P.F. Strawson’s critique of Hume’s bundle theory of the ‘self’

Reghu asked:

How far is Strawson’s theory of Persons a critique of Hume’s theory of Self?

Answer by Geoffrey Klempner

I was looking for my old copy of P.F. Strawson’s Individuals: An essay in descriptive metaphysics (1959) which discusses Strawson’s view of ‘Person’ as a ‘primitive concept’, in Chapter 3. Then I did a search on the internet and found his paper, ‘Persons’ published in the Minnesota Studies in the Philosophy of Science Vol. II (1958) at http:mcps.umn.edu/assets/pdf/2.7_Strawson.pdf. This looks to be a fuller treatment.

Here’s a quote from the concluding section:

“What I have been mainly arguing for is that we should acknowledge the logical primitiveness of the concept of a person and, with this, the unique logical character of certain predicates [viz. psychological predicates]. Once this is acknowledged, certain traditional philosophical problems are seen not to be problems at all. In particular, the problem that seems to have perplexed Hume does not exist — the problem of the principle of unity, of identity, of the particular consciousness, of the particular subject of ‘perceptions’ (experiences) considered as a primary particular. There is no such problem and no such principle. If there were such a principle, then each of us would have to apply it in order to decide whether any contemporary experience of his was his or someone else’s; and there is no sense in this suggestion.”

In the five decades since those words were written, it is fair to say that not many philosophers have taken up Strawson’s idea that the concept of a person is ‘primitive’. Various approaches have been taken to the problem of identifying/ defining ‘self’, the most popular being materialism. According to materialists, there are not two ‘kinds’ of ‘basic particulars’ (to use Strawson’s terminology), spatio-temporal continuants and persons but only one: spatio-temporal continuants, i.e., material entities. Some of these material entities are so internally configured that they have the property of being conscious. The problem of accounting for consciousness is still a major topic of debate at the present time, as is the problem of accounting for the criteria of personal identity.

I have recently expressed scepticism about materialism (Philosophy Pathways 213 Special mind-body dualism issue) but for the sake of this answer I will leave the question open.

First point of disagreement: I don’t read Hume’s ‘perplexity’ as a failure on his part. He is being ironic, at the expense of the Cartesian dualist. When I look into myself (says Hume), no ‘self’ (no soul substance) is to be found. However, Hume has a perfectly workable (in fact, brilliant) alternative theory of how the self is constructed, as what one might term a ‘virtual object’ of knowledge. Here is my take on this:

There is no immaterial ‘self’, but there are ‘ideas’, conceived as discrete, incorrigible mental events. These events form ‘bundles’ according to the following logical principle: If mental event A is co-present with B, and B is co-present with C, then A is co-present with C. (Co-presence is the ‘primitive’ concept.) When are two mental events co-present? An example would be, if I currently have a mental picture of the Eiffel Tower, and simultaneously hear the singing (Humean ‘impression’) of a blackbird outside my window. Let’s say that at the same time, I also experience a twinge of back pain. The three mental events form a ‘bundle’.

Over time, new mental events are added to the bundle and other mental events are discarded. What we term ‘personal identity’ is just the continuity of the bundle — like a flock of birds growing and shrinking as birds join or leave the flock. Memory plugs the gaps of unconsciousness when no mental events are occurring. (Memories are just more ‘ideas’ on Hume’s theory.) That’s it. The self, and personal identity, are ‘virtual’ in the sense explained: they are merely logical constructions. There is no mental ‘reality’ beyond the flux of impressions and ideas.

The weakest link in Hume’s theory is the Cartesian notion that mental events are ‘incorrigible’, a notion critiqued by Wittgenstein in his argument against a Private Language in Philosophical Investigations. But Strawson doesn’t like Wittgenstein’s solution to the problem of explaining how self-ascription of psychological states (such as ‘I am in pain’) is possible. According to Wittgenstein, ‘My back hurts,’ is not a statement with truth conditions but merely an ‘expression of pain’. Strawson calls this the ‘no-ownership view’, which he finds counter-intuitive. Surely if I say my back hurts and you say, ‘GK’s back hurts’ we are saying the same thing about the same thing? We are both making statements with the same truth conditions.

How would Wittgenstein respond to this criticism? Say what you like, it makes no difference to what is actually taking place. There is a huge asymmetry between the first-person and third-person case, but if you want to paper over the crack and call it ‘saying the same thing (is in pain) about the same thing (GK)’ you can, and no-one will contradict you. That’s how we actually talk, that is our ‘conceptual scheme’ to use Strawson’s terminology. Wittgenstein would add that it is part of our ‘form of life’ that we don’t stop to puzzle over the asymmetry of the first- and third-person case — until we are tempted into doing philosophy.

Asserting that the concept of a person is ‘primitive’, in effect ruling out any further attempts at analysis or theory, achieves nothing of substance. The deep philosophical problem remains, how it is possible that there could be such a thing as ‘being in pain’ or ‘being a person’.

As I said, that’s just my take. It remains the case that P.F. Strawson’s Individuals is one of the most important works of 20th century analytic philosophy, which ought to be on every Metaphysics reading list.

One thought on “P.F. Strawson’s critique of Hume’s bundle theory of the ‘self’

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.