Critique of Pure Readymade

More than any other art historian, critic, or philosopher, Thierry de Duve has most vigorously assumed the task of completing Marcel Duchamp’s work for him. De Duve’s scholarship aims to make sense of how the artist’s intentionally delayed messages—such as the notion that it is the spectator who is responsible for completing a work of art—have been and ought now to be received. Since the publication of his major monograph, Kant after Duchamp, this art historian-philosopher has emphasized that following Duchamp’s invention of the readymade, the Kantian judgment of taste, “This is beautiful,” should be replaced with the no less transcendental statement, “This is art.” His latest book, Aesthetics at Large: Art, Ethics, Politics, refines his aesthetic critique toward new ethical and (explicitly center-left) political ends. True to form, Aesthetics at Large is a sprawling intellectual work. It assembles over two decades of the author’s previously published or presented essays. These range in scope from close readings of Immanuel Kant’s Critique of Judgment, to a quasi-manifesto on the purpose of public art museums, to more familiar analyses of Duchamp’s reception by artists in the 1960s and 70s. Consequently, Aesthetics at Large resists simple summaries and solicits instead (as David Carrier has admiringly suggested) quarrels.1 For the majority of de Duve’s art historian readers—readers perhaps less confident evaluating competing interpretations of continental and analytic philosophy—I suspect that his latest book will generate more questions than quarrels. This too is meant admiringly.

As such an art historian reader, I want to raise two questions regarding de Duve’s Kant after Duchamp approach. The first concerns his art historical and art theoretical account of the readymade, and how this interpretation of the readymade is said to impact one’s reading of Kant. The second question concerns de Duve’s philosophical argument that the word art was—or is—a proper name. Both postures were first developed in Kant after Duchamp and have since become central postulates in Aesthetics at Large. Given that Art, Ethics, Politics is only the first volume of Aesthetics at Large, an anticipated three-part series, I hope that both lines of inquiry will permit readers to take stock of some of de Duve’s most original claims as well as allow for art historians to remain engaged with his ongoing project as he settles more firmly into the realm of philosophical aesthetics.

At the heart of de Duve’s Kant after Duchamp approach is his refusal to take the term art for granted. It is this insistence that contemporary art historians must provide a coherent account of their work’s objects that situates de Duve’s scholarship within a self-critical tradition of art historical theory extending back, at least, to Erwin Panofsky. Specifically, de Duve has endeavored to figure out what we really mean when we declare, “This is art,” now that this statement conjures not only poems, paintings, and performances, but also urinals (Duchamp’s infamous readymade, Fountain). For de Duve, at the time of the readymade’s invention, a work like Fountain could solicit only one of two judgments: that it ought to count as a work of art, or that it ought to count as nothing much at all because an upturned urinal did not fit within any preexisting category of the fine arts. In its circumvention of these traditional discipline-specific criteria for determining an object’s belonging to any specific category of art, Fountain made perspicuous the conditions for judgment that underpin our nomination of any object as art at all, or in de Duve’s terms, as a work of art in general. Furthermore, given that Duchamp did not create but merely selected the urinal he submitted for display, de Duve argues, “the readymade virtually erases the distinction between making art and estimating its quality.”2 Which is to say that the judgment of art (as opposed to the judgment of a work of art’s quality in relation to criteria stipulated by the work’s kind) arises as the critical question posed by Duchamp, and subsequently, by a generation of ambitious post-Duchampian artists. “The readymade is nothing but the statement ‘this is art’ as it is affixed to any ‘this’ whatsoever.”3 In other words, when we judge a readymade to be art, we end up pronouncing a judgment that assents to the artist’s original ascription by repeating it. We thereby ratify the work’s performative contents (“This is art”) and do so with the expectation that others ought also to view the work as such.

Duchamp thus represents the paradigmatic figure in both de Duve’s historiography of modern art and the aesthetic critique that de Duve draws therefrom. From an art historical perspective, Duchamp’s readymades marked a shift from an academic artworld rooted in discrete artistic disciplines toward the art-in-general artworld that still exists today, wherein we supposedly no longer necessarily speak of the arts in the plural and can instead refer to art or nonart in the singular. From the perspective of philosophical aesthetics, the readymade occasioned de Duve’s updating of Kant’s Critique of Judgment. In de Duve’s words, Kant after Duchamp asserts “that the sentence ‘This is art,’ with which readymades are baptized and countless other works made in the wake of Marcel Duchamp’s readymades are appreciated, is an aesthetic judgment in the Kantian sense” (AL, 15). In more concrete terms, de Duve rereads Kant’s “Analytic of the Beautiful” by replacing the word beautiful with the word art, such that judgments of art are also said to be rooted in personal feeling and pronounced in the same universal voice. For Kant, when we judge an object to be beautiful, we are describing a pleasure taken in an object without any requirement that we subsume it under a concept in our understanding. What he means by this is that we can, for instance, judge a sunset to be beautiful even without recognizing that the spectacle is indeed a sunset, and without any expectation that our best available formal descriptions of this particular sunset should somehow pick out essential features of beauty from which we could develop rules to guide our future judgments of taste.

For de Duve, the same is also true of the judgment, “This is art.” Just as beauty picks out particular objects in the presence of which we experience a feeling of the beautiful, so too does art pick out particular objects in the presence of which we experience a feeling of art. In this way, de Duve’s theory is particularly well suited for considering how we can judge new forms of art to be such, even when they have exceeded our earlier expectations of what had counted as art. His answer: “When an object—any object whatsoever—is plucked from the great no-man’s-land of art-in-general and yields an aesthetic judgment … this object, in being dubbed ‘art,’ gets called by a name that, like proper names, only refers to things and doesn’t have any fixed or determinable meaning aside from the subjective aesthetic significance attached to the experience” (AL, 42). Now, grammarians may bristle at de Duve’s likening of art to proper names, but his analogy derives from his strong reading of the work of philosopher Saul Kripke, for whom proper names function as rigid designators, indicating that proper names are terms that “serve to fix the reference and not to pronounce a meaning” for the objects to which they refer (KD, 57). To use a favored example in philosophical debates over proper names, the name Aristotle identifies a great philosopher, but this name would still refer to this same identity, even if in some other possible world Aristotle were to have become a great poet, instead. This is to say that the name Aristotle refers distinctly to the definite descriptions that we may use in our actual world to help us identify the same historical figure. I will have more to say about de Duve’s recourse to Kripke later on, but for now, it is sufficient to recognize that de Duve’s likening of the term art to the Kripkean concept of proper names is theoretically necessary for his rereading of Kantian aesthetics, in which we substitute the word art for the word beauty.

To understand why, we might first consider the same object, such as Leonardo da Vinci’s Mona Lisa, once judged as beautiful and then again judged as art. In judging Leonardo’s painting to be beautiful, Kant believed that we spoke “as if beauty were a characteristic of the object and the judgment were logical (namely, a cognition of the object through concepts of it).”4 In judging the same work to be art, in contrast, we do not typically speak “as if [art] were a characteristic of the object.” It is not as if art is a characteristic of Leonardo’s achievement. It simply is his achievement. By contrast, even if Leonardo’s art were not so achieved, and thus not all that beautiful, it would still have been art, or at least conceivably art. This distinction between the contents of both judgments creates a tension in de Duve’s self-described “mechanical and stupid” reading strategy of substituting art for beauty. (As this example drawn from Kant’s Critique makes plain, de Duve’s Kant after Duchamp approach is in reality neither mechanical nor stupid). However, as de Duve recognizes, in all earlier and most competing theories of art, the judgment “This is art” logically entails the application of determinate concepts, such as the concept of the sister arts or of specific mediums. As such, a new conceptualization of art would be required for de Duve’s substitution to come off. Furthermore, Kant claimed that judgments of beauty refer exclusively to a specific class of feelings experienced before certain objects, or rather, a specific mode of liking these objects: disinterested liking. De Duve, in contrast, recognizes that our judgments of art can refer to a nearly indeterminate range of feelings. He marshals reactions of both delight and disgust as potential feelings felt before Duchamp’s Fountain, which means that judgments of art are not judgments of taste, or at least, not exactly. They do not specify some specific mode of liking or disliking, precisely because they name a particular mode of liking and disliking. All these differences between art and beauty notwithstanding, art’s function as a proper name preserves its logical availability as a substitute for beauty in de Duve’s critique because art’s usage as a rigid designator renders the term conceptually indeterminate all the way down. When we judge something to be beautiful, we are pointing to an object and stating that we count it as an instance of beauty: This here is beautiful. We need not refer to any prior definition of beauty, but reflect only upon our current subjective experience. For de Duve, the same is true for art. When we call an object art, we are asserting that others ought also to recognize the object as such, and we do so without giving any definite determinate meaning to the concept of art, precisely because art, if used as a proper name, only means “to fix the reference and not to pronounce a meaning”: This here is art. Hence Fountain’s exemplarity, because the historical significance of the work centers on this very question of whether we ought to point to a manufactured urinal and call it art.

To be sure, the all-or-nothing situation of the readymade in 1917 as de Duve has described it may not be theoretically necessary for accepting his indeterminate, purely reflexive usage of the word art. However, the status of the work as a treasure in the art historian’s personal art collection surely depends upon its unprecedented registration of the epochal shift that inspired de Duve to update Kantian aesthetics in a move from nature to culture (beauty to art), and from the specific plural to the generic singular (arts to art as such). It is for this reason that art historians might begin their engagement with de Duve’s Kant after Duchamp approach by assessing the persuasiveness of his interpretation of the readymade.

In that vein, my first question concerns de Duve’s account of Fountain, Bottle Rack, In Advance of a Broken Arm, and other so-called pure or unassisted readymades. Pure readymades are said to be unmodified manufactured objects that have merely been selected as art. In de Duve’s words, “[Unassisted] Readymades are works of art that condense all the artist’s decisions into one single choice” (AL, 154). For this condensation to be realized—for the status of the artist and the spectator to become logically equivalent—a pure readymade would need to be a ready-made object to which no other aesthetic decision ought to be claimed beyond its baptism (to use de Duve’s preferred expression) with the judgment, “This is art.” However, as the field of Duchamp studies has also exhaustively demonstrated, none of the artist’s so-called unassisted readymades actually fit this shorthand description.5 His ready-made objects were variously inscribed, punned, tilted, nailed down, strung up, or manufactured to specification. Were one to imagine a readymade without any of these additional features (the features that provide the artist’s readymades with their ongoing art historical interest), its actualization would still require at least two aesthetic decisions, and not just one: (1) the choice of this ready-made object as art (a shovel as opposed to a pitchfork), and (2) the choice to present the object as art in any given way (on the studio floor or gallery wall or on a pedestal). In Aesthetics at Large, de Duve refers to the second of the two aesthetic decisions in his description of Marcel Broodthaers’ reception of post-Duchampian art (56-57), and he seems to consider it as a necessary condition for “genuine art” in his provocative chapter on art and identity politics (72). However, his most succinct theoretical descriptions of the readymade reduce the post-Duchampian paradigm to a shift inaugurated by just one aesthetic choice: choice tout court. Indeed, de Duve’s most full-throated claims for Duchamp’s exemplarity depend explicitly upon this radically reductive formulation, because it is only the unassisted readymade that could virtually “erase the distinction between making and appreciating art” and thus close the gap between artist and viewer (153).

Let us consider, in this regard, two key passages from Kant after Duchamp. In the first, de Duve treats Fountain as the “exemplary example” of the judgment, “This is art,” because the work crystallizes the jurisprudential conditions that underpin anyone’s usage of the word art in the singular. “It is either art at large or nothing” (48-49). Fountain focalizes these conditions because, as mentioned earlier, it did not fit within any particular sub-category of art, by which de Duve means specific mediums, such that there would be no other available name to affix to the work beyond that of art as such. If I were to point to a painting and call it a work of art, and someone were to ask me why, I could always respond by saying that this work is a painting and that painting is a kind of art. For de Duve, Fountain interferes with this syllogistic reasoning—which is really just a deferral of the underlying skeptical challenge to any account of modern art’s identity—because a manufactured urinal is not any particular kind of (fine) art. Furthermore, Fountain “compels you to call it art, as if you were its author.”6 The act of designating Fountain as art resembles the work’s authorship because the crux of its creation is said to inhere to this very act of designation. It is with this second line of reasoning that we can begin to observe how the unassisted readymade is crucial for de Duve’s all-or-nothing account of art at large. Any additional assistance on the part of the artist would open up any number of substantive distinctions between his judgments and the spectator’s judgments, and these would provide the viewer with the opportunity to determine the status of the object independently of his or her aesthetic experience before it. To be more precise, if we take Duchamp’s designation of a urinal as art as Fountain’s only aesthetically significant feature (which I do not imagine de Duve to believe), then a given viewer’s judgment that it ought to count as art would indeed resemble its authorship. If, instead, we insist that the title, its clever signature, the urinal’s inversion, and the urinal’s submission (in good or bad faith) can all count toward our judging the urinal as a work of art and toward our subsequent interpretations (even if only to confirm the rote conventionality of art as an institution), then the spectator’s judgment that the urinal is a work of art only passingly resembles the author’s reflexive act of serving as the first judge of his own work. More importantly, this critical encounter with Fountain would no longer count as an all-or-nothing situation, because even if Fountain lacks the determinate concept of medium, any close observer could still uncover other conventional registers of artistic intentionality as conceptual resources for their judgment that the urinal was meant to be viewed as art.

A second passage from Kant after Duchamp helps to clarify this final point. Here, de Duve returns to the unassisted readymade as the paradigm for the readymade in the singular. Similar to his book’s introductory efforts to theorize a non-conceptual definition of art that would be capacious enough to accommodate the sheer variety of its objects (poems, paintings, performances, urinals, etc.), de Duve declines any effort to consolidate Duchamp’s disparate readymades under the concept, readymade.

We could designate them, one by one, but we could in no way collect them within a single category making up the concept of the readymade—in the singular—in that these fifty or so objects do not necessarily have characteristics in common at the level of medium, form, or style; in that all are not straight manufactured objects declared works of art; in that certain ones, called aided or assisted, imply a manipulation modifying the object, while still others have remained at the stage of an imaginary project … and finally, in that in the Green Box the word “readymade” often has a textual existence semi-independent of its referents, above all when connected to a qualifier (388).

De Duve offers no explanation as to why this list does indeed resist conceptualization, as opposed to disclosing any number of potential conceptual through lines, such as Duchamp’s apparent insistence on only treating objects, and not people, as readymades. Instead, de Duve pivots to argue that “one unassisted readymade—an ordinary object that nothing in its appearance distinguishes it from its nonartistic counterpart … is enough to observe … that the convention or condition tested was indeed that works of art are shown in order to be judged as such” (388-89). Already, the definitional portion of this claim should give us pause. We might first object that de Duve has modified his account of the unassisted readymade inaptly, because it would allow for indiscernible replicas of ordinary objects alongside “straight manufactured objects declared works of art.” A readymade is not a readymade because of the object’s appearance, but rather because of its identity. Fountain is not indistinguishably like a urinal; it is one. Second, and more importantly, even if we were to accept this particular definition, we would still have to insist that none of Duchamp’s so-called unassisted readymades ever fulfilled the definition. In this case, de Duve marshals “the bottle rack, the snow shovel … [and] … the famous urinal” as his examples, and thus we can restrict ourselves to these three (388).  Duchamp’s bottle rack was meant to be inscribed, and was also lost before the original could ever be designated as art; the shovel was strung up; and the urinal was upturned and signed. Thus, if anyone were to look closely enough, these so-called unassisted readymades do appear distinguishable from their nonartistic counterparts. And if all of Duchamp’s unassisted readymades are in fact minimally modified ordinary objects, then an additional conceptual through line opens up for the readymade. Duchamp’s readymades appear to test whether infrathin adjustments to how we ordinarily treat objects (including textual objects and preexistent art objects) ought to count as establishing aesthetically significant differences in how we ought to regard them.

Of course, the objection might be made that I have been approaching the readymade the wrong way round and that minimal decisions like titling, or signing, or submitting are not and were never meant to be seen as aesthetically meaningful. However, this objection would not support de Duve’s all-or-nothing binary (i.e., “This is art” or “This is not art”). In Kant after Duchamp, and again in Aesthetics at Large, de Duve explicitly aims to challenge Duchamp’s anti-aesthetic reception, including the artist’s own occasional anti-aesthetic statements. Thus, the dismissal of the artist’s acts as aesthetically irrelevant is incompatible with de Duve’s broader convictions as an art critic or historian. His point is that, after the readymade, viewers of art are now entitled, and even obliged, to evaluate all registers of the artistic intentionality presented to them in a potential work because these will serve as objective grounds for our aesthetic experiences, whatever they may be. Therefore, if it can be determined that the readymade only solicits the public’s assent that it ought to count as a work of art, provided it is intentionally presented to them as art, then the post-Duchampian artworld differs from the theoretical artworld described by de Duve, if only by an infrathin margin. I may be as responsible as the artist for the decision to count a work of art as art, but I am not responsible for its presentation or even for its having been initially chosen as such. Once confronted with the perceptible signs of Duchamp’s prior intentions, the person who continues to call Fountain nothing, as opposed to expressing dismay by calling it a bad or illegitimate work of art, is either speaking figuratively, in which case she roughly means the latter (if in a slightly different tenor), or she is mistaken.

Curiously, given de Duve’s thoroughgoing investigations into the contents of our possible answers to the questions, “What is art?” and “Why is this art?”,  he rarely reverses course to consider what someone might mean by calling an object nothing, or how she might respond to the follow-up questions: “What is not art?” or “Why is this not art?” Certainly, his work has considered the relationship between art and nonart in the modernist epoch, but his archaeology of art in the singular has returned only briefly to the moment in which Kant was actually writing his still-influential Critique of Judgment.7 His stated reason for this relative inattention is that Kant’s concept of art was still necessarily beholden to theories of the fine arts, which makes it unlike our current art-in-general system. This may be true of Kant’s concept of fine art, but not, or at least not necessarily, of his ideas about art in general (“Von der Kunst Überhaupt”), in which case art in the singular is defined in contradistinction to products of nature, products of science, and products of craft. For our purposes, we might focus on Kant’s view that works of art in general, unlike objects of nature, “are made through a power of choice that bases its acts on reason,” a definition that could surely apply to the readymade.8

To be clear, de Duve is right to insist on making distinctions between early modern preoccupations and ours. In Kant’s day, for instance, cultural connoisseurs were less concerned with evaluating art that did not seem to compare to any of the fine arts (like a readymade) than they were with considering natural accidents that looked a lot like art but were really not art.9 Similar to our confrontations with a urinal in a gallery, aesthetic feelings were in and of themselves incapable, for example, of determining whether dendrites (a particular variety of picture stones with tree-like figures on their surface) ought to remain displayed as objects of wonder or reorganized as samples of geological processes; these landscape-like cross sections of stone were considered wonderful precisely because the aesthetic delight taken in their appearance so nearly resembles the delight taken in looking at a landscape painting.

If someone were to bring me one of these wondrous dendrites and ask if it were art, and I replied that it was nothing, I would mean something quite different from what the 1917 jury meant when it rejected Fountain as nothing. I do not, for instance, see the picture stone as an act in bad faith, or as a joke at my expense, because I do not see it as meaningful in the first place. (I would, of course, be partly mistaken, because I could still analyze the cross section for its natural origins, but just not interpret it for human reasons.) On the other hand, were I to adopt the stance of some early modern collectors and declare that the stone is indeed a work of art by Nature, then I might end up replacing the word beautiful with the word art and remain perfectly consistent with the overarching idea of Kant’s “Analytic of the Beautiful,” but not with de Duve’s update to it, because my conviction would be grounded only in the object’s seeming purposiveness and not in my interpretation of an actual artist’s motives. In the case of the dendrite, we encounter a remarkable instance in which the utterance “This is art” might be true to our aesthetic experience, and yet also patently false.10

In summary, without the ready-made object’s selection as art and its intentional presentation as art, I am unclear as to how the readymade would have ever posed any art critical problems, because it would never have raised the possibility of substantive competing judgments. Even Clement Greenberg’s oft-repeated claim that a blank canvas tacked onto a wall might count as a potential modernist picture would fail to satisfy the singular condition of the pure readymade because this virtual work would stem from at least two discrete decisions: the selection of the ready-made textile—not to mention its dimensions—and the minimal, yet undoubtedly aestheticizing, move to affix it to the wall.11 De Duve appears to recognize this in Kant after Duchamp, which is why he calls the blank canvas a “specific readymade,” in contrast to the so-called unassisted readymade with its more radical reduction. This is why he must also speak of there being no possible art object with which to compare a urinal in 1917, even though within the field of applied arts, as opposed to the fine arts, one could have recovered any number of criteria to judge this object, if only Fountain had been submitted as such. In this case, the work might not have been refused just as nothing, but more scandalously as a plagiarized something. Furthermore, the earliest critical responses and formal accounts of Fountain did indeed find a conventional category of artistic work against which to compare Duchamp’s urinal: sculpture, which was also the medium that the artist had affixed to the work’s submission in his own correspondence.12 Here, Fountain would still not have been rejected as nothing, but rather as an illegitimate something.13

What all of this may mean is that the pure readymade—the kind that “reduces the work of art to its enunciative function and [is simultaneously] the ‘result’ of this operation”—can count as no more than the limit condition of any possible artworld (KD, 389). Its mythic status within Duchamp’s oeuvre should also impact our philosophical understanding of post-Duchampian aesthetics. If, for instance, I take a solitary walk through the woods and admire a flower’s beauty, I am treating nature as if it were a product of artistry. If, upon leaving the woods, I continue my solitary walk into a public restroom wherein I admire a urinal as art, I am either overestimating my own wit or I am evaluating someone else’s craftsmanship. If a stranger in the restroom overhears my designation and assents to it, he is either agreeing with me that the urinal is a work of art—though someone else’s—or we have entered into a different form of complicity wherein he judges the situation at hand to be the source of art and not the urinal itself. Unlike any readymade that I am likely to encounter in de Duve’s ideal public art museum, the pure readymade’s absolute conflation of the artist-spectator’s judgment of a ready-made object’s purposiveness with the object’s very reconstitution as purposive without purpose would render it unrecognizable as a work of art. Or, if it were to count as such, then the unassisted readymade would exemplify a radical version of psychological idealism, and not Kant’s transcendental critique.

Nevertheless, it might be claimed that Duchamp’s actual readymades still presaged the same paradigm shift that de Duve has outlined, by virtue of their pointing toward these virtual limit conditions imposed by the unachieved pure readymade. This would mean that Duchamp could still have forecasted the replacement of more specific Kantian judgments (“This is beautiful”) with more generic statements (“This is art”), because his works sufficiently deracinated artistic practices from their prior conventions in order to establish what de Duve calls the art-in-general system.14 As mentioned earlier, this system is said to have broken from the earlier system of fine arts in which art was determined on the basis of discrete art forms. In the visual arts (unlike, say, the literary arts) this has typically meant specific mediums, which—following the readymade—gave way to the singular idea of art as such. In de Duve’s Kantian phrasing, the readymade becomes a “perceptible aesthetic idea [that refers] to art as the unperceivable idea of reason … Concretely: when he made readymades public, intimating that they should go by the name of art, Duchamp did not attribute the concept art to them … Rather, he presented readymades to the viewer’s appreciation in the name of the idea of art—of art itself” (AL, 146-47). De Duve’s treatment of the word readymade as an aesthetic idea in the Kantian sense is primarily a philosophical claim, which I feel ill-prepared to evaluate. However, this theoretical interpretation also results in a radical expansion of the critical purchase made by the term readymade in art historical contexts, such as when applied to post-Duchampian conceptual art and the so-called linguistic turn. These historiographical entailments are worth considering because they suggest to me that even if readers find de Duve’s Kantian analogy to be theoretically persuasive, they may still find his aesthetic critique ill-equipped to answer extra-philosophical questions.

Just a few sentences later, for example, de Duve writes, “Words are readymades: writers choose their words, they do not make them; and writers are readers” (148). The argument here, which is part of de Duve’s analysis of text-based conceptual art practices, is somewhat difficult to parse. We might initially be inclined to dismiss it as wrong out of hand because, of course, not all ready-made items are readymades proper, and thus there is little reason to accept that all words are readymades simply by virtue of their being selected as opposed to their being made. (This account would be especially problematic for the interpretive arts.) On the other hand, we might try to accept de Duve’s claim that all words are indeed readymades, while also maintaining our ordinary belief that not all words (such as those you are currently reading) are art. Whereupon we would be left with a world in which some readymades are art and some readymades are nothing. However, this would mean that our judgments regarding a ready-made object’s status as art would yet again depend upon the object’s intended presentation to us as art and not solely upon its mere selection. This latter possibility is logically implicit in de Duve’s own critique. Had Duchamp not “presented readymades to the viewer’s appreciation in the name of the idea of art,” then there would still have been ready-made objects (such as Mott Factory urinals), but there might not have been readymades (these same objects presented as art). In Kantian terms, the readymade’s distinction between art and nonart must therefore be determinative, because our judgment that Fountain and the readymade ought to count as instances of art requires our conceptual understanding that the same activity—namely, choosing ready-made things—can be put to different uses.15 So long as you are theoretically prepared to subsume toilets to the laws of the signifier (and hence, to entertain de Duve’s analogy between picking out urinals and picking out words), then you are also prepared to judge a readymade as art, and to do so in the absence of any particular aesthetic experience before Fountain.

De Duve’s underlying claim that Duchamp’s readymades charted the movement from specific (painting) to generic (art itself), may in the end be a more compelling account of the history of the readymade’s reception than a theoretical account of the readymade, because his emphasis on the readymade’s genetic connection to art in the singular has meant his downplaying of the specific significance that Duchamp (and many later critics) attached to the term readymade. In other words, before the artist determined that ready-made objects could be judged as works of art, he had first to name them readymades. “The word ‘readymade’ did not appear until 1915, when I went to the United States. It was an interesting word, but when I put a bicycle wheel on a stool, the fork down, there was no idea of a ‘readymade,’ or anything else. It was just a distraction.”16 In 1913, then, his Bicycle Wheel was not yet a readymade, because there was no idea of a “readymade.” It was not art either, because without this idea of the readymade (rather than the idea of art itself), Duchamp could only judge the object to be a distraction. In 1917, Fountain was rejected as a sculpture, and perhaps rightly so. (De Duve has persuasively argued that readymades have much more to do with the history of painting than they do with any other specific medium.) It is untrue, or at least not necessarily true, that Fountain was then subsequently accepted as art as such, or as nothing at all. It was accepted as art as a readymade.17

In Duchamp’s transatlantic passage, ready-made-ness shifted from a general descriptor of the articles in the artist’s studio to an artistic genre descriptive of those very same articles presented as art. If, as de Duve has taught us, the readymade never completely exited from the realm of aesthetic judgment, and instead moved that realm away from the aestheticist taste for beauty, then it does not necessarily follow that the only other available criterion for our judgments would be the one co-identical to the work’s aspiration to be accepted as art. Were this to be true, one might have expected a diminishing interest in readymades, or at the very least, an only nominal interest. Instead, as early as 1916, Duchamp’s readymades became a kind of art, in competition with (his) paintings. And though this competition promised to conclude itself once the artist recognized that readymades could also be paintings, and vice versa, no art historian has failed to notice the difference between a readymade (Fountain) and a painting (Nude Descending a Staircase, No. 2) when one was meant. The only way that the readymade could have ever succeeded in its campaign to be judged as art was by establishing itself as a new kind of art, under which it was properly described. One criterion for evaluating this new kind of art, if not the criterion retained by Duchamp, was neither beauty nor art as such, but rather interest. (Was this not his point when he baptized a financial instrument [Monte Carlo Bond, 1924] as art?)18

Thus, the visitors to Duchamp’s Bourgeois Gallery show in 1916 tended to pay no attention to his earliest presented readymades, because they thought, as had the artist himself only three years prior, that his contributions in the vestibule were no more than mere distractions. Once critics gave a name to the game, interest could be piqued and the public could now judge those things hanging on a hat rack, and maybe also the hat rack itself, as potentially art.19 No matter what the readymade’s relationship to or against the history of painting may have been, and in spite of its purported solicitation of only the most unspecific of aesthetic judgments (the feeling of the idea of art itself), it nevertheless came to constitute one of the first new genres (if not the first new genre) in art history since Watteau’s fête galante, such that its existence has never been left completely unspecified (i.e., it had certain fulfillable criteria that Duchamp, his earliest champions, and later art historians could recognize or dispute, and it has since gone on to acquire distinct standards of achievement wherein the relative merits of, say, a Duchamp versus a Koons can now be weighed).

Treating the readymade as an artistic genre has the additional benefit of acknowledging its intentional circumvention of specific mediums as the privileged loci for artistic conventions, without also losing sight of its now century-old conventionality. Like the history of satire or portraiture, the history of the readymade simply tracks across media. This generic conception of the readymade would also encourage us to recover an alternative materialist definition of art already in circulation in the nineteenth century, which departs from the limited, medium-specific definition that de Duve finds enshrined in the system of the fine arts, and also in his own idealist definition of art in the singular. In the 1835 edition of the Dictionnaire de l’Académie Française, art was defined as a “method for making a work, for carrying out or executing something according to certain rules.”20 Any reader of Kant after Duchamp or of de Duve’s earlier Duchamp monograph, Pictorial Nominalism, will know with algebraic clarity just how well Duchamp’s readymades adhere to this conceptual definition.

By defining art as a rule-bound cultural technique, this 1835 definition might also allow us to circumvent the institutional definition of art against which de Duve has long contrasted his Kant after Duchamp approach, because it would treat a work and its organizing conventions as materially immanent to one another. It also opens up a different set of ordinary responses to the question “What is art?” than those provided by de Duve in the introductory chapter of Kant after Duchamp. Readers of this work will recall that de Duve sets off with a remarkable survey of various disciplinary approaches to that question, ranging from alien anthropology to critical sociology, to critical philosophy, until he finally arrives at the hypothetical responses of non-specialists. These mere art lovers are said to lack a fixed intellectual method, and so their answers end up amounting to their humble catalogs of specific works of art. These responses should be viewed as no more than a rhetorical heuristic, because they already presuppose de Duve’s particular definition of art and exclude a range of other possible answers. For instance, a non-specialist might just as easily develop a list of the creative activities that she associates with art, in which case her definition of art would not congeal into her preexistent collection of art objects and would instead appear more like a dynamic field of practice. This enlarged picture of art making would also admit into the art-in-general system a range of artistic activities that have since departed entirely from the jurisprudential conventions first brought to the fore by Fountain’s submission in 1917. In the non-critical realms of hygiene (art therapy) and art-based recreational activities (“Drink and Draw”), the underlying aim of art is to be made and not necessarily to be beheld.21

By defining art as a meaningful form of making, our new imaginary respondent to the question, “What is art?,” may indeed be unaware of Marcel Duchamp’s substitution of choosing for making. However, once faced with Fountain, she could also decide that Duchamp’s conflation of both terms is illegitimate. Moreover, she could decide against accepting this conflation while still maintaining that his work ought to count as art. This is a possibility that de Duve never seriously entertains, because his view of the art-in-general system depends upon his acceptance of Duchamp’s unassisted readymades as already furnishing the exemplary examples of just such an equation.

From an art historical perspective, so far, I have questioned de Duve’s recruitment of the so-called pure readymade as the paradigmatic example of post-Duchampian art and his Kantian critique of post-Duchampian aesthetics because I doubt that a truly unassisted readymade has ever, or could ever, meaningfully satisfy what de Duve values most in the judgment, “This is art.” In Kant after Duchamp, de Duve stressed that Duchamp’s Fountain was exemplary because “the testing device [a urinal called art] was obviously designed to be itself submitted to a test” (91). The obviousness of Fountain’s designs is precisely what distinguishes post-Duchampian art from either the pure readymade or from the objects of pure judgments of taste. When an exquisite seashell washes ashore and I call it beautiful, I am describing a pleasure that I take in intentions that merely appear to lie behind the shell’s design and, maybe also, the shell’s presentation to me.22 When I discover an exquisite drawing while rummaging through the contents of a junk sale, and pronounce it beautiful, I am naming a pleasure taken in the actual appearance of some unknown author’s underlying intentions as a draftsman, even in spite of its having been presented to me as rubbish. When I visit the modern art galleries in the National Gallery of Canada (where de Duve collaborated with curators to “de-institutionalize” its Duchamp displays [AL, 54]) and call a shovel leaning against a wall a work of art, I need not be describing any aesthetic feeling at all, because I could just as well be voicing my intuition that if this shovel were not meant to be in the gallery, then it would not. Or maybe I might express some doubt, and instead of declaring the shovel to be nothing, I might ask, “Is this art?,” in which case I would hope to be provided with some art historical explanation, rather than someone’s subjective account of his or her aesthetic experience.

This brings me to more uncertain territory and to my second line of questioning for de Duve’s Kant after Duchamp approach. De Duve turns not only to the history of art but also to analytical philosophy to justify his substitution of the word art for the word beauty. Were we to agree with him that “the reception history of the readymades seems to impose [mechanically replacing the word ‘beautiful’ with the word ‘art’]” (AL, 153), we might still ask ourselves whether this substitution is of critical use to us as art historians, art critics, and/or art lovers. As mentioned earlier, Kant’s account of beauty is at its core an analysis of a specific form of pleasure (i.e., this sunset is beautiful because it pleases me in just this way). Under de Duve’s redescription, this same analysis is now updated for a term that expresses any feeling whatsoever, and thus expresses nothing in particular about the varieties of human feeling. This indeterminate art feeling is said to be formed through a spontaneous comparative evaluation of the object presented to us as art and our inherited mental collection of works of art (i.e., this object is a work of art because I feel that it ought to count alongside other examples of art that I have previously encountered or have been told about).23 De Duve mentions enthusiasm and disgust as potential art feelings, but surely anyone who has ever visited a modest regional museum or a seasonal art festival has also found themselves saying “This is art” in an affectless deadpan, until a work worthy of the exclamation “This is art!” comes into view. By collapsing such an incredible range of emotion into the single category of the aesthetic experience of art for art’s sake, de Duve’s critique will not be of much use should we agree to count a work as art, but then disagree as to its value. How, for instance, would my judgment that an instance of installation art is indeed a work of art answer Rosalind Krauss’s judgment that the installation is meretricious? By characterizing the judgment of art as an aesthetic judgment through and through, de Duve may also downplay the possibilities of erring in one’s judgment—such as the unwarranted esteem a post-Duchampian critic might direct toward some randomly encountered, upturned, and graffitied urinal never intended as art—and the conceptual resources that one might rely upon in order to revise one’s view.

More importantly, while I readily accept that the predicative statement “This is art” entails some sort of judgment, I worry that de Duve’s particular intension-less interpretation of the word art (art is “made of things and not of meanings” [AL, 43]), when used in this statement, must necessarily admit non-aesthetic judgments. For instance, I might be taken on a tour of an art museum’s storage facility, in which I encounter crates of packed works of unknown mediums, or walls lined with flat files, or stacks of rolled canvases. I could in this moment declare “This is art” with complete conviction, and command anyone and everyone’s agreement, but do so without passing an aesthetic judgment. Curiously, if this statement did indeed refer back to some original aesthetic feeling—just ask anyone who has had the opportunity to visit a major museum’s storage facility—it would be a feeling much closer to the Kantian sublime than of the beautiful, and also one completely divorced from the Duchampian question of any individual work’s legitimacy as art.

Rather than clearing up one of de Duve’s most counterintuitive claims about post-Duchampian art in Kant after Duchamp—that it fulfilled (and so fulfills) philosopher Saul Kripke’s concept of proper names—Aesthetics at Large may only add to a reader’s confusion through its introduction of a new cluster of art categories (art altogether, art as such, art-in-general, and art itself), all of which appear to demonstrate just how easily art can be used non-rigidly, even in the circumscribed context of aesthetic reflection. De Duve first turned to Kripke because he needed a philosophical foundation for his usage of art as an indeterminate concept. Specifically, he wanted to find a way to define art that made no claims about art’s meaning and would also not be reliant on institutional theories (i.e., this is art because the artworld calls it art). Kripke claims that we can “call something a rigid designator if in any possible world it designates the same object.”24 The name Aristotle thus fixes a reference to a particular identity, even if in some counterfactual scenario a previously clear identifying description of the philosopher might become unavailable.25 Somewhat more controversially, Kripke also extends his concept of rigid designation to natural kind terms, such as gold and meter. In Kant after Duchamp, de Duve claimed that it was “preferable to speak of art as a proper name in the narrow sense rather than as a rigid designator … because of the essentialism to which Kripke’s category ultimately leads” (57-58). In Aesthetics at Large, however, de Duve offers a brief apology for the unclear ways in which he had previously analogized art to proper names—though he only cites Barry Schwabsky’s criticism—but defends the overall correctness of his usage of Kripke’s ideas by comparing art to the rigid designation of natural kind terms.26 “When Kripke claims that names of colors, measures, natural kinds, and substances such as gold, are rigid designators, it is precisely on the basis of samples being compared with a paradigm sample acting as a standard of comparison” (150-51). For Kripke, a kind term’s rigidity connotes that if we were to consider another possible world in which the United States had adopted the metric system, then we would still mean for the name meter to fix a reference to the same specific measurement in our actual world (the meter’s essential feature), even though in yet another possible world, a meter might just as well have been called art or be set to a different standard.

This is not de Duve’s meaning of paradigm sample when he speaks of art, because he still rejects Kripke’s essentialism. Precisely, his historiographical account of the transition from a fine arts artworld to an art-in-general artworld requires that we appreciate the conditionality of the usage of art as a rigid designator. Duchamp’s achievement was to announce a paradigm shift in our judgments of art by creating a work that expanded the term’s possible extension. Were there to be a similar shift in the metric paradigm, we would use meter to refer to an entirely different object. Moreover, Kripke’s account of natural kind terms affirms the mind-independent existence of natural objects. What this means is that if I offer you a piece of fool’s gold as real gold, I am either ignorant of gold’s identity and require better training in sorting different metals, or I am learned and aim to defraud you. Neither my error in judgment nor my mercenary behavior has any real consequences for the identity of gold itself. The paradigm sample is unaffected by my judgment. By contrast, if I offer you a picture stone as a work of art, I am either ignorant of the stone’s true origins and I am thus mistakenly treating a natural accident as if it were a painting, or I do understand the object’s true nature, and intend instead to expand the application of the term art. Unlike natural kind terms, art appears uniquely, or at least distinctly, liable to shifts in our shared criteria, given that what counts as art—following de Duve’s critique—is dependent upon our aesthetic judgments and not on the perfectibility of science. The urinal selected for Fountain has an existence that is mind-independent of both Duchamp and the beholders who ratify his judgment that the same object ought to count as art. Its status as art is, according to the Kant after Duchamp approach, entirely recognition-dependent. To put things in more Kripkean terms, in another possible world, the Mott Factory urinal would still exist as a urinal, but not as art. By contrast, if no one ever designated gold as gold, gold would still be gold, and it would necessarily remain gold in all possible worlds.

But even granting that de Duve’s paradigm sample of art is in fact Kripkean—which as a nonexpert I readily admit to be plausible—I would still have trouble determining from de Duve’s writing what object art rigidly designates. (I have no such difficulty with gold, or with meter, or with Aristotle.) De Duve claims that in the judgment “This is art,” art fixes the reference “to all the other things equally designated by you” as art in your own imaginative collection of works (KD, 59). De Duve’s claim is more precisely that art, as a natural kind term, picks out exemplary examples of art, and (without providing a justification) he restricts these exemplary examples to specific works of art. The potential problems that this restriction might pose for a Kripkean account of art should be obvious. In other possible worlds—crucially in worlds that have also been described by de Duve—the exemplary example of art might not be Fountain (a specific work), but rather, the readymade (a paradigm or genre of art). To offer one additional example, for Rosalind Krauss the exemplary example of art in an era of art-in-general would not be Fountain, or the readymade, but instead the specific medium, whatever that may be in what she diagnoses as the post-medium condition.

The various art categories developed in Aesthetics at Large are meant, it would seem, to offer greater theoretical precision because they allow for de Duve’s readers to distinguish between his various usages of the term art, and consequently, to not confuse every appearance of the word art with his larger philosophical claims regarding the term’s status as a proper name or rigid designator. As if in anticipation of questions regarding the validity of this maneuver, de Duve has claimed that his specification of a term’s uses does not count as providing the word with conceptual definitions, thus there is no contradiction in claiming that art functions as a rigid designator while also using art to refer to objects non-rigidly. I find this reasoning difficult to follow, given that one of the conventional uses of dictionaries is to define a term precisely by specifying its uses.27 Nevertheless, it should be highlighted that only the word art in de Duve’s term of art, art altogether, conceivably fixes the particular reference that de Duve needs for his claim that art is a rigid designator. Art altogether refers to that universal collection of art objects that serve as our basis for comparison when we wish to count this (an object presented to us) as a work of art. Alternatively, the art in art as such points to a collection of art-related feelings, such as enthusiasm or revulsion, which would therefore mean that the art in art as such is purely connotative. Finally, the art in art itself points to the regulative idea that brings these art-related objects and art-related feelings together in aesthetic harmony (AL, 40-49). These terms will be of use to readers already prepared to grant de Duve’s difference between a word’s usages and its definitions, but they will likely remain points of confusion for those of us still grappling with how to make sense of the claim that art functions as a proper name.

Even if this new terminological armature were to have helped me understand de Duve’s continued nomination of art as a rigid designator, I would still have follow-up questions regarding the logic of his argument. Say, for instance, that you are sitting in a classroom, setting up materials to begin an oil painting, when a fellow student enters the room and asks, “Is this Chemistry?” If you respond, “This is Art,” you would have successfully used art as a proper name or rigid designator, as I understand the concept, but without any necessary reference to art altogether. One might complain that this example does not exactly correspond to the substance of de Duve’s thesis, which seems to depend upon a conditional usage of the statement, “This is art” (“when you show your appreciation of something” [AL, 151]). However, if de Duve’s thesis that art is a rigid designator really is conditional upon its usage in the context of aesthetic judgment, then his aesthetic critique is only ostensibly a quarrel with earlier theories of art, which made no such restrictions. Furthermore, even if we were to accept de Duve’s conditional parameters, there would still remain the possibility of using art to refer to objects non-rigidly. Suppose, for example, that you are a devotee of pictorial modernism and decide to visit an art museum with your Renaissance specialist friend. After your friend indulges you with your admiration of the museum’s Pollocks, she asks you to join her to look at some Titians. Immediately upon entering the Titian galleries, and even before she has time to examine the paintings in any detail, your friend declares, “This is art.” In this case, the speaker is certainly passing an aesthetic judgment on both the Titians and the Pollocks. However, she is using art aesthetically as an honorific that is connotative through and through. Inasmuch as she has made this judgment by following her prior preference, and even before having had the time to direct her gaze toward any particular object for praise, her usage of art—in the Kantian sense—is hardly disinterested. In de Duve’s nomenclature, she would have been using art to mean art as such, indicating that she does not mean to fix a reference to art altogether, and instead wishes to express her own partiality as an aristocrat of taste. Now, say I object to her snobbishness, and she draws me closer to inspect a favored work. Pointing out a particularly lush passage of brushwork, she doubles down: “This is art.” In this case, art belongs yet again to an aesthetic judgment, but it fixes a reference to Titian’s quality of execution, perhaps in comparison to the seeming artlessness of Pollock’s dripping procedure. In both cases, art picks out exemplary examples, but the paradigm of exemplarity shifts in each instance: the first refers to the quality of artistic works in toto, and the second refers to their local quality of execution. Unlike meter, which refers to the same standard of measure in all possible worlds, art can be used within the restricted context of aesthetic appreciation in the sentence “This is art” to refer to different categories of objects, and it can do so even when we remain in the very same room and discuss the very same painting.

I suspect that all my questioning boils down to one of the central preoccupations of de Duve’s intellectual project in Kant after Duchamp and Aesthetics at Large: the difficulty of bypassing our quarrels over what art means, even in spite of—or perhaps as a result of—our post-Duchampian condition. Against avant-garde-inflected theories of art, de Duve’s latest book offers a spirited defense of art’s autonomy, but does so explicitly in terms of his Kantian theory of art’s capacity to fix a reference. Nevertheless, when he uses connotative language like genuineness or treasure to describe individual works of art, or when he raises the specter of a Norman Rockwell blockbuster at his ideal art museum, it becomes clear that de Duve’s most important insights about what makes art meaningfully art are irreducible to the name by which we call it. Duchamp taught de Duve to prioritize the ratification of objects as art over the longstanding disputes surrounding art’s meaning or making as the fundamental concern for criticism after the readymade, but the artist’s work and de Duve’s singular analysis of its art historical reception might encourage readers to arrive at the exact opposite conclusion. In this regard, I would encourage fellow art historians to pay particular attention to the ninth chapter of Aesthetics at Large: “Kant’s ‘Free Play’ in Light of Minimal Art,” in which de Duve offers a brilliant philosophically-informed intervention into critical debates on the meaning of Robert Morris’s Untitled (Three L-Beams). On other occasions, though—such as when de Duve muses that “everyday language is a spontaneous artist” simply because of the referential structure of the expression, “you and me” (AL, 84)—this art philosopher’s ideas will demand that his readers ask themselves whether he is just momentarily speaking in metaphor or if he has stopped referring meaningfully to art altogether. In this case, however, I am glad to refer to the book’s powerful thesis for my answer: This is art.