to the vast majority of mankind--of all that is supported by
demonstration. For the power to prove implies a sense of the need of
proof, and things which the majority of mankind find practically
important are in ninety-nine cases out of a hundred above proof. The
need of proof becomes as obsolete in the case of assumed knowledge,
as the practice of fortifying towns in the middle of an old and long
settled country. Who builds defences for that which is impregnable
or little likely to be assailed? The answer is ready, that unless
the defences had been built in former times it would be impossible to
do without them now; but this does not touch the argument, which is
not that demonstration is unwise, but that as long as a demonstration
is still felt necessary, and therefore kept ready to hand, the
subject of such demonstration is not yet securely known. Qui
s'excuse, s'accuse; and unless a matter can hold its own without the
brag and self-assertion of continual demonstration, it is still more
or less of a parvenu, which we shall not lose much by neglecting till
it has less occasion to blow its own trumpet. The only alternative
is that it is an error in process of detection, for if evidence
concerning any opinion has long been denied superfluous, and ever
after this comes to be again felt necessary, we know that the opinion
is doomed.

If there is any truth in the above, it should follow that our
conception of the words "science" and "scientific" should undergo
some modification. Not that we should speak slightingly of science,
but that we should recognise more than we do, that there are two
distinct classes of scientific people corresponding not inaptly with
the two main parties unto which the political world is divided. The
one class is deeply versed in those sciences which have already
become the common property of mankind; enjoying, enforcing,