To: Mark S.
From: Geoffrey Klempner
Subject: In what sense is Hume a sceptic about causation?
Date: 19th March 2010 12:37

Dear Mark,

Thank you for your email of 14 March, with your essay for the University of London BA Modern Philosophy: Descartes et. al. module, in response to the question, 'In what sense, if any, was Hume a sceptic about causation?'

This is an excellent answer to the question. I liked our observations concerning the various 'mitigating factors' regarding scepticism: naturalised epistemology, pragmatism, contextualism etc.

I agree that Hume is, in your terms a 'mitigated' sceptic about the truth of causal determinism, and also about individual claims that c is the cause of e, and an 'unmitigated' sceptic about the notion of a 'necessary connection' between cause and effect.

However, I do have problems with Hume's use of the term 'sceptical'. Though it may have corresponded to the usage of the day, his attacks on metaphysical claims would not be described as 'scepticism' according to modern usage. If I claim that the notion of X is incoherent, I am not being 'sceptical' about the existence of X: I am simply denying it. You can be sceptical about the existence of God (given that there is more than one possible definition of God) but if you regard the notion of time travel as involving a logical contradiction, then you are not a sceptic about time travel, except in a loose sense, as 'one who denies the logical coherence of time travel'.

I may have expressed before my view that the notion of the truth of a statement of 'constant conjunction' -- the truth of an unrestricted universal generalization -- is itself something that an opponent of metaphysics could well have doubts about. However, to my recollection, nowhere in the Enquiry or Treatise does Hume consider this from the point of view of what would now be called 'realism' or 'anti-realism' about the truth or meaning of universal generalizations. If this is the cure, how much worse must be the disease? Yet, for Hume, the idea that there exists, in reality, an unperceived and unperceivable link between cause and effect apart from what we can observe is indefensible. It is we who supply that link, in 'spreading our minds' over objects.

It is surprising, therefore, that in his 'unmitigated scepticism' with regard to necessary connection, the thought of what exactly it would mean to assert or believe the 'truth' of an unrestricted universal generalization never seems to have occurred to Hume. One possible explanation is that Hume is so thoroughly entrenched in an anti-realist mindset that he doesn't even see the need to make the case.

Saul Kripke in his book 'Wittgenstein on Rules and Private Language' compares Wittgenstein's account of following a rule to Hume's critique of the notion of causation as a 'sceptical solution to a sceptical paradox'. As with last time, this isn't something you would necessarily know about, but I would bet that the allusion was in the examiner's mind when the question was composed. That would be something worth following up.

One thing you could also have done is compare other kinds of thing that Hume expresses unmitigated scepticism about: the self as an entity, the 'continued and distinct' existence of the objects of perception. How do these compare with his scepticism with regard to necessary connection? For starters, Hume is so dismayed by his conclusions regarding external objects, that the only solution seems to be to go off and play his game of backgammon. He doesn't at all like the conclusion he is driven to, yet he insists on it because it follows from his own principles concerning ideas and impressions.

In the case of the self, Hume is more inclined to see this as an illusion which we can dispense with. There will still be David Hume, the physical person (modulo the problem of external objects) and his mental states. No need for an additional internal 'owner' of those mental states.

Again, with necessary connection, Hume seems to think that we are much better off without this piece of metaphysical baggage.

What Hume does offer, in a kind of 'mitigation' of this unmitigated scepticism is his theory of fictions. You remark about his projectionism, the idea of the mind 'spreading itself' on objects. It is the same process of composing fictions which accounts for our notion of the distinct and continued existence of external objects, and, possibly also the self (although I'd have to search hard to find a reference for this).

In this regard, Hume is seen by commentators as having anticipated to some extent Kant's treatment of causation, external objects and the self. What were for Hume, 'necessary fictions' become for Kant 'synthetic a priori principles of the understanding'. Again, something that you will be able to follow up when you commence your studies of Kant.

All the best,

Geoffrey