„In every area of thought we must rely ultimately on our judgments, tested by reflection, subject to correction by the counterarguments of others, modified by the imagination and by comparison with alternatives. Antirealism is always a conjectural possibility: the question can always be posed, whether there is anything more to truth in a certain domain than our tendency to reach certain conclusions in this way, perhaps in convergence with others. Sometimes, as with grammar or etiquette, the answer is no. For that reason the intuitive conviction that a particular domain, like the physical world, or mathematics, or morality, or aesthetics, is one in which our judgments are attempts to respond to a kind of truth that is independent of them may be impossible to establish decisively. Yet it may be very robust all the same, and not unjustified.
To be sure, there are competing subjectivist explanations of the appearance of mind-independence in the truth of moral and other value judgments. One of the things a sophisticated subjectivism allows us to say when we judge that infanticide is wrong is that it would be wrong even if none of us thought so, even though that second judgment too is still ultimately grounded in our responses. However, I find those quasi-realist, expressivist accounts of the ground of objectivity in moral judgments no more plausible than the subjectivist account of simpler value judgments. These epicycles are of the same kind as the original proposal: they deny that value judgments can be true in their own right, and this does not accord with what I believe to be the best overall understanding of our thought about value.
There is no crucial experiment that will establish or refute realism about value. One ground for rejecting it, the type used by Hume, is simply question-begging: if it is supposed that objective moral truths can exist only if they are like other kinds of facts--physical, psychological, or logical--then it is clear that there aren't any. But the failure of this argument doesn't prove that thereobjective moral truths. Positive support for realism can come only from the fruitfulness of evaluative and moral thought in producing results, including corrections of beliefs formerly widely held and the development of new and improved methods and arguments over time. The realist interpretation of what we are doing in thinking about these things can carry conviction only if it is a better account than the subjectivist or social-constructivist alternatives, and that is always going to be a comparative question and a matter of judgment, as it is about any other domain, whether it be mathematics or science or history or aesthetics.“
Mind and Cosmos: Why the Materialist Neo-Darwinian Conception of Nature Is Almost Certainly False