INTRODUCTION

The central aim of this essay is to set out and defend the idea that indexicality may be a feature not only of referring expressions but also of predicate expressions. Section 2 examines the notions of reference, predication and indexicality, and explains how they will be understood in what follows. Section 3 asks, on this basis, what an indexical predicate or predicate component would look like. It also asks whether we can identify any in English and suggests that we can find at least one specimen. Section 4 argues that the phenomenon is more widespread than might at first appear. Finally, Section 5 considers ramifications of the idea, among them some of the uses to which indexical predication may be put and some of the reasons why it may be important for us. I hint at the end of this section at what seems to me potentially one of the most interesting applications, namely in giving an account of the means by which we represent items which are themselves representational. So, I suggest, the idea of indexical predication holds out the possibility of illuminating the structure of indirect speech reports and of our thoughts about others' thoughts. But full development of these ideas lies beyond the scope of this essay.

I do not seek to provide at any point a full formal treatment of the idea of indexical predication.

INTRODUCTION

This essay applies to the elucidation of indirect discourse an idea which has already been defended elsewhere: that of indexical predication (see Essay 9). The resulting view has, I suggest, the advantages of Davidson's proposal in “On Saying That” without its disadvantages (Davidson 1984: essay 7, first published 1968). Indeed, if Davidson had revised this earlier proposal about indirect speech in the light of the ideas sketched in “Quotation”, the upshot would resemble what is suggested here (Davidson 1984: essay 6, first published 1979; Rumfitt 1993: 431–432). Davidson's earlier proposal is that the ‘that’ of indirect discourse refers indexically to the particular utterance which follows it. The alternative view to be defended here is that a that-clause refers indexically to some non-particular item of which the particular utterance is an instance.

The essay, however, does not focus on the nature and identity conditions of these non-particulars or on the project of providing some formal theory of indirect discourse, although it offers just a little on these topics in the final section. Rather, as the title suggests, the aim is to approach the study of reports of speech through the idea of indexical predication and thereby to situate indirect discourse in a wider context. Doing so brings out motivations for the general view which would still be operative even if some particular implementation is found unacceptable.

Section 2 summarises the main relevant ideas about indexical predication.

INTRODUCTION

Holism is an alluring and heady notion. Many claims that this or that notion is ‘holistic’ have been made, and in particular the idea that semantic notions such as meaning or content are ‘holistic’ has seemed plausible to a good number of philosophers. It is not obvious, however, that they have all meant the same thing, and Fodor and Lepore, in Holism: A Shopper's Guide, have performed a useful service (Fodor and Lepore 1992). They offer an account of what holism is, together with an outline attack on it and an attempt to undermine the major arguments for it. There is, however, further clarification to undertake since, despite their best efforts, the exact content of the thesis remains in doubt. This is in part because Fodor and Lepore (sometimes with and sometimes without argument) proceed as if certain distinct positions are in fact equivalent. The upshot of this is that plausible versions of semantic holism are run together with implausible versions, and the innocent are made to suffer with the guilty.

The structure of the rest of this essay is as follows. The next section aims to disentangle various ways in which ‘semantic holism’ may be understood and to explain how some of them seem to get run together by Fodor and Lepore. Section 3 is primarily ad hominem.

INTRODUCTION

We are often in the position of knowing something about another's state of mind and wanting to know more. For example, I may know of my friend Maria (hereafter ‘M’) that she believes that p1, p2 … and pn, and has set herself to reflect on whether q; I would like to know what conclusion she has reached. Or I know something about her projects and wonder what particular moves she will take to fulfil them. Or I know that she will shortly receive certain information and wonder how she will feel about it. Or I know that she had certain visual experiences and wonder what kinds of afterimages, if any, they produced. Quite often, also, when we are curious in these sorts of ways about others' mental states we can come up with an answer. How do we do it?

Psychological states are extremely various, as are also the kinds of interactions they have with each other. I myself am extremely doubtful that there is any one story to be told about how we proceed in all cases. So the purpose of this essay is not to answer in full generality the question of how we form views on others' thoughts, but only to explore how we may articulate one particular story about how we do it in some cases, namely the cases I mentioned first, where what interests me is what M has concluded about whether q, in the light of her other beliefs.

TWO STRATEGIES

In this essay I want to examine two contrasted models of what we do when we try to get insight into other people's thoughts and behaviour by citing their beliefs, desires, fears, hopes, and so forth. On one model we are using what I shall call the functional strategy, and on the other what I label the replicative strategy. I shall argue that the view that we use the replicative strategy is much more plausible than the view that we use the functionalist strategy. But the two strategies issue in different styles of explanation and call upon different ranges of concepts. So at the end of the essay I shall make some brief remarks about these contrasts.

The core of the functionalist strategy is the assumption that explanation of action or mental state through mention of beliefs, desires, emotions, and so on is causal. The approach is resolutely third-personal. The Cartesian introspectionist error – the idea that from some direct confrontation with psychological items in our own case we learn their nature – is repudiated. We are said to view other people as we view stars, clouds or geological formations. People are just complex objects in our environment whose behaviour we wish to anticipate but whose causal innards we cannot perceive. We therefore proceed by observing the intricacies of their external behaviour and formulating some hypotheses about how the insides are structured. The hypotheses are typically of this form: ‘The innards are like this.

This collection assembles a number of papers published over the past fifteen years which deal with interrelated topics in philosophy of mind and philosophy of language. In two cases (4 and 8) it seemed worthwhile to incorporate some second thoughts into the papers themselves, since the revisions were self-contained and not lengthy. Otherwise the papers appear very much in their original form, except for some minor stylistic tidying.

The papers were written to be self-standing but they dovetail in various ways. Themes touched on in earlier pieces reappear for further treatment later. Ideas developed in some detail at one point are, at other places, merely summarised as the basis for further explorations. These overlaps are, I hope, not excessive and I have not tried to remove them. The Introduction sums up the themes of the various pieces and tries to indicate how they fit together.

The papers have benefited in many ways from discussions with colleagues and from their comments. In particular, I would like to thank Martin Davies, whose invitation to contribute to a collection he was assembling got me thinking about “mental simulation” again in 1994. The paper which was the most direct outcome of his suggestion (Heal 1994) is not included here, since doing so would have increased the amount of repetition. But without the spur presented by that initial invitation, and other encouragement and comment, this collection would not exist.

INTRODUCTION

The central aim of this essay is to articulate and recommend the idea that the simulationist view about what is involved in grasp and use of psychological concepts may be understood in two different ways, namely as an a priori claim about the relations of certain personal-level cognitive abilities or as an a posteriori hypothesis about the workings of sub-personal cognitive machinery. I hope also to show that this contrast has important implications for the simulation debate in general, in that it suggests there is a potentially serious confusion lurking in the way the ‘theory versus simulation’ question is often posed and discussed. The stimulus for the attempt to clarify these matters has been the desire to respond to some of the points made in a recent paper by Stich and Nichols (1997), which is in turn a response to an earlier paper of mine (this volume, Essay 5). So, in the final sections of this essay, 1 shall comment briefly on some aspects of this exchange. But one major issue raised there, namely the role of (some notion of) rationality in defining the domain of simulation, is not fully considered here. I seek rather to provide the context in which a more adequate discussion of that topic might occur.

Put slightly more fully, my central suggestion is this.

INTRODUCTION

‘We would maintain that, for any cognitive capacity, demonstrating that that capacity is cognitively penetrable indicates that the capacity derives from an information base rather than from an off-line simulation’. So write Nichols, Stich, Leslie and Klein (1996), echoing some earlier claims of Stich and Nichols (1995a, 1995b). But are they right? I shall argue first that they are not, and that a simulationist may admit the possibility of cognitive penetration. This discussion will comprise the next three sections of the essay, Section 2 setting the scene and Sections 3 and 4 presenting the main considerations against the quoted claim.

But following that I shall ask whether cognitive penetration of the kind outlined is, in fact, likely to be the source of error in the cases cited by opponents of simulation, and I shall suggest that it is not. In understanding the causes of these errors it is probably more important to grasp what simulation can and cannot plausibly claim to deal with and certain potential complexities of its operation. The domain of simulation is considered in Section 5, where I urge that rational connection of content is the matter that simulation should centrally be expected to cope with. Section 6 summarises possible sources of error that a simulationist can acknowledge without embarrassment and offers some comments on the errors discussed in the literature.

INTRODUCTION

Some, the theory theorists, say that when we make judgements about the psychological states of others and use such judgements to predict or explain, we employ some theory about the psychological. But others, the simulationists, say that we possess no such theory, or at least none complete enough to underpin all our competence with psychological notions. Rather, they say, what we do in such situations is ‘simulate’ others' mental states and processes in ourselves and thus gain insight into what others are likely to do.

My aim in this essay is first to offer an argument in favour of simulationism and second to suggest possible limits to the simulationist strategy. I shall suggest that simulation must be central as far as dealing with the contents of others' mental states is concerned, but is much less clearly of relevance in dealing with non-content. Thus, philosophers and psychologists should not oppose simulation to theory, but rather should ask what is the appropriate realm of each and how they interact.

The topic throughout is the nature of the fully developed adult competence with psychological notions, in the context of predicting others' future psychological states and actions on the basis of knowledge about their current psychological states. I shall not discuss the (it seems to me) importantly different question of how we arrive at judgements about others' thoughts, feelings and so on from knowledge of their placement in the environment or bodily behaviour.

INTRODUCTION

The so-called simulation theory of how we arrive at (some) judgements about others' thoughts is taken by many to be plausible. (See the volumes edited by Davies and Stone [1995a and 1995b] and by Carruthers and Smith [1996] for an introduction to the literature.) Goldman has contributed notably to this debate and also to the discussion of related questions (Goldman 1995, 2000a, 2000b). Among such related questions are these: Can the ideas invoked in the simulation theory of the heuristics of psychological judgements throw any light on our possession of psychological concepts? And can they throw any light on the nature of the mental? Goldman has trenchantly criticised one version of the idea that simulationism can help us in these areas (Goldman 1995). It may be that one could begin to muster some response to some of his objections (see Section 3 below), but I shall not be attempting to revive the kind of view which he attacks, since I share his scepticism about it. The purpose of this essay is, however, to suggest that there is another, and very different, way in which ‘simulationist’ ideas can be deployed to illuminate at least some aspects of our possession of psychological concepts. Whether this alternative approach has anything to offer on the nature of the mental itself is more obscure. I believe that it may, but will have only a little to say on this topic here.