Hume on miracles, part 2 Notes for October 28

Main points

We started with a question about the argument in part 1: does it prove too much? That is, if we accepted Hume’s conclusion about miracles, would we also be committed to dismissing testimony and observations that we really should believe? If so, we should not accept Hume’s conclusion about miracles.

Then we turned to the argument in part 2. I said that part 2 advances an inductive argument for the conclusion that testimony about miracles is always tainted.

I thought our discussion of these questions was excellent. (I didn’t write enough down, so I’m going to miss a lot of names: sorry!) It’s been good all term, but we had an unusually good presentation of both sides today. Well done!

Does Hume’s argument prove too much?

We used a particular example: Rutherford’s incredible experience with the atom. The question was whether Hume’s conclusion that we should dismiss reports of the miraculous out of hand would also apply to Geiger and Marsden’s observations. Should Rutherford have dismissed what Geiger told him out of hand too? If so, something must be wrong with Hume’s argument because our understanding of the atom was advanced by Rutherford’s accepting what Geiger told him.

There is one obvious difference between Rutherford’s experiments and miracles. The experiments can be replicated. This suggests that they conform to the laws of nature. It also means that lots of people can see them.

But there is a problem. Remember that we’re talking about the argument from part 1. The conclusion of that argument, as Greg noted, is that reports of the miraculous should be dismissed out of hand, without further investigation. So Hume owes us an explanation of why we would take an experimental result like this one seriously in the first place. Why would we think there is something to investigate rather than dismissing the result out of hand?

In the particular case at hand, I think Gabe is probably right to say that Rutherford wasn’t confronted with an observation that contradicted what he took to be a law of nature. They had a first draft of a theory of atoms, but not much more than that. So Hume might legitimately say that he would not be committed to the wrong answer in this case. He only calls for rejecting observations that conflict with laws of nature but that is not relevant to this case.

Is that enough to get him out of trouble? Brian didn’t think so. Suppose you have an observation that really does seem to conflict with what you take to be a law of nature. There should be two options. Either your observation shows that you had a mistaken understanding of the laws of nature or you were mistaken in thinking that you observed what you thought you did. Brian thinks Hume is committed to thinking there is only one option, the second one. But that’s a mistake. Sometimes, the first option is the correct one.

I agree that Hume doesn’t give much guidance in a case like this. Since miracles are defined as violations of the laws of nature, every reported miracle involves a third option. It is that the observation involves a violation of the laws of nature without calling those laws into question. They’re still the laws of nature, it’s just that a supernatural being violated them. It’s that sort of unambiguous claim that he wanted to dispute.

Whether that is enough to answer Brian is something I will leave open.

Part 2

I said that this part contains an inductive argument against the testimony on behalf of miracles. By calling it an inductive argument, I mean that he argues from what we know about testimony concerning miracles from the past to a conclusion about all testimony about miracles. All testimony about miracles that we know about is tainted, so we should conclude that testimony about miracles that we don’t know much about is tainted too.

This argument is different from the argument from part 1. That argument took testimony about miracles for granted, treating it as a proof. The point of that argument was that even if we accepted this testimony without question, we still would not have enough evidence to believe that a miracle happened.

I should add that it is not obvious how, exactly, the two parts are related to one another and people who study Hume have different interpretations. I’m right, of course. But others might reasonably disagree.

This page was written by Michael Green for Problems of Philosophy, Philosophy 1, Fall 2009. It was posted October 29, 2009.
Name of website