To: Matan A.
From: Geoffrey Klempner
Subject: Pythagorean reincarnation theory and personal identity
Date: 23 September 2003 11:31

Dear Matan,

Thank you for your e-mail of 16 September, with your second essay for the Ancient Philosophy program, in response to the question, 'Critically discuss the Pythagorean theory of reincarnation. What issues does the idea of reincarnation raise concerning the nature of the self, and the definition of personal identity?'

After your introduction (fine for a 2000 word essay but too long for an essay of 800 words!) you pose two questions, 'first, what makes up the person - the self - that survives the death of the body?...second, what keeps its personal identity over a succession of incarnations?'

However, it is not clear that in your formulation you have succeeded in capturing two questions here, rather than only one. We are looking for a difference (if indeed there is one) between asking about the 'nature of the self', and asking about the 'definition of personal identity'. The fact that you give exactly the same answer to both questions - 'memory' - suggests that you have not succeeded in conceptualizing the crucial distinction.

The point will be clearer if we first look at John Locke's account of personal identity, which you briefly allude to towards the end of your essay. In a famous section of his 'Essay', Locke takes the reader through a thought experiment where the souls of a prince and a pauper are switched, but where the prince's soul simultaneously acquires the pauper's memories and the pauper's soul acquires the prince's memories. Locke says that the soul switch is completely *irrelevant* to the personal identities of the prince and the pauper, because memory is the only thing that counts. The prince wakes up still thinking he is the prince and the pauper still thinks he is the pauper, despite the soul switch. And Locke says they are both right! (So you are wrong when you say that according to Locke, 'the soul of a man alone, divorced from the matter, is to define personal identity'.)

Let's ask, what according to Locke's view of personal identity, is the 'nature of the self'? The self is not a soul, but then what is it? Locke doesn't have an answer to give, and this is significant.

The issue, which Locke fails to address, is what it is that makes a memory genuine. You are right to raise problems with the reliability of memory, but the problem is not simply that we can never be sure when a memory claim is correct or not. Let's say Pythagoras remembers, when he was Euphorbus, having a conversation with a pretty girl in an olive grove. We can never know whether this incident was real or imagined, because there is no corroborating evidence. But that's not what raises problems for personal identity. Let's take an incident which is well corroborated. Euphorbus once gave a famous speech in the market place about valour (I am making these examples up). Pythagoras 'remembers' making the speech. We all know the speech took place. But what distinguishes this from merely *seeming* to remember? It seems that the only thing sufficient to make the difference is that *Pythagoras was really there*. But now the definition of personal identity in terms of memory has become completely circular! (This is the essence of the objection raised by Butler to Locke's account of personal identity.)

For Pythagoras to 'really have been there' giving the speech it is not sufficient that he *seems to remember* being there, and that his seeming memory corresponds with the facts. An extra ingredient is required, which is the real continuity of something which holds, or embodies, or encodes the memory. (I am being deliberately vague here, because this is the big question.) A metaphysical account of the nature of the self is what you would need to fill this gap, by providing an account of the difference between a genuine memory, and a merely seeming memory which corresponds with the known facts.

Arguably, the concept we are looking for is the notion of a *cause*. A genuine memory is distinguished from a true but merely seeming memory by the kind of causal link involved. A Cartesian immaterial soul would serve this purpose. The problem is, as Locke realized, that we have no way to prove that a Cartesian soul is indeed the ultimate causal basis for memory. Hence the soul switch thought experiment.

In the unit on Pythagoras, I suggest an answer to this metaphysical question, in terms of the theory of 'attunement', and I connect this with contemporary accounts of the self as a 'program' which can be downloaded from one body and uploaded into another body. However far fetched these ideas may seem, Pythagoras must have had some notion of what 'made' the soul, which explained how it was possible for Pythagoras to correctly remember, and not merely seem to remember, things that Euphorbus did. Exactly what that notion was, we can only speculate.

All the best,

Geoffrey