Condorcet wrote in 1794, ‘The sole foundation for belief in the natural sciences is this idea, that the general laws dictating the phenomena of the universe are necessary and constant. Why should this principle be any less true for the development of the intellectual and moral faculties of man than for the other operations of nature?’ In similar vein, Buckle speaks of the ‘undeviating regularity of the moral world’, which is part of ‘one vast scheme of universal order’. These sentiments catch what I take to be central to passive conceptions, the idea that human action is a natural and determined phenomenon. Our next task is to find the best account we can of causal explanation. Naturalism is the thesis that there is only one mode of explanation, and determinism the thesis that any fact which has an explanation is, together with some other fact, an instance of a natural law. It is therefore vital to understand the notion of a law. I shall present the task as one in epistemology. What is a natural law and how do we know when we have found one?

I grasp these nettles in hope, since I do not at all deny that the moral world has its regularities. Action has antecedents which sometimes wholly and always partly determine it in accordance with natural laws. My doubts about the orthodox rubric for social science will not be followed by a rampage in hermeneutics. None the less I accept neither naturalism nor determinism. To explain x is to find a y such that x because y. Sometimes ‘x because y’ is equivalent to ‘x is caused by y’, but, I shall argue, not always. The equivalence holds for Plastic Man but not for Autonomous Man, whose actions will be presented later as wholly explicable but only partly determined. A lesser aim of the chapter is to destroy any presumption that ‘because’ is always causal.

The job of epistemology, in Kantian terms, is to show how scientific knowledge is possible. On the one hand we have beliefs about the world's furniture and how it works, on the other we have criteria which beliefs must meet to be justified.