To: Max W.
From: Geoffrey Klempner
Subject: Locke on personal identity
Date: 12th January 2011 12:02

Dear Max,

Thank you for your email of 2 January, with your essay for the University of London Modern Philosophy module, in response to the question, 'Describe and critically evaluate Locke's account of the identity of persons.'

This is a very good answer to the question and also an impressive piece of research that more than makes up for the deficiencies of your previous essay on this topic. If you could muster this kind of expertise in a one hour exam question, you would probably get a 70+ mark for this.

Personal identity is a fascinating topic which is still gripping today, as you state. Let's start with the question of substance which was the topic of most of my response last time. Locke is sceptical about substance, and expresses this scepticism in various ways. Probably the most accurate view of what Locke actually believed would be that Newtonian corpuscles are the true 'material substance'. Kant takes a stronger line, arguing for the a priori indestructibility of 'substance' in the Transcendental Analytic. However, from the perspective of Locke's 'historical plain method', if the question is about human understanding and the concepts which we are able to form, then the corpuscular theory is just a theory, not something that is bound up with the very possibility of deriving knowledge from experience.

Maybe we are complex arrangements of corpuscles, nothing more, or maybe not. Maybe there is an immaterial soul. But if the soul exists, we can't know that for sure or argue for its existence on the evidence of our experience.

This would be my defence of Locke's alleged 'inconsistencies' in his account of the relation between the concept of a person and the notion of substance. Following the path of strict analysis, his job as underlabourer-philosopher is simply to articulate our concepts. The concept of a person would be the same in a corpuscular world, or a Cartesian dualist world, though only in the latter can there be hope of a resurrection. (Dennett in 'Consciousness Explained' counters this with the account of the self as essentially a program, a mathematical entity and therefore capable of indefinite 'resurrection' in fresh brains/ bodies.)

However, the main question concerns the shortcomings of Locke's account of personal identity as such, and not the question of how he might have conceived of the relation between 'person' and 'substance'.

Here are some points you could have made (I'm not saying you should have made them!):

The first is one that (I think) I mentioned last time, concerning the criterion for the distinction between false and veridical 'memory'. One way to put the objection is in terms of Butler's charge of circularity, but that doesn't get to the heart of the issue. The issue concerns the nature of memory as such, and what is needed for a memory to be 'true'. Butler's point is taken that one needs to distinguish between 'my remembering my F-ing' and 'my remembering that F-ing took place', where in the latter case it might have been someone other than me who F-ed. But what makes it true that it was indeed I who F-ed?

In 'Sameness and Substance' David Wiggins argues that it is continuity in an organized material package of all that is required to causally account for the persistence of memory. This revolutionizes Locke's account because it logically forces us to bring in the material aspect. (Even if this is not a living human body, perhaps only the vital part e.g. the brain.) This is a failure on Locke's part with regard to the concept of memory on which his account is based.

On page 2, you note correctly Locke's 'purpose, which was to show that consciousness provided a forensic basis for identity'. However, I do think it is misleading to describe this as a 'further' purpose. On the contrary, the very idea that personal identity is a forensic notion motivates Locke's account from the start. He isn't saying, 'Here's a good definition of personal identity. And by the way, this works very well when one considers the forensic aspect.' Personal identity according to Locke IS a forensic notion. That is how we discover that what is essential to personal identity is consciousness.

As you and many commentators have noted, this gets Locke into all sorts of difficulties. However, my intuitions tend to deviate somewhat in Locke's favour. There are two aspects to this. The first is the possibility of loss of memory, amnesia. I think there is something morally worrying in the idea of punishing a total amnesiac for a crime which so far as they are concerned was committed by someone else. You have to think of cases. Some will favour punishment (as in the drunkenness scenario) while others do not (say, permanent brain damage after a car accident).

The second aspect is more controversial, and relates to the two notions of 'remembering' (remembering F-ing and remembering my F-ing). Leaving aside the question of the causal aspect to memory, it could be argued that Locke is wrong to regard continuity of consciousness/ memory as sufficient for personal identity because it ignores something which is in some sense up to the agent, namely, my decision to 'identify' with a past self. 'I am a different person now,' could perhaps in some circumstances be meant literally. Again, one has to think of possible scenarios. Consider the WWII concentration camp guard who for the last 30 years has lived as an devout priest, tortured by memories of what was done by an individual with whom he feels no meaningful sense of connection. 'That wasn't me.' As in the case of amnesia, there would be serious difficulties in proving this, but that shouldn't obscure the logical point that arguably an aspect of personal identity is the act of 'self-identification'. (It is a question on which I am agnostic at the present time.)

All the best,

Geoffrey