On the Naturalistic Fallacy

In addition to denying the is/ought gap (see my previous post), those who attempt to argue for the existence of a scientific morality often deny the so-called naturalistic fallacy. This is the fallacy of defining moral concepts in non-moral terms, as Sam Harris does when he says that “good” just means “that which increases the overall well-being of conscious creatures.” Other examples of the fallacy include defining the good as happiness, or as what helps promote human flourishing, or (to use a supernaturalistic example) as what God commands. And the reason this is a fallacy is that, no matter what one picks as the definition of “good,” someone can ask whether that thing is actually good without thereby making any kind of mistake.

Harris, as might be expected, specifically denies this: “it makes no sense at all,” he says, “to ask whether maximizing well-being is ‘good’.” Yet at one point in The Moral Landscape, he himself ends up asking this very question! In endnote 50 to the second chapter (p. 210), Harris considers a challenge to his view which asks whether “it would be ethical for our species to be sacrificed for the unimaginably vast happiness of some superbeings.” That is, if by eating us, their gain in well-being was greater than our loss (so that the overall amount of well-being in the universe increased), then wouldn’t it follow that our becoming their dinner would be a good thing? Harris has no alternative but to admit that it would: “Provided that we take the time to really imagine the details (which is not easy), I think the answer is clearly ‘yes’.”

Now, whether you agree with Harris or not (I sure don’t), the larger point is that there is a real question here. And Harris in effect admits as much when he grapples with the issue. He says that we first must “really imagine the details” and then only “thinks” the answer is yes. But if “good” just meant “whatever increases the overall amount of well-being,” then there should be no hesitation about this. Given that the aliens devouring us increases well-being, it would immediately follow that it was good. What could be more obvious? (Compare: if I were to ask you whether a particular man who has never been married is actually a bachelor, you wouldn’t have to “really imagine the details” of his unmarried existence and then answer that, as a result, you think he is a bachelor.) The fact that this thought experiment presents a real challenge to Harris shows that he doesn't think of "good" as synonymous with "increases well-being."

The naturalistic fallacy is a fallacy because moral terms do not mean the same as anything non-moral. That’s just not how moral concepts function. When someone claims that something is good, they do not mean only that it increases well-being (or that it promotes human flourishing, or that it was commanded by God). I also think it’s important to point out that the approach taken by those like Harris and Carrier (as discussed on the previous post) is unnecessary. I suspect that the only reason it ever occurred to them to argue as they do is because of apologists who claim that atheism cannot justify morality. But in order to answer these apologists, one does not need to claim that morality is objective, much less that it can be made scientific.

Moreover, atheists in particular might think twice before arguing against the naturalistic fallacy, for it is closely related to one of the most important arguments used by nonbelievers, the Euthyphro problem. This is the problem raised regarding to the Divine Command Theory, which states that what God commands is by definition good. It seems perfectly reasonable to ask of a divine command theorist whether it would be good for God to, say, command torturing babies. But that’s just another instance of asking whether something that is defined by some as good is actually good.

Franz Kiekeben is a former lecturer in philosophy and the author of two books on atheism, The Truth about God, and Atheism: Q & A. He has also written for Skeptic magazine and published academic articles on determinism and on time travel.