To Criticize the Critic

To the Editors:

In concluding his review of my Concise History of Avant-Garde Music (NYR, September 28) Robert Craft remarks on “the urgent necessity of establishing standards in music criticism.” Even more urgent, his piece suggests, is the need for some principles in the criticism of criticism.

Your readers will not need me to point out Mr. Craft’s misreadings of certain passages from my book. I would be tempted to ascribe such misconstructions to illiteracy did they not come from so well-used a pen as Mr. Craft’s; as it is, I must put them down to wilfulness.

More serious, because less obvious to the reader of the review, are those many occasions where Mr. Craft uses the old trick of selective quotation to make nonsense of what I wrote, sometimes adding nonsense of his own for good measure. I cannot expect you to allow me the space to set the record straight on every point, so I will confine myself to the matter of serialism, since on this Mr. Craft chooses to review not only my book but also my competence to write an article which he has not read.

He quotes me as saying that: “The series is thus a sort of hidden theme.” He then picks holes in this on the grounds that “in Schoenberg’s music, a series is not ‘a sort of hidden theme,’ but an overt one.” This is not quite true: the series, as an ordered collection of pitch classes, cannot be a theme, though it may be presented as one. If we complete my original sentence we find the whole question clarified: “The series is thus a sort of hidden theme: it need not be presented as a theme (though, of course, it may be), but it is a fount of ideas and a basic reference” (p. 87).

And I have not yet exhausted the errors in one short paragraph of Mr. Craft’s review. He challenges my statement that the series “can be used to generate melodies and harmonies” because, he explains, “all melodic and harmonic constructions must derive from the series.” Well, if Mr. Craft can demonstrate such derivation throughout, let us say, Schoenberg’s A Survivor from Warsaw (not The Survivor of Warsaw as he would have it), or Berg’s Violin Concerto, or Boulez’s Second Piano Sonata, then I and the whole world of musicology will be greatly indebted to him. And if these are not serial works, what is?

Let me turn briefly to the three examples Mr. Craft chooses from among my “great many simple errors of fact.” Perséphone, he says, is not a ballet, and yet Stravinsky in his own notes on producing the work recommends that “the staging could be worked out entirely in choreographic terms” (I quote from Dialogues and a Diary by Igor Stravinsky and one Robert Craft). Secondly, there are of course eight instruments required for Pierrot lunaire but only five players; Mr. Craft is ingenuous in his complaint. I must, however, thank him for correcting me in the matter of Hindemith’s death, which happened in 1963 not 1964: an important point.

I have not chosen here to concern myself with questions of critical judgment on which Mr. Craft takes a different view. Those who read the whole book rather than Mr. Craft’s edited excerpts will find my comments justified. And in any event, since the book is, according to Mr. Craft, “never redeemed by novelties of perspective or musical percipience” it follows that he is arguing with views which are already familiar.

Paul Griffiths

Oxford, England

Robert Craft replies:

Mr. Griffiths’s “old trick” is that of trying to patch up one casualty among several dozen and of excusing his failure to deal with the others by assuming a “lack of space.” But the “nonsense” is his own, not the result of distorting quotation, and his inability to explain away the many muddles to which I called attention is evident in his straw-clutching correction of the preposition and article in the Schoenberg title.

Mr. Griffiths’s letter further demonstrates that what he actually says is often quite different from what he believes himself to be saying. That Perséphone might be staged “in choreographic terms” does not make this melodrama a ballet, any more than Balanchine’s ballet on the same composer’s Violin Concerto changes the classification and form of that work. Nor can the instrumental ensemble of Pierrot Lunaire be described as “a small band of flute, clarinet, two strings, and piano.” The Concise History does not mention “five players,” and some readers—the book is introductory—will not be aware of the third string instrument and of the other two winds. Reviewers are obliged to expose these and similar faults—as well as to help prevent the perpetuation of wrong dates, such as that of Hindemith’s death, which will be repeated from the Concise History, just as Mr. Griffiths has himself repeated other mistaken dates (pages 48 and 64) from secondary sources.

Mr. Griffiths’s new account of certain aspects of serial composition only shows that he would have been wiser to leave bad enough alone. Instead of “clarification,” his complete sentence provides a pleonasm—that a thing “need not be…(though, of course, it may be)”—and a misattribution (a series is not a “fount of ideas” until a composer with ideas makes it one). More important, a series is not a “basic reference,” since this implies that an extra-serial element is the origin of the referral, whereas every note in, for example, Schoenberg’s Quintet is serially determined. Mr. Griffiths’s letter disingenuously switches the discussion from the comparatively simple procedures introduced by Schoenberg in the early 1920s, however, to a period so much later in development that, in relation to the Boulez Sonata, at least, the expressions “themes” and “melodies” are anachronistic. Nevertheless, the assertion that the series cannot be a theme is false, and does not supersede the statement in Schoenberg’s 1941 lecture on the subject: “…a part of the theme, or the entire theme, consists simply of a rhythmization and phrasing of a basic set.” Nor does Mr. Griffiths’s updating jargon, “an ordered collection of pitch classes,” change this.

Since Mr. Griffiths feels himself to have been shortchanged by my specimen list of his “simple factual errors,” may I add to it, this time from the area of his serialist expertise? He says that

the first three bars [of a Babbitt quartet] are occupied [?] entirely with the minor third and related intervals.

But the only harmonic intervals in the first two bars are octaves. (What are the “related intervals”?) And, of the opening of Webern’s Symphony, the historian writes that

in the four-part canonic structure[,] the second and fourth lines mirror the first and third respectively, in instrumentation as well as in melodic shape.

But the instrumentation is not “mirrored.” It follows, in both canons, the changes of timbre and modes of articulation of, respectively, the original-order and reverse-order first voices. “And if these are not serial works, what ARE?”

More pernicious than errors of this kind are the vapid comparisons (“[Webern’s] Concerto…is a twentieth-century Brandenburg…”), the fostering of an inane concept of the creative musical mind (“other works of Debussy…are probably based directly on nature”), and the obliviousness even to the necessity of logical argument. Here is the first sentence of the book:

If modern music may be said to have a definite beginning, then it started [!] with this flute melody, the opening of the Prélude à “L’après-midi d’un faune” by Claude Debussy (1862-1918).

If A, then B. But the step is far from self-evident, and the claims of other contenders—Parsifal, for one—cannot be ignored.