Edited by

Andrew Brook

Carleton University, Ottawa


Richard DeVidi

University of King's College, Halifax


Some intricate, very subtle differences between awareness of self and awareness of other things and the relationship of these differences to some equally intricate features of reference to self using pronouns such as 'I', 'my', 'me' and 'mine' have been the subject of a fairly well defined and autonomous dialogue within analytic philosophy for about thirty-five years now. The investigation began with seminal papers by Hector-Neri Castañeda (1966, this volume; 1967; 1968),(1) took on important new dimensions this Shoemaker (1968, this volume), generated further ground breaking work by Perry (1979, this volume) and Evans (1982, this volume), and continues to produce important new work to this day. Shoemaker called his paper 'Self-reference and self-awareness'. It would be hard to think of a more perfect title for the topics under discussion in this volume, so as well as republishing his seminal paper, we have also gratefully borrowed his title.

At the beginning of his 1966 paper, Castañeda says that his topic is "almost brand new". Though his paper is without question highly original and of first rate importance, Castañeda's claim is an exaggeration. No idea in philosophy (and precious few anywhere else) is ever constructed entirely from whole cloth. With respect to self-awareness and self-reference in particular and indicators more generally, Frege was an important precursor as is well-known, especially (1918). The Wittgenstein of The Blue and Brown Books was another; this has been clear at least since Shoemaker's paper. Less well-known is that Kant also uncovered some of the characteristic peculiarities of self-reference and self-awareness, indeed anticipated claims about them that had to wait till the recent work we've been describing to get a full articulation.

In this collection, we start by exploring the precursors of the recent dialogue on self-reference and self-awareness just mentioned. Next comes four papers that have become 'classics' in this research, indeed laid the foundation for all the subsequent work. The final section offers five examples of the subsequent work, all prepared for this volume.

Two of the precursors, Kant and Frege, are the topic of the two papers that make up Section I, one by each of the editors. Andrew Brook argues that Kant not only anticipated but clearly articulated the phenomenon that Shoemaker calls self-reference without identification, that is to say, that one can refer to oneself, and to refer to oneself as oneself, without (otherwise) identifying oneself. Brook argues, moreover, that, unlike contemporary theorists up to probably Peacocke (this volume), Kant had a theory to explain this peculiarity of self-awareness, a theory that also, among other things, explains when references to oneself are and are not immune to the error of identifying someone else as oneself (though Kant seems not to have noticed this immunity to error itself).(2)

In his paper on Frege, Richard DeVidi urges that

It was not necessary to commission a separate treatment of Wittgenstein's 1930s contributions to our topics. Shoemaker's paper, which had to be part of the volume in any case, contains an excellent discussion of the most interesting ideas in Wittgenstein's middle period work.

Wittgenstein made some well-known remarks on the use of 'I' and related matters in his later work, too, remarks in which many interpret him as denying that when we use 'I', etc., in the relevant ways, we are not using them referentially at all. There is no discussion of these remarks in this volume. To be sure, Peacocke says some relevant things near the end of his paper but this omission may still surprise some people. It reflects the following considerations. First, it is singularly difficult to get a clear, uncontroversial account of what Wittgenstein was trying to show us. If anything, it is even harder to see what the considerations were that, in his view, supported his view. Second, his remarks have persuaded almost no one who is not strongly drawn to his point of view in general. For example, not a single contributor to this volume holds that the relevant uses of 'I', etc., are nonreferential. Indeed, the so-called 'non-cognitivist' approach has had almost no influence on the work that is the topic of this volume.

Section II contains four papers that have achieved the status of classics in work on our topics. The paper that launched the whole recent discussion was Hector-Neri Castañeda's famous work, '"He": A study in the logic of self-consciousness'. In this and other papers of the same period, Castañeda argues that certain uses of 'I' and cognates are ineliminable. In particular, no description, not even one containing (other) indexicals, can be substituted for these uses. This is one of the two great claims that launched the recent discussion.

The second is the idea of immunity to error through misidentification of another as oneself. It was first articulated by Sydney Shoemaker in 'Self-reference and self-awareness', where it is a central theme. This immunity goes with another central peculiarity of self-awareness, what, as we saw earlier, Shoemaker calls self-reference without identification.

In 'The essential indexical', John Perry picks up the first of these claims, that nothing can be substituted for certain uses of 'I' and cognates and articulates it as the idea of the essential indexical. Building his account on what he needs to know in order to find out that a person making a mess in a store is him, Perry argues that without knowing that some part of a description applies to himself, without therefore knowledge of himself (and as himself), even a complete description of the person who is in fact Lingens and his location, features, etc., would be utterly useless to Lingens in trying to find out where he is.

The chapter from Gareth Evans' 1982 book, 'Self-identification', is a massive, sprawling meditation on many aspects of self-reference and self-awareness. One suspects that Evans might have done some more work on it had he lived long enough. At any rate, it resists summarization. Shoemaker's two themes, self-reference without identification and immunity to error through misidentification and their relationship to one another, play a central role in the chapter. So does the idea that one must have available more indexicals than 'I' and its cognates. In particular, one must be able to locate oneself in the objective world (so be able to use 'here') and know when a given time has arrived (so be able to use 'now').

In Section III, we turn to recent work on self-reference and self-awareness. All but one of the papers was prepared for this volume. (The exception is Peacocke's paper, which is also being published in xxx. Millikan's paper is a revision of her well-known 1990 paper.) One of the first things one notices about these five papers is how pervasively the four 'classic' works of Section II dominate recent work. Every paper makes reference to at least two of them. All the papers of Section III criticize papers in Section II to some degree. One of the former, Ezcurdia's paper, also criticizes Millikan's Section III paper. Some will find the criticisms persuasive. Others might suspect that the original authors could handle the criticisms if they had the chance. Whatever, all five papers seem to us to raise significant issues and back their claims with considerable force of argument. In fact, it would extraordinarily interesting to work through the various criticisms, set them against the claims they aim to criticize, and try to adjudicate the disputes.

Ruth Millikan's criticism in 'The myth of mental indexicals' can be summarized quite succinctly: there is something in reference to self that is essential but it is not an indexical. Why? The relevant uses of 'I' do not have central features of indexicals. Indexicals have no constant reference. 'I' as used by a given person always does. Moreover, any element that is indexical in uses of 'I' does not yield the information crucial to predicting what the user of 'I' will go on to do when. And so on. In short, the essential element is not indexical.

In 'Thinking about myself', Maite Ezcurdia criticizes two authors who, she says, argue that there need not be any presentation of self when one refers to oneself using 'I' and cognates. Mellor is one. Millikan, the Millikan of 1990 and this volume, is the other. Mellor holds that the job that a representation of self would do is in fact done by context. If so, Ezcurdia argues, we would have no way to discriminate between an indexical referring to me and an indexical referring to the place I'm at. Since we can discriminate these things, Mellor must be wrong. In Ezcurdia's view, Millikan's emphasis on the relationship between an indexical and a referent specified by the semantic rules for the kind of indexical in question goes with a "Millian" view of indexical reference, in which the semantics of an indexical are exhausted by its naming function and mode of presentation plays no role. Ezcurdia argues that this view leaves Millikan with no way to distinguish between, for example, a use of 'I' presenting an object that happens to be myself and a use of 'I' presenting (what I know to be) myself.

Melinda Hogan and Raymond Martin go after the notion of immunity to error through misidentification in 'Introspective misidentification: an I for an I'. They do so primarily by presenting four putative counter-examples to the view, The counter-examples are all of the form: I am aware of someone having a feeling in the way that I am aware of feelings that I am having but it may in fact not be me who is having them. They consider objections to their counter-examples and reply to the objections. Two of the counter-examples are enmeshed in current controversies about identity of persons over time.

In 'First-person reference, representation independence, and self-knowledge', Christopher Peacocke separates out a special kind of awareness of self in which one knows something about oneself, e.g., that one is seeing something, when this could not be inferred from anything about how the object of this act is represented. Peacocke calls this representationally independent uses of the first person. (His distinction is very subtle and more interesting than this gloss.) What interests him is that often representationally independent uses of 'I' and cognates give one knowledge, knowledge, for example, that it is oneself who is seeing something. He offers what he calls a delta account to explain this, according to which a subject having a property is sufficient in the relevant cases for a subject to refer to itself and specifically, to ascribe the property to itself. He then applies the account to an illusion of a "transcendent subject" that he claims to find in Kant, Schopenhauer and the early Wittgenstein, perhaps without recognizing that his delta account is itself very Kantian (see Brook, this volume) and concludes with a critical look at "no-ownership" theories and theories that the relative uses of 'I' and cognates are not referential.

[the summary of Bill's paper needs work. He leads from Perry and introspection. That needs to be worked in. Suggestions welcome]

Evans (1982, 205; this volume, xxx) urges that we cannot give a complete account of self-reference and self-awareness without having a complete account of the mind. Doubtless he is right, though he is also right to hold that progress can be made in the meantime. In the final paper of the volume, William Seager takes some steps in the direction of developing the kind of general theory that Evans had in mind. He urges that from the perspective of a representational theory of mind, it is natural to arrive at a dual view of selfhood, a dual view of what it is that realizes personhood and carries sameness of personhood across time. He then applies the distinction to a wide range of phenomenon, all the way from Williams' puzzle cases about personal identity to such mundane things as the relative cognitive impenetrability of illusions such as the Müller-Lyer arrowhead illusion and the surprisingly tenacious conflicts between, e.g., our considered values and what we desire that we all find in ourselves from time to time.

The work of the past thirty-five years stimulated by Castañeda's and Shoemaker's seminal papers has rich precursors and has generated a wealth of research of the greatest depth and intricacy. We hope that this volume represents something of this depth and richness.

Andrew Brook

Carleton University

Richard DeVidi

University of King's College

1. Note that we did not refer to 'I", etc., indicators. That we did not may seem a bit odd to some. It is an indications of intricateness and complexity of the recent work on self-reference that even something as basic as whether these pronouns and their cognates are indicators is a matter of debate. For example, one of Castañeda' three seminal papers is called Indicators and Quasi-Indicators. In that paper, he distinguishes pronouns such as 'I' from indicators; according to him, they are just quasi-indicators.

2. As this introduction to the paper on Kant illustrates, we have kept our introductions to the various papers in this volume short. Most of the papers in this volume are so densely and intricately structured that there is no substitute for reading the papers themselves.