Personal identity

DOI: 10.4324/9780415249126-V024-1
Version: v1,  Published online: 1998
Retrieved March 25, 2019, from

3. Psychological and mixed criteria

I take the bionic brain example to undermine not just the brain criterion, but also the narrow version of the psychological criterion – I survive with a bionic brain, yet the cause of my psychological continuity is abnormal. Hence, it might be thought, the combined effect of these conclusions is to push us towards the wide version of the psychological criterion. Indeed some readers may have wondered why I made no appeal in the first place to examples such as Bernard Williams’s brain-state transfer device or Derek Parfit’s teletransporter in order to establish the wide version of the psychological criterion.

What are these examples? First, Williams describes a device which can wipe a brain clean, while recording all the information stored in the brain. This machine can then reprogramme another clean brain. The machine can thus preserve psychological continuity in the absence of any physical continuity. Second, Parfit has made much use of the Star Trek fantasy of teletransportation. In this scenario, a physical and psychological blueprint is made of a person. The person is then painlessly destroyed. The blueprint is transmitted to another location where, out of different matter, an exact physical and psychological replica of the original person is made.

Some philosophers have claimed that, in both these examples, the original person is identical to the later person, and so concluded that physical continuity is not necessary for personal identity over time. Other philosophers have taken the opposite view, and concluded that physical continuity is necessary for personal identity over time. But these are judgments in which theory is being allowed to dictate intuitions. In truth, there simply is no general agreement about whether the original is the same person as the replica, in cases in which there is psychological continuity but no physical continuity.

Further, there appears to be a decisive objection to the wide version of the psychological criterion – the criterion which holds that personal identity over time consists in psychological continuity, however that continuity is caused. (A defender of the wide version would agree with my verdicts about the counterexamples to the body and brain criteria, and would also regard the normal operation of the brain-state transfer device and the teletransporter as identity-preserving.) Imagine that I step into the teletransporter booth. My psychophysical blueprint is constructed, and sent to another location, where a replica is created. Unfortunately, the machine malfunctions and fails to destroy me. I step out of the booth, intrinsically no different from when I went in. In this case, we have no hestitation in judging that I continue to exist in the same body, and therefore that the replica is not me. But both me-later and my replica stand to me-earlier in the relation of psychological continuity. If the cause of that continuity is deemed to be irrelevant to personal identity, as in the wide version of the psychological criterion, then it ought to be the case that both later candidates have an equal claim to be me. Yet, as we have seen, we strongly believe that I am identical to the later person who is physically and psychologically continuous with me. Consequently, the wide version of the psychological criterion cannot be correct.

I conclude that the best account of personal identity over time is provided by the mixed criterion. We have seen that neither continuity of body nor brain (nor, by extension, the continuity of any other human organ) is a necessary condition for personal identity over time. But we should not conclude from this that psychological continuity, whatever its cause, is sufficient for personal identity over time. As just noted, that thesis does not accord with our intuitions. The most consistent and plausible view that can be recovered from our core set of common-sense judgments appears to be the following: psychological continuity is necessary for personal identity over time; a sufficient condition of personal identity over time is not psychological continuity with any cause, but psychological continuity with a cause that is either normal or continuous with the normal cause (this is why I continue to exist with a bionic brain). One might well ask: why do we have just this concept of personal identity and not some other (such as the wide version of the psychological criterion)? Here, as with other conceptual analyses, there may be no non-trivial answer to this question.

Our discussion thus far has made a certain simplification. The counterexample to the wide version of the psychological criterion exploited the fact that relations of psychological continuity are not logically one-one. In that example, one of the streams of psychological continuity did not have a normal cause. However, it is logically possible for a person at one time to be psychologically continuous with two or more later persons, even when both streams of psychological continuity have their normal cause (that is, the continued existence of the brain hemispheres). But the relation of identity is logically one-one: I cannot be identical to two distinct people. It seems, therefore, that the sufficient condition for personal identity endorsed in the previous paragraph will have to be modified, unless either such branching is impossible or the possibility of branching can be redescribed so that it does not conflict with our sufficient condition. The problem raised by the possibility of branching continuities is known as the problem of fission.

Citing this article:
Garrett, Brian. Psychological and mixed criteria. Personal identity, 1998, doi:10.4324/9780415249126-V024-1. Routledge Encyclopedia of Philosophy, Taylor and Francis,
Copyright © 1998-2019 Routledge.

Related Articles