ese, we must admit that Caesar is
mortal. We may make a list of possible inferences of this kind, and
point out under what circumstances the truth of two statements implies
the truth of a third, and under what circumstances the inference cannot
be made. Our results can be set forth in a system of symbols. As in
mathematics, we may abstract from the particular things reasoned about,
and concern ourselves only with the forms of reasoning. This gives us
the theory of the _syllogism_; it is a part of logic in which the
mathematician is apt to feel very much at home.

But this is by no means all of logic. Let us consider the following
points:--

(1) We are not concerned to know only what statements may be made on
the basis of certain other statements. We want to know what is true
and what is false. We must ask: Has a man the right to set up these
particular statements and to reason from them? That some men accept as
true premises which are repudiated by others is an undoubted fact.
Thus, it is maintained by certain philosophers that we may assume that
any view of the universe which is repellant to our nature cannot be
true. Shall we allow this to pass unchallenged? And in ethics, some
have held that it is under all circumstances wrong to lie; others have
denied this, and have held that in certain cases--for example, to save
life or to prevent great and unmerited suffering--lying is permissible.
Shall we interest ourselves only in the deductions that each man makes
from his assumed premises, and pay no attention to the truth of the
premises themselves?

(2) Again. The vast mass of the reasonings that interest men are
expressed in the language that we all use and not in special symbols.
But language is a very imperfect instrument, and all sorts of
misunderstandings are possible to those who express their thoughts in
it.

Few men know exactly how much is implied in what they are saying. If I
say: All men are mortal, and an angel is not a man; therefore, an angel
is not mortal; it is not at once apparent to every one in what respect
my argument is defective. He who argues: Feathers are light; light is
contrary to darkness; hence, feathers are contrary to darkness; is
convicted of error without difficulty. But arguments of the same kind,
and quite as bad, are to be found in learned works on matters less
familiar to us, and we often fail to detect the fallacy.

Thus, Herbert Spencer argues, in effect, in the fo