To: Anna H.
From: Geoffrey Klempner
Subject: Hume's account of the self vs Cartesian dualism
Date: 4th January 2011 13:10

Dear Anna,

Happy new year!

Thank you for your email of 22 December, with your second essay for the Philosophy of Mind program, in response to the question, 'Do you agree with the philosopher David Hume that 'I never can catch myself at any time without a perception, and never can observe any thing but the perception'? Examine Hume's account of the self showing the main features that distinguish it from Cartesian dualism.'

In your essay, you show a well tuned appreciation of the mechanics of Hume's theory of ideas, and the difficulties which this raises for the notion of a substantial 'self'. As you argue, Hume's unsuccessful search for the self leads only to failure. Philosophy, it seems, is unable to justify by reason that which we are all fully conversant with, namely we are persons with an identity over time, subjects of experience who have thoughts, feelings and perceptions.

However, there are grounds for hope. As you argue, there does appear room here for a distinction between an empirically based notion of an enduring self, and the discredited Cartesian notion of a soul or non-material substance. In your last sentence, you suggest a possibility that Hume appears to have overlooked: 'that sense perceptions are only foundation stones and not the sole elements in the narrative of our existence'.

In other words, we can be good empiricists and embrace a notion of a self, not as a Cartesian soul substance, but as a construct based on and validated by experience.

How would this work, exactly?

In the program, I offer a kind of justification for Hume's theory, in terms of the 'flock of birds' analogy which you refer to at one point. From a purely logical point of view, if any two experiences A and B are 'present' another (a term one can use is 'co-present'), then any third experience C which is co-present with A is necessarily co-present with B and vice versa. In other words, the very idea of two or more experiences 'occurring' at the same time implies a kind of identity which is purely logical, the identity of a set or class.

At any given time, on this theory, I am aware of all 'co-present' ideas or impressions, the sum total of which logically constitute my present 'self'. (To the objection that we have illegitimately imported the notion of 'I' when all there is, is the set of ideas/ impressions, we can reply that a more accurate statement would be that 'There is awareness of all co-present ideas or impressions...'.)

The difficult part comes when we try to include the passage of time. On this theory, each co-present set of ideas/ impressions leaves behind a memory image, which is then gathered up by the 'next' co-present set, forming, in your words, 'links in a chain'. In the program I talk of 'weaving' overlapping strands together to create the story of a self. The self just 'is' the sum total of overlapping strands, just as a cloth is nothing but the threads the compose it.

At first sight, this seems defensible, and consistent. There is a 'self', not an object of perception as such but a logical construct of mental items. So Hume is right to assert that nothing is perceived apart from perceptions. Yet this theory would also allow him to speak of the 'self' as a legitimate construct.

I am reminded of the infamous remark by Margaret Thatcher: 'There is no such thing as society.' What she meant was that a society is just a collection of individuals, and is nothing apart from those individuals. (She didn't much care for 'sociology'!) However, the question at issue -- assuming the truth of the claim that a society is a class or collection of individuals from a logical point of view -- is whether there are useful things to say about that class or collection in addition to all that can be said about the individuals that compose it. The point here is about reductionism. Even if the objects of biology are the objects of physics (i.e. made of atoms and molecules) it does not follow that you can translate the laws of biology into the laws of physics.

Transposing this to Hume, one might argue that there is room for acknowledging Hume's premise, that all there is to the self is a collection of mental items, while denying the implication that there is nothing that one can say about the subject or self, that cannot be said about the individual items that compose it. In other words, the possibility one might see here is the idea of a 'narrative of our existence' which is more than just an account of all the items that make up the contents of consciousness. Such a narrative is provided by folk psychology. One talks of persons who have beliefs and desires, who feel pleasures and pains, who form intentions to act, who engage with one another in dialogue, offer justifications, praise, blame etc.

The unanswered question is whether the materials which Hume allows are sufficient to enable us to do this. Arguably, what Kant saw, and Hume missed, is that the very notion of a world of external objects, as well as the notion of a subject of experience which traces a path through that world, is a presupposition of our being able to describe our experiences. In other words, there is an a priori aspect to the very notion of a self, or a world, which cannot be reduced to what is empirically 'given' in an uninterpreted stream of ideas an impressions.

All the best,

Geoffrey