I

Some years ago I was flattered to receive an invitation to address a conference at the Arnold Schönberg CenterFootnote ¹ – flattered, but also troubled, because I have acquired a dubious reputation where Die Wiener Schule is concerned. As someone perhaps best known for a book about Stravinsky, I felt a bit like ‘Zomeone from ze ozzer camp’, as (according to Robert Craft) Richard Hoffmann exclaimed when he opened the door to Schoenberg's house in Los Angeles and beheld Ingolf Dahl.Footnote ² When a recent review of my Oxford History of Western Music noted quite matter-of-factly that ‘Schoenberg’, in my telling, ‘seems to be the object of a personal vendetta’,Footnote ³ it came as no surprise. Many other reviewers have found in my treatment of twentieth-century music a tone that ‘becomes ever harsher, more forceful, and stridently moralistic, […] particularly as it pertains to Schoenberg and his immediate followers’.Footnote ⁴ With respect to something I shall discuss below, namely Webern's orchestral transcription of the six-part ricercar from Bach's Musikalisches Opfer, the same reviewer accused me of coming, like Shakespeare's Marc Antony, ‘to bury Webern, not to praise him’, the point of my analysis being that ‘Webern, in highlighting through his orchestration of the Ricercar a certain motivic coherence which he finds in the work, is distorting Bach, offering not “a revelation, but a revision of Bach's priorities”, and committing a “pseudohistorical fiction”’.Footnote ⁵

Guilty as charged, your honour. I did say all those things, but the accusation of harbouring a moral animus could only come from someone who regards the role of historian to be that of an advocate or at least an exegete, and who in the absence of advocacy can see only hostility. I would rather think that we all recognize the pursuit of understanding to be our common purpose, and that in seeking understanding we come neither to bury nor to praise. Maybe the cold war is over by now where Schoenberg and Stravinsky are concerned. Maybe we can all agree that the primary weapon in pursuit of understanding is scepticism towards all claims; that the process of advancing understanding involves the testing of all claims against evidence; that the best outcome is mehr Licht; that in order to let the light in one must dispel obstructions; that the primary obstruction in the way of scholarly light is unexamined myth; and that myths arise when the words of historical actors are taken to be the words of oracles rather than witnesses, and when their intentions are equated with their achievements.

For all that, ‘pseudohistorical fiction’ is indeed the subject of this essay, and in particular the way in which pseudohistorical fiction can serve a perfectly legitimate purpose for composers, performers, and listeners – perhaps even for critics. It is only historians who are pledged to expose what they take to be pseudohistory. But in the case of Schoenberg, Berg, and Webern the pseudohistory is now part of history and needs to be understood in the light of the legitimate purposes that it once served for these great musicians, and what it enabled them to achieve.

In constructing their idiosyncratic view of the past, the Wiener Schule was not unique, nor was it even unusual. I was momentarily surprised to find, in a book I was recently given for review, the following comment about the article on ‘Chromaticism’ that Pierre Boulez wrote around 1958 for the Fasquelle Encyclopédie de la musique, in which he waxed enthusiastic about the madrigals of Cipriano de Rore and Carlo Gesualdo: ‘It appears as nothing short of amazing’, the author, Glenn Watkins, exclaims (no doubt thinking of ‘Schoenberg est mort’), ‘that a leader of the radical avant garde could simultaneously be making a call for an awareness of the musical past.’Footnote ⁶ But of course it only appears that way. As Watkins goes right on to show, nothing could be less amazing. What else have avant gardists ever done, throughout the history of music? The self-conscious avant-garde stance implies a high awareness of history. Without it the avant garde could have no consciousness of itself as such. Historical awareness is something all composers have been cursed with since the middle of the nineteenth century, thanks in the first place to the neo-Hegelian historians of that time (with Franz Brendel at their head) and in the second place to their antagonists, headed by Eduard Hanslick, the first professor of music history at the University of Vienna. Brendel's historiography underwrote the propaganda of the Neudeutsche Schule (which was christened by Brendel himself), and the historiographical notions of the Neue Wiener Schule, as most authoritatively enunciated by Webern, who earned his doctorate in musicology at the same university under Guido Adler, Hanslick's successor, were adapted directly from Brendel's precepts.

But appeals to ancient precedents to justify stylistic, technical, and aesthetic innovations in music have a much longer pedigree than that. The radical Florentine humanists who first theorized and then midwifed the birth of opera at the end of the sixteenth century claimed to be reviving the drama of ancient Greece, just as Wagner would claim 250 years later. Caccini's Nuove musiche were founded on the example of a vividly imagined vecchia musica, an invented musica antica. Monteverdi justified his seconda prattica by claiming that, whereas the musicians of the ars perfecta could claim a distinguished heritage going back to ‘the first to write down music for more than one voice, later followed and improved upon by Ockeghem, Josquin des Prez, Pierre de la Rue, Jean Mouton, Crequillon, Clemens non Papa, Gombert, and others of those times’, he, Monteverdi, could claim a heritage that began with Plato.Footnote ⁷ And Plato beats Ockeghem any day of the week.

Nor did such appeals end with the Neue Wiener Schule. It has long been a commonplace that organizations specializing in the performance of very new music also perform very old music. The first concert given by Boulez's Domaine musical in January 1954 began with Bach's Musikalisches Opfer (not in Webern's orchestration, yet nonetheless anything but a fortuitous choice) and continued with Nono, Stockhausen, Webern, and Stravinsky. The second concert, in February, began with isorhythmic motets by Du Fay and ended with madrigals by Monteverdi and Gesualdo – works that, by Boulez's explicit stipulation, had ‘a particular relevance for our time’.Footnote ⁸ The Los Angeles Evenings on the Roof concerts, begun by Peter Yates and continued by Lawrence Morton as the Monday Evening Concerts, featured a similar mixture. Stravinsky heard a lot of early music at these events, beginning in 1944, and immediately began imitating it in such works as his Mass (1944–8), the Cantata of 1952, and In Memoriam Dylan Thomas (1955). When Robert Craft joined Stravinsky's ménage in 1948, he immediately began conducting for Morton, and the pieces he rehearsed and performed at the Monday Evening concerts were the ones he immediately began recording for Columbia records – the legendary Webern set, Stockhausen's Zeitmasse and Boulez's Marteau sans maître, quantities of Schoenberg, Berg, and Stravinsky, but also Gesualdo, Schütz, Monteverdi, and Bach.

An example from my own lived and formative experience is the Group for Contemporary Music, the first university-based new music ensemble (the first of countless such American groups), formed by the composers Charles Wuorinen and Harvey Sollberger at Columbia University when I was an undergraduate there and an avid attender of their concerts. The first programme, on 22 October 1962, began with Christes Crosse, a formidable demonstration of metric proportions from Thomas Morley's Plaine and Easie Introduction to Practicall Musick of 1597. Thereafter the opening slot was always reserved for old music with ‘a particular relevance for our time’. At the second concert it was ‘Music from the Mensural Codex of Nikolaus Apel (late fifteenth century)’, the title taken directly from the modern edition in the series Das Erbe deutscher Musik; at the third it was Wuorinen's Bearbeitungen über das Glogauer Liederbuch, after another volume of Das Erbe deutscher Musik, titled in German for some obscure but nevertheless characteristic reason. At the fourth and fifth concerts the opening pieces were what we would now call ars subtilior chansons by ‘Matheus de Perusio’ and Anthonello de Caserta (their names taken again directly from the modern editions in which the directors found them), including Matteo's Le greygnour bien, the redoubtable opening selection in Willi Apel's French Secular Music of the Late Fourteenth Century, published in 1950 by the Medieval Academy of America with a preface by Paul Hindemith, a book that was a veritable bible – all right, a pony – for generations of graduate students learning medieval notation. It was an inevitable choice at a time when academic composition was itself an ars subtilior, obsessed with rhythmic complication.

II

Mentioning Hindemith is a reminder that the Wiener Schule's appeals to historical precedents were matched – indeed, more than matched – by those of their principal antagonists, in the first place by the composer Schoenberg derided as ‘der kleine Modernsky’ with his ‘echt falsches Haar wie eine Perücke […] ganz der Papa Bach’.Footnote ⁹ So I will begin with Stravinsky, and with some trepidation, because Stravinsky can so easily take over any essay of mine; but I shall try to keep him in perspective. And the first thing I would point out about Stravinsky is that, like the Wiener Schule, he justified his work by appealing to old music at all phases of his career, not just the one we now call ‘neoclassical’.

In fact, during his earliest period, the so-called ‘Russian’ period, he appealed to the earliest music – the music of Russian, or ur-Slavic, prehistory. Before Stravinsky, most Russian composers who quoted folk songs quoted the long, lyrical songs known in Russian as protyazhnïye pesni (‘drawn-out’ songs) – the beautiful tunes we remember from the work of Chaikovsky and the group the Wiener Schule would have called das mächtige Häuflein, either quoted directly from the singing of peasants (as particularly in the work of Balakirev) or from published transcriptions (as particularly in the work of Rimsky-Korsakov) or expertly imitated (especially by Borodin). Stravinsky quoted and imitated lyrical songs too, especially in Firebird, his first ballet. But in The Rite of Spring he turned (very likely at the suggestion of his collaborator, the painter and archaeologist Nikolai Roerich) to what was considered the much older repertory of kalendarnïye pesni or calendar songs, work and holiday songs that ritualized and memorialized the ancient pagan agrarian calendar.Footnote ¹⁰ He mined them particularly from a book of Lithuanian wedding songs collected by a priest named Anton Juszkiewicz, Roerich probably having told him that sacrificial rituals had lasted longest in White Russian and Baltic territories on the western fringe of the Russian Empire. Stravinsky probably did not know that they were wedding songs, since the Polish and German title pages did not identify them as such, but it made no difference to the style of his ballet, since many wedding songs were fashioned like calendar songs out of tiny repetitive modules of restricted ambitus. The fantastically influential ostinato-driven style of The Rite of Spring was founded on a repertoire of folk songs chosen for their antiquity. Their presumed status as authentic remnants of pagan culture lent Stravinsky's music an irresistible aura of authenticity that certainly abetted its – and his – rise to eminence, and eventual dominance, in the world of modern music.

Another particularly ancient vein on which Stravinsky consciously drew was that of children's songs, in several opuses from his so-called ‘Swiss years’, corresponding to the duration of the First World War. These are the sort of homely two-or-three-note chants children seemingly spontaneously adopt – or transmit to one another – in every culture. These chants are often thought of as a sort of Ur-Musik. As one recent researcher, Pieter van der Merwe, has put it, ‘The best place to reconstruct the evolution of melody is the nursery.’Footnote ¹¹ But that is a preposterous claim, being just an inversion of the old biologist's bromide about ontogeny recapitulating phylogeny, in which the scholar presumes to deduce phylogeny from ontogeny, as if the first music produced by infants represents, in actual historical terms, the infancy of music. There is no evidence at all to substantiate such a purely conceptual assumption. A range of music from children's chant all the way to Brian Ferneyhough and Helmut Lachenmann can be observed at the present moment. A comparable range could have been observed during the infancies of Ferneyhough and Lachenmann, when those worthies were themselves presumably confined to children's chanting. As early as there are written records of music, there is evidence of a broad stylistic range.

The same caveat applies to the assumption that the simple, melodically restricted, and rhythmically repetitive kalendarnïye pesni on which Stravinsky drew to evoke primitive humanity in The Rite of Spring are older than the far more elaborate and ‘artistic’ protyazhnïye on which the nineteenth-century Russians liked to draw. The equation of simpler with older – or, for an assumption that has also influenced a lot of modern music, the supposition that pentatonic is more ancient than diatonic – amounts to no more than arbitrarily turning an observed (and interpreted) horizontal axis of simple-to-complex ninety degrees to produce an imaginary vertical axis of early-to-late. As many readers will recall, the earliest actual document of musical notation known to history, the ‘Hurrian’ (Sumero-Babylonian) hymn transcribed from a cuneiform tablet by Anne Draffkorn Kilmer and Richard L. Crocker in 1974 and dated somewhere between 1225 and 1400 BCE, in no way conforms to these conjectural models of musical prehistory. It is polyphonic, for one thing, and fully diatonic (i.e., with semitones).

In fashioning the music of The Rite of Spring, then, Stravinsky was relying on a pseudohistorical fiction. And he was altogether correct to do so, since what he was after was not historical accuracy but historical verisimilitude – ‘truthiness’, as it is now called in the United States, after a popular fake-news comedian named Stephen Colbert. The ring of truth was all he wanted or needed, and when we listen, that's all we need as well.

During his neoclassical period Stravinsky competed for truthiness with Schoenberg, and their battle over it is a well-known story. But Stravinsky's latest phase, the so-called serial one, was just as strongly bolstered by alliances with ancient prototypes – far more ancient than the ones his neoclassicism invoked. In December 1952, asked by a reporter from the New York Herald Tribune to name his favourite composer, Stravinsky took his interviewer by surprise, the way he loved to do, and started rhapsodizing about the music of Heinrich Isaac.

The volume in question, edited as a doctoral exercise by Louise Cuyler, had been published in 1950 by the University of Michigan Press. Stravinsky had been playing it over, usually in tandem with Robert Craft at the keyboard, for about a year by the time he made this declaration of love; in fact he was playing it in January 1952 at the same time that he was attending Robert Craft's rehearsals of Schoenberg's Septet-Suite, Op. 29, and composing the Cantata, the first work in which he employed serial, if not yet twelve-note, techniques.Footnote ¹³ Mention of Isaac of course brings Webern to mind, since Webern had also edited a volume (the second) of Isaac's mammoth collection of polyphonic mass propers as his doctoral dissertation under Adler in 1908. It was published a year later in Adler's series of monuments Denkmäler der Tonkunst in Österreich. In 1956 Universal Edition sent Stravinsky a photocopy of this volume, which Stravinsky had requested (according to his biographer Stephen Walsh) ‘in order to study its elaborate canonic settings of the mass liturgy’.Footnote ¹⁴ But Stravinsky's acquaintance with a wide variety of medieval and Renaissance music, and his creative appropriations from it, go back a good bit further, probably to the year he spent at Harvard University in the Charles Eliot Norton Chair of Poetry (1939–40), during which he delivered the lectures that were subsequently published as Poétique musicale. While at Harvard he met Archibald T. Davison, professor of choral music, who, together with Willi Apel, an émigré from Nazi Germany then employed at Harvard as a lecturer, was in the process of compiling the Historical Anthology of Music, a two-volume compendium of examples for use in music history classes, along the lines of Arnold Schering's Geschichte der Musik in Beispielen and other German textbooks of that type.

‘HAM’ served as a trusty vade mecum to generations of American college music majors from the time of its publication in 1946. That Stravinsky had it, read it, and used it is evident to anyone else who ever had it, read it, and used it, because so many of the examples of early music to which he made archly knowing reference in his books of conversation with Robert Craft can be found in this lowly source or others equally lowly. They include some very esoteric items, like Par le deffaut, a chanson by Baude Cordier; or the ‘beautiful seconda Stravaganza’ by Giovanni de Macque; or ‘that marvelous second Agnus Dei (the three-voice one) in [Josquin's] Missa l'Homme armé’, which reflected Apel's special interest in the sort of musical oddities that would appeal to a modernist, who would have regarded them as pioneering or prescient – even prophetic inasmuch as they seemed to forecast his own coming.Footnote ¹⁵ And they also included the first Gesualdo madrigal Stravinsky ever copied out, according to Craft, namely Io pur respiro from the sixth book.

During his American years Stravinsky collected pretty much the whole corpus of standard textbooks on early music. When I was working in his archive, during the temporary custodianship of the New York Public Library in 1983, I came across a letter from Faber and Faber to Stravinsky, dated 18 November 1960, confirming shipment of Carl Parrish's A Treasury of Early Music, another anthology that anyone who studied music in college or university in the 1950s or 60s would have used. In the very first piece of writing Robert Craft ever published on Stravinsky, an essay on the Mass published in 1949, he raised what he called ‘the problem of assimilation’ of appropriate models for a work intended for actual Catholic devotional use. ‘The Credo’, he wrote,

That bit of pseudomusicological nonsense had to be something Stravinsky himself imparted to Craft, I thought, and, recalling my own scholarly training, I had a hunch that the delightfully abstruse Latin term must have come from the chart, adapted from Dom Gregory Suñol's Introduction à la paléographie musicale gregorienne, that Gustave Reese had printed in his Music in the Middle Ages of 1940, a somewhat more advanced textbook used by graduate students.Footnote ¹⁷ (I had already suspected that the opening of Stravinsky's ballet Orpheus (1946–7) was inspired by Reese's table of Greek scales.)Footnote ¹⁸ I wrote to Craft, with whom I was then in frequent correspondence, and he wrote back confirming that ‘Stravinsky did own Reese, Apel and Davison, Parrish and Ohl [Masterpieces of Music before 1750], […] and in fact every anthology of “old” music that he could find. Also every recording, from Anthologie Sonore down or up and including private ones.’Footnote ¹⁹

Almost every piece Stravinsky composed in America, beginning with Orpheus and the Mass, both completed in 1948, bears some resonance from early music – that is to say, from ‘pre-Bach’ or pre-Baroque repertoires. Sometimes the connection does indeed seem to be a pseudohistorical validator of Stravinsky's stylistic departures, the most obvious case being that of the Canticum sacrum ad honorem Sancti Marci Nominis of 1955, the first piece of Stravinsky's to use – albeit intermittently – a full Schoenbergian twelve-note row. It was composed for performance in the Venetian basilica of San Marco, and the non-serial sections of the piece are obviously patterned on earlier music for that great church. At its first performance, at San Marco on 13 September 1956, it shared the programme not with other religious works by Stravinsky, as originally requested by Alessandro Piovesan, the impresario of the Venice Biennale, but rather with a group of compositions that, in Stravinsky's words, would ‘effectively situate’ his new piece: a ricercare by Andrea Gabrieli performed on four trombones, a Latin psalm setting by Schütz (a pupil of Giovanni Gabrieli), a motet by Gesualdo, and the setting of Lauda Jerusalem from Monteverdi's Vespers collection of 1610. (The Gesualdo piece was jettisoned at Piovesan's request in favour of Giovanni Gabrieli's impressive concertato motet In ecclesiis. Piovesan argued that the substitution would produce a programme that was ‘much more organic, more stylistically perfect’, but Craft surmised that the real reason was that ‘Venice would not hear of a Neapolitan in the precincts of San Marco’.)Footnote ²⁰

Whether the static pseudocounterpoint of quick repeated brass chords in the Canticum sacrum should be regarded as an obeisance to Monteverdi's stile concitato or to the Gabrielis' spatially oriented chori spezzati, or more simply and practically as an accommodation to the extremely live acoustics in San Marco (measurable on recordings as a six-second reverb), Stravinsky's work is indeed emblematic of our theme. In Glenn Watkins's well-chosen words, it is

III

Stravinsky's response to the past was worth a fairly detailed look because it provides us not only with an embodiment of our theme but also with an appropriate foil against which to examine the very different response of the Neue Wiener Schule. And that examination, I hope, will justify my title. For ‘alte Musik’ as imagined by the Austro-German musicians of the Wiener Schule's generation is quite different from ‘early music’ as we use the term today, especially (but not only) in the Anglo-American cultural sphere. Indeed, the two terms imply opposite perspectives. To speak of ‘old music’ is to look back on it from the actual or conceptual present, the period we inhabit. Early music means ‘young music’, not old, and to speak of it is to look forward to the present from an imagined past, or even to attempt to inhabit that imagined past without reference to the present.

A practical example of that difference can be observed by recalling the career of the lutenist Thomas Binkley, one of the great early music performers of the twentieth century. He was an American midwesterner, born in Cleveland, Ohio, who graduated in 1954 from the University of Illinois, where he was introduced to early music by George Hunter (1918–2011), a Hindemith pupil and a great unsung pioneer of the movement, who ran the campus Collegium Musicum. Binkley then went to Munich to study musicology and, together with another alumnus of Hunter's collegium, the string player Sterling Jones, founded a performing group which, following normal German usage, he called the Studio für alte Musik. However, by the time the group issued its first recording (the first of about fifty), featuring Binkley, Jones, the British tenor Nigel Rogers, and the singer Andrea von Ramm (who was a native German speaker, but not a native German, having been born and brought up in then-independent Estonia), Binkley had rechristened it the Studio der frühen Musik, which at the time was not idiomatic in German.

Why the change? I wish I could have interviewed Binkley on this score, but I never thought to do so while I knew him. It seems to me a virtual certainty, however, that it had to do with his perception of the differences to which I am calling attention. Even before so-called early music performers began challenging so-called mainstream performers in the standard concert repertory, the term ‘early music’ implied not only a repertoire (increasingly broad and ultimately impossible to define) but also a performance practice (much easier to define). And that performance practice went back to Stravinsky in stark opposition to the performance practice to which the composers of the Wiener Schule remained faithful to the end of their lives.

It is true, of course, that Stravinsky was far longer-lived than the composers of the Wiener Schule, and that even Schoenberg, the longest-lived of the three, did not live into the period when the early music performance style became well enough defined to describe as such, or into the period defined by Stravinsky's so-called third or ‘serial’ phase. But Stravinsky's ideas on performance were given a particularly forceful, conspicuous, and influential formulation in his Poétique musicale. (Indeed, the sixth lecture in that ghostwritten series, on performance, is the only one that, I am confident, reflected Stravinsky's own personal convictions.)Footnote ²² That lecture was delivered in 1940 and first published in 1942; and Schoenberg gave a forceful if indirect rebuttal to it a few years later, which shows that Stravinsky's ideas and Schoenberg's did coexist in time in active contention, and that comparing them is no anachronism. Here, first, is Schoenberg's. It comes from a three-page typescript, composed in English in 1948, entitled ‘Today's Manner of Performing Classical Music’. It was tidied up by Dika Newlin in 1950 and published for the first time in 1975 in the expanded edition of Style and Idea, edited by Leonard Stein. It is a vociferous protest, and it begins with a fusillade of startling assertions:

Although this screed was composed in America in 1948, it reflects a view more accurately associated with the Germany of 1928 or so, when Amerikanismus, which meant commercialization and mechanization as well as modernization, took a lot of credit as well as blame for changes that were taking place in all of the arts. Those who favoured it called it not Amerikanismus but neue Sachlichkeit (another term that Schoenberg hated with a passion). The remarks about the body and about drugs, of course, as well as Schoenberg's use of the word ‘primitive’, added a hint of racial revulsion, which can be read in its natural habitat, so to speak, in the responses to Krenek's Jonny spielt auf and other Zeitopern by Hindemith, Weill, Max Brand, and the transplanted American George Antheil.Footnote ²⁴ The opposition to Amerikanismus remained alive – anachronistically alive, I would say – not only in Schoenberg's thinking and writing, but also in the fetishized writings of T. W. Adorno, and has therefore survived as a sort of fossil into the discourse of our own time.

Having seen how Schoenberg accounted for what he despised in the performance practice of his later years, let us now see how he accounted for the practice it had supplanted. ‘Music should be measured’, he conceded:

Schoenberg's reference to blood is surely an allusion to Brahms, who is said to have said that he provided metronome marks only when ‘good friends have talked me into putting them there, for I myself have never believed that my blood and a mechanical instrument go well together’.Footnote ²⁶ It may seem paradoxical that Schoenberg should endorse the notion that tempo should be dictated by the pulse rate when he so bitterly denounces the musical impulses of ‘a dancer inspired by his body’: but that dancer has been narcotized by the music of a dance band and can no longer respond spontaneously to emotional stimuli. Schoenberg's allegiance is clearly to what Wilhelm Worringer called vitalism, the notion that art expresses the life force, the source both of our pulse and of our emotions.Footnote ²⁷ In his transcription of Bach's Saint Anne Prelude and Fugue, Schoenberg imposes his blood pulse on the music with a bilingual multitude of tempo variations:

There is every reason to suppose that, in adding all of these obviously non-authorial directions, Schoenberg meant not to supersede Bach's own performance practice but to restore it. The main reason for supposing this, of course, is that Bach was for Schoenberg the fountainhead of the German musical spirit, and that spirit was the very embodiment of vitalism.

The impulse Wilhelm Worringer opposed to the vital is what he called the geometrical. He associated it primarily with the art of early Christianity, but also with the modern art of his day. In a long essay that I published more than twenty years ago,Footnote ²⁸ I associated that geometrical impulse with the early theorists of post-World-War-I anti-Romantic modernism, especially Ezra Pound and T. E. Hulme, thence with the musical practices of Toscanini and then Stravinsky, and finally with those of ‘early music’ as exemplified in the performances of the 1960s to the 1980s, the ideal to which Thomas Binkley seemed to pay tribute when he changed the name of his performing group from Studio für alte Musik to Studio der frühen Musik. (This seems especially evident when you consider that the group's speciality, particularly at the outset, was the music of the very periods that furnished Worringer with his chief examples of geometrical art.) I won't rehearse the whole argument again here, but that vital/geometrical distinction – rather than Amerikanismus or the music of dance bands – is where I would locate the source of the performing style that Schoenberg deplored. Regardless of its source, it is the philosophy of performance Stravinsky espoused in the Poétique musicale, where one can find an abundance of explicit endorsements of the same performance practice that Schoenberg excoriates, and many references to old music (and, in particular, to Bach) as exemplifying the geometrical in contradistinction to the vital.

Stravinsky, you will recall, subdivided ‘performance’ into two subcategories, namely execution and interpretation. In one especially notorious sentence Stravinsky defined ‘the idea of execution’ as ‘the strict putting into effect of an explicit will that contains nothing beyond what it specifically commands’.Footnote ²⁹ This accords, of course, with the Toscaninian injunction to play ‘com'è scritto’, which means precisely the suppression, exactly as Schoenberg complained, of ‘all unnotated change of tempo and expression’. Stravinsky made it entirely explicit when he wrote that

When one listens to his recordings, especially the recordings he made of music by other composers (for example, his incredibly dry and unyielding performance, with his son Soulima, of Mozart's Fugue in C minor for two pianos, K426, recorded in 1938), one sees that Stravinsky meant what he said. And that Schoenberg, speaking for his ‘school’, also meant what he said is evident from, say, Webern's recorded performance of his own arrangement of Schubert's Deutsche Tänze, D820.

Here too the context in which Stravinsky's earliest twelve-note compositions were performed becomes pertinent. When Movements for piano and orchestra had its première at Town Hall in New York on 10 January 1960, it was accompanied by works by Bach, Monteverdi, Schütz, and Gesualdo – almost precisely the same company the Canticum sacrum had kept in Venice in 1956 – all conducted by Craft. Paul Henry Lang, reviewing the concert in the New York Herald Tribune, criticized these performances almost exactly as one might have imagined Schoenberg criticizing them. It was vitalism vs geometry redux, with the pulse once again the main point of contention. ‘So detached and bare’ were Craft's performances, Lang wrote, ‘that a stethoscope could not have picked up a heartbeat in Schütz or Monteverdi’.Footnote ³¹ But it was the performance of Gesualdo's chromatic motet Aestimatus sum that really provoked the critic:

Clearly Robert Craft was a bit ahead of the game in so modernizing his approach to early music. Understandably stung by Lang's harsh words, he defended himself in a letter to a friend against Lang's accusation that his performances were ‘cold, cold, cold’ by claiming that ‘they are, I think, quite “cool” (which means very hot)’ – in the sense, that is, of sophisticated, up-to-date, ‘with it’.Footnote ³³ Love his performances or hate them, one has to agree that Craft was correct by the lights of his time.

But the vagaries of performance fashion were only an epiphenomenon, a surface ripple atop the real gulf that divided and distinguished the attitude of Stravinsky from that of the Wiener Schule, and this is the real distinction between ‘alte Musik’ and ‘early music’. The difference between the two represents an even more profound rift in the aesthetics of the twentieth century – namely that between familiarization and what, in the twentieth century, became known as defamiliarization (from the Russian ostraneniye, as coined by the Russian Formalists; compare the German Entfremdung). That is why ‘early music’ is so heavily invested in what Hans Ferdinand Redlich so unforgettably derided as musealer Klangmaterialismus.Footnote ³⁴ Its proponents never called it that, of course. They called it ‘authenticity’. But the sense of authenticity promoted by ‘early music’ derives its aura from a sense of difference – and distance – from what is standard and, by implication, unconsidered. ‘Alte Musik’, as understood at the time of the Wiener Schule, wished to hear Bach performed on the same instruments as Brahms, with the same nuances and the same articulations, because the sense of authenticity that ‘alte Musik’ values derives its aura from a very considered sense of kinship. Klangmaterialismus is interested in the sounding surface. Timbre, texture, articulation – all of these matter at least as much as pitch or rhythm. At its most extreme, as enunciated, say, by Malcolm Bilson, ‘early music’ puts the instrument before the performer.Footnote ³⁵ That was Klangmaterialismus at its very limit (and, as devotees of early music know very well, there has been retrenchment since). ‘Alte Musik’, like the Wiener Schule, is idealistic, not materialistic. What matters is what lies below the surface – what Schoenberg, once he had learned the word in English, loved to call the ‘subcutaneous’ level, where motives lurk.

IV

When Stravinsky turned to the past he looked for difference. He looked for ‘early music’. When the Neue Wiener Schule turned to the past they looked for sameness. They looked for ‘alte Musik’. For Stravinsky the past was a tasting menu. He rummaged and browsed – often quite literally, as I have been relating, in standard reference sources – in search of frissons. And when he found them he did not proclaim a direct historical line from, say, Gesualdo to himself. He was open to everything, invested in nothing. What he was looking for in his eclectic and voracious way was in fact an escape from his immediate heritage, of which he had grown ashamed.

Not so Schoenberg. A moment of unintentional hilarity in Schoenberg's writings – alongside the many moments of intentional hilarity for which I cherish them – comes in ‘Nationale Musik’, an essay in two parts dating from 1931, in which he asserted with truly frightening aggressiveness the hegemony of German music and his own position at its head. ‘Remarkably’, he wrote, ‘nobody has yet appreciated that my music, produced on German soil, without foreign influences, is a living example of an art able most effectively to oppose Latin and Slav hopes of hegemony and derived through and through from the traditions of German music’. That's not the funny part. That's the scary part, Schoenberg at his least lovable. The funny part is when he lists his teachers: ‘primarily Bach and Mozart, and secondarily Beethoven, Brahms and Wagner’. After listing the technical devices he learned from each of them, he adds, ‘I also learned much from Schubert and Mahler, Strauss and Reger too.’ And then the sentence that always makes me laugh: ‘I shut myself off from no one.’Footnote ³⁶

There you have German universalism at its most explicit; and I raise the matter here only because that kind of parochialism is nowhere more rife at present than in the Anglo-American academy, which is why I have been fighting it so hard. But it certainly characterized the view of the Wiener Schule towards ‘alte Musik’ when it came to making appropriations. Consider: Schoenberg made arrangements of Bach, Handel, Monn, and Brahms (all right, Johann Strauss too); Webern of Bach and Schubert (and Johann Strauss too); Berg also did his Strauss arrangements for the Verein für Musikalische Privataufführungen; in his Violin Concerto he quoted Bach, and in his Lyrische Suite he quoted Wagner. These composers cut themselves off from many, indeed from most, reminding me of the old Jewish saying, attributed to Reb Eliyahu ben Shlomo Zalman, the so-called Vilner gaon (the Bible prodigy from Vilna), that the best way for a man to preserve his purity is never to leave his house.

One can easily multiply quotations from Schoenberg, Berg, and Webern testifying to their insularity. Memoirs by Webern's pupils abound especially in exasperating anecdotes. To one he says, ‘If we want to understand philosophy we must turn to the ancient Greeks, and if we want to understand music we must turn to Mozart, Haydn, Beethoven, and the other great masters of the Austro-German tradition.’ To another pupil he says that ‘leading composers of other lands are but pale reflections of Germanic masters: Berlioz a French Beethoven, Tchaikovsky a Russian Schumann, Elgar an English Mendelssohn.’ To a third, Humphrey Searle, whom I name in order to identify him as English, he said, looking up from the score of Bruckner's Seventh Symphony, ‘Could your Elgar write an arch of melody like that?’Footnote ³⁷ The pupils were by turns shocked and amused, but always tolerant of the great musician's foibles, and I understand that perfectly well. It will touch off no vendetta in me. As one of the pupils wrote, ‘One may look upon this as a most confining attitude on Webern's part – not a single reference to Berlioz, to Verdi, to Mussorgsky, or even to the Liszt–Strauss vein in German music. But Webern's horizon was wholly filled by the music from Bach, through the Viennese classics to Wagner, Brahms, and Mahler; he found his complete personal affirmation as composer therein.’Footnote ³⁸ Whatever attitudes produced his music were the right attitudes for that purpose, and are in that sense immune from critique.

But we have a different purpose, that of understanding Webern's, Berg's, and Schoenberg's relationship to the past in its cultural context and in its contribution to musical discourse, and that does require critique, especially as their actual compositions – as opposed to their commentary – participate in the discourse. I have in mind the references one finds to the BACH cipher in such works as Schoenberg's Variations, Op. 31, where it sounds forth as a proclamation; or his Suite for piano, Op. 25, a work that deserves to be compared with Stravinsky's contemporaneous ‘back to Bach' exercises, where it is built into the note row (at the end, in reverse, so that it will be quoted literally in untransposed retrogrades), or in Webern's String Quartet, Op. 28, where it wholly pervades the writing because the entire row consists of two transpositions of the cipher surrounding a retrograde inversion of it (though, as Webern boasted in the analysis that he prepared for Erwin Stein, ‘only secretly because […] the original form never occurs in this brazen transposition', which is to say, without transposition).Footnote ³⁹

These are all proprietary references. They celebrate heirship, ownership, a sense of belonging, not exotic discovery – and that sense of direct heirship, of being in a line of succession, is founded, as any musicologist today will tell you, on a pseudohistorical fiction. The idea of a straight evolutionary line ‘from Bach, through the Viennese classics to Wagner, Brahms and Mahler', to recall the words of Webern's pupil, Arnold Elston, has been roundly refuted by modern scholarship, most pointedly by Daniel Heartz, who in an explosive four-page article of 1969, ‘Approaching a History of 18th-Century Music', showed succinctly but conclusively that the German tradition that supported Bach and Handel ended with them, whatever the Italianate admixtures, and the tradition that fed the Viennese classics actually originated in Italy, in comic opera. The line asserted by proponents of German universalism as pure and unbroken was in fact an arbitrarily spliced product of hybridization. We all know that now. No serious scholar would today endorse the ideas that sustained the historical self-image of the Wiener Schule, as promulgated, for example, in Webern's Der Weg zur neuen Musik.

But what of that? Neither Schoenberg nor Webern invented the pseudohistorical fiction to which they subscribed; and, as far as they were concerned, it was historical fact, as well as an article of devout belief. That belief is what shaped their approach to ‘alte Musik’, and therefore we have got to understand it, if we are to do justice to our theme. Who did invent it, then? Who did the splicing? Was it Beethoven, with all the Bachian fugues and Handelian pastiches of his post-Napoleonic phase? Was it Mendelssohn, who in 1829 brought the vocal works of Bach back into the public sphere and made Bach, through the new genre of German festival oratorio, a national emblem for all German speakers, not just Lutherans? Was it Franz Brendel, who in 1859 decreed that all who served the progressive ideals of German art (even if, like Berlioz and Liszt, they allowed themselves to be born elsewhere) were members in spirit of the universal Neudeutsche Schule? Actually, I would name as chief author of the myth of unbroken continuity the Neudeutsche Schule's bitterest enemy, Johannes Brahms, because Brahms performed the act of hybridization not just in words or in pastiche but in actual creative deed. In his Variations on a Theme by Haydn (1873), and later in the finale of his First Symphony (1876), Brahms adapted Bachian compositional techniques to Viennese classical forms.

In the last of the Haydn variations (and the fact that the theme may not actually be by Haydn has no bearing on this argument) Brahms fashioned the first phrase of the theme into a ground bass. In so doing he did something that Haydn never did in any variation set. In fact, I think it is pretty clear that Brahms was explicitly invoking Bach's famous organ Passacaglia in C minor, because, like Bach in that piece, Brahms suddenly and quite unorthodoxly transposed the ground bass to the soprano voice in preparation for the climax. By thus making Haydn shake hands with Bach, Brahms closed the historical gap to which Heartz drew such forceful if belated attention, and provided the precedent that made the pseudohistorical fiction true for his successors.

Schoenberg recognized this, of course: both in ‘Brahms the Progressive’ and in his radio talk on the Cello Sonata, Op. 99, he educed what I like to call his technique of micromanaging motives (or ‘developing variation’, to use his term) out of themes by Brahms.Footnote ⁴⁰ And then, in his arrangement of the six-part ricercar from Das musikalisches Opfer, Webern used his trademark pointillistic instrumentation not merely for colouristic purposes, as is often assumed, but rather analytically, to extract hidden motives that in effect turn Bach into a Brahms – or a Schoenberg – in training.Footnote ⁴¹ What might, on the kaleidoscopic surface of Webern's transcriptions, appear to be a very orgy of Klangmaterialismus is in fact at the opposite extreme, dredging up the most arcane and abstract relationships that testify to the music's ideal structure, the occult level at which lay the true kinship between Bach and himself. Webern in effect put the two ends of the invented tradition in contact, thus conclusively unifying and concretizing it. Although he would of course have claimed – and believed – that he was only making more explicit that which was already fully implicit in Bach's own writing, Webern's arrangement counts – in my view, anyway – as a Verbesserung, an improvement, to use the word Schoenberg applied to his Concerto for String Quartet with respect to its model, Handel's Concerto Grosso, Op. 6, no. 7, exactly in the sense Schoenberg intended it: Webern's orchestration realized qualities in Bach's work that were latent, but fell short of full expression. Bach in Webern's hands became like Haydn in Brahms's hands: fully in touch at last with the implications of the tradition in which they participated.

V

Of course one did not use a word like Verbesserung about Bach. As his biographer reports, Schoenberg ‘was furious if one mentioned Handel in the same breath as Bach'.Footnote ⁴² In a letter to Berg he wrote of his Quartet Concerto that ‘it will be a very good piece and that won't be Handel's doing'.Footnote ⁴³ And in the preface to the published score he wrote of Handel's ‘insufficiency with respect to thematic invention and development [that] could satisfy no sincere contemporary of ours'. (What a contrast with every attitude implied by the term ‘early music', whereby it is assumed, however untenably, that modern taste should be set aside in favour of historical authenticity!) And he added in conclusion that he had achieved a ‘solidity of form and intensification of motivic development' that rendered Handel worthy of comparison with Mozart or Brahms.Footnote ⁴⁴ Webern could have said as much about his Bach orchestration, but didn't. Why? Because, unlike Handel, Bach had never been contaminated by loose Italian or theatrical ways – had never, so to speak, left his house – and had retained his purity, and his density.

Interestingly enough, one can find an altogether comparable assessment of Handel and Bach in the writings of Stravinsky. In Expositions and Developments, the third so-called conversation book with Craft, there is a sustained sacrifice of Handel on the altar of Bach, ending with the withering observation that ‘Handel's inventions are exterior; he can draw from inexhaustible reservoirs of allegros and largos, but he cannot pursue a musical idea through an intensifying degree of development'.Footnote ⁴⁵ Coming from one who in the very same book of memoirs said of the ‘Pergolesi' he was given to arrange for Pulcinella that ‘I looked, and I fell in love',Footnote ⁴⁶ the derogation of Handel must be another symptom of der alte Modernsky's need to forge a retrospective solidarity with the Neue Wiener Schule – a need that also found expression in Stravinsky's orchestration of the Choral-Variationen über das Weihnachtslied ‘Vom Himmel hoch da komm' ich her', quite clearly an emulation, in its intensification of the already very dense counterpoint of the original, of Schoenberg's and Webern's Bach arrangements.

The issue of Klangmaterialismus – an issue that is crucial, as I am claiming, for the distinction between ‘early music' and ‘alte Musik' – is curiously relevant to Schoenberg's arranging process. It provided him with a critical tool. Schoenberg's Bach transcriptions – the chorale preludes ‘Komm, Gott, Schöpfer, heiliger Geist' and ‘Schmücke dich, O liebe Seele', and the St Anne Prelude and Fugue, all done between 1922 and 1928 – use a slightly updated Wagner orchestra (updated by the addition of two pitched percussion instruments – celesta and xylophone – that were unavailable in Wagner's day) and employ quite conventional techniques of sumptuous scoring. The idea, as Schoenberg put it in a letter to the conductor Fritz Stiedry, was not ‘any “full flavor” that comes of varied color; rather colors help to clarify the movements of the parts, and, in a contrapuntal texture, that is very important!'. Schoenberg, who rarely let an opportunity go by to take a shot at performers (excepting those in his own coterie), added that he felt it was his duty towards Bach to transcend the limitations of his instrument. ‘Whether the Bach organ was capable of [clarifying the movement of the parts] we do not know. Present-day organists are not; that I do know (and it is one of my starting points).'Footnote ⁴⁷

And yet Schoenberg's transcriptions employ massive octave doublings – at times as much as quadruple doublings – such as might have seemed called for as a way of evoking the serried ranks of pipes that might be coupled when performing Bach's originals on a late-model nineteenth-century organ. These transcriptions have therefore seemed to me to be a protest or a polemic against the exactly contemporaneous Orgelbewegung or Orgel-Erneuerungsbewegung that sought to defamiliarize the music of Bach – or at least to alienate it from what was viewed as ‘Romantic' performance practice – by seeking to emulate its original timbres by restoring or copying the instruments on which it was originally played. Such an effort could only diminish Bach in the eyes of his self-proclaimed heirs. In the words of Adorno, it turned Bach into ‘a composer for organ festivals in well-preserved Baroque towns', rather than the fountainhead of contemporary music.Footnote ⁴⁸ The Orgelbewegung, in short, was one of the earliest harbingers of the twentieth-century ‘early music' boom, and one of the most egregious early manifestations of musealer Klangmaterialismus. Schoenberg's opposition to it is altogether understandable in this light, since it sought precisely to dismantle the sense of direct inheritance of Bach in which Schoenberg and his pupils so fervently believed. So, big and lush as they are, Schoenberg's Bach transcriptions are not ‘Klangmaterialistic'. The Bach they presented to their beholders was a familiar Bach. The defamiliarizers of the Orgelbewegung were the klangmaterialists, since they were relying on a novel sound to accomplish a revisionary task.

But now compare Schoenberg's free transcriptions of Handel and Monn. They are riots of colour, and riots of laughter, and are so obviously send-ups that only the forbidding name of the composer (and also, possibly, the ferocious difficulty of the solo parts) has prevented their becoming pop-concert favourites. Their gaudy scoring really amounts to graffiti – something that can be easily illustrated, I think, by looking at the orchestral percussion parts, and particularly at the use of the xylophone. The transcription of the St Anne Prelude and Fugue employs a huge orchestra: thirty-one wind and brass parts that must be balanced by a massive complement of strings, but only six percussion instruments are called for besides timpani. In the String Quartet Concerto after Handel the orchestra is much smaller, with only fourteen wind and brass parts, implying a correspondingly smaller complement of strings. But there are eight percussion parts besides the timpani, including piano instead of celesta. In the Cello Concerto after Monn the wind and brass contingent is smaller yet – only thirteen parts to balance against the strings. But the percussion (besides timpani) has grown to ten instruments, including both piano and celesta. One can say that the weight of the percussion writing varies inversely with the respect Schoenberg had for the original composer.

As for the xylophone, it is the most garish colour in Schoenberg's palette, one traditionally associated with cartoon caricature (as in Saint-Saëns's Danse macabre or his ‘Fossils', along with other skeletons I could name) or expressionistic horror (recall the third act of Wozzeck, or Schoenberg's own Begleitungsmusik zu einer Lichtspielszene). In the St Anne fugue the xylophone gets to double the huge unison run at the very end, and is otherwise confined to a grand total of four notes to reinforce sforzandos. In the Handel Concerto it has a few notable solos, predictably including the answer to the notorious fugue subject – itself a joke – that consists of nothing but diminutions of a repeated note. In the Monn Concerto, especially the outer movements, it never shuts up, and gets many solos.

I am particularly conscious of them because of the peculiar circumstances under which I came to know the piece. I have heard most of the existing recordings – two under Robert Craft, an amazing performance by Yo-Yo Ma, and a more recent one by Mark Drobinsky, recorded in Yekaterinburg, Russia. But this is a piece one really has to experience live, which I have had the good fortune to do only once. That was in Russia too, during my student year there in 1971–2. Schoenberg's Concerto was given its Soviet première that season at the Small Hall of the Moscow Conservatory, with a wonderful cellist named Natalia Gutman – then a young but already famous ‘laureate of international competitions', as the standard poster phrase went, now known as an outstanding exponent of the late-Soviet modernists Shnitke, Denisov, and Gubaidulina. I knew Natalia Gutman through mutual friends I had made at the Conservatory, and went to the concert as the guest of the percussionist Mark Pekarsky, who was playing the xylophone in the orchestra. I don't think I stopped laughing from one end of the piece to the other, not only as a result of the antics Schoenberg had composed into the cello part – if you've seen it you know that there is no other piece in which the cellist's left arm has to do so much leaping up and down – but also, and mainly, from the non-stop activity in the kitchen (to use the term Russian percussionists used). I was not the only one laughing. The whole audience was responding to the piece as if they were at a comedy club. And it was not the unfriendly laughter of the shocked bourgeoisie; it was the especially hip or knowing laughter of conservatory students who knew their music from the inside. After the concert I went backstage to offer congratulations, and told Mark how funny it had been. ‘Yeshcho by!' he shouted in reply (the Russian for what in English would now be ‘Duh!'). He told me that it was he who had talked Natasha into performing the piece, precisely because he knew he would be the centre of attention.

Was this the reaction Schoenberg would have wanted? Yeshcho by! Schoenberg's glorious sense – dare I say, his Jewish sense – of humour must be the best-kept secret in all of Western music. The fact that it usually goes completely unrecognized speaks many volumes about the culture of modernism – or what it became in the later twentieth century, when all of us were growing up. I am always amazed to witness the long, serious faces all around me at performances of Pierrot lunaire, and I would like to close with an example from Pierrot, because it relates equally to Schoenberg's humour and to his attitude towards ‘alte Musik’, and ties together all the themes I have been broaching.

VI

When considering the relationship between the composers of the Neue Wiener Schule and ‘alte Musik', scholars often speak, at times in awe, of the prevalence of canons and other contrapuntal tours de force in their compositions. One of the densest of these constructions is ‘Der Mondfleck', the eighteenth melodrama in the Pierrot cycle. It is the one in which Schoenberg piles a strict canon for the violin and cello on top of a freer canon for the clarinet and piccolo plus a fugue for the piano (and piles the Sprechstimme part on top of all of that), and in which the two instrumental canons reverse direction halfway through and reproduce themselves in retrograde. I do love to quote all the solemn, awestruck commentary this piece has elicited, my favourite perhaps being Charles Rosen's exclamation that ‘Der Mondfleck' is ‘one of the most elaborate canons worked out since the end of the fifteenth century'.Footnote ⁴⁹

But how elaborately ‘worked out’ is a canon or a fugue that is written in an idiom that recognizes no distinction between consonance and dissonance, so that, harmonically speaking, literally anything goes? The essence of counterpoint has always been its ‘dissonance treatment'. That, and that alone, is where skill is required and displayed. What makes Bach's Musikalisches Opfer and Die Kunst der Fuge such astonishing tours de force is not just the complexity of the texture but the fact that that complexity is achieved within such exacting harmonic constraints. Take away the constraints, and you have rendered the tour de force entirely pointless. If he had had enough instruments at his disposal, Schoenberg could have piled six more canons on top of the texture as written, and it would have been no harder to calculate.

I have already made this point in print, both in the Oxford History of Western Music and in an article published in 2004.Footnote ⁵⁰ And of course I have been roundly rebuked for my insolence. The pianist Ian Pace, who thinks himself a very loyal Schoenbergian, protested at my ‘crude formulation':

These strictures are familiar enough. Adornian high dudgeon is as conventional a playing card as any other. But if my observations on atonal canonic or fugal writing give evidence of a middlebrow attempt to marginalize all atonal music, then Arnold Schoenberg must share the dock with me. In the shorter of two essays titled ‘Composition with Twelve Tones' (1947) he wrote that ‘even the writing of whole fugues is a little too easy under these circumstances' (that is, circumstances of emancipated dissonance) and recognized ‘no merit in writing canons of two or more voices'. Instead, he maintained, ‘composing of these forms should only be undertaken for some special reason’.Footnote ⁵²

That certainly applies to ‘Der Mondfleck', where the special reasons, as in most vocal music, are to be sought in the text. Kathryn Puffett has analysed what Schoenberg called ‘das Verhältnis zum Text'Footnote ⁵³ in that piece and determined that the retrograde occurs ‘at the exact moment Pierrot notices a white fleck of moonlight on the back of his coat'. She interprets the combination of wind and string parts moving backwards while the piano and voice parts continue to move forwards as representing Pierrot ‘walking forwards while looking back over his shoulder' as he tries in vain to rub out what he thinks is a spot of plaster.Footnote ⁵⁴ So far so good, and so far so droll; but why all the contrapuntal complication? I think it is part of the joke: an even better part, in fact. The text is all about Pierrot's efforts to rub out the spot of moonlight – all frenzied but pointless activity. That is a perfect description of an elaborate contrapuntal texture with emancipated dissonance. Or to put it the other way around, an elaborate contrapuntal texture with emancipated dissonance is a perfect metaphor for the urgent but ineffectual efforts that Pierrot is making. From a bogus masterpiece of counterpoint, ‘Der Mondfleck’ suddenly becomes a genuine masterpiece of self-mocking irony.

That self-mocking irony is one of Schoenberg's most attractive traits, and it is rarely given sufficient acknowledgment in our pompous and ponderous literature. In an especially unbuttoned mood Schoenberg is reputed to have told a pupil one day that ‘Now that I have emancipated dissonance, anyone can be a composer'. Do I believe that to be literally true? No more than I believe literally that ‘anything goes', harmonically, in atonal or twelve-note music. A composer as craftsmanly as Schoenberg creates ad hoc harmonic norms, and they are easily discoverable in ‘Der Mondfleck', the augmented triads at the beginning of the piano fugue being especially telling in this regard. Ian Pace is therefore quite right to assert that even without a traditional criterion of consonance in ‘Der Mondfleck’, ‘there is still much differentiation between different levels of dissonance (for example in terms of predominance of harmonies made up from whole- or semi-tones at different moments)’. Analysis is still fruitful, and can uncover the consistent harmonic relations that produce the syntactical coherence and the well-sounding surface that listeners intuitively sense and admire in Pierrot. In twelve-note music, of course, there are other, perhaps more objective criteria to judge the correctness of the writing.

But there are still problems of perception that cannot be willed away by critical fiat, and they bedevil analysis as well as naïve listening. It is time to acknowledge them. I once hazarded the claim that any analytical method worthy of the name should offer a reliable tool for textual criticism.Footnote ⁵⁵ I made that claim in a polemical context, with reference to an ongoing exchange with Allen Forte over what I saw as the sterility of his ‘set-theoretical’ analytical method and, in particular, its pitfalls as applied to Stravinsky's Rite of Spring.Footnote ⁵⁶ I still think my claim is valid, and by now I can cite examples. Knowledge of Stravinsky's octatonic usages, the reality of which were questioned long ago by Forte and quite recently in a bewilderingly impertinent article by Dmitri Tymoczko,Footnote ⁵⁷ enabled me to conjecture a textual correction at the end of the first of Stravinsky's Pribaoutki of 1914, which subsequent inspection of the sketches confirmed.Footnote ⁵⁸ But there are notorious places in the music of Schoenberg where misprints have remained uncorrected for many decades, because no one is in possession of a reliable analytical criterion. The most famous one is towards the end of the first piano piece in Opus 11, where analysts relying on what they have regarded as internal evidence (but which was really the pseudo-internal evidence of their own theoretical prejudices) have invariably proposed the wrong correction.Footnote ⁵⁹

There are similar longstanding misprints in ‘Der Mondfleck’ too. Some of them can be putatively corrected on the basis of the canonic writing. Kathryn Puffett reports, and a mere glance at the score will confirm (after ninety-five years and countless printings!), that ‘there are several discrepancies in the canons, […] in both pitch and rhythm, mostly in the retrograde half of the piece'. She further reports that ‘in 1977, when the Schoenberg Archive was still in Los Angeles, Leonard Stein wrote to me describing Schoenberg's conducting score, on which most of the discrepant notes had been corrected in pencil'.Footnote ⁶⁰ I think it worth emphasizing that, according to what Stein reportedly told Puffett, Schoenberg corrected most of the discrepant notes. Why not all? Is that evidence of intention or of inattention? If we cannot tell (or if, as seems at least possible, Schoenberg could not tell), then how important, after all, are the details of pitch content in this music?

Charles Rosen raised the point years ago in his little Schoenberg monograph, when he wrote that the ‘new eminence of color, texture, and dynamics entails, as a consequence, something that most respectable musicians have been reluctant to admit because it is conceived as something disreputable: the downgrading of the importance of pitch’.Footnote ⁶¹ We are still reluctant, and music theorists, being perhaps the most respectable of all musicians, have been most reluctant of all; indeed, we still tend in our theoretical literature to fetishize pitch relations, and this has led, in the case of the music of the Neue Wiener Schule, to many errors, unnecessary denials, and false claims.

So here is where we have the most to learn, perhaps, from studying the Neue Wieners in conjunction with ‘alte Musik’. Rosen made the very sensible suggestion that Schoenberg relied as much as he did on canons, despite his own recognition that they were no longer the compositional achievement that they once had been, because during his pre-serial atonal period they provided a guide to writing in the absence of conventional norms, similar to the guide that the series would eventually provide in twelve-note music.Footnote ⁶² (And this, by the way, recapitulates the phylogeny of pervading imitation in Franco-Flemish polyphony in the fifteenth century; it mainly occurred at first in the tenor tacet sections of masses and motets, where it provided a scaffolding that elsewhere the cantus firmus had supplied.) Canonic writing also provided a criterion of correctness that reassured the composer, however arcane or even specious the criteria of harmonic correctness might be under conditions of emancipated dissonance.

Therefore I think we might extend Rosen's point as a way of understanding why Webern continued to rely heavily on canonic writing, and even touted it as a compositional achievement, when writing twelve-note music, where there was once again a built-in criterion of correctness on which both analysts and performers could rely (though not infallibly) in critiquing the text.Footnote ⁶³ Webern's boast, at the end of Der Weg zur neuen Musik, that the second movement of his Symphony, Op. 21, has a unity that ‘even the Netherlanders' could not have managed, is often regarded as naïve, at least when compared with Schoenberg's wily urbanity.Footnote ⁶⁴ In twelve-note music, canons do not even have whatever merit they might possess in ‘Der Mondfleck'. And Webern could never be accused of self-mocking irony. But if you count moral probity as a musical merit, then Webern's insistence on multiple or even redundant criteria of correctness makes not only aesthetic but also ethical sense, adding yet another dimension to his musical utopia – a utopia defined, and validated, by ‘alte Musik’, if not by ‘early music’.