One of the few claims accepted by nearly everyone writing about the nature of value is that the evaluative features of things supervene on their natural features. Take any two persons, actions, characters, or states of affairs that are identical in all of their naturalistic features: every naturalistic feature that is a feature of the one is a feature of the other, and vice versa. These two persons, actions, characters, or states of affairs must be identical in evaluative respects as well. There can be no evaluative difference without a naturalistic difference. So, at any rate, it is said.

Those who flout the supervenience requirement when they make evaluative judgements are supposed thereby to reveal themselves to be incompetent in their use of evaluative terms. The supervenience of the evaluative on the natural thus purports to operate as a conceptual constraint on evaluative judgement. This too is accepted by nearly everyone writing about the nature of value. Given that supervenience operates as a conceptual constraint on evaluative judgement, it follows that the right way of thinking about an evaluative theory, at least in its most abstract form, is as a mapping of natural features onto evaluative features. An evaluative theory, in its most abstract form, is thus simply a long list of supervenience conditionals, conditionals such as “If objects have natural features N then they have evaluative features E,” “If objects have natural features N* then they have evaluative features E*,” and so on.