We can see at once how much wider this conception of causation is than the modern conception. If we take Mill's definition of a cause as the best expression of modern scientific ideas, we find that he defines a cause as the "invariable and unconditional antecedent of a phenomenon." This cuts out final causes at once. For {270} the final cause is the end, and is not an antecedent in time. It also does not include formal causes. For we do not now think of the concept of a thing as being part of its cause. This leaves us with only material and efficient causes, and these correspond roughly to the modern notions of matter and energy. Even the efficient causes of Aristotle, however, appear on further consideration, to be excluded from the modern idea of causation. For, though the efficient cause is the energy which produces motion, modern science regards it as purely mechanical energy, whereas Aristotle thinks of it, as we shall see, as an ideal force, operating not from the beginning but from the end. But it must not be supposed that, in saying that the modern idea of causation excludes formal and final causes, we mean that Aristotle is wrong in adding them, or that the modern idea is better than Aristotle's. It is not a question of better and worse at all. Modern science does not in any way deny the reality of formal and final causes. It merely considers them to be outside its sphere. It is no business of science whether they exist or not. As knowledge advances, differentiation and division of labour occur. Science takes as its province mechanical causes, and leaves formal and final causes to the philosopher to explicate. Thus, for example, formal causes are not considered by science because they are not, in the modern sense, causes at all. They are what we have called reasons. If we are to explain the existence of an object in the universe it may be necessary to introduce formal causes, concepts, to show why the thing exists, to show in fact its reasons. But science makes no attempt to explain the existence of objects. It takes their {271} existence for granted, and seeks to trace their history and their relations to each other. Therefore it does not require formal causes. It seeks to work out the mechanical view of the universe, and therefore considers only mechanical causes. But Aristotle's theory, as being philosophy rather than science, includes both the principles of mechanism and teleology.