Does Art Pluralism Lead to Eliminativism?

I. Introduction

Philosophers of art often pose questions about art as if it were a single, well-defined thing. This presupposition is monism about art. For thinkers who value art, monism can seem like the only option. So, despite the collapse of every proposed definition of ‘art’, they might feel that the only alternative to some definition is to abandon art to the abyss. In an earlier paper, we argued that pluralism offers an alternative to monism and utter nihilism. As we put it,

Christopher Bartel and Jack Kwong argue that this middle way is closed.² Although they endorse all the reasons for pluralism, they argue that these reasons should lead us to eliminate art from our conceptual repertoire and ‘art’ from our vocabulary.

In Section II, we introduce two kinds of pluralism about art. In Section III, we rebut the objections that Bartel and Kwong pose to pluralism. In Section IV, we turn to our title question and ask whether pluralism should lead us to abandon art talk altogether. Our answer is maybe. You can keep the word ‘art’ if you like, but only in a loose, everyday sense. It does not do substantive philosophical work.

II. Two Kinds of Pluralism

The core idea of pluralism is that there is no single definition or way of thinking about art that serves all the purposes that philosophers and other scholars have. Instead, there are a number of different conceptions or more refined categories. Each of these is useful in some contexts and for some enquiries, but none of them is useful in all contexts or for all enquiries.

We distinguish art concepts in terms of different functions and activities. Aesthetic art applies to ‘artifacts satisfying some aesthetic function; for example, affording some aesthetic attitude, experience, interest, value’. Thinking in these terms supports enquiry into perception and the value of art experience. Communicative art applies to ‘artifacts that are […] vehicles for the communication of certain contents; for example, representational, semantic, or expressive content’. Thinking in these terms is useful for some cultural and cognitive enquiries, and it reveals different values of art experience. Conventional art applies to ‘artifacts recognized, accepted, targeted, governed by artworld conventions, institutions, and practices’. Thinking in these terms is useful for curation, as well as thinking about legal and economic issues. Historical art applies to ‘artifacts emerging from, belonging to, embedded in, art-historical traditions or narratives’ (ACP, pp. 91–92). Thinking in these terms is useful for historical enquiry and anthropology.

We do not mean this list to be definitive. One might add skill art, which applies to artifacts produced by subtle, non-algorithmic skill. This would capture art objects as products of artistry, practices that (as one says) are more art than science. Alternately, one might object to one of the concepts on the list. Regardless, the exact list is not the point. Rather, pluralism just requires that there are multiple concepts that capture different aspects of art talk such that each is useful for some enquiries but none of them suffices for all enquiries.³

It is no mistake that each of the art concepts corresponds to a class of would-be definitions of art. For example, conventional art is the category identified by institutional theories of art. Many of the objections to institutional theories, which make problems for them as definitions, are no problem for conventional art as its own category. Rather, they underscore the fact that the conventional art concept is not useful for every purpose or every enquiry. A pluralist is free to think in terms of conventional art in contexts where institutional theories are apt.

There is another form of pluralism that starts by noting that different art forms vary in important ways. As Peter Kivy comments, in theories of art ‘the overriding concern was, and continues to be, the search for sameness; and that search blinded the philosophical community to a bevy of questions […] involving the arts not in their sameness but in their particularity’.⁴ Theories of art constructed with visual arts like painting in mind were difficult to apply to absolute music, and theories that made sense of music were hard to extend to the visual arts. Whereas Kivy sees the enquiry into differences as a parallel project – one that has no consequences for definitions of art – Dominic McIver Lopes argues that there is no substantive theory to be given of art altogether.⁵ Instead, all of the substantive work is done by accounts of the separate arts: painting, sculpture, absolute music, and so on. In contrast to our art concept pluralism, let’s call this art media pluralism.⁶

Note that the arguments for art concept pluralism and art media pluralism are compatible. Although reiterating those arguments is beyond the scope of this paper, we are now willing to accept both. According to this double pluralism, ‘art’ talk is rather loose. In every enquiry, the substantive context determines both the more specific concept at work and some specific art forms.⁷ Our understanding of the individual arts will depend on our implicit art concept, and the details of each art concept may vary depending on which arts are under consideration.⁸

III. Some Objections to Pluralism

Bartel and Kwong argue that pluralism of either form ultimately fails. To start with art concept pluralism, they demand to know what holds the various art concepts together. Why is aesthetic art an art concept rather than just a free-standing concept that applies to objects of aesthetic appreciation? They write that, ‘for art concept pluralism to be successful, we would need some explanation of what makes any proposed concept count as an art concept’ (PED, p. 100).

Our earlier work offers this response: ‘The constraints of art-historical or art-critical inquiry constrain what can count as an art concept’ (ACP, p. 95). Yet one can then push the challenge back to the level of enquiries: what makes some history art-historical or some criticism art-critical? There is no substantive insight from adding that they are history and criticism that we talk about using the word ‘art’. Yet this does explain the association in a weak sense. All the refined art concepts were developed, as one might say, in the neighbourhood of ‘art’ talk. This is underscored by the fact that, although the concepts differ in application to some edge cases, their extensions are largely overlapping.

Bartel and Kwong object that this overlap presents the pluralist with a dilemma.⁹ Either this area of overlap can be identified in some independent way or there is no deep connection between the so-called art concepts after all.

Consider the first horn of the dilemma, that the overlap is identified as art independently of the separate aspects. As Bartel and Kwong explain, this ‘seems to indicate that the concept art in the unqualified sense is already well understood’ (PED, p. 104). Saying that there is a separate, well-understood conception of art would be a retreat to monism.

So take the second horn of the dilemma, that there is no absolute consideration that makes the intersection of the various art concepts important. Bartel and Kwong write that ‘if some other works had been treated as canonical, other concepts for art would have been more appropriate’ (PED, p. 104). As pluralists we accept this point, but we see no harm in doing so. There is a history to the use of the word ‘art’, a causal story about why contemporary English speakers apply it to canonical artworks. Likewise for cognate words in other languages. Given that prior usage, each of the art concepts fulfils purposes that enquirers sometimes have when talking about those things. This causal history does not indicate that there is some single conception of art corresponding to our loose use of the word, and if history and usage had been different then our talk might not have picked out our current art concepts.

Of course, the developed art concepts were not simply already implicit in earlier loose talk about art. Instead, they are more specific and refined language developed to serve some of the purposes of the original imprecise term. They are what Rudolf Carnap called explications.¹⁰ When philosophers aimed to develop monist art theories, they articulated conceptions that failed as analytic definitions of ‘art’ but which nevertheless supported rewarding enquiries into some aspects of the objects and practices that elicited ‘art’ talk. In more recent discussions, explication is sometimes called conceptual engineering. Instead of trying to clearly express concepts we already have, the conceptual engineer constructs new concepts that can more successfully serve some of our purposes. Each of the elaborated art concepts is an engineered concept that is helpful for enquiries into some questions that, historically, were posed in terms of art.

Bartel and Kwong see the second horn of the dilemma as more problematic because they want more from an account of art than pluralism can deliver. They write that a ‘plausible account of art (and the art concept) […] would […] allow us to make substantive generalizations about (for example) the history and sociology of artistic practice, aesthetic appreciation, and evaluative discourse’ (PED, p. 104). Schematically, their demand is that philosophers should produce law-like generalizations of the form All art is G. We are unsure whether substantive generalizations should be demanded from philosophy of art – as opposed to a rich exploration of details and differences – but the connection to Carnap may be informative here. On Carnap’s account, generalizing is central to the practice of explication and conceptual engineering. One of the tests of an explication, he thought, was how many law-like generalizations it made possible. Even so, a crucial feature of a more precise, engineered concept is that typically it will not apply to exactly the same things as the original, imprecise concept. For a term that serves many different purposes, no single explication will serve all the purposes that the original term did. So there will not be law-like generalizations ranging over the whole extension of the original term, but instead just narrower generalizations ranging over the extensions of the various explicata. So, even if we demand generalizations, we should not demand ones that apply to All art…

For the pluralist, each of the art concepts does some but not all of the work that enquirers set out to do when they start thinking about art. Rigorous generalizations, insofar as they can be discovered, will be in terms of the specific concepts rather than in terms of art simpliciter. So the pluralist should reject Bartel and Kwong’s insistence that philosophy of art should deliver conclusions about all art.

For media pluralism, Bartel and Kwong pose a similar dilemma. Either there is some central reason why all the art forms should count as arts or it is largely just a historical accident that they are grouped together. Again, the pluralist is free to accept the second horn of the dilemma. There may be internal similarities that explain why visual arts like painting and sculpture are grouped together, but there is no strong analogy that captures all and only the arts. There are also historical reasons, when considering particular periods when painters and musicians influenced one another, to consider music and painting together – but again these connections will not bind together all and only the arts. These different kinds of connection are especially salient when we combine media pluralism with concept pluralism. The internal similarities are aesthetic or perhaps communicative, while the interaction in particular periods is historical.

One difference for media pluralism is that the particular art forms did not develop as explications for the imprecise concept art. Concepts of painting, sculpture, and other media predated the modern conception of fine art. Nevertheless, this provides no reason to accept Bartel and Kwong’s demand that philosophy of art should allow us to make generalizations about all art, rather than just about particular media.

IV. Should We Eliminate Art Talk?

Where does this leave us, as pluralists who accept that there can be no philosophical analysis of art as such? Removing ‘art’ from our lexicons would admittedly require some effort. Arguably, it would also incur some cost.

For their part, Bartel and Kwong argue that the cost would be negligible. Like us, they recognize that there is no satisfactory analysis of art. However, they argue that there is a central concept that groups all of the art concepts together: each corresponds to a different way of valuing objects. They summarize their view as the ‘proposal […] that there is a plurality of substantive generalizations that one can make about the ways in which objects can be valued, and that the recognition of this point would eliminate the art concept as well as the need to define art’ (PED, p. 111). We find this proposal unpersuasive. Ways of valuing is at once too broad and too narrow to do all the philosophical work that shelters under the umbrella of art. First, not all ways of valuing correspond to an art concept. For example, collectors paid a great deal of money for Bored Ape NFTs, but surely that is no answer to someone wondering whether they are or were art. One could say that this financial valuing is different from art-related valuing, but such a move is not available to Bartel and Kwong. Just as they deny that there is any way to identify which concepts are art concepts, they must also deny that there is any way to identify which valuing is art valuing. Second, an art concept need not correspond to any way of valuing at all. For example, we can imagine a community that uses artifacts that serve an expressive function without placing any special value on those things. Although a communicative art concept would apply to these artifacts, they would not be valued as such. So the pluralists’ attention to concepts and media allows us to track distinctions that would be lost if we had only Bartel and Kwong’s general laws about ways of valuing.

Annelies Monseré, responding to Lopes’s account, argues that ‘“art” functions as a framework concept’ in a way that makes it indispensable.¹¹ In a short span, she lists several such functions. For example, she argues that ‘it is hard to convey cultural studies, art education, and art practice that do not refer to “art”’.¹² The idea is that critical and scholarly discourse uses the word ‘art’ and that it would be difficult to accomplish the same communicative ends without the word. Perhaps this is so. However, scholars may just be using ‘art’ in the loose, everyday way. Bartel and Kwong contend that ‘a closer look almost always reveals that the scholars and practitioners who operate within these areas are concerned with different classes of objects’ (PED, p. 110). Here they are simply disagreeing with Monseré on the facts. Without a more detailed examination of practice, it is unclear how much scholarly work about art relies on undifferentiated ‘art’ talk.¹³

Furthermore, Monseré notes that the word ‘art’ appears in official laws and policies. She writes that ‘[t]he concept also has legal implications: consider, for example, art vandalism, plagiarism, and copyright’.¹⁴ However, this shows at most that laws and policies have been written with a presumption of art monism. If this is true and there were a definition of art to be had, philosophers could do a service by articulating it. Given the fact of pluralism, though, the laws and policies cannot be saved by careful philosophy. Either legal invocations of art talk mean ‘art’ in some special legal sense or they are just as loose and imprecise as our everyday talk. So, unlike Monseré, we do not think that scholarly discourse, norms, and policies provide a compelling reason to retain ‘art’ talk.

This brings us back to our title question: As pluralists, should we purge the term ‘art’ from our discourse?

Maybe.

This answer of maybe is not, or at least not primarily, an expression of uncertainty. Rather, the point is that we are faced with a choice. Pluralism does not force us to abandon the word. Perhaps the casual, conversational use is valuable in our culture or in the culture we would like to have. The upshot of pluralism is that we cannot expect ‘art’ to carry any weight in rigorous philosophical debate. If we ask whether some object is or is not art, there is not a single non-figurative, precise meaning that makes it yes or no.¹⁵ Art can be the beginning of a philosophical conversation, but it cannot be the end of it.

Competing Interests

The authors have no competing interests to declare.