Friday, October 11, 2013

Hume and scepticism about causation

To: Craig S.
From: Geoffrey Klempner
Subject: Hume and scepticism about causation
Date: 31st December 2010 12:08

Dear Craig,

Thank you for your email of 15 December, with your essay for the University of London BA Modern Philosophy: Locke, Berkeley and Hume module, in response to the question, 'In what sense, if any, was Hume a sceptic about causation?'

(Yes, I'm here in my office, on the very last day of the year. However, you needn't feel sorry for me, I don't have too much to do today!)

This essay has given me some pause for thought. I have always held that Hume's first, 'philosophical' definition of causation is in fact a very strong claim, when interpreted in a realist manner (there is no evidence that Hume was anything but a realist about truth, in the contemporary sense of 'realism' as opposed to 'anti-realism'). If it IS true that the stone caused the window to break, then at all times and places, to infinity, ceteris paribus, stone throwing is followed by window breaking. This is something no-one can ever know. I can't think of any necessity that you'd want that was stronger than this. But that's because I agree fundamentally with Hume that there isn't anything *else* to causal necessity than the truth of a universal generalization.

The big catch in all this (as many have pointed out, e.g. Anscombe) is defining the 'ceteris paribus' clause. It can't be done without a residual fringe of vagueness. So what becomes of the alleged 'universality' then? My solution would be to go to the extreme: repeat the big bang (or the start of the universe, wherever you want to start it) an infinite number of times (as in Nietzsche's eternal recurrence, originally a Stoic doctrine). Every time, the stone breaks the window, the entire course of the history of the universe is the same. No ceteris paribus problem because *everything* is the same, modulo where we are in universal time. The only drawback with this is that you need to assume the truth of determinism; whereas arguably there could be causal truths even in a universe where there were occasional deviations from determinism.

I have yet to see any case made that there is something 'underneath' that makes universal causal truths true, the 'secret connexion' or whatever. Like Hume (or like my 'Hume', if that's not what Hume claimed) the idea still makes no sense at all to me.

I do understand, however, that there is lots of stuff we don't know about how objects are able to exert influence on one another, the unknown 'powers and principles' that I think Hume is appealing to. This is an empirical matter. Newton thought that gravity had to be taken as an ultimate, as you say, he has no 'hypothesis' to explain it. But that's no longer the case. There are (as I understand) several candidates for an explanation of why gravity exists and how it works.

All this is fully consistent with the universalist definition of causation. There isn't an extra 'super-power' which enables all the various powers and principles to do their work. Their working in the way that they do is just how the world works, how causation happens.

Another example which I have been thinking about recently: Descartes in the 'Meditations' argues that material objects continue to exist from one moment to the next only because of God's continual exertion of his creative power. I think Descartes saw something here that many have missed. You need an explanation of why an object which exists at t, also exists at t' no less than you need an explanation of why an object continues at constant speed in a straight line unless acted upon by a force, i.e. an explanation of the phenomenon of inertia. In both cases, this is a fact or law of nature, which could have been otherwise, and the physicist has the option of taking this as an ultimate (as in Newtonian gravity) or seeking a deeper explanation. 'Existential inertia' is not true a priori.

You say one thing which I definitely disagree with: 'First, an assertion that there definitely is no real connection is too dogmatic, too out of keeping with the non-committal scepticism he shows elsewhere, for example in his views about the existence of external objects.' I take it that you are referring to the section, 'On Scepticism With Regard to the Senses'. Non-committal? Hume despairs of his conclusion, his only recourse (famously) is to go off and play a game of backgammon. This is where philosophy leads you. Believe what you like when you are not doing philosophy.

I have something to say about this in my new blog Hedgehog Philosopher, Day 14:

As I see it Hume's agenda is to limit the pretensions of reason to make room for his theory of human nature. This was Hume's great discovery. He was to be the Newton of the mind. To say, for example, that a material object 'continues' to exist, 'distinct' from the act of perceiving it, IS to make a statement about human psychology.

Where Hume's treatment of causation differs is in the availability of two definitions, a philosophical definition and a psychological one. He badly needs a philosophical definition of 'continues to exist' but never provides one. However, arguably, Kant provided the missing account in his transcendental deduction of the category of substance.

All the best,