stephan, hindemith's marienleben 1922-48 (music review 15, 1954)

7/28/2019 Stephan, Hindemith's Marienleben 1922-48 (Music Review 15, 1954)

1/7

274 THE MUSIC REVIEWMannes"; often, as here, Berg succeeds in presenting Wedekind's non-emotionalLulu, particularly in his treatment of the exchange between Schwarz and Lulu .(act one/I), "Can you speak the truth?". "I don't know". "Do you believein a creator?". "I don't know", etc.)

III. CoNCLUSIONSI believe that I have shown in sections I and II that Berg himself pre

cipitated a large-scale dramatic confusion when basically (either consciously orunconsciously) altering the constitution and character of the opera's Luluwithout duly amending the text and dramatic action. While bearing in mind .the tantalizingly incomplete last act which, if finished, might have made manyevents clearer, I think there is little doubt that Lulu, as it stands, is not aconvincing musico-dramatic entity; what goes on in the orchestra pit and on thestage fail to match, and one suspects that the third act, however conclusive,would not have succeeded in balancing the two levels. This inconsistency,perhaps, would be an altogether shattering flaw were it not for the fact that themusic, unlike the drama, is satisfyingly consistent on its own masterly level;and I suggest that the problem of Lulu's at present ill-integrated and alarm-ingly erratic characterization could be partially overcome by deriving the truestate of dramatic affairs from the music rather than the libretto and its origins.In short, the next production of Lulu (and is there any reason why CoventGarden should not take a few practical steps in this direction?) must jettisonWedekind and preconceived notions founded on Earth-Spirit and Pandora'sBox, and concentrate on Berg's transformed and developing Lulu drawn fromthe opera. I t is useless, in my view, to attempt to reproduce Wedekind'sstatic "innocent corruptress" who plays "a n entirely passive role" when, inmusical terms, she is most active and experienced in her feelings (as in "MeinesMannes" quoted above). Since Berg allows his Lulu to indulge in emotionaldisclosures which would not disgrace a tragic operatic heroine of a less eccentricdisposition, the producer of the opera must respond by creating a stage figure

II(I

and environment from which these disclosures may naturally proceed. Thetrouble with Hartleb's Essen production was not that Lulu developed from jact to act, bu t that she developed from impossible beginnings. By permittingLulu to develop at all, Hartleb was half-way towards the realization of Berg's Iulu, bu t by too great an insistence on her as a "completely natural andunintellectual creature" he was still trying to cling to Wedekind's coat-tails;and, as I have suggested, Berg's and Wedekind's Lulus are almost mutuallyexclusive. Or, more accura tely, Berg's music excludes \Vedekind. IIf nothing else, this rudimentary comparison between \Vedekind's text and Berg's opera proves that a large part of the difficulties of producing a convincing ILulu is due to Berg's own curious approach and attitude to his heroine (an linconsistency that extends besides to other characters , Alwa especially). _One ma y also reasonably conclude that the course his adaptation eventually .pursued was forced upon him by the very nature of the ar t he practised. 4 iMusic ceases to be music when it ceases to express feelings; and to write an .f 'opera about a heroine virtually without any feelings wou ld no t only have beenbeyond Berg's creative powers, but, as a creative musician of genius, beneatlihis contempt.

- ~ ' .}

Hindemith 's Marienleben (1922-1948)An assessment of its two versions

BYRUDOLF STEPHAN

ALL composers who-at the turn of the century-felt unable to confine themselves to the traditional features of style, were compelled to organize theirmaterial afresh either by breaking consciously with tradition or by continuingindependently on the line of its chief stimuli. Such purposes were mainlyserved by the miniature forms of Lied and piano piece.1 Once cyclic form ha dbecome pro blematic with the disintegration of tonality, it is not surprising tonotice that the first larger compositions to put a new style into practice werecycles of songs in which the unity of the poetic idiom fostered the musicalcontinuity.

In that sense Arnold Schonberg's Buch der Hiingenden Garten, op. 15 (completed in 1908, after poems by Stefan George) is the earliest specimen of auniformly organized new style whose formal logistics result from the dialecticrelationship between motives (not melodies) and their harmony. The implicitmusical abundanc e of Schonberg's music-which could be neither understandable nor agreeable to the poet-has transformed the poems by liquidating theirstylization, thereby depriving them of their rigidity. Pierrot Lunaire, op. 21(1912), shows more extr emis t principles of composition. Schonberg's tendencyto objectify the morbid imagination of A. Giraud and Otto Erich Hartleben(his German translator) leads him to conjure up old musical patterns. Thisis not synonymous with a subtractio n of all subjective matter; on the contrary,subjectivity is it s sole guarantor. The ancient static forms are therefore notsuperimposed on an alien musical language, bu t are adjusted to the musicalmaterial, i.e. they have become dynamic.Approximately a decade later Hindemith follows an analogous trail in hissong cycles, Die junge Magd, op. 23 (on poems by Georg Trakl), and DasM arienleben, op. 27 (on poems by Rilke). In op. 23 direct expression prevails,whereas in op. 27 deliberate formalization becomes apparent. Pierrot, as wellas Marienleben,-in contrast to the earlier works which ha d been conceivedfrom the angle of sonority-show a strong tendency towards polyphony, insome nu mbers completely determining the manner of the part-writing. Thatthe character of polyphonic style diverges in each composer is partly conditioned by the difference of tradition to which each feels indebted: Schonbergstarts from Mahler, while Hindemith stems from Reger, the classicist.2 Therefore Hindemith's polyphonic style emerges more clearly from Bach thanSchonberg's, and concertante elements become more and more important fromthe string Quartet, op. 22, onward until they are fully developed i n his compositions of the middle "Twenties", chiefly in opp. 31-39. The motoric impulse

' Bart6k's Bagatelles, op. 6, an d SchOnberg's Three piano Pieces, op . I I, were composed in rgo8.:a For an explanation of the term "Neo-classicism" cf. the attempt at a definition in DeutscMUniversitiitszeitung VII, 1952, H. t6/I7.

II


2/7


3/7

278 THE ~ I U S I C REVIEWrather "operatic" version with orchestral accompaniment. The composeradded a comprehensive preface to this new version in which he invites th einterested stud ent "to immerse himself in the problems which are hiddenbeneath the surface and not within easy reach".

How far off seem the days when Hindemith, asked to explain or to analysethose of his compositions which were then being performed at the festival ofDonaueschingen in I922, was only willing to give the following terse reply:" . . . I cannot give analyses of my works, because I do no t know how to explain a pieceof music in a few words; I prefer to write a new piece in that time. Besides, I believethat my stuff is easy to understand for people with ears an d therefore an analysis issuperfluous. One cannot help people without ears with such makeshift expedients.I also refuse to write down a few themes only as they always tend to give a wrongimpression''.5

That was more than 30 years ago. In I937 the composer published the theoretical part of his treatise, Die U n t e r w e i s ~ m g im Tonsatz, Volume I, in which hedeveloped his opinions on the basis of arguments mainly derived from acousticsand from Tonpsychologie. These opinions have not remained uncontradicted,especially when, from problematic premises, conclusions have been reachedwhich may be called dangerous without undue exaggeration.6 Since thenHindemith's musical idiom has submitted to a large degree to these selfimposed demands.

In the preface to the new version of the j\ I arienleben th e original version issubjected to severe self-criticism. The vocal part is said to have been "dictated by considerations of a non-vocal nature" and to have contained "progressions difficult and sometimes impossible to execute". 7 Hindemith thencontinues by asserting that the song cycle has become "an object ofutility", "nothing to fuss about", that it contains "badly balanced chromatics,complicated intervallic skips and tonally incommensurable features . .. .Mistakes of the latter kind are excused by ignorance of the later "discovered"laws, promulgated in th e Unterweisung, errors of the former type are explainedby "inexperience in vocal composition".8 The proof deduced for this contention carries little conviction, however, as th e ability to intone difficultintervals remains in the first place a task for the ear and only in the secondinstance a matter for the vocal chords (preface, pages iiifiv). In the thirdparagraph Hindemith explains the structural plan of the new version in whichsongs I-4. s-g, Io-12 and I3-IS have been grouped together. Every suchgroup should contain a dynamic and expressive climax. According to Hindemith the dynamic climax of the whole opus is "Die Hochzeit zu Kana" (no. g)and he indicates "Pieta" as its expressive counterpart. In the following paragraph each item is discussed and the reasons are given which have led to its

6 ."-'leue A1usikzeitung, year 43, 1 9 2 2 , p. 329. Cf. H. Schole in Die cUusik, XXX, 1938, Vol. 2, p. 528-35; Rene Leibowitz in L'Arche, III,Nov., 1946, H. 21, p. III , alsop. I I5ff an d J. Handschin, Der Toncharakter, 1948, p. 13oj.7 F. \Villms, however, writes in his purely apologetic study: " . . . As far as purely technicalproblems of performance are concerned, they are of no exceptional difficulty, either for singer oraccompanist . As regards difficult intervals, Hindemith makes lesser demands on the singer'spowers of intonation than Strauss or Schreker . . .".8 At that time Hindemith had composed many vocal compositions: operas and songs, half ofwhich ha d a!Ieady been published, viz. opp. 6, 9, 12, 13, 14, 20, 21, 23/I-2.

'I..

'I''f

;I

tr.

HINDEMITH'S "MARIENLEBEN" {I922-I948) 279reVISIOn or re-writing. In the main th e revision aims at more comfortable J /singability and at more clearly perceptible tonality.In the fifth section of the preface Hindemith discusses the character of keyswhich he develops on the basis of original, bu t totally arbitrary premises.He bases everything on a row, no. I, arrived at in the Unterweisiing, which hasalso determined the sequel of fugues in Ludus tonalis.9 The starting point ofthat row is "E " which in the Marienleben becomes the keynote. To this"E " tonality as to every other key, Hindemith attributes a certain "emotionalclimate" and an "ideological category". In proportion as the relations of th edifferent keys to the central tonality become weakened {all of them are rathernaively determined from the angle of row I), the composer allots a more distant"ideological category". As he himself has admitted the arbitrariness of theargument (he calls it an "agreement"), nothing more shall be said about it here.

For reasons of elucidation the analysis of the subject of no. I4, "Vom Tode,tJariii" II, may be quoted in the new version (preface, page ix).

". . . The tonal contemplation embodied there could lead approximately to th efollowing sequel of thoughts and emotions: we feel th e unification with infinity (C, ba r1) which, with its unavoidable inexorability (C sharp, 2-3) an d also with its infinitemildness (diffuse G, 4), infuses in us the sensation of minuteness. Although we havefaith in fate (D, 7) we are nevertheless tortured by a little feeling of incomprehensiveness (B flat, 8). The pious one will see in th e saviour (E, g-10) an d in his once mortalmother (B, II} his leaders into th e region of the purity of death (E flat, II-rz)".What hearer without absolute pitch and without having memorized a table of

key interpretations is able, with the best will in the world, to feel according tothis prescription? Hindemith, forestalling such remonstrances, adds immediately; "... I don't expect over-enthusiastic acceptance of my tendency toburden the sound with so much ideological matter". The preface is woundup by some remarks about new music, of a confessional kind and therefore no twithout interest for the understanding of the composer's basic approach.

The single items may now be classified with regard to the degree to whichthey have been subjected to revision. l

I. Taken over unchanged from the first version into the new one: no. I2("Stillung Mariii mit dem Auferstandenen").II . Revisions:{I) Single notes only (re-arranged, eliminated or added) in nos. 4,5, I I;

(2) Thoroughly revised, bu t retaining the original thematic matter:nos. I, 2, 6, 8, IO, I3, I4, IS;(3) Changed structure, with thematic "substance" retained fromversion I ; no. 9 ("Hochzeit Zi t Kana").

III. Entirely new compositions: nos. 3 and 7The fundamental tendency of these "emendations" emerges clearly fromgroup Iljr . The simplification in harmony and tonality shows at the beginning

Vol. I, new edition, 1940, p. 75 !f. and p. II9.

J


4/7

280 TH E MUSIC REVIEWof no. 4 ("Maria Heimsuchung"), where a complicated bitonal harmonic process has been reduced to a relationship of fifths with the help of the old musicalpanacea, "pedal point".

Ruhig bewegt fJ., SO)

Later in this song the composer has interpolated some bars by forcibly alteringthe vocal part. In no. 5, "Argwohn Joseph's", an analogous situation is to befound, whereas in "Pieta" (no. n) only a few notes have been modified.

The items of group II/2 show in principle the same type of changes. AsHindemith believes he has discovered more serious shortcomings in the firstversions of these items, the modifications are naturally more frequent and moreconsequential. For instance, in no. I, the well balanced passage "I n dieserNacht ... no w becomes a mere transition and the magical dissonant tripletsabove the chord of fourths are eliminated without further ado.Ex . 6af1tJ in dieser Nacht wird demJCnaben die Kul - le r ge fl ~ ~ . , & .,.. b. -;;-:-..... = - ~ ~ ~ ...- ~ ~ - ~ ~ n : J ;:n-

j ~ p Li&." pp p --===v : : : : = : : : : . ~ w- ~ # 1 1 ?

Ii1

'I.

tI

H I N D E M I T H ' S "MARIENLEBEN" (I922-I948) 28ISimilarly, the passage, "Ach ... sie fuhlten sich ... (op. 27, ba r 52jJ.),so impressive with its unresolved suspensions, has been dropped. In no. 8 therecitativo section has been reduced i n size, and in nos. 4 and IO single sectionsof melody have been transposed, thereby creating simpler tonal relationships.In no. I4 the unfettered metre has been regularized, a feature which has evidently not been felt as a similar disturbance in other places. To no. 9 thecomposer now attributes a central importance within the framework of thecycle, so that it had to be completely revised (preface, page vi).

In this paper the two versions will be compared at some points,without, however, tou ching upon the problems of modem Lied composition ingeneral.As ma y be seen from the table above, most of the songs of the new versionare reVISIOns. I f in group II/I things are so simple that the above remarksmay suffice, nos. 6 and IO from group II/2 warrant closer inspection. No. IO,although in principle less radically altered, could not be included in groupII/I because of the piano postlude.

The words of "Verkundigung uber den Hirten" (no. 6) are the most complicated of the whole cycle from a formal point of view. A glance at its rhymingscheme, however, reveals its poetic language to be formally organized despiteits apparent complication. The structure of it s contents (nowadays oftenfashionably labelled Sinnstruktur in Germany) is however relatively simple.Four parts are clearly distinguishable: (I) Evocation of the shepherds; (2)Self-description by the angel ("Seht, ich bin ein neuer, steigender Stern") ;(3) Direct address to the shepherds by the angel ("Ihr Unerschrockenen"); and(4) The announcement of wisdom made manifest ("Doch dieses war ... ) .The structure in op. 27 does not conform to this simple sequence bu t is moreconcentrated. Part I corresponds to p. 28, part II to pp. 2g-3I, ba r 6, part

Ex.Bbfl

t) "(l II ..

Kna. ben

ge bo renII. /'

I


5/7

282 TH E MUSIC REVIEWII I to p. 31, ba r 7-34, bottom. The transition from the third to the fourthpart in this composition is not entirely unambiguous. Although in part twoHindemith acknowledges the contrast between the luminosity of the starand the darkness of the glances riveted on it, yet he joins the themes expressing this contrast together by the same final cadence. On page 34 (top) thecomposer repeats the first half of the music of part two and of the second half(in the text "Doch dieses war ... ) he repeats the somewhat altered finalphrase. A shortened repeat of the music of part one continues on the text "Was .ist ein Darnicht uns .. . . So the decisive words "Gott fuhlt sich ein in einer]ungfrau Schooss" come to achieve greatest effect, as the character of part one

( ~ e p r ~ s e n t i n g an ~ v o c a t i o n ) is in the nature of a fanfare. I f the formal organrzahon of the Smnstruktur presents itself as a simple sequence, the composition's scheme ma y be understood roughly as A, B, C, B1, A1; where parts B1and A1 appear to have been considerably curtailed. In the new version theformal layout is approximately the same, except for the fact that parts B1 andA1 appear in motivic preparation. Part one ends differently because it has beentransposed a fifth higher and to allow part two to appear in the key of the firstversion. The principal subject of part two, intoned by the voice, is identicalwith the first version in the first ten bars, but is subsequently changed, therebyconsid_erably weakening the effect of that section. The "motoric" accompaniment IS now replaced by purely chordal support, the bass part of which revelsin intervals of the second and in pedal points-so characteristic of the laterstyle of Hindemith.

Ex.7Jj_

I" ' Seh t , _ iCh bin ein neu . .fl Von hier an sehr bewegt CJ ,a>

Op.27 '"' - ,.p s e n r p r ~ staccato1>?1 -

fl Lebhaft! ..J , : i ; O I > ~Newversion ... I,

T,....,...,_k)

-- stei - gen-der St.ern.-- 1"!"1"'1 -... ...

-""- ......~ . . J . _ J ~ . J J I f. I

...

This part has been extended by motivic interludes, seemingly based onnineteenth-century models.The middle section is new and totally deprived of the colouristic verve of

the original. The melody of the vocal part flows in chromatic or diatonic scalemotion, supported by the bass, running in part parallel to it at the distance ofan octave.

l_

tJ

HINDEMITH'S " l i !ARIENLEBEN" (1922-1948)

lr F FUn___lh r er - s ch ro c- ke - n en , o_ wiiss - tet ihr,fl '"""31 ~

l"'-- ~ ., ., ., 71 ....... ........cresc.

i bi ... ...Ex.Sb,., ~ - - - -I" ' lh r Un er - t ~ b r o c k e . non 0 wiiss- tet ihr,___/)

I" ' ~ 'ff _.,_ , ., 'ff .. ., .1!..== pplfl

I"' _ wie jetd a.ut eu remSchauen . den Ge - sicb - telfl --I " ' ~ ~ ~ # ~ ll ... ---I-' 'f--" ill- - - -'I

Even if the structure shows only slight traces of interference, the two versionsof the song reveal strong discrepancies in character, effected by the change inthematic substance and by the resulting differences in the technique of"motivic" treatment. In "Pieta" the poem runs on in simple stanzas. However song-character is not hereby achieved; nor, evidently, is it desired, as theorganization of sentences tends to overlap the stanza form. The tasteless text"0 hast du dies gewollt, du hiittest nicht durch eines Weibes Leib entspringen durfen."

(0 if you wanted this, you should no t have sprung from a woman's womb.)is separated, musically, at the punctuation sign which is quite feasible. Thisdoes not prevent Hindemith from fitting in song-like correlations. Save forquite small variants necessitated by the text (op. 27, ba r 12, 14, new version,6, 14, 17-18), the first two stanzas are musically identical and the fourth stanzaconforms to them as far as musical substanc e goes. Slight changes are unavoidable as the distribution of syllables at the beginning of stanza four doesnot conform to the pattern of the first two. Formal divergencies do not existbetween the two versions; except for the postlude added to the new versiononly of importance for the next song which should contain the expressiveclimax of the whole work. For this reason the composer has diminished the

:I!


6/7


7/7

r286 THE MUSIC REVIEWheterophonic treatment of the accompanying melodies in the piano part-hascaused him any particular annoyance.

What is the result of having compared both versions? We have noticedhow in the first version of the Marienleben several strata of style tend to overlap. Symptomaticall y enough Hindemith endeavours to smooth ou t allextremes. Thereby, in the first place, all rigidly neo-classical passages,whose harmonies are criticized as being arbitrary, 12 have been rooted out. Inmusic of neo-classical bent no t only does the harmony appear questionablebut also the employment of conventional formulae of melody and of formalpatterns. Exactly these, however, have been retained in the new version andeven enriched through cliches of harmony. By this means neo-classicism seemsto me to have excelled itself (if possible) in false positivism. That Hindemithfinds himself with these neo-classic tendencies in sudden proximity to Puccini,especially in his version for orchestra, is not further surprising (new version,no. 7, bars 57tf-). Yet also th e spontaneous expression, existent in the firstversion, has been modified.

Hindemith has retreated to the middle way of least resistance. Hisrenunciation of any radical spirit speaks clearly through his more recent compositions and is pressed into a direct formula in the final section of th e prefaceto the new version of the lv!arienleben as well as in his Schonberg-analysisin th e Unterweisung. 13 This continuous increase in animosity against anydifferentiated modernity of style could not but influence his own compositions,and so it may be time to query the technical mastery of Hindemith which isreceiving so lavish praise nowadays. Technical skill in disposing musicalmaterial does not alone constitute mastery, as it is by necessity a replica ofthose works of the past on which it has been modelled. To remain within thelimits of the achieved leads immediately to falsification and in this sense therehas never really existed an y such "Ars antiqua" as seems to be Hindemith'sevident goal. I t is therefore not surprising to find in the new version of theMarienleben sections which are simply badly composed (cf. Ex . Sb, the whole ofno. 3, especially bars 71-87 jj.). These sections represent nothing bu t theirown "jargon" which is made to stand for itself without any higher artisticintention. To this may be added the inconsistencies of the new version. Howis it possible that the composer can tolerate a certain feature of style in onesong, whereas he feels compelled to criticize it in another; as for instance th einstrumental treatment of the vocal part which annoys him in nos. 6 and 7of op. 27, but which he retains unaltered in no. 14 of version II ? How is itthat no. 12-although it cannot boast of either an exemplary "harmonic bent"or of being a model sequence of harmonic r a d a t i o n ~ o u l d have been retained

u In the first edition of Hindemith's Unterweisung one finds on page 25 I !f. a list of his composi-tions in which "the realization" of the composer's opinions on the technique of composition-asexpressed in the book-may be studied. This list contains, in chronological order, the opusnumbers: II/ I-4, 22, 23/2. 24/2. 25/I, 2, 3, 44/4. 45/I. 49. 50, th e oratorio Das Unaufh&liche.etc. Evidently the strictly neo-classical wor.ks--eharacterized above-are missing here. Onpage 25 2 Hindemith defines the new version of the Marienleben "as the practical elucidation ofhis theories"".11 Cf. Unterweisung. new ed., 1940, p. 254

1I

HINDEMITH's "MARIENLEBEN" (1922-1948) 287unchanged? Not even th e historic importance of the first version can beapproached b?' the new one. The somewhat motley style of op. 27 may have 7b e e ~ more unified, _but at the e x p ~ n s e of spontaneity. In place of spontaneity lone IS now faced With the concoctions of an unstable theory and with a handfulof practical formulae which are used by Hindemith and his imitators withinthe c o n ~ n e s . of respectable academicism. Yet Hindemith himself proved againand agam, m former years, that it is possible to compose new music of greatartistic value. J

(Translated by Hans F. Redlich.)

stephan, hindemith's marienleben 1922-48 (music review 15, 1954)

Documents