Seeing Ought from Is

     The other day I found myself in an argument with a religious person who claimed that atheists cannot support any concept of morality, because you can't derive an ought from an is. I was familiar, of course, with the history of this idea, first articulated by the famous Scottish philosopher David Hume, and sometimes smooshed together with G.E. Moore's naturalistic fallacy. But I'd not encountered it quite so plainly adopted as an argument against atheism, and it rather surprised me, because I should have thought that anyone familiar with the ought-is problem should also have known that philosophers have developed quite a few different ethical theories that do not depend on the God postulate.
     Now, the is-ought problem does seem to me to be insurmountable, at least formally, but that doesn't particularly bother me. We know, since Gödel, that in any sufficiently robust logical system there will exist unprovable truths. (If you cannot prove "This statement cannot be proved", then it's true. But if you can prove it, then your system allows you to prove a falsehood, and you have bigger problems.)  So it may well be that values can be true but unprovable.
     Besides, there are lots of reasonable things we can infer about morality, even if we cannot prove that it objectively exists. By definition, morality is concerned with choice, more specifically the business of approving one choice as preferable to another. The very idea that we might prefer one alternative over another implies a host of pragmatic considerations. For example, if we posit that there exists a moral distinction between two alternatives, that it matters which one we choose, and we do not know which one we should choose, then it seems extremely likely that finding out what we ought to do is at the very least an instrumental good. (It is conceivable, of course, that some mysterious absolute morality places the highest value on ignorance of morality, in which case we ought not make such inquiries, but morality's just not playing fair with us in that case.)
     Similarly, a pragmatic approach to uncertain morality leads to several other prudent rules of thumb. Generally speaking, a good functional strategy in almost any game is to make the move that leaves you the greatest range of options for the next turn. In day-to-day life, this means you'll want to avoid choices that can't be undone later, unless you're pretty sure you won't want to undo them.
     As well, recognizing that other minds exist and are themselves engaged in the business of trying to figure out what to do, we might do well to take their perspectives into account, and there is some empirical evidence to suggest that that cooperation gives us a wider range of options than conflict does. So even if it turns out that this mysterious absolute morality really just wants us to hate and kill each other, we're likelier to figure that out by putting our heads together than by going at it alone.

     So, without knowing for certain any particular value is absolute, we can derive some pretty useful moral principles just from the presupposition that some sort of moral value exists. But we cannot establish with any certainty that there is any such thing, and that was the core of this person's argument.

     However, as it turns out, there is an argument frequently used in apologetics for why one ought to believe, even in the absence of evidence, and surprisingly, while it's invalid for the purpose Pascal intended (showing that it's prudent to believe in the Biblical God just in case), it turns out to be quite valid for acting as if there is such a thing as morality in the absence of evidence that there is.

     Consider the two possibilities: Either there exists some kind of moral value, or there does not. To put it another way, either there is a objective standard in the universe by which choices may be said to matter, or nothing matters at all. You can choose to behave in accordance with either belief. If you behave as if something matters, and you're wrong, well, it doesn't matter. It only matters if something matters, so prudence would dictate that we should act as if something matters, even if we aren't sure it does matter.

     That does not actually tell us what matters with any certainty, and I guess that makes some people pretty uncomfortable. But morality isn't supposed to be comfortable.