Please humor my curiosity and answer the following survey questions (more questions after the fold). You don’t have to be a member of SHAFT to take the survey, only a visitor of this site.
Jon has done a very nice job of motivating the Humean problem, so I will mostly leave this to him. Plus I am not a Humean.
But I think James is missing the point of Hume’s argument. Let’s back up a bit: To say that nature is uniform is to say that the future will resemble the past (if, in the past, X has been followed by Y, we can expect future Xs to be followed by Ys). Now Hume is not denying that, in the past, the future has resembled the past. Hume is not denying that nature has, in the past, been uniform. And James is quite right, had there not been “constant conjunction” (Xs followed by Ys) in the past, we’d be living in a very different universe (or may not be living at all).
But Hume is making a different point. He has no doubt that empirical reasoning tends to work. His question is whether scientific / empirical understanding is properly knowledge. Can you prove its foundational assumptions? His answer is: no. Why? Because even if you marshall forth the entire history of the universe being uniform as evidence, all of your empirical evidence to which we could appeal is either past or present. You can say, “Hey look, it keeps working! Xs keep being followed by Ys!” But this does not give us necessity. It does not tell us anything necessary about the future. There is no necessity in the future resembling the past. It might not. Sure, it always has before, but the laws of nature could change tomorrow. That is conceivable, it is logically possible.
Hume’s objection is not rooted in primitive science, and the problem is not resolved because of our more sophisticated view of things. This is not the sort of problem that science can “outgrow” because it is a foundational epistemological issue, not a question of how much data we have or something like that. It doesn’t matter how slick your science is, Hume’s point is that the uniformity of nature is (a) not an a priori truth (it is not necessarily the case) and (b) any attempt to prove it empirically invariably turns into a circular argument (where you presume the very thing you are seeking to prove).
Jon has done a very nice job of motivating the Humean problem, so I will mostly leave this to him. Plus I am not a Humean.
But I think James is missing the point of Hume’s argument. Let’s back up a bit: To say that nature is uniform is to say that the future will resemble the past (if, in the past, X has been followed by Y, we can expect future Xs to be followed by Ys). Now Hume is not denying that, in the past, the future has resembled the past. Hume is not denying that nature has, in the past, been uniform. And James is quite right, had there not been “constant conjunction” (Xs followed by Ys) in the past, we’d be living in a very different universe (or may not be living at all).
But Hume is making a different point. He has no doubt that empirical reasoning tends to work. His question is whether scientific / empirical understanding is properly knowledge. Can you prove its foundational assumptions? His answer is: no. Why? Because even if you marshall forth the entire history of the universe being uniform as evidence, all of your empirical evidence to which we could appeal is either past or present. You can say, “Hey look, it keeps working! Xs keep being followed by Ys!” But this does not give us necessity. It does not tell us anything necessary about the future. There is no necessity in the future resembling the past. It might not. Sure, it always has before, but the laws of nature could change tomorrow. That is conceivable, it is logically possible.
Hume’s objection is not rooted in primitive science, and the problem is not resolved because of our more sophisticated view of things. This is not the sort of problem that science can “outgrow” because it is a foundational epistemological issue, not a question of how much data we have or something like that. It doesn’t matter how slick your science is, Hume’s point is that the uniformity of nature is (a) not an a priori truth (it is not necessarily the case) and (b) any attempt to prove it empirically invariably turns into a circular argument (where you presume the very thing you are seeking to prove).