Necessary connection: background

Class notes for 18 January

Main points

I began by revisiting a question I had asked about the Uniformity Principle: why is it the case that reason determines us to make causal inferences only if it uses this principle?

I said that I was no longer confident in the answer I had given last week and proposed another one. I’m having third thoughts, see below.

Then I described what I see as various differences between the way that probabilistic arguments work when they are to be used by reason and they way they work when they are used by custom or habit. When our thinking is governed by habit or custom, the Uniformity Principle is confirmed by a million experiments and we draw distinctions between good and bad inferences. When our thinking is governed by reason, the Uniformity Principle must be proven by something more than a probable argument and all inferences based on past experiences are tied for last.

It isn’t clear to me why Hume thinks he can say both things.

Finally, I went over an outline of 1.3.14 and gave a quick run through Malebranche’s occasionalism. I said that there were theological arguments for his occasionalism, based on premises about God’s omnipotence, and non-theological ones that use God’s omnipotence to explain the various causes and effects that are obviously part of our world.

Hume went Malebranche one better: we don’t understand power, either in natural or supernatural bodies.

Reason and the uniformity principle

I tried to recount what I see as an error that I made last week. Specifically, I asked why reason has to use the Uniformity Principle and I proposed that the answer is that the Uniformity Principle guarantees a certain conclusion, one that cannot be conceived to be otherwise. This is because the Uniformity Principle concerns how things must be.

However, I did not find that satisfying. Among other things, knowledge, the product of demonstrations or intuition, is what cannot be conceived to be otherwise. Reason consists in comparing things and finding either the constant or inconstant relations among them. ( The only reason I had for thinking that the conclusion of reason had to be certain was my story about the Uniformity Principle.

I proposed another answer to the question of why reason has to use the Uniformity Principle. It is that only the Uniformity Principle could explain the necessary connection between cause and effect. The idea was that the Uniformity Principle describes how things must be: the same as they were.

But, having thought about it, the textual evidence is underwhelming. It’s true that the paragraph immediately before the Uniformity Principle is about the idea of necessary connection. But there appears to be a break after that, a switch to a different question about how we make particular inferences.

My third (and I hope final) thought is that I have been paying too much attention to the word “must” in the Uniformity Principle. Instead, perhaps I should just think of it as a principle that purports to explain the relevance of past experiences to the future. It would perform the same function, perhaps, if it said that the future will be the same as the past.