John Biro
IV. PERSONAL IDENTITY
The most general philosophical question about the mind had always been an ontological one: What is it? But Hume’s eschewal of speculative metaphysics leads him to substitute for this question two others to which there are clear answers: what kind of thing is my belief about when I believe that I am a self, something that can be reidentified as the same thing at different times; and, what is the source of my belief that I am such a thing? It is important to see that these questions are being asked from the first-person point of view, thus rendering irrelevant the easy answer that to believe in one's identity is to believe that one is, or is at least associated with, an enduring body. Re-identification of a body depends only on general criteria of identity applicable to physical objects. While Hume is interested in those criteria, and indeed appeals to considerations involving them in attempting to clarify the concept of identity in general, when he comes to the topic of personal identity and tries to explain "the identity, which we ascribe to the mind of man," he knows that he cannot simply rely on those criteria, since they involve activities (remembering and associating, for example), and hence the identity, of a mind, the very thing whose identity is at this point in question (T 1.4.2, 200-204; 1.4.6, 253-9). For talk of a mind doing something to make sense, there must be a temporally extended item of some sort denoted by the term mind (and by the pronoun I), one to which the predicate "same at time2 as at time1" can be applied. [16] Given this, before an account of my belief in the identity of material objects in terms of various mental activities attributed to me can be intelligible to me, I must believe that I am a self: I must believe that I am a subject to whom such activities, taking place over time, as activities must, can be attributed. Thus, as Hume recognizes, the first belief standing in need of an analysis and a genetic account is the belief one has in one’s own identity. [17] He therefore gives an account of what one believes when one has that belief and of how one could come by a belief of that sort. That in his analysis he must make use of the more general concepts which
[16] Hume’s general account of identity involves thinking of the
mind as surveying objects and "tracing( the succession of time"
)T 1.4.2, 201) -- something that seems to presuppose its identity
over time. So, it may seem, does talking of one’s believing
and re-identifying, as I just have. It may seem to beg the question against
the very sceptic Hume is often taken to be, namely, one who doubts that
there is a self. Does such talk not already imply, by virtue of
its grammar alone, that there is one?
One may be tempted to say that if the sceptic were right, "he"
could not state "his" view. However this may be, Hume’s
discussion is better seen, as I have suggested here, as one about what
kind of thing the self is, rather than one about whether there is one.
(The similarity with his treatment of belief in body, discussed earlier,
should not be overlooked.) Once we understand his answer -- essentially,
that the self believed in is a complex object united by certain crucial
relations -- we can also see that there is no puzzle about how it can be
the subject of so-called intentional verbs such as those used to attribute
actions. For additional discussion of these issues, see my "Hume on
Self-Identity and Memory," The Review of Metaphysics 30(1976):
19-38, and "Hume’s Difficulties with the Self," Hume
Studies 5(1979): 45-54.
[17] This is why Hume insists on the need to account for one’s
belief in one’s self first: that belief is required for one’s
belief in external objects. See T 1.4.2, 189.
[p47]
themselves presuppose having such a belief can obscure this all-important point, as can, on occasion, Hume’s language. But when he asks whether "in pronouncing concerning the identity of a person, we observe some real bond among his perceptions" (T 1.4.6, 259), he must be taken to be talking about a person pronouncing on his own identity on the basis of observing his own perceptions. There is no observing another’s perceptions, as there is another’s body. So, for an answer to the question that must be most basic -- How do I come to think of myself as a self? -- I must turn inward, I must look to see what there is in my experience to lead me to think of myself as the same person or mind over time. That I do so is a datum, one Hume is seeking to explain within the new scientific framework he has adopted. [18]
Thus Hume should be seen as having a theory about a certain fundamental belief that underlies and is presupposed by all other beliefs. The theory has two parts: an explication of what I think when I think of myself as a self (or a mind, a person); and an explanation of how I can come to think that I am such a thing on the basis of my experience. The answer philosophers, especially those in the Cartesian tradition, commonly give to the first question is that to think that one is a self is to think that one is a simple substance, one that endures essentially unchanged in spite of many accidental changes, in particular, changes in what perceptions one has. On this view, I am the owner of the many experiences I undergo, but I am distinct from those experiences, and what I am, in the metaphysically relevant sense, is independent of what they are.
But why should we believe such philosophers? Hume’s negative arguments are intended to show that there is no good reason to do so. No demonstrative argument can prove the existence of such an entity (any more than of any other), and no evidence can be found in experience, the only source of non-demonstrative evidence for anything. [19]
Had Hume stopped here, we could say with some justice that his position was a sceptical one, though it would still be worth noting that his would have been a scepticism only about the particular philosophical doctrines he was examining. From his arguments against that doctrine, however, nothing follows concerning the prospects for constructing some other theory about the content and source of one’s belief in one’s identity. And, indeed, Hume does not
[18] The relation between the first-person and the second-person elements
in Hume’s method is discussed below, Part VI, and in my "Hume’s
Methods," a paper presented to the 1992 Hume Conference.
[19] In the second book of the Treatise, concerned with the passions
(with what we would today call moral psychology), Hume insists that "the
idea of ourselves is always intimately present to us" (T 2.2.4,
354). The appearance of inconsistency here evaporates once we remind
ourselves that the kind of self-awareness Hume requires in his account
of the passions does not entail anything about the nature of the object
we are aware of. We should distinguish the question of what the self is
from the question of what role it plays in our emotional life. The
two questions are independent; having answered the first, Hume can quite
consistently expect the reader to be aware of this answer in giving his
answer to the second. For more on the passions, see Terence Penelhum, "Hume
s Moral Psychology," this volume.
[p48]
stop with negative arguments against the substantial view of the self. He goes on to give an alternative account of how the belief in personal identity can be based on experience.
On Hume’s alternative analysis -- his famous "bundle" theory -- a mind (self, person) is a collection of perceptions related to each other in certain ways so as to constitute a complex entity to which identity of one sort, though not of another, may be intelligibly and truly ascribed. The sort of identity that is appropriate to such an entity is what Hume calls "imperfect identity," thus distinguishing it from "perfect," (or "strict") identity, a property only simple and unchanging entities possess (T 1.4.6). Having argued that nothing in one’s experience answers to a belief that one is such a simple and unchanging entity -- "when I enter most intimately into what I call myself, I always stumble on some particular perception or other" (T 1.4.6, 252) -- Hume goes on to show that the same experience can nevertheless serve to explain how one comes to believe in one’s identity over time. This is a belief each of us has, and it is central to the common-sense picture of the world that we all accept when uncorrupted by bad philosophy. It is a belief implicated in all our other beliefs, a belief without which arguably even the sceptic’s position could not be understood.
What, above all, unites the perceptions that collectively constitute a mind or self is memory, and the natural relation of causation with which memory is inextricably bound up. Memory is in one way the more fundamental here, since without it the natural relation would not arise: "Had we no memory, we never shou’d have any notion of causation, nor consequently of that chain of causes and effects, which constitute our self or person" (T 1.4.6, 261-2). One reason for this priority of memory is that Hume’s account of causation requires that I remember the constant conjunctions between a pair of events if experience of such a conjunction is to lead me to think of them as cause and effect – that is, leads me to expect the second to always (where this means, necessarily) follow on the first. The mere occurrence of such a constant conjunction in my experience would not suffice. Suppose that my experience did include such repeated conjunctions of two events, A and B, but that it did not also include perceptions that are rememberings of previous co-occurrences of the pair. This might be sufficient for giving rise, upon a fresh experience of A, to an expectation of B. (There are reasons to
[p49]
doubt this: What would be the mechanism? What, in the absence of memory, would distinguish the umpteenth experience of A from the first?) Still, even that expectation would not be enough to give rise to the idea of necessary connection, whose genesis Hume is trying to explain. I may still lack the felt "determination of the mind," which, in that explanation, serves in lieu of an impression of necessary connection and gives rise to the idea of such a connection. Thus my expectation of B, while caused by its constant conjunction with A, would, in the absence of a memory of that constant conjunction, fail to be an expectation of an effect.
It is, then, the presence of memories among my perceptions that is the ultimate source of the idea that I am a temporally extended being. These memories need not be veridical: what matters is that they are what philosophers since Franz Brentano have meant by intentional, in the sense of referring to, being about, other things -- here other perceptions -- experienced at an earlier time. In the full story, forward-looking perceptions -- anticipations -- also play a role, as does the inertial tendency we have already seen at work elsewhere. "But having once acquir’d this notion of causation from the memory, we can extend the same chain of causes, and consequently the identity of our persons beyond our memory, and can comprehend times, and circumstances, and actions, which we have entirely forgot, but suppose in general to have existed" (T 1.4.6, 262). [20]
One of the chief insights emerging from this discussion of personal identity, with implications that go far beyond that topic, is that an entity of the sort Hume takes the mind to be (complex, dynamic, ever-changing) can be thought of as an active agent in the formation of our beliefs about everything (including even, as we have seen, the formation of the belief in its own identity). A generalization of this insight underlies virtually all of Hume’s analyses of the concepts we employ in thinking about the world and our relation to it. Most important, it drives all Hume’s hypotheses about how we come to believe what we believe, whatever the content and object of our belief. From the earliest parts of the Treatise, with its picture of complex ideas being generated from simple ones, through the account of the nature of belief (as well as of belief formation, of belief transition by way of the principles of association, of causal beliefs as expectations produced by experience and habit), to the practical philosophy (where almost every interesting principle of
[20] For further details, see my "Hume and Cognitive Science,"
History of Philosophy Quarterly 2 (1985): 257-74, reprinted in Historical
Foundations of Cognitive Science, ed. J-C. Smith (Boston, 1990).
[p50]
moral psychology, and of ethics, politics, and aesthetics involves it), the constant activity of this mind is what dominates the story and ties it together into a unified and coherent whole.
[end]