Tuesday, June 28, 2011

Non-causal scientific explanation

Some scientific explanation is non-causal. For instance, we can explain the hyperbolic cosine shape that a rope suspended at both ends makes as the shape that minimizes total gravitational energy subject to the constraints. In some cases, the non-causal explanation can perhaps be turned into a causal explanation: presumably if we model a rope system suspended at both ends, and have some friction to damp things, the causal evolution of the system will eventually lead to a hyperbolic cosine profile. But (a) it is not clear that all cases of non-causal scientific explanation can be reduced to causal explanation, and (b) even when such a reduction is possible, we lose something of explanatory value.

But it is very puzzling how these sorts of non-causal explanations explain. We might invoke laws of nature that say that things behave in such-and-such a manner, but when these laws are non-causal, their explanatory oomph is puzzling. It is interesting that theists can solve this problem, by allowing non-causal explanations to be reduced to causal explanation in terms of God setting up a world where such-and-such patterns will be instantiated, but still allowing the non-causal element—say, the variational principle that says that gravitational energy will eventually be close to a local minimum—to play a genuine role in the explanation. For we may suppose that God intentionally sets things up so that they follow variational principles. Even in cases where the variational principles can be reduced to causal explanations, we can suppose that the causal structure was so set up by God as to make the variational principles true, and hence the variational principles genuinely enter into the explanations.

4 comments:

Andrew said...

Alex,

I find this very interesting. Do you have other examples of non-causal scientific explanations?

I am guessing what you mean by "causal explanation" is what Aristotle means by "efficient cause." Is that correct?

Also, I think that teleological explanations are (if the above is correct), using your terminology, 'non-causal' explanations. Analogously, to try to reduce them to 'causal' explanations eliminates some of the explanatory oomph they provide (and frankly, WHY we use them in the first place).

Alexander R Pruss said...

I think evolutionary explanations aren't merely causal--there is an added schematic element. I am here thinking of Kitcher's work on evolutionary explanation.

Jonathan Livengood said...

I find (b) very puzzling. What is it that you think is lost when a (supposedly) non-causal explanation is reduced to a causal one?

Incidentally, I think the reason it is puzzling how non-causal explanations explain is that there are no non-causal explanations. ;)

Matthew said...

If the question is *how* a non-causal explanation explains, we are asking for an explanation of an explanation. I can see why theism might be thought to provide such an explanation (or at least to locate the principle which would explain in something real). But I don’t really see why this is especially true for non-causal explanations. Don’t we face the same difficulty if we ask how a causal explanation explains?

(I understand ‘to explain’ to mean ‘to derive from a principle’. You might mean it slightly differently.)