Foot's approach is everywhere influenced by the later work of Wittgenstein, though she rarely deals explicitly with materials treated by him. She is an atheist.
It is on this question - the "why be moral?" question (which for her may be said to divide into the questions "why be just?", "why be temperate?" etc.) - that her doctrine underwent a surprising series of reversals. In "Moral Beliefs" she had argued that the received virtues - courage, temperance, justice and so on - are rationally cultivated, and that it was thus rational to act in accordance with them. The "thick" ethical concepts she emphasized (without using this expression) in her defense of the cognitive character of moral judgment were associated with such rationally-cultivated traits, i.e. virtues; this is how they differ from randomly chosen descriptions of action. The crucial point was that the difference between "just action" and "action performed on Tuesday" (for example) was not a matter of "emotive" meaning, as in Ayer and Stevenson, or a secret imperatival feature, as in Hare.
Fifteen years later, in the essay "Morality as a System of Hypothetical Imperatives" she took this back when it came to justice and benevolence, that is, the virtues that especially regard other people. Though everyone has reason to cultivate courage, temperance and prudence whatever he or she desires or values, still, the rationality of just and benevolent acts must, she thinks, turn on contingent motivations. Though many found the thesis shocking, on her (then) account, it is meant to be in a certain respect inspiring: in a famous reversal of a remark of Kant, she says that "we are not conscripts in the army of virtue, but volunteers"; the fact that we have nothing to say in proof of the irrationality of at least some unjust people, should not alarm us in our own defense and cultivation of justice and benevolence: "it did not strike the citizens of Stalingrad that their devotion to the city and its people during the terrible years of the siege was contingent".
Her book "Natural Goodness" attempts a different line. The question what we have most reason to do, is tied to the idea of the good working of practical reason. This in turn is tied to the idea of the species of an animal as providing a measure of good and bad in the operations of its parts and faculties. Just as you have to know what kind of animal you are dealing with to decide whether its eyesight is good or bad - veterinarian optometrists would surely use different machinery! - so also the question whether someone's practical reason is well developed depends on the kind of animal he is. (This idea is developed in the light of a conception of animal kinds or species as implicitly containing "evaluative" content, which may be criticized on contemporary biological grounds, though it is arguable, even on that basis, that it is very deeply entrenched in human cognition.) In our case what makes for a well constituted practical reason depends on the fact that we are human beings, characterized by certain possibilities of emotion and desire, a certain anatomy and neurological organization, and so forth. Once this step is made, it becomes possible to argue for the rationality of moral considerations in a new way. We begin with the conviction that justice is a genuine virtue. Thus the conviction that the well constituted human practical reason operates with considerations of justice, and that taking account of others in that sort of way is "how human beings live together," understood in a sense that is compatible with it that often they don't. (Just as the fact that "human beings have n teeth" is compatible with it that often they don't.) There is nothing incoherent in the thought that such forms of practical calculation might characterize some kind of rational being; similarly, of course, there is nothing incoherent in the idea of a form of rational life within which such considerations are alien, and only possibly imposed by damaging and disturbing the individual person. Our conviction that justice is a virtue and that considerations of justice are genuine reasons for action is the conviction that the kind of rational being that we are, namely, human beings, is of the first type. Of course it might be suggested that this is precisely not the case, that human beings are of the second kind, and thus that the justice and benevolence we esteem are artificial and false (just as on another sort of view machismo and femininity are discounted as artificial and false, matters of "mere convention" that put one off of the main things.) Such a position was maintained by the Platonic "immoralists" Callicles and Thrasymachus, and perhaps by Nietzsche. In the case of Nietzsche this is apparently to be shown by claiming that justice and benevolence can only be inculcated by warping the emotional apparatus of the individual. Foot's book ends by attempting to defuse the evidence Nietzsche brings against what might be called the common sense position.