It is possible to accept Hume's theory in principle {82} without pledging oneself to all his incidental contentions. Causation, as a general law, may be known only by experience, whether we can or cannot think of it as a pure abstraction. And we may interpret it in terms of unconditional antecedence and consequence, while discarding his apparent assumption of an inscrutable connection between the two; a mysterious necessity for the production of the one by the other, for which it is felt that a reason exists, but for which our reason cannot account. It is inconceivable that our knowledge of any given sequence could be increased, except by the disclosure of intermediate sequences, making their continuity, in space and time, more absolute than we had before perceived, until the whole process has been resolved into a transference of momentum from one molecule to another--a change for which, according to Hume, no reason can be given. Nor, on his principles, would it help us to explain such transferences by bringing them under the law of the Conservation of Energy. For, although this would be a great triumph for science, his philosophy demands a reason why the quantity of energy should remain unalterable for ever. It is a mistake, shared by Hume with his opponents, to suppose that the common sense of mankind ever saw more than invariable sequence in the relation of cause and effect, or ever interpolated a mysterious power between them. In the famous verse, "Let there be light, and there was light," it is the instantaneity of succession, not the interpolation of any exerted effort, that so impresses the imagination. And when Shakespeare wants to illustrate logical compulsion in conduct, his reference is to an instance of invariable succession:-- {83} This above all,--to thine own self be true; And it must follow, as the night the day, Thou canst not then be false to any man. Indeed, I think it will be found on examination that when we associate the idea of power, or of necessity, with causal sequences, it is not in connection with a case of causation here and now, but rather in reference to similar effects that may be expected from the same cause elsewhere or at another time. And that "custom," by which Hume seeks to explain our belief in the "power" of the cause to produce its effect as well as the "necessity" of the connection between them, rather acts negatively by eliminating all other antecedents as possible causes than positively by setting up a habit of thinking about a particular antecedent and consequent at the same time. And that is why a burnt child needs no repetition of the experiment to be convinced that contact with fire was the cause of its pain. The very novelty of the experiment was enough to eliminate any explanation other than that of contact with the flame.