20 March 1998

I'm stuck. I don't know how to continue my thought experiment! It's not the first time this has happened. This is the way I seem to approach any philosophical question. 'First, think of a thought experiment.' The last thing I'd ever consider doing is attempt a piece of sober philosophical analysis, or asking myself, 'What do I mean by...?' And how my students hate it! 'You can't just start off without defining your terms! How do you know that you're not talking nonsense, or contradicting yourself? What is the point of all this philosophy of science fiction, anyway?' And so on. — Well, what is the point?

I've always liked Wittgenstein's formula: 'I am teaching you to move from a piece of disguised nonsense to a piece of patent nonsense.' (From the Philosophical Investigations. That may not be an exact quote.) In yesterday's thought experiment of the buried Ancients, the 'disguised nonsense' was supposed to be the idea that conscious beings could go through their entire lives learning about the world, but never applying that knowledge in a practical context. In attempting to fill in the details of the imaginary world I had constructed, the nonsense was supposed to become clearly apparent. Now, I'm not even sure that it does. I feel like saying, 'Look, you can see that this is absurd!' To which, of course, someone who was sceptical about the point I was trying to make would say, 'It doesn't seem absurd to me.' Stand off.

Can anyone help me out with this?

Let's put the thought experiment on hold for the moment and try a different approach. We are looking at the process of learning, not in a psychological way but in terms of logic, as a contribution to the branch of philosophy known as 'epistemology' (13 March). To do the psychology of learning you conduct experiments. Get people to learn things, measure how fast they do it, and so on. My project is altogether different. I am trying to find out what learning is in the most general, logical terms. Perhaps terms that would apply in conceivable circumstance, every possible world (although I'm not sure about this). In terms of this project, it would be a discovery of some importance if one could prove — which I haven't been able to do up to now — that you need to be a physical agent in order to learn things. Or, for that matter, that you don't need to be physical agent, which I can't prove either. Maybe I'm on the wrong track altogether.

The American philosopher W.V.O. Quine once described a picture of human knowledge as a network of interconnecting beliefs, where only the edge of the network made contact with sense experience. The logical connections between beliefs then transmitted the results of that contact throughout the network, and ultimately to its very core, the fundamental propositions of science, such as the law of the conservation of energy, or even, possibly, the very laws of logic themselves. My quarrel with this picture is that it describes a passive process. The external world makes its impression on us and we react by forming beliefs. I don't like it. The picture is wrong, and wrong in a way that leads us into serious philosophical difficulties.

The difficulties have to do with the traditional problem of scepticism (13 March). Now Quine has his own way of dealing with the worries of the sceptic. It's beautifully simple, in its way. You just refuse to recognise the legitimacy of the sceptic's demand for a proof of legitimacy! We don't have to 'prove' our 'title' to the beliefs we hold. Learning about our world, responding to sense experience, theorising, are natural processes and the only job of the epistemologist — the philosopher inquiring into human knowledge — is to describe those processes in general terms. (See Quine's essay, 'Epistemology Naturalised'.) — Actually, I suspect that all Quine has done is propose a magical solution to an insoluble problem. At least, the problem is insoluble on Quine's picture.

My alternative picture of the learning process is one of active attunement to one's surroundings. All learning is more or less 'groovy' (17 March). Now, of course, it's been said before: 'Belief is for the sake of action.' But it is not enough to say that. All believing is, I want to say, in some sense acting. All judgement is practical judgement. A person who knows what to dowith the belief XYZ doesn't hold the same belief as someone who does not know what to do with XYZ, for whom XYZ is merely 'a piece of knowledge'. Knowledge doesn't come in 'pieces', or 'articulated' for that matter. We actively learn, we attune ourselves.

In measuring my distance from Quine I'm miles away from hitting the nail on the head. I'm still waving my hammer in the air. I can't even see a nail. — Is there a nail? Has anyone out there seen a nail?