Realism and Anti-Realism

Stuart Brock, Edwin Mares

• Publisher: Routledge
• Year: 2014
• eBook ISBN:
  9781317494263
• Topic: Philosophy
• Subtopic: Epistemology in Philosophy

There are a bewildering variety of ways the terms "realism" and "anti-realism" have been used in philosophy and furthermore the different uses of these terms are only loosely connected with one another. Rather than give a piecemeal map of this very diverse landscape, the authors focus on what they see as the core concept: realism about a particular domain is the view that there are facts or entities distinctive of that domain, and their existence and nature is in some important sense objective and mind-independent. The authors carefully set out and explain the different realist and anti-realist positions and arguments that occur in five key domains: science, ethics, mathematics, modality and fictional objects. For each area the authors examine the various styles of argument in support of and against realism and anti-realism, show how these different positions and arguments arise in very different domains, evaluate their success within these fields, and draw general conclusions about these assorted strategies. Error theory, fictionalism, non-cognitivism, relativism and response-dependence are taken as the most important positions in opposition to the realist and these are explored in depth. Suitable for advanced level undergraduates, the book offers readers a clear introduction to a subject central to much contemporary work in metaphysics, epistemology and philosophy of language.

1	Introduction

Questions about the ultimate reality of things sometimes seem like silly questions to non-philosophers, but to philosophers they are questions of the utmost importance and deepest significance. It is not easy or straightforward to know when some contentious realm of entities is real, or to understand and appreciate what is at issue between those on each side of the dispute. Thus, the questions posed by those who originally framed the realist–anti-realist debate centuries ago – most notably, the nominalists and idealists – have dominated the attention of philosophers ever since. Indeed, questions about the plausibility and character of realism and its alternatives are at the heart of all metaphysical disputes today.

Our aim in this book is to make clear to any intelligent reader what is really at stake in the contemporary realism debate. As far as we know, no other work has really succeeded in doing that. This is unfortunate, because the primary literature is notoriously difficult to grapple with. Consequently, anyone confronting this literature for the first time is likely to be deterred by the abundance of jargon and the almost universal presupposition that the reader has significant background knowledge of the main issues. The uninitiated could be forgiven for giving up quickly after forming the impression that they are somehow uninvited guests at an exclusive club. It is hoped this book might succeed in making the club a little more ecumenical. This book is for those who want to direct their attention squarely towards questions of realism and anti-realism for the first time. It is meant as a substantial introduction to the realism debate, and a good springboard to the most important technical papers on the topic. With this goal in mind, let us move directly to our main question.

1.1 What is realism?

Perhaps the most striking feature of the realism literature is its lack of agreement over what the basic terms “realism” and “anti-realism” mean. Let us start therefore by stipulating what we mean by “realism”. Realism about a particular domain is the conjunction of the following two theses: (i) there are facts or entities distinctive of that domain, and (ii) their existence and nature is in some important sense objective and mind-independent. Let us call the first thesis the “existence thesis” and the second thesis the “independence thesis”.

Some further elaboration is necessary. Notice that our characterization of the existence thesis involves quantification over entities and facts. (There is a difference: entities are the referents of the singular terms in a language, while facts are aspects of the world represented by whole declarative sentences in a language.) This is awkward, to say the least. And this awkwardness explains why many who have attempted to explain the existence thesis – and distinguish it from the independence thesis – have quantified only over individuals. But to ignore the realist’s commitment to facts is to miss something very important about the realist’s distinctive claim. For it is possible to be a realist about some domain without thinking that there are any special entities distinctive of that domain. And the facts in question are not just trivial negative facts to the effect that such things do not exist. One might believe it is a fact that everything that goes up must come down without believing in gravitational fields; one might believe it is a fact that, in the novel Wuthering Heights, Heathcliff is haunted by a ghost on the windy moors without believing in Heathcliff or ghosts; one might believe it is a fact that Senator John Kerry could have won the US presidential election without believing there is a possible world in which (something appropriately related to) Kerry does win; one might believe it is a fact that one ought to assist people in need without believing that such actions have the property of being morally good. In our view, those who believe such things (without believing that such facts depend on us) count as realists about the relevant domain – scientific realists, realists about fiction, modal realists and moral realists – but they do not count as realists about scientific entities, fictional objects, possible worlds or moral properties.

Notice also that our brief description of the independence thesis alludes to an important sense in which a domain is objective and mind-independent. But what sense is the important sense? The sense-data of the phenomenalists, the ideas of the idealists, the monads of the Leibnizian, the phenomenal world of the Kantian and the verifiable world of the logical positivists have all been alleged to exist. But such entities are all, to a greater or lesser extent, subjective and mind-dependent, and so lack reality in our fundamental sense. For a domain to be objective and mind-independent in our sense it must not exist in any of these ways. Facts about the domain must be out there to be discovered rather than constructed; entities within the domain will be detected rather than constituted by our mental states. This is, very roughly, what it means for a realm of facts and things to exist objectively and mind independently.

We started this section by noting that there is a lack of consensus about the right characterization of realism. Consequently, the reader should be unsurprised to discover that many philosophers have qualms about our particular account. Nonetheless, our understanding of the notion has a fine pedigree, and we think every philosopher working in the field would agree that ours is at least one way legitimately to unpack the notion, even if they disagree about whether it is the best way. To be sure, consider what others have said about the issue.

Realism [is] a claim about what entities exist and a claim about their independent nature.

(Devitt [1984] 1991: 14)

To assert that something is somehow mind-independent is to move in the realist direction; to deny it is to move in the opposite direction … Many philosophical questions have the following general form: Is such-and-such mind-independent in so-and-so way? Given specifications of such-and-such and so-and-so, one may call someone who answers “Yes” a realist.

(Williamson 1995: 746)

There are two general aspects to realism, illustrated by looking at realism about the everyday world of macroscopic objects and properties. 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)

So our construal of realism represents one well-established tradition. But our characterization was chosen not only because of its lineage, but also because we believe it is the most useful way to represent realism and anti-realism. It unifies realist debates across different domains, and allows us to represent clearly contemporary positions commonly held to be opposed to realism. To illustrate, consider Figure 1.1, showing the various salient positions, each of which will be explained in the chapters to come.

Figure 1.1

Realism is the shaded region in the first quadrant, the upper-right-hand region bounded by the existence and independence axes. Anti-realist positions – positions opposed to realism – include error theory, fictionalism, instrumentalism, non-factualism, idealism, verificationism, social constructivism and response-dependent theories. Each anti-realist position is located in either the second or fourth quadrants. As Figure 1.1 makes clear, idealism, verificationism, social constructivism and response-dependence on the one hand, and the other forms of anti-realism on the other, have less in common with one another than they do with the realist theories they oppose. According to idealism, verificationism, social constructivists and response-dependence theorists, there are positive facts about the relevant domain; individuals of the relevant kind exist, but they are not objective – they depend on us. According to other forms of anti-realism, however, there are no individuals or facts of the relevant kind, and their non-existence does not depend on us; they would still fail to exist even if there were no minds at all. Finally, it is important to note that there is no territory to be staked out in the third quadrant. Why? The existence thesis and the mind-independence thesis are not independent of one another. In order for an individual or fact to be dependent on our minds, it must exist; mind-dependence therefore guarantees existence.

Of course, this characterization of realism and its alternatives is more than just a little metaphorical. We shall attempt to make precise our understanding of the two theses that constitute realism – and evaluate them – in Part I. Before we do that, however, it is important to contrast our characterization of realism with some alternatives.

1.2 Supplementing and modifying the canonical characterization

Some philosophers will take issue with our account of realism (and, conversely, anti-realism) on the grounds that it is either incomplete or mischaracterized. Realism, as they conceive of it, is not just a claim about the existence and independence of a certain sort of fact or entity. The account, it is maintained, must at best be expanded and supplemented to include further theses, or at worst be completely recast.

Sometimes it is said that realism is partly an issue about our potential for knowledge. This epistemic constraint on our account of realism might be spelt out in a number of different, non-equivalent, ways. For example, a number of philosophers suggest that the following thesis should be a component of any satisfactory elucidation of realism.

• The confidence thesis. Although it may in some circumstances be difficult, we are always capable of coming to know about the existence and nature of the domain we are realists about. That domain is epistemically accessible to us (cf. Wright 1992: 2; Haldane & Wright 1993: 4).

This thesis sets the realist in direct opposition to the sceptic. Realism, on this conception, is the optimistic alternative to a sceptical position. More commonly, though, realism is given a pessimistic epistemic gloss. Jack Smart, David Papineau, Philip Pettit and Amie Thomasson (among others) thus partially characterize realism by way of the following thesis.

• The insecurity thesis. It is possible to be in ignorance or error about the domain we are realists about. In order to avoid such mistakes, one must make appropriate contact with the domain in question, and there is no guarantee that anyone will succeed in doing that.

The confidence and insecurity theses are not, despite initial appearances, contrary to one another (indeed, many philosophers embrace them both), for while the insecurity thesis may provide succour to the sceptic, it does not automatically pave the way to scepticism. It cannot do so without also denying the confidence thesis.

Nothing in our characterization of realism is inconsistent with these claims. We believe each thesis is plausible in its own right. And often anti-realist arguments will assume that the realist is committed to one or other of the two claims. Nonetheless, philosophers inclined to supplement the account with either of these theses will find the map of the terrain provided in this book frustratingly incomplete. Some explanation for this lacuna is therefore necessary.

While introducing an epistemic component to an account of realism is not uncommon, it is still controversial (cf. Lewis 1984; Devitt 1991). Reasons to avoid including such an element in our characterization of realism are not hard to find. If we could know incorrigibly and infallibly of our own existence, as Descartes supposed we could, would this undermine our very reality? That sounds silly, but the insecurity thesis seems to entail as much. If there were objects existing outside the light cone – objects that cannot be observed, and hence cannot be known about – would this undermine their reality? Again, that sounds silly, but the confidence thesis seems to entail as much. Another reason for the controversy is that these epistemic theses do not fit well with the independence thesis. For the confidence and insecurity theses make claims not just about an independent reality, but also about our minds. The worse our mental capacities become – the closer our minds approximate that of a scallop – the less plausible the confidence thesis becomes. And the better our mental capacities become – the closer our minds approximate that of an omniscient being – the less plausible the insecurity thesis becomes. For this reason, we thought it wise to dodge trouble by not including these tenets in the canonical account of the realist–anti-realist divide. For those readers who seek more, we hope that this partial map still proves useful.

Another common amendment to the canonical characterization is to scrap completely the existence and independence theses in favour of a semantic thesis (or cluster of theses). Such a move became popular in the twentieth century, when analytic philosophy took a linguistic turn. This was not so much because it was thought that the two canonical theses were incorrect, but rather because it was felt that providing an emphasis on truth (and meaning) would be a more illuminating way of understanding the dispute between realists and anti-realists.

Philosophers such as Michael Dummett who attempt such a re-characterization of the traditional divide, suggest that realists about a domain will claim that distinctive statements about that domain are either true or false, independent of our means for coming to know which. This way of marking the boundaries of any realist–anti-realist dispute suggests that realists and anti-realists have different conceptions of truth itself. Realists, for example, tend to embrace a correspondence theory of truth, according to which a proposition is true if and only if it corresponds to the facts. Anti-realists, on the other hand, are thought to hold a contrary theory: perhaps a theory that equates truth with warranted assertability, practical utility or something else altogether.

This contemporary reconstruction of the debate is very interesting indeed, and we take some time to consider its merits in Chapters 4–6. Whatever its merits, though, we resist giving any sort of semantic spin on our official account of realism and anti-realism for two reasons. First, to do so would be to prejudice the issue. The semantic gloss is at least as controversial as the epistemic glosses considered earlier. While contemporary philosophers such as Thomas Kuhn, Hilary Putnam, Dummett and Crispin Wright all give an account with a semantic spin, other equally important figures working in the area, such as David Lewis, Michael Devitt and Paul Horwich, vigorously deny any connection between the problem of realism and the problems of truth and meaning. Secondly, we think the questions “How should a theory of truth affect one’s stand on the problem of realism?” and “How should one’s stance on the realism issue affect ones accepted theory of truth?” are very interesting questions indeed. Answers to these question are not, and should not, be trivial. But they would be – or at least would be more likely to be – if we insisted on defining realism in terms of a theory of truth.¹

1.3 Looking ahead

This book is split into two parts. Part I is an examination of the realist and anti-realist debate abstracted away from any particular application of it. In Chapters 2 and 3 we explain in more detail local realism and anti-realism and look at the motivations that might support one position over the other with regard to a particular subject matter. In Chapter 2 we look specifically at the existence thesis and in Chapter 3 we consider the independence thesis, ex...

Table of contents

• Cover Page
• Half Title page
• Title Page
• Copyright Page
• Contents
• Acknowledgements
• Introduction
• Part I
• Local realism and anti-realism: the existence axis
• Local realism and anti-realism: the independence axis
• Idealism
• Kantianism
• Verificationism
• Part II
• Colour
• Morality
• Science
• Mathematics
• Possible worlds
• Fictional characters
• Notes
• Bibliography
• Index