In comments starting here Reshef asks some questions and I sometimes get close to responding relevantly and then sometimes don't. After which he offers a kind of summary of what he's been trying to get at:
I’ve been trying to look through the eyes of the unhappy, or of the happy for that matters. Through their eyes human nature is morally significant. They are responding morally to it. The happiness/unhappiness is their response.I want here to get clearer about what I have been saying (what I said in that old paper still seems right to me) and what Reshef is saying. I say "It is not that just anything can be given a moral application" and "It would be a mistake to claim that just anything could be brought into a moral relation with our lives." This certainly sounds like a denial by me that anything whatever could have a morally significant place in one's life. But that isn't what I mean. What I mean is that, although a very wide range of things (including both physical objects, ideas, and sentences) can be morally significant, as can be shown by various examples, these examples do not show that absolutely anything whatsoever could be morally significant. Perhaps it can be, but (as far as my investigation goes) that remains to be seen.
What I fear, again, is that if we say that not anything could be brought into a moral relation with our lives, we will deny ourselves access to these happy/unhappy points of view: to these moral reactions (also reactions to human nature). I’m not saying that the happy or the unhappy is right. I’m not so much asking this question. And I agree that not everyone will agree. I agree that not on every view of what moral thinking consists in human nature can be a moral issue. I am just worried of a kind of meta-ethics that does not leave room for views, or attitude to life (because I'm not sure we should call happiness or unhappiness “views”), in which human nature is or can become a moral issue.
Reshef seems to be saying that someone might have an ethical view, or attitude, according to which everything one cares about is morally significant, precisely because one cares about it. And this (the thing cared about) might be anything at all.
This is a view that I find hard to get in focus, but I don't think I'm ruling it out as a possibility at all. I'm just not endorsing or adopting it.
Am I perhaps trying to do meta-ethics without ethics, and is that a tenable distinction? And what about the points made by Cora Diamond that I quoted here? Not to mention the paper by Anne-Marie Søndergaard Christensen that I mentioned here. There are reasons to think that I might need to change my tune and not just stick with what I wrote twenty years ago. But at the moment it still seems OK to me.