Personal identity

2. Psychological continuity theories

John Locke (1689) makes an argument very like the one offered above to defend the conclusion that personal identity should be defined in terms of sameness of consciousness rather than in terms of the sameness of either immaterial soul or physical substance (i.e. the human body). Locke uses imaginary cases in which sameness of consciousness is separated from both sameness of body (e.g. a case where the consciousness of a prince enters the body of a cobbler, bringing with it all of the prince’s memories and erasing all those of the cobbler) and sameness of soul (e.g. someone with the soul of Nestor or Thersites at Troy but no consciousness of any of their actions), and concludes that in each case we will see that the person goes with the consciousness. The person in the cobbler’s body with the prince’s psychological life is the prince, not the cobbler, he says, and the person with Nestor’s soul but no consciousness of his life is not Nestor.

While Locke’s arguments have a great deal of intuitive appeal, it is not clear exactly what his positive view amounts to because it is not evident what he means by ‘sameness of consciousness’. This phrase is usually taken to stand for autobiographical memory, and Locke’s view is interpreted as the view that someone’s ability to remember past experiences makes them the person who had those experiences. There are a host of very serious objections that can be raised against this simple memory theory. It seems to imply, for instance, that if I claim to remember leading the troops at Waterloo that makes me Napoleon, or that a murderer who forgets committing a murder did not commit it. Because of these difficulties, there is general consensus that the view as it stands is not viable. Over the last several decades, however, a number of philosophers have updated and revised Locke’s original insight to address these problems, yielding new and more complex psychological accounts of personal identity.(For examples and descriptions of this view see e.g. Lewis 1976; Parfit 1984; Perry 1976; Shoemaker 1984; Noonan 1989.)

Updated psychological continuity theories of identity define personal identity in terms of overlapping chains of psychological connections (i.e. the connection between a memory and the experience of which it is a memory, between an intention and the action that carries it out, or between different temporal portions of a continuing belief, desire, or trait) that are appropriately caused. These theories thus differ from the memory theory attributed to Locke in three significant ways: first, they include psychological connections other than memory; second, they require these connections to be appropriately caused; and third, they require overlapping chains of connections rather than direct connections.

The first change is made in the interest of generally increasing plausibility. Psychological theorists argue that memory is not the only important psychological connection, and so other kinds of connections should be included in an identity criterion. The second change is made to deal with problems of delusional memory-like experiences of the sort described above. Psychological theorists must avoid the consequence that someone who has delusions of having led the troops at Waterloo, no matter how sincere, is therefore identical to Napoleon. The most straightforward way to avoid this result is to say that only genuine memories constitute personal identity. The problem with this strategy, however, is that the most obvious way of distinguishing between genuine memories and delusions is by saying that in the former case but not the latter the putative memory is of an experience that the rememberer actually had. This approach, however, defines personal identity in terms of genuine memory and genuine memory in terms of personal identity, making the view viciously circular. To avoid this difficulty psychological theorists use the requirement of appropriate cause to distinguish between delusions and the kinds of memory experiences that actually contribute to identity. Napoleon’s memory-like experiences of Waterloo are causally connected to the events at Waterloo in a way that the delusional memory-like experiences are not. Psychological theorists call appropriately caused memory-like experiences ‘quasi-memories’. Those who make use of them point out that someone might in theory have a nondelusional quasi-memory of an experience they did not have (e.g. if neurosurgeons were able to transplant memory traces). This means that quasi-memory does not presuppose identity, so that a definition of identity in terms of quasi-memory (and quasi-intentions, quasi-beliefs, etc.) avoids circularity.

The third change is made to avoid problems with transitivity. A man of 40, while taking a vacation, might remember a vacation he took when 20, and a man of 70 remember the vacation of the man of 40 but have no recollection of the vacation the 20-year-old took. On a straight memory theory, this makes the man of 70 the same person as the man of 40, and the man of 40 the same person as the man of 20. From these facts and the transitivity of identity (if A=B and B=C then A=C) it follows that the man of 70 is the same person as the man of 20. However, since the memory theory says that someone must remember an earlier person’s experiences in order to be that person, it also implies that the man of 70 is not the same person as the man of 20. This absurd conclusion can be avoided if identity is defined in terms of overlapping chains of psychological connections rather than only in terms of direct connections. This alteration allows the view to say consistently that the 70-year-old is the same person as the 20-year-old.