must be, uniformity and regularity in human affairs; or, as Mr. Buckle expresses it, that social phenomena 'are the results of large and general causes which, working on the aggregate of society, _must_ produce certain consequences, without regard to the volition of the particular men of whom the society is composed.' Now, though free agency may co-exist with _invariable_ regularity, it obviously cannot co-exist with _necessary_ regularity, which, consequently, is incompatible likewise with moral responsibility. If men are compelled by the force of circumstances, or by any force, to move only in one direction, they cannot be responsible for not moving in a different direction. Nor is it more to the purpose to undertake a subtle analysis of the nature of causation, and to explain that it does not, properly speaking, involve compulsion, but simply means invariable antecedence. Let it be that a cannon-ball does not really knock down the wall against which it strikes, and that it would be more correct to say that the ball impinges and the wall falls; though, seeing that the wall would not have fallen unless the ball had impinged, the distinction is too nice for ordinary apprehension. As understood, at any rate, by the joint headmasters of the new school, causation does involve compulsion. 'Men's actions,' say they, 'are the product not of their volition, but of their antecedents,' and 'result from large and general causes which must produce certain consequences.' Neither, if this be so, is it of any avail to suggest that, possibly, the large and general causes in question may be of only temporary operation. 'It may be that the rules,' in accordance with which the sun has hitherto risen every morning since the creation of the world, 'will hold good only for a time.' It may be that the springs, whatever they are, by which the universe is kept in motion, may require to be periodically wound up like the works of a clock, and that, unless this be done, 'on some particular day out of many billions,' the sun may fail to rise, just as the clock, if suffered to run down, would stop on the eighth day. The conjecture would, of course, be not less applicable to social than to natural laws. It is conceivable that the large general causes assumed to regulate human actions might lose their efficacy at the end of a certain cycle, when mankind might either have to recommence a social revolution similar to the one just completed, or might have to begin an entirely different revolution under entirely different laws. Be it so. Still, if the causes, as long as they remained in operation, possessed a compulsory character--if, during the continuance of the supposed cycle, men were bound to act in a certain way in accordance with certain laws, and irrespectively of their own volition--what would it matter that those laws were not eternal and immutable? For the time being men would no more be free agents than the hands of a clock, while the clock was wound up.