In response to:

The Fallen Angel? from the April 8, 1999 issue

To the Editors:

Marilyn McCully’s review of Rosalind Krauss’s The Picasso Papers [NYR, April 8], entirely negative except for a token appreciation of Krauss’s already well-accepted semiotic analysis of collage cubism, does not meet the minimum requirement for intellectual debate, namely that the opponent’s argument be understood.

At the outset of the review, McCully sketches Krauss’s position by citing a long passage from the book which in fact is Krauss’s description of a commonly held, indeed utterly banal opinion: that Picasso, in his turn from collage to neoclassicism, abandoned modernism. (This is a view Krauss regards as exactly half right, as we will see in a moment.) McCully omits the part of the passage which would have made clear Krauss’s resistance to the simple opposition of true and false Picasso, modernist and betrayer: “For Uhde, as for many art historians after him,…it seemed perfectly apparent that the [World] war [I] years had simply split his [Picasso’s] work in two.”

No matter. McCully proceeds to make the counterargument, dear to art historians who are not inclined toward theory, that in fact there was no break but (to borrow the title of a recent Stuart Davis monograph) an “amazing continuity.” McCully goes on to propose various missing links.

One missing link is photography, which McCully claims was a resource for Picasso throughout the period and not, as Krauss maintains, a dire threat—but this is a debate I will not enter. Another missing link concerns Picasso’s work from the war years. McCully declares that “the main importance” of Picasso’s Harlequin (1915) is that “it anticipates the scale and spatial considerations of some of Picasso’s ‘neoclassical’ paintings.” This is the kind of vague assertion that gives formalism a bad name, and it stretches continuity to the breaking point. Krauss takes the very same painting as an occasion to refute Cocteau’s claim that he triggered Picasso’s neoclassical phase in all its Commedia dell’Arte panoply by procuring him a harlequin suit (pp. 89-95). And McCully, citing Richardson’s Picasso biography, approves of Krauss’s refutation of Cocteau’s claim. Which makes it ironic that she goes on to make Cocteau’s fundamental argument—that the picture did in fact lead to neoclassicism.

The real irony of the review, however, is that McCully’s effort is not so different from Krauss’s, at least at first glance. Krauss too wants to discover the continuity, or rather the hanging-together, of Picasso’s neoclassical and collage-cubist work. That McCully does not see Krauss as a fellow gap-bridger may be because the continuity Krauss proposes is not a formal or stylistic one, not a matter of such things as “scale and spatial considerations.” Rather, it is a matter of something deeper, of a dialectical relationship between opposites, or two halves of a whole, or two sides of a coin, to use the metaphor that Krauss brilliantly develops, drawing on Gide’s Counterfeiters.

Modernism, Krauss argues, is as much pastiche as collage, as much neoclassicism as cubism, because, as she concludes, “the newly liberated circulation of the token-sign always carries as its potential reverse an utterly devalued and empty currency” (p. 241). In other words, Picasso’s pastiche of classical styles, following hard on the flowering of collage, revealed the free-floating, nonreferential signifier of collage as no longer autonomous but meaningless instead. This is the bad dream or (Krauss’s last words) the “guilty conscience” of modernism. So Krauss does not distinguish modernism from its betrayal, as McCully would have us believe, but rather insists that modernism is a double structure, always potentially self-betraying, especially when it is at its best.

Of course, this kind of dialectical argument does require a clear break or obvious opposition between Picasso’s collage and his neoclassicism. So McCully’s obsession with finding stylistic continuity between the two, if it were fruitful, might actually ruin Krauss’s argument. As it is, however, this obsession simply leaves McCully manifestly uninterested in how Krauss negotiates a break that McCully does not recognize. Thus the word “dialectic,” which is at the heart of Krauss’s argument, does not cross McCully’s lips. And yet it seems McCully realizes the centrality of dialectic to Krauss’s argument, for she makes sure to dismiss the two dialectical models at the heart of the argument, Freud’s reaction-formation and Adorno’s pastiche, by referring in each case to a single dissenting work.

This kind of dismissal reveals the abyss between McCully and Krauss, indeed between two kinds of art historians working today—those who believe that theoretical perspectives drawn from other disciplines, from linguistics to economics to psychoanalysis, have their uses, and those who believe visual material is best approached directly, in the absence of any mediation. For Krauss, what counts is not the empirical soundness or current popularity of any given theory but rather the quality of interpretive work that it allows. For McCully, since theory is not necessary, only theories of unquestioned validity (name one!) should be applied.


Not only is The Picasso Papers the first serious attempt to deal with Picasso’s neo-classical work as something other than simple failure or welcome relief (depending on one’s attitude to “modernism” typically understood). For those interested in Krauss’s work, this book represents a simultaneous extension and self-criticism of her earlier semiotic work on collage cubism, a salutary darkening of its triumphal overtones, an admission that the free play of the signifier has its costs. What Krauss seems not to have decided, for all her citations of Adorno, is whether those costs are indeed dialectically necessary, or “‘truly’ endemic,” to modernism. This is an opinion she assigns (in a footnote) to Stanley Cavell without making clear whether she agrees. This ambivalence, this suspension, is a key moment in Krauss’s text, lending it a complexity that might well confound a less than sympathetic or dedicated reader. And yet such suspension is characteristic of the productive tension in Krauss’s writing, and is its virtue.

Harry A. Cooper

Associate Curator of Modern Art

Harvard University Art Museums

Cambridge, Massachusetts

Marilyn McCully: replies:

Harry A. Cooper takes issue with my review of Rosalind Krauss’s The Picasso Papers on two principal points: first, that I have failed to understand Krauss’s dialectical method of argument, and, second, that I represent a school of art history that refuses to allow the application of theoretical models to the discipline. Cooper suggests that while I realize “the centrality of dialectic to Krauss’s argument,” I do not appreciate her point about the “dialectical relationship between opposites” that, supposedly, connects Picasso’s cubism-collage and his neoclassical pastiche.

The fact that opposites are related is, of course, self-evident—it was evoked centuries ago in the myth of the fallen angel—without recourse to either Marxist dialectic or Freudian reaction-formation. However, while Cooper—though not Krauss—describes the view that Picasso abandoned modernism “in his turn from collage to neoclassicism” as banal, Krauss’s entire argument rests on the crude opposition of true modernism and pastiche (modernism’s “guilty conscience”). She makes this view clear when she cites Adorno’s dialectical opposition of Schoenberg and Stravinsky and associates Picasso’s true coin with the former and his false coin (or the obverse of the true coin—the metaphor gets a little mixed) with the latter.

Krauss relies for much of her argument on Adorno (as well as, pace Cooper, on Uhde); she presents the views of both Adorno and Uhde entirely uncritically. Cooper cites another instance where, he says, she assigns an opinion to an authority “without making clear whether she agrees.” He describes this as a key moment in her text, “lending it a complexity that might well confound a less than sympathetic or dedicated reader.” It seems bizarre that Cooper identifies as a “virtue” what he sees as the ambivalent procedure of citing others—ostensibly with approval—while leaving open the option of dissociating oneself from the opinion if one is criticized.

There is, however, no ambivalence in Krauss’s search for dialectical opposition in Picasso’s work from around 1910 through the 1920s. Her selective reading of the evidence and misleading descriptions of non-cubist works emphasize this opposition; indeed, the vigor of her interpretation makes Cooper’s struggle to distance her from the concept of a “rupture” in Picasso’s work rather pointless. To suggest, in these circumstances, that Krauss wants, as I do, “to discover the continuity, or rather the hanging-together, of Picasso’s neoclassical and collage-cubist work” strains the notion of continuity to the breaking point. A more careful examination of the technique, context, and chronology of Picasso’s entire body of work during the period under review suggests, as I wrote, not an artificially constructed dialectical opposition but a continuing investigation of representation that took a number of different approaches. These “different systems of representation,” as I wrote, can be accurately understood as compatible with one another—by contrast with the view that one of them should be seen as the expression of guilty secrets emanating from the dark side of Picasso’s personality, the consequence, as Krauss put it, of a “structure of duplicity.”

If these two ways of examining an artist’s work represent two different approaches to art history, this does not mean a history based on full examination of evidence excludes theory. Theories of many different kinds are essential for art historians and can be among their most valuable tools. There is no need in art-historical practice for there to be any opposition—whether dialectical or of any other kind—between interpretation based upon the work of art itself and the application of theoretical models. Without theory, art history can become sterile descriptive formalism, or simply archival documentalism.


However, before any theory can be brought into play a careful examination both of the works themselves and of a wide range of documentary sources must be undertaken. And if the results of this examination undermine a theoretical hypothesis, the proposed theory must not be allowed to override the results of scholarly investigation. If there is a polarization in art history today, it is because a theoretical approach has too often ignored the evidence. (In the case of Professor Krauss’s book, to take only one example, I showed that she ignored important evidence about Picasso’s use of photographs.) It is vital for the future of the discipline that the art historian should pay as much attention to historical accuracy and critical analysis of sources as to adopting theories from related disciplines and forcing art history into their mold. Most troubling of all for art history would be Cooper’s notion, ascribed by him to Krauss, that “what counts is not the empirical soundness…of any given theory but rather the quality of interpretive work that it allows.” To draw on theories without critically considering their vulnerability to factual and logical analysis seems an abdication of judgment that any serious discipline must try to avoid.

This Issue

October 7, 1999