To: Tony L.
From: Geoffrey Klempner
Subject: Williams' body-swap thought experiment
Date: 20 July 2006 10:04

Dear Tony,

Thank you for your email of 8 July, with your essay for the University of London Diploma in Philosophy Introductory module, in response to the question, 'Is Williams's thought experiment best described as two persons swapping bodies? If not, why not?'

This is a very good piece of work, which on the whole gives an accurate representation of Williams's argument. You have not followed Williams's treatment of the thought experiment slavishly but contribute views of your own, which is also good.

The 'punch line' of Williams's article occurs in the paragraph which begins on p. 165 and continues to p. 166. William's main objective in this article is to question the assumption that so called 'first personal' and 'third personal' views of personal identity problem cases coincide with 'mentalistic' considerations, and considerations of bodily continuity respectively.

It is not quite accurate, therefore, to say that the 'purpose of the experiment is to clarify what constitutes personal identity'. It would be closer to say that it's purpose is to bring out our philosophically pre-reflective beliefs about personal identity. It turns out - and this is the core of Williams's case - that different descriptions of the very same experiment elicit contradictory beliefs. This poses a worrying challenge for any philosopher attempting to analyse the notion of personal identity in terms of necessary and sufficient conditions.

However, the essay title is squarely focused on the question of personal identity. Regardless of the variations in how we may be tempted to describe the situation, what IS the best description of the case from a philosophical point of view?

A point to make, however, is that the philosophical answer must respect our pre-reflective intuitions, or, when it clashes with them, it must be possible to offer a suitable explanation of why our intuitions are 'wrong'.

If you agree with the swapping bodies description, then you need to explain why we get confused when the experiment is described in the second way. If you disagree, then you need to explain why we get confused when the experiment is described in the first way.

You have chosen neither option, preferring to leave the matter unresolved, which is OK. As you point out, Williams has himself given some support to this by suggesting that the experiment has been made deliberately 'neat' in order to make the question easier to answer.

I think that you may have been a little unfair (unless I am misreading you) to Williams in your discussion of the second way. The question, as Williams poses it, 'Will I be tortured?' is a question A is answering now. Whatever the A-body-person will say when he has no memories, or has B's memories, is irrelevant. This will happen to ME, one is tempted to say, pointing to one's own body. That's Williams's gut feeling, and he expects it to be our gut feeling too.

This contrasts with the first way of describing the thought experiment, which focused on outcomes, and what the resulting A-body-person and B-body-person say after the experiment has actually been performed.

Is Williams right about this gut feeling? There are other cases, which Williams does not discuss in this extract which put greater pressure on the 'gut feeling' - for example, bodily fission, where I discover that I am not human but a Martian who is about to undergo his yearly fission (one half will get 10000 galactic credits and the other half will be tortured).

Your characterization of the 'conventionalist' approach is inaccurate. The point of the conventional answer is not that we 'wait until the case arises and see how we feel', but rather that we chose the best, or least worst candidate in a similar way to a decision that one might make about a legal inheritance in a court of law. It is not about feeling but making a decision because a decision has to be made. I can't decide about my feelings on the conceptually questionable prospect of torture on that basis.

As I implied above, I don't think Williams is trying to push a particular view of the 'necessary and sufficient conditions for personal identity' in this extract. But the examiner is, effectively, asking for this. While refusing to decide is a defensible option - at the end of the essay you give your own description of the situation in terms which avoid answering the question - I would prefer a straight answer.

My straight answer would be that it is. The memory criterion, suitably reinforced with the 'right kind' of causal continuity, has to be respected. Your examples of ASBOs and disability allowances backs this up. A disability allowance is given for cases of need, and this is the basis on which 'desert' is decided. Whoever ends up with the disabled body deserves the allowance. By contrast, an ASBO is a punishment meted out to an individual who is guilty of antisocial behaviour, a matter decided just as Locke said it should be, in terms of a person's character and consciousness.

However, I would qualify this answer in the light of possibilities - like fission - which Williams has not considered in this extract. Elsewhere in 'Problems of the self' Williams uses these kinds of case to discuss the worrying prospect of that the notion of a person could become a 'universal' rather than an 'individual' concept.

All the best,

Geoffrey