Antirealists don't care if you dropkick random people

This post got a lot of traction on twitter:

I can’t speak for other atheists, but I can say how I can speak for myself: I don’t think we evolved emotional reactions that track objective moral truths. I enjoyed one exchange that followed from this tweet. Someone replied:

I don't believe in objective moral truths; I think our emotional reactions are a product of natural selection towards behaviors that facilitated survival

To which the original poster responded:

This is a classic case of normative entanglement. No, we don’t think it’s objectively wrong to dropkick random people. This question entangles a normative claim with a metaethical claim, such that to say “no” could give the impression that you don’t think it’s wrong to dropkick random people. But saying “no” to this question does not mean that you don’t think it’s “bad to dropkick random people,” it only means that you don’t think it’s objectively bad.

I’ll illustrate why this is a bad question with two examples. A gastronomic realist thinks there are facts about what food is stance-independently good or bad independent of our taste preferences. For example, you may prefer the taste of Pepsi over Coca-Cola. But if so, you could be incorrect. Coca-Cola could be objectively tastier, regardless of your preferences. Suppose a gastronomic realist sees atheists having a pizza party and asks them:

How do atheists explain the fact that we just happened to evolve emotional reactions that track the objective gastronomic truths?

One of the atheists responds:

I don't believe in objective gastronomic truths; I think our emotional reactions are a product of natural selection towards behaviors that facilitated survival

The gastronomic realist then responds:

So you don’t think it’s objectively disgusting to put pineapple on pizza?

Suppose they say “No.” What follows from this? Does it follow that they have no taste preferences at all? That they are utterly indifferent to what’s on their pizza? Pineapple. Salted caramel. A pile of worms. Since they don’t think there are objective facts about what food is good or bad, surely they have no reason to prefer one pizza topping over another. This, of course, is absurd. People can and do have subjective food preferences and they act on those. People can likewise have non-objective moral standards, and judge and act in accord with those standards. They don’t have to think their moral standards are objectively correct to think things are morally right or wrong, any more than you have to think there are objective facts about what food is good or bad to have food preferences.

Here’s a second way to illustrate the problem. Suppose you’re a moral realist, and I ask you:

“Do you think it’s magically wrong to dropkick random people?”

If you say “yes,” I can interpret this as a concession that you think morality is magical. I could then mock you for believing in magic. If you say “no,” I could then recoil in horror and act like you just acknowledged you have no objections to dropkicking random people. Neither response would be appropriate. Why? Because I’m embedding a metaethical question in a normative question, such that any answer with respect to the former carries pragmatic implications and Bayesian information about your stance on the latter, and you have no way of individually addressing each of the two components of the question with a simple “yes” or no.” I discuss the latter problem here. This table illustrates how a “yes” resolves in one, definitive direction, while a “no” is ambiguous between three possible responses. Going back the question posed in the tweet, this results in the following:

Saying “Yes” puts you in the top left category. Saying “No” puts you in any one of the other three categories. Asking a question with an asymmetric and ambiguous response pattern like this is a bad way of exposing problems with moral antirealism. Moral realists should stop asking this sort of question.

I’ve said all of these things many times before. I am largely posting this to document a very clear case of normative entanglement. I really hope antirealists start pointing out the problem whenever people ask questions like this. I will keep pointing to examples like this until doing so catches on.