If you’re following the debate over whether science can take over morality it’s worth remembering Korsgaard’s distinction between explanatory and normative accounts of ethics. An explanatory account is an account which explains why we have certain moral convictions and what those convictions are; a normative account is an account that justifies having those convictions rather than some other ones. It’s easy to mix those together accidentally, Korsgaard writes, and indeed I believe that’s what most people writing about this debate do. The scientists and science-fans assume that an explanatory account is desirable, while the philosophers are focused entirely on a normative account.
Of course an explanation of morals is not going to lead to a justification of them. This is Hume’s famous is/ought gap: you can’t derive an ought from an is. Hume’s argument is gigantic. One of the most punchy criticisms of Sam Harris says that deriving an ‘ought’ from an ‘is’ is like adding two even numbers and obtaining an odd one – you don’t have to check the working to know you’ve made a mistake somewhere. But is Hume’s argument really decisive?
Let’s take Dan Dennett’s response to Hume: that if you can’t derive an ‘ought’ from an ‘is’, just what can you derive an ‘ought’ from? Look at it this way. We have ‘oughts’ in our lives, moral prescriptions and rules that govern our consciences. But by definition everything that is in our lives ‘is’: it exists in the real world! So what are these ‘oughts’ based on, if not something about the real world?
Secondly, take Philippa Foot’s argument from her essay “Morality as a System of Hypothetical Imperatives” (although she recanted much of that later on.) Can’t we just agree that there are no ‘oughts’ that impose duties on us? Sure, you can’t derive an ‘ought’ from an ‘is’, but that doesn’t matter because all our ethical duties originate from our built-in feelings. In short, there is nothing on the other side of the is-ought gap.