Carl Hempel has proposed the following model for explanations by reasons.

• (R) (i) A was in a situation of type C.

• (ii) A was a rational agent at the time.

• (iii) In a situation of type C, any rational agent will do X.

• (iv) Therefore, A did X.

According to Hempel (R) overcomes the shortcomings of alternative accounts of explaining by reasons such as William Dray's model, in that the so-called evaluative principle is replaced by an empirical law. It is William Dray's contention that to understand the action of a rational agent is to see that his action was appropriate in his situation. Following Dray this then means that in explaining such an action one makes use of a normative ‘principle of action', not an empirical law, which has the form ‘when in a situation of type C1… Cn the thing to do is X'.