Tuesday, August 11, 2009

Presentism, Truthmakers and Theoretical Virtues

I’m just back from the BSPC where I gave my Truthmaking for Presentists. I got loads of good questions of and comments, but wanted to comment on one of them. In the paper I argue that truthmaker theory can be contingently true and hence that it’s no immediate objection to my view if there are possible presentist scenarios in which my account doesn’t work, so long as we can reasonably believe that they are not actual. (The scenario in question discussed in the paper is a presentist world with an infinite past.)

One of my commentators, Caspar Hare, was pushing me on whether I thought it was similarly okay for the truthmaker principle to simply be true now. He was expecting me to say yes – but I say no, it must be true always; and he was rightly pushing me to say why one should demand that the principle be true at all times but not at all worlds when one holds a view that treats times and worlds analogously in holding that only one of each (the present time, the actual world) is real.

For me the reasons to be a truthmaker theorist concern theoretical virtues. Truthmaker theory is the theory that all the brute truths are truths about what there is, and this is theoretically more virtuous than theories that take as brute truths not only about what there is but also truths about what there was, could be, should be, etc, as well as truths about how things are, what laws hold, etc, etc. If truthmaker theory is true, God’s language only needs names and an existence predicate: that’s ideologically simpler, and hence more virtuous (other things being equal), than theories which require God to be able to make predications, express tensed facts, etc.

I think that in the absence of further information, if the only reason to believe a theory is that it is theoretically virtuous, then we should take that theory to be at best contingently true. There’s no inconsistency in a theoretical virtue selecting a necessarily true theory, but we’d need some reason to think that it does in any particular case. In general the world needn’t have cooperated with what is theoretically virtuous: being guided by simplicity, parsimony etc might have taken us badly wrong – we just hope it doesn’t actually do so. That’s why I only hold truthmaker theory to be a contingent truth. (It’s also why I hold the contingency of composition, and the contingency of whether there is a fundamental level.)

But it seems to me that satisfaction of the theoretical virtues makes a demand on how the world is across times, not just at the present time, even if the present is all that is real. Suppose we’re faced with two theories, T1 and T2. T1 says that there are always 10,000 things. T2 says that there are now only 100 things but that at all other times there are billions. It seems to me that the presentist doesn’t get even a pro tanto reason to accept T2 on the basis that it says that there are fewer things in reality. (I’m assuming quantitative parsimony is a virtue – replace talk of number of things with number of kinds of things if you don’t.) Sure, the present is all that there is, and T2 says that there are presently 100 things whereas T1 says that there are presently a hundred times as many things. But the presentist should still be moved, I think, by the fact that according to T2 reality just was, and is soon to be again, massively more unparsimonious than T1 says it is. The fact that those times aren’t real shouldn’t make us worry any less about that. (By contrast, if T3 and T4 agree on what actually exists but T4 says there are more possibilia than T3 does, this only gives us even a pro-tanto reason to prefer T3 if we are realist about possibilia.)

I think that what’s driving this thought is an idea that Kit Fine pushes in his fantastic paper ‘Tense and Reality’: that everyone should think of non-present times as part of the same ‘all encompassing reality’ whereas only the realist about possibilia should think that about worlds. Now Fine takes that as a reason to deny that the present time is privileged, and so to be a non-standard realist if you’re going to be a realist about tense; but I think that one can still accept that thought and accept a privileged present. One just needs to accept that what’s going on at other times is relevant to how we judge theories in a way that what’s going on at other worlds is not (at least, not in the same way).

So parsimony doesn’t tell the presentist to accept T2 over T1, and if the truthmaker principle is only true now it doesn’t tell the presentist to be a truthmaker theorist. The virtues of truthmaker theory are only obtained if it is always true; by contrast, whether it is necessarily true are neither here nor there, as far as securing those virtues is concerned.