deprived of free will, or that his freedom of will is more or less restricted, and that he is therefore irresponsible. In my own opinion, that proposition would be by no means an easy one to establish. I fancy that a man may be insane and yet capable, within very wide limits, of being good or bad, and that therefore we must at any rate hold that he has still some power of free will. The bearing of this upon the question of moral responsibility brings us within sight of some delicate problems. But, however this may be, the criterion by which we shall have to judge whether we are believers in free will or determinists will be the same. The problem is essentially, is this man accessible to the motives by which normal men regulate their conduct? or does he so far deflect from the typical constitution, however that constitution may be precisely defined, that his conscience or his affections or his intellectual powers are unable to act according to the general laws of human nature?

Having said so much, I think that I may proceed to this conclusion, that the theory of heredity can make no real difference whatever to our problem. There is a difficulty for the metaphysician—the difficulty which is involved in discussions between materialists and idealists, determinists and believers in free will. I do not deny the existence of that difficulty. I only say that the question of heredity is altogether irrelevant to the difficulty. The desire to treat ethical problems by the methods of science may predispose a thinker to materialism, and may at the same time lead him to attach particular importance to the doctrine of heredity. But that doctrine only takes note of facts which every theory has to state in its own phraseology, and do not alter the ultimate problem.

Let us, in fact, go back to our murderer. I am not responsible, he says, because I am determined by the processes in my brain. I am a mere machine, grinding out one set of actions or another as external accidents set my wheels and pulleys in motion. If that argument be fatal to moral responsibility, or to the belief that any truly moral action exists (a point which I do not argue), it will no doubt remove the moral element from the treatment both of murderers and madmen. They might still require different measures, just as we treat a machine differently when we consider that it is not of the normal construction, or that its various parts have somehow got out of gear, so that we can no longer, for example, expect that the mainspring will transmit its motion to the wheels. But, in any case, if the dependence upon the body be a fatal objection to morality in the highest sense, the circumstance that the body is made upon the plan of previously existing bodies makes no additional difficulty. If we could suppose every brain to be started afresh by a fortuitous concourse of atoms, the difficulty would be neither increased nor diminished. The problem, are we automatic? and the validity of the inference, is morality meaningless? are questions altogether independent of the question, what particular kind of automata are we? and do we or do we not resemble a previous generation of automata?

If, however, we reply to the criminal that he is not a machine or an automaton, but a responsible, reasoning, and thinking being, we do not get away from the facts. We then assert that he is responsible because he possesses a certain moral constitution. But whatever words we may use to express the facts correctly, we must still allow that there is such a correlation between soul and body (if those old-fashioned words be admissible) that the health of his moral constitution depends at every instant upon the health of his nervous system and his brain. It may be shattered or destroyed by an injury; and, if this be so, what does it matter whether the injury—say the defective shape of the skull, which causes pressure on the brain—is due to some accident or to a connate malformation due to his parents? The difficulty, if it be difficulty, is that the want of responsibility is due to some cause, accidental relatively to him; and it matters not whether that cause be in his parents' constitution or in some other combination of circumstances. In any case, we have to suppose, whatever the relation of mind and body, we must at least assume that a man is born with some character. Like everything that exists, he has certain definite qualities which he did not make for himself, and upon which his subsequent development depends. And, if that be once admitted, the whole difficulty still occurs, and the question as to whether the origin of these innate qualities be derived from his parents or from a something else is a mere matter of detail.

In fact, the confusion seems to me to arise from the vague phraseology which induces us to accept, virtually at least, the mental attitude of Dean Swift in Er the Pamphylian. We speak as if the man were an independent entity, lying somehow outside the chain of cause and effect, and arbitrarily plunged into it; nay, as if even his inner constitution were something superinduced upon his nature. It is really an absurd abstraction to distinguish between the man and his character, as though he meant a something existing without a character, and afterwards run into a mould by fate. The character is the man in certain relations, and he can never exist without it, any more than a piece of matter can be outside of all particular times and places. If the doctrine of free will and moral responsibility be so interpreted as to imply our acceptance of such fallacies, I can only say that it appears to me to be irreconcilable with the most undeniable facts. But I am very far from supposing that any intelligent supporter of the doctrine would state it in such a form. He would admit as fully as I do the facts, and, if they can be admitted and reconciled to the doctrine of moral responsibility, certainly the doctrine of heredity can be so reconciled. The only peculiarity of the doctrine is, that it has called attention to an order of facts which must in any case be recognised by every philosopher; and that it helps, therefore, to disperse a fallacy which only requires articulate statement to show its radical want of logic or even conceivability. We are, beyond all doubt, affected somehow, and affected profoundly, by our environment; and this particular form of relation to other beings has no more bearing upon the problem than the other forces which have been recognised ever since speculation began.

There is, however, another side from which I must briefly consider the question of heredity; and it is a side which, I think, is really more important, because it involves issues of facts, and has suggested some more reasonable prejudices. It is, undoubtedly, very common that when a theory has obtained a certain currency it should be applied rashly beyond its proper limits. When the speculations of Darwin encouraged us to believe that the natural selection might be analogous to artificial selection, that different species of animals have been produced as varieties of dogs and pigeons have been produced by breeders, it was, at least, tempting to apply the same formula directly to other cases. Some men of science have endeavoured to show that genius or criminality is hereditary; and that, if one man writes a great poem and another picks a pocket, it is always in virtue of their hereditary endowment. Within certain limits, this statement is not surprising, and I shall be very glad when men of science can tell us what those limits may be. Without being a man of science, I fully believe that our congenital characteristics form, as I have said, certain impassable limits to our development. One baby is a potential Shakespeare,