What makes me the person I am

1. “Who Am I?” — Qualitative Identity and Defining Properties

In the analytic tradition, “the problem of personal identity” is usually taken to mean a question of numerical identity over time: what makes X at one time the same person as Y at another? But the title also fits a set of questions — at least as interesting — which concern what may be called qualitative identity. A person’s qualitative identity comprises his defining properties (DPs): these are properties1 that he must mention in a full answer to the question “Who am I?”, taken in a special sense which can be discerned by contrast with the ordinary sense of the third-person “Who is X?”. If you and I are watching a ceremony and I, pointing to one of the participants, ask “Who is she?”, my purpose is likely to be to find out that person’s role in the ceremony: the question and the appropriate answer are relative to my purpose, which is set by the context.2 In the sense relevant to DPs, “Who am I?” is not thus relative to context and purpose: rather, in answering the question I identify properties of mine that determine my purposes.3 There is, however, no simple asymmetry between the first-person and the third-person questions; the third-person question can, although it rarely does, take this sense, and conversely the first-person question can be relative to purpose and context, as for example where roles in a game are being assigned and one of the players is unsure of his role. Nevertheless the special sense is more prominent in the first-person case.

DPs “determine” purposes in a sense that covers relations of justification, constitution and causation. Suppose it is one of my DPs that I am a devout Catholic: then my being one justifies me in having — in the sense of being a practical reason for me to have — the purpose of going regularly to church; it is a practical reason for me to fulfil that purpose; and, assuming that I am practically rational, it is a theoretical reason to believe that I have and fulfil that purpose. Also, my having the purpose of going regularly to church is likely to be part of what it is — although it is not a necessary condition — for me to be a devout Catholic, and my being one may cause me to form the purpose of visiting Rome. To say that DPs determine purposes does not imply that a person must first have DPs before he can have any purposes: I may acquire a DP, and may even do so deliberately, in the course of pursuing an existing purpose. If I am set on making money, I may decide that the best method is not only to become a stockbroker but also to adopt that job as part of my qualitative identity.

The justification of purposes is one role played by DPs in practical reasoning. More generally, if F is a DP of mine, that I have F is likely often to be an important premiss, explicit or implicit, in my practical deliberation. Importance can be explicated in various ways, depending on the model adopted of practical reasoning. If the structure of a person’s practical reasons is conceived by analogy with standard models in epistemology,4 it might be viewed as a hierarchy of reasons for action that rest on a set of foundational reasons — the person’s ultimate ends — or as a web of reasons. On the former view, the importance of a premiss will be the greater, the nearer it is to the foundation; on the latter view, the nearer it is to the centre of the web.

A wide variety of properties can be DPs.5 People are often defined by their job, or even their former job (retired civil servant), or by their occupation in a broader sense (housewife, writer). Someone may be defined by nationality, by a connection with an institution (alumnus of Yale), or by relations — notably ones involving commitments or close attachments — to individuals (father of so-and-so). Holding a certain belief — in particular a religious or political belief — can be a DP, as can having a certain trait of character. A DP can be a physical property — being very short, for example, or being beautiful.

The distinction between a person’s DPs and those of his properties that do not define him cross-cuts certain other distinctions, in particular those between necessary and contingent properties and between individuating and non-individuating properties. As to the necessary and the contingent, it may be a DP of mine that I am an artist, but I might not have been one; conversely, it may be a necessary property of mine that I have a certain genetic code, but this is unlikely to appear in my answer to the question “Who am I?” As to individuating and non-individuating properties, I am not the only artist; conversely, even if I have the property of being the only man wearing a black hat in Piccadilly at noon on 1 July 2001, this is unlikely to be one of my DPs. It is a confusion of immaturity to try to base one’s qualitative identity on uniqueness and thus on a contrast between oneself and everyone else.6

The concepts of identity (from now on, except where a distinction is needed, “identity” will mean qualitative identity) and of DPs, and the special sense of the question “Who am I?”, can be defined in terms of each other, but they appear to form a primitive triad:7 an explanation in terms of the determination of purposes, practical reason and the contrast with necessary and individuating properties, or in terms of other concepts such as commitment (see below), is not readily sharpened or expanded into a definition or even into a set of necessary and sufficient conditions. In particular, importance in practical reasoning seems, at any rate on a narrow conception of practical reasoning, not to be necessary for a property to be a DP. If, for example, the DP in question is susceptibility to a mood, its connection with action will not primarily be rational in any narrow sense. My gloominess may affect my actions — and the connection between mood and action may be rational in the broad sense that my behaviour is appropriate to someone who is gloomy — but in deciding what to do I shall not normally reason from the premiss that I am gloomy (although I may do so if, for example, I am trying to find ways of changing my character). Also, in the case of some DPs — such as that of being a contemplative hermit — there is no connection with action in any narrow sense. Nor is importance in practical reasoning sufficient by itself for a property to be a DP. For example, the premiss that I desire to stay alive is likely to be near the foundation or centre of my structure of practical reasons, but the property of desiring to stay alive will not normally be a DP of mine. My answer to the question “Who am I?” is likely to presuppose, rather than mention, this property.

Even if these concepts cannot be defined in other terms, distinctions can be drawn within them, in particular between a person’s actual identity, the identity he strives to have (projected identity) and the identity he believes he ought to have (normative identity).8 There is a parallel distinction between DPs. These identities interact: for example, actual identity constrains both the formation and the achievement of projected identity. Someone with the actual identity of a Chinese peasant farmer is unlikely ever to have heard of the British Parliament, and hence will be unable to strive to have the identity of a member of the House of Commons. Even if he does have the relevant concepts and thus is able to form that projected identity, he is unlikely to be able to achieve it. The obstacle to achievement may be metaphysical rather than causal: time being irreversible, an Indian with the actual identity of being one of Midnight’s Children9 — i. e. one of those born at the moment when India became independent in 1947 — will be unable to achieve the identity of a medieval warrior, even if he is confused enough to strive to have that identity. Conversely, projected identity affects actual identity: if I constantly strive for an identity beyond my reach, I may acquire the actual identity of a loser.10 Projected identity may also affect actual identity through interpretation.11 Suppose that hitherto I have spent my time going to fashionable parties. Disgusted with the triviality of my life, I resolve to become a novelist and, like Proust, to incorporate my experiences of high society in my novels. I accordingly reinterpret my life to date not as a waste of time but as a necessary period for the collection of material. The reinterpretation may involve self-deception — I might pretend to myself that I had only ever gone to the parties with the intention of collecting material — but need not do so. It might be suggested that such cases of interpretation connect projected identity not with actual identity but with beliefs about it: as will be argued later, there is a distinction between who I am and who I think I am. But, as will also be argued, this distinction becomes blurred where the ascription of an identity involves interpretation.

Overemphasis on the interaction between actual and projected identity can amount to bad faith, in a sense similar to Sartre’s.12 One form of bad faith is the assumption that my actual identity fixes my projected identity: I am a coward, and so can never strive to be brave. Another form is the converse assumption: my resolution to be brave shows that I have never really been a coward, despite evidence to the contrary.

2. Living without an identity

Whether a property is a DP is a matter of degree:13 this is confirmed by the connection between DPs and proximity — itself a matter of degree — to the foundation or centre of the structure of a person’s practical reasons, and by the fact that DPs often involve attachments and commitments, which can grow and fade. Correspondingly, identity comes in degrees, and it is a commonplace of developmental psychology that it passes through phases and is subject to crises.14 At all events, the orthodox psychological view — blurred however by a tendency to merge, sometimes deliberately, the identity a person has with the identity he believes himself to have15 (see below) — appears to be that any normal person has an identity. This need not be a form of historical parochialism. Although it seems that the concept of, and still more a preoccupation with, identity is modern — they are arguably products of phenomena characteristic of modern liberal societies, namely high social mobility and the erosion of hierarchies and authority16 — a distinction should be drawn between a concept’s articulation and its applicability. The proposition that members of a rigid archaic society did not articulate the concept of identity does not imply that the concept has no application to them: it is consistent with the plausible view that the concept’s application was too obvious to need articulation. People in those days knew who they were.

The question remains, whether it is possible to live without an identity. In the case of projected and normative identity, the answer seems to be yes: I could live from day to day without ever considering who I want or ought to be. But it also seems possible, if more difficult, to live without an actual identity, or at most with only a low degree of actual identity: such a condition might be described as one of radical irony or detachment.17 Consider a woman who says of her work as a philosopher: “It’s fun, and I have to earn a living, but I’d have been just as happy working in the City”; of her husband: “I get on with him, but who knows how long we’ll stay together?”; of her attendance at synagogue: “It’s just a habit I picked up in childhood”; and so on. It might be said that, by default, she has the identity of an ironist; but that would be inaccurate if, say, her irony came in phases between periods of commitment, or if she had certain commitments and attachments which were nevertheless too weak to ground an identity. It is a further question whether the ironist’s life is desirable.18 The prosaic answer is that it has good and bad points:19 irony is a protection against ossification of the self, bad faith and the bloodshed of “identity politics”,20 but creates a risk of anomie and low self-esteem21 and restricts one’s ability to enjoy those goods that presuppose commitment. In principle, the best of both worlds is attainable whether or not one is an ironist.

3. Plural identities

It is uncontroversial that a person can have various DPs over time and at a time: someone might simultaneously or sequentially be defined as a Buddhist, an invalid and a sculptor. Whether one can likewise have a plurality of identities, as it is fashionable to maintain,22 is a matter of stipulation. As to identity over time (note that the subject is still qualitative, not numerical, identity), there is a clear sense in which my identity as a small boy is different from my identity as an old man,23 but a broad concept of identity includes these and other narrower identities by encompassing my whole life: in this broad sense, my identity is only complete at my death.24 Intermediate concepts can be formulated to cover periods of my life of various durations. There are likewise broader and narrower concepts of identity at a time: my identity at time T may be conceived as the set of all my DPs at T, or as the maximal coherent set of my DPs at T (or, where there is more than one such set, the set of those sets), or I may be conceived as having several, possibly overlapping, identities at T, each being a subset of my DPs at T. Identity, whether over time or at a time, may be conceived as vague, in the sense that two identities are the same provided that enough, but not necessarily all, of the DPs that compose one also compose the other.

Coherence, as a relation among DPs or identities, can be defined in various ways.25 Two of my DPs may be said to lack coherence with each other where my having both entails, or implies, or makes it likely, or causes it to be the case, that I shall make judgments which cannot all be true, or justified, or that I shall act, or be disposed to act, to achieve purposes which cannot jointly be fulfilled. An example is the pair of DPs: being a monk and having a strong sexual drive. (Incoherence among one’s DPs can be an advantage: if the lustful monk is expelled from the monastery, he can console himself with the increased opportunities for sexual encounters.26) At the limit, such incoherence becomes incompatibility: it is impossible for me to have both properties as DPs. The simplest case of incompatibility exists where it is impossible for me to have properties F and G regardless of whether either is a DP of mine; a more interesting case is one in which my having F as a DP involves a commitment to act in a way that precludes me from defining myself by G (example: being a rabbi, being a Nazi). In a thinner sense, two of my DPs may be said to be incoherent, or at any rate non-coherent, if they are irrelevant to each other (example: being a gardener, being cheerful). Relevance of course is relative to specified respects and is usually a matter of degree (cheerfulness may reinforce the gardener’s intention to continue weeding when it starts to rain).

In another sense, DPs may also contribute to the coherence of a person’s actions: where a premiss attributing a DP to me is close to the foundation or centre of my structure of practical reasons, my actions will cohere with each other in the sense that the same premiss figures, directly or indirectly, in the justification of all or many of them. There is a connection between this form of coherence and one concept of integrity, in that actions justified by such a premiss can be said to flow from my identity. Integrity here is a matter of the relation between each such action and the premiss attributing the DP, not of the coherence among the actions that consists in their having a common justification. Whether such coherence can itself be claimed to be a form of integrity is more doubtful: the claim arguably confuses integrity and integration.27

4. Choice of identity

It is a current cliché — arguably attributable to the factors mentioned above as possible causes of the modern preoccupation with identity — that we invent ourselves.28 This sometimes seems to mean little more than that we can change our hair-styles and the music we listen to, but it can also be interpreted to mean that we can choose our DPs and identities.29 This issue can be usefully discussed on the basis of an intuitive distinction between those things we can and those we cannot choose, without going into the general issue of free will. As regards DPs, there are two questions: for a given property F, can I choose whether or not I have F? And, assuming that I do have F, can I choose whether or not F is a DP of mine? There is no general answer to the first question. For some properties, there is no choice: I cannot choose whether or not to be able to fly. In other cases the answer depends on the circumstances: if I am a bright university graduate, my range of choices of career will be larger than if I have a severe mental handicap. The answer to the second question depends both on me and on F. I may be able to choose whether to accept or reject F as one of my DPs, but I shall be unable to do so if I lack the concept of a DP. If I have the concept, but am an unimaginative and conformist member of a society in which F is used to stereotype people, it may still never occur to me that I could reject F as a DP: in that sense I shall be unable to choose. As to F, the choice will often be available even if I cannot choose whether or not I have F: I might be paraplegic, and there might be nothing I can do to regain the use of my legs, but it is up to me whether I define myself by my paraplegia30 (this sort of thought motivates talk of “people with a disability” in place of “the disabled”).

In the case of some properties, the choice whether or not to have them as DPs is either unavailable or constrained. If I have an obsessive-compulsive disorder, the condition may so dominate my life that having it is bound to be a DP of mine: if I deny this, I am lying or deceiving myself. However, the distinction between actual, projected and normative DPs is important here: although I am unable to choose to reject having the disorder as one of my actual DPs, I may still strive, and believe that I ought, to rid myself of it. It might be suggested that I only ever have a choice over my projected and normative DPs, but that is inaccurate: I often can choose my actual DPs, although I seldom can bring it about that I instantaneously acquire one. This is a consequence of the fact, discussed below, that change of DPs is normally a gradual process. One might thus say that the choice is usually of “future actual” rather than “present actual” DPs. Frequently the choice of actual DPs will be by way of projected DPs: if I strive to have F as a DP, I may in time succeed.

The opposite situation is one in which F is a property that I cannot choose to have as a DP: to use an earlier example, the property of being the only man wearing a black hat in Piccadilly at noon on 1 July 2001 is too trivial and fleeting — absent special circumstances — to define me.31 This point is reinforced by the connection between DPs and practical deliberation: the proposition that I have this property is unlikely to be found near the foundation or the centre of my structure of practical reasons. F may also be precluded from being one of my DPs by virtue of its incompatibility with some DP that I already have. This raises the fact that I do not choose my DPs from scratch:32 some will be acquired through upbringing and social circumstances. Nevertheless I shall often be able later to reject my acquired DPs.33 A further constraint on my choice of properties as DPs will be my ethical judgments: this is a truism in the case of normative DPs, but it also applies to projected and actual DPs.34

As already noted, often part of what it is to have a DP is to have a commitment.35 The connection with commitment is closest in the case of projected DPs. In particular, choosing a property as a projected DP often partly consists of making a commitment: for me to choose the property of being a poet as a projected DP, it will not normally be enough for me to scribble a few verses when I have nothing better to do. The making of a relevant commitment is, however, neither necessary nor sufficient for a property to be chosen as a projected DP: I may choose the projected DP of being a poet while yet my striving to be one is too heavily qualified by reservations to amount to a commitment; conversely, I may be committed, say, to finishing a report in time for the next board meeting without any of my DPs being at stake. A more general account of the choice of projected DPs can be given in terms of the connection already described between DPs and practical reasoning: in choosing F as such a DP, I give to the proposition that I have F a place near the foundation or centre of my structure of practical reasons.

Choice of identity is choice of DPs writ large: how large depends on the concept of identity at issue. The broader the concept, the less likely it is that one’s identity is chosen — at any rate if this means chosen explicitly and in one go. In rare cases a child might choose his lifelong identity, but more often identity emerges gradually from more limited choices.36 The extent to which the formation of identity from such choices is a conscious and intentional process varies between people; in so far as it is, it may be guided by an ideal of coherence, both among one’s DPs and among actual, projected and normative identity.37 Coherence is a minor ideal:38 the possible advantages of incoherence among one’s DPs have already been mentioned, and in any case the ideal is, at least primarily, formal rather than substantive: a wicked person may be coherent in his DPs and identities.39 It might also be objected, in Faustian vein, that the ideal is undesirably static, but that would be to conceive of coherence in restrictive terms: coherence can be conceived dynamically, as stretching over time and accommodating change.40 Thus the accounts suggested earlier, in terms of compossibility, might be extended to provide for example that, where F is a DP of mine at time T1 and G a DP of mine at T2, F and G cohere pro tanto if, first, my having F is partly constituted by my having the purpose PF, second, my having G is partly constituted by my having the purpose PG and, third, my acting at T1 in a way that fulfils PF triggers, or can be reasonably foreseen to trigger, a sequence of events that makes it possible for me to fulfil PG at T2. Even if coherence is conceived restrictively, the objection of stasis is weakened by the fact that a person’s DPs are always liable to be unsettled by contingent events, and that accordingly a constant effort may be needed to maintain coherence or, where necessary, to re-establish it.41

That choices of DPs and identities are constrained, or even impossible, follows from certain versions of an axiom of social psychology, that identities are socially determined.42 This can mean various things, for example that social forces cause me to have an identity; or that they cause me to have the particular identity I have; or that that the concept of identity is itself a social product (this might be derived from a more general thesis about the social basis of concepts); or that my having an identity, or the particular identity I have, entails that I am or have been, or at least could be, a member of a society. This claim of entailment is true at least as regards identities comprising certain DPs, notably those of having certain social roles,43 but it is also confirmed by other DPs which presuppose social institutions, language in particular: examples include traits of character such as being witty and being articulate. The range of DPs that presuppose social institutions is wide: a man may have the DP of being placid, and it it is true that some non-social creatures are placid, but it is plausible to hold that the placidity of men differs from that of, say, cows and that the former has a social dimension.

The thesis that identities are socially determined is false if it is read as excluding the possibility that I define myself as rejecting society. Clearly I may define myself as rejecting the bigoted society in which I was brought up,44 but I may even define myself as rejecting society altogether, by becoming a hermit. Against this it might be said that in becoming a hermit I take myself to be joining some other society,45 possibly that of God and the saints; but the reply is, first, that this need not be so — I might for example be a nature mystic — and second that, even if it is so, I need not (although in practice it is likely that I shall) define myself by membership of that society. Granted, to define myself as rejecting society, I need intellectual equipment that I can presumably only acquire by having been a member of a society, but again it does not follow that being or having been a member of the society in question, or of any society, is ever a DP of mine. Even if it is one of my actual DPs, I can reject it as a projected DP, and if I abide by my decision it is likely to cease to be an actual DP.

5. Who I am and who I think I am

It is one thing to say that I may be able to choose my DPs and identity, another to say that there is no distinction between, on the one hand, the DPs and identity that I have and, on the other, those that I believe myself to have. The latter claim is false; in that sense, realism is the right view of DPs and identity.46 So far as concerns DPs, there are — as there were in the discussion of choice — two issues, one as to my having a property F and the other as to F’s being a DP of mine. In all or almost cases I can be mistaken as to whether I have F. Arguably there are some properties — those to which I have certain forms of privileged access — about which I cannot make such a mistake, but it is doubtful that these are fitted to be DPs: the usual candidates are such properties as feeling pain now and seeming to see a patch of yellow.47 Assuming I truly believe that I have F, I may still be mistaken as to whether F is one of my DPs. Suppose I believe myself to have the actual DP of being a family man, and that I am one in the sense that I have a family of whom I am fond: my belief will nevertheless be false — it is likely to be an instance of self-deception — if I constantly put my career first, I am unfaithful to my wife and I forget my children’s birthdays. The connection between DPs and practical deliberation is again illuminating here: being a family man fails to count as one of my DPs because the proposition that I am one is insufficiently close to the foundation or centre of my structure of practical reasons.

Mistake is also possible in the case of one’s projected and normative DPs. A woman might believe herself to have the projected DP of being an actress: she recognises that she never gets any acting work — and thus that being an actress is not an actual DP of hers — but she nevertheless believes that she strives to be an actress. In fact this belief is false: she neglects to look for auditions; when she hears about one, she usually fails to apply; and if she does apply, she usually fails to turn up. Again this is likely to be a case of self-deception: to think of oneself as an actress is more gratifying than to resign oneself to being a temporary waitress. As to normative DPs, the question is not whether I can be mistaken about the DPs I ought to have — which is one version of the question of realism in ethics — but whether I can be mistaken about those I believe I ought to have. The answer is yes: the mistake might consist of confusion or, again, of self-deception — as for example where I want to convince myself that my normative beliefs coincide with those held by someone of whom I am in awe.

The distinction between belief and reality is blurred, however, by the fact that the ascription of DPs often involves interpretation.48 Again there is a question both as to my having F and as to F’s being one of my DPs. Suppose I claim to be kind. You cast doubt on this by citing cases in which I have treated people harshly. I reply that harshness was demanded in the circumstances; for example, plain speaking was needed to shake someone from an illusion that would have caused him unhappiness if allowed to persist. You say that, even if this was so, I need not have been quite so brutal — and so on. There is no simple fact of the matter here. Suppose now that I persuade you of my kindness: it may still be a matter of interpretation whether kindness is a DP of mine. In particular, we may still disagree as to whether my disposition to treat people kindly is pervasive enough and often enough realised to constitute one of my actual DPs. The debate will involve the identification of patterns in my behaviour: various patterns can be discerned in a single stretch of behaviour, and it will be a matter of interpretation which ones are given emphasis. This is not to say that the identification of patterns is a merely subjective matter: some will not be there to be identified. A familiar analogy is the formulation of mathematical functions to describe a curve: an infinite number of functions will fit the same curve, but an infinite number will not.

It is clear from this discussion that I am fallible as to whether F is a DP of mine; but it may be asked whether there is some sense, falling short of infallibility, in which I have authority on the matter.49 That there is such a sense might be thought to be implied by the connection drawn, at the start of the paper, between DPs and the first-person question “Who am I?” The implication was not intended. One sense in which it might be claimed that I have such authority is this: if you and I both know that I have F, and we consider the question whether F is one of my DPs, I am in a better position than you to give the right answer. This claim is false; which of us is in the better position will depend on the property at issue. I may be where F’s being or not being a DP of mine is primarily an inward matter, for example where F is the property of striving for inner serenity in the bustle of life. You may be where F’s being or not being a DP of mine is largely revealed in my outward behaviour, for you may be in the better position to survey my behaviour and to discern patterns in it: the example of my kindness might be such a case. Another sense of “authority” does not compare one person’s epistemic position with another’s: it might be claimed that I have authority as to my DPs in that, if I know that I have F, and I consider the question whether F is one of my DPs, the answer I reach is more likely than not to be true. This claim is also false; to continue with the same example, the circumstances may make it very hard for me to tell whether my disposition to treat people kindly is so pervasive and so often realised as to constitute one of my actual DPs. The claim might now be modified to say that the probability that the answer I reach is true exceeds a certain threshold. But this claim is still false if the threshold is high enough — if it is above 50 per cent, the position is the same as before — and it becomes uninterestingly true if the threshold is low enough. It is a barren exercise trying to identify a number that will apply to all or most properties: to the extent that a number or range of numbers can be identified at all, it will vary between cases.

This section began with an affirmation of realism about DPs and identity, in the weak sense of a distinction between the DPs and identity that I have and those that I believe myself to have. A strong realism contends that whether or not I have a given DP or identity is mind-independent. This might mean various things, but it is implausible on its face. Clearly it is false when stated thus generally — for some DPs, such as holding a certain belief, are mental properties — but it is an unintuitive thesis even in the case of other DPs. One argument against strong realism invokes the view mentioned earlier, that identities are socially determined. As noted then, this view also can be interpreted in various ways: on certain interpretations it entails that, if there were no societies, there would be no DPs or identities. It is plausible to hold that if there were no minds there would be no societies. Assuming a suitably restricted transitivity,50 it follows that if there were no minds there would be no DPs or identities. This undermines the plainest versions of strong realism.

6. Self-reflection

The fact that one can be mistaken about one’s DPs and identity, and the fact that there is scope to choose them, motivate reflection about them, but it is debatable how much reflection of this kind is desirable.51 The modern preoccupation with identity, noted above, receives its most intense philosophical expression in existentialism: Heidegger writes that “Dasein… is ontically distinguished by the fact that, in its very Being, that Being is an issue for it”, and Sartre that “the being of consciousness is a being such that in its being, its being is in question”.52 Such claims can be read as supporting the view that one should keep one’s identity under constant scrutiny, and on that reading they can produce impatience: this seems to be a recipe for permanent adolescence. The brusque opposing view — characteristic of old-fashioned British public schools — is that you should decide who you want to be and then get on with it: once I have decided that I want to be a doctor, I shall make myself unhappy, and may thwart my aspiration, if I keep reconsidering the decision and contemplating other careers. Concern with one’s identity, it might be added, is a luxury of affluence: it does not bother anyone who has serious things to worry about.

This response is overstated. While it may be true that reflection on one’s identity is characteristic of adolescence, it is often a necessary step towards psychological maturity.53 Questions of identity may also properly arise in later life, for example when one loses one’s job or one’s spouse or one’s religious faith.54 In any event, affluence does not remove all serious worries — it creates some — and often part of what makes a worry serious is its bearing on identity: the worst part of a grave illness may be that it prevents me from continuing the career that has hitherto defined me.55 In so far as a concern with identity is attributable to aspects of modern liberal society, affluence seems less important than the factors already mentioned — high social mobility and the decline of hierarchies and authority. (This is not to deny that, if I am so poor that I must spend my whole time searching for food, I am unlikely to think much about my identity.) There is not a general right view as to how much one should reflect on one’s identity: it depends on personality56 and on personal and social circumstances.57

7. Change of identity

A person’s DPs and (except for the broadest concept of identity over time: see above) his identity are liable to change: this can be depressing or alarming, particularly in so far as identity is constituted by value-judgments,58 for a change of identity can thus seem a form of apostasy. If my identity as a composer is partly constituted by the judgment that a composer’s life is the best life I can lead, and circumstances move me to abandon that life in order to make money as a lawyer, it is likely to be cold comfort to be told not to worry because in time my values as a composer will fade and I shall acquire the identity and values of a lawyer.59 The degree of concern is liable to be greater if I take a realist view of values; for then, to the extent that the lawyer’s values conflict with the composer’s, I shall regard my future identity as embodying an error. One’s reaction to a future change of identity will depend, however, on the degree of satisfaction with one’s present identity: if I am miserable as a composer because my work is never performed, I may resign myself to, or even welcome, the prospect of ridding myself of a composer’s values. Likewise, I may desire to change my actual identity where it subsists with judgments that condemn it (I despise myself for having the identity of an alcoholic) or where judgments condemning part of the way of life embodied in my actual identity themselves partly constitute that identity (my identity as an alcoholic may in part be constituted by self-contempt). In retrospect a change of identity dreaded in advance may be judged to have been for the best: if I become a happy and successful lawyer, I am likely to be glad no longer to be a miserable and unsuccessful composer.

Change of DPs is normally a gradual process: a property does not suddenly become one of my DPs; if it is one it will remain one for an extended period; and a DP is acquired against a background of other DPs which remain in place. In these respects, DPs provide continuity of identity. Even if I suddenly acquire a property, it normally does not suddenly become a DP of mine. Suppose that my wife dies, so at the instant of her death I acquire the property of being a widower: it is nevertheless likely to take time for widowerhood to become an actual DP of mine. Part of the pain of grief arises from the difficulty of adjusting one’s identity to the absence of the loved one (and another, later, part from the recognition that the adjustment has been made60). Sometimes, however, a DP can be suddenly gained. A striking achievement may both occur at an instant and give the person in question a new actual DP at the instant it occurs: being the first to run a four-minute mile might be an example. An example of the sudden acquisition of a projected DP is the case of a woman who goes to a political rally and is inspired there and then to devote her life to the cause. More common is the sudden loss of a DP (some DPs are as reputations are said to be — gained slowly, lost quickly), for I shall suddenly lose a DP whenever I suddenly lose the property that is the DP. Reverting to the example of the widower: if, while my wife is alive, it is an actual DP of mine that I am married to her, I shall lose this DP at the instant of her death. Russell’s notorious bicycle ride is arguably another example of the sudden loss of an actual DP;61 but it is possible that, even if Russell suddenly stopped loving Alys, the property of loving her had previously and gradually ceased to be a DP of his — if it had ever been one. In any case, the event that he describes as suddenly occurring is his recognising that he no longer loved her, not his ceasing to love her: many apparent cases of the sudden gain or loss of a DP are in fact cases where someone suddenly becomes aware that he has already gained or lost one.62

There are various reasons for the resistance of DPs to sudden and frequent change. One is the role of DPs in practical deliberation: given that propositions ascribing them tend to be near the foundation or centre of a person’s structure of practical reasons, a change is liable to entail extensive and disruptive revision of the structure. Another reason is that certain DPs — notably those of holding certain beliefs or of having certain traits of character — can only be changed indirectly: I cannot simply decide to believe in God63 or to be kind-hearted. Also, at least some DPs must of their nature be possessed for an extended period:64 the property of having a given character trait is again an example; another is any DP involving a commitment. I cannot be kind-hearted, or committed to a religious life, merely for an instant. It seems in fact that no property possessed for an instant is fit to be a DP. There was an instant at which Armstrong stepped on to the moon, but the relevant DP here is not the fleeting property of stepping on to the moon but the enduring one of being the first person to have stepped on to it.

The point made earlier, about the formation of identity from choices, applies equally to the changing of DPs: the extent to which it is conscious and intentional varies between people, and in so far as it is it may be guided by an ideal of coherence. One model of coherence which may be proposed in this context is the concept of narrative: the idea is that transitions between DPs can be intelligible as stages in the story of a life, or of part of a life.65 The proposal faces a dilemma. Narrative may be conceived thinly, as no more than a chronology of events — in which case it is not a source of intelligibility.66 Alternatively, a thicker concept can be used, which includes structures from certain forms of literature — but then it is unclear why narrative should be an appealing model of coherence: why should we want or expect our lives to be homologous to novels, comedies or epics? Moreover it seems that, where a narative does make intelligible a transition between DPs, it often does so by ascribing causes or reasons which could be invoked in the absence of the narrative: to that extent, the concept is doing no work. Rather than providing a general account of coherence through change of DPs, the concept of narrative seems best suited here to the more limited role of providing a framework for the reinterpretation of earlier DPs in the light of later ones, or of actual in the light of projected DPs (see the example of the novelist who reinterprets his earlier life as a period of collecting material). The broader task seems better accomplished by a dynamic concept of compossibility, as sketched above.

8. Qualitative and numerical identity

The paper ends as it began, with the distinction between qualitative and numerical identity of persons.67 The former is neither necessary nor sufficient for the latter. That it is not necessary is shown by the ironist, who can be reidentified over time although having no, or only a low degree of, actual qualitative identity, and (for all but very broad concepts of qualitative identity over time) by the young boy and the old man who are numerically identical to each other but differ in their actual, projected and normative identities. It might be argued that some radical changes of qualitative identity break the thread of numerical identity: Parfit’s example of the nineteenth-century Russian, who as a young man regards his socialist ideals as essential to him but abandons them in middle age, might be described in these terms;68 but it seems that such a description would either be hyperbolical or presuppose a revisionist concept of personhood. That qualitative identity is not sufficient for numerical identity follows from the fact that not only are DPs distinct from individuating properties, as noted earlier, but two people can share all the DPs that constitute their qualitative identity. You and I may both be English aristocratic hypochondriacs who work in banking (actual identity), strive to become rich enough to retire as country squires (projected identity) and believe that we ought to have had military careers (normative identity).

Qualitative and numerical identity are nevertheless connected in significant ways. Standard accounts of numerical identity analyse it in terms of continuity, bodily or psychological,69 but it is sometimes sustained in part by continuity of qualitative identity: part of what makes the old man numerically the same person as the young boy may be a gradual and overlapping change of DPs. “Sustained in part” might here be glossed in terms of a condition more complex than a necessary or sufficient condition — possibly, in some cases, an “inus” condition in Mackie’s sense (an insufficient but non-redundant part of an unnecessary but sufficient condition).70 The relations between continuity of qualitative identity and the other forms of continuity differ from case to case; where the DPs include psychological properties, the same factors — say, a gradual change of character — may partly constitute both psychological continuity and continuity of qualitative identity. Conversely, numerical identity at least sometimes partly sustains qualitative identity. It was suggested earlier that a DP cannot be had for only an instant: possession of a DP therefore presupposes numerical identity through the minimum period.71 There is a further connection. The existence of a person who can be reidentified over time entails a degree of integration of his properties, specifically those of having certain mental states:72 as integration declines, the person subsides into madness and at a certain point ceases to exist.73 It is plausible to hold that such integration in its higher degrees may partly consist of coherence among DPs and qualitative identities.

9. References

  • T. Adorno [1973], The Jargon of Authenticity, Routledge & Kegan Paul, London.
  • W. Alston [1989], “Varieties of Privileged Access”, in Epistemic Justification, Cornell University Press, Ithaca.
  • S. Asch [1946], “Forming Impressions of Personality”, Journal of Abnormal and Social Psychology 41.
  • L. Athens [1995], “Dramatic Self Change”, Sociological Quarterly 36.
  • R. Audi [1993], The Structure of Justification, Cambridge University Press, Cambridge.
  • D. Bakhurst [2000], “Ethical Particularism in Context”, in B. Hooker & M. Little (eds.), Moral Particularism, Clarendon Press, Oxford.
  • M. Benjamin [1990], Splitting the Difference: Compromise and Integrity in Ethics and Politics, University of Kansas Press, Kansas.
  • P. Berger [1966], Invitation to Sociology, Penguin Books, Harmondsworth.
  • L. Berk [1997], Child Development (4th edition), Allyn & Bacon, Boston.
  • O. Black [1994], “Ends, Desires and Rationality”, International Philosophical Quarterly 34, 1.
  • O. Black [Unpublished], “Ethics, Identity and the Boundaries of the Person”.
  • S. Boër and W. Lycan [1986], Knowing Who, MIT Press, Cambridge MA.
  • W. Bouwsma [2000], The Waning of the Renaissance, Yale University Press, New Haven.
  • R. Brown [1986], Social Psychology: The Second Edition, The Free Press, New York.
  • W. Bruford [1962], Culture and Society in Classical Weimar, Cambridge University Press, Cambridge.
  • S. Butler [1933], The Way of All Flesh, Dent, London.
  • W. Collins [1908], Poor Miss Finch, Charles Scribner’s Sons, New York.
  • J. Dancy [1993], Moral Reasons, Blackwell, Oxford.
  • D. Davidson [1980], Essays on Actions and Events, Clarendon Press, Oxford.
  • D. Davidson [1984], Inquiries into Truth and Interpretation, Clarendon Press, Oxford.
  • D. Davidson [2001], Subjective, Intersubjective, Objective, Clarendon Press, Oxford.
  • K. Durkin [1995], Developmental Social Psychology, Blackwell, Oxford.
  • E. Erikson [1968], Identity: Youth and Crisis, Norton, New York.
  • S. Evnine [1989], “Understanding Madness?” Ratio, new series, 2.
  • O. Flanagan [1991], Varieties of Moral Personality, Harvard University Press, Cambridge MA.
  • O. Flanagan & A. Rorty (eds.) [1990], Identity, Character, and Morality, MIT Press, Cambridge MA.
  • S. Gallagher & J. Shear (eds.) [1999], Models of the Self, Imprint Academic, Thorverton.
  • P. Gilroy [1996], “British Cultural Studies and the Pitfalls of Identity”, in H. Baker, M. Diaware & R. Lindeborg (eds.), Black British Cultural Studies: A Reader, University of Chicago Press, Chicago.
  • J. Glover [1988], I: The Philosophy and Psychology of Personal Identity, Allen Lane, London.
  • E. Goffman [1963], Stigma: Notes on the Management of Spoiled Identity, Prentice Hall, Englewood Cliffs.
  • E. Goffman [1969], The Presentation of Self in Everyday Life, Allen Lane, London.
  • E. Gosse [1949], Father and Son, Penguin Books, Harmondsworth.
  • J. Habermas [1976], “Können komplexe Gesellschaften eine vernünftige Identität ausbilden?” in Zur Rekonstruktion des Historischen Materialismus, Suhrkamp Verlag, Frankfurt.
  • R. Harré [1989], “The ‘Self’ as a Theoretical Concept”, in Krausz [1989].
  • P. Harvey [2000], An Introduction to Buddhist Ethics, Cambridge University Press, Cambridge.
  • M. Heidegger [1962], Being and Time, Harper Collins, San Francisco.
  • E. Higgins [1987], “Self-Discrepancy: A Theory Relating Self and Affect”, Psychological Review 94.
  • M. Hogg & G. Vaughan [1998], Social Psychology (Second Edition), Prentice Hall, Harlow.
  • W. James [1950], The Principles of Psychology, Dover Publications, New York.
  • F. Jenkins [1999], “Sense and Sensibility: the Integrity of Vision and Response”, in Montefiore & Vines [1999].
  • G. Keller [1878-9], Züricher Novellen, in Gesammelte Werke, vol. 2, Knaur, Berlin.
  • M. Krausz (ed.) [1989], Relativism: Interpretation and Confrontation, University of Notre Dame Press, Notre Dame.
  • M. Kuhn [1960], “Self-attitudes by Age, Sex, and Professional Training”, Sociological Quarterly 1.
  • R. Laing [1965], The Divided Self, Penguin Books, Harmondsworth.
  • D. Lewis [1973], Counterfactuals, Blackwell, Oxford.
  • A. Lindesmith, A. Strauss & N. Denzin [1999], Social Psychology (8th edition), Sage, Thousand Oaks.
  • P. Linville [1987], “Self-complexity as a Cognitive Buffer Against Stress-related Depression and Illness”, Journal of Personality and Social Psychology 52.
  • J. Locke [1965], An Essay Concerning Human Understanding, Dent, London.
  • A. MacIntyre [1985], After Virtue, Duckworth, London.
  • J. Mackie [1974], The Cement of the Universe, Clarendon Press, Oxford.
  • J. Macquarrie [1973], Existentialism, Penguin Books, London.
  • G. Márquez [1997], Love in the Time of Cholera, David Campbell, London.
  • R. Martin [1998], Self-Concern, Cambridge University Press, Cambridge.
  • G. Mead [1934], Mind, Self & Society, University of Chicago Press, Chicago.
  • A. Montefiore & D. Vines (eds.) [1999], Integrity in the Public and Private Domains, Routledge, London.
  • P. Morris [1976], Sartre’s Concept of a Person: An Analytic Approach, University of Massachusetts Press, Amherst.
  • A. Nehamas [1985], Nietzsche: Life as Literature, Harvard University Press, Cambridge MA.
  • P. Noordhof [2001], “Believe What You Want”, Proceedings of the Aristotelian Society 101/3.
  • M. Nussbaum [1994], The Therapy of Desire, Princeton University Press, Princeton.
  • F. Olafson [1967], Principles and Persons, The Johns Hopkins Press, Baltimore.
  • D. Parfit [1984], Reasons and Persons, Clarendon Press, Oxford.
  • M. Philp [1999], “Citizenship and Integrity”, in Montefiore & Vines [1999].
  • A. Phoenix [1999], “Personal Growth and Racialized Identities: The Inextricable Linking of the Personal and the Social in Everyday Practices”, in D Messer & S Millar (eds.), Exploring Developmental Psychology, Arnold, London.
  • J. Plamenatz [1975], Karl Marx’s Philosophy of Man, Clarendon Press, Oxford.
  • R. Putnam [1990], “The Moral Life of a Pragmatist”, in Flanagan & Rorty [1990].
  • P. Railton [1988], “Alienation, Consequentialism, and the Demands of Morality”, in S Scheffler (ed.), Consequentialism and Its Critics, Oxford University Press, Oxford.
  • W. Ralegh [1961], “All the World’s a Stage”, in E Chambers (ed.), The Oxford Book of Sixteenth Century Verse, Oxford University Press, London.
  • J. Raz [2000], “The Truth in Particularism”, in B Hooker & M Little (eds.), Moral Particularism, Clarendon Press, Oxford.
  • P. Ricœur [1992], Oneself as Another, The University of Chicago Press, Chicago.
  • A. Rorty [1988], Mind in Action, Beacon Press, Boston.
  • A. Rorty [1989], “Relativism, Persons, and Practices”, in Krausz [1989].
  • A. Rorty [1999], “Integrity: Political Not Psychological”, in Montefiore & Vines [1999].
  • A. Rorty & D. Wong [1990], “Aspects of Identity and Agency”, in Flanagan & Rorty [1990].
  • C. Rovane [1998], The Bounds of Agency, Princeton University Press, Princeton.
  • S. Rushdie [1995], Midnight’s Children, David Campbell, London.
  • B. Russell [1967-9], The Autobiography of Bertrand Russell, George Allen and Unwin, London.
  • J. Sabini [1995], Social Psychology: Second Edition, Norton, New York.
  • J-P. Sartre [1969], Being and Nothingness, Methuen, London.
  • W. Shakespeare [1951], As You Like It, in P. Alexander (ed.), William The Complete, Shakespeare Works, Collins, London.
  • M. Snyder [1974], “Self-monitoring of Expressive Behaviour”, Journal of Personality and Social Psychology 30.
  • M. Snyder [1979], “Self-monitoring Processes”, in L Berkowitz (ed.), Advances in Experimental Social Psychology, vol. 12, Academic Press, New York.
  • R. Stainton Rogers, P. Stenner, K. Gleeson & W. Stainton Rogers [1995], Social A, Psychology Critical Agenda, Cambridge: Polity Press.
  • R. Stalnaker [1991], “A Theory of Conditionals”, in F. Jackson (ed.), Conditionals, Oxford University Press, Oxford.
  • G. Strawson [1999], “The Self”, in Gallagher & Shear [1999].
  • H. Tajfel [1981], Human Groups and Social Categories, Cambridge University Press, Cambridge.
  • H. Tajfel & J. Turner [1979], “An Integrative Theory of Social Conflict”, in W. Austin & S. Worchel (eds.), The Social Psychology of Intergroup Relations. Brooks/Cole, Monterey.
  • C. Taylor [1975], Hegel, Cambridge University Press, Cambridge.
  • C. Taylor [1985a], Human Agency and Language, Cambridge University Press, Cambridge.
  • C. Taylor [1985b], Philosophy and the Human Sciences, Cambridge University Press, Cambridge.
  • C. Taylor [1989], Sources of the Self, Cambridge University Press, Cambridge.
  • C. Taylor [1991], The Ethics of Authenticity, Harvard University Press, Cambridge MA.
  • R. Travisano [1981], “Alternation and Conversion as Qualitatively Different Transformations”, in G. Stone & H. Farberman (eds.), Social Psychology Through Symbolic Interaction (2nd edition), John Wiley, New York.
  • L. Trilling [1955], The Opposing Self, Harcourt Brace Jovanovich, New York.
  • L. Trilling [1972], Sincerity and Authenticity, Harvard University Press, Cambridge MA.
  • A. Trollope [1992], The Eustace Diamonds, David Campbell, London.
  • E. Tugendhat [1986], Self-Consciousness and Self-Determination, MIT Press, Cambridge MA.
  • S. White [1991], The Unity of the Self, MIT Press, Cambridge MA.
  • D. Wiggins [1998], Needs, Values, Truth (3rd edition), Clarendon Press, Oxford.
  • K. Wilkes [1999], “GNOTHI SEAUTON (Know Thyself)”, in Gallagher & Shear [1999].
  • B. Williams [1973], “Personal Identity and Individuation”, in Problems of the Self, Cambridge University Press, Cambridge.
  • R. Wollheim [1984], The Thread of Life, Harvard University Press, Cambridge MA.
  • W. Wordsworth [1982], “Surprised by joy — impatient as the Wind”, in P. Sheats (ed.), The Poetical Works of Wordsworth, Houghton Mifflin, Boston.

  1. Followers of Heidegger might argue that talk of properties is out of place, on the basis that properties characterise the “present-at-hand” whereas identity falls within the realm of Dasein: Heidegger [1962], p. 200. The terminology of properties here is not intended to carry ontological weight. For the suggestion that existentialist thought is in any event best shorn of its ontology, see Olafson [1967], ch. 5. ↩︎

  2. This is the core of Boër and Lycan’s analysis of “knowing who”: [1986], part 1. ↩︎

  3. In this respect, but not in others, the concept of DPs is similar to Sartre’s concept of a “fundamental project”: [1969], part 4, ch. 1. See notes 32 and 39 below. ↩︎

  4. This approach is taken by Audi in [1993], ch. 13. ↩︎

  5. Compare Railton on the various “constituents of identity”: [1988], pp. 128-9. A similar picture is presented in Rorty [1988], p. 293. For patterns in the variety of properties that people specify in answer to the question “Who am I?”, see Kuhn [1960]. ↩︎

  6. The confusion is perpetrated by Jacques in Keller [1878-9]. On the aspiration to uniqueness, see Rorty [1988], pp. 90-91, and Tugendhat [1986], pp. 260-61. ↩︎

  7. Brown takes a bluff approach: “Identity is a concept no one has defined with precision, but it seems we can move ahead anyway, because everyone roughly understands what is meant” ([1986], p. 551). For some sketches of attempted definitions, see Erikson [1968], p. 208. ↩︎

  8. Higgins makes a parallel classification, of self-schemas rather than identities, in [1987]. Distinctions between types of identity abound in the psychological and sociological literature, for example between personal and social identity (Brown [1986], p. 551, Hogg & Vaughan [1998], pp. 364-5); personal, circumstantial and structural identity (Lindesmith et al [1999], p. 14); circumstantial and situational identity (Berk [1997], p. 426); social, personal and ego identity (Erikson [1968], p. 50, Goffman [1963], pp. 2, 56, 105); identity as subjectivity, as sameness and as solidarity (Gilroy [1996], pp. 228-9); and natural identity, identity of roles and ego identity (Habermas [1976], pp. 94 ff.). Some of these distinctions in fact concern the identity a person believes himself to have, rather than the identity he has in fact; as noted in the main text below, this contrast is often blurred. There is also little consistency among the systems of classification: for example, Brown holds that the individual has several social identities but only one personal identity, whereas Hogg and Vaughan hold that the individual can also have several personal identities. ↩︎

  9. See Rushdie [1995]. ↩︎

  10. A passage in Sartre suggests that my projected identity may be that of a loser: [1969], p. 472. ↩︎

  11. See Berger [1966], ch. 3; Martin [1998], ch. 5; Nehamas [1985], chs. 5-6; Ricœur [1992], sixth study; Sartre [1969], part 4, ch. 1; Heidegger [1962], secs 65 and 74. Heidegger writes: “Anticipation of one’s uttermost and ownmost possibility is coming back understandingly to one’s ownmost ”been“” (p. 373). ↩︎

  12. Sartre [1969], introduction, ch. 2. ↩︎

  13. Compare Asch’s distinction between central and peripheral traits: [1946]. Rorty and Wong ([1990]) describe identity in terms of central traits, and represent centrality as a matter of degree. ↩︎

  14. The notion of an identity crisis is associated with Erikson: see [1968]. There are overviews of the literature in Durkin [1995], ch. 15, and Lindesmith et al [1999], chs. 8 and 11. ↩︎

  15. For example Durkin writes of “achieving identity, in the sense of acquiring a set of beliefs about the self”: [1995], p. 290. ↩︎

  16. See Benjamin [1990], p. 7; Berger [1966], ch. 3; Plamenatz [1975], ch. 12; Taylor [1989], chs. 2 and 14, and [1991], ch. 5; Trilling [1972], ch. 1. ↩︎

  17. Lizzie, in The Eustace Diamonds, is arguably a radical ironist: see Handley’s introduction in Trollope [1992]. ↩︎

  18. On the benefits of irony, and their representation in the writings of Wilde, Nietzsche and Austen, see Trilling [1972], ch. 5, and the essay on Mansfield Park in Trilling [1955]. ↩︎

  19. On good and bad aspects of identity, see Erikson [1968], p. 41. ↩︎

  20. For a Buddhist view of the roots of conflict in the “”I am“ conceit”, see Harvey [2000], p. 240. ↩︎

  21. The connection between identity and self-esteem is central to Tajfel’s social identity theory: see Tajfel [1981] and Tajfel and Turner [1979]. For a survey of recent work in this area, see Hogg & Vaughan [1998], pp. 363 ff. ↩︎

  22. See for example Phoenix [1999]. ↩︎

  23. “[H]uman beings are not born once and for all on the day their mothers give birth to them, but… life obliges them over and over again to give birth to themselves” (Márquez [1997], p. 199). ↩︎

  24. There are versions of this idea in Heidegger [1962], sec. 46 et seq, and Nehamas [1985], chs. 5-6. ↩︎

  25. One approach might be to extend White’s concept of an “ideal reflective equilibrium” of desires ([1991], chs. 7-11); White says that we are identified with our desires in such equilibrium (p. 242). Another might be to extend Rovane’s concept of overall rational unity within a person’s rational point of view: [1998]. See also Wollheim [1984], ch. 6, on co-existence and conflict of desires. ↩︎

  26. Compare Linville [1987] on the protection given by one’s having various self-schemas. See also Phoenix [1999] on the “syncretic” formation of identity. ↩︎

  27. On the relation between the two, see the articles by Montefiore, Cohen, Rorty and Jenkins in Montefiore & Vines [1999]. ↩︎

  28. For a theorised version of this claim, see Stainton Rogers et al [1995], chs. 3-5. In some versions, the notion of self-invention is a manifestation of the ideal of authenticity, as to which, see Adorno [1973]; Heidegger [1962], sec. 42; Taylor [1991]; Trilling [1972]; and Tugendhat [1986], ch. 10. ↩︎

  29. On the development of the idea of the self-defining subject, see Taylor [1975], ch. 1. Rorty suggests that choices of identity are attributable to individuals rather than to persons: [1988], p. 91. ↩︎

  30. Compare Sartre’s claim that “the formula ”to be free“ does not mean ”to obtain what one has wished“ but rather ”by oneself to determine oneself to wish“”: [1969], p. 483. See also part 2, ch. 1; Heidegger [1962], sec. 60; and Tugendhat [1986], p. 307. Wilkie Collins’ Poor Miss Finch is someone who joyfully defines herself by her disability: Collins [1908]. ↩︎

  31. Compare Tugendhat [1986], p. 243, on predicates that do not provide self-understanding. ↩︎

  32. Compare Heidegger on “facticity” and “thrownness”: [1962], p. 174. In Sartre there is an uneasy relation between facticity and radical freedom to choose one’s fundamental project: [1969], part 2, ch. 1, and part 4, ch. 1. ↩︎

  33. This is a central theme of those two records of liberation from aspects of Victorian religion, Butler [1933] and Gosse [1949]. ↩︎

  34. See Black [unpublished]. ↩︎

  35. On the connection between commitment and identity, see Taylor [1989], chs. 2 and 11, and White [1991], ch. 10. ↩︎

  36. There are descriptions of this process in Macquarrie [1973], pp. 185-6; Raz [2000], pp. 73-4; and Wiggins [1998], pp. 382-3. ↩︎

  37. See Erikson [1968], pp. 161-3, 209-11, 217; Higgins [1987]; Putnam [1990]; Sartre [1962], part 4, ch. 1. Bildung, as revered in classical Weimar, is an ambitious version of coherence, incorporating the notion of a full cultivation of one’s powers: see Bruford [1962], which also traces the development of the ideal of Bildung in later writers such as Schopenhauer. For the ideal as it appears in the writings of Nietzsche, see Nehamas [1985], chs. 6-7. For a recent version of the ideal, see Glover [1988], part 2. ↩︎

  38. See Rorty [1999]. ↩︎

  39. Compare Nehamas [1985], pp. 191 ff. For a parallel view of integrity, see Flanagan [1991], ch. 4, and Philp [1999]. Nehamas and Philp suggest that the ideals which they discuss have a substantive element. For a broader view of the philosophical tradition, stemming from Plato, which connects inner harmony with the choice of the good, see Morris [1976], p. 150, where it is invoked to question Sartre’s concept of a fundamental project (see notes 3 and 32 above). ↩︎

  40. See Flanagan’s description of identity as a “dynamic integrated system” ([1991], p. 135) and Jenkins’ claim that “”integration“ is not a state but an activity” ([1999], p. 127). ↩︎

  41. For a parallel point on equilibrium of desires, see Black [1994]. ↩︎

  42. See Hogg & Vaughan [1998], ch. 10, and Lindesmith et al [1999], ch. 8. The classic statement is Mead [1934]. A similar idea is central to the quite different tradition of social thought stemming from Hegel: see Taylor [1975], ch. 14. ↩︎

  43. On the rise in the sixteenth century of the modern concept of society, and on the connection drawn then between social and histrionic roles, see Trilling [1972], ch. 1, and Bouwsma [2000], ch. 9. The connection is famously expressed in Jaques’ speech beginning, and in the poem attributed to Ralegh entitled, “All the world’s a stage” (Shakespeare [1951], 2. 7, Ralegh [1961], p. 499), both of which can be read, in the terms of this paper, as representing plural identities over time. In modern social psychology the concept of a role is associated with the work of Goffman (see [1963] and [1969]), but in his hands it arguably supersedes rather than explicates the concept of identity at issue here; see, however, Sabini [1995], pp. 200-202. For reservations about the use of the concept of a role in an account of identity, see Tugendhat [1986], lecture 12. ↩︎

  44. For surveys of the psychological literature on motives and strategies for, and constraints on, changing one’s social group, see Brown [1986], ch. 15, and Hogg & Vaughan [1998], ch. 10. ↩︎

  45. Compare Mead [1934], pp. 167-8. ↩︎

  46. Flanagan is a realist about identity: [1991], pp. 133-9. ↩︎

  47. See Alston [1989]. ↩︎

  48. See Taylor [1985a], chs. 1, 2 and 4; [1985b], ch. 1; and [1989], ch. 2. ↩︎

  49. Compare Rorty [1988], pp. 187-9, where it is argued that there is no special epistemic access to one’s sense of identity. ↩︎

  50. Unrestricted transitivity does not hold for counterfactual conditionals: Stalnaker [1991], p. 38; Lewis [1973], pp. 32-35. ↩︎

  51. On benign and malign forms of self-examination, see Wollheim [1984], ch. 6. ↩︎

  52. Heidegger [1962], p. 32 (Heidegger’s italics); Sartre [1969], p. 74. ↩︎

  53. See note 14 above. ↩︎

  54. James writes of similar cases: “The problem with the man is… what being he shall now resolve to become” (James [1950], vol. 1, p. 288). Putnam pursues James’ idea in her discussion of “critical moments”: cfr. Putnam [1990]. ↩︎

  55. The peculiar horror of an identity crisis is caught by a phrase of Trilling’s: unlike some obvious misfortune which leaves identity intact, it is not met by an “opposing self”: [1955]. See also Laing’s description of “ontological insecurity” ([1965], ch. 3), Travisano’s distinction between “alternation” and “conversion” ([1981]) and Athens’ schema for dramatic self-change, particularly the phase of “fragmentation” ([1995]). On the difference between identity as it is experienced in crisis and in times of stability, see Erikson [1968], pp. 19-22, 165. ↩︎

  56. Compare Snyder’s scale of high and low “self-monitors”: [1974], [1979]. ↩︎

  57. Macquarrie suggests that the existentialist style of thought emerges when man finds his securities threatened, and that it has thus flourished in countries that have suffered social upheavals: [1973], p. 60; see also p. 270. Compare Olafson’s suggestion that existentialism embodies a “voluntarisation” of ethics caused by the transition from pre-industrial to industrial society: [1967], ch. 9. ↩︎

  58. On the relations between identity and ethical judgments, see Black [unpublished]. ↩︎

  59. White [1991], ch. 10, discusses breaks in one’s conception of the good. ↩︎

  60. Wordsworth’s “Surprised by Joy” poignantly captures an experience that contains elements of both phases of grief: [1982], p. 541. ↩︎

  61. Russell [1967-9], vol. 1, p. 147. ↩︎

  62. Compare Erikson [1968], p. 80: “While such realignments may seem to appear [sic] suddenly, they develop slowly”. ↩︎

  63. For a recent discussion of the limits to believing at will, see Noordhof [2001]. ↩︎

  64. Compare Nussbaum’s observation that “the activities and relationships that human beings usually value most have, in more or less every case, a temporally extended structure”: [1994], p. 208. ↩︎

  65. The concept of narrative has been prominent in recent work in this area: see for example MacIntyre [1985], ch. 15; Martin [1998], ch. 5; Nehamas [1985], ch. 5; Ricœur [1992], fifth and sixth studies; Taylor [1989], chs. 2-4 and 17, and [1991], ch. 9; Wollheim [1984], ch. 3. On the connection between narrative and coherence, see Dancy [1993], p. 113, and Bakhurst [2000]. On the various degrees to which people conceive of their lives in narrative terms, see Strawson [1999]. ↩︎

  66. Compare Rorty’s criticism of the proposal to define the concept of a person in terms of narrative: [1989]. ↩︎

  67. Versions of the distinction are drawn in Flanagan [1991], pp. 134-5; Railton [1988]; Ricœur [1992], fifth study; Rorty [1988], chs. 1 and 2; and Tugendhat [1986], appendix to lecture 12. Harré connects this distinction to the one between identity and beliefs about identity: [1989], p. 389. ↩︎

  68. Parfit [1984], pp. 327-8. Parfit distinguishes between persons and selves in describing the case. He writes: “We may regard some events within a person’s life as, in certain ways, like birth or death. Not in all ways, for beyond these events the person has earlier or later selves… The young Russian socialist regards his ideals as essential to his present self. He asks his wife to promise to his present self not to act against these ideals. And, on this way of thinking, she can never be released from her commitment. The self to whom she is committed would, in trying to release her, cease to exist”. Earlier Parfit makes a similar point, but using only the concept of a person: “On one view, certain kinds of qualitative change destroy numerical identity. If certain things happened to me, the truth may not be that I become a very different person. The truth may be that I cease to exist, and the resulting person is someone else” (p. 202). Parfit’s sense of “qualitative identity” is not, however, the one at issue in this paper, but the thinner one in which two things are exactly alike. Rorty describes such situations by distinguishing the person from the individual: [1988], p. 51. ↩︎

  69. Williams [1973] is an account in terms of bodily continuity, Locke [1965], 2. 27. 9, is one in terms of psychological continuity. ↩︎

  70. Mackie [1974], p. 62. ↩︎

  71. Compare Wilkes’ argument for the need for a temporally extended self: [1999], pp. 27-9. ↩︎

  72. On Davidson’s view, mental states themselves are integrated with each other in the sense that they are attributed holistically and on the basis of normative principles of rationality: [1980], [1984], [2001], passim. ↩︎

  73. Compare Evnine [1989], which makes a parallel point about the existence of mental states. ↩︎