In order for the duty of friendship to be practically possible, and for gratitude and beneficence to be unencumbered virtues, Kant need not have held that human beings are basically good. He need only have understood that they are social beings, with desires for both independence and connection, autonomy and affiliation, and purposes that are not always merely their own. I have argued that because he did not, his moral theory is flawed in three important respects.In Kant's theory, morality is (...) only possible because humans are rational, and necessary because (absent morality) they are self-interested individuals, egoistically motivated, distrustful and isolated. When this view is applied to particular questions of friendship and the virtues, it becomes a distorting medium. It is my contention that a more adequate theory of character would result from the application of Kantian moral principles to a sounder, more social conception of human nature.In conclusion, I wish to make it clear that I am not arguing that human beings are basically good, only that they are essentially social. In his political writings Kant has emphasized the social dimension of reason itself, and I believe he would agree with Thomas Nagel's recent claim about human nature and morality: To say that altruism and morality are possible in virtue of something basic to human nature is not to say that men are basically good. Men are basically complicated; how good they are depends on whether certain conceptions and ways of thinking have achieved dominance, a dominance which is precarious in any case. Thomas Nagel, The Possibility of Altruism (Princeton: Princeton University Press, 1970), p. 146. (shrink)