, Musiques de Russie et Musiciens de France, 1903.

VARIOUS WORKS BY OTHER MASTERS 329

felt, whatever be the material to which they are applied. With Tchaikovsky (1840-93) Russian music became cosmopolitan. His pianoforte concerto The Russian, in B i? minor, has already been mentioned. The second and third concertos in G and E b, and the Fantaisie de Concert in G, Op. 56, fell far below that high mark. Among his six symphonies, the fifth and sixth are the most important, and the latter of them (known as the Pathetique} is the most famous. Next to these symphonies, or rather beside them, we may place the Poemes symphoniques, Romeo et Juliette and Francesco da Rimini. The plan of the latter pieces, it has been said by Mr. Ernest Newman, * fulfils very happily one of the main requirements of good programme music that the various points shall not be not only dramatic but musical, lending themselves naturally to musical treatment at the same time that they speak connectedly to the intellectual ear '.

Among Tchaikovsky's achievements may further be included an overture entitled 1812, another overture, Hamlet, the Poeme symphonique, Manfred, a symphonic Ballade, Der Wojewode (after Mickiewicz), three string quartets, and a pianoforte trio, A la memoire d'un grand artiste (Nicholas Rubinstein). His lesser works of unequal merit suites for stringed instruments, numer- ous small pianoforte pieces, 117 songs, many set to inferior verse J have all, more or less, a fascination of their own. It seems to be the rule with the Slavs, * that the power of creating intrinsic interest is considerable, but that the faculties which are needed for concentration and systematic mastery of balance of design are proportionately weak % and this applies to Tchai- kovsky in almost the same degree as to all the rest. His operas, Eugene Onegin and La Dame de Pique excepted, met with little success outside Russia they contain much graceful and at times interesting and original music but the composer approached the theatre as a novice, and his stage instincts, if

1 And generally sung to perfunctory German translations.

2 C. Hubert H. Parry, Summon/ of Musical History, p. 89.

330 THE ROMANTIC PERIOD

he ever had any worth cultivating, remained in an undeveloped condition.

Edvard Grieg, born 1843, a Romantic of the Romantics, had the good fortune to light upon topics of a fresh and fascinating nature, in the folk-songs and dances of his native Norway ; and he had the instinct to treat them adequately, without disguis- ing or unduly accentuating their characteristic features. His Norwegische Volksweisen, Op. 17 and 66, contain all the germs of his music : the tunes, plaintive or crude, as the people sing and play them the drone bass (which is implied), the chromatic inner parts, which he supplies, the use of some quaint fragment of the tune by way of introduction or coda, the studied com- pactness and concentration, the sudden and unexpected contrasts. That he should have been able at all to weld these tiny phrases, and fuse them so as to serve for the thematic material of pieces in large form, such as his pianoforte concerto, already mentioned, his string quartet, and the sonatas for violin and for violoncello, speaks highly for his genius. We owe to Grieg a number of the most beautiful of modern songs, and a host of charming lyrical pieces for pianoforte solo. Mention must also be made of Max Bruch, born in 1838, a master of choral as well as instrumental effect, and the writer of some very effective violin concertos and of Felix Draeseke, born in 1 835, an accomplished theorist and critic, and a gifted composer whose music is full of original and romantic ideas.

Before we pass on to the consideration of organ music, mention must be made of the little-known pianoforte pieces, chiefly Etudes, by C. V. Alkan, who died in 1880, and of the clever transcriptions of certain movements from Beethoven's string quartets (in the manner of Liszt's partitions de piano) by Tausig and Saint-Saens. Alkan's fitudes the work of a specu- lative and eccentric rather than an essentially musical talent are technically magnificent in so far as the treatment of the instrument is concerned ; the inventiveness in virtuosity is very considerable, though musically, that is to say, meledically and

VARIOUS WORKS BY OTHER MASTERS 331

harmonically considered, they are somewhat barren. Alkan's most important Opus is marked 39. It is made up of twelve impressionist Etudes of inordinate length : I, is entitled, ' Comme le vent/ II, 'En rhythme molossique,' III, ( Scherzo diabolico ' ; Nos. IV to VII are meant for a symphony ; Nos. VIII to X for a concerto ; No. XI for an overture. The twelfth Etude, called Le Festin d'Esope, in E minor, is a veritable tour de force it consists of a set of curiously characteristic variations on a theme of eight bars akin to that of No. 6 of Liszt's f Paganini Caprices ' in A minor, and of Brahms' ' Paganini Variations ' in the same key remarkable for an almost farcical humour and for ingenuity of contrapuntal device. If well played the total effect of this grotesque piece is astonishing from the virtuoso's point of view and almost, if not entirely, satisfactory from the musician's. There is no actual indication of a plot, but the comical effects tell their own tale. Other numbers worth attention are Le Chemin de fer, Op. 27, Trois Etudes pour les deux mains separees et reunies, Op. 76, and twelve Etudes,

Op-351-

The organ music of the nineteenth century owed much of its impetus to Mendelssohn, who infused new life into the forms of prelude and fugue. His so-called organ sonatas do not essen- tially belong to the sonata order, having little about them of its typical character or its principles of design ; still they rank among his best works and occupy an important place in the literature of the organ. Schumann's six fugues on the name B-A-C-H have already been mentioned ; Liszt's ambitious fantasia on Ad nos, ad salutarem undam, published in connexion with the 1 Illustrations ' to Meyerbeer's Le Prophete, and his B-A-C-H Fugue must also be mentioned. Finally there remains a mass of organ music by Joseph Rheinberger which consists of two concertos with orchestra, twenty-two trios, twelve ' Meditations,' twenty solo sonatas, &c. Rheinberger's operas and his numerous

1 Huit prieres pour orgw ou piano a darter de pedales, Op. 64, have been admirably transcribed for pianoforte solo by Jost? Vianna da Motta.

332 THE ROMANTIC PERIOD

symphonic and choral works have almost entirely disappeared. An early work, a deft and fresh pianoforte quartet in E \), Op. 38, had some vogue, and is still welcomed in amateur circles.

Few of the elder Wesley's works for the organ are in print. The best of those contained in Vincent Novello's collection of Select Organ Pieces consist of a f Slow Air in D/ f , a Fugue in D, the transcription of a choral fugue for four voices, Sicut eratf in C, a Voluntary and Fugue in B b, and a Fugue in C minor on a partially chromatic subject. Of these, the last two, especially the Fugue in C minor, show considerable skill and originality.

Of Sebastian Wesley's works for the organ, fourteen numbers have been edited by his pupil, the late Dr. Garrett, ' for modern Pedal organ/ They consist of single pieces elegiac cantabile voluntaries, andante, or grave and andante, and produce their impression by persistence of mood and without any particular con- trapuntal subtlety, somewhat after the manner of Spohr, though generally freer in treatment, broader in melody, and less cloying in harmonization. There are a few instances of incongruous pianoforte technique, as for instance the Andante in F (No. 5), and the fourth and fifth variations on the National Anthem (No. 10) ; but apart from these the workmanship is sound and musicianly, with bold sweeping melodic outlines and a strong and characteristic handling of the bass. Among the finest numbers may be cited the opening Andante in C (No. i), the Introduction and Fugue in C $ minor (No. 9), and the masterly settings of Psalm-tunes (Nos. 1 1 to 14) which close the volume.

CHAPTER XIV

THE ROMANTIC OPERAS OF WAGNER AND THE INCIPIENCY OF THE MUSIC-DRAMA

So far as the musical stage is concerned Wagner : sums up and completes the ideas and aspirations of Romanticism. He expresses them in Tannhauser and Lohengrin, transcends them in Tristan, departs from them in Der Ring and Die Meister singer, and returns to them in Parsifal. The gradual transformation of the opera seria, semi-seria, or buffa into the current modern equivalents, the development of German ( Singspiel ' into ' Ro- mantische Oper ' and finally into the music-drama, both alike mark a change in the relative position of the two operatic factors to which nothing in artistic history supplies an exact parallel. By degrees the play asserts its full rights, operatic conventions recede, and the music becomes pliant ; until at last dramatic illusion is attained by means of a compromise between the imitative arts on the one hand and music on the other.

Dramatic poet by instinct, by training supreme master of musical effect, Wagner was gradually led towards a new manner of blending music with the drama. Eight early operas or musical plays conceived between 1833 and 1848 saw the light in pairs, with an interval of about five years between each group of two Die Feen and Das Liebesverbot, Rienzi and the Flying Dutchman, Tannhauser and Lohengrin. Siegfrieds Tod and Friedrich Rothbart did not go beyond the stage of elaborated sketches for a musical drama (Siegfried), and a spoken tragedy

1 Richard Wagner was born in 1813 thirteen years before Weber's death and fourteen years before the death of Beethoven. He died in 1883.

334 THE ROMANTIC PERIOD

(Rothbart). In 1848-9 Wagner carried a number of dramatic sketches with him, sketches for the tragedy Rothbart, for Die Franzosen vor Nizza and Wieland der Schmied, the operas, Jesus von Nazareth, the music-drama, as well as for other less- matured dramatic ideas, such as Achilleus. The development of his dramatic and musical capabilities took place with logical consistency, even when the process was quite instinctive, from one positive experience to another. There is nothing like this in the history of any musician, and it can be explained only by the extraordinary energy of Wagner's character, which kept him isolated from the world and wholly surrounded by the atmosphere of his own deeds and aspirations.

In the libretto to the Hollander Wagner begins to pay attention to poetic qualities regardless of operatic considerations. He had begun, he tells us, by trying to acquire the faculty of musical expression in the way in which one learns a language. A man speaking in a foreign tongue over which he has not yet acquired complete control must consider its peculiarities in every sentence that he utters ; if he wishes to be understood he must always be thinking of the expression, and this will influence him in the choice of what he shall say. Wagner, however, was an apt pupil, and could soon declare : f By this time I had finished learning the language of music. I am now able to use it like my own mother tongue/

Again, Wagner maintains that legendary subjects are to be preferred to historical ones, inasmuch as the substance of a legendary story is so readily intelligible that there remains plenty of space for the full expression of the inner motives of the action. For instance, the story of Der fliegende Hollander is set forth in the simplest way possible details resembling the intrigue of every-day life are excluded, whilst stress is laid on those aspects which serve to accentuate the expression of emotion. In Tannhduser the action springs mainly from the inner motives of the characters, and even the final catastrophe is essentially lyrical. In Lohengrin the interest is concentrated

THE ROMANTIC OPERAS OF WAGNER 335

on a psychological process in the heart of Elsa. Thus the lyric spirit pervades the whole, and the total effect depends upon close connexion of the play with the music each factor being modi- fied in turn by the other.

Questions of aim and method arise here. If in an opera close and direct expression is desired, the use of formal musical design seems to stand in the way, for during the process of musical exposition the action is apt to be retarded, whereas it would seem to be a necessity that the music moves simultaneously with the action ; and the difficulty from beginning to end consists in the proper adjustment of speed, the give and take between the motions of each collaborator. Every true melodic subject has its inner law of growth and expansion, and this musicians are loth to infringe for the sake of histrionic effect; on the other hand, operatic music must be true to the situation. It is, there- ! fore, the principal convention in the Wagnerian drama that J musical sounds may be accepted as symbolical. Music for the theatre must be regarded from a standpoint other than that of chamber or concert music ; for as soon as dramatic presentation and stage effect are elements in the artistic whole, the appeal is not exclusively to the auditor's sense of musical balance and proportion, but it is also addressed to other forms of consciousness. It follows that the standard of absolute self-contained formal music cannot be fairly applied. Dramatic music is meant to arouse, stimulate, or exhaust emotion it does not aim at delight in purely musical expression in just balance of statement and restatement. And because it illustrates or emphasizes, or fully expresses something more or less extraneous, it ought to be frankly accepted and judged as a kind of rhetoric.

To suppose that Wagner ever was guided by some abstract theory would be entirely erroneous. With him theory and practice advanced together, or rather his artistic instincts led the way and his theoretical opinions acted as support and rearguard. With his divine discontent and self-sufficing strength he, the great learner, was ever striving after something fresh and new.

336 THE ROMANTIC PERIOD

Every work marks a step in the development of his genius, and the distance traversed from the first romantic opera, Die Feen (1833), to the last music-drama, Parsifal (1883), is perhaps greater than the distance ever before covered by any great artist. Wagner's individuality was first revealed in Eine Faust- Ouvertiire (1839-40) (pp. 106-10 ante), then in Der fliegende Hol- lander (i 841 ), and so onwards. The three operas of earlier date, Die Feen, Das Liebesverbot, and Rienzi, do not demand close examination, though the third, Rienzi, der letzte der Tribunen, is of vast dimensions a grand tragic opera in five acts in the manner of Spontini, and with sundry traces of Meyerbeer. All three resemble the types of opera which prevailed in their time, and were it not for their authorship, the first two, at least, might rest in oblivion. Das Liebesverbot was withdrawn from the stage after two performances : Die Feen, a weightier and more important work, was never heard until after the master's death. His first operatic victory was won with Rienzi, which contains some noble passages, such as the Introduction to Act IV, and particularly the Introduction and the Prayer in the fifth act. Its remarkable success at Dresden in 1 842 was fully justified.

It is curious to note that these early pieces possess the stamp of theatrical rather than of musical originality. The grip of the dramatist is unmistakable ; there is a keen instinct for general effect, there is frequent evidence of a practical acquaintance with the stage, but the musical details, both in the action and in the orchestra, are often raw and blatant. One point, however, stands forth conspicuously : Wagner always succeeds in his fusion of dramatic and musical elements, and invariably contrives to get the result that he wants. And this gift remained with him throughout his wonderful career. As he approaches maturity the technique of the musician and the power of the dramatist is everywhere seen to expand with the complexity, the subtlety, and the intensity of his aims ; but from the first he approaches his hearers on every side and excites them with the cumulative appeal of all arts in combination.

THE ROMANTIC OPERAS OF WAGNER 337

The libretto of Die Feen is an arrangement of Gozzi's La donna serpente. Das Liebesverbot is based on Shakespeare's Measure for Measure. Bulwer's novel, Rienzi, the last of the Tribunes, suggested the characters and the plot of the third opera. The poem of Der fliegende Hollander it must not be called a libretto is derived from Heine's account1 of the Ahasuerus of the ocean. The materials for Tannhduser and for Lohengrin were collected from the wide field of German mediaeval ballads and epic poems, and from certain modern romantic stories by Tieck and Hoffmann 2. > Now and then the music of Die Feen is reminiscent of Weber and Marschner, as Das Liebesverbot contains echoes of Auber and Bellini. In Die Feen the composer's sudden change of aim and of style comes as a surprise. It is the only one of his works planned in two acts, and the only one that is tainted with what has been called ' an open championship of the rights of the senses/ In this matter it is difficult to criticize ; but as, throughout the opera, the music is, on the whole, the predomi- nating factor, its effects may be judged from a musical point of view ; and in that respect they appear just as little deserving of censure as anything of Auber or Bellini. As is the case in some of Marschner's less important operas, a certain lack of melodic distinction is noticeable in Die Feen the musical phrases are effective and by no means weak or commonplace, yet they might be signed with a name other than Wagner's. The pianoforte score of Die Feen was published in 1888 3; but' of Das Liebes-

1 Memoiren des Serrn von Schnabelewopski in Heine's Salon.

* Tieck's rhymed Erziihlung ' Tannhauser/ and Hoffmann's novel ' Der Kampf der Sangcr* (Serapionsbruder, ii. i).

3 The following pieces will be found fairly representative : Act I. Ouverture and Ballet, in E major; a characteristic Tenor Aria in C minor; Quartet in B b. Act II. Introduction and Chorus of Warriors, D minor (powerful and very effective) ; a touching Aria in F minor; a capital comic duet in C for Soprano and Bass ; a fine Scena and Aria for Soprano in D, and a grand Finale. Act III. Terzett in C, and Finale in E minor and major, remarkable for its use of the trombones. Extensive and good use is made of Ritornellos before and after the principal arias and ensembles.

*• DASNREUTHER Z

338 THE ROMANTIC PERIOD

verbot only one complete number and certain slight fragments have hitherto appeared in print l.

The imposing spectacular and musical pomp of Spontini's Olympic and of Meyerbeer's Huguenots is at least equalled in Wagner's Rienzi. The subject, first attracted him by the superb opportunities that it offers for the display of operatic pageantry on a grand scale, and by the presence of certain lyrical elements, such as the chorus of the Messengers of Peace, the Battle Hymns, the Church's call, and the Excommunication. When Wagner wrote Rienzi, grand opera loomed large before him ; and it was the object of his ambition not merely to produce a copy, but to outvie the original on its own ground and in its every detail. Yet already in Rienzi stress is far more consistently laid on the drama than in the case of any contemporary grand opera.

Derfliegende Hollander was originally meant to be performed in one act, as a long dramatic Ballade, and not as a conglomerate of operatic pieces. Reference to the score will show that the division into three acts is made by means of crude cuts, and of new starts equally crude2. The music grew out of Senta's Ballade in the second act, which Ballade, as it were, forms the musical nucleus and contains the principal thematic germs (symbolical Leitmotive) which permeate the entire work. Far more distinctly than in Rienzi (1839), we may recognize in the Hollander (1841) the true incipiency of the music-drama. In the poem of The Flying Dutchman Wagner treats the legendary subject on its own merits, with the total effect in view, and with little regard to any operatic scheme of recitative, aria, and ensemble, though, to a considerable extent, their forms and even their cadenzas are still present. * There are moments when the music rises to an extraordinary pitch of vivid picturesqueness and expressiveness. The whole of the overture is as masterly a

1 They consist of a vivacious carnival song in D, f , specimen bars of which, together with two other short quotations, may be seen in Mr. Wm. Ashton Ellis' English version of Glasenapp's Life, vol. i. pp. 184-5.

3 This has been set to rights at Bayreuth.

THE ROMANTIC OPERAS OF WAGNER 339

musical expression of omens and the wild hurly-burly of the elements as possible, and carries out Gluck's conception of an overture completely ; Senta's ballad is one of the most charac- teristic things of its kind in existence, and hits the mood of the situation in a way that only a man born with high dramatic faculty could achieve ; and the duet between Senta and the Hollander is as full of life, and as fine in respect of the exact expression of the moods, of the situation, and as broad in melody, as could well be desired V The instrumentation of the entire score was twice retouched in 1846 and in 1852 and the close of the overture completely rewritten.

Originally the legend of Tannhauser and the Hill of Venus, and that of the contest of the Minnesingers at the Wartburg, were not connected. The fusing and welding of these materials is Wagner's own. Tannhauser has undergone more change and transformation than any other of Wagner's productions. We can but touch upon a few salient points. At the close of the third act, both action and music were altered (1845-7) with the intention of making things clear to the sensuous perception of the audience in lieu of an appeal to their imagination ; and the entire scene in the Venusberg, Act I, was completely transformed for performance at Paris in 1861. Wagner immediately realized the difficulty of adapting French verse to the prevailing square rhythms of the German music, and he seems to have felt no hesitation in making extensive changes to triple time, both in the scene in the Venusberg and the Ballet that frames it. In the course of revision both the Ballet and the scene came to be expanded to more than double their original dimensions. And, together with the great expansion, there came an equally great change of style a change so great that one cannot help deploring the interval of fully sixteen years which intervenes between the old Tannhauser and the new (1845-61). The new music was composed to French rhymed verse (by M. Nuitter), and all that remained of the old was

1 C. H. H. Parry, The Art of Music, p. 350. Z 2

340 THE ROMANTIC PERIOD

carefully revised so as to meet the exigencies of French accentuation *.

The Lohengrin legend tells of a knight from oversea, who reached the banks of the Scheldt in a skiff drawn by a swan. There he fought for a noble maid and was wedded to her, but when she asked whence he came and desired to know his name he was forthwith obliged to depart. Wagner takes up this legend one of the many mythical stories with a religious colouring that cluster round the traditions of the Holy Grail at the point of its contact with History in the first half of the tenth century. He develops the historical aspect side by side with the supernatural, and thus contrives to present an unrivalled picture of Teutonic mediaeval manners and belief. Lohengrin presents the ideals of the later Middle Ages so completely that, for emotional essentials, it would seem idle to go back to documents, and we may add that this is the last of his pieces which Wagner called a Romantische Oper. The copiousness of resource displayed, the power and variety of dramatic and musical detail, are astounding. The whole work is a single organism, the soul of music clad in a body of dramatic action.

It may be stated here that for complete comprehension of Wagner's intentions in Hollander, Tannhduser, Lohengrin, and especially in the later music-dramas, it ought always to be borne in mind that on the stage, the power to declaim and put dramatic meaning into the delivery is in the forefront, and the singer's task is little more than that of assisting and intensifying the expression of emotion. In other words, the hearer's atten- tion is meant to be drawn and directed more to the dramatic whole than to the musical details 2. If this be understood, it

1 The Parisian version is of course adopted at Bayreuth as ' the sole authentic one' in spite of the fact that translation back into German has inevitably brought about certain discrepancies between text and music. On the whole, however, the new Tannhdustr is superb and perfectly convincing. Wagner dropped the sub-title Romantische Oper and called the new version Handlung, i. e. action.

a But every singer ought to be able to sing a fact overlooked by the majority of people who are allowed to take part in Wagnerian performances.

THE ROMANTIC OPERAS OF WAGNER 341

will be readily conceded that since the poetical subject is every- where amenable to the governance of music, the latter, no matter how complex, need not be cast in the mould of conven- tional operatic forms, the declamation need not spoil the vocal melody, and the melody, vocal or orchestral, need not interfere with the progress of the action. Thus, without consciously striving to deepen the musical expression, Wagner, in accordance with the peculiar nature of his subjects and of certain histrionic details connected with their due presentation, did in point of fact develop a new melodic idiom ; and so step by step, particularly with the inception of the music-drama, considerably enlarged the scope and power of his music.

Lohengrin, already, shows great concentration in the scenic arrangements. Its precursor, Weber's Euryanthe, was laid out in three acts, with two changes of scene in each. Lohengrin also has three acts, but each has only one set scene an immense gain in the direction of perspicuity and sustained interest. The choruses, in their prodigious variety, from mere ejaculatory utterance to the most expansive lyrical effusion, are very important factors in the development of the dramatic action. To take but one instance, the beautiful chorus in eight parts which precedes and accompanies that miracle of scenic effect, Lohengrin's arrival in the first act, is perhaps the finest example extant of a dramatic chorus springing directly from and entirely belonging to the plot. The instrumentation of Lohengrin exhibits the highest instinct for beauty of tone. *" To any one who hag neither seen nor heard Wagner's scores, neither studied their consummate workmanship nor felt their scenic power, it is not so easy to convey a notion of his extraordinary doubling of the great symphonist with the great dramatist. . . . The orchestra is divided into three main constituent bodies, with subsidiary groups of three. This ternary system has the advantage, among other things, that the whole chord can be given and held in the same scale of colour. . . . Wagner also makes frequent use of the distribution of the strings into separate bodies. In a

342 THE ROMANTIC PERIOD

word, instead of treating the orchestra as an almost homogeneous mass, he parts it into tributary streams and brooks ; at times to change the metaphor he spins it out to the finest parti- coloured threads, and casts their spools first here then there, now weaving them together, now dividing, until their wondrous ravelling has formed a tissue of priceless lace V

We have already mentioned the occasional touches of Weber, Marschner, Auber, and Bellini which are apparent in Wagner's earliest works. As he comes nearer to maturity Italian and French melody predominates Rienzi, and even ten years later the Finale to the first act of Lohengrin, recall Spontini. In the Hollander the melody leans either towards the tersely rhythmical folk-song (e. g. the Ballade, the spinning chorus, and the sailors' choruses) or the broad cantilena in which emotion is paramount. In Tannhauser, and still more in Lohengrin, the melodic ebb and flow is regulated by the action, which in turn is enforced by characteristic harmony and instrumentation. Finally, in the music-dramas Tristan und Isolde, Der Ring des Nibelungen, Die Meistersinger, Parsifal, the vocal melody often springs from the words ; it is frequently independent of the orchestra, in some cases, indeed, it is but an intensified version of the actual sounds of the German language, and it becomes lyrical only when the situation demands lyrical ardour.

In the hands of the dramatist, music possesses an inestimable advantage in its capacity to convey the mood of an entire scene or act at once and in an unmistakable manner. A few bars suffice to indicate a mood, and, once established, the expression of such a mood can be sustained for as long a period as may be desirable. A series of scenes or an entire act can be so laid out as to be governed by one or more musical moods, each at will developed, focused, and brought to a climax. The greatest scenic contrasts may thus be risked without fear of failure such as the sudden transformation in Tannhauser from the lurid light of the Venusberg to sunshine and open air,

1 Liszt, Tannhauser et Lohengrin a Weimar, 1850.

THE ROMANTIC OPERAS OF WAGNER 343

the reappearance of Venus after Tannhauser's recital of his pilgrimage to Rome, the appearance of Elsa on the balcony after the scene of the conspiracy in the second act of Lohengrin, or the happy contrast between the two sections of the third act of Die Meistersinger; while, for similar examples of homogeneous development, we may take any of the three acts of Tristan und Isolde, the first or third act of Die Walkiire, or the first act of Siegfried. In all such cases music makes for simplicity in dramatic construction, whilst it furnishes the fullest and deepest expression. Witness the opening of the second and third act of Tristan, and the third act of Die Meister singer.

Every medal has its reverse. Wagner, whose work at the dawn of the twentieth century is acclaimed with indiscriminate admiration all the world over, was, i a the third quarter of the nineteenth, the best abused man in Europe. Violent and rancorous attacks upon him found admission into the columns of German, French, and English journals. Leading musical and theatrical critics were bitterly hostile. Musicians, the veteran Spohr excepted, stood aside in the difficult position of Moliere's Bridoison : ' Ne sachant pas trop que dire pour exprimer sa fa9on de penser.' Playwrights, actors, singers, put forward the most inept professional comments. At best poets were ready to admit Wagner's musical attainments, composers had no objection to his dabbling in poetry, whilst sober-minded people among the laity felt uneasy and held aloof. Thus during the greater part of his lifetime Wagner was placed in an anomalous position; that of an idealist, a passionate poet, confronted with the journalists, the miscellaneous public, the host of professionals connected with the opera and the operetta. It requires a long period of cure to eradicate from the body of art the poison of a bad tradition. In his own words, ' It was like having to walk against the wind with sand and grit and foul odours blowing in one's face/ But time has brought its revenge. The present generation of professional musicians is making the most minute study of Wagner's scores, both from

344 THE ROMANTIC PERIOD

the dramatic and the musical point of view; public perform- ances of his work are still on the increase, and are steadily improving in quality ; while if we put aside sundry attempts to find f hidden meaning ' in the dramas, it may be said that even the futilities of an overgrown Wagner literature1 appear to have their use, inasmuch as they frequently arouse and stimulate enthusiasm.

About the time of the composition of Lohengrin Wagner's mind was agitated by the question whether he ought to continue as dramatist or musician or both. As has already been said, he tried historical subjects, Friedrich Rothbart and Jesus von Nazareth the latter a tentative effort in the direction of Parsifal, for which a vast number of notes were taken and elaborate sketches made 2. Both of these subjects were conceived as spoken plays. Ultimately his musical instincts gained the upper hand, and he came to the conclusion that, in his own case at least, perfect emotional expression was possible only when the idea occurs simultaneously to the poet and the musician. Accordingly he discarded Barbarossa and Jesus von Nazareth, and went on with the story of Siegfried's death which ultimately grew into Der Ring des Nibelungen. So by degrees he approached the music-drama. Before it was reached, however, an immense amount of mental fermentation was at work as may be traced in the mass of theoretical writing which he put forth between 1849 and 1852. His great problem, ' the problem of the art-work of the future ' as he called it somewhat like the social problem of Comte was to inquire, first how the scattered elements of modern existence generally, and of modern art in particular, could be united so as to form an adequate expression of the whole; and secondly, what hope of a reaction in favour of higher forms of life, than our present industrialism, would the creation and acceptance of

1 ' Lorsque celui qui parle ne comprend pas et celui a qui Ton parle ne comprend non plus, alors c'est de la mdtaphysique ' (Memoirts de Voltaire, p. 151). s They have now been published.

I

THE ROMANTIC OPERAS OF WAGNER 345

such a work of art hold out ? His views of artistic possibilities being thus ultimately connected with those of social regenera- tion— art reform with social reform he might well venture to take a plunge edans I'improvisation risquee des theories sublimes/ if only to clear his own mind of doubts and cobwebs l.

There is many an utterance in Wagner's writings of 1849- 53, which appears but as a comment upon certain experiments in the execution of Tannhauser and Lohengrin. His writings at that important period of transition are little more than a forcible reaction against obstacles in the way of his impulse to produce he admits 2 some obscurity, some want of definiteness in the use of philosophical categories he rightly calls it confusion. In quite early days, and even later up to the end, his writings represent the extreme sensitiveness of the modern man occasional lassitude alternating with crudely vigorous effort. In consequence his utterances are at times fanatical in tone, at times needlessly protracted. If we take his prose works as a whole, and appraise them with regard to style, we must admit that Nietzsche's words are final. ' These products of Wagner's genius excite, produce unrest ; there is an irregularity of rhythm in them, which makes them, as prose, confusing. The discourse is frequently broken ; a sort of aversion on the part of the writer lies like a shadow over them, as if the artist were ashamed of conceptual demonstration. What perhaps most offends those who are not quite at home in them, is an expression of authoritative dignity, which is quite peculiar to them, and difficult to describe. It seems as if Wagner often felt he was talking before enemies for all those writings are in a talking, not a writing style, and they will be found to be such when they are read aloud before enemies to whom he refuses familiarity, and

1 Letter to Uhlig, May 1853. ' Nur insofern kann ich mit einiger Befriedigung auf meine in den letzten Jahren gespielte Litteratenrolle xuriickblicken, als ich f iihle, daes ich mir selbst dabei vollkommen klar geworden bin.'

8 Introduction to vols. iii and iv of his collected writings.

346 THE ROMANTIC PERIOD

for this reason he shows himself reserved and supercilious. But not unfrequently the violent passion of his feelings breaks through the assumed impassibility; then the heavy artificial periods, loaded with qualifying words, disappear, and sentences and whole pages escape him which are amongst the most beautiful that German prose possesses V

The main object Wagner had in view was, as he put it, ' to reconcile the claims of poetry and music with the claims of that most contestable, most equivocal institution of our day, the opera.' Or in other words, and broadly stated, it was his aim to reform the opera from Beethoven's point of vantage. Can the modern spirit produce a theatre that shall stand in relation to modern culture as the theatre of Athens stood to the culture of Greece ? This is the complex problem that he set himself to solve. Whether he touches upon minor points connected with it ; speaks of the performance of a play or an opera ; proposes measures of reform in the organization of existing theatres ; discusses the growth of operatic music up to Mozart and Weber, or of instrumental music up to Beethoven ; treats of the efforts of Schiller and Goethe to discover an ideal form for their dramatic poems ; whether he sweeps round the problem in wide circles ; comparing modern social and religious institu- tions with ancient, and seeking free breathing space for his aspirations, he arrives by either method at the same ultimate result his final answer is in the affirmative. Starting from the vantage ground of symphonic music, he asserts that we may hope to rise to the level of Greek tragedy ; our theatre can be made to embody the modern ideal of life. From the opera at its best a drama can be evolved that shall be capable of express- ing the complex relations of modern life and thought. In the first of his speculative, semi-prophetic books, Die Kunst und die Revolution (1849), he points to the theatre of Aeschylus and Sophocles, searches for the causes of its decline, and finds them identical with the causes that led to the decline of the ancient

1 G. A. Right's translation.

THE ROMANTIC OPERAS OF WAGNER 347

state itself. An attempt is then made to discover the principles of a new social organization that might bring about a condition of things in which proper relations between art and public life might be expected to revive. These and similar ideas are further developed in Das Kunstwerk der Zukunft, which followed in 1850 a book which, despite its difficulty, is well worth attention. The main argument is as follows: Poetry, imitation, and music were united in the drama of the Greeks ; the drama disappeared with the downfall of the Athenian State; the union of the arts was dissolved, each had an existence of its own, and at times sank to the level of a mere pastime. Attempts made, during and since the Renaissance, to reunite the arts, have been more or less abortive, though most of them have made some advance in technique or in width of range. In our day each ' separate branch of art ' has reached its limits of growth, and cannot overstep its limits without incurring the risk of be- coming incomprehensible, fantastic, absurd. At this point each art demands to be joined to a sister art poetry to music, imitation to both ; each will be ready to forgo its special pretensions for the sake of an * artistic whole,' and the musical drama may become for future generations what the drama of Greece was to the Greeks.

Wagner's next work, Oper und Drama (by far his largest critical and theoretical treatise), contains little of this revolution- ary and philosophical ferment. It is set forth in three divisions, of which the first contains an historical criticism of the opera, the second consists of a survey of the spoken drama, and the third is an attempt to unite the results obtained and so construct a theory of the musical drama. In the opera, Wagner asserts, the means of expression (music) have been taken for the sole aim and end, while the true aim (the drama) has been neglected for the sake of particular musical forms. Mythical subjects are best, and Beethoven's music indicates the ideal language in which they are to find expression.

These and other assertions of Wagner's tending in the same direction have already been discussed. One further point, how-

348 THE ROMANTIC PERIOD

ever, requires elucidation his use of alliterative verse in Der Rinff, and of a combination of alliteration, assonance, and rhyme in Tristan and Parsifal. Poets of the Middle Ages, to attain regularity of rhythm, constructed their verses according to some semi-melodious chant or fixed melody ; the great variety of Greek metres arose under mimetic influence, springing from the pantomimic action of a dance combined with the choral song. German poets have imitated, as well as their language permits, every possible metre, but no one can deny that the complex rhythms, upon which they pride themselves, exist far more for the eye than for the ear. Take the most common form of verse in modern German plays accentual iambics is it not torture to hear the sense of the language forced and twisted to suit the purpose of this metre ? Sensible actors, when iambics first came into use, were afraid of sing-song, and recited the lines as prose 1.

French poets, who do not base their rhythms upon recurrences of stress, and who measure their verse by the number of syllables that it contains, believe rhyme to be indispensable. Now if we examine the relation of music to verse, we find the curious fact that musicians declaim iambics, and indeed every species of verse, in every sort of time ; as for the rhyme at the end of a line, music usually engulfs it ! and the cases wherein the musical rhyme actually corresponds to the rhyme in the verse are for the most part accidental or at any rate few and far between. A musician can do more with iambics than the actors did : he must treat them as prose and stretch or compress them to fit his melody. Seeing that modern versification offers such small attraction, Wagner was led to ask himself what sort of rhythmical speech it might be that would best admit of musical diction, and the answer was not far to seek. When we speak under the

1 ' Talma, in remarking to me that a French actor has difficulties to surmount which an English has not, began with pointing out the necessity he lies under of breaking the joints and claws of every verse, as of pigeons for a pie, and of pronouncing it as if it were none at all ; thus undoing what the writer had taken the greater part of his pains to accomplish' (Landor, Imaginary Conversations, The AbU DeliUe and Walter Landor).

THE ROMANTIC OPERAS OF WAGNER 349

pressure of some strong emotion, we drop conventional phrase- ology ; we enforce accents with a raised voice ; our words become strongly rhythmical. In the early days of the Teutonic languages, such a manner of speech was in use for artistic purposes ; it is the alliterative verse of the Eddas and of Beowulf. The condensed form and the close relative position of the accented vowels in alliterative verse give to it an emotional intensity, which renders it peculiarly musical: while, in like manner, assonance and rhyme can be contrived so as to suit the musician's requirements. The verse, then, with Wagner, is conceived and executed in the spirit of musical sound, and there is neither place nor scope for subtleties of diction ; music can supply all that is needed. Firm and concise, abounding in strong accents, the alliterative lines of his verse, notably in Der Ring, seem to demand music; indeed musical emphasis and prolongation of sound render them more readily intelligible and more impressive.

The entire music-drama is musical in spirit and in detail. The mythical subject, chosen because of its essentially emotional nature, the division into scenes, and their sequence; verse, declamation, the orchestra, preparing, supporting, commenting, enforcing, recalling ; all these factors are imbued with the spirit of music. The pathos of dramatic speech is positively fixed by the musician's technique, and their interrelation is a direct appeal through the senses to the emotions.

Artists connected with the opera scene painters and stage managers, dancers, choristers, actors, the members of theorchestra and the Capellmeisters, have been roundly scolded by Wagner for this or that reason, but all owe a debt of gratitude to him. He has made their task more difficult, but infinitely more interesting. Even apart from the stage, at every good perform- ance of music on a large scale, Wagner's spirit is present. The leading conductors, whether they care to acknowledge the fact or not, are under his spell: and who can name a composer (Brahms perhaps excepted) who has not to some extent felt his weight and in some measure submitted to his influence ?

CHAPTER XV

MUSICIANS AS WRITERS ON MUSIC

IT remains to trace the Romantic masters' efforts in criticism and to mention some of the good work which they have done in musical philology, historical research, the editing of classics and the like. The most significant feature in the mental activity of nineteenth-century musicians is the fact that the spirit of J. S. Bach has become a living influence. The Romantic element in Bach, already pointed out in Chapter I, found response in the mind and heart of Schumann, Mendelssohn, Chopin, Brahms, and Wagner. Bach's earnestness and consistency became the ideal of all serious-minded composers, and his contrapuntal technique gave a fresh impulse to poly- phonic treatment in choral and orchestral composition. The study of his works, with their solidity, their variety, and their elasticity of form, acted as a steadying and staying power, and it may be hoped that it will ultimately serve as an antidote to the incoherence and laxness of structure which came as the attendant disease of programme music.

E. T. A. Hoffmann, a writer of imagination and poetical insight, was the first to recognize Beethoven's genius. His reviews of Beethoven's fifth and sixth Symphonies, of the Trios, Op. 70, the Fantasia, Op. 80, together with his so-called Phantasiestiicke Ritter Gliick and Don Juan, all of which belong to. the early decades of the nineteenth century, are still worth reading. 'When we speak of self-dependent music,' says Hoffmann in an article on Beethoven *, * do we not intend

1 Kreisleriana, 4.

MUSICIANS AS WRITERS ON MUSIC 351

instrumental music exclusively? Is not instrumental music the most romantic of arts, the one truly romantic art ? Is not the infinite its sole object ? There are secrets which only sounds can reveal, and under whose weight words break down.' His estimate of the C minor symphony is broad, sympathetic, and tersely expressed : ( Critics have often complained of a lack of unity in Shakespeare, and failed to realize that a fine tree, with its leaves, blossoms and fruit, may spring from a single seed. So they might fail to comprehend the clearness of vision, the high seriousness and complete self-possession l which mark the genius of Beethoven and stamp his art/ Valuable remarks on music occur incidentally in Hoffmann's tales ; and it is well known that several of the tales and many a stray aphorism made a strong impression on both Schumann and Wagner. For instance ' A fantastic description of a piece of music is admissible only in so far as it is understood to be metaphorical/ Certain titles adopted by Schumann are borrowed from Hoff- mann's works : Nachtstiicke, Kreisleriana, Phantasiestiicke ; Hoffmann's Serapionsbriider and Schumann's Davidsbiindler are closely akin. Wagner's indebtedness to Hoffmann's story Der Krieg der Sdnger has already been pointed out j and, in like manner, the influence of Hoffmann's Meister Martin der Kiifer is perceptible in Die Meistersinger von Nurnberg. Moreover, Hoffmann's opinions on the subject of Poetry and Opera, as set forth at length in a brilliant article entitled ' Der Dichter und der Komponist 2,' are the immediate precursors of Wagner's article ' Ueber das Operndichten und Komponiren.' The very language strikes one as Wagnerian : ' Ja, in jenem fernen Reiche, das uns oft in seltsamen Ahnungen umfangt da sind Dichter und Musiker die innigst verwandten Glieder einer Kirche : denn das Geheimniss des Worts und des Tons ist ein und dasselbe, das ihnen die hochste Weihe erschlossen.' And again : * Eine wahrhafte Oper scheint mir nur die zu sein, in

1 ' Die hohe Besonnenheit.' ' Serapionsbruder, i.

35* THE ROMANTIC PERIOD

welcher die Musik unmittelbar aus der Dichtung als nothwen- diges Erzeugniss derselben entspringt V

Weber did not aim at literature when he wrote his e Drama- tisch-musicalische Notizen ' little articles on new operas. His object was to attract attention and induce sympathy with the works which he was about to conduct for the first time at Prague or Dresden. Marschner (Heinrich IV. und IfAubigne), Meyerbeer (Abimelek), Hoffmann (Undine), Fesca, and other composers of operas, profited by his generous advocacy. Exam- ples of Weber's technical strictures have already been given in the extracts from the review of Hoffmann's Undine 2. That they are always apt and to the point goes without saying ; yet there is little of enduring value in Weber's literary remains, except in his last publication a small pamphlet that accompanied his directions as to the tempi in Euryanthe : ( On Tempo in music and on its metronomic indications/ This is a little masterpiece, a landmark in the history of style. Weber's views regarding ' modification of tempo ' are exactly those of Wagner, as set forth in the tatter's essay on Conducting. There is another curious and highly significant point of agreement between the two masters : ' I look upon any one,' says Weber, ' who per- forms a piece from one of my operas at a concert as my personal enemy' (Letter from London, 1826).

Schumann was proprietor, editor, and chief writer in Die neue Zeitschrift fur Musik from its foundation in 1834 to 1853, when he took leave of his readers with the prophetic article on Johannes Brahms. Shortly before his death he revised a number of the essays, reviews, and reports, and in 1852 republished them in four small volumes. The Zeitschrift was issued to a few hundred subscribers. But in book-form, since

1 ' Yes, in that far-off land which we often reach in dreams, poet and musician are closely related members of one church : and in their highest moods the secret of word and tone is revealed to them as an identity.' And again : ' It appears to me a true opera is one in which the music emanates from the poem as an inevit- able product.'

' See p. 25.

MUSICIANS AS WRITERS ON MUSIC 353

about 1860, Schumann's writings have been widely read and have had great influence. It is the rare union of literary gifts with the insight of a composer of genius that renders Schumann's writings unique. At once enthusiastic and humorous, bold in imagery, and whimsical in phrase, the style is redolent of Jean Paul Richter and E. T. A. Hoffmann not involved like Jean Paul's, however, but made up of short vivid sentences, always fresh and breezy. Like his forerunners, the brothers Schlegel, Tieck, Jean Paul, Hoffmann, Schumann protested against all kinds of pedantry and formalism ; like them, he was ever ready to hail anything that showed a touch of individuality and of genuine human nature. His weightier reviews, such as those of Berlioz' Symphonic fantastique, Meyerbeer's Huguenots, Men- delssohn's choral works and overtures, Liszt's etudes, and a large number of Chopin's pieces, are of permanent value. A note of perfect sincerity pervades them, and the style has a rare charm, even when it deals with mere technicalities. In some of the earliest articles (1834-6) Jean Paul and Hoffmann are closely imitated. Later on, the quaint ironical devices, disguises, and noms de guerre are dropped, and Schumann writes like a serious artist addressing his equals. Everywhere, even when he laughs or plays tricks, his earnestness is felt to be present. He was a perfect hater of shams. And though he dealt gently and kindly with all manner of ephemeral productions he never lost sight of a high standard of excellence. The essay on Brahms, with which he closed his career as a journalist, is written with the same care as the essay on Chopin with which he began it1.

Mendelssohn's letters, many of which were apparently written to be read outside the family circle, demand some notice. They are full of facts and precepts valuable to practical musicians and to students of contemporary musical history; they show a delicate gift for reproducing impressions received from nature

1 A good English translation of his works would be a boon. Madame Raymond- Bitter's version is incomplete and miserably inadequate.

DANKRECTHXB A a

354 THE ROMANTIC PERIOD

and art, and are expressed in a tone of genial good humour. Mendelssohn always expresses himself with the crispness and precision of a man who thoroughly knows what he is talking about.

There is no such thing as a Wagnerian system of aesthetics. Das Kunstwerk der Zukunft and Oper und Drama excepted, Wagner's writings are occasional pieces without any calculated continuity. The earliest Romantic pieces, i. e. the Parisian feuilletons1, recall the manner of E. T. A. Hoffmann. Then came the books which reflect the revolutionary ferment of 1848-52 and show a leaning towards Hellenism ; next follow the mature pieces written at Munich and Triebschen (i 864-70), Ueber Staat und Religion, Deutsche Kunst und deutsche Politik, Ueber das Dirigiren (1866), and Beethoven (1870). These are the result of insight, wide culture, and wide experience. Finally we may mention the occasional contributions to the Bayreuther Blatter, such as Ueber das Dichten und Komponiren, Ueber die Anwendung der Musik auf das Drama, Wollen wir hoffen ?, in which the master talks leisurely and confidentially to his friends. They are singularly charming and instructive.

Wagner's criticism is always valuable, even if it is but a side- light or indirect comment on his own practice. Now and then his judgement seems somewhat strained and beside the mark as when he maintained that in Beethoven's pth Symphony instrumental music has burst its confines and said the last word possible, when he denounced the oratorio as a feeble hybrid without proper raison d'etre, or when he refused to credit Jews with creative ability. But his width of view and his absolute sincerity are everywhere apparent. Next to the valuable essay on Conducting a treatise on style in the execution of classical music we must rank Beethoven, an exposition of the author's thoughts on the significance of Beethoven's music. This work contains his contributions towards the metaphysic of music, if indeed such a metaphysic can be said to exist. It is based on 1 See W. A. Ellis' translation of the prose works, vol. viii.

MUSICIANS AS WRITERS ON MUSIC 355

Schopenhauer's famous theory ; which that philosopher candidly admitted to be incapable of proof, though it satisfied him. Wagner accepts it, and supplements it, by way of analogy, with quotations from Schopenhauer's Essay on Visions, the doctrine of which is at least equally problematic l.

The history of music in Paris from 1835 to 1863 might be traced in the feuilletons which Berlioz wrote for the Journal des Debats. Though he was a journalist of genius and well aware of the fact, Berlioz all along protested against his weekly task, and eloquently complained of it as downright slavery 2. The celebrated Mtmoires are a brilliant plaidoyer, but not a record of fact. 'Ma vie est un roman qui m'interesse beaucoup,' he says in his letters 3. Indeed his own standard seems to have been Dtsordre et Gtnie in literature as in music. The fantastic child of a fantastic time, he never passed beyond the period of storm and stress belonging to his youth and early manhood. His critical remarks, often penetrating, do not spring from a consistent principle, but strike or retort as the occasion suggests. The famous attack on Wagner (Journal des Dtbats, Feb. 8, 1860) and Wagner's dignified reply (Feb. 22) only serve to accentuate the fact that Berlioz could not be just to Wagner without disavowing part of his own work. Comparatively few articles can be classed as literature. The enthusiasm expressed in the more elaborate the essays on Spontini, on Beethoven's symphonies and sonatas, on Gluck's Alceste and Orphfa, Weber's Oberon and Freischiitz was doubtless genuine ; but now and then, notably in the essay on Spontini, it seems as though Berlioz was writing de parti pris with more fervour than penetration. Many of the smaller pieces, brilliant fireworks for

1 See the writer's translation of Ueber das Dirigiren, 1887, and Beethoven with a supplement from Schopenhauer, London, 1 880.

8 ' II faut pourtant m'obstiner a 4crire pour gagner mes miserable* cent francs et garder ma position arm^e centre tantde drdles qui m'aneantiraient s'ils n'avaient pas tant peur. La violence que je me fais pour loner certains ourrages, e«t telle que la ve"rit£ suinte k travers mes lignes, comme dans les efforts extraordi- naires de la presse hydraulique 1'eau suinte & travers le f er de 1'instrument.'

3 Lettres intimes, p. 127.

A a 2

356 THE ROMANTIC PERIOD

the most part biting, satirical, ironical were issued in book form as Les Soirees d'orchestre, A travers chants, and Les Grotesques de la musique *. A number of filoges de complaisance arrachees a sa lassitude were not reprinted, but the rancorous attack on Wagner et la musique de I'avenir was included in A travers chants. A series of letters describing his tours in Germany and visits to England and Russia are incorporated in the Memoires. Taken altogether there is not much beyond amusement to be gained from Berlioz' f euilletons. He had little to teach 2 ; and his teaching was too often phrased in terms of contempt. The two following sentences will convey some idea of his position and his attitude : ( La musique pure est un art libre, grand et fort par lui-meme.J f Les theatres lyriques sont des maisons de commerce, ou cet art est seulement tolere, et contraint d'aiUeurs a des associations dont la fierte a trop souvent lieu de se revolter/ In point of style the demon romantique occasionally led him to bizarre exaggeration, and to a confusion between the grandiose and the great. He seemed to consider Beethoven as older Berlioz. But many pages are full of wit and charm ; and this is particularly the case with a large portion of the Memoires. There is something musical about the tempo, the rhythm, and the cadence of Berlioz' best sentences. Like his own music his prose is always rhetorical, sometimes eloquent, sometimes violent, sometimes even grotesque. His humour occasionally degenerates into buffoonery, his wit too often takes the form of parody or sarcasm. One only of his literary efforts is really a landmark in the History of Music: the Grand trait e d' instrumentation, Op. 10, with its sequel Le Chef d'orchestre, which, taken all round, is an exhaustive, and in the full sense of the word, masterly work. Thus, Berlioz on Instrumentation, Weber on Tempo, and Wagner on Conducting, are the three practical treatises (classics in their way) that

1 A volume of his miscellaneous articles has recently been collected under the title of Musique et Musiciens.

2 ' Esthe'tique ! Je voudrais bien voir fusilier le cuistre qui a invent^ ce mot-la ! '

MUSICIANS AS WRITERS ON MUSIC 357

represent the refined sense of style and instrumental colour prevalent in the nineteenth century.

Liszt habitually wrote in French1. His enthusiastic admiration for Wagner is recorded in the best of his literary works, a short pamphlet entitled Lohengrin et Tannhausef a Weimar. It made a great stir and was very helpful in the furtherance of Wagner's aims. Next to this masterpiece of sympathetic criticism, we may rank a delightful little essay on John Field, written to serve as an introduction to an edition of Field's nocturnes. The more ambitious efforts, Frederic Chopin, and Les Bohemiens et leur muisique enHongrie improvised contributions to the Gazette musicale were rewritten with the collaboration of Princess Sayn-Wittgenstein, and spoilt in the process. Both contain much irrelevant detail couched in hyperbolical language. Les Bohemiens, in book form, was published together with the revised edition of the Rhapsodies hongroises. In the main it consists of a laboured attempt to prove the existence of some- thing like a gipsy epic in terms of music an attempt which was met with ridicule in Hungary itself the fact being that Hungarian gipsies merely play Hungarian popular tunes in a fantastic and exciting manner peculiar to themselves, but have no music that can properly be called their own. Liszt's book on Chopin contains much that is delicately appreciative and valuable as a record at first hand of Chopin's methods as a player and composer ; unfortunately, it also contains many mis- statements of fact, and a good deal of verbiage. Other pieces of interest are the articles on Robert Franz* songs, on Wagner's Fliegende Hollander, and on Berlioz' Symphony, Harold en Italic. German critics, not without good reason, have spoken of the correspondence between Liszt and Wagner as worthy to rank with that between Goethe and Schiller. These letters, Wagner's especially, are full of interesting passages on problems of music and literature. Liszt's literary efforts, whatever their short-

1 The complete collection of his writings is to be found in a German version by Lina Kamanii, six volumes.

358 THE ROMANTIC PERIOD

comings, proved a stimulating force, the effect of which is still felt.

The manner of Hoffmann's imaginative criticism was success- fully imitated by Ambros, the historian of music, whose Cultur- historische Bilder aus dem Musikleben der Gegenwart, a col- lection of admirable essays, 1860-5, stands forth conspicuously among the doings of the lesser men. Ambros was well equipped as a musician and gifted with some of Hoffmann's insight and felicity of speech, which he further qualified with Jean Paul's fantastic imagery and verbal wit. Even in comparison with Schumann his articles hold their own, and in point of detail he occasionally surpasses his master. Billow's analysis of Wagner's Eine Faust-Ouvertiire and Draeseke's articles on Liszt's 'poemes symphoniques ' deserve mention as models in their way J.

With advancing years the historical tendency gained in strength and widened in scope. The success of the Bach revival prepared the way for editions of the works of Palestrina, Lasso, Purcell, Handel, Gluck, Mozart, Beethoven, Schubert, and also of a number of Bach's immediate precursors, such as Schiitz, Sweelink, Frescobaldi, Frohberger, Buxtehude, Reinken, Kerl, George and Theophilus Muffat, Fux, Couperin, Rameau, and both Alessandro and Domenico Scarlatti.

Antiquarian research brought to light the splendid collection of early English music for the Virginals known as the Fitzwilliam collection, Denys Gaultier's La Rhetorique des dieux, and a vast number of miscellaneous Italian, Spanish, and German pieces for the lute. Extensive collections of folk-songs and dances were made; and good historical and biographical work was done.

Among the leading books of antiquarian research, history, and biography may be mentioned :-Kiesewetter's Geschichte der euro- paisch-abendlandischen oder unsrer heutigen Musik (1834-46);

1 Ferdinand Killer's lucubrations were characterized by Wagner as ' Literatur ' meaning waste paper. And the same holds good of Riehl's once celebrated CharacterWpfe, 1853, and La Mara's Musikalische Studienkopfe.

MUSICIANS AS WRITERS ON MUSIC 359

Coussemaker*s Lea Harmonies des XII6 et XIII" sticks, and L'Art harmonique aux XII8 et XIII6 siecles (1865) ; Ambros, Geschichte der Musik, of which vols. ii and iii are the most valuable (the third volume of Ambros' History, extending down to Palestrina, appeared 1868, and the work was completed, in rather perfunctory manner, by W. Langhans) ; Ritter's Geschichte des Orgelspiels (1884); Weitzmann's Geschichte des Clavier spiels (I863)1; Wasielewski's Die Vtoline und ihreMeister (1869). The standard biographies are PohPs Haydn, Jahn's Mozart, Thayer*s Beethoven (to be read in Belter's much augmented German edition); F. W. Jahns' C. M. von Weber in seinen Werken (a thematic catalogue, chronologically arranged, the comments on which form the most trustworthy treatise on Weber's works), 1871; Chrysander's Handel, Spitta's /. S. Bach, and Glasenapp's Life of Wagner as rewritten by Mr. W. Ashton Ellis. Among technical treatises Helmholtz' Die Lehre von den Tonempfin- dungen, and Riemann's researches into the nature of rhythm, and his ingenious solutions of difficult rhythmical problems, are important. In Russia much energy has been devoted to the history and theory of ecclesiastical music. Dimitri Rasumovsky author of Der Kirchengesang in Russland, Die patriarchali- schen Stinger, Diakone und Unterdiakone (1868), and Unter- suchungen iiber die Lesung der Znamja-Notation (1884) began to lecture at the Conservatoire of Moscow on the music of the Orthodox Church in 1866. Yurij v. Arnold published Die alten Kirchenmodi, historisch und akustisch entwickelt, 1879 ; Theorie des altrussischen Kirchen- und Volksgesanges, 1881 ; Die Harmonisierung des altrussischen Kirchengesangs, 1886. Smolensky, Rasumovsky's successor as lecturer at Moscow, wrote a Kursus des kirchlichen Chorgesangs (1887), and ABC- Buch des Gesanges nach der Znamja-Notation des Alexander Minez (1868), which is said to be an important work with regard to the history of the melodies belonging to the Russian Church.

1 See third edition, edited and enlarged by Max Seiffert and Oskar Fleischer, 1899. An admirable piece of work in course of publication.

360 THE ROMANTIC PERIOD

Instrumental music does not form part of the Greek orthodox service, which is exclusively choral. And in this connexion the ten volumes of sacred music a capella by Dimitri Bortniansky, which have been edited by Tchaikovsky, deserve mention. Bortniansky (1779-1828) was a pupil of Galuppi, and to his vapid Italianisms, quite as much as to the Italian opera, may be traced many of those curious southern idioms which so often and so incongruously occur in the melody of later Russian masters.

CHAPTER XVI

SUMMARY AND CONCLUSION

A BRIEF summary will be sufficient to recall to the reader's mind the salient points in this survey of the Romantic move- ment. In Weber's time musicians came under the spell of Romantic literature and learnt to look at their art from the Romantic standpoint. Increased facility of international inter- course was a powerful source of change. Thus it has come about that at the present day exotic rhythms, harmonics, and even melodies are found to be admissible and sometimes welcome elements in the musical speech of Western Europe. Indeed, if a lexicon of musical diction were compiled, it would have to include many curious rhythms as well as melodic and harmonic deviations from the normal language. In opera as well as in instrumental music poetical suggestion by musical means became one of the chief aims. In music for the orchestra and the pianoforte characteristic titles, mottoes, superscriptions, were employed. Gradually ' poetical intentions ' took the lead ; and composers began to accept relaxations of the laws of structure. In symphonic music design on purely musical lines was gradually set aside to make room for a kind of impressionism, wherein unity was sought not so much in well-balanced musical develop- ment as in extraneous considerations, such as the sequence of ideas in a poem, the incidents in a story, or the variety of colours in a landscape. After a time illustration became the ideal and symphonic music was transformed into programme music. At first the aim seemed to be freedom in matters of form only. But with this freedom the door was opened to sheer eccentricity and ugliness of theme. Yet in the long run the common sense

352 THE ROMANTIC PERIOD

of musical art showed that these extravagances were mere incidents which did not prevent the attainment of more pliant and varied forms, together with increased power and beauty of emotional expression.

The entire process, in instrumental music, was a change from the formal to the characteristic a movement away from the precise symmetry of the sonata and the symphony and tending towards the Characterstiick and the Po&me symphonique. In dramatic music, it was a movement away from the conven- tions of the older opera and a tendency towards the freedom of the music-drama. Expressive consistency, at times of a very subtle sort, was retained and depended in songs, chansons, Lieder ohne Worte, nocturnes, and other short characteristic pieces. A taste for excessive emotionalism was developed, and composers took pains to attain the clearest articulation of details. Everywhere in the short lyrics of the time there is the charm of novelty, 'the magic touch of Romanticism, the addition of strangeness to beauty/ Technically the principle of tonality was expanded, and new departures in key distribution and an increasing use of chromatic harmonies and complicated discords became prevalent, especially in Chopin, Berlioz, and Wagner. Liszt, in particular, endeavoured to find new cadences to serve for the close of his pieces.

Weber's Freischiitz marks the triumph of early Romanticism, his Euryanthe the transition to the continuous music of later times. Characterization, subtle devices of instrumentation, and local colour, played an important part in his work. In France the storm and stress of Romantic literature found an echo in the opera, and in Berlioz* orchestral pieces. Italy began with sentimental cantilena and ended with a marked increase in dramatic effect. The tendency towards closeness of characteriza- tion affected the oratorio and the cantata, which were rejuve- nated, and somewhat secularized in the process. Instrumental music in the concert-overture and the symphony endeavoured to reproduce moods and impressions derived from literature

SUMMARY AND CONCLUSION 363

or from natural phenomena. The beginnings of illustration proper, apart from the ostensible writers of programme music, appeared in Mendelssohn's octet and the cantata Die erste Walpurgisnacht. Attempts at direct alliance of music with poetry and painting were made by Berlioz and Liszt, who frankly employed the means of musical expression for purposes of illustration. The programme, suppressed or implied, made its appearance in the concertos of Spohr and Weber. The instinct for concentrated expression produced Schumann's string quartets. Weber in the ball-room and Chopin in the salon infused the Romantic spirit into the dance. The growth of instrumental technique is exemplified in Paganini, Chopin, and Liszt ; the last of whom invented the ' Dramatic Fantasia J and produced faithful transcriptions of symphonies, overtures, and songs for the pianoforte. Berlioz and Wagner extended the technique of orchestration. National elements, Polish, Hunga- rian, Norwegian, Spanish, Russian, came into play. Schumann was successful in obtaining a perfect equipoise between verse and music in the Lied. Impressionism and word-painting made their appearance in Liszt's Lieder and Berlioz' chansons; the dramatic and histrionic element prevailed in the vocal ballade ; and music, even apart from the stage, was pressed into the service of melodrama.

One result which followed from this attempt to make music representative was the development of a new kind of comedy, its lighter form in Offenbach and Sullivan, its more elaborate in Wagner's Meistersinger and Verdi's Falstaff. At the same time national opera, based on folk-tunes and dances, appeared in Poland, Bohemia, and Russia. Finally Wagner gathered together the various Romantic tendencies ; tested them through a period of experiment, speculation and theory, and ultimately reached the music-drama.

The main problem set before us by the Romantic movement is that of an alternative between programme music, with concomitant laxness of structure, and self-dependent music, it

364 THE ROMANTIC PERIOD

may be in conformity with an unwritten programme, but firmly and consistently designed on musical lines. The case for self- contained instrumental music cannot be better stated than in the words of Schopenhauer : c If we look at pure instrumental music, we shall see that in the symphony of Beethoven there reigns the greatest confusion, beneath which nevertheless there is the most perfect order : the most violent strife, that in the next moment grows into loveliest concord : it is rerum concordia discors, a true and complete image of the essential nature of the world, that rolls on in the immeasurable complication of countless shapes, and supports itself by constant destruction. At the same time all human passions and emotions speak from this symphony : joy and sorrow, love and hate, fear and hope, in countless gradations ; all however in the abstract only, and without any particularity ; it is merely the form of emotion, a spirit world, without matter. It is true, however, that we are inclined to realize it while listening, to clothe it in our fancy with flesh and blood, and to see in it the various scenes of life and nature. Yet on the whole, this neither facilitates its comprehension, nor enhances its delight, giving rather a hetero- geneous and arbitrary alloy : it is therefore better to receive it directly and in its purity V

Perhaps the following considerations may be taken to repre- sent fairly what can be said of music with an implied or an avowed programme. In characteristic overtures such as Mozart's Don Giovanni, Beethoven's Coriolan, Leonora, Egmont, and the overtures of Weber, the feeling for musical symmetry and proportion is completely in accord with the tendency to express the particular mood or moods indicated by the titles. But in the poeme symphonique, symmetry and proportion are made to depend, not on purely musical, but on more or less extraneous considerations. Under such circumstances, so long as the com- poser develops his subjects and figures on musical lines and makes no attempt to deal with concrete facts of any kind,

1 Schopenhauer, Die Welt als Wttle und VorsteUung, ii. chap. 39,

SUMMARY AND CONCLUSION 365

musical design may still be direct and definite. Conversely the question may be asked : if a poem or a picture forms the programme, can the music be actually identical with the material contents of the verse or with the scene upon the canvas ? It can be analogous, but that is all. Yet so long as the moods of a poem or picture are truly rendered in terms of music, this very analogy will illuminate the musical form with its own beauty and suggestiveness. Programme music, at its best, means symbolism that is to say, expression gained by the use of symbolical phrases which are treated by some intellectual process necessarily logical. But the genius of symphonic music cannot rest satisfied with signification. Beethoven's musical design is complete in itself; it appeals to us not by what it signifies, but by what it is, and its fullest expression never obscures or weakens its architectonic structure. On the dramatic side, Wagner's method is at present the most completely organized system for purposes of musical expression. On the instrumental side illustration, apart from design, is in pursuit of a false ideal : it is the satyr Marsyas, imitating on his flute the music of his native uplands, and doomed to destruction if he challenges the golden lyre of Apollo.

INDEX

ABT, 314. ADAM, g. ALKAN, 330-1. AMBROS, 358. ARNOLD, von, 359. AUBER, 9, 68, 337. Operas, 36-7, 57, 60.

BACH, J. S., i, 6, 13, 24, 150, 240, 266, 350.

BALAKIREV, 7, 321.

BALFE, 78.

BEETHOVEN, i, 6, 7, n, 13, 77, 106-7, 112, 117, 123, 228-9, 237, 240, 266, 287, 346-7, 354, 355.

356, 364- BELAIEP, 328.

BELLINI, 9, 44, 47, 184, 259, 337. Operas, 58-60.

Influence on Chopin, 60, 254. BERLIOZ, 2, 7, 9, n, 13, I4~i5» l8» 50-6, 65, 112, 173, 198, 260, 289,

353, 357, 362-3. Operas, 50- 6, 69-70. Overtures, in, 135-8. Symphonies, 111-35, 145, J48. Enfance du Christ, 11, 71, 170, 190. Damnation de Faust, 166-70. Requiem, 174-85, 190. Te Deum, 185-90. Songs, 278-81. Criticisms, &c., 355-6.

BIZET, 49, 72, 323.

BOEHH, 6.

BOIELDIEU, 6, 9.

BOITO, 63.

BORODINE, 7, 139, 236, 321-8.

BORTNIANSKY, 359-60.

BBAHMS, i, 7, 14, 191, 199, 209,

229, 267, 271, 315, 350. BRUCH, 330. BRUCKNER, 7. BULOW, 289, 319-20, 358. BUXTEHUDE, 6.

CANTATAS. See Oratorios.

CATCH, 301-2.

CHAMBER Music, 234-6, 311-13,

314-15, 320, 322, 329, 332. CHERUBINI, 77, 174. CHEZY, Frau von, 21. CHOPIN, 6, 12, 14, 24, 60, 153, 1 86,

317, 327, 350, 353, 357- Concertos, 230-1. Pianoforte Music, 240, 250-8, 261,

268, 269.

CHRYSANDER, 359. CONCERTOS, 225-34, 311-13, 320>

329, 330. CORNELIUS, 73-4, 289, 316-19.

Cui, 7, 321. CZERNY, 263-6.

DALAYRAC, 9.

368

INDEX

D'ANGLEBERT, 6. DARGOMIJSKY, 321, 322. DAVID, Felicien, 9, 122, 138-41. DAVID, Ferdinand, 153. DEITERS, 359. DELIBES, 72. DONIZETTI, 44, 49, 63. Operas, 58-60. DRAESEKE, 137, 330, 358. DUSSEK, 6, 237.

EDITIONS OF CLASSICS, 358-60. ELLIS, Mr. W. A., 338, 354, 359. ERKEL, 74. ERNST, 13, 225.

FESCA, 14, 352.

FIELD, 232, 233, 238-9, 255, 357.

FLEISCHER, 359.

FLOTOW, 67, 68.

FRANZ, 13, 271, 357.

Songs, 271, 275-8. FROHBERGER, 6, 150.

GADE, 7, 14, 289, 310-11.

GALUPPI, 360.

GARRETT, 332.

GAULTIER, 358.

GAUTIER, 278.

GLASENAPP, 359.

GLAZOUNOV, 139.

GLEE, 301-10.

GLINKA, 9, 74-6, 321.

GLUCK, i, 14, 19, 339, 355.

GOETHE, 3, 106, 165, 244, 281, 357.

GOETZ, 73, 289.

GOLDMARK, 49.

GORDIGIANI, 314.

Goss, 301, 303-9.

GOSSEC, 174.

GOUNOD, 49, 314.

Operas, 71.

Oratorios, 71. GRETRY, 9. GRIEG, 225, 330.

GRIMM, 1 4.

HALEVY, 9, 49, 63. HATTON, 301, 314. HAUPTMANN, 153, 242. HAYDN, n, 24. HAYES, 302. HEBBEL, 32-3. HEINE, 105, 221, 241, 337. HELLER, 14, 289. HELMHOLTZ, 359. HENSELT, 225, 231-2.

HlLLER, 233, 358.

HOFFMANN, 4, 14, 32, 243, 337,

352-3, 354, 358. Operas, 24-6.

Criticisms, &c., 350-2, 357. HUGO, Victor, 9, n, 142, 149, 279,

281. HUMMEL, 225, 242, 266.

ISOUARD, 9.

JAHN, 359. JAHNS, 359. JENSEN, 289, 316. JOACHIM, 13, 225, 229. JOUY, 43.

KIEL, 289, 315. KIRCHNER, 289, 316.

KORNER, 17.

KOUKOLNIK, 74. KREUTZER, 67. KUCKEN, 314.

LABITZKY, 238.

LABLACHE, 58.

LACHNER, 314.

LALO, 72.

LANGHANS, 359.

LANNER, 238.

LASSEN, 314.

LEITMOTIF, 142, 191, 200, 338.

LERMONTOV, 312.

LESUEUR, 68.

INDEX

LlADOV, 328.

LISZT, 7, n, 12, 13, 14, 34, 122,

173, 220, 246, 255, 289, 331, 342,

353, 362-3. Symphonies, 141-8. Podmes symphoniques, 149, 155. Smaller Choral Works, 171-2, 219. St. Elisabeth, 190-6. Thirteenth Psalm, 196-9. Masses, 199-209. Christus, 209-19. Concertos, 225, 232-3. Pianoforte Music, 258-70. Songs, 281-5. Melodramas, 287. Criticisms, &c., 357. LOEWE, 13, 286-7.

LORTZING, 67-8. LULLY, 71, 122.

MARPUBG, 14.

MARSCHNER, 8, 14, 273, 337, 352.

Operas, 27-31. MASSENET, 49. MASSES, 1 1- 1 2, 174-209.

Graner Fest-Messe, 12, 199-204.

Messe des Morts (Berlioz), 1 1, 174- 85.

Ungarische Krvnungs-Messe, 12,

204-9.

MATTHESOK, 14. MEHUL, 6, 186. MELODRAMA, 72, 78, 287-8, 313,

319- MENDELSSOHN, 5, 7, 10, 14, 137,

253, 289, 310, 350, 353, 362-3. Overtures, 80-5. Symphonies, 90, 104-6. Oratorios, II, 156-60. Lobgesang, 1 60. Motets, Cantatas, &c., 161-2. Concertos, 225-30. Chamber Music, 234-5. Pianoforte Music, 239-40. Songs, 274-5.

MENDELSSOHN (con*.)

Melodramas, 287.

Organ Music, 331. MERCADANTE, 49. MEYERBEER, 8, 9, 14, 138, 338, 35o, 353-

Operas, 45-9. MIKULI, 258. MONIUSKO, 9, 74, 77. MONSIGNY, 9. MOSCHELES, 225.

MOZART, i, 13, 14, 19, 24, 47, M7i 259, 266, 364.

MttLLER, 32. MUSORGSKY, 321, 328.

NEWMAN, Mr. E., 329. NICOLAI, 67-8, 70. NIETZSCHE, 345. NUITTER, 339.

OFFENBACH, 73, 78, 363. ONSLOW, 312. OPERAS :

Abimelek, 49, 352. Adelson e Salvina, 59. Africaine, L\ 49. Atda, 49, 63-6. Alceste, 6, 355. Anna Hole tut, 59, 60.

Hal masque", Le, 60. Ballo in Maschera, Unt 60. Barbier von Bagdad, Der, 73-4. Barbiere, II, 38, 39. Bartered Bride, The, 9, 74. Beatrice et Benedict, 70. Benvenuto Cellini, 9, 50-6, 73. Bohemian Girl, The, 78. Box and Cox, 78.

Capuletti e Montecchi, I, 59. Carmen, 72, 323. Cenerentola, La, 38. Chanson de Fortunio, La, 73.

DANNREUTHER

Bb

INDEX

OPERAS (cont.). Cid, Der, 74. Colporteur, Le, 312. Comte d'Ory, Le, 38. Conies d* Hoffmann, Les, 73. Corsaro, II, 60. Crodato in Egitto, H, 49.

Dame Blanche, La, 6. Dame de Pique, La, 329. Demon, The, 312. Dinorah, 49. Djamileh, 72. Domino Noir, Le, 57. Dom Pasquale, 63. Dow Sebastien, 49. Don Carlos, 49, 60. Donna del Logo, La, 38. D«c Foscari, I, 60.

Eclair, L', 49. ^

Elisabetta, Eegina d'Inghilterra,3%. Emma di Resburgo, 49. Entfuhrung ausdem Serail, Die, 19. Ernani, 9, 60.

Granata, L', 49. Afortf, Z/', 49. Eugene Onegin, 329. Euryanthe, 5, 16, 18, 19, 20-2, 32, 34, 341, 362.

Fair Maid of Perth, The, 72. Falstaff, 63-6.

(Gounod), 71.

(Spohr), 23. Favorita, La, 49, 58, 60. Feen, Die, 333-8. Fidelio, 8, 17. Figaro, 19.

Figlia del Reggimento, La, 60. Fliegende Hollander, Der, 23, 31,

333-9, 357-

Forza del destino, La, 49, 60. jFra Diawlo, 57. Franfoise de Rimini, 72.

OPERAS (cont.). Freischutz,Der, 5, 8,17-20,355,362.

Ladra, La, 38. Genoveva, 16, 31-5. Giovanna d'Arco, 60. Giuramento, II, 49. Gondoliers, The, 78. Grande Duchesse, La, 73. Guillaume Tell, 9, 38-44. Gunlod, 74.

9, 74, 77. Hamlet, 76. .Hans Heiling, 8, 27. Heinrich der Vierte, 27. Huguenots, Les, 46-8, 353.

Iphigdnie en Tauride, 19. Italiana in Algeri, L', 38. Ivanhoe, 79.

Ji3aw rfe Paris, 6. Jephthas Tochter, 48. Jessonda, 23. Joseph, 6. Judith, 9, 74. Jww, La, 9, 49, 63.

Konigin von Saba, Die, 49. Kreuzfahrer, Der, 24.

Lakme, 72.

Liebesverbot, Das, 333-8. Lohengrin, 8, 21, 23, 47, 63, 67,

200, 333-43. Lombardi, I, 60. Lucia di Lammermoor, 43, 44, 58,

60.

Lucrezia Borgia, 58, 60. Luisa Miller, 60. Lurline, 78.

Macbeth, 60. , Le, 57.

INDEX

OPERAS (cont.). Margherita ffAnjou, 49. Marino Falieri, 60. Maritana, 78. Martha, 67. Mamadieri, I, 60. Matrimonio Segreto, H, 38. Me'decin malgre lui, Le, 71. Meistersinger, Die, 5, 8, 43, 73, 343,

351, 363- Merlin, 49. Merry Wives of Windsor, The, 67,

68, 70. Mignon, 72. Mikado, The, 78. Mireille, 71. .flfos^ w Egitto, 38.

Portici, La, 36-7, 57.

Nabuchodonosor, 60. Nachtlager von Granada, Das, 67. Nonne sanglante, La, 71. JVbrmo, 9, 58, 59.

06m>n, 5, 19, 23, 355. Olympie, 43. Orpfcfe, 355. Orphe'e anx enfers, 73. , 44, 59, 63-6.

Parsifal, 9, 224, 257, 333, 342. Philemon et Baucis, 71. Pfc»ftr«, Le, 57. Pirata, 11, 59. Pottw of Evil, The, 77. Preciosa, 16, 19, 20. Prince Igor, 322-8. Prince Kholmsky, 75. Pme rfe TroiV, La, 69. Prophete, Le, 46, 48, 63. Puritani, I, 9, 43, 58, 62.

Eheingold, Das, 200. .B/enzi, 333-8. Rigoletto, 9, 43, 60-3, 66.

OPERAS (cont.). Ring des Nibelungen, Der, 8, 342,

344, 349«

Robert le Diable, 8, 9, 45. Rogneda, 77. Roi d" Ys, Le, 72. Roi fa dit, Le, 72. Rome'o et Juliette, 71. Romilda e Constanza, 49. Roussalka, 322. Russian et Ludmilla, 9, 74.

71.

Semiramide, 38. Semiratnide riconosciuta, 49. SiVjre Corinthe, La, 38. Siegfried, 5. Sonnambula, La, g, 58. Sylvana, 19.

Taming of the Shrew, The, 73.

TancraW, 38.

Tannhauser, 23, 31, 63, 196, 200,

333, 339-43-

TVmpkr unrf di> Judin, Der, 27-3 1 . rmn'ate, La, 60-3, 66. Trial by Jury, 78. Tristan und Isolde, 8, 129, 257,

343-

Trovatore, 77, 60-3, 66. Troyens, Les, 69-70. Turandot, 16, 19, 20.

Undine, 24. 6.

Vampyr, Der, 8, 27. F^pre* Siciliennes, Les, 60. Fe«<a7e, La, 6, 37, 43. Fi> pour le Tsar, La, 74-6.

WaldmSdchen, Das, 19. TTa//ture, Die, 343. Widerspenstigen ZtihnniHci, Der, 73-4-

373

INDEX

OPERAS (cont.). Yeomen of the Guard, The, 78. Zelmira und Azor, 23.

ORATORIOS, CANTATAS, &c. : Christus (Liszt), 12, 209-19. Christus (Mendelssohn), 160. Christus (Lachner), 315. Comala, 311. Creation, u.

Damnation de Faust, 168-70. Elijah, II, 158-60. Erlkonigs Tochter, 311. Enfance du, Christ, n, 71, 170." Golden Legend, 314. Gott Ttnd die Natur, 48. Last Judgment, u. Liebesmahl der Apostel, Das,

323-4-

Light of the World, 314. Lobgesang, 160. Martyr of Antioch, 314. May Queen, 311. 3fors Ft to, 71. JVtfwX 315.

Paradise and the Peri, 162-4. Prodigal Son, 314. Redemption, 71. S#. Elisabeth, 12, 190-6. Sfe. PottZ, u, 157-8. Seasons, II.

Stem ww Bethlehem, Der, 315. Walpurgisnacht, n, 161-2.

ORGAN Music, 331-2. OSTROVSKY, 77. OVERBECK, 4. OVERTURES :

Beethoven, 364.

Berlioz, in, 135-8.

Goetz, 315.

Mendelssohn, 10, 80-5.

Mozart, 364.

Schumann, 10, 85-90.

Sterndale Bennett, 311.

OVERTURES (cont.). Sullivan, 314. Tchaikovsky, 329. Wagner, 106-10. Weber, 10, 364.

PAGANINI, 13, 225, 363. PALESTRINA, i. PARAPHRASES, Les, 327-8. PARRY, 7, 18, 45-6, 329, 339. PEARSALL, 301. PERGOLESI, 6. PICCINNI, 8. POEMES SYMPHONIQUES, n, 77, 319, 329. 362. Liszt, 149-155, 321-2.

POHL, 359.

PROGRAMME Music, 111-55, 363- 365-

PURCELL, 301.

RACHMANINOV, 232. RAFF, 233, 289, 320. RASUMOVSKY, 359. RAUPACH, 32. REISSIGER, 67. RHEINBERGER, 331-2. RICHTER, Jean Paul, 241, 243, 249, 353-

RlEHL, 358. RlEMANN, 359.

RIES, 233.

RlETZ, 153.

RIMSKY-KORSAKOV, 7, 139, 321. RlTTER, 359. ROMANI, 59.

ROMANTIC ideal in Music, 5-12. ROMANTIC movement in Literature,

3-5, U, 15- ROSSINI, 9, 22, 47, 64, 147, 314.

Operas, 37-44, 64.

Stabat Mater, 184, 221. ROUSSEAU, 2, 14, 20, 287.

RUBINI, 58.

RUBINSTEIN, 7, 139, 233, 266, 289, 312-3.

INDEX

373

SAINT-SAENS, 49, 122, 139, 148.

S A YN-WlTTGENSTEIN, Princess, 357.

SCHEFFER, 146. SCHILLER, 3, 11, 149, 357. SCHLEGEL, 4, 243, 353. SCHOPENHAUER, 355, 364. SCHUBERT, 5, 6, 20, 32, 255, 266,

271.

SCHUMANN, 5, 7, 10, 11, 12, 13, 14, 24, 46-8, 106, 172, 186, 253, 264, 289, 310, 331, 350, 351.

Genoveva, 31-5.

Overtures, 85-90.

Symphonies, 90-104.

Paradise and the Peri, 162-4.

Smaller Choral Works, 164-5.

Faust, 165-8.

Concerto, 230.

Chamber Music, 234-5.

Pianoforte Music, 240-50, 269.

Songs, 272-5.

Melodramas, 287.

Criticisms, 46-8, 352-3. SCHUTZ, I, 6. SCRIBE, 36,45, 46/57. SEIFFERT, 359. SEROV, 9, 74, 76-7.

SlNGSPIEL, 1 6.

SMETANA, 9, 74, 77. SMOLENSKY, 359. SOLOVIEV, 77. SONGS, 271-87.

Schumann, 272-4.

Mendelssohn, 274-5.

Franz, 275-8.

Berlioz, 278-81.

Liszt, 281-5.

Wagner, 285-6.

Loewe, 286-7.

SOUMET, 59.

SPOHR, 6, 7, n, 24, 47. 225, 363. Operas, 23-4.

SPONTINI, 6, 36, 43. 336, 342, 355. STANFORD, 7, 35. STCHERBACHEV, 328.

STERNDALE BENNETT, 7, 14, 78,

233, 289. 310-12. STEVENS, 310. STRAUSS, J., 238. STRAUSS, R., 7, 288. / SULLIVAN, 78-9, 289, 363. SYMPHONIES :

Berlioz, n, 111-35.

Borodine, 321-2.

Gade, 311.

Goetz, 315.

Liszt, 141-8.

Mendelssohn, 90, 104-6.

Raff, 320.

Rubinstein, 312.

Schumann, 10-11, 90-104.

Sterndale Bennett, 311.

Tchaikovsky, 329.

TAUSIG, 266.

TCHAIKOVSKY, 225, 229, 315, 323,

329-30, 360. TELLEFSEN, 359. THALBERG, 13. THAYER, 359. THOMAS, 49, 72. TIECK, i, 32-3, 278, 337, 353. TRESSAN, Count, 21.

VERDI, 9, 44, 64, 363. Operas, 60-6. Requiem, 221-2. Smaller Choral Works, 222-3. Quartet, 235-6.

VlANA DA MOTTA, 25. VlEUXTEMPS, 13, 225.

VIRTUOSI, influence of. 13.

VOGLER, 49. VOLKMANN, 289, 314-5.

WAGNER, i, 2, 5, 7, 8, 13, 21, 22 23, 34, 36, 49. So. 57, 64, 69, 76, 129, 135, 147, 153, 154, 155, 172, 191, 200, 217, 271, 286, 350-1. 355, 357, 358, 362-5-

374

INDEX

WAGNER (cont.). Famt Overture, 106-10. Smaller Orchestral Works, no,

136. Liebesmahl der Apostel, Das,

223-4.

Songs, 285-6. Sketches, 333-4, 344. Operas, 333-43- Criticisms, &c., 58, 135, 154, 155,

218, 256, 344-9. 354-5- WALLACE, 78. WASIELEWSKY, 359. WEBEB, 2, 5, 7, 13, 14, 24, 25, 47,

68, 80, 266, 273, 337, 362-4.

WEBEB (cont.).

Influence of, 7-10.

Operas, 16-23.

Concertstiick, 80, 225-8.

Concertos, 228-9.

Pianoforte Music, 237-8.

Criticisms, &c., 352. WEITZMANN, 359. WESLEY, S., 13, 78, 289-90, 332. WESLEY, S. S., 13, 78, 289-300, 332.

WlENIAWSKI, 13.

ZELTEB, 287. ZUMSTEEG, 287.

Oxford : Printed at the Clarendon Press by HORACE HART, M. A.

PLE CARDS C

*"• 9 1992

-Q

oo

3

•'

'

'