g the causes of unintentional actions or results.

Again, of good and bad things there are three classes. For they can
exist either in men's minds or bodies, or they may be external to both
of these materials, then, as far as they are subordinate to argument,
all the parts must be carefully turned over in the mind, and
conjectures bearing on the subject before us must be derived from each
part.

There is also another class of arguments which is derived from traces
of a fact, as a weapon, blood, an outcry which has been raised,
trepidation, changes of complexion, inconsistency of explanation,
trembling, or any of these circumstances which can be perceived by our
senses, or if anything appears to have been prepared, or communicated
to any one, or if anything has been seen or heard, or if any
information has been given.

But of probabilities some influence us separately by their own weight,
some, although they appear trifling by themselves, still, when all
collected together, have great influence. And in such probabilities as
these there are sometimes some unerring and peculiar distinguishing
characteristics of things. But what produces the surest belief in a
probability is, first of all, a similar instance, then the similarity
of the present case to that instance sometimes even a fable, though it
is an incredible one, has its influence, nevertheless, on men's minds.

XII. _C. F._ What next? What is the principle of definition, and what
is the system of it?

_C. P._ There is no doubt but that definition belongs to the genus,
and is distinguishable by a certain peculiarity of the characteristics
which it mentions, or else by a number of common circumstances, from
which we may extract something which looks like a peculiar property.
But since there is often very great disagreement about what are
peculiar properties, we must often derive our definitions from
contraries, often from things dissimilar, often from things parallel.
Wherefore descriptions also are often suitable in this kind of
address, and an enumeration of consequences, and above all things, an
explanation of the names and terms employed, is most effectual.

_C. F._ You have now then explained nearly all the questions which
arise about a fact, or about the name given to such fact. The next
thing is, when the fact itself and its proper title are agreed upon,
that a doubt arises as to what its character is.

_C. P._ You are quite right.

_C. F._ What divis