Abstract It is often thought that the boundaries and properties of art-kinds are determined by the things we say and think about them. More recently, this tendency has manifested itself as concept-descriptivism, the view that the reference of art-kind terms is fixed by the ontological properties explicitly or implicitly ascribed to art and art-kinds by competent users of those terms. Competent users are therefore immune from radical error in their ascriptions; the result is that the ontology of art must begin and end with conceptual analysis. Against this tendency towards concept-driven ontology, I offer a trio of objections derived from: (1) the cultural and temporal variability of concepts of art, (2) the systematic tendency, on the part of would-be ontological assessors, to err on the side of familiar categories or, conversely, to exaggerate minor differences between familiar and unfamiliar practices, and (3) the influence artworld precedents exert over expert and folk concepts alike. These considerations, I argue, mandate an epistemic humility that is simply unavailable to the concept-descriptivist. 1. Introduction It is sometimes thought that the boundaries and properties of art-kinds are determined by the things we say and think about them. This is because art-kinds are thought to be paradigmatic instances of social kinds, which means that their ontology is not grounded in homeostatic property clusters, an underlying real essence expressible by microstructural properties, or causal powers capable of entering into powerful empirical generalizations. Unlike natural kinds such as bismuth or potassium, no particular arrangement of particles can explain what makes dance or sculpture art-kinds, nor will it explain where printmaking ends and photography begins. And if the ontology of art and art-kinds depends on what we say and think, then it seems to follow that we cannot be substantially wrong about the ontological properties we ascribe to them, so long as ‘we’ is suitably specified. ‘Painting’ is thus taken to refer to the more-or-less two-dimensional application of pigment to a surface, usually canvas, paper, wet lime plaster, or wood, while ‘sculpture’ refers to the more-or-less three-dimensional manipulation of matter, usually in ceramic, metal, stone, or wood, and so on for other art-kinds. The result is that paintings and sculptures are concrete objects with relatively strict identity and persistence conditions, while dances, novels, and musical works are abstracta of one stripe or another and can survive much more radical changes. At its extreme, this tendency to privilege concepts manifests itself as what I call conscriptivism, a portmanteau of ‘concept-descriptivism’: Conscriptivism The reference of art-kind terms is fixed by the ontological properties that competent users of the terms explicitly or implicitly ascribe to them.1 According to the conscriptivist, the reference of any art-kind term is grounded in a concept of the ontological kind involved as it is manifested in, for example, discourse about the art-kind among competent users. The ontology of art is thus relegated to an exercise in bare conceptual analysis with no meaningful chance of failure, provided the right experts are consulted. Although conscriptivists remain few and far between, philosophers of art are nevertheless quite fond of claiming that cultures (including our own) can only be said to have art if they first possess a concept of art, as evidenced (perhaps) by the presence of such a word in the language, or by the existence of a class of objects that functions culturally just as art does in our culture (or in some other culture).2 While this commitment to concept-dependence3 is distinct from the commitment to conscriptivism, both have their roots in concept-first ontology, the idea being that art-ontological puzzles are best resolved by conceptual analysis. I argue that it is this move which we ought to avoid, because our concepts of social kinds make for untrustworthy guides. So do our concepts of natural kinds, of course; the difference is that, where natural kinds are concerned, the physical world acts as a readymade system of checks and balances on our theorizing. When it comes to social kinds, the sources of reality checks are less obvious. In what follows, I will argue that appealing to our concepts is bound to seriously misrepresent the ontology of art and art-kinds. This is because a growing body of anthropological, art-historical, and psychological evidence indicates that our concepts of art and art-kinds reflect entirely arbitrary historical interests with a limited range of application. Nor is the problem limited to ‘folk’ theories or concepts4—I will argue that the sources of error are pervasive, and depend on basic features of human culture and psychology. The result is that even artworld ‘experts’ are liable to deploy flawed concepts and intuitions, and are no more likely to deploy those concepts in a manner that accurately reflects their ontology, whatever it may be. The concepts of competent users thus have minimal evidentiary value for the ontology of art. I will begin, in Section 2, by introducing the problem of ‘art’s’ conceptual instability, which gives us reason to doubt that ‘art’s’ content is stable across the uses to which it was put in the history of our own culture, let alone other cultures. If concepts are to guide ontology, then ‘art’s’ history offers us an embarrassment of riches. Against the response that what we are interested in is the concept that competent users of the terms in our present culture deploy, Section 3 will present the problem of conceptual imperialism. ‘Conceptual imperialism’ names the tendency of would-be ontological assessors to err on the side of familiar categories, leading them to assimilate unfamiliar practices under the banner of familiar concepts or to exaggerate minor differences between familiar and unfamiliar practices. What results are serious mistakes about the extension of the concept being deployed. Nevertheless, it remains open to the conscriptivist to reply that conceptual instability and imperialism simply disqualify susceptible speakers as competent users of the relevant kind-terms. Section 4 responds to this objection by introducing the problem of conceptual inclination, which concerns the influence artworld precedents exert over folk and expert concepts alike. Ultimately, I argue, the concepts and intuitions to which the conscriptivist would have us defer reflect conventionally salient features of our thinking, rather than ontological bedrock. Consequently, we should not trust bare conceptual analysis to reveal art’s ontology. 2. The problem of conceptual instability The first problem for conscriptivism concerns its ability to secure the source of its privileged concepts. This is what I call the problem of conceptual instability. The idea is that the diachronic consideration of our concepts of art and art-kinds yields a proliferation of different, often contradictory data points. A growing body of anthropological and historical evidence indicates that the concepts underpinning the various practices we regroup under the banner of ‘art’ are not stable across time, let alone across human cultures. Our artistic practices have changed a great deal over time, not just in terms of techniques but also with respect to their functions and the place they occupy in our societies, and this has led to changes in our artworld concepts. Consequently, there are bound to be at least as many concepts as there are cultures and times, if not as many as there are users of the relevant kind-terms. The result is that the conscriptivist cannot secure the uniqueness of the reference of her terms except by arbitrarily restricting herself to particular expressions of the relevant concepts which share all or most of their content. The case for ‘art’s’ instability begins with two hugely influential essays by Paul Oskar Kristeller, who argued that the modern ‘system of the arts’ emerged in the mid-eighteenth century through the work of Charles Batteux.5 According to Kristeller’s story, the ancients had concepts of ars and techné, but these did not specifically denote the ‘fine arts’ and instead included all manner of craft or scientific activities. The mediaevals likewise had no concept of the fine arts and insisted in applying ‘art’ to crafts such as shoemaking, juggling, and arithmetic. This conflation persisted through the Renaissance until the modern writers began to group some artefact-making activities (especially painting, sculpture, and architecture) together and distinguishing them from crafts. Still, it was Batteux who first ‘correctly’ grouped together the arts which define our system of the arts today: poetry (including literature), painting, sculpture, music, and dance. These he grouped together on the grounds that their primary goal is pleasure and from them he distinguished the ‘mechanical’ arts, which serve practical needs such as food and shelter, and a third category combining the two.6 Although Kristeller’s view still holds considerable sway, there is good reason to suspect that it is just another ‘Just So’ story. For one thing, as James I. Porter has pointed out, alternative analyses of the eighteenth century are available.7 More importantly, much of the historical evidence that Kristeller adduces in support of his claim that, prior to Batteux, there was no concept of ‘fine art’ or any systematic distinction between fine art and craft, is simply false. So, for example, Kristeller accuses Plato in the Cratylus of numbering the imitation of animal noises among the arts.8 But, as James O. Young has pointed out, the discussion of animal noises in the Cratylus has nothing at all to do with art, let alone fine art: Plato is simply arguing that one does not name animals by imitating their noises! Young’s careful work reveals that Kristeller routinely mistakes discussions of imitation and imitative practices for discussions of artistic practices. Worse still, Young observes that both Aristotle and Plato did identify music, dance, poetry, painting, and sculpture as arts and distinguish them from other art-forms, and that Batteux himself explicitly credits Aristotle with this grouping.9 Kristeller’s contention that pre-modern peoples did not share our concept of art is thus grounded in a systematic conflation of pre-modern talk of ‘the arts’ with the fine arts.10 To this list of errors we might also add that Batteux did not actually isolate the fine arts from considerations of morality and utility.11 The result is that there is no sound historical reason to believe that our ancestors had a wildly different concept of the fine arts; what they may have had, instead, is a broader notion of the arts in general—certainly one that is broader than our notion of fine art; so much the better for the conscriptivist. Kristeller was not wrong on every count, however. For one thing, we should be suspicious of the idea that the concepts of individual art-kinds are stable over time, so that, e.g. Greek music or theatre operated just as ours do today, or even as they did in mediaeval Europe. I will deal with this subject in more detail in Section 3; for the time being, let us assume this view is substantially correct. Even if individual art-kinds exhibit diachronic stability, it should be clear that the class of the fine arts is not especially stable.12 Art-kinds pass in and out of it, depending on prevailing cultural attitudes. Contemporary culture, for instance, has added photography and film, while gardening exists only at the periphery of fine art today (at least in the West). More importantly, Kristeller was right to focus his attention on eighteenth-century Europe because the practice of art-making and the cultural attitudes surrounding art underwent rapid changes at the time. As Peter Kivy has put it, ‘the eighteenth century witnessed a veritable explosion of separate, self-contained works on the subjects of the fine arts, beauty, sublimity, taste, criticism, the standard of taste, and much more—works unprecedented in Western intellectual history’.13 It is no coincidence that it is in the eighteenth century that aesthetics was elevated to the status of a canonical philosophical discipline, that works of visual art acquired titles and were exhibited en masse, and that aesthetics and the philosophy of art were first distinguished from one another. Kristeller’s cardinal sin was to identify the eighteenth century with the first moment in history when a consensus over ‘art’s’ conceptual content was possible; what actually happened in eighteenth-century Europe is that people finally began to philosophize about art in the full knowledge that that was what they were doing.14 They began to deploy theories of art in order to reflect upon the value of art and artistic practices. Recent evidence suggests that Kristeller misidentified the conceptual change that dominated the eighteenth century; what was new was not which kinds of things people classified as ‘art’, but rather the cultural role those things started to play for them. The eighteenth century saw the crystallization of the concept of an artwork as a special kind of entity requiring a special kind of attention. An example will serve to make this clear. According to Lydia Goehr, the seventeenth and eighteenth centuries saw a number of important changes to musical practices, culminating in the nineteenth century in a new conception of musical works as standalone ends in themselves. This change was driven primarily by the development of accurate notations, as well as the concepts of composers, perfect compliance, performance-of-a-work, and scores.15 Before the shift, composers enjoyed very little creative freedom and a composer’s identity mattered less than the occasions for which or persons by whom works were commissioned.16 Musical performances were closely tied to particular extra-musical goals, e.g. as background accompaniment (where false starts were common), and still depended to a large extent on words. Passages were frequently re-used to suit similar occasions and there was a great deal of ‘creative’ overlap between composers and pieces. The absence of complete and institutionalized systems of notation also meant that performers were given a great deal of leeway to complete or interpret passages and that music was not typically made or expected to outlast its performances.17 But after the shift, musical performances increasingly became the direct focus of public attention and appreciation, freeing them from service to particular extra-musical goals and from their dependence on words. Compositions became self-sufficient entities, ‘works’ to be published, repeatedly performed, and enjoyed for their own sakes. None of this is to say that musical works were not sometimes appreciated for their own sakes prior to the nineteenth century, that there were no functioning concepts of composition, performance, or notation prior to the eighteenth century, or that musical works did not exist before then.18 The point, rather, is that the musical work-concept (as we understand it today) began to regulate musical practice—and our conception of musical practice—in a significant way at the end of the eighteenth century. What changed was the significance we attributed to musical works and to various aspects of their production. These conceptual changes are significant where conscriptivist ontology is concerned because they serve to ground new kinds of claims about the first-order properties of musical works. Once musical works became systematically scored, the dominant conception shifted from thinking about them as occasion-specific performances to thinking of them as repeatable and multiply-instantiable entities with much stricter identity and persistence conditions governed by compliance with the score. If the conscriptivist thesis is to be believed, then this change in the uses to which competent speakers put musical terms reflects a change in the ontological properties which they explicitly and implicitly ascribed to them. This means that when we enter the musical ontology room, we have at least two sets of properties to consider: those encoded by competent users of musical terms prior to the development and adoption of a complete system of notation, and those encoded afterwards. The conscriptivist faces a dilemma: either one set of speakers is incompetent, or pre- and post-notation practices belonged to different art-kinds entirely. If the former, then how should we choose between them? If the latter, then she faces the additional problem that most speakers do not typically draw this distinction in their music-talk—that is to say, most ordinary users of music-terms are incompetent because they are not aware of the difference that notation made. These problems are not limited to our concept of musical works. Tiffany Stern has likewise observed, of theatrical productions, that their practice prior to the late eighteenth century departs radically from what we have come to expect.19 Consider Shakespeare’s plays, for which there was, contrary to our current practice, no ensemble revision. Actors practised individual rehearsal (‘study’), but received no direction on how to behave when not speaking (there were no producers or directors in the modern sense). Nor were actors given much indication of the play’s content: there was no complete typescript to which they could refer—they were only given their cues and lines. Finally, the play’s opening night would have been the actors’ first opportunity to get a sense of the endeavour as a whole, but it was also a trial run of the play’s plot and writing, which were subsequently subject to major rewrites.20 The result is that, contrary to our ordinary intuitions, Elizabethan theatre was not a text-based art-kind, and neither, perhaps, were any theatrical productions prior to the crystallization of a work-concept in the nineteenth century.21 That is not to say that there existed no playscripts prior to the nineteenth century; rather, the point is just that the script did not yet exert the kind of regulative force we routinely assign to it today. And as with the musical case above, it is a simple matter to see how the conceptual shift between the Elizabethan era and our own could result in competent users of theatrical terms encoding different sets of identity and persistence conditions—and maybe even classifying them as different sorts of entities altogether (e.g. action types vs. events). Generalizing from these two sets of cases to the artworld more broadly, the lesson here is that the differences between an old and a new artistic practice are not always readily apparent because artworld conventions develop incrementally over time and are historically contingent. As Michael Baxandall has observed, the inferences we draw about artworks are based on our understanding of the work’s historical context and the interests driving artists at the time, and these are in turn based upon our understanding of our own historical situation and interests.22 As a result, our concepts are liable to reflect anachronistic ontologies. The trick is to discover when and where the analogies to our own time break down, and to revise our reconstructions in light of those facts. In other words, there is no fully independent standard to which we can appeal, and so we should be wary of leaning too heavily on our concepts. 3. The problem of conceptual imperialism So far, I have argued that the conceptual content of ‘art’ and individual art-kinds is unstable across time, and that this fact should caution us against top-down approaches to art’s ontology which begin by taking some concept in hand and then squint at its entrails in the hope of divining something about art’s nature. For such a strategy to have any hope of working, we would have to repeat it for as many of the historically-situated concepts of art as we can discern, and the result will be far too many distinct ‘concepts’ to secure the reference of art and art-kind terms. But it remains open to the conscriptivist to argue that what she is actually concerned to describe are concepts with a much more limited range of application: our concepts, properly indexed. In fact, this is exactly how Amie Thomasson has conceived of her own project. To put the point in her terms, the descriptivist (my conscriptivist) can respond that what we are interested in are the application and co-application conditions of ‘art’ explicit or implicit in the categorial intentions of the competent users of the term who ground its reference in English (or in our artworld or culture). Even so, I will argue that the project stumbles on what I call the problem of conceptual imperialism, which describes two related tendencies: (1) the tendency to assume that objects and practices with which we are unfamiliar fit into the conceptual categories that are most familiar to us, and (2) the tendency to exaggerate the differences between practices when one of these is unfamiliar to us. It is, in effect, a way of begging the question in favour of the categories that are most familiar to us, and adds two synchronic problems to the conscriptivist’s original diachronic problem. First, it introduces a new source of erroneous ontological ascriptions, namely, our approach to understanding unfamiliar practices. Second, we routinely apply ‘our’ concepts far beyond ‘our’ practices, exporting them to cultures across the globe. Conscriptivists cannot simply restrict themselves to describing ‘our’ concepts because our use of them is bound by no such restrictions. Conceptual imperialism is at its most obvious in the way we talk about the artistic practices of other cultures. Consider Bill Holm’s seminal Northwest Coast Indian Art: An Analysis of Form. In this landmark monograph—often called the bible of northwest coast art—Holm presents an analysis of the formal characteristics of some four hundred objects created by the Kwakiutl people of the Pacific northwest and tries to derive the aesthetic and artistic principles underpinning their creation. And yet, his preface contains this startling admission: Ideally, a study of this sort should lean heavily on information from Indian artists trained in the tradition that fostered the art. Unfortunately, I was unable to locate a qualified informant from the area covered, i.e., the coastal region from Bella Coola to Yakutat Bay. That there may be some still living is not questioned, but contemporary work seen from the area reveals a lack of understanding by Indian craftsmen of the principles that are the subject of this study.23 This peerless scholar of indigenous art began his investigation with a set of aesthetic principles in hand and, finding them disconfirmed by the testimony and work of contemporary practitioners, naturally concluded that these practitioners were not ‘qualified.’24 Holm’s preface to the 50th anniversary edition goes some way towards explaining this failing. The book has its origins in a research paper written for his MFA on the characteristics of the form-line system in Pacific northwest coast two-dimensional art. When Holm decided to publish it several years later, he realized that he had no documentation; to remedy this deficiency, his former advisor suggested that he record the characteristics of 392 artifacts on Keysort cards.25 Holm also expresses regret for developing a misleading vocabulary which erroneously tied stylistic choices to representational intentions on the artist’s part—so that, for example, his salmon-trout’s head is better thought of as an elaborated inner ovoid, since it does not actually represent a fish’s head. Indeed, Holm confesses that the term comes from Tlingit weavers who, of course, are not members of the Kwakwaka’wakw Nation, let alone the Kwakiutl clan.26 I do not mean to fault Holm’s intentions, which I think were beyond reproach; nor do I mean to take issue with the aesthetic patterns he discerns in Kwakiutl artefacts, which are there to be seen (or not). My point is that even experts are susceptible to conceptual imperialism; guarding against it requires us to pay careful attention to our methods and to the tests we devise for our hypotheses. Part of the problem here is Holm’s insistence on the cultural purity of the practices he aims to study, rather than pinpointing the conventions governing current Kwakiutl artistic practices and situating them relative to those which appear to have governed their past practices, and those of their neighbours. The fact that Holm’s search for informed practitioners came up short should come as no surprise, since he seems to have been looking for his subject in No True Scotland (or, perhaps, the Tlingit Nation).27 Holm is hardly unique in this respect. Larry Shiner tells a similar story, according to which an Alaskan State Arts Council representative finds himself in the position of ‘constantly explain[ing] to Alaskan [First Nations] that ivory carving and beadwork can be supported “as art,” but kayak or harpoon making cannot’.28 If indigenous Alaskans are to be believed, then the close relationship they perceive between their carving and kayak-making ought to prompt the State Arts Council to re-evaluate either their ideas about art’s proper extension, or their designation of carving and beadwork as ‘art’. Even works from the history of our own culture are susceptible to this kind of misrepresentation. As Susan Feagin has observed: ‘Altar-pieces don’t transform spaces the same way when they are hung on walls in museums. Neither do paintings originally produced for chapels in churches and cathedrals when they are also hung in museums.’29 Many of the works from our history, Feagin argues, are site-specific; this means that when galleries, museums, or textbooks mediate our engagement with these works, we experience them as caricatures deprived of their cultural significance and their historical place. This poses a fairly straightforward epistemic problem for appreciation and interpretation. But it can also pose an ontological problem, because it invites us to apply art-kind terms without regard for factors that might well influence their identity and persistence conditions, or their kind-classifications. Just consider Tracey Emin’s groundbreaking Everyone I Have Ever Slept With 1963–1995 (1995), a tent with the appliquéd names of 102 people with whom the artist shared a bed (literally understood) in the designated period. Without access to the salient contextual details, it would be all too easy to mistake Everyone’s vehicular medium for a relatively uninspired work of sculpture, rather than an important piece of conceptual art. But we can also err too far on the side of caution and exaggerate the differences between cultural practices. This is the moral of Denis Dutton’s critique of Lynn M. Hart’s infamous analysis of the jyonti paintings of Uttar Pradesh. Hart argued that the aesthetic principles of jyonti painting are different in kind from those of the Western world because the images and patterns are based on religion, ritual, and myth, and the paintings are produced primarily for use in religious contexts.30 As Dutton observes, however, Hart errs in two respects: first, she ignores the fact that all of the same observations are applicable to much of the Western artistic tradition; second, she assumes that because these works are painted their proper Western analogues are paintings. When we actually compare the practices at issue, however, it becomes clear that the appropriate comparison classes are the domestic and dowry arts.31 According to Sidney Kasfir, this kind of misplaced concern actually betrays a simple double standard: we insist on interpreting the works of other cultures primarily in terms of their political, religious, and social functions, all the while conveniently ignoring the fact that our own artistic practices share parallel histories of political significance, religious patronage, and social utility.32 At its extreme, Shiner observes that this imperialistic tendency can have the absurd result that ‘carvings intended to be Art in our sense, i.e., made to be appreciated solely for their appearance, are called “fakes” and are reduced to the status of mere commercial craft’.33 Similarly, Kasfir notes that ‘when a contemporary [African] carver from another ethnic group (or “tribal style area”) intentionally takes up this same style, the resulting object is said to be a fake because, it is claimed, there is conscious intent to deceive.’34 By insisting on cultural purity or art’s functionlessness we rewrite art history to suit our preconceptions, and in doing so we paint an impoverished picture of the variety of art-kinds. In the absence of such preconceptions, an Aleut iqyax might well introduce a new art-kind (e.g. boat-building) or challenge the properties we ordinarily attribute to an extant one by showing, e.g. that the restriction of ‘sculpture’ to non-functional artifacts obscures a close relationship between the Aleut practices of carving and tool-making, or by introducing sculptural entities—viz. boats—with unusually permissive identity and persistence conditions. But burdened by these false ideas about art’s proper extension, we cannot hope to accurately describe the diverse range of artworld practices. The moral of the story is just this: it does not make much sense to speak of ‘our’ artworld as though the pronoun could do more than limit the range of its temporal application. Today’s artworld is a pan-cultural syncretic amalgam of far too many different conventions, practices, and works. The conscriptivist cannot rely on the categorial intentions of competent users of the relevant terms who ground their reference in ‘our’ artworld because ‘our’ artworld is the world’s artworld. To do so is to risk having those categorial intentions reflect entirely anatopistic criteria, resulting in mistaken classifications and the relegation of new and different art-kinds to inapt old categories. 4. The problem of conceptual inclination At this point, the conscriptivist may well object that the failings I have described are not so much problems with deferring to our concepts as they are failures to defer to accurate concepts or real expertise. According to this line of argument, the flawed uses to which art-kind terms are put by speakers in the throes of conceptual imperialism should disqualify them as competent speakers in the first place. But to insist on regimenting ‘competence’ in this way runs the risk of disqualifying us all as competent users of ‘art’ and art-kind terms. Simply put: we are all susceptible to these basic errors and prejudices, many of which derive from universal and well-documented biases and heuristics such as status quo bias and the availability heuristic.35 If competence requires immunity from error and prejudice, then nobody will turn out to be a competent user of art-kind terms. This is because our concepts of ‘art’ and art-kinds encode and reflect content that is conventionally salient, but not necessarily ontologically complete or even correct. The ubiquity of our testimonial incompetence is captured by what I call the problem of conceptual inclination. The problem here lies with the source of our ontological ascriptions: in deferring to what we say and think, the conscriptivist trusts ontology to the whims of conventionally-reinforced patterns of speech and thought. Before engaging in conceptual analysis, then, the conscriptivist must first establish that the things competent speakers say and think reflect more than mere historical accidents. We can best illustrate the impact of conceptual inclination by considering women’s role in art history, where the structural factors which have erased them and their contributions have resulted in the ongoing cultural transmission of a skewed canon of art. This erasure is not necessarily total, since women and other extra-canonical agents could still exert some influence over the precedents underpinning the conventional attitudes and practices that compose the canon.36 Nevertheless, these lacunae continue to exert considerable force over our artworld concepts today; they reflect historical accidents and conventional patterns of thinking, not just bare ontology. Allow me to explain. Since the example set by Linda Nochlin’s groundbreaking essay ‘Why Have There Been No Great Women Artists?’, feminist scholarship has argued that art history and art theory are fundamentally flawed. This is because they purport to be neutral and universal, but are in fact based on biased and narrowly applicable criteria.37 The result is a narrow conception of artistic value—embodied in the art-historical canon—which in turn circumscribes our concepts and ontologies. From the outset, women’s access to the artworld was constrained by the traditional system of transmitting professions from father to son. In order for a woman to receive even basic artistic training, her father would have had to take an exceptional interest in his daughter’s education. Even so, marriage might easily disrupt a fortunate woman’s career path; this is because until recently women were expected to abandon the amateurish pursuits of youth to concentrate on child-bearing, child-rearing, and household maintenance, or to help their husbands in their business endeavours. While this social attitude is especially well-documented for the nineteenth century, when modest proficiency in several art-forms was considered a sign of a well-educated woman, it is one based on long historical precedent. Aspiring female artists effectively faced a choice between career and matrimony.38 Any woman who successfully negotiated these obstacles faced additional stumbling blocks. Her work might, for instance, be widely attributed to her father, mentor, or other male contemporaries, thereby erasing her from art history (as was the case with, e.g. Artemisia Gentileschi and Marie-Denise Villers). Similarly, female artists were often restricted in the subject matter it was culturally acceptable for them to depict (viz. still lifes and scenes of animals); unfortunately, these subjects were considered the purview of amateurs, not masters. When the academies began to take over artistic education in the seventeenth and eighteenth centuries, female students were barred from attending life drawing classes lest they should chance to see any nudity. Later, when women were allowed to attend such classes, the models had to be partially draped.39 Prevented from developing their skills in the most prized kinds of depiction, all of which involved the human form, women had to seek their subjects elsewhere in the lesser fields of genre painting, portraiture, and still life. Considered by itself, the erasure of some groups’ contributions to the artworld does not yet pose an ontological problem. After all, the fact that paintings by Renaissance women were under-appreciated until recently does not mean that they were perceived as non-artworks. The trouble where ontology is concerned is that the weight of precedent disposes the artworld to preserve and to reproduce existing conventions. Conventions are by their nature conservative, and the art-historical canon gives us a necessarily incomplete picture of artworld practices. This means that the judgement of whether a new practice belongs to an art-kind depends, to a large extent, on whether other practices like it have been so-judged in the past. Consider, for instance, the early efforts to justify photography’s inclusion among the arts, which made the case on the grounds that photography shares its goals and methods with the other pictorial arts and that, contrary to appearances, it requires the skilled manipulation of its vehicular medium.40 These efforts ultimately succeeded in extending the cardinality of art-kinds by one, but the work required to perform this extension also served to legitimate and to reinforce existing criteria for art-kind membership. The result is that perfectly good candidate art-kinds (e.g. calligraphy, gardening, needlework, textile arts, etc.) are left out of the canon entirely, and this absence affects the space of possibilities we are willing to entertain. As Peggy Brand has put it, what we have inherited is ‘an artworld whose conventions have been established and perpetuated by a relatively elite group. ... What has come down to us is an art of exclusion’.41 The relative absence of women and of certain kinds of practices from the canon’s ranks poses a problem for the conscriptivist because it affects the space of possibilities and the set of relevant alternatives which even competent users of ‘art’ and art-kind terms are willing to entertain. The low visibility of some candidate art-kinds such as calligraphy and weaving directly influences their plausibility qua art in the eyes of members of the dominant culture. And that, in turn, affects the set of first-order ontological properties we think determine the boundaries of individual art-kinds. One might think, for example, that film and photography are exceptional among the arts in featuring a mechanical process of reproduction (perhaps resulting in a type ontology); one might even go further and argue that this precludes them from counting as art-kinds in the first place. But to do so would be to ignore the long history of mechanical reproduction in other candidate art-kinds, from the use of moulding in smithing and sculpture to Jacquard punch-cards in weaving, not to mention the art of print-making itself. The art-historical canon occupies an important social role at the heart of the artworld: it is not just a compendium of greatest hits; it also supplies the primary text for the artistic education of a culture (experts and folk alike). To see the extent of the canon’s influence, one need only open any art history textbook: it gives the academic discipline of art history the bulk of its subject matter, kindling student interest in particular artists, styles, traditions, and works, and plays a central role in disseminating the history of art to the broader public (especially through the intercession of critics, galleries, and museums). It furnishes criticism with a reference point for the evaluation of new works, and helps to define the parameters of genre and style. In a word, the art-historical canon selects from among the field of artworks those which are deemed worthy of special attention, and serves the function of turning critical, economic, historical, and even popular interest upon those works and others like them.42 Consequently, the more central a place that canonical value judgements occupy in one’s artistic education, the more salient the ontological properties they encode become for identifying new artistic practices. This is borne out by evidence from social psychology. I have already mentioned Tversky and Kahneman’s work documenting the pervasive influence of the availability heuristic, and Samuelson and Zeckhauser’s work on status quo bias. To this extensive body of evidence we can also add Robert Zajonc’s work, which found that mere exposure to a stimulus is sufficient to enhance a subject’s attitude towards that stimulus.43 A subsequent meta-analysis of 208 studies showed that Zajonc’s exposure effect is both robust and reliable.44 In a more recent follow-up study, James Cutting likewise found that subjects’ preferences for certain artworks is a function of their familiarity with the work in question: aesthetic preference is a function of frequency of appearance, not canonicity, prototypicality, or the subject’s expertise (except insofar as these contribute to the frequency of the subject’s exposure to the work in question).45 The canon’s influence populates the realm of ontological possibilities; it gives us the list of entities whose ontology we are supposed to investigate in the first place. The fact that the canon does not paint a complete picture of the variety of artistic practices therefore poses a significant problem, because it excludes some practices and works which might otherwise prompt us to revise our ontological ascriptions. 5. Conclusion There is no doubt that our art-historical canons presuppose some general understanding of what makes an artefactual practice an art-kind. Likewise, there is no doubt that our use of ‘art’ and art-kind terms presupposes a general understanding of what makes something art in the first place. In posing this trio of problems for conscriptivism, my point has not been to deny these facts, or to suggest that our commitment to descriptive adequacy is hopeless or misguided. My goal has simply been to show that we should not mistake our use of ‘art’ and art-kind terms for a reliable guide to ontology. We cannot simply read a kind’s ontology from the uses to which we put its associated terms; our language reflects historically-situated and conventionally-reinforced preferences, not bare ontology. Human beings share a cognitive apparatus which makes extensive use of common heuristic devices, and these routinely encode systemic biases into the concepts we deploy. Once our conceptual ghosts have been properly exorcised, we can instead lavish our attention on the practices that gave rise to those concepts and intuitions in the first place. Doing so will not guarantee the accuracy of our theorizing—nothing short of discovering a common microstructure will do that—but it does hold out the promise of minimizing errors due to conceptual instability, imperialism, and inclination by making our methods and the sources of our ascriptions explicit. Our concepts, intuitions, and practices are useful starting points for social ontology, but they should not be mistaken for a neutral foundation. Their proper use requires a healthy dose of epistemic humility, and a measure of skepticism. Simply to defer to our concepts, as the conscriptivist would have us do, is to stake everything on what are effectively fake views.46 Footnotes 1 See, for example, Amie Thomasson, ‘The Ontology of Art and Knowledge in Aesthetics’, Journal of Aesthetics and Art Criticism 63 (2005), 221–229, at 223; ‘Debates about the Ontology of Art: What Are We Doing Here?’, Philosophy Compass 1 (2006), 245–255; and Ordinary Objects (New York: OUP, 2007), 189–190, and also Andrew Kania, ‘The Methodology of Musical Ontology Descriptivism and its Implications’, BJA 48 (2008), 426–444. 2 See, for example, Noël Carroll, ‘Historical Narrative and the Philosophy of Art’, Journal of Aesthetics and Art Criticism 51 (1993), 313–326; David Clowney, ‘Definitions of Art and Fine Art’s Historical Origins’, Journal of Aesthetics and Art Criticism 69 (2011), 309–320; Jeffrey Dean, ‘The Nature of Concepts and the Definition of Art’, Journal of Aesthetics and Art Criticism 61 (2003), 29–35; Denis Dutton, ‘Mythologies of Tribal Art’, African Arts 28 (1995), 32–43 and ‘But They Don’t Have Our Concept of Art’, in Noël Carroll (ed.), Theories of Art Today (Madison, WI: University of Wisconsin Press), 217–240; David Novitz, ‘Art by Another Name’, BJA 38 (1998), 19–32; and Richard Wollheim, Art and Its Objects, 2nd edn (Cambridge: Cambridge University Press, 1980). On the art-historical side, see Paul Oskar Kristeller ‘The Modern System of the Arts: A Study in the History of Aesthetics (I)’, Journal of the History of Ideas 12 (1951), 496–527 and ‘The Modern System of the Arts: A Study in the History of Aesthetics (II)’, Journal of the History of Ideas 13 (1952), 17–46; Larry Shiner, The Invention of Art: A Cultural History (Chicago: University of Chicago Press, 2011); and Susan Vogel, Baule: African Art, Western Eyes (New Haven, CT: Yale University Press, 1997). Stephen Davies makes, but does not endorse, similar claims in his ‘Non-Western Art and Art’s Definition’ in Carroll, Theories of Art Today, 217–240. 3 The term comes from Dominic McIver Lopes ‘Art Without ‘Art’’, BJA 47 (2007), 1–5, at 3. An act of Φ-ing is concept-dependent iff Φ-ing requires a concept of Φ. 4 Julian Dodd, ‘Adventures in the Metaontology of Art: Local Descriptivism, Artefacts and Dreamcatchers’, Philosophical Studies 165 (2013), 1047–1068. 5 Kristeller, ‘The Modern System of the Arts: (I)’ and ‘The Modern System of the Arts: (II)’. 6 Kristeller’s references are to Batteux’s Les Beaux arts réduits à un même principe, nouvelle edition (Leiden: Elie Luzac, 1753), 12. 7 Clement Greenberg, for example, thought the artworlds of the seventeenth and eighteenth centuries were dominated by literature—see James I. Porter ‘Is Art Modern? Kristeller’s “Modern System of the Arts” Reconsidered’, BJA 49 (2009), 1–24, at 4. 8 Kristeller, ‘The Modern System: (I)’, 504. The discussion of animal noises in the Cratylus can be found in Plato, ‘Cratylus’, in John M. Cooper and D. S. Hutchinson, (eds.), trans. C. D. C. Reeve Plato: Complete Works (Indianapolis, IN: Hackett Publishing Company, Inc., 140 (423c). 9 James O. Young, ‘The Ancient and Modern System of the Arts’, BJA 55 (2015), 1–17, at 4–6. 10 Young, ‘The Ancient and Modern System of the Arts’, 10. 11 See Porter, ‘Is Art Modern?’, 13 and Young, ‘The Ancient and Modern System of the Arts’, 3. 12 Kristeller contends that it stabilizes after the eighteenth century, which is not quite right, but he may be right that it acquired more stability after that point. 13 Peter Kivy, ‘What Really Happened in the Eighteenth Century: The “Modern System” Re-examined (Again)’, BJA 52 (2012), 61–74, at 70. 14 Porter, ‘Is Art Modern?’, 14 and Kivy, ‘What Really Happened in the Eighteenth Century’, 64 and 73. 15 Lydia Goehr, The Imaginary Museum of Musical Works: An Essay in the Philosophy of Music (Oxford: OUP, 1992), 103. 16 Goehr, The Imaginary Museum of Musical Works, 176. Note that Goehr’s claim is actually that musicians only gained creative freedom at the end of the eighteenth century. Given her subsequent remarks about the freedom performers enjoyed and the underdeveloped nature of scores, however, it seems clear that she actually has composers in mind. 17 Goehr’s defence of each of these properties can be found in The Imaginary Museum of Musical Works, 148–155, 179–202. 18 In ‘The “Great Divide” in Music’, BJA 45 (2005), 175–184, James O. Young adduces evidence for a work-concept as early as the late seventeenth century. I take it that this is compatible with Goehr’s point, which concerns the moment when these emergent ideas finally become dominant and widespread in European culture. The decisive cut-off, on her account, comes with Beethoven and the Romantic era, rather than the Classical era of Haydn and Mozart. For Goehr’s analysis of seventeenth and eighteenth century counterexamples, see The Imaginary Museum of Musical Works, 176–204. 19 Tiffany Stern, Rehearsal from Shakespeare to Sheridan (New York: Oxford University Press, 2000). 20 These observations come primarily from Stern, Rehearsal from Shakespeare to Sheridan 5–8, and 46–123 and 240–290. 21 See, for example, James Hamilton, The Art of Theatre (Chichester, UK: Wiley-Blackwell, 2007) and David Osipovitch, ‘What is a Theatrical Performance?’, Journal of Aesthetics and Art Criticism 64 (2006), 461–470. 22 Michael Baxandall, Patterns of Intention: On the Historical Explanation of Pictures (New Haven, CT: Yale University Press, 1985), 109. For more on the pragmatic value of Baxandall’s heuristic principles and his ‘inferential’ approach to criticism, see David Davies, ‘The Function of Generalization in Art History: Understanding Art across Traditions’, Arts and Literary Studies (2016), 8–19. 23 Bill Holm, Northwest Coast Indian Art: An Analysis of Form (Seattle, WA: University of Washington Press, 1970), vii. Emphasis mine. 24 Ibid. 25 Bill Holm, Northwest Coast Indian Art: An Analysis of Form, 50th Anniversary Edition (Seattle, WA: University of Washington Press, 2014), xxii. 26 Ibid., xxi. It is also worth noting that Holm’s focus is exclusively on the art of the Kwakiutl clan, not (as might be inferred from the title) all of the Nations making up the rest of the Pacific northwest coast, nor even the other sixteen clans making up the Kwakwaka’wakw Nation. 27 Ibid. ‘No True Scotsman’ is the name of an informal fallacy in which one attempts an ad hoc rescue of a generalization which has already been refuted. So, e.g. when the generalization “All Scotsmen can toss a caber” is met with the counterexample “My brother Duncan can’t toss a caber,” the fallacious response would be a rejoinder to the effect that “Fine, but all true Scotsmen can toss a caber.” 28 Larry Shiner, ‘“Primitive Fakes”, “Tourist Art”, and the Ideology of Authenticity’, Journal of Aesthetics and Art Criticism 52 (1994), 225–234, at 225–226. 29 Susan Feagin, ‘Paintings and their Places’, Australasian Journal of Philosophy 73 (1995), 260–268, at 265. 30 Lynn Hart, ‘Three Walls: Regional Aesthetics and the International Art World’, in George E. Marcus, & Fred R. Myers (eds), The Traffic in Culture: Refiguring Art and Anthropology (Berkeley, CA: University of California Press, 1995), 127–150. 31 Dutton, ‘But They Don’t Have Our Concept of Art’, 218–220. 32 Sidney Kasfir, ‘African Art and Authenticity: A Text with a Shadow’, African Arts 25 (1992), 40–53, 96–97 and Shiner, ‘“Primitive Fakes”’, 231. Stephen Davies likewise illustrates the interdependence of Balinese artistic practices and Bali’s tourist industry in ‘Balinese Aesthetics’, Journal of Aesthetics and Art Criticism 65 (2007), 21–29. 33 Shiner, ‘“Primitive Fakes”’, 226–227. 34 Kasfir, ‘African Art and Authenticity’, 45. 35 For the availability heuristic, the locus classicus is Amos Tversky and Daniel Kahneman, ‘Availability: A Heuristic for Judging Frequency and Probability’, Cognitive Psychology 5 (1973), 207–231; for status quo bias, see William Samuelson and Richard Zeckhauser, ‘Staus Quo Bias in Decision-Making’, Journal of Risk and Uncertainty 1 (1988), 7–59. 36 So, for example, women might exert some influence indirectly by means of their amateur productions, expressions of judgements of taste, occasional patronage, and the initiation and reinforcement of their children’s artworld participation. 37 See, for example, Peggy Brand, ‘Glaring Omissions in Traditional Theories of Art’, in Noël Carroll (ed.), Theories of Art Today (Madison, WI: The University of Wisconsin Press, 2000), 175–183, at 184. 38 These obstacles come from Linda Nochlin, ‘Why Have There Been No Great Women Artists?’, in her Women, Art, and Power: and Other Essays (New York: Harper & Row, 1988), 145–178, at 156 and 164–166. 39 Ibid., 158–159. 40 See, for example, Peter Henry Emerson, Naturalistic Photography (London: Creative Camera, 1889), 328 and Alfred Stieglitz, ‘Pictorial Photography’, Scribner’s Magazine (1899), 528–537. 41 Brand, ‘Glaring Omissions in Traditional Theories of Art’, 177. Anneliese Monseré makes much the same point in response to Gaut’s cluster theory of art in her ‘Non-Western Art and the Concept of Art: Can Cluster Theories of Art Account for the Universality of Art?’, Estetika 49 (2012), 148–165, at 159. 42 Hubert Locher, ‘The Idea of the Canon and Canon Formation in Art History’ in Matthew Rampley, Thierry Lenain, Hubert Locher, Andrea Pinotti, Charlotte Schoell-Glass and C. J. M. Zijlmans (eds), Art History and Visual Studies in Europe: Transnational Discourses and National Frameworks, Vol. 4, (Boston, MA: Brill, 2012), 29–40, at 33–34. 43 Robert Zajonc, ‘Attitudinal Effects of Mere Exposure’, Journal of Personality and Social Psychology Monographs 9 (1968), 1–27. Cited in Dominic McIver Lopes, ‘Aesthetics and Philosophy of Art’, in Herman Cappelen, Tamar S. Gendler and John Hawthorne (eds), The Oxford Handbook of Philosophical Methodology (New York: OUP, 2016), 657–670. 44 Bornstein, ‘Exposure and Affect: Overview and Meta-Analysis of Research, 1968–87’, Psychologial Bulletin 106 (1989), 265–289, at 268. A recent study by Aaron Meskin, Mark Phelan, Margaret Moore and Matthew Kieran, however, suggests that mere exposure did not increase liking for bad art; see their ‘Mere Exposure to Bad Art’, BJA 53 (2013), 139–164. 45 James Cutting, ‘Gustave Caillebotte, French Impressionism, and Mere Exposure’, Psychonomic Bulletin and Review 10 (2003), 319–343. Cited in Lopes, ‘Aesthetics and Philosophy of Art’. 46 This research was supported by a grant from the Social Sciences and Humanities Research Council of Canada. I would also like to thank Emily Carson, David Davies, and Sherri Irvin for helpful comments on early drafts © British Society of Aesthetics 2018. Published by Oxford University Press on behalf of the British Society of Aesthetics. All rights reserved. For Permissions, please email: firstname.lastname@example.org This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
The British Journal of Aesthetics – Oxford University Press
Published: Apr 14, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera