Musa 12042

BEN EARLE

TWELVE-NOTE MUSIC AS MUSIC: AN ESSAY IN TWO PARTS

I. Theory

Can it make sense to call music unmusical?1 If the perennial complaint ‘It isn’tmusic!’ is more than a contradiction in terms, it presumably expresses a judge-ment that the music in question should be described differently: as ‘noise’,perhaps. Even to raise this issue is to court accusations of conservatism. Pre-scriptive use of the word ‘musical’ has been aggravating composers for over acentury. Music will win its freedom only when it ceases to be ‘musical’, declaredFerruccio Busoni in 1907. For him, the usage was tautologous: ‘Musical is:that which sounds in rhythms and intervals’. The noise compositions ofthe futurists were still a few years off. Indeed, Busoni had as his target amore subtle distinction. In German, he explains, musikalisch denotes a listenerwho is particularly discriminating in respect of music’s technical construction.Unmusikalisch refers to composers or compositions reckoned technically lacking(Busoni 1962, pp. 86–8 [translation modified]).

Passing from the dawn of musical modernism to postmodernism, we find – inplace of Busoni’s Germans – the British philosopher Roger Scruton. Like Busoni’sunmusikalisch, Scruton’s contrast between ‘sound’ and ‘tone’ opens up a domainthat is neither ‘musical’ nor ‘noisy’. Scruton refers on the one hand to discretesounds, which, though they follow one another in sequence, appear unconnectedone to the next; and on the other to sounds heard to possess some kind of innerforce or ‘tone’, such that they appear to be mutually responsive (Scruton 1997, pp.19–20). The distinction itself is unexceptionable. It becomes problematic in so faras Scruton associates music exclusively with tone. He has difficulties with post-tonal repertoire in general, and when he encounters compositions in which toneappears to have been thoroughly disaggregated into its components, he is at a loss.In the twelve-note melodies of Alban Berg, identities of rhythmical shape ‘canmake one set of pitches into the answering phrase required by almost any other’.But ‘when the rhythmic foreground is dissolved – as in the serialized rhythms ofLuigi Nono – the result is a kind of punctilious shapelessness in which, in theabsence of tonality, we search in vain for musical relations’ (Scruton 1997, p. 39).Nono’s twelve-note music is evidently unmusical.

Let us look more closely. The rhythms of the seventh movement of Nono’scantata Il canto sospeso (1955–6) are indeed serialised, as are the pitch classes (butnot their registration). Ex. 1a gives bars 457–461 as three-part counterpoint,without instrumentation, but including dynamics (restricted to ppp, p and mf,

DOI: 10.1111/musa.12042

Music Analysis, 34/i (2015) 91© 2015 The Author.Music Analysis © 2015 John Wiley & Sons Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UKand 350 Main Street, Malden, MA 02148, USA

but not serialised).2 Heard as they stand in Ex. 1a (especially played back via themusic-typesetting software used to create it), these bars confirm Scruton’s cri-tique. The pitches merely follow one another; they are not mutually responsive.But in Nono’s realisation (Ex. 1b), the effect is very different. Though Scrutonsets tone colour ‘to one side, as a secondary characteristic of the musical object’(p. 78), it is tone colour in Ex. 1b – along with a specifically non-musicalelement, the verbal text – that brings the notes to life. The G on ‘-glia’ at the endof bar 458 completes the setting of ‘figlia’ begun at the start of the bar: it is notan atomised sound, but descends from the G�, which in turn is linked to theprevious E. Similarly, the falling fifth at ‘Liubka’ (with its Puccini-style suddenhush on the high A) is precisely heard to fall. The A and D do not merely followin sequence – they respond to each other.

Scruton appears wrong to deny the existence of ‘musical relations’ in Nono.More importantly, it does not cross his mind that the contrast between ‘sound’and ‘tone’ might in fact be meaningful. Scruton wishes to provoke, notably in hisjudgement that ‘until we have furnished ourselves with an account of our centralinstances of the art’, the question of ‘whether this or that modernist orpostmodernist experiment is a work of music is empty’ (1997, p. 17). Yet fromthe perspective to be developed here, works of musical modernism will indeed betaken to be best understood in the light of their relations to historical precedents,with regard to which Scruton’s distinction between sound and tone names anessential characteristic.

It is not only the verbal text that lends the seventh movement of Il canto sospesoits celebrated poignancy.3 But when Massimo Mila praised the cantata for its‘capacity to institute a coherent continuity between one note and another’, hesurely did not mean to imply that a cantabile ran from the first bar to the last(Mila 1960, p. 309). As Helmut Lachenmann suggests, the linearity of the solosoprano’s contributions in the seventh movement stands in contrast to the‘punctual structure of dispersed lines’ that surrounds it (De Benedictis and

Ex. 1a Luigi Nono, Il canto sospeso (1955–6), bars 457–461, analytical reduction.© Ars-Viva-Verlag GmbH. Reproduced by permission of Schott Music Ltd.

92 BEN EARLE

Music Analysis, 34/i (2015)© 2015 The Author.Music Analysis © 2015 John Wiley & Sons Ltd

Mosch 2002, p. 253). We hear isolated moments of continuity, appropriatelysuggestive of subjectivity on the point of evanescence. Again this is to accept oneof Scruton’s principal claims: that in ‘the world of sound’, we ‘discover ...through music, the very life that is ours’ (1997, pp. 13–14). The point is that inNono this discovery is primarily negative. In the words of Theodor Adorno,Ex. 1b gives us a life that ‘does not live’ (1974, p. 19).

Nono’s work is characteristic of late modernism in this regard. The notion ofa music that places tone in expressive relation to sound is hardly unique to Ilcanto sospeso. Indeed, if Scruton’s equation of music with tone stands for therestricted musicality deplored by Busoni, then Nono’s placing of tone in relationto sound may be representative of musical modernism in general. As Jean-François Lyotard puts it, ‘When Cézanne picks up his paint-brush, what is atstake in painting is put into question; when Schönberg sits down at his piano,what is at stake in music; when Joyce grabs hold of his pen, what is at stake inliterature’ (1988, p. 139). In modernism, Lyotard suggests, the goal of art is nolonger that of ‘“pleasing” through the beautiful’, but of ‘“pleasing/displeasing”through the sublime’. A composition will be on the right track ‘if it obliges theaddressee to ask about what it consists in’ (ibid.). The claim that a work oftwelve-note music is musical – that it is in fact music – would thus inherently putinto question what is meant by these terms.

Ex. 1b Nono, Il canto sospeso, bars 457–461, short score. © Ars-Viva-VerlagGmbH. Reproduced by permission of Schott Music Ltd.

TWELVE-NOTE MUSIC AS MUSIC: AN ESSAY IN TWO PARTS 93

Music Analysis, 34/i (2015) © 2015 The Author.Music Analysis © 2015 John Wiley & Sons Ltd

Who Is ‘Musical’?

What they come down to is social class. Such is the conclusion to be drawn fromthe work of one writer particularly keen to bandy about a notion of the musical,the British critic Hans Keller. If commentators on music are unable to differen-tiate between analysis and description, Keller declares, this is a function of ‘thebasic issue ... of musicality versus unmusicality in our time’ (Keller 1977, p. 114).Musicality is an instinctive, emotional understanding. And to acquire it one hasto ‘grow up with, and into music’, and do so ‘in the right social context’ (pp.112–13). Some today have evidently been lucky enough to spend their formativeyears in adequately cultured environments. But ‘[t]he days are long past ... whenpeople were only open to what, potentially, they could understand’. In a societycatered to by mass media, stupid, talentless people (Keller’s adjectives) no longerknow their place. Neither they nor their populist masters will recognise ‘the factthat all men are not equal’ (pp. 115–16).

Keller writes from a position of unabashed elitism. Yet one cannot dismiss histhought as mere ideology. When Scruton points out that a musical culture exists‘by virtue of the surplus which creates the conditions of leisure’, and that ‘thedevelopment of our music ... depended ... on the bourgeois’, he is not wrongeither (Scruton 1997, p. 466). Let us look one last time at Ex. 1b. On what kindof knowledge does the assertion depend that sections of this music possess ‘tone’,or (which would seem to come to the same thing) a measure of subjectiveexpressivity? Is this the kind of knowledge imparted in university analysis semi-nars? The meaningfulness of Nono’s music would appear to depend on the sortof recognition detailed above. Yet as Scruton suggests, its ground is a knowledgeexpressed precisely ‘in acts of recognition’, and ‘not in theories’ (ibid., p. 18).

Pierre Bourdieu explains that students acquire cultural capital in two modes:the domestic as well as the scholastic. It is not just that the self-assured bourgeoislistener, who has been immersed in an atmosphere of high culture at home sinceearly childhood, will tend to feel more confident with difficult modernist worksthan the listener of working-class or petit bourgeois origins, who is less likely tohave been furnished with such a cultural head start and, being thus belatedlydependent on what is taught at school, will have ‘more “classical”, safer culturalinvestments’ (Bourdieu 1986, p. 65). As Bourdieu points out, a domesticallyacquired cultural capital in relation to music engenders a distinctive attitude. Hisexample is that of Roland Barthes, who assures us that the singing of hisfavourite, Charles Panzéra, was ‘exclusively bourgeois ... i.e., in no way petit-bourgeois’ (Barthes 1991, p. 273). The presence of ‘grain’ in Panzéra’s voice –a trace of the materiality of the French language in vocal production – testifies tothe fact that an amateur singer of the mélodies of Fauré or Duparc might aspireto perform them too (ibid., p. 262). By contrast, the grainless voice of thetechnically irreproachable Dietrich Fischer-Dieskau keeps amateurs at anunbridgeable distance. Fischer-Dieskau suits listeners who merely listen. His ‘art... corresponds perfectly to the requirements of an average culture’, which

94 BEN EARLE


reduces music to ‘what is said of it, predicatively, by the Academy, by Criticism,by Opinion’ (ibid., p. 273). It is a performance style suited to the primarilyscholastic mode of cultural acquisition typical of those lacking a bourgeoisformation (Bourdieu 1986, p. 76).

The above account of Ex. 1b, which leaves to one side the pedagogicallysanctioned route of note counting in favour of observations about aspects ofNono’s work not notated in the score, would seem fatally class marked, displayingnot just a confidence in handling difficult repertoire but also a presumed insider’sknowledge of how the music ‘goes’. Similarly, when Adorno, in a posthumouslypublished note from 1945, brandishes a notion of the musical, class distinctionswould seem clearly in play. Adorno takes issue with Busoni. The violinist whopraised as ‘so musical’ one of the Italian’s compositions was no fool: ‘There is, asit were independently of the musical quality, such a thing as speaking the languageof music or not. Bad composers such as Tchaikovsky, Puccini or Rachmaninovspeak it – Elgar or Sibelius do not’ (Adorno 2006a, p. 158).

Adorno here criticises a specifically British taste. Praise for Sibelius’s incom-petent music amounts to a destruction of ‘the criteria of musical quality thathave endured from Bach to Schoenberg’ (Adorno 2011, p. 336). If the Britishlike Sibelius, it is because they espouse a ‘Western’ aesthetic, which assessesmusic ‘according to the consistent exchange principle, which values any entityin terms of another [als ein Für anderes]’. That the work of Sibelius or Elgar isthreadbare or feeble is of no concern, since, unlike Central European listeners,the British are not interested in art for its own sake (Adorno 1976, pp. 172–3[translation modified]). Adorno may recognise the ideological character of the‘Central European’ position, its ‘fetishism of ... a man-made social product’(ibid., p. 173), but from the perspective of class, he simply transfers a distinc-tion normally regarded as operating within a single nation onto the relationshipbetween nations. The ‘aesthetic disposition’, an ability to concentrate onmatters of form alone, Bourdieu observes, ‘presupposes the distance from theworld ... which is the basis of the bourgeois experience’ (Bourdieu 1986, p.54). If this is the prerogative of Central European listeners, then Westernlisteners are reduced to petit bourgeois status, unable properly to separateworld and work.

But what does Adorno mean by ‘speaking the language of music’? Here, paceScruton, we may find a means to theorise ‘tone’ – or at least to identify somemodes of musical ‘movement’. Adorno once drew attention to bars 87–94 fromthe first movement of Brahms’s Fourth Symphony (1884–5) as they sound in the1951 recording by Arturo Toscanini and the NBC Symphony Orchestra:

In an extraordinarily lyrical eight-bar passage in B major a wind antecedentcontrasts with a consequent on the strings. The latter naturally has a richer soundthan the cooler wind antecedent. However, Toscanini hurls himself with suchenthusiasm onto the strings that the consequent ceases to be heard as thehalf-theme, the response, that it is in Brahms’s work. Instead, thanks to itsexaggerated and oversweet tone, it is turned into the main event and reduces the



exposition of the theme to a secondary issue, thereby throwing the entire formalpattern out of kilter. (Adorno 1999, p. 46)

The philosopher does not merely dislike what Toscanini makes of these bars. Hethinks the conductor’s reading objectively wrong. Toscanini demonstrates a lackof feeling for the syntax of Brahms’s work: his performance is unmusical. Adornogoes on to wonder whether the Italian ‘really had any understanding of phrasingat all’ and points out further instances of his failure to grasp the ‘musical sense’of various passages, or more strongly, to ‘do justice to the objective process ofcomposition’ (ibid., pp. 50–1).

Objectively Incomprehensible

Analysts of post-tonal music would do well to ponder these ideas. Adorno’sconception of the linguistic character of music is a theory of the phrase, anunderstanding of music as articulated at the kind of medium scale generallyneglected by analysts of this repertoire. Scruton furnishes a telling example. It isonly on the basis of his attenuated view of musical syntax as ‘a context-dependent affinity between tones’ (Scruton 1997, p. 186) that he can justifiablycomplain, in respect to bars 40–42 from the second movement of Hindemith’sConcerto for Trumpet, Bassoon and Strings (1949–52), of ‘each bar leadingsmoothly into its successor, yet the whole thing [being] a kind of nonsense’ (ibid.,p. 179). For Scruton, these bars illustrate a musical parallel to the ‘intransitivity’of verbal syntax, in which ‘[a] word may be acceptably joined to its successor,and the successor to its successor, and yet the result be ill-formed’ (ibid.).Something closer to the opposite of this would be more accurate, however, for atthe level of the phrase Hindemith’s music moves conventionally. It is the relationof one bar to the next, or one pitch to the next, that is more problematic.

Hindemith’s melodic line is ungainly. And while one may analyse the openingand closing harmonies of bars 40–46 in terms of ‘pillar chords’ in F� major,between these (particularly at bars 43–45) ‘the harmonic construction ... can besomewhat looser’, as David Neumeyer euphemistically puts it (Neumeyer 1986,p. 42). Yet Hindemith’s ‘wrong notes’ are part and parcel of his music’s mildlyparodic waltz character (see Ex. 2a). Scruton appears oblivious to this aspect ofthe extract. More significantly, his choice to cite only bars 40–42 means that hecannot show how they form part of a larger structure. The way in which bar 41answers bar 40 is unmistakable and presumably accounts for Scruton’s ‘each barleading smoothly into its successor’. But to grasp the musical sense of bars40–46, we need first to take account of bars 32–39, a Schoenbergian sentence.4

A look at the melodic line will suffice (Ex. 2b). At bar 34 the opening two-barunit is set to repeat, but (with possibly humorous intent) the first violins get stuckon the material of bar 32, before Hindemith moves to a more developmentalthree-bar continuation at bars 36–38, followed by a cadence in bar 39 (basedonce again on the material of bar 32). With the repetition of bars 32–39 at bars

96 BEN EARLE


40–46, various details of scoring, dynamics, registration and harmony arechanged. The second repetition of the phrase’s opening bar, due in bar 43 as aparallel to bar 35, does not reappear, hence the seven-bar phrase. But the pointis that bars 40–46 are heard precisely as a (varied) repetition.

Ex. 2a Paul Hindemith, Concerto for Trumpet, Bassoon and Strings (1949–52),second movement, bars 40–46, short score. © Schott Music GmbH & Co. KG.Reproduced by permission of Schott Music Ltd.

Ex. 2b Hindemith, Concerto for Trumpet, Bassoon and Strings, second movement,bars 31–39, melody only. © Schott Music GmbH & Co. KG. Reproduced bypermission of Schott Music Ltd.



For Adorno, stylistically conservative music like this could only have con-firmed Hindemith’s ‘petit bourgeois’ social character (1970–86, vol. 17, pp.239–46). Nevertheless, the analysis sketched above may be taken as movingtowards the ‘objective’ sense of this extract. The word is meant, of course, toraise eyebrows. For Julian Horton,

[t]he fact that the cognitive strategies underpinning perception may be rooted intonal paradigms tells us something about the cultural pervasiveness of thoseparadigms, but very little about the objective basis of musical perception. Weshould not, in other words, confuse cultural memory with cognitive objectivity; todo so is to succumb to a kind of musical false consciousness. Such arguments tendto mistake subjective analytical engagements for objective judgemental criteria.(2005, p. 256)

To make the Adornian judgement of a piece’s ‘objective incomprehensibility’ –one evidently not tenable in respect of Ex. 2 – is to fall into error, Horton claims,unless the reading can be accepted as ‘universally true’ (ibid.). Adorno mighthave permitted himself a wry smile. As he puts it, ‘the reproach that a statementis “too subjective”’ is standard in such cases. ‘If this is brought to bear’, hecontinues, ‘with an indignation in which rings the furious harmony of all rea-sonable people, one has grounds, for a few seconds, to feel self-satisfied’ (1974,p. 29). For Horton, an objective statement is one uncoloured by subjectiveidiosyncrasy; it may thus be expected to receive universal assent. But for Adornoit is a feature of the contemporary world that ‘[t]he notions of subjective andobjective have been completely reversed’:

Objective means the non-controversial aspect of things, their unquestionedimpression, the façade made up of classified data, that is, the subjective; and theycall subjective anything which breaches that façade, engages the specific experi-ence of a matter, casts off all ready-made judgements and substitutes relatednessto the object for the majority consensus of those who do not even look at it, letalone think about it – that is the objective. (Ibid., pp. 69–70)

Yet more snobbery, one might retort. Adorno’s position is certainly high-handed. In philosophical language, it is ‘transcendental’. As Brian O’Connorexplains, the negative dialectic involves the claim that a philosophy that ‘deniesor excludes any of the conditions identified [by Adorno] as the conditions for thepossibility of experience’ is doomed to incoherence (O’Connor 2004, p. 15).Genuine experience, the recuperation of which may be taken as the central aimof Adorno’s life’s-work, is structured as ‘mediation’. It is a reciprocal, dynamicyet ‘indetermining’ relation between subject and object (ibid., p. 18). Famously,the philosopher insists on the ‘primacy of the object’, the claim that objects ingeneral cannot be fully subsumed under concepts. At the same time, he is clearthat the experiencing subject is not merely passive. An object may be determi-native, ‘but its determination is articulated by the subject’ (ibid., p. 73). Genuineexperience involves openness to the necessity of conceptual adjustment and

98 BEN EARLE


awareness of the inadequacy of our concepts in relation to reality. Objectiveknowledge is indeed possible, but ‘this ... will be a nontotalistic knowledge thatexcludes the assumption that knowledge means total encapsulation by thesubject of the object’ (ibid., pp. 4–5).

Adorno sets himself on a collision course with ‘positivistic’ modern ration-ality, which he considers ‘isomorphic with the economic structures of [latecapitalist] society’ (O’Connor 2004, p. 3). The kind of writing that takesobjectivity to equate with success in classification remains subjective inasmuchas it is involved in the business of fitting objects into schemas rather thanattending to their particularity. By the same token, complaints about the sub-jective character of what Adorno would regard as objective judgements miss thepoint that genuine experience has a mediatory structure. That an element ofsubjectivity is involved in genuine aesthetic judgement by no means impliesrelativism:

Anyone who, drawing on the strength of his precise reaction to a work of art, hasever subjected himself in earnest to its discipline, to its immanent formal law, thecompulsion of its structure, will find that objections to the merely subjectivequality of his experience vanish like a pitiful illusion: and every step that he takes,by virtue of his highly subjective innervations, towards the heart of the matter, hasincomparably greater force than the comprehensive and fully backed-up analysesof such things as ‘style’, whose claims to scientific status are made at the expenseof such experience. (Adorno 1974, p. 70)

As O’Connor explains, crucial to Adorno’s aesthetics is the idea that, in theirresponses to artworks, audiences may experience the possibility of a non-instrumental engagement with the world. What Adorno calls ‘mimesis’ is anaffective relation to the other, in which, rather than attempting to dominateobjects, audiences try to imitate them (O’Connor 2013, pp. 149–54). Adornoopposed the separation of knowledge and affect: ‘Once the last trace of emotionhas been eradicated, nothing remains of thought but absolute tautology’. Thekind of perception that belongs to genuine experience maintains a degree of‘anticipatory desire’ based in memory. Rather than tabooing memory as‘unpredictable, unreliable, irrational’, Adorno insists that consciousnessneeds a historical dimension if it is to be able to ‘establish that relationbetween objects that is the irrevocable source of all judgement’ (Adorno 1974,pp. 122–3).

The three-level model Adorno puts forth in his posthumous notes on perfor-mance practice is informed by this same kind of thinking. Objectivity in perfor-mance requires subjective intervention. What is notated (the ‘mensural’ level)has to become articulated, which is to say, meaningful, or ‘gestural’ (this is the‘neumic’ level), by way of the player’s spontaneity. But spontaneity is notenough. The analytical activity which must precede adequate performance needsto be guided by awareness of the unnotated context of interpretation (the levelof the ‘tone-lingual’). The latter is the presence of history in musical language.



To speak the language of music – to be capable of adequate performance (orindeed composition or listening) – is to be able to anticipate what should comenext (Adorno 2006a, passim).

Horton’s dismissal of cultural memory, and his concomitant attempt to dis-sociate aesthetic judgement from historical understanding, are at one with hiscommitment to that most positivistic of music-analytical tools, Allen Forte’stheory of pitch-class sets. Faced with Berthold Hoeckner’s reading ofSchoenberg’s Die Jakobsleiter (Hoeckner 2002, pp. 189–223), impeccablyAdornian in its intermingling of constructive and expressive elements (the sig-nificance of which resides precisely in their irreducibility one to another), Hortonbrings things down to the facts. Such considerations ‘really reduce to a mixtureof thematic cross-referencing and the heterophonic presentation of set-class6–Z13’ (2005, p. 256).

Back to Schoenberg

It is depressing to think that, despite the disciplinary self-criticism of the 1990s,such attitudes are still abroad. Yet it is not hard to locate their source. Here is thelate Milton Babbitt speaking ‘to a class of college sophomores’ about the openingof Schoenberg’s Fourth Quartet, Op. 37 (1936). He presents the first violin lineof bars 1–6 (see Ex. 3), stripped of rhythm, dynamics and articulation, andcomments:

After stating his first hexachord, which he articulates very obviously,[Schoenberg] then lays out the next six notes of his set in a particular registraldistribution which is the only one out of the six factorial (6!) registral distributionswhich will define a particular relationship between the two hexachords which isfundamental to the structure of the piece, the counterpoint of the piece, theprogression of the piece. (Babbitt 1987, p. 23)

He continues:

Do you see what Schoenberg has done? He’s made it clear ... that the secondcollection is an inversion of the first collection, which is by no means trivial. By sodistributing it in register, he’s telling you that there’s going to be a particularrelationship between the hexachord and its inversion which will create aggregates,that is, complete twelve-tone collections in the segments of the two sets will be soused. (Ibid.)

For Babbitt, ‘[T]his is the way Schoenberg’s music goes. It’s not the only way itgoes; it’s one of the many ways it goes’ (ibid.). But no, one wants to reply. Thisis not even one of the ways that Schoenberg’s music ‘goes’. Babbitt regards hisexample as an instance of ‘directed motion’ – namely, ‘the idea of constantlyintimating where you are going ... so that when you arrive in certain places in thepiece, you have the sense of having gotten someplace which has already beenpredicted’ (ibid.). When the Fourth Quartet gets to the upbeat to bar 7, we are

100 BEN EARLE


Ex. 3 Arnold Schoenberg, String Quartet No. 4, Op. 37 (1936), bars 1–31. Copy-right © 1939 by G. Schirmer, Inc. (ASCAP). All rights reserved. InternationalCopyright Secured. Used by permission



Ex. 3 Continued

102 BEN EARLE


Ex. 3 Continued



to hear the second violin’s transposed inversion of the first violin’s pitches in bars1–6 as having been prepared by the registral distribution Babbitt has isolated.But again, no. As Robert Pascall has pointed out, bars 63–9 relate to bars 1–62

as contrasting idea to basic idea within the antecedent of a period (Pascall 2002,p. 238). Schoenberg’s twelve-note music is music.

It is telling that Arnold Whittall, praising Babbitt’s analytical exposition as‘masterly in its musicality’, does so in relation to his comments on tonal music(Whittall 1988, p. 421): repertoire – to employ Babbitt’s terminology – in whichaudiences recognise ‘communal’ elements from work to work (those of tradition)which are allegedly not present (or insufficiently present) in ‘contextual’ (atonaland twelve-note) compositions, where formal principles hold just for the indi-vidual work in question (Babbitt 1987, pp. 9–10). This is the crux, the point atwhich anxieties about social snobbery find their focus. In a recent book, MichielSchuijer not only demonstrates the historical derivation of Forte’s method fromthat of Babbitt, but also argues in favour of pitch-class set theory from asomewhat politicised perspective: ‘Whoever rejects its formalism andreductionalism, and finds its language inappropriate, should bear in mind thatthese aspects serve the purpose of equal opportunity. They enable anybody whois interested, and perseveres to develop an expertise in some of the art music ofthe post-1900 era and to hand it down to others’ (Schuijer 2008, p. 277).5 Tocomplain about the unmusicality of this approach, so well-suited to the acqui-sition of cultural capital in its scholastic mode, is merely to enact a classdistinction.

But this cannot be the end of the story. Taking issue with an intemperatereview of his 1982 book The Harmonic Organisation of ‘The Rite of Spring’ by theBritish composer Anthony Milner, in which he found himself vilified for his useof a computer, Forte expressed amazement:‘[s]urely everyone knows that manypeople now routinely work with computing devices’ (1985, p. 46). In a footnote,Forte explains that ‘although I have written many computer programs in con-nection with analytical and theoretical studies’, the book in question ‘wasexecuted entirely by hand methods, save for an electric typewriter and a high-tech electric eraser’ (ibid., p. 58 n. 42). Nevertheless, it should not be forgottenthat pitch-class set theory ‘was actually devised for computer-aided analysis’(Schuijer 2008, p. 237): ‘“PC set”, “normal” or “prime form”, and “intervalvector” have become household concepts for many music theorists, but in themid-1960s they met a very specific need: the need for definitions of musicalrelations that a computer could recognize’ (p. 248). ‘So what?’, one can imagineForte replying. In 1985 he was already welcoming ‘the advent of the microcom-puter’, envisioning ‘a powerful set-complex analyser with artificial intelligenceaspects’ (Forte 1985, pp. 55–6). The basis to the kind of objection made byMilner needs to be carefully spelled out.

Nicholas Cook has invoked the experience of listening ‘to a synthesizedperformance of Chopin’s E minor Prelude, in which every note is equally longand equally loud’. As Cook explains, the effect of the synthesiser’s performance

104 BEN EARLE


would be to make us negatively aware of how much the effect of a goodperformance relies on factors of ‘temporal and dynamic shaping’ that are notnotated in the score (Cook 1998, p. 61). In an earlier account of this sameexample, the synthesiser was a Martian. Its hypothetical present-day audience,Cook suggested, ‘would be inclined to think of someone who played Chopin’smusic in this manner, as being not so much unmusical as mentally deranged’(1990, p. 122).

Consider a class of undergraduates learning to analyse post-tonal music. Forall the democratic credentials of pitch-class set theory, in this environment thedistinction between domestic and scholastic acquisition of cultural capital mayhave little purchase. Most, if not all, of these students will be (or have been)instrumentalists or vocalists, studying with teachers who will be trying (or havetried) to cultivate in them a sense of where the unnotated factors observed byCook would be appropriately employed. The ability to do this well, Cooksuggests, ‘is a major part of what being a good pianist means’ (1998, p. 61). Itis what musicians in the Western tradition call ‘musicality’. So the question toBabbitt, Forte and their disciples is: Why, when we expect our students to playmusic – including that of the Second Viennese School – like human beings, dowe ask them to analyse its post-tonal manifestations as if they were Martians?Why not encourage them to bring their musical experience to the exercise?

Schuijer is no uncritical partisan. He freely admits the theory’s positivisticbasis and argumentative circularity (2008, pp. 250, 89 and 270). But against thecomplaint that one cannot hear pitch-class relations, that they have no bearingon musical experience, he mounts a determined defence. The crucial issue, ofcourse, is segmentation. In the absence of criteria for the ‘verification’ of pitch-class sets, analysts have tended to employ their musical intuition. Yet whenpitch-class set analysis carried out in this manner is apparently successful and apreviously unsuspected relation of equivalence is discovered between what, froma ‘musical’ perspective, would seem two self-evidently isolatable units, Schuijerremains unconvinced. ‘Intuitive’ or ‘hand-made’ segmentation, while appearingas a humanising correction to the theory, stands in tension with the ‘scientificspirit’ in which it was conceived (ibid., p. 24). To suggest that listeners ought torecognise that their intuitive grasp of the coherence of certain works of musicrests on what are for them as yet undiscovered pitch-class set relations may be tomisunderstand the project entirely. For why should a theory be based on listen-ers’ expectations? ‘PC set theory is not a theory of music cognition ... . As aconsequence it should not be judged from this viewpoint’ (ibid., p. 269).

Forte’s labels may sometimes have a certain heuristic usefulness, but in so faras we are concerned with twelve-note music as music, pitch-class set theoryshould detain us no further. Isolating and comparing elements independently ofthe ‘movement’ expressed by a composition, Scruton observes, the theory analy-ses not music, but sounds (1997, p. 415). But how shall we proceed instead?Scruton, for one, is an unregenerate tonalist, though what he means by ‘tonality’can appear obscure, since the term is explicitly able to encompass Luigi



Dallapiccola’s twelve-note opera Il prigioniero (Florence, 1950), ‘whose tonalorder is constituted entirely on the surface and in defiance of the traditional lawsof harmony’. It turns out that, for Scruton, ‘tonality’ is synonymous with musicalsense. ‘Tonality’ permits us to hear music as music: ‘as moving with the forceand logic of gestures which are mutually aware and mutually accommodating’(ibid., pp. 271–2). As the example of Il prigioniero suggests, atonal music may alsobe ‘tonal’.

Scruton finds Schoenberg’s twelve-note music more problematic. Of theViolin Concerto, Op. 36 (1934–6), he insists that ‘[w]hen we hear movement inatonal music, it is precisely not the serial ordering that we are hearing’ (1997, p.304). That he finds it necessary to make such an elementary observation speaksvolumes about the manner in which this repertoire has tended to be discussed inEnglish. But just how is Schoenberg’s twelve-note music musical? It is strikinghow rarely this question has been asked.6 For Scruton, the opening of theConcerto ‘moves through tonal space with ... the logic of a tune, meditating onneighbouring semitones’. Attending to this music as music, the listener ‘will berelying on the tonal implications of the ascending and descending semitone – theperceived character of leaning that these intervals derive from the tradition oftonal harmony’ (ibid., pp. 304 and 305). The analytical problem here is akin tothat encountered above in respect to Hindemith. Ironically, it is shared withthose US theorists of twelve-note music against whom Scruton takes himself tobe arguing. As Christopher Wintle explains, ‘form and phrase in Second Vien-nese School music’ is ‘a topic that never really crossed the Atlantic when itmattered’ (2003, p. 294). The term ‘phrase’ is one Babbitt refuses to employ(Babbitt 1987, p. 138).

Schoenberg as Music

It is pitiful to have to observe that, despite the rivers of ink since spilt on thisrepertoire, one of the most musical accounts in English of Schoenberg’s twelve-note music remains the analyses of passages from the Piano Concerto, Op. 42(1942), published by Heinrich Jalowetz 70 years ago. Having stressed the over-riding importance of ‘[t]he melodic element’ as ‘the basic force’ of Schoenberg’smusic, Jalowetz proceeds to phrase analyses of bars 1–39 and 264–285, theopenings of the first and third movements. In the case of the latter, Jalowetz tellsus, ‘directly audible ... interrelations ... arise from the operation of the primarylaws inherent in the language of music.’ This repertoire may be discussed‘according to methods that may be regarded as universally valid’ (Jalowetz 1944,pp. 401–2).

Few today would have such confidence. As Henry Klumpenhouwer observes,among the features of ‘Adorno-style analytical studies many music theorists willfind troubling’ is an evasion of ‘any discussion of analytical methodologicalissues’ (2001, p. 392). In the case of work by Alastair Williams (1997), ‘thoseaccustomed to dealing with serial theory, set theory or generalised interval

106 BEN EARLE


systems’ will find the analyses presented ‘technically naïve or even impression-istic’ (Klumpenhouwer 2001, pp. 390–1). In response, let us consider Ex. 3again. Wintle’s comments on ‘phrasing and form’ derive from a report on ananalytical lecture in which Alexander Goehr, while apparently followingSchoenberg in referring to the opening theme of the Quartet as a two-partstructure (bars 1–6 and 7–9),7 also took ‘the theme’ to run to bar 16. Wintleagrees with this revision of what is purportedly Schoenberg’s own analysis,‘for the “proper” first subject’, he avers, ‘falls clearly into a small ternary form’(with the repeat at bar 10) (2003, pp. 294–5). Babbitt too hears bar 162 as ‘theend of the first section of the piece’ (1987, p. 73), while Joel Lester anticipatesWintle’s ‘ternary’ reading (Lester 1989, pp. 192–3). Yet none of these commen-tators give much sense of how one is to grasp the music of bars 16–31.8

One solution is provided by Pascall. As we saw, he reads the opening themeas a period. There is an antecedent at bars 1–9 (basic idea, bars 1–62; contrastingidea, bars 63–9) and a consequent at bars 10–24 (basic idea, bars 10–163;contrasting idea, bars 164–24). Pascall gives two reasons why bars 164–24‘should become perceptually attached to the preceding music’. They ‘offer avariation of bars 63–9 together with the four-eighth note + quarter note figurefrom bar 21–3 [in the upper strings at bars 173–181]’ (Pascall 2002, pp. 238–9 n.32). And they employ the same row form as the previous contrasting idea. Butagainst Pascall’s reading one must set the cadential quality of bars 14–16: a‘musical “full stop”’, as Wintle describes it, followed by a ‘brief silence that ...puts the structural division beyond doubt’ (Wintle 2003, p. 295).

To sit on the fence of formal ambiguity would be to conceal the real difficul-ties of bars 16–31. A division at bar 25 is clear: Adorno calls this a ‘secondarytheme’ (2006b, p. 79).9 The question of what to do with bars 164–24 remains.Proponents of the ternary reading will presumably speak of a transition, perhapsdivided at bar 212. The periodic reading does not fare much better. Pascall tooquickly labels the passage ‘a variation of bars 63–9’ (2002, p. 239 n. 32), since thecello line at bars 16–212, the Hauptstimme at this point, refers explicitly to bars63–9 only in its final descending fourth. The cello’s gesture in bar 17 is morelikely to remind listeners of the first violin’s music at bar 10, while its ensuingmusic (up to the E�) is not obviously related to any of the previous melodic lines,none of which share its dotted rhythms. As for bars 212–24, they simply do notfit under Pascall’s rubric. Their function is that of a transition; in Schoenbergianlanguage a ‘liquidating’ transition that transforms its ‘motival characteristics’(the triplets of the imitative entries) into the accompanying lines of bar 25onwards (Schoenberg 1967, p. 179). Even this process is far from straight-forward. The forte interruption of the viola in bar 24 may be easy to explainmotivically. But what is its musical sense? Does it anticipate the crotchet tripletsof bars 303–312?

Despite its references to traditional style – Wintle speaks of a ‘Mahlerian-march type’ (2003, p. 296) – the continuity of Schoenberg’s music is fracturedand sometimes very hard to grasp. As Wintle continues, ‘Goehr is certainly more



right to talk of an ill-at-ease quality than are all the positivistic analysts who faileven to sense it’ (ibid., p. 297). This instance of high modernism makes aninstructive contrast with the late modernist Nono of Ex. 1b. Whereas in Il cantosospeso the listener clings to moments of melodic continuity that emerge unpre-dictably from the otherwise pointillistic texture, in the first movement of theFourth Quartet one has the sense of a musical discourse poised on the edge ofincomprehensibility, liable to collapse at any moment ‘from the intentional to thematerial realm’, as Scruton has it: from continuity to discontinuity, tone tosound (Scruton 1997, p. 281).

Klumpenhouwer rejects as fruitless the Adornian dismissal of ‘certain domi-nant uses of music-analytical methodology’ (2001, p. 391). If pitch-class setanalysis does indeed constitute an ‘obvious’ expression of ‘identity thinking’,which ‘falsely annuls all contradictions, oppositions, remainders and differences,smudging the distinction between concepts on the one hand and their referentson the other’, the same must be said of any other analytical method, for ‘identitythinking, as Adorno reminds us, is fundamental to all styles of consciousness andcognition under capitalism’ (ibid., pp. 388 and 391). Klumpenhouwer’s positionis not just a counsel of despair; in its failure to refer to ‘mimesis’, it is amisrepresentation. As he recognises from his reading of Shierry Weber Nichol-son, genuine aesthetic experience involves the interaction of subjective andobjective poles (Nicholsen 1997, pp. 1–58). Indeed, Klumpenhouwer eloquentlyexplains why, from an Adornian perspective, the procedures of ‘contemporarymusic analysis’ need to be jettisoned, bound as they are to formalism as to a‘creation myth’. For ‘traditional formalism is reliant on an absolute discontinuitybetween aesthetic subjects and objects, along with commitment exclusively toabandon the realm of the subject in favour of investigations of the logic thatcontrols the object realm’ (Klumpenhouwer 2001, pp. 393–4). But whatKlumpenhouwer takes as objective is, for Adorno, nothing of the sort.

Mimesis

For Adorno, music speaks. It is articulated in a manner analogous to thestructures of verbal language. This is ‘a universal characteristic of music’(Adorno 1998, p. 1). Like Jalowetz, Adorno displays an apparently recklessabsolutism. Only by overshooting the object can thought be free to determine it.There can be no possibility of relativism, which equates to bourgeois positivism:not just a refusal of experience (a separation of subject and object), but a ‘bad’absolutism, a jealous guarding of scraps of knowledge as if they were possessions(see Adorno 1974, pp. 126–8 and 2006a, p. 54). The speech character of theopening bars of Schoenberg’s Fourth Quartet is objective: not a construction,but something that imposes itself on the experiencing subject. It may be too easyto dismiss as ‘unmusical’ those analysts who disregard this aspect of Op. 37. Thework of art is not timeless, after all, but ‘unfolds’ historically (see Adorno 2006a,pp. 194–5). Yet if the standard positivistic approach to the Fourth Quartet

108 BEN EARLE


genuinely reflects its practitioners’ experience of this music (which is difficult tobelieve), one can only concur with Silvina Milstein’s diagnosis of an ‘abyssseparating the post-War generation from Schoenberg’ (Milstein 1992, p. xii).

The twentieth century, so the art historian T. J. Clark has suggested, was ‘feltby artists and intellectuals ... to be ... the end of something called bourgeoissociety’ (Clark 2013, pp. 16–17). Both Keller and Barthes express this veryclearly; one might characterise Adorno’s entire music-critical corpus – not tospeak of such texts as Dialektik der Aufklärung and Minima Moralia – as ameditation on this same phenomenon. Pitch-class set analysis shows us what themusic of the twentieth century would sound like once all traces of bourgeoisculture had definitively disappeared. The very essence of musical modernism’sself-understanding lay in its complex engagement with the bourgeois canon ofthe eighteenth and nineteenth centuries. Unable to recognise this, Forte’smethod treats its object as something archaic, alien and incomprehensible, notjust aesthetically but also historically. For the bourgeois subject, as Barthesvividly illustrates, the musical work of art was, above all, a means to identifica-tion (1991, pp. 261–2). By contrast, Schoenberg’s music – so Adorno has it –‘becomes highly allergic ... to every tendency to ingratiate itself with the listeneras well as the latter’s efforts to ingratiate himself with it, to all identification andempathy’ (1967, p. 158). A ‘scientific’ approach might thus seem to be invited.But this opposition is too polarised. As Clark recognises, the ‘deepest and mostpersistent note’ in modernism belongs to those artists who recoiled from a belief‘in some version of modernity’s movement forward, toward rationality or trans-parency or full disenchantment’ (2013, pp. 14 and 19). If the first movement ofthe String Quartet No. 3, Op. 30 (1927), ‘achieves a musical cubism’, operating‘with completely rigid, pure, and in a sense geometrical symmetries’,Schoenberg’s American twelve-note works ‘pose the question of how construc-tion can become expression’ (Adorno 2006b, p. 77).

The thematic elements in the first movement of Schoenberg’s Fourth Quartetrelate stylistically to the composer’s pre-atonal music, especially the FirstChamber Symphony, Op. 9 (1906) (ibid., p. 80). Yet Adorno’s understanding ofthis music is not restricted to its ‘surface’ characteristics. Ethan Haimo, buildingon the work of Babbitt, David Lewin, Martha Hyde, Andrew Mead and others,observes how ‘[b]y Op. 31 [the Variations for Orchestra of 1928] the set hadbecome a referential idea, not merely for local melodic figures, but for theharmony, the form, the metre, as well as, in specialized situations, middlegroundmelodic control’ (Haimo 1990, p. 41). Schoenberg had ‘transformed’ the serialidea ‘until it embraced every dimension of the musical fabric’ (ibid., p. viii). ForAdorno, this had been the goal all along. He locates the origin of twelve-notetechnique in a search for ‘a common denominator for all musical elements’. Theproblem is that ‘this unification threatens to undermine every single musicaldimension.’ Indeed, ‘[t]he failure of the technical artwork can be confirmed in alldimensions of its composition’ (Adorno 2006b, pp. 45 [translation modified], 65[translation modified] and 57).



Adorno’s sociological reading of the American Schoenberg is well known. Theheroic failure of the composer’s attempt to instil the empty clatter of the twelve-note mechanism with expressivity produces an irreconcilable ‘struggle betweenalienated objectivity and limited subjectivity’ that is its ‘truth’ (Adorno 2006b, p.81). In contrast to the administered society of late capitalism, of which twelve-notetechnique is a formal analogue, Schoenberg’s later music is to be praised for notsuppressing its antagonisms (ibid., p. 75). Less widely appreciated – in musico-logical circles, at least – is the more philosophical lesson Adorno draws here. If thequality of ‘speaking’ is objectively present in the later Schoenberg, the process bywhich the subject comes to appreciate this is by no means passive. For it is notconceptless. Without the terms ‘phrase’, ‘sentence’, ‘period’ and so forth, wecannot communicate a sense of how this music goes. Nor is this yet ‘mimesis’.Adorno’s use of the term is negative. It is precisely in the moments of falteringsyntax identified above, when our concepts are resisted by the very music thatsuggested their use, that we genuinely encounter the work of art. Shocked by thesedepartures from musical convention, listeners find themselves embodying a modeof subjectivity that is receptive rather than classificatory. Such moments are ofmetaphysical significance; they constitute ‘real experiences’, ‘true happiness’,even (O’Connor 2013, p. 170). Adorno characterises the return of experience ‘inthe rebellion of music against its own meaning’ as a mythic rebirth: ‘The man whosurrenders to tears in music that no longer resembles him at the same time allowsthe stream of what he himself is not – what was dammed up back in the world ofthings – to flow back into him.’ A music that, in its flashes of objective incompre-hensibility, reaches beyond ‘meaning and subjectivity’, uncovers an archaic levelof experience that betokens reconciliation (Adorno 2006b, pp. 98–9).

II. Analysis

In his sociological work, Adorno can appear to occupy two incompatible posi-tions: on the one hand, the forbidding theorist of the inhuman, administeredsociety; on the other, the lucid public intellectual urging West Germans tocultivate autonomous individuality (O’Connor 2013, pp. 13 and 130–5). Asimilar contrast can be found in his writings on the Second Viennese School.Adorno is at once the philosopher of redemption in music beyond sense and ‘thefaithful repetiteur’ – Der getreue Korrepetitor – encouraging performers towards amore musically meaningful phrasing (see Adorno 1970–86, vol. 15, pp. 157–402). If this music is poorly played, moments of faltering syntax will be indis-tinguishable from any other. Indeed, Adorno declares, ‘senselessness in thepresentation of new music is almost universal’ (2006a, p. 122). Take the openingof Schoenberg’s Five Orchestral Pieces, Op. 16 (1909), conducted by RafaelKubelik. In the six bars shown in Ex. 4a, Adorno writes,

The three-bar main theme is followed by a dissolved, also three-bar consequentthat is extremely opaque [ein äußerst schwer durchzuhörender Nachsatz]. The

110 BEN EARLE


dynamics are evidently indicated ‘subjectively’, i.e. according to the playingtechniques of the instruments, e.g. an accompanying trill in two flutes is f, againstwhich the principal voice in the clarinet is p; equally a distinct middle voice inthe horn is p, a non-melodic lower voice in the bassoon f. The final bar isa solo rhythm played by a muted horn, marked f. R. Kubelik follows theindications most obediently in the recording. But because the diving clarinetfigure, which leads into the horn rhythm, genuinely comes out p in its awkwardregister, a dynamic hole results between it and the muted horn, despite thegood ‘connection’; as a result of the unmediated dynamic difference, onecan no longer perceive that rhythm as what it is, namely the melodic continu-ation of the clarinet. Thus the sense of the entire passage, which is delicateenough in any case, becomes incomprehensible – and at the same time thatof the entire exposition, which depends on the relationship between anteced-ent and consequent. The only thing that might help would be dynamic retouching,i.e. to have the clarinet play loudly enough for the horn to follow on seamlesslyfrom it. But exactly this – contrary to the letter of the notation – presupposesanalysis; the diving demisemiquavers in the clarinet are thematic (from thecounterpoint to the main theme). And it is precisely this step that was not takenby the musical, faithfully vigilant conductor. (ibid., pp. 146–7 [translationmodified])10

Ex. 4a Arnold Schoenberg, Five Orchestral Pieces, Op. 16 No. 1 (1909), bars 1–6,full score. © C. F. Peters Leipzig, 1912. Reprinted by kind permission of PetersEdition Limited, London



When Babbitt looks at this same passage, his concern is to demonstrate howSchoenberg is ‘composing with the tones of the motive’ (Babbitt 1987, pp.157–8). The penultimate sentence of the extract indicates that Adorno too takes‘analysis’ to involve microscopic attention. Schoenberg himself used to analysemusic at this level, of course, yet ‘his actual Formenlehre’, as William E. Caplin putsit, ‘has little connection to melody and motive’ (1998, pp. 259–60 n. 12). In atonalrather than tonal repertoire, such a distinction is harder to maintain. Many of theconventions on which traditional Formenlehre depends, most obviously the inter-dependence of formal function and tonal harmony, are no longer operative. In theprototypically Schoenbergian opening bars of the early song ‘Lockung’, Op. 6,No. 7 (1905), Adorno writes, the apparently unrelated ‘groups’ of material are tobe understood in terms of a ‘general unity of motivic-thematic relations’. But atthe same time, he suggests, ‘the groups are also syntactically linked’ (Adorno1967, pp. 153–4). The latter point is crucial. The present argument rests on theconviction that when it comes to the sense of Schoenberg’s post-tonal music, thekind of Formenlehre preoccupations Adorno demonstrates in respect to Op. 16 No.1 are of greater importance than any motivic links.

One may quibble with Adorno’s ‘antecedent’ and ‘consequent’, since there islittle sense of periodic construction in Ex. 4a. These bars nevertheless fall into twothree-bar units, of which the second stands in gestural response to the more lyricalfirst. As Adorno suggests, this contrast informs ‘the entire exposition’ (bars 1–25),given in Ex. 4b in short score. The pair of materials appears three more times: firstat bars 7–14 (dividing after bar 9), then at bars 15–22 (dividing after bar 19). Thethird appearance is less obvious, beginning simultaneously with the end of thesecond (bar 22) and lasting until the end of the example (dividing after bar 23).And it is not just the substance of such an approach that is important, but itspotential audience. When Adorno claims that Webern’s Six Bagatelles, Op. 9(1909), cannot be meaningfully presented in the absence of prior analysis (Adorno2002, p. 168), his intention is not that the members of any string quartet braveenough to tackle the work should launch themselves on a hunt for some abstractpitch-class set unity. Analysts working in the tradition of Babbitt and Forte speakonly to each other. But phrase analysis could benefit readers outside as well asinside the academy. It is directed primarily at performers.

Forward to Skalkottas

We do not have to take Adorno’s word for it that analysis is necessary inpost-tonal repertoire. Evidence is abundantly available in many of the recentlyreleased recordings of music by Schoenberg’s pupil Nikos Skalkottas (1904–1949), whose twelve-note compositions are the object of a major recent study byEva Mantzourani. After decades of neglect by record companies, today a rep-resentative selection of Skalkottas’s music in all genres is ready to hand. It is amixed blessing. To listen to the Malmö Symphony Orchestra under NikosChristodoulou in the 32-minute Largo sinfonico is to wonder why anyone should

112 BEN EARLE


have bothered with it in the first place.11 From performances like this one learnsexactly what it means not to speak the language of music. Skalkottas wasevidently a monomaniacal dilettante, sitting up night after night in an indifferentAthens to notate immense scores according to esoteric schemes (in the case ofthe Largo sinfonico, an ordered superset of sixteen distinct 12-note sets) – com-positions that give the visual impression of music but in reality are mere frame-works from which sense is absent.

Ex. 4b Schoenberg, Five Orchestral Pieces, No. 1, bars 1–25, reduction. © 1922 C.F. Peters. Reprinted by kind permission of Peters Edition Limited, London



Listeners will need to look beyond the BIS recordings to discover how impres-sive Skalkottas’s music can sound.12 But this unfortunate situation ought topresent an opportunity. Analysis can demonstrate the musical character ofSkalkottas’s music, the basis on which it might be made to speak. Unhappily,however, there is little in Mantzourani’s book that will be of help in this regard.The most successful section is the opening 80-page ‘biographical study’. In anunsparing account, Mantzourani shows that if the story of Skalkottas’s life was‘largely ... lonely, depressing, difficult and ultimately tragic’, the composershould nonetheless be seen as ‘more the architect of his own misfortune than onewho suffered unreasonably from life’s vicissitudes’. For Skalkottas, the rapid andfurtive composition of an extensive body of music, most of which remainedunpublished and unperformed during his lifetime, could be viewed ‘almost as anact of resistance to the vulgarity and constraints of the despised environment thathe felt surrounded him’ (Mantzourani 2011, pp. 75–6).

The rest of Mantzourani’s hefty book is given over to analysis. There are 100pages on Skalkottas’s twelve-note technique and a further 150 that provide ‘casestudies’ of his ‘twelve-note development’. Evidently modelled after Haimo(1990), this design causes problems, particularly with regard to the availability ofmaterials. Recordings of many of Skalkottas’s compositions may now be at hand;the same cannot be said of their scores.13 Even where these are available, theymay not be useful. Those of the Second String Trio (1935) and First Piano Suite(1936) are reproductions of the composer’s barely legible manuscript. For muchof the time the reader of this book is confronted with detailed accounts of musicthat can be heard but not seen. And sometimes not even that: while the FirstSymphonic Suite is clearly one of Skalkottas’s most imposing creations, onewonders about the author’s purpose in devoting an entire chapter (pp. 245–73)to a work that has been neither recorded nor published.

Why is Mantzourani pushed to such obscurity? There are a number of moreeasily accessible atonal scores from Skalkottas’s Athens period, including somethat have even gained a toehold in the repertoire (for example, the Concertinofor oboe and piano accompaniment of 1939), to which she pays no attention.But while these works may be atonal, they are not twelve-note. AlthoughSkalkottas’s best-known music remains his tonal Greek Dances, the composer’smusicological reputation rests on his having been a serialist. Mantzouraniobserves that ‘the vast majority of his surviving works use predominantly twelve-note idioms’ (2011, p. 6). In the last decade of Skalkottas’s career, however, thisis not the case. Not only are there a number of tonal compositions from theseyears, but the atonal music of this period often has an ambiguous relation tododecaphony. Mantzourani writes of a ‘free dodecaphonic technique’ (ibid., pp.275–317) but gives no sense of how music thus composed might relate to worksthat are not primarily twelve-note. Happiest with the ‘strict’ compositions thatwere the first Skalkottas completed following his return to Athens, she is con-cerned to a striking degree in her book with works composed in just a couple of(admittedly fecund) years, 1935–6.14

114 BEN EARLE


Mantzourani would like to appear open to the expressive content ofSkalkottas’s work. At the end of her ‘Biographical Study’, she declares that‘[i]rrepressible activity, constant forward motion, and an agitated, agonizedquest for something unattainable are the characteristic elements that govern thismusic’ (2011, p. 77); in the ‘Epilogue’ she writes of ‘immediacy of sound,expressive power and sheer, unbridled energy’ (p. 338). She takes the unusualstep of threading through her book the stanzas of Constantine P. Cavafy’scelebrated poem ‘Ithaca’. Presumably the experiential riches to which the poetrefers are intended, mutatis mutandis, to strike a chord with readers working theirway through her text. But alas, it is the work of a thoroughgoing positivist.Outside the encomia just cited, the experience of listening to Skalkottas’s musicis barely even a topic for discussion.

Emblematic of Mantzourani’s approach are the tables that pepper her text.Future writers on Skalkottas’s twelve-note music will be able to turn to themwith confidence: the basic work of row identification has been reliably carriedout. As a means of conveying what is ‘really going on in Skalkottas’s music’(2011, p. xxiii), however, tabular presentation of data, in its detemporalisedabstraction, is unsuitable. Mantzourani contends that ‘deciphering and mappingout the technical parameters’ of the composer’s twelve-note technique ‘is nec-essary for any informed future discussion of Skalkottas’s dodecaphonic music’(ibid., p. 6). Against such a statement one should place the words of thecomposer’s principal Berlin teacher. Schoenberg was particularly impressed bythe Octet (1931) for string quartet with flute, oboe, clarinet and bassoon: ‘Onecan see the systematic way with which he build his themes and sentences, theelaboration of the motivic work, qualities with which I judge my students’ works.This is an outstanding work’ (cited on p. 38). Mantzourani is not entirelyunresponsive to this aspect of Skalkottas’s music. Throughout her study shemakes use of the Schoenbergian vocabulary of motive, phrase, period andsentence. Yet her employment of these terms is not always very successful. Letus consider her response to Ex. 5, which shows the first 24 bars of the Octet’sopening movement.

Skalkottas as Music

In such a busy polyphonic idiom, the clear articulation of phrases and themes isimperative if the music is not to descend into incoherence. If bars 11–12 soundno more transparent in the old Melos Ensemble recording than they do in thenew BIS version, that is not necessarily the fault of the players.15 As JuditAlsmeier has pointed out, identification of themes in this movement is renderedproblematic by Skalkottas’s techniques of ‘dissolution’. Strictly speaking, bars174–20 should be compared with bars 4–62, since at bars 174–20 Skalkottasbrings back both the wind and string parts of the earlier passage (with the scoringreversed). But it will be sufficient to compare the first violin at bars 174–20 withthe same instrument’s music at bars 04–41. Transposed up an octave, the later



Ex. 5 Nikos Skalkottas, Octet (1931), first movement, bars 1–24. © UniversalEdition A.G. Wien. Reproduced by permission of Universal Edition (London) Ltd.

116 BEN EARLE


Ex. 5 Continued



Ex. 5 Continued

118 BEN EARLE


Ex. 5 Continued



violin line retains its rhythmic character (until bar 193), but the B� and B� in bar18 are substituted for the expected E� and C (placed two octaves lower in thesecond violin). Then, at bars 194–20, while the ordered pitch classes of bars23–41 are retained, their rhythmic presentation is transformed (see Alsmeier2001, pp. 95–9).

Mantzourani identifies an opening ‘period structure’, which she calls ‘A’,based on the first violin’s opening twelve-note melody, and a ‘B’ section at bars74–11, which presents ‘a new thematic idea in the bassoon ... based on a newtwelve-note set’. An ‘A1’ section is ‘characterized by a dense, antiphonal texturebetween the winds and strings’; finally, ‘section C’ (beginning at bar 233)features a melody that is not quite twelve-note (2011, pp. 200–1). Problemsbegin with the initial identification of a period. The segmentation of bars 04–73

into a three-bar antecedent and four-bar consequent is unconvincing, since froma Schoenbergian perspective the two phrases lack the contrasting ideas thatought to appear in both. Rather than viewing bars 04–73 and 74–11 as distinctsections, we might instead grasp the whole of bars 04–23 as a single thematicunit: a ternary shape that opens with the kind of immediate repetition charac-teristic of a sentence. Mantzourani is keen to show how the bassoon line at bars74–11 completes the chromatic while incorporating an ‘interpolation’ in bar 10(pp. 101–2). She does not ask herself whether bar 11, rather than marking theend of a section, might initiate a repetition of the material introduced by thebassoon at bars 74–10. It is bars 74–173 that can be heard to open as a period.16

Certainly the antecedent (bars 74–10) possesses the right kind of two-partstructure, even if it is only three bars long. The contrast between bars 74–91 and92–10 is clear. But most prominent in the two recordings at bars 11 and 12 is thematerial heard on flute and oboe (bar 11) and then clarinet and first violin (bar12), which, as Mantzourani points out (p. 101), echoes the bassoon’s ‘interpo-lation’ of bar 10. So when the viola enters in bar 13 with a version of thebassoon’s ‘contrasting idea’ from bar 9, it comes from nowhere, having beenstripped of the ‘basic idea’ that should have preceded it. In order to make senseof this passage, performers need to realise, first, that at bar 11 the bassoon bothremains the Hauptstimme and launches a consequent phrase to its antecedent atbars 74–10; and second, that on the final quaver of bar 11, the Hauptstimmepasses to the cello (as indicated by the forte dynamic), which repeats the rhythms(and the pitch classes) of the bassoon’s basic idea before passing the line to theviola.17

Mantzourani asks whether this movement might be a rondo or a ternary form,but plumps finally for sonata on account of the music’s ‘tonal’ structure. Misled,presumably, by the least convincing aspect of Milstein’s work on Schoenberg,Mantzourani finds it appropriate to make quasi-Schenkerian analyses of ‘[l]arge-scale movement in the bass line’ (2011, p. 202). In the case of Ex. 5, she cannotclaim that the pitches she extracts from the cello line have any tonal function. Yetthe look of the repeated Gs and Cs in bars 13–16 was apparently irresistible.Thus the ‘A1’ section gives us a ‘cadence’ in C (evidently the G–C between the

120 BEN EARLE


final pitch at the end of bar 20 and the bass of the chord in bars 21–22), and the‘A’ section has a ‘cadence’ on G (the repeated G at bar 73). As non-functional (andgenerally inaudible), these identifications are irrelevant to the movement’s form.They also cast doubt on Mantzourani’s feeling for the sense of Skalkottas’s music.Consider the role of the clarinet in bars 15–17. Taking up the viola’s material frombars 13–14, it departs from the primarily conjunct character of the viola’s line bymeans of octave displacement. Yet in bar 16, instead of taking this developmentfurther, the clarinet makes the very classical gesture of waiting. It liquidates themotive of bar 15 in a semiquaver figuration before leading the ensemble in aquasi-half-close (its opening A–B� transposed up an octave) before the return ofthe material of bars 04–41. From a tonal perspective, the quality identified at theend of the ‘B’ section (bars 74–173) should be dominant, not tonic.

There is another source for Mantzourani’s tonalism. In a short piece from1954, Hans Keller hailed Skalkottas as a genius on the basis of a tape of the 1953premiere of the Second Piano Concerto; he had no score (Keller 1954).18

Though he was never to publish a more detailed appraisal of Skalkottas’s music,Keller was happy to pronounce the composer not just a genius, but a ‘symphonicgenius’, the only one since Schoenberg to have successfully built on the latter’sexample (Keller 1994, p. 191). Keller’s understanding of Schoenberg wasspecial: ‘Harmonically ... his dodecaphony is composed against the background... of well-defined, well-implied, but violently suppressed ... tonal expectations,whose replacement ... makes instinctive sense, and this is why and how histwelve-tone harmony “works”’ (Keller 1981, p. 7). Mantzourani has taken upthe challenge (Mantzourani 2013; compare with Mantzourani 2011, pp. 214–21). The first movement of Skalkottas’s Third String Quartet is composedagainst the ‘unifying background’ of a C major that is ‘frequently suppressed anddisguised’ (2013, p. 54). In the opening paragraph of this sonata movement,shown in Ex. 6, C major is nowhere to be seen. Undeterred, Mantzouraniemphasises the harmonies at bars 111–3 and 143, suggesting that the ‘texturallayout’ of bars 11–14, ‘aurally emphasising the triadic content of severaltetrachords[,] ... outlines a harmonic movement from an A-major/A�-diminishedto [a] C� minor triad’ (ibid., p. 47).

Mantzourani convincingly points out how the diminished seventh of bars9–10 resolves to bar 11. On the other hand, her suggestion that the cello’s D� atbar 143 ‘functions as a leading-note to the following e1’ is specious, while theemphasis on the C� minor triadic component of the harmony at bar 143 seemsmisguided, not just because it involves turning a deaf ear to the D� in the bass,but also because it distorts the musical sense of the passage. If the bass comes toa rest towards the end of Skalkottas’s first phrase, it is at bar 134, which ismarked, significantly, with an accent. The warm C�9 sonority is sustained overthe bar line, suggesting an expansion of the notated common time to a 5/4 metre.Following this, the two chords at bar 142–3 act not as the conclusion to aharmonic progression, but as a gesture of dismissal: a 2/4 bar that clears the airbefore the start of the second phrase at bar 144.



Ex. 6 Nikos Skalkottas, String Quartet No. 3 (1935), first movement, bars 1–42.Copyright Associated Music Publishers, Inc. (BMI). All rights reserved. Interna-tional Copyright Secured. Used by permission

122 BEN EARLE


Mantzourani’s search for triadic elements has clouded her musical response,and Keller is to blame. ‘[T]oo proud to descend to mere helpfulness’, as RobinHolloway puts it (Holloway 2003, p. 412), Keller’s refusal to parse the ‘fore-ground’ of his examples was justified by a belief that ‘musical’ listeners would

Ex. 6 Continued



grasp it instinctively, and that if others did not, analysis was de facto wasted onthem (Keller 1977, pp. 114 and 117). But analysis of the ‘foreground’ oftwelve-note music was always what was required. In order to make this scoreexemplify Keller’s claim that Schoenbergian structures sometimes begin ‘beforethe beginning’, Mantzourani suggests that the ‘exposition proper’ of the materialof Ex. 6 starts only at bar 144; at the same time, she designates bars 144–232 asa consequent to the antecedent at bars 1–143 (Mantzourani 2013, pp. 47 and44). The latter reading is appropriate: the slow unfolding of harmonies at bars1–10 constitutes the ‘basic idea’, the more rapid homophonic movement of bars11–143 the ‘contrasting idea’. Both then return: the unfolding at bars 144–183,the more rapid movement at bars 184–232. By bar 23 the consequent of theopening period would seem to be complete, yet there are a further four barsbefore the next obvious formal break at bar 28. Mantzourani treats bars 18–27as a period in themselves, the first violin outlining an antecedent at bars 184–232

and a consequent at bars 233–27. This is unsatisfactory, since the new period notonly lacks the ‘basic idea’/‘contrasting idea’ duality; it also cuts across thepreviously established period at bars 1–232. As for bars 28–421, Mantzouranioffers ‘[v]aried and condensed repetition of antecedent’ (bars 28–331) and‘[t]ransition and chordal cadence’ (bars 334–421) (ibid., pp. 44 and 48). Neitherlabel is very helpful.

As far as pitch is concerned, Skalkottas proceeds here in a strictly linearfashion, typical of his music of 1935. Each instrument at bars 1–143 introducesits own row, which it then repeats, without transposition, inversion or retro-gression until the end of the example. Skalkottas relies on texture and aboveall rhythm to sustain interest. Though Mantzourani barely mentions thisfeature, acceleration (first encountered in bars 9–10) could well be taken asthe ‘idea’ of Ex. 6. The process, which reaches a climax of intensity at bar 37,is intriguingly non-uniform. At bars 144–163 the opening chromatic tetrachordunfolds (in inversion) at twice its previous speed; the following whole-tonetetrachord appears ready (at bars 164–171) to double the speed once more. ButSkalkottas holds back the arrival of the D� by two crotchets, producing anelongated C� (and then E� in parallel), the music at once pressing forward andheld back.

Mantzourani declares that in ‘Skalkottas’s dodecaphonic works based on alimited number of sets, the periodicity of the twelve-note set-groups largelyconstitutes the basis of the phrase structure’ (2011, p. 132). Yet in Ex. 6 thecomposer sometimes avoids a serial/phrasing homology. We noted earlier howthe two chords at bar 142–3 lead into the phrase that immediately follows; they areconstructed from order numbers 11 and 12 of the first quartet of rows. At bars231–2 this disjunction is more evident. Musically, the first violin’s syncopated Eand C� are bound to the following G, not to the preceding F, while the accom-paniment to the final two pitches of the first violin’s row emphasises the con-nection to bars 233–27 in that it introduces the pairs of quavers heard in bar 24.Mantzourani refers to bar 27 as a ‘cadence’; yet, as at bars 12 and 22, the phrase

124 BEN EARLE


proper comes to an end when the bass reaches D� (previously spelled C�) at bar263. Bar 272–4 would be better heard as pressing forward to the quavers of bar 30.

The sudden reappearance at bars 28–29 of a texture and rhythmic style akinto those of the movement’s opening is the most striking example of holding backin Ex. 6. After the quavers of bars 30–332, Skalkottas halts his music’s momen-tum once again. Like bars 30–332, the following section in syncopated chordsopens in the manner of a sentence. A model (bars 333–34, parallel to bar 30) isimmediately followed by its varied repetition (bars 34–352, parallel to bar 31).But where, at the point of expected continuation, the quavers peter out infragmented repetition, the syncopated chords explode into a climax, with imi-tative entries from all four instruments. This is another place where phrasestructure and twelve-note order do not coincide: when the first violin arrives atits top G (bar 372) to begin the final statement of its row, the other instrumentsare still completing the rows begun at bar 33. The final phrase of the extractemerges into clarity only at bars 384–393, with their climactic statements of thewhole-tone and diminished-seventh tetrachords, leading to an unmistakablereturn to the texture of bars 11–12.

From Skalkottas to Dallapiccola

Keller’s admiration for Skalkottas was probably genuine. The form of the ThirdQuartet’s Allegro moderato would have met with his approval inasmuch as itexemplifies the ‘Mannheim’ reversed recapitulation (Mantzourani misses this),which, along with the return to the material of the opening at the start of thedevelopment (bars 824–903), produces the sonata rondo effect he regarded as aformal compensation for the loss of tonality. And as Mantzourani rightly pointsout (Mantzourani 2013, pp. 54–5), the movement’s subordinate theme createsthe kind of contrast with the music of Ex. 6 that Keller regarded as a sine qua nonof symphonism (Keller 1994, pp. 188–91). But just how ‘Schoenbergian’ Kellerwould have called this music surely turns on the issue of development. ForMantzourani, development, or developing variation, appears to be demonstratedby the discovery, in apparently unrelated materials, of invariant segments(Mantzourani 2013, pp. 51–3). But for Keller, development refers to an audibleand dynamic process: thematic evolution grasped in the context of music of a‘forward-urging’ metrical type (Keller 1994, p. 208). If, at the start of his StringQuartet No. 2, Op. 10 (1907–8), Schoenberg ‘begins before the beginning’, it isbecause he extends the developmental character of his music from the motive tothe phrase. By contrast, however complex the motivic transformations one mightfind in Webern, the latter is not ‘a contraster and developer’, but a ‘stater andvarier’ (Keller 1994, pp. 207–8 and 186). From a Schoenbergian perspective,the first movement of Skalkottas’s Third Quartet contains a remarkable quantityof exact repetition.19 As the above analysis of Ex. 6 shows, Skalkottas is a ‘staterand varier’ at the local level too. Bars 1–41 of the Third Quartet are a miniaturetheme and variations.



Mantzourani declares that Skalkottas’s version of twelve-note technique was‘developed ... before attending Schoenberg’s masterclasses’ and ‘remained con-ceptually unchanged both during and after his studies with Schoenberg’ (2011,pp. 3–4). It is plain that Skalkottas used twelve-note technique in a mannerdifferent to that of his teacher. Yet Alsmeier is surely not incorrect to view thecontinuous transformation of material in the first movement of the Octet asSchoenbergian (Alsmeier 2001, p. 99). In contrast to the more relaxed idiom ofthe Third Quartet, the hectic atonal polyphony of the Octet could well be takenas Schoenbergian in origin too. Mantzourani courts the error of identifying thesecomposer’s styles with their means of generating pitches. Perhaps the title of herbook is not quite right. The text is not, after all, primarily concerned with thetwelve-note music of Nikos Skalkottas, but with the pitch-class manipulations heemployed while composing it.

The problem reappears in Brian Alegant’s similarly titled study of Skalkottas’scontemporary Luigi Dallapiccola (1904–1975). Alegant is concerned ‘to dem-onstrate convincingly the influence of both Schoenberg and Webern’ onDallapiccola’s compositional practice (2010, p. 9). He is particularly interestedin ‘cross partitions’, whereby the order numbers of a series are arranged to formsimultaneities, usually tetrachords or trichords – 43 or 34 configurations, inAlegant’s nomenclature – the vertical arrangement of which may be altered:Alegant refers to such reorderings as ‘[s]lot-machine permutations’ (p. 21). ‘It islikely that Dallapiccola’s initial exposure to cross partitions came through thestudy of Schoenberg’s twelve-tone works’, Alegant suggests, ‘especially the Op.33a Klavierstück’ (p. 23). But he immediately backtracks, commenting that ‘itdoesn’t matter whether Dallapiccola came to cross partitions on his own orwhether he appropriated them from Schoenberg, Berg, or another composer’ (p.24). This lack of certainty does not, however, prevent Alegant from referring toSchoenberg whenever he comes across this technical device.

One doubts that Dallapiccola needed to analyse Op. 33a before turning hisrows into chords. Nor indeed was Schoenberg the only composer to employcross partitions before Dallapiccola. In the first edition of his book on Berg(which Dallapiccola had read by 1943), Willi Reich demonstrates how, in Lulu(1928–35), Berg derives the ‘portrait harmonies’ from the opera’s principalseries (Reich 1937, pp. 112–13; see also Dallapiccola and Mila 2005, p. 64).So why not refer to cross partitions as Bergian? That Alegant hardly refers toBerg is surprising, given that commentators on Dallapiccola’s music of the1940s have so often emphasised the manner in which the Italian – like Berg –manages here to create a twelve-note idiom of quasi-diatonic euphony (forexample, Morgan 1991, p. 256 and Whittall 2008, pp. 160–2). For Alegant,committed to a set-theoretical approach, the traditional elements ofDallapiccola’s music are necessarily uninteresting, even though it is evidentlytheir presence that has brought his music of the 1940s a degree of recognitionthat the later work has conspicuously lacked. It does not matter to Alegant thatwhile Dallapiccola’s early twelve-note music sometimes sounds a bit like Berg

126 BEN EARLE


and his work of the 1950s often a great deal like Webern, at any period of hiscareer his music rarely sounds like Schoenberg at all. For a commentator whoequates music to the techniques by which it is made, such issues may indeedbe ‘irrelevant’ (Alegant 2010, p. 190). Yet they can lead us to the essence of acomposer’s achievement.

In the course of a chapter on Dallapiccola’s Parole di San Paolo for mediumvoice and ensemble (1964), Alegant points to instances of ‘formal parallelism’(p. 250) and ‘recapitulation’ (p. 257). The music at bars 32–39 returns largelyunchanged at bars 51–57; bars 38–39 are anticipated at bars 17–20. ‘[S]uchliteral repetition was for the most part strenuously avoided by the serial com-posers in the Second Viennese School’, Alegant tells us (p. 252). But as thestudent of Dallapiccola’s music of the 1930s and ’40s cannot fail to notice,strophic repetition is this composer’s principal formal means. In Keller’sterms, Dallapiccola is precisely not a Schoenbergian; like Webern, he is a‘stater and varier’. But unlike Webern – and unlike Skalkottas too –Dallapiccola is not a ‘Viennese’ composer in anything more than a superficialsense. The syntax of his music has French or Russian sources: Debussy, Ravel,Stravinsky.

Consider Ex. 7, the second of the Quattro liriche di Antonio Machado (1948),which Alegant discusses in detail (pp. 24–8). The obvious analytical point here(which Alegant does not make) concerns the changing relationship of voice andpiano. The first eleven bars of the example present a refrain-verse structure: thepiano has the ‘refrain’, fashioned from two 43 cross partitions (P0 and R6).20 Bars43–46 are followed by a ‘verse’ consisting of an unaccompanied vocal line(beginning with P11) that slowly describes a melodic arch. Characteristically,Dallapiccola proceeds to a repetition of these materials. At bars 51–52, we heara transposition down a tone of bars 43–44 (bar 51 has the rhythm of bar 45rather than that of bar 43); the imitative lines of what Alegant calls the following‘stretto’ (p. 24) relate closely to the contour of bars 47–49, though this music hasnow migrated from voice to piano.

The closing beats of the stretto (bar 541–3) initiate a passage of ambiguity. Thesecond repetition of the refrain-verse structure (the third of the song’s fourstrophes) is disguised: the harmonies of bars 43–46 return at bars 543–55,transposed from the succession P0–R6 to P5–R11. But the rhythmic character ofbars 43–46 and 51–52 is lost, along with their melodic contour (transformed by‘slot machine’ permutation). Meanwhile, the voice has not waited for the pianoto finish, entering with a line that inverts the arch of bars 47–50. After the formalclarity of bars 43–53, the sense of bars 54–56 appears less certain. One landmarkis the piano’s descent at bars 554–56 (prepared by the repeated G at bar 552–3),which echoes the vocal descent of bar 54. But the vocal line at ‘Después soñéque’ presents an unfamiliar contour, unsettling any sense of recognition at thispoint. Then, suddenly, at bar 57, everything locks into place. The refrainand verse, introduced separately and then unstably combined, now find equi-librium. The piano’s music has not regained the rhythms of bars 43–46 or



Ex. 7 Luigi Dallapiccola, Quattro liriche di Antonio Machado (1948), No. 2, complete.© Copyright 1950 by Edizioni Suvini Zerboni S.p.A. Milan. Reproduced bypermission

128 BEN EARLE


51–52, but the pitches, and above all the melodic contour, of bar 57 are those ofbars 43–44. The voice, too, returns to its starting point: as at bars 47–51, P11 isfollowed by I6.

Alegant refers to ‘a three-part, a–b–a design, with four-voiced homophonicpassages flanking a brief canonic episode’ (2010, p. 24). The strophic form ismissed. His handling of the song’s pitch structure is perfectly efficient. But in thiscontext, efficiency is not enough. It is one thing to isolate and compare crosspartitions; the analyst’s concern must surely be to explain their arrangement. Anunwillingness to rise to the level of interpretation marks Alegant’s study as awhole. In the chapter on Dallapiccola’s octatonic practice (pp. 109–54), thequestion of how octatonicism might interact with other kinds of harmony isscarcely even broached. Alegant may strew his text with avowals of enthusiasm.Ex. 7, for instance, is ‘exquisitely constructed and expressive’ (p. 24). But inorder to elucidate this music’s expressivity, Alegant would need to leave behindhis theoretical safety net and comment on the song’s temporal progress from theperspective of his musical experience. He does possess a term for those aspectsof Dallapiccola’s music that are unreachable via his music-theoretical concepts:the strikingly static ‘soundscape’. Alegant finds in Ex. 7 ‘an atmosphere that isintimate and delicate, tentative and weightless’ (p. 27). Up to a point, this is

Ex. 7 Continued



appropriate. Yet Dallapiccola’s is not ‘texture’ music. There is more to Ex. 7than ‘soundscape’ or ‘atmosphere’.

The Italian is above all a vocal composer, a setter of words. Alegant gives textsand translations for the works he tackles; he is otherwise ‘happy to let othersspeak on the cultural, literary, metaphysical [and] philosophical ... influencesthat shaped Dallapiccola’s music’ (2010, p. 2). To imagine that one might givean adequate interpretation of Ex. 7 on the basis of a reading of its text as ‘adream-inspired fantasy’ (p. 24) is naïve. Alegant has an odd idea of the Machadosongs in general. His observation that their ‘ethereal lyricism’ stands opposed tothe ‘stridency and intensity’ of Dallapiccola’s music elsewhere (p. 54) suggeststhat he may not have listened to the cycle all the way through. The third song isa thoroughly ‘strident and intense’ outburst, setting one of the poems Machadowrote after the death of his teenaged bride, Leonor.

Señor, ya me arrancaste Lord, you took from melo que yo más quería. what I loved best.

Oye otra vez, Dios mío, Hear once again, my God,mi corazón clamar. my heart cry out.

Tu voluntad se hizo, Señor, Your will was done, Lord,contra la mía. against mine.

Señor, ya estamos solos Lord, we are alone,mi corazón y el mar. My heart and the sea.

This is the core of the cycle. The other poems (not contiguous in their originalpoetic context) are arranged by the composer to form a progression that,as Roman Vlad noted long ago, stands in close relation to the religiouscrisis evinced by the libretto of Il prigioniero (see Vlad 1957, pp. 40–1).21 AsAlegant notices too, the opera and the song cycle ‘share many structural andsonic characteristics’ (Alegant 2010, p. 132). The joyous greeting to spring inthe first song relates poetically and musically to the opening of the opera’sScene 4.22 After the catastrophe registered in the third song, however, theunion of God and nature in the ‘¡Aleluyas blancas / de los zarzales floridos!’(‘White alleluias / of the brambles in flower!’) can no longer be celebrated. Toa slow transformation of the material of the first song, marked con amarezza(‘with bitterness’), Dallapiccola sets the couplet ‘La primavera ha venido /Nadie sabe cómo ha sido’ (‘Spring has come / Nobody knows how ithappened’).

The second song, placed between joy and catastrophe, is uneasy in tone:

Ayer soñé que veía Yesterday I dreamed that I sawa Dios y que a Dios hablaba: God and that I spoke to God:y soñé que Dios me oía ... and I dreamed that God heard me ...Después soñé que soñaba. After that I dreamed that I was dreaming.

The focus of the poem is the third line, marked musically by the singer’sextraordinary ascent to a pianississimo C�. The poet relates a moment of directcommunication with God, of ecstatic temporal suspension, corresponding to the

130 BEN EARLE


moment in Dallapiccola’s song when, as we saw, the musical sense seems tocome undone. Yet it was only a dream, or rather a dream within a dream. Theangular setting of ‘Después’ already starts to break the spell, while therecapitulation at bar 57, with its sudden clarification of texture, can now be seenas a return to reality, a disillusionment.

Alegant prefers the post-1950 compositions, especially those from the years1956–60, which ‘arguably represent the most fertile period in Dallapiccola’sdevelopment’ (2010, p. 47). The periodisation derives from the composer, whoobserved how on each occasion that he composed a tonal piece based on themusic of an earlier Italian composer – Sonatina canonica for piano (1942–3),Tartiniana for violin and orchestra (1951) and Tartiniana seconda for violin andpiano (1955–6) – he subsequently ‘took a noteworthy step forward along thepath of twelve-tone music’ (p. 12). As Alegant points out (pp. 12–13), theboundaries between ‘phases’ are not watertight, nor is there any tonal compo-sition to usher in the period 1960–72. And his characterisations of phases one,two and four would be more convincing if they paid decent attention toDallapiccola’s large-scale compositions of these years. The two hours of his finalopera, Ulisse (Berlin, 1968) are discussed on just a handful of pages, while Job,the 35-minute sacra rappresentazione of 1950, is hardly mentioned. Alegant’scontention that the works of the second phase are primarily miniatures (pp. 46and 86) neglects the 30-minute Canti di liberazione for chorus and orchestra(1951–5), the pre-eminent achievement of the early 1950s. His treatment of thesecond phase adds little to the accounts of Babbitt (2003, pp. 64–5 and 1987,pp. 38–41), David Lewin (1993, pp. 1–15) and Thomas DeLio (1985, pp.186–95).

It is no coincidence that the compositions of the third phase – An Mathilde forfemale voice and orchestra (1954), Cinque canti for baritone and eight instru-ments (1956), Concerto per la notte di Natale dell’anno 1956 for soprano andchamber orchestra (1957–8), Requiescant for mixed chorus, children’s chorusand orchestra (1957–8) and Dialoghi for cello and orchestra (1959–60) – shouldbe at once the most unfamiliar of Dallapiccola’s postwar music and that withwhich Alegant is happiest, for it was at this point that the composer advanced(if that is the right word) furthest into ‘rigor and systematization’ (Alegant 2010,p. 74) involving the parameters of duration and instrumentation, as well as pitch.Dallapiccola had signalled a break with his previous idiom in the Goethe-Lieder,for mezzo-soprano and three clarinets (1952–3). As he put it a decade later, ‘Iconsider the Goethe-Lieder a work of extreme importance in my development.For the first time I was able, and with much clarity, I believe, to write a metricalnotation which corresponded exactly to what I had in mind’ (Nathan 1966, p.302). Did the composer genuinely believe that before 1952 notation hadimpeded the communication of his ideas? As Alegant does not make properlyclear (2010, pp. 38 and 88), the ‘floating rhythm’ of the opening of the seventhand last of the Goethe-Lieder (Ex. 8) is achieved not by an avoidance of down-beats (for poetic stress is aligned with the bar lines), but rather, first, by the



avoidance of metrical regularity in the vocal line, and secondly, by the placementin the clarinets of various multiplications of the durations of the voice’s openingcrotchets (by 0.75, 0.33, 0.5, 0.66 and 0.5) so as to obscure any hierarchicalsuperiority of the notated metre. It is hard to imagine that the complexity ofEx. 8 merely corresponds to music that was pre-existent in the composer’s mind.Perhaps Dallapiccola really meant the opposite of what he said, namely that inthe Goethe-Lieder, he found himself for the first time not struggling to find ametrical notation for music he could already hear, but writing music unimagi-nable in its details prior to notation.

He was doubtless pleased by the look of Ex. 8, which is more complex than itneeds to be (see especially the B� clarinet in bars 4–5). More charitably, whatDallapiccola ‘had in mind’ could be interpreted as an approximation of the‘internal rhythm’ that from the first, as Alegant points out (2010, p. 49), he hadadmired in Webern. The debt to Webern in Ex. 8, with respect to pitch con-struction, rhythmic/metric style and scoring, is unmistakable. In the historicallyrichest section of his book (pp. 47–83), Alegant shows how, armed with theanalyses of René Leibowitz (1949), Dallapiccola created his own versions ofWebern’s ‘soundscapes’, employing many of the same technical means:RI-symmetrical rows, palindromes, derivation, four-voice arrays and – mostobviously – canons. These elements are particularly prominent in Alegant’sfavourite pieces of the later 1950s, of which he gives a thorough account,including a technical description (not properly an analysis) of the whole of AnMathilde (pp. 155–225). As with Mantzourani on Skalkottas, so with Alegant onDallapiccola: this book will be an indispensable first port of call for anyoneworking on this repertoire in future. It is, however, only intermittently musical,and then weakly so.

Ex. 8 Luigi Dallapiccola, Goethe-Lieder (1952–3), No. 7, bars 1–5. © Copyright1953 by Edizioni Suvini Zerboni S.p.A. Milan. Reproduced by permission

132 BEN EARLE


Ulisse

Alegant is concerned that readers be able to hear the pitch-class invariances heuncovers. A ‘[s]ummary of tetrachordal ideas’ in Parole di San Paolo is offered‘as an ear-training aid’ (2010, p. 281 and p. 309 n. 38). Appreciation ofmusical sense is one thing; aural skills another, we might respond. But perhapsthe situation is not quite so clear-cut. In Schoenberg or Skalkottas, a tradi-tionally conceived ‘Austro-German’ syntax finds itself stretched to breakingpoint; the necessity to explain that this syntax nevertheless constitutes thesubstance of their work – Schoenberg’s ‘what it is’ rather than ‘how it is done’(Schoenberg 1964, p. 164) – only indicates how poorly this repertoire hasbeen served in English. If the same might be said with regard to the ‘French’syntax of Dallapiccola’s music of the 1940s, his work of the 1950s and ’60s isanother matter. Already in late Webern, Adorno thought, a fetishism of therow had led to the extinction of musical meaning (Adorno 2006b, pp. 86–7).If the work of Webern and the post-Webernians is indeed devoid of musicalsense, one might think its employment for the purposes of ear training per-fectly appropriate. But Adorno surely exaggerates. If Alegant’s methodologymakes it impossible for him to account for the expressivity he finds inDallapiccola’s music of the 1950s and ’60s, it is a sign not that this expres-sivity is non-existent, but that we need to look elsewhere for an interpretativeapproach. Certainly the intellectual framework of Romano Pezzati’s full-lengthstudy of Ulisse (2008) could hardly contrast more vividly with that of Anglo-American positivism.

In the final lines of his libretto, Dallapiccola has Ulysses pronounce aninversion of Machado’s despair: not ‘Señor, ya estamos solos mi corazón y elmar’, but ‘Signore! / Non più soli sono il mio cuore e il mare’ (see Ex. 9). Theopera he regarded as the summa of his achievement was intended to set to rest theanguish of his earlier work. ‘I should like some day, after all the question marks– mine and others’, to succeed in expressing a “certainty”’, he wrote(Dallapiccola 1987, p. 104). Following Dante in Canto XXVI of Inferno,Dallapiccola sets his final scene (an Epilogue) not on Ithaca but at sea. Everunsatisfied, Ulysses has set sail once more: we see him alone in a small boatunder the stars. His anguished monologue reaches its climax in the overt expres-sion of a quest for meaning: ‘Trovar potessi il nome, pronunciar la parola / chechiarisca a me stesso così ansioso cercare; / ... Se una voce rompesse il silenzio,il mistero ...’ (‘If I could but find the name, speak the word that would explainto me this anxious searching; / ... If a voice would but break the silence, themystery ...’). All of a sudden a voice does speak – or so we are to believe, forUlysses experiences the epiphany cited above: ‘Lord! / No longer are they alone,my heart and the sea.’

Most telling in Ex. 9 is not, perhaps, the moment when God ‘speaks’ (bars1023–10251), but what happens immediately before and after the passagemarked come una parentesi. As Dietrich Kämper was the first to reveal,



Ex. 9 Luigi Dallapiccola, Ulisse (1956–68), Act II, bars 1020–1037, short score.© Copyright 1971 by Edizioni Suvini Zerboni S.p.A. Milan. Reproduced bypermission

134 BEN EARLE


Ex. 9 Continued



Dallapiccola constructed the opera from ten series, all of which (as the composernoted) employ permutations of the same two hexachords (Kämper 1984, pp.159–60; see also Nathan 1977, p. 46).23 At bars 10253–10262, the strings statethe ‘Ur-series’ ‘Mare I’ at I0, as 62 cross partitions. The third chord repeats thepitch-class content of the first, initiating a statement of I0 that is completed in adisordered manner by Ulysses and the orchestra in bars 10293–10301. Ulysses’pitches in these two bars are simultaneously the second trichord of the ‘Calypso’row at I1. He sings only the second and fourth trichords (at bars 10293–10301

and 10312–1033); the others are heard in the orchestra.24 The six 62 cross-partitions at bars 10302–1033 have been identified by Anthony Sellors as,

Ex. 9 Continued

136 BEN EARLE


successively, two statements of ‘Mare III’ at I0 and one at P1, a statement of‘Mare I’ at I6 and finally statements of ‘Mare III’ at P2 and P1 (Sellors 2001, vol.2, p. 569). At bars 1034–1035 solo instruments outline ‘Mare I’ at R5. And inbars 1036–1037 Dallapiccola combines a twelve-note collection derived from a(014) trichord (lower system) with its retrograde (middle system).25

This kind of note counting can be carried out for the opera as a whole (seeSellors 2001, vol. 2, pp. 621–69). But it is essentially pointless, since, as Sellorsremarks, ‘it is not the series which assert themselves in Ulisse, but the shapes (ormotives) which Dallapiccola constructs from them’ (ibid., vol. 1, p. 83). Thechords at bars 10253–10262 are also the opera’s opening music. Their combi-nation with Ulysses’ vocal line and its accompaniment at bars 10293–1033 isdirectly foreshadowed in the Prologue at bars 725–77. The music of bars 1034–1035 occurs frequently; the final two bars are also familiar (see Sellors 2001, vol.1, pp. 97–100). It might seem standard practice for a composer to round off alarge-scale work by returning to some of its most prominent materials. YetUlysses has just experienced the revelation of divine presence. Dallapiccola’smusic carries on as if nothing had happened.

Pezzati’s solution to this problem involves comparisons between Ulysses’ finalvocal line and Calypso’s melody (her row at I8 and P6) as she sings it in thePrologue at bars 114–161 and 803–841 (Exs 10a and 10b). Abandoned on theisland of Ogygia, Calypso addresses her absent lover by way of Machado: ‘Sonsoli, un’altra volta, il tuo cuore e il mare’ (‘They are alone, once more, your heartand the sea’). Pezzati notes the ‘metrical irregularity’ at ‘mare’ in Ex. 10a andsuggests that this ‘rhythmic caesura’ remains unresolved until the 6/4 bar ofEx. 10b. In the latter case, ‘cuore’ and ‘mare’ (for Pezzati symbolic, respectively,of Ulysses’ conscious and unconscious mind) are bound together in a singlephrase; in Ex. 10a, they were separated. In Ex. 9 ‘mare’ is once again distinctfrom ‘cuore’ (which is spoken rather than sung). There is a ‘suspension of time’of the kind ‘implicit in a dramatic conception in which, amid the folds of thechronological succession of facts, the way to a possible now of revelation canopen at any time’ (Pezzati 2008, pp. 275–6).

Pezzati’s book is introduced by Mario Ruffini as ‘unprecedented’ in itsdetailed analytical attention to a single opera, and ‘without equal in the field ofdodecaphonic studies’ (Pezzati 2008, pp. xvii and xviii). Such pronouncementsindicate an ignorance of musicological literature in general and that on Ulisse inparticular (e.g. Sellors 2001 and van Hees 1994). Pezzati is in fact an unsteadyguide to Dallapiccola’s twelve-note technique. Especially disappointing ishis refusal to identify the series in use at bars 23–6 of the Prologue: ‘nooriginal, determined series can appear at this opening, whose pluri-directionaland multidimensional structure admits no definition of the web of thedodecaphonic fabric’, he writes (2008, p. 81) of a passage that deploys ‘Mare I’at P8, R5, RI5 and I10. The deployment takes the form of bar 1034 in Ex. 9(identical to bar 4 of the Prologue). Pitches overlap, to be sure, but they appearone at a time and in strict order.



Later on Pezzati correctly writes out the pitch-class succession of bars 23–41 ofthe Prologue, yet he still does not identify it. Instead, he sows confusion (2008,pp. 309–11) by means of an unhelpful derivation of this series (‘Mare I’ at P8: G�,G, A, B�, E, E�; C, B, D, D�, F, G�) from its presentation as two 62 cross partitions(at I0) in bars 1–2. If the opera’s opening chord is understood as a verticalisation

Ex. 10a Dallapiccola, Ulisse, Prologue, bars 11–16, short score. © Copyright 1971 byEdizioni Suvini Zerboni S.p.A. Milan. Reproduced by permission

Ex. 10b Dallapiccola, Ulisse, Prologue, bars 80–84, voice only. © Copyright 1968 byEdizioni Suvini Zerboni S.p.A. Milan. Reproduced by permission

138 BEN EARLE


of the ordered hexachord C, C�, B, B�, E, F, then its second chord may be viewedas a transposed inversion of the first: G, F�, A�, A, E�, D. Thus while the firsthexachord of P8 appears as a transposition up a semitone of the opera’s secondchord, the second hexachord of P8 appears to involve a reordering of the firstchord, a transposition up a semitone of the succession B, B�, C�, C, E, F (not C,C�, B, B�, E, F). Pezzati relies on the fact that the pitch-class content (but not theorder) of the two hexachords of ‘Mare I’ is identical under inversion (P0,1–6 = I5, 7–12). It would have been better simply to spell this out.

His account of the dodecaphony of Ulisse as existing on two levels – on the onehand, as a ‘continuum’; on the other, as motives, in particular three trichordal‘fundamental figures’ (Pezzati 2008, p. 100) – is misleading. Except at the levelof an abstraction, twelve-note technique provides no more of a ‘continuum’ inthis work than it does in any other. The reader of Sellors’s dissertation will learnto recognise upwards of 50 motives in Ulisse. The effect is a hyper-Wagnerianism, for ‘while Wagner’s motives are ever-present in the musicaltexture, they cannot be used to explain every note found in it’ (Sellors 2001, vol.1, p. 55). Pezzati recognises a few of these motives (2008, pp. 101–5), but hisconcentration on the three trichordal figures tends to obscure the most distinc-tive feature of the opera’s formal construction. Nor is his handling of the figuresappropriate, for he hypostatises their significance in a manner that does notsufficiently discriminate between moments when the trichords are leitmotivicand when they are not.

Pezzati labels the chromatic trichord Dallapiccola takes over from the Goethe-Lieder (see again Ex. 8) ‘Ulysses/Nobody’, the major/minor thirds that openCalypso’s melody ‘Calypso/Ulysses’, and the ‘Viennese’ triad of superimposedaugmented and perfect fourths ‘Ithaca/not-Ithaca’. The third of these identifi-cations is most problematic. The sonority is ubiquitous in Ulisse. For Sellors, it‘does not carry any particular symbolism: it is effectively the basic harmonic unitof the entire opera’ (2001, vol. 1, p. 92). What makes Pezzati’s identification ofthe Viennese triad with Ithaca’s presence and non-presence so unconvincing isnot just the interpretative knots into which he ties himself (on which it would bekinder not to dwell) but, first, that Dallapiccola appears to have composed aquite different ‘Ithaca’ motive, and second, that Pezzati’s ‘Ithaca/not-Ithaca’chord (or rather, two of them, superimposed) does in fact gain an evidentleitmotivic content, although in reference not to Ithaca but to Ulysses himself.26

‘Sound’, Transcendence, Death

If Pezzati’s book is worth bothering with, it is for its idiosyncratic concern withthe ‘sound’ of Dallapiccola’s music. Ulisse belongs to a particular strand inmodernist opera, running from Wagner’s Tristan und Isolde (Munich, 1865) toNono’s Prometeo (Venice, 1984), characterised by its ‘internal’ character. Inthese works the real drama takes place in the orchestra pit (Pezzati 2008, pp.98–9). Such Schopenhauerianism does violence to Dallapiccola’s score, which is



frequently illustrative of the stage action in a traditionally mimetic manner (seeMila 1975). For those listeners who find the dramatic and musical propulsion ofAct II a relief after the static tableaux of Act I, Pezzati’s distaste for the conven-tionally operatic elements in Ulisse will appear perverse. But one cannot watchUlisse as if it were a melodramma, Pezzati insists (2008, p. 183). Twelve-notetechnique is properly a search for ‘a primordial and undifferentiated structure ofsound ... on which memory can draw freely, without citing or imitating passagesor styles traceable to works of the past’ (pp. 252–3).

In Ulisse this search is at once musical and religious. The theology ofDallapiccola’s opera derives from Saint Augustine, for whom there are, ofcourse, five physical senses, but also five ‘spiritual’ ones. Ulysses searches formeaning in the physical world: ‘Guardare, meravigliarsi, e tornar a guardare’(‘To look, to wonder and to return to looking’) runs the refrain of Calpyso’sopening scene. But mere looking cannot lead to truth, Pezzati observes (2008, p.132). When God spoke to Augustine, it was via a ‘spiritual’ sense of hearing, forGod is ‘within’ (Augustine 1992, p. 201). At bars 1023–10251 in Ex. 9, Ulyssesstands outside narrative and representation. In an ecstatic moment of listening,‘the memory of the unspeakable Word’ is revealed to him (Pezzati 2008, p. 7).Pezzati sets himself the analytical task of demonstrating that this G�, the musicalsign of the ‘absent-present word’ (p. 269), was always already sounding through-out the opera. Dallapiccola indicates something of the kind when he arranges theinstrumental cut-off points at bars 1023–10251 (meant to be heard as sforzandi)in the form of the opera’s Hauptrhythmus. In terms of its rhythmic profile, the G�has been present since bars 92–93 of Act I. But Pezzati goes further, assertingthat the pitch classes F�, G� and C� form ‘a spatio-temporal conjunction’ on thebasis of which ‘the serial framework of the entire opera is generated’ (p. 81).

It is possible that Dallapiccola intended F� as the ‘symbolic note of the entireopera’ (Pezzati 2008, p. 177). But too often Pezzati’s attempts to find signifi-cance in this pitch, along with C� and G�, appear arbitrary. In fact, the interpre-tative key to the opera is staring him in the face, though he fails to see it. Frombar 943 of Act II onwards (the Epilogue begins at bar 912), Dallapiccolaorganises an apotheosis of the kind of mosaic construction employed throughoutthe work. The material consists now not only of motives already used in Ulisse,but also of quotations from Dallapiccola’s previous compositions. A self-identification with the protagonist is a feature of all the composer’s stage works.But this appeal to past musical deeds also finds a parallel in Augustine. The sainttoo is looking for God, whose residence, he concludes, is in memory. But where?‘There is no place, whether we go backwards or forwards; there can be noquestion of place.’ God is found in memory in so far as he transcends it(Augustine 1992, p. 201).

Unhappily, Pezzati is no steadier a guide to Dallapiccola’s self-quotationsthan to his twelve-note technique. The references to the Cinque canti at bars955–957 and 966–9691 are missed; nor does Pezzati recognise the quotationsfrom the third of the Canti di liberazione at bars 9892–997, 10082–1015 and

140 BEN EARLE


1026–10292. The last of these – marked come una parantesi in Ex. 9 – is especiallysignificant. Missing from the original 1968 edition of the vocal score (the versionwith German translation), these bars were inserted in the full score of 1971; theyappear in the 1970 vocal score (with English translation) only in an appendix.Their significance lies in the text set at this point in the Canti di liberazione:‘Vocasti et clamasti’. ‘You called and cried out loud’, Augustine writes, ‘andshattered my deafness’ (ibid.). It is a secret reference: Augustine’s words are notheard in Ulisse. But secrecy is the point. There is no need for Pezzati’s F�–G�–C�framework, nor for his desperate attempt to suggest that, following Ulysses’epiphany, the final moments of the opera occupy ‘another time’ (2008, p. 333).From a musico-dramatic perspective, these bars are an anticlimax. It is indeed asif nothing has happened, for if we have witnessed Ulysses’ moment of revelation,we have not ourselves participated in it. A strictly transcendent event, it occurswithin Ulysses and nowhere else. As Augustine reminds us, citing I Corinthians2:11, ‘[N]o one “knows what is going on in a person except the human spiritwhich is within”’ (Augustine 1992, p. 180).

Pezzati employs Dallapiccola’s declaration that the ‘unity of the musicaldiscourse’ in a twelve-note composition will always be ‘an interior matter’ asconfirmation that the basis of his creativity was ‘that deep and hidden place inman where the sense of beauty and the voice of God are already present andoperative’ (Pezzati 2008, p. 285). Dallapiccola’s music must be grasped as‘sound’, ‘leaving aside its serial elaboration’ (p. 207). Yet ‘sound’ always wantsto achieve meaning: it seeks words. At the same time, words seek their ‘origin’,their theological ‘true meaning (the name)’ in ‘sound’ (pp. 19–20). ‘While inMonteverdi voice and instruments converge unambiguously in the soundingarticulation of the words ... , such that words may assimilate sound (secondaprattica), in Dallapiccola, instead, a reciprocal movement occurs: it is words thatarise from sound – from the same sound – which belongs to them originally anddiffuses itself in the melodic movement and sounding space’ (p. 32).

There is a place for mysticism in the analysis of Ulisse. But Pezzati’s notion ofa ‘pure signification’, an ‘original space of sound, intermediate between ...musical structure and linguistic structure’, which ‘cannot be observed by logicalcomprehension, but rather heard in the movement of the soul’ (2008, p. 35),offers little by way of a practical guide to a far from straightforward opera.Sellors declares that the motivic mosaic of Ulisse ‘does not sound like a string offragments, any more than a piece of Mozart does’ (2001, vol. 1, p. 85, emphasisin original). This too looks like an evasion. On a smaller scale, Sellors can beless confident. Of Ex. 10a, he writes, the ‘four chords ... are meaninglesswithout the co-ordinating impulse provided by the melody’. And further:‘[P]layed by themselves, the chords do not seem to generate any sense offorward motion. What forward motion there is is generated by the vocal line’(ibid., vol. 1, p. 196). In the absence of vocal melody, passages like Ex. 10awould teeter on the edge of a ‘sound’ quite different to Pezzati’s: that ofScruton’s unmusical succession.



So how is Ex. 10a musical? We can dismiss Pezzati’s talk of a ‘discord’between Calypso’s melody and the previous music owing to an alleged lack ofa direct twelve-note link, along with his attempt to interpret the melody interms of his F�–G�–C� structure (Pezzati 2008, pp. 90–1).27 But his earliercomments about the ‘schizophrenic’ character of most analyses of twelve-notevocal music, in which ‘humanistic and spiritual’ concerns are incongruouslysuperimposed upon technical observations, are well taken (p. 22). As Pezzatirecognises, verbal and musical sense in Ex. 10a is inextricable. The melody isdivided at bar 133 by the rest, by the following change in register and by themanner in which the falling semitone at ‘volta’ lowers the tension of thesecond segment of the first phrase. In bars 133–161 the similarly appoggiatura-like semitone at ‘cuore’ rises rather than falls; more generally, these bars aremarked by a broadening, ‘as if the images of “heart” and “sea”, despite theirseparation due to the rests, or precisely on that account, might stand out inanother space-time capable of interrupting the temporal succession expressedby “once more”’ (p. 91).

The orchestra of Ulisse is no mere accompaniment, but ‘an extension of themelodic field, its reverberation in sonic space’ (Pezzati 2008, p. 95). Pezzatinotes some invariance in Ex. 10a: between Calypso’s opening trichord and theaccompanying B�, D and D� in viola, double bass and horn (both (014)), andbetween the vocal line at ‘il tuo cuore’ and the accompanying chord (both(0123)) (p. 97). He might have mentioned the ascent in the oboe, E�–E–F�,which (pace Sellors) does lend these chords a certain direction. And finally, thereis the issue of metre. If the melodic pitches of Ex. 10a possess Scruton’s ‘tone’,this can be attributed to the poetry as much as to the music. The ‘broadening’of the second phrase is not only rhythmic, but results from a weakening of theupbeat/downbeat distinction clearly projected at ‘Son soli’ (note the tenutoaccent) and ‘un’altra volta’. Though the phrase ‘e il mare’ is appropriatelynotated in terms of poetic stress, the loss of the minim tactus as a result of thecrotchet quintuplets causes the word ‘mare’ to stand apart temporally, much asPezzati suggests.

The sea is a constant point of reference in this opera. With regard to itssymbolic value, a clue is given by the sombre colour of the muted brass in thechord that enters at bar 134. In the Spanish poetic tradition, to which Machadosurely responds, the meaning is evident:

Nuestras vidas son los ríos Our lives are riversque van a dar en la mar, that run into to the sea,qu’es el morir which is death28

Later in her scene (bars 65–68), Calypso will ask: ‘Che bramare può l’uomo senon sfuggir la morte?’ (‘For what can man yearn but to escape death?’).29 Butalready at bar 15 of this opera, poetry and music combine to give a sense of theexistential abyss that is the source of Ulysses’ anguish, to be assuaged only in hisfinal epiphany.

142 BEN EARLE


NOTES

1. Following the establishment of a theoretical framework in Part I, Part II ofthis essay will be a review of Eva Mantzourani, The Life and Twelve-NoteMusic of Nikos Skalkottas (Farnham: Ashgate, 2011). xxvi + 414pp. £80(hb.) ISBN 978-0-7546-6310-3; Brian Alegant, The Twelve-Tone Music ofLuigi Dallapiccola (Rochester, NY: University of Rochester Press, 2010). x+ 326pp. £55 (hb.). ISBN 978-1-58046-325-6; and Romano Pezzati, Lamemoria di Ulisse: Studi sull’Ulisse di Luigi Dallapiccola, ed. Mario Ruffini(Milan: Edizioni Suvini Zerboni, 2008). xxvi + 374 pp. €25.00 (pb.). ISBN978-88-900691-3-0.

2. For more details see Nielinger (2006), pp. 125–9.

3. Nono sets texts by condemned resistance fighters of World War II, pub-lished in Malvezzi and Pirelli (1954). That of the seventh movement reads,in full: ‘Addio mamma, tua figlia Liubka se ne va nell’umida terra’ (‘Fare-well, Mother, your daughter Liubka is going into the damp earth’).

4. See Schoenberg (1967), pp. 20–4 and 58–81.

5. For the roots of pitch-class set theory in Babbitt’s ‘source set’, see Schuijer(2008), pp. 95–6.

6. Whittall, in a recent introductory text, at least raises the issue. Yet hissection headings, ‘Schoenberg as music’ and ‘Serialism as music’, aredeceptive, the discussion in both cases swiftly retreating from the concreteto the abstract. See Whittall (2008), pp. 32–6.

7. Schoenberg’s own analytical comments in fact refer just to bars 1–7 as ‘themain theme’. There is no talk of division into parts (see von Rauchhaupt1971, pp. 59–60).

8. Of these four commentators, the only one to pay any attention to thismaterial is Babbitt; yet, characteristically, he homes in on a tiny detail (thechords in the upper instruments at bar 164–172) that is of little musicalsignificance. See Babbitt (1987), pp. 74–5.

9. With more subtlety, Pascall (2002, p. 239) views bars 25–26 as an ‘antici-patory curtain’ to a ternary theme beginning in bar 27.

10. The reference is presumably to Kubelik’s 1953 recording with the ChicagoSymphony Orchestra, most recently re-released in the 55-CD box setMercury Living Presence 2, Mercury Living Presence 4785092 (2013).Adorno uses the 1912 edition (reproduced in Ex. 4a) rather than the 1922revision, which alters some dynamics and adds Hauptstimme markings.

11. Nikos Skalkottas: Violin Concerto, Largo Sinfonico, 7 Greek Dances, BIS,BIS-CD-904 (1998).



12. It is worth seeking out the 1979 performance of the Overture for LargeOrchestra (The Return of Ulysses) (1942–4/1949) by the Danish RadioSymphony Orchestra under Miltiades Caridis, on Greek Symphonies/Griechische Sinfonien Live, Koch Schwann Musica Mundi CD 311 110 H1(1990), or the 1962 premiere of the Violin Concerto (1938) with TiborVarga and the North German Radio Symphony Orchestra Hamburg underMichael Gielen, currently available (as of 16 May 2014) as ‘Tibor Vargaspielt Nikos Skalkottas: Konzert für Violine und Orchester (Uraufführung15.5.1962)’ at http://www.youtube.com/watch?v=mNqpNqfunF8. In thechamber music, the performance of the Third Quartet (1935) by theDartington String Quartet remains unsurpassed, on Nikos Skalkottas:Octet, Eight Variations on a Greek Folk Tune for Piano Trio, String QuartetNo. 3, EMI ALP 2289 (1966) / Argo ZRG 753 (1973).

13. Of the works studied by Mantzourani, the following are apparently notpresent in any UK library: the First String Quartet (1928), the SecondSonatina for Violin and Piano (1929), the First Piano Concerto (1931), theThird Sonatina for Violin and Piano (1935), the Third String Quartet(1935), the First Symphonic Suite for large orchestra (1935), the SecondPiano Concerto (1937–38), the first and third movements of the ThirdConcerto for Piano, Ten Winds and Percussion (1939), and the OuvertüreConcertante from the Second Symphonic Suite for large orchestra (1942–4/1946/1949). Of these, the orchestral scores have never been published.The slow movement of the Third Piano Concerto appeared in 1954 asAndante sostenuto in an edition by Walter Goehr (London: Universal).

14. As well as those mentioned above, there is a Fourth Sonatina for Violin andPiano and a Concertino for Two Pianos and Orchestra (both 1935).

15. The 1966 Melos Ensemble performance has reappeared in an eleven-CDset, Melos Ensemble: Music among Friends, EMI 9185412 (2011). The BISrecording, which couples the New Hellenic Quartet with Scandinavianwoodwind players, is on Skalkottas: Chamber Music, BIS-CD-1124 (2001).

16. Mantzourani appears undecided about the status of bars 12–173. In herbook (Mantzourani 2011) the B1 section is at bars 74–11, but in her earlierarticle it is at bars 74–17 (Mantzourani 2004, p. 82 n. 6).

17. Partly responsible for the confusion must be the editorial forte in the fluteand clarinet at bar 11 in the published edition (London: Universal, 1967),omitted in Ex. 5.

18. The performance, by George Hadjinikos with the North GermanRadio Orchestra Hamburg under Herman Scherchen, has since sur-faced on Berg: Kammerkonzert, Skalkotas [sic]: Klavierkonzert n. 2,Schönberg: Klavierkonzert Op. 42, Arkadia CDGI 768.1 (1993).

19. Bars 761–822 return almost unchanged at both bars 1492–1553 and 184–901, bars 194–28 return unchanged at bars 1684–176 and the opening bars

144 BEN EARLE


http://www.youtube.com/watch?v=mNqpNqfunF8

of the recapitulation (1214–1293) are surely closer to the presentation of thesubordinate theme in the exposition (at 423–501) than anythingSchoenberg would have countenanced.

20. Alegant points out that the cross-partition employed at bars 1–2 is unusualin that ‘the contiguous notes of the row appear horizontally instead ofvertically’ (2010, p. 24).

21. For the poems, see Machado (1947), pp. 134, 162 and 194. This is theedition in Dallapiccola’s library.

22. See the clarinet at bars 8863–8871 of the opera, where the Prisoner, in hisillusory escape, breathes a ‘spring air’.

23. As Alegant points out (2010, pp. 84–6), there is in fact just one hexachordin play in all ten series, Forte’s 6–5 [012367].

24. At bars 10253–1026, order numbers 1–3 are the melody notes in thestrings’ hexachords; in bar 1030, order numbers 7–9 are heard at B, B �and F.

25. B is omitted from the retrograde’s second trichord, and E is displaced fromthe third trichord to join the fourth.

26. For leitmotivic instances of this ‘name-chord’, always at the same pitchlevel, see the Prologue on the final quaver of bar 56; Act I, bars 267, 302and 799; and Act II, bars 510 and 856–858. For the ‘Ithaca’ motive, seeAct I, bars 812–813 and 1070–1071.

27. A twelve-note link is not hard to find. The trichord A–A�–C at bar 11 is atonce order numbers 1–3 of the statement of the ‘Calypso’ row at P8 thataccompanies the singer’s version of the same row at I8, and the final (014)trichord of a derived set, explicated above in relation to bars 1036–1037 inEx. 9.

28. The three lines of poetry are from the Coplas por la muerte de su padre ofJorge Manrique (c. 1440–1479).

29. Again the source is Spanish: Dallapiccola is paraphrasing Unamuno(1972), p. 9.

REFERENCES

Adorno, Theodor W., 1967: Prisms, trans. Samuel and Shierry Weber (Cam-bridge, MA: MIT Press).

——, 1970–86: Gesammelte Schriften, ed. Rolf Tiedemann, 20 vols (Frankfurt amMain: Suhrkamp).

——, 1974: Minima Moralia: Reflections from Damaged Life, trans. E. F. N.Jephcott (London: NLB).



——, 1976: Introduction to the Sociology of Music, trans. E. B. Ashton (New York:Seabury Press).

——, 1998: Quasi una Fantasia: Essays on Modern Music, trans. RodneyLivingstone (London and New York: Verso).

——, 1999: Sound Figures, trans. Rodney Livingstone (Stanford, CA: StanfordUniversity Press).

——, 2002: Essays on Music, ed. Richard Leppert, trans. Susan H. Gillespie et al.(Berkeley and Los Angeles: University of California Press).

——, 2006a: Towards a Theory of Musical Reproduction: Notes, a Draft andTwo Schemata, ed. Henri Lonitz, trans. Wieland Hoban (Cambridge:Polity).

——, 2006b: Philosophy of New Music, trans. Robert Hullot-Kentor (Minneapo-lis: University of Minnesota Press).

——, 2011: ‘Gloss on Sibelius’, trans. Susan H. Gillespie, in Daniel M. Grimley(ed.), Jean Sibelius and His World (Princeton, NJ: Princeton UniversityPress), pp. 333–6.

Alegant, Brian, 2010: The Twelve-Tone Music of Luigi Dallapiccola (Rochester,NY: University of Rochester Press).

Alsmeier, Judit, 2001: Komponieren mit Tönen: Nikos Skalkottas und Schönbergs‘Komposition mit zwölf Tönen’ (Saarbrücken: Pfau).

Augustine, Saint, 1992. Confessions, trans. Henry Chadwick (Oxford: OxfordUniversity Press).

Babbitt, Milton, 1987: Words about Music, ed. Stephen Dembski and Joseph N.Straus (Madison: University of Wisconsin Press).

——, 2003: The Collected Essays of Milton Babbitt, ed. Stephen Peles (Princeton,NJ: Princeton University Press).

Barthes, Roland, 1991: The Responsibility of Forms, trans. Richard Howard(Berkeley and Los Angeles: University of California Press).

Bourdieu, Pierre, 1986: Distinction: a Social Critique of the Judgement of Taste,trans. Richard Nice (London: Routledge & Kegan Paul).

Busoni, Ferruccio, 1962: ‘Sketch of a New Esthetic of Music’, trans.Theodore Baker, in Three Classics in the Aesthetic of Music (New York:Dover), pp. 73–102.

Caplin, William E., 1998: Classical Form: a Theory of Formal Functions for theInstrumental Music of Haydn, Mozart, and Beethoven (New York: OxfordUniversity Press).

Clark, T. J., 2013: Picasso and Truth (Princeton, NJ: Princeton UniversityPress).

Cook, Nicholas, 1990: Music, Imagination and Culture (Oxford: ClarendonPress).

——, 1998: Music: a Very Short Introduction (Oxford and New York: OxfordUniversity Press).

Dallapiccola, Luigi, 1987: Dallapiccola on Opera, ed. and trans. RudyShackelford (London: Toccata Press).

146 BEN EARLE


—— and Mila, Massimo, 2005: Tempus aedificandi: Carteggio, 1933–1975, ed.Livio Aragona (Rome: Accademia di Santa Cecilia; Milan: Ricordi).

De Benedictis, Angela Ida and Mosch, Ulrich (eds), 2012: Alla ricerca di luce echiarezza: L’epistolario Helmut Lachenmann–Luigi Nono (1957–1990) (Flor-ence: Olschki).

DeLio, Thomas, 1985: ‘A Proliferation of Canons: Luigi Dallapiccola’s“Goethe Lieder No. 2”’, Perspectives of New Music, 23/ii, pp. 186–95.

Forte, Allen, 1985: ‘Pitch-Class Set Analysis Today’, Music Analysis, 4/i–ii, pp.29–58.

Haimo, Ethan, 1990: Schoenberg’s Serial Odyssey (Oxford: Clarendon Press).Hoeckner, Berthold, 2002: Programming the Absolute: Nineteenth-Century German

Music and the Hermeneutics of the Moment (Princeton, NJ: Princeton Univer-sity Press).

Holloway, Robin, 2003: On Music: Essays and Diversions, 1963–2003(Brinkworth: Claridge Press).

Horton, Julian, 2005: ‘Schoenberg and the “Moment of German Music”’, MusicAnalysis, 24/i–ii, pp. 235–62.

Jalowetz, Heinrich, 1944: ‘On the Spontaneity of Schoenberg’s Music’, MusicalQuarterly, 30/iv, pp. 385–408.

Kämper, Dietrich, 1984: Gefangenschaft und Freiheit: Leben und Werk desKomponisten Luigi Dallapiccola (Cologne: Gitarre + Laute).

Keller, Hans, 1954: ‘Nikos Skalkottas: an Original Genius’, The Listener,52/cxxxiv, p. 1041.

——, 1977: ‘Description, Analysis and Criticism: a Differential Diagnosis’,Soundings, 6, pp. 108–20.

——, 1981: ‘Schoenberg’s Return to Tonality’, Journal of the Arnold SchoenbergInstitute, 5/i, pp. 2–21.

——, 1994: Essays on Music, ed. Christopher Wintle (Cambridge: CambridgeUniversity Press).

Klumpenhouwer, Henry, 2001: ‘Late Capitalism, Late Marxism and the Studyof Music’, Music Analysis, 20/iii, pp. 367–405.

Leibowitz, René, 1949: Schoenberg and His School: the Contemporary Stageof the Language of Music, trans. Dika Newlin (New York: PhilosophicalLibrary).

Lester, Joel, 1989: Analytic Approaches to Twentieth-Century Music (New York:Norton).

Lewin, David, 1993: Musical Form and Transformation: Four Analytic Essays(New Haven, CT: Yale University Press).

Lyotard, Jean-François, 1988: The Differend: Phrases in Dispute, trans. GeorgesVan Den Abbeele (Minneapolis: University of Minnesota Press).

Machado, Antonio, 1947: Poesie, ed. and trans. Oreste Macrí (Milan: Il Balcone).Malvezzi, Piero and Pirelli, Giovannio (eds), 1954: Lettere di condannati a morte

della resistenza europea (Turin: Einaudi).



Mantzourani, Eva, 2004: ‘Nikos Skalkottas: Sets and Styles in the Octet’,Musical Times, 145/mdccclxxxviii, pp. 73–86.

——, 2011: The Life and Twelve-Note Music of Nikos Skalkottas (Farnham:Ashgate).

——, 2013: ‘Hans Keller, Nikos Skalkottas and the Notion of SymphonicGenius’, Tempo, 67/cclxiii, pp. 33–59.

Mila, Massimo, 1960: ‘La linea Nono’, La rassegna musicale, 30/iv, pp. 297–311.——, 1975: ‘L’“Ulisse”: Opera a due dimensioni’, in Fiamma Nicolodi (ed.),

Luigi Dallapiccola. Saggi, testimonianze, carteggio, biografia e bibliografia(Milan: Edizioni Suvini Zerboni), pp. 31–41.

Milstein, Silvina, 1992: Arnold Schoenberg: Notes, Sets, Forms (Cambridge: Cam-bridge University Press).

Morgan, Robert P., 1991: Twentieth-Century Music: a History of Musical Style inModern Europe and America (New York: Norton).

Nathan, Hans (ed.), 1966: ‘Luigi Dallapiccola: Fragments from Conversations’,Music Review, 27/iv, pp. 294–312.

Nathan, Hans, 1977: ‘Dallapiccola’s Working Methods’, Perspectives of NewMusic, 15/ii, pp. 34–57.

Neumeyer, David, 1986: The Music of Paul Hindemith (New Haven, CT andLondon: Yale University Press).

Nicholsen, Shierry Weber, 1997: Exact Imagination, Late Work: On Adorno’sAesthetics (Cambridge, MA: MIT Press).

Nielinger, Carola, 2006: ‘“The Song Unsung”: Luigi Nono’s Il canto sospeso’,Journal of the Royal Musical Association, 131/i, pp. 83–150.

O’Connor, Brian, 2004: Adorno’s Negative Dialectic: Philosophy and the Possibilityof Critical Rationality (Cambridge, MA: MIT Press).

——, 2013: Adorno (London and New York: Routledge, 2013).Pascall, Robert, 2002: ‘Theory and Practice: Schönberg’s American Pedagogical

Writings and the First Movement of the Fourth String Quartet, op. 37’,Journal of the Arnold Schönberg Center, 4, pp. 229–44.

Pezzati, Romano, 2008: La memoria di Ulisse: Studi sull’Ulisse di LuigiDallapiccola, ed. Mario Ruffini (Milan: Edizioni Suvini Zerboni).

Reich, Willi, 1937: Alban Berg: Mit Bergs eigenen Schriften und Beiträgen vonTheodor Wiesengrund-Adorno und Ernst Krenek (Vienna: Herbert ReichnerVerlag).

Schoenberg, Arnold, 1964: Letters, ed. Erwin Stein, trans. Eithne Wilkins andErnst Kaiser (London and Boston: Faber).

——, 1967: Fundamentals of Musical Composition, ed. Gerald Strang and LeonardStein (London and Boston: Faber).

Schuijer, Michiel, 2008: Analyzing Atonal Music: Pitch-Class Set Theory and ItsContexts (Rochester, NY: University of Rochester Press).

Scruton, Roger, 1997: The Aesthetics of Music (Oxford: Oxford University Press).Sellors, Anthony, 2001: ‘Expressing a Certainty: Musical and Poetic Imagery in

Dallapiccola’s Ulisse’, 2 vols. (PhD diss., University of London).

148 BEN EARLE


Unamuno, Miguel de, 1972: The Tragic Sense of Life in Men and Nations, trans.Anthony Kerrigan (Princeton, NJ: Princeton University Press).

van Hees, Julia, 1994: Luigi Dallapiccolas Bühnenwerk ‘Ulisse’: Untersuchungen zuWerk und Werkgenese (Kassel: Gustav Bosse Verlag).

Vlad, Roman, 1957: Luigi Dallapiccola, trans. Cynthia Jolly (Milan: EdizioniSuvini Zerboni).

von Rauchhaupt, Ursula (ed.), 1971: Schoenberg, Berg, Webern: DieStreichquartette der Wiener Schule; Eine Dokumentation (Munich: VerlagHeinrich Ellermann; Hamburg: Deutsche Grammophon Gesellschaft).

Whittall, Arnold, 1988: [Review of Babbitt, Words about Music], Music & Letters,69/iii, pp. 418–21.

——, 2008: Serialism (Cambridge: Cambridge University Press).Williams, Alastair, 1997: New Music and the Claims of Modernity (Aldershot:

Ashgate).Wintle, Christopher, 2003: ‘An Island Formation in Schoenberg’s Fourth String

Quartet: Notes from a Diary’, in Alison Latham (ed.), Sing, Ariel: Essays andThoughts for Alexander Goehr’s Seventieth Birthday (Aldershot: Ashgate), pp.291–8.

NOTE ON CONTRIBUTOR

BEN EARLE is Lecturer in Music at the University of Birmingham. His publica-tions include articles on the twelve-note music of Arnold Schoenberg, LuigiDallapiccola and Humphrey Searle. His monograph Luigi Dallapiccola andMusical Modernism in Fascist Italy was published by Cambridge University Pressin 2013.

ABSTRACT

Analysts of twelve-note music tend to treat this repertoire as if it were composedof numbers rather than notes. The first part of this article asks how they mighttreat this music as if it were music. A theoretical position is developed via closereadings of the work a number of theorists and philosophers, notably RogerScruton and Theodor W. Adorno. In Part II, the potential of the positiondeveloped in Part I is tested in a review of three books on twelve-note music byNikos Skalkottas and Luigi Dallapiccola.



Documents

Musa 12042