neration to generation.
So equally with murders. Even if there were no doubt that the percentage
of such crimes in England had long continued the same, still that fact
would prove nothing as to the uniform reproduction of crime, if it could
be shown that the percentage had ever varied anywhere else--in France or
Italy, for example, or in Dahomey. For it would be mere childishness to
point to the different conditions of England and Dahomey, and to plead
that no more was intended to be said than that, with uniformity of
circumstances there would also be uniformity of results. So much no one,
in the least competent to discuss the subject, would for a moment dream
of disputing. But in political affairs there cannot be uniformity of
circumstances. The aggregate of circumstances from which spring human
motives cannot, from the nature of things, ever be repeated; and, though
a few general causes may continue permanently in operation, they cannot
continue to produce the same identical results; for even though they
could themselves remain stationary, it would be impossible that their
operation should not be affected by the constant change going on around,
or should not partake of an otherwise universal forward movement. In
political affairs there cannot possibly be any recurrence of identical
phenomena; nor can there, except within a very limited period, be any
occurrence of very similar phenomena. But recurrence (and not merely
recurrence, but complete and invariable recurrence) is the very
foundation of science. Without it there can be no scientific laws, and
without such laws--_i.e._, without records of past recurrences--there
can be no sure predictions as to the future. It is only because certain
motions of certain bodies have hitherto been observed to take place with
invariable regularity, that they are expected to continue to do so, and
it is upon that assumption only that we venture to predict that the sun
will rise to-morrow morning, or that an eclipse will take place next
year. But if no event recorded in history has ever yet been known to
occur twice under precisely the same conditions, and as a consequence of
the same causes, what ground can there be for predicting whether or when
any such event will occur again? What possibility is there of
constructing a science of history, when history supplies no materials
for either foundation or superstructure?

There is nothing in this conclusion in the slightest degree opposed