Roughly, values are what we are disposed to value. Less roughly, we have this schematic definition: Something of the appropriate category is a value if and only if we would be disposed, under ideal conditions, to value it. It raises five questions. (1) What is the favourable attitude of ‘valuing’? (2) What is the ‘appropriate category’ of things? (3) What conditions are ‘ideal’ for valuing? (4) Who are ‘we’? (5) What is the modal status of the equivalence?

By answering these questions, I shall advance a version of the dispositional theory of value. I begin by classifying the theory that is going to emerge. First, it is naturalistic: it advances an analytic definition of value. It is naturalistic in another sense too: it fits into a naturalistic metaphysics. It invokes only such entities and distinctions as we need to believe in anyway, and needs nothing extra before it can deliver the values. It reduces facts about value to facts about our psychology.

The theory is subjective: it analyses value in terms of our attitudes. But it is not subjective in the narrower sense of implying that value is a topic on which whatever we may think is automatically true, or on which there is no truth at all. Nor does it imply that if we had been differently disposed, different things would have been values. Not quite – but it comes too close for comfort.

The theory is internalist: it makes a conceptual connection between value and motivation.