> The foundation of law is not scientific exactness or scientific empiricism. It is the mechanism by which a state establishes norms.

Exactly. So it sounds like you're agreeing with me that qualification of a particular effect as "harm" is not a matter of "medical expertise", but is rather a question of subjective norms that is in fact on the opposite side of the is-ought gap from the side at which expertise is applicable.

> A law against murder does not stop murder, but it does tell you that society does not appreciate it.

Well, not exactly. This presumes that "society" in the abstract (a) actually has a general consensus on the question, and that (b) the rules imposed by the legal system reflect that broad consensus, rather than reflecting the values or intentions of the people administering the legal system, without necessarily aligning with those of the general public.

There are a lot of questions that do have broad consensus across society, but also a lot of subjective questions that different people answer very differently. And I think that the level of consensus that actually exists in terms of considering things causing physical injury or pain as "harm" is far, far greater than the level of consensus on treating anything that causes emotional stress as "harm".

I don't think that the "negative response" criteria that you're articulating is sufficient to reveal an underlying normative consensus: I would not presume that most people would equate harm with any kind of negative reaction. For example, I would personally not consider something harmful merely on account of being annoying, insulting, or even morally questionable (though there's often overlap in the last case).

They are saying that judgements of what qualifies as harm is something like a judgement of what is good, or what is right or wrong. That’s not the same thing as evaluating whether something causes pain. You can measure whether something caused pain, sure. (Well, the sort of limitations you mentioned in measuring pain exist, but as you said, they are not a major issue.)

“Harm” isn’t the same thing as “pain”.

I would say that when I bite my finger to make a point, I experience pain, but this doesn’t cause me any suffering nor any harm. If something broke my arm, I claim that this is harm to me. While this (“if my arm were broken, that would be harm to me”) might seem like an obvious statement, and I do claim that it is a fact, not just an opinion, I think I agree that it is a normative claim. It is a claim about what counts as good or bad for me.

I don’t think normative claims (such as “It is immoral to murder someone.”) are empirical claims? (Though I do claim that they at least often have truth values.)