ticular
facts_), such as a saying, a momentary act. It is enough that several
persons should have been present when the fact occurred, that they
should have recorded it, and that their writings should have come down
to us. We know what were the words which Luther uttered at the Diet of
Worms; we know that he did not say what tradition puts in his mouth.
This concurrence of favourable conditions becomes more and more frequent
with the organisation of newspapers, of shorthand writers, and of
depositories of documents.

In the case of antiquity and the middle ages historical knowledge is
limited to general facts by the scarcity of documents. In dealing with
contemporary history it is possible to include more and more particular
facts. The general public supposes the opposite of this; it is
suspicious about contemporary facts, with reference to which it sees
contradictory narratives circulating, and believes without hesitation
ancient facts, which it does not see contradicted anywhere. Its
confidence is at its greatest in respect of that history which we have
not the means of knowing, and its scepticism increases with the means of
knowledge.

VI. _Agreement between documents_ leads to conclusions which are not all
of them definitive. In order to complete and rectify our conclusions we
have still to study _the harmony of the facts_.

Several facts which, taken in isolation, are only imperfectly proved,
may confirm each other in such a manner as to produce a collective
certainty. The facts which the documents present in isolation have
sometimes been in reality sufficiently near each other to be connected.
Of this kind are the successive actions of the same man or of the same
group of men, the habits of the same group at different epochs separated
by short intervals, or of similar groups at the same epoch. It is no
doubt possible that one of several analogous facts may be true and
another false; the certainty of the first does not justify the
categorical assertion of the second. But yet the harmony of several such
facts, each proved imperfectly, yields a kind of certainty; the facts do
not, in the strict sense of the word, prove, but they _confirm_[175]
each other. The doubt which attached to each one of them disappears; we
obtain that species of certainty which is produced by the
interconnection of facts. Thus the comparison of conclusions which are
separately doubtful yields a whole which is morally certain. In an
itin