Don't Mind the Gap
The is ought gap isn't a problem
Ever since Hume, lots of people have been very worried about the is/ought gap — the notion that there’s an unbreachable divide between descriptive claims and normative claims. The basic idea is that, because you can’t derive an ought from an is, there can’t be moral facts — after all, we can’t prove a prescriptive fact from the merely descriptive claims we know about the world.
However, I don’t think that this is a problem at all. To see this, let’s consider the two types of moral realism.
One type is moral naturalism. Moral naturalism says that moral facts are some complex type of descriptive fact. So, much like water is just H20, the moral facts are just some type of complex fact about what is. Maybe morality is what impartial ideal observers would approve of. Or perhaps morality is some irreducible higher-order property — this is what Cornell Realists think.
If one is a moral naturalist, it’s not hard to see why the is/ought gap isn’t a problem. After all, on this account, you can get an ought from an is. Oughts are just a type of is statement. Much like you can deduce facts about water from facts about hydrogen and oxygen, the same is true of getting an ought from an is. Thus, on this account, there is no fearsome, unbreachable gap.
But what about moral non-naturalism which says that moral facts are not reducible to descriptive facts? On this account, knowing all the descriptive facts about the world wouldn’t tell you the relevant moral facts. Moral facts are thought to be somewhat like the laws of logic on this account — not contingent on any physical facts. This would say that moral facts are not descriptive — they’re distinctly about what should be rather than what is.
But for moral non-naturalists, there’s no problem either. This is because moral non-naturalists don’t get an ought from an is. Rather, they argue that we have some way of knowing about non-natural facts. Much like one doesn’t derive the laws of logic from what is, the non-naturalist about morality thinks that you don’t get an ought from an is — there are just certain necessary oughts that aren’t caused by the descriptive facts.
There’s no is/math problem, even though complex mathematical facts don’t come from descriptive facts about the world. This is because mathematical facts don’t need descriptive facts as their source; moral naturalists would say the same is true of normative facts.
Thus, either you can get an ought from an is or you don’t need to because the oughts are necessary and unrelated to physical facts. Either way, Hume’s guillotine need not decapitate our moral realism.
the is ought problem doesn't defeat moral realism, but it certainly gives some doubt to the idea.
i don't buy moral naturalism—it seems viscerally intuitive that prescriptive facts are separate from descriptive ones—after all, the latter can be observed while the former can't. there is no moral particle we can put under a microscope or infer from stuff we can see from under a microscope.
if, as I believe, morals truly are simply the result of evolution imperfectly attempting to solve coordination problems, you get quite an unglamorous view of morality.
this applies to the laws of logic and math as well. despite our complex understandings of math and logic, these concepts seem to break down at the margins. for instance, russell's paradox and certain aspects of quantum mechanics seem to be in contradiction with paradigmatic logical intuitions.
it is, however, even worse for prescriptive facts than it is for logical ones. logical facts are far more agreed upon than prescriptive ones; logical facts infer and can be inferred from observed phenomena; and prescriptive facts seem often contingent on logical facts (e.g. utilitarianism assumes the aggregation principle).
nowadays i find myself leaning towards emotivism for the reasons stipulated above.