SELF

In its normal use the English expression "self" is not even quite a word, but something that makes an ordinary object pronoun into a reflexive one (e.g., her into herself ). The reflexive pronoun is used when the object of an action or attitude is the same as the subject of that action or attitude. If I say Mark Twain shot himself in the foot, I describe Mark Twain not only as the shooter but as the person shot. In this sense "the self" is just the person doing the action or holding the attitude that is somehow in question. "Self" is also used as a prefix for names of activities and attitudes, identifying the special case where the object is the same as the agent: self-love, self-hatred, self-abuse, self-promotion, self-knowledge.

"The self" often means more than this, however. In psychology it is often used for that set of attributes that a person attaches to himself or herself most firmly, the attributes that the person finds it difficult or impossible to imagine himself or herself without. The term identity is also used in this sense. Typically, one's sex is a part of one's self or one's identity; one's profession or nationality may or may not be.

In philosophy the self is the agent, the knower and the ultimate locus of personal identity. If the thought of future reward or punishment is to encourage or deter me from some course of action, I must be thinking of the person rewarded as me, as myself, as the same person who is now going to endure the hardship of righteousness or pass up the enjoyments of sin in favor of this ultimate reward. But this same self comes up in much more mundane transactions. If I pick up the cake and shove it in this mouth rather than that one, is it not because I think it will be me, the very same person who picks up the cake, that will have the pleasure of tasting it?

A straightforward view of the self would be that the self is just the person and that a person is a physical system. This view has been challenged on two fronts. First, the nature of freedom and consciousness has convinced many philosophers that there is a fundamentally nonphysical aspect of persons. The second challenge stems from puzzling aspects of self-knowledge. The knowledge we have of ourselves seems very unlike the knowledge we have of other objects in several ways, and this has led some philosophers to rather startling conclusions about the self. In his Tractatus, Ludwig Wittgenstein tells us that "I am my world" and that "'the world is my world'" (1961, 5.63, 5.641). This should lead us to the rather surprising conclusion that I am the world, or that at least Wittgenstein was. He draws at least one conclusion that would follow from this: "at death the world does not alter, but comes to an end."

The contemporary philosopher Thomas Nagel has been led to a possibly less radical but still quite dramatic view. According to Nagel, when he says "I am Tom Nagel," at least in certain philosophical moods, the "I" refers to the "objective self," which is not identical with but merely contingently related to the person Tom Nagel. This self could just as well view the world from the perspective of someone other than him (Nagel, 1983). We need to discuss the puzzling features of self-knowledge that give rise to such views.

Self-Knowledge

"Self-knowledge" seems to have a straightforward meaning: cases of knowledge in which the knower and the known are identical. But this does not seem sufficient. The philosopher Ernst Mach once got on the end of a bus and saw a scruffy, unkempt, bookish-looking sort of person at the other end. He thought to himself,

1. That man is a shabby pedagogue.

In fact, Mach was seeing himself in a large mirror at the far end of the bus. He eventually realized this and thought to himself:

• (2) I am that man.
• (3) I am a shabby pedagogue.

Now consider Mach at the earlier time. Did Mach have self-knowledge? In our straightforward sense it seems that he did. He knew that a certain person was a shabby pedagogue and, furthermore, that person was him. The knower and the known were the same. But this is not what we mean by self-knowledge. Self-knowledge is something Mach really had only when he got to step (3), when he would have used the word I to express what he knew.

Self-knowledge seems peculiar. First, it seems "essentially indexical." Statement (3) expresses self-knowledge because of the word I ; it is hard to see how Mach could have expressed self-knowledge without using the first person. If he said "Mach is a shabby pedagogue," he would be claiming to know only what everyone else may have known. It does not seem that there is any objective characterization D of Mach, such that knowing that he is a shabby pedagogue amounts to knowing that D is a shabby pedagogue (Castañeda, 1966, 1968; Perry, 1990, 1993).

Secondly, we seem immune to certain sorts of misidentification with respect to self-knowledge. If we learn, in certain ways, that someone is in pain, then we cannot miss the fact that it is we who are in pain. That is, if Mach discovers that he has a headache in the ordinary way that a person discovers she has a headache, he can scarcely be wrong about who has the headache, if the range of choices is "I/you/that man," and so forth. Of course he can be wrong if the range of choices is "Mach/Freud/Wittgenstein," and so on, for he might not realize which of those people he is (Shoemaker, 1984).

Third, self-knowledge seems to play a unique cognitive role. If Mach desires that he do so and so, and believes that he can do so and so by executing such and such a movement, then he will execute that movement without further ado (Perry, 1990).

Agent-Relative Knowledge

At least some of these peculiarities of self-knowledge can be explained by taking self-knowledge to be a species of agent-relative knowledge. There are two quite different ways of cognizing objects (people, things, places, and times). We can think of them via their relationship to us, the role they are playing in our lives at the moment of thought: the object I see; the present moment; the place I'm at; the person I'm talking to. We need to think about things in the first way, when we are picking up information about them perceptually or interacting with them, since ways of knowing and acting are tied to these agent-relative roles. I can learn about the here and now by looking; I can learn about the person I am talking to by asking questions, and so forth.

But these agent-relative roles cannot be our only ways of thinking about objects of more than passing interest to us. Different objects play the same agent-relative roles at different times, and at any given time many of the objects we wish to retain information about will not be playing any agent-relative role for us. And we cannot accumulate information along such roles. Suppose I am in Tokyo on Tuesday but return to Palo Alto on Friday. From the facts that on Tuesday I truly thought "Japanese is the official language here " and on Friday I truly thought "Senator Stanford used to live near here " it does not follow that there is some place where Japanese is the official language and near which Senator Stanford used to live.

In order to retain and accumulate information about objects, to construct and maintain a coherent picture of the world, we need to have a way of conceiving of objects as existing independently of us, as occupying and then ceasing to occupy various agent-relative roles. That is, we need objective ways of thinking about objects. We keep track of them by names or descriptions that do not depend on their relationship to us: Cordura Hall, 4 p.m., June 23, 1995, the southernmost town in Santa Clara County, Aurora Fischer. These serve as our fundamental ways of thinking about those objects. Recognition consists in connecting our objective ways of thinking of objects with the roles those objects play at a given moment. Consider the knowledge I might express with "Today is July 4." This is knowledge that a certain day, objectively conceived ("July 4"), is playing a certain role in my life; it is the present day, the day on which the thinking and speaking take place. This kind of knowledge, "knowing what day it is," is quite crucial to successful application of other, more objective knowledge. If I know that the party is on July 4 and know that today is July 4, then I will form the right expectations about what the day will be like.

Similarly, I may be in Kansas City and know that Kansas City is a good place for a steak dinner. But if I do not know that I am in Kansas City, if I do not realize that Kansas City is playing the "here" or "this city" role in my life at this moment, I will not be able to apply the knowledge that Kansas City is a good place for a steak dinner.

And again, I may know that Aurora Fischer has important information about my schedule, but unless I realize that the person I am talking to is Aurora Fischer, I will not apply this information and say, "Can you tell me where this afternoon's meeting is?"

These kinds of knowledge are, like self-knowledge, "essentially indexical." We use now and today to express our knowledge of what time it is and here to express our knowledge of where we are. These locutions are not reducible to names or objective descriptions, just as I was not. I cannot express what I say when I say, "The meeting starts right now" by saying "the meeting starts at D " for any description D of the present moment.

We are also immune to certain sorts of misidentification when we use certain methods of knowing. There is a way of finding out what is going on around one, namely opening one's eyes and looking (Evans, 1985). Now when one learns what is going on in this way, one can hardly fail to identify the time at which this is happening as now and the place as here. And finally, the forms of thought we express with now and here seem to have a unique motivational role. If I want to do something here and now, I will simply do it.

Self-Knowledge as Agent-Relative Knowledge

"Self" is really the name of such an agent-relative role, that of identity. As with other agent-relative roles, there are special ways of knowing and acting that are associated with identity. If Mach had wished to know, during the interval while he was confused, if the shabby pedagogue he was seeing had lint on his vest, he would have had to walk over to him and look. If Mach had wanted to know if he himself had lint on his vest, he could have simply lowered his head and looked. Had he done this, he would have had no doubt about whom the lint was on. If Mach found lint and wanted to brush it off, he would engage in self-brushing, a quick movement of the hand across one's front that each of us can use to remove lint from our own vest and no one else's.

Unlike the other agent-relative roles, identity is permanent. I will talk to many people, be in many places, live through many times in the course of my life. But there is only one person I will ever be identical with, myself. Hence, accumulation along "I" is valid, unlike accumulation along "here" or "now" or "that man."

Earlier we rejected the straightforward account of self-knowledge, as knowledge about a person by that very person. Now we can put forward an alternative. Self-knowledge is knowledge about a person by that very person, with the additional requirement that the person be cognized via the agent-relative role of identity. This agent-relative role is tied to normally self-informative methods of knowing and normally self-effecting ways of acting. When these methods are employed, there will be immunity of misidentification as to who is known about, or who is acted upon.

This role can serve as a person's fundamental concept of himself or herself. In this way our self-conceptions have structures that are different from our conceptions of other individuals of importance to us. If we understand the special way in which a person's self-knowledge is structured, we do not need to postulate anything but the person himself or herself for the knowledge to be about.

See also Consciousness; Freedom; Identity; Indexicals; Mach, Ernst; Nagel, Thomas; Personal Identity; Philosophy of Mind; Reduction; Self-Knowledge; Wittgenstein, Ludwig Josef Johann.

Bibliography

Barnes, A. Seeing through Self-Deception. Cambridge, U.K.: Cambridge University Press, 1997.

Bermúdez, J. The Paradox of Self-Consciousness. Cambridge, MA: MIT Press, 1998.

Brewer, B. "Bodily Awareness and the Self." In The Body and the Self, edited by J. Bermúdez, A. Marcel, and N. Eilans. Cambridge, MA: MIT Press, 1995.

Cassam, Q. Self and World. Oxford: Oxford University Press, 1997.

Cassam, Q., ed. Self-Knowledge. Oxford: Oxford University Press, 1994.

Castañeda, H.-N. "'He': A Study in the Logic of Self-Consciousness." Ration 8 (1966): 130–157.

Castañeda, H.-N. "On the Logic of Attributions of Self-Knowledge to Others." Journal of Philosophy 65 (1968): 439–456.

Evans, G. "Understanding Demonstratives." In Collected Papers. Oxford: Clarendon Press, 1985.

Evans, G. The Varieties of Reference. Oxford: Clarendon Press, 1982.

Gallois, A. The Mind Within, The World Without. Cambridge, U.K.: Cambridge University Press, 1996.

Gertler, B., ed. Privileged Access: Philosophical Accounts of Self-Knowledge. Aldershot, U.K.: Ashgate, 2003.

Hill, C., ed. Philosophical Topics: Introspection 28 (2000).

Lowe, E. J. Subjects of Experience. Cambridge, U.K.: Cambridge University Press, 1996.

Moran, R. Authority and Estrangement: An Essay on Self-Knowledge. Princeton, NJ: Princeton University Press, 2001.

Nagel, T. "The Objective Self." In Knowledge and Mind, edited by C. Ginet and S. Shoemaker. New York: Oxford University Press, 1983.

Neisser, U., and D. Jopling, eds. The Conceptual Self in Context: Culture, Experience, Self-Understanding. Cambridge, U.K.: Cambridge University Press, 1997.

Perry, J. The Problem of the Essential Indexical. New York: Oxford University Press, 1993.

Perry, J. "Self-Notions." Logos 11 (1990): 17–31.

Shoemaker, S. The First-Person Perspective and Other Essays. Cambridge, U.K.: Cambridge University Press, 1996.

Shoemaker, S. Identity, Cause, and Mind. Cambridge, U.K.: Cambridge University Press, 1984.

Strawson, G. "The Self." Journal of Consciousness Studies 4 (1997): 405–428.

Wittgenstein, L. Tractatus Logico-Philosophicus. Translated by D. F. Pears and B. F. McGuinness. London: Routledge and Kegan Paul, 1961.

Wright, C., B. Smith, and C. Macdonald, eds. Knowing Our Own Minds. Oxford: Clarendon Press, 1998.

John Perry (1996)

Bibliography updated by Alyssa Ney (2005)