ea not five miles by four.

It is for the purpose of considering these human motives and personal
accidents that I offer these pages; for if we can comprehend Erlon's
error, we shall fill the only remaining historical gap in the story of
Waterloo, and determine the true causes of that action's result.

[Illustration]

There are two ways of appreciating historical evidence. The first is the
lawyer's way: to establish the pieces of evidence as a series of
disconnected units, to docket them, and then to see that they are
mechanically pieced together; admitting, the while, only such evidence as
would pass the strict and fossil rules of our particular procedure in the
courts. This way, as might be inferred from its forensic origin, is
particularly adapted to arriving at a foregone conclusion. It is useless
or worse in an attempt to establish a doubtful truth.

The second way is that by which we continually judge all real evidence
upon matters that are of importance to us in our ordinary lives: the way
in which we invest money, defend our reputation, and judge of personal
risk or personal advantage in every grave case.

This fashion consists in admitting every kind of evidence, first hand,
second hand, third hand, documentary, verbal, traditional, and judging the
general effect of the whole, not according to set legal categories, but
according to our general experience of life, and in particular of human
psychology. We chiefly depend upon the way in which we know that men
conduct themselves under the influence of such and such emotions, of the
kind of truth and untruth which we know they will tell; and to this we add
a consideration of physical circumstance, of the laws of nature, and hence
of the degrees of probability attaching to the events which all this mass
of evidence relates.

It is only by this second method, which is the method of common-sense,
that anything can be made of a doubtful historical point. The legal method
would make of history what it makes of justice. Which God forbid!

Historical points are doubtful precisely because there is conflict of
evidence; and conflict of evidence is only properly resolved by a
consideration of the psychology of witnesses, coupled with a consideration
of the physical circumstances which limited the matter of their testimony.

Judged by these standards, the fatal march and countermarch of Erlon
become plain enough.

His failure to help either Ney or Napoleon was no