Cause and correlation Notes for October 20

Main points

We make inductive inferences to identify causes and effects. Hume claims to have shown that we cannot do this using reason. Instead, he claims, these inferences are the product of custom or habits.

We spent a fair amount of time talking about his negative case. Then we turned to a problem for his positive account, the definition of cause. The problem is that Hume’s definition appears to be incapable of distinguishing between cause and correlation.

Hume’s negative case one more time

Hume doesn’t think that you can literally see that a cause made the effect necessary. I think he’s right about that. Everyone else disagreed, but I think you all actually meant something slightly different. I don’t think you meant that you think you can see why the cue ball makes it necessary that the 8 ball move, such that it cannot do anything else. I think you meant that you don’t think it’s possible for the 8 ball not to move, given the laws of nature.

I think you may well be right about that and that Hume might have agreed with you.

One thing our conversation exposed is that Hume means something fairly specific and controversial when he uses the term “possible.” He means if you can conceive or imagine something, then it is possible.

I think everyone can imagine the 8 ball not moving. As Hume uses the word “possible,” you think it’s possible that it won’t move. But you were using the word “possible” in a different way. You meant something like “possible, given the laws of nature.” In that sense, it may well not be possible for the 8 ball to do anything other than move.

At bottom, Hume wants to establish two things.

1. There is an interesting problem for inductive inference: it relies on the uniformity principle and the uniformity principle cannot be established by a non-circular argument.
2. We don’t really understand why things must be as they are. That includes the laws of nature and the causal relations that are governed by those laws.

We were talking about the second point today. At some point in the conversation, Hume will have to insist that you can’t explain why the laws of nature make it impossible for the 8 ball to do anything but move. You’re just reduced to saying that this is the way it has always been. That’s not a very deep understanding.

Hume’s positive account

Hume offered several definitions of the terms “cause” and “effect.” He presented them as being identical when, clearly, they are not. It’s confusing.

Nonetheless, the central idea is clear enough. It is that “cause” and “effect” are equivalent to things that are constantly conjoined with one another. If A’s are always followed by B’s, then A is a cause of B and B is an effect of A.

I said that this is not enough to distinguish cause from correlation or coincidence. We considered how Hume might respond to some of the examples that support this objection. Then we closed with some remarks about the next section. There, Hume will insist that our reasons for believing in the laws of nature are much better than our reasons for believing in miracles. How he can say that consistently with his skepticism about inductive inference is an open question.