Realism is here meant to be a thesis about (some portion of) the world and our sensitivity to it, as opposed to a thesis about language and our use of it.¹ In general, realism is the claim that something exists objectively, that is, its being is independent of what particular people think. Whether cats, daisies, volcanoes, dinosaurs, centaurs, colours, justice, beauty, or any supreme being exist is, according to realism, independent of what people think, that is, the realist admits the possibility that, always or at some point in time, we may all be wrong about what exists. Anti-realism about an alleged entity is the rejection of realism about that entity.

Given that the realism/anti-realism debate has been understood in various ways, it is important to first clarify, as possible, what will be at issue here. My aim in this chapter is to arrive at a characterization of the debate that both realists and their opponents can accept. By doing so I also aim to dismiss scepticism concerning the very significance and worth of the debate between aesthetic realism and anti-realism.² I begin by briefly characterizing the general debate concerning common-sense realism about the external world, in order to see what light this can throw on the local debate in aesthetics and the philosophy of art (including the philosophy of literature) with which this study is concerned.

Realism is, under one traditional conception, contrasted with idealism, the doctrine that reality is fundamentally mental.³ According to realism, the world is fundamentally non-mental, that is, its features are not the result of anyone’s imagination, except, of course, in the unproblematic sense that created (imagined) items, such as pieces of music, paintings, works of literature or artefacts like hammers have features that their creators imagined, intended to produce, and successfully produced. My aim in this book is to defend the view that aesthetic reality is objective (and non-mental) in the sense that is it is independent of particular aesthetic attributions or judgements. Here are some quotations from contemporary discussions on general realism ⁴:

Realism [is] a claim about what entities exist and a claim about their independent nature. (Devitt 1984, 14)

The leading idea is that the world consists of objects whose existence, nature, and relations are fixed independently of what we happen to think, feel, or desire. (Loux 2002, 252)

There are two general aspects of realism […]. First, there is a claim about existence. Tables, rocks, the moon, and so on, all exist, as do the following facts: the table’s being square, the rock’s being made of granite, and the moon’s being spherical and yellow. The second aspect of realism about the everyday world of macroscopic objects and their properties concerns independence. The fact that the moon exists and is spherical is independent of anything anyone happens to say or think about the matter. (Miller 2002, 1)

To sum up: realism is the thesis that a certain portion of reality is objective, that is, it exists and its existence is independent of what we might think, feel, be able to grasp, or wish. No amount of thought or feeling (let alone any wishful thinking) on the part of a subject will make or change the way such reality is. And reality (such as aesthetic reality) transcends our knowledge, that is, it is possible, and even likely, that we do not know everything about it.⁵ The interest of this general debate for aesthetics (and for the philosophy of art⁶) should be clear: whether or not we take the world to contain, say, aesthetic properties or values will have a bearing on what works of art are. A work of art may not simply be identified with a physical object (say, a painted canvas), but with an object with aesthetic properties and aesthetic value as well, from which it cannot be separated without losing his nature as an aesthetically valuable object. If the properties and value an object is said to possess exist objectively, then any description of the object in question which does not consider these properties and this value will be inappropriate or at least incomplete. Describing a beautiful painting as a painted canvas does not capture its essence as an aesthetic (artistic) object, its nature as a work of art.

Under another (not incompatible) traditional conception, realism is a claim about objective truth.⁷ As van Inwagen notes (1993, 56), an area of thought and discourse is objective in this sense when the truth of our beliefs and assertions depend not on our thoughts, feelings, and wishes but on their objects, on the things our beliefs and assertions are about or represent.⁸ One image sometimes invoked to pick out this relation between our beliefs and assertions and what they are about is that of a map.⁹ A map is supposed to ‘get the territory right’¹⁰ (in other words, it is to represent the territory), and so, the realist claims, do our beliefs and assertions. Realism thus is the claim that a way of thinking and speaking is representational in this sense: it aims to be an accurate and reliable map of an area of reality. It not only claims that the world is independent of the human mind, but it also takes the world to function as a standard for the truth of our beliefs and assertions.¹¹ Furthermore, realism claims that reality (or the world) might transcend, and that it plausibly sometimes does transcend, our ‘best efforts’ to know it (Loux 2002, 253). For the realist truth is ‘epistemically unconstrained correspondence’ (Loux 2002, 258), a conception which implies not only that truth is a matter of fit (2002, 252) with respect to reality, but also that reality is independent of our best tools for knowing and of our best knowledge (2002, 253). Even if we know very little, or nothing, about an area of reality, that has no bearing on the objective existence of that portion of reality.

In twentieth-century analytic philosophy, this metaphysical realism has been construed in a novel way via a debate in the philosophy of language in which Dummett’s work has played a prominent role.¹² For Dummett metaphysical questions are, at bottom, semantic questions, that is, questions about language and meaning.¹³ As M. J. Loux (2002, 258) explains, Dummett rejects the claim that meaning concerns a relation of correspondence ‘between statements and mind-independent states of affairs’, claiming instead that meaning is best conceived of as an epistemic notion, that is to say, the meaning of a statement depends on what counts as evidence for that statement. Therefore, according to this view, truth cannot, contrary to what the realist maintains, be independent of, and plausibly sometimes transcend, the possibility of knowing it. As Loux (2002, 259) notes, for Dummett truth is ‘warranted or justified assertability’ (i.e., what can warrantedly be said), not correspondence with a mind-independent reality. By contrast, as we saw, for the realist statements can be true even when it is impossible to know whether they are true. Whether or not the ancient Greek playwright Aeschylus was killed by a tortoise dropped on his head by an eagle is, for the realist, true or false independently of the evidence or justification we might have concerning what really happened. In other words, a statement is true (or false) independently of verification. This is the thesis I wish to defend for the aesthetic case.

Whether or not the realism/anti-realism debate can adequately be approached via a debate in the philosophy of language will not concern me here. I will also try to remain neutral on whether the realism debate can be reduced to a debate about objective truth.¹⁴ My interest is not in any question which is, at least strickly speaking, semantic, such as the question of whether a disputed class of statements is genuinely assertoric.¹⁵ Rather, my interest is in an ontological question¹⁶: whether a certain area of thought and discourse relates to a domain that exists objectively, that is, independently (in a sense to be explained) of human minds. Before addressing the ontological question, however, we should consider and evaluate one specific semantic approach to the debate concerning realism and anti-realism which has brought a new focus to the ontological debate (§1.2). And we should contemplate also another option, which consists in the very rejection of the ontological debate: ‘quietism’ (§1.3).

Crispin Wright (1992, 1996a, b) has proposed that the realism/anti-realism debate be construed, and conducted, in a new way. In particular, Wright has proposed a minimalist (‘non-metaphysically committed’) conception of truth-aptitude and of truth, coupled with a pluralistic account of the concept of truth, so that the realism/anti-realism debate be carried out ‘by examining the substance’, or ‘metaphysical weight’, of the truth-predicate in each disputed area of discourse. So his method begins by analysing language and concluding about reality. Wright’s proposal is that just because in some area of thought and discourse there is the possibility of truth and falsehood (‘the sky is normally not yellow’) that is not sufficient to ‘secure’ realism about that area. The fact that the sky is not yellow normally does not secure realism about colour. (I dispute this.) For Wright, other ‘realism-relevant cruces’ in each area must be considered and the realism/anti-realism debate is significant only beyond discussions about truth.

This proposal has an important consequence for realism in the aesthetic realm, namely that aesthetic realism is not secured by truth-aptitude or truth alone. That is, even if aesthetic statements are shown to be genuine assertions, and even if some of them are shown to be true, that is not sufficient to secure aesthetic realism.¹⁷ So it is important to give some attention to Wright’s proposal. I will summarize Wright’s project and address, in particular, the realism-relevant constraint of ‘cognitive command’, and consider its relevance to the defence of aesthetic realism in particular.¹⁸

Wright’s main motivation for proposing a ‘shift’ in the realism/anti-realism debate is dissatisfaction with the available anti-realisms, in particular expressivism and error theory.¹⁹ Two theses are central to the approach Wright recommends. One is minimalism about truth and truth-aptitude. As Wright explains the thesis,

it is necessary and sufficient, in order for a predicate to qualify as a truth-predicate, that it satisfy each of a basic set of platitudes about truth: the platitudes, for instance, that to assert a statement is to present it as true; that ‘S’ is true if and only if S (the Disquotational Scheme); that statements which are apt for truth have negations which are likewise; that truth is one thing, justification is another; that to be true is to correspond to the facts; and so on. (Wright 1996a, 864)

But these minimal ‘platitudes’, Wright claims, ...