seen that in
the instance adduced, and indeed in most others, where we reason on
things, that precision can only be attained through the medium of
number; for these mineral substances, although similar in external
character, may contain very different proportions of the precious
metals, and their actual value can only be estimated by comparison; that
is, by an analysis, founded in knowledge, to ascertain the per centage
of gold or silver, which must be expressed in numbers: and the
comparison that is instituted concerning general or abstract terms,
must have for its basis the establishment of their legitimate force and
meaning.

When we consult authorities on this subject, and particularly Dr.
Johnson's dictionary, we find that he has given eleven different
significations of the term _reason_, which he defines to be "the power
by which man _deduces_ one proposition from another, or proceeds from
premises to consequences." There is, however, much ambiguity in this
statement; and it would perhaps be impossible, in reasoning concerning
things, (which is to be considered as the most perfect example of this
process,) to adduce an instance, in which one proposition is strictly
_deduced_ from another.

Every proposition is distinct, and independent: numbers, which are
definite, may be added together, and the sum-total exhibited, or a
lesser number subtracted from a greater, and the remainder shown. It is
difficult to say what is really meant by the words "deduces one
proposition from another." On examination, it will be found that every
simple proposition contains some fact or dictum, something set up or
laid down, _aliquid propositum_; and that nothing can be _deduced_ from
it, more than the meaning which the words constituting such proposition
legitimately convey: indeed, it must be evident, that any deduction from
a simple proposition would destroy its force. The sum of our knowledge
consists of individual facts, which are in themselves distinct, as much
as a flock of sheep is the aggregate of the different animals that
compose it; and it is only a misapplication of language, to affirm that
we are able to deduce one proposition from another. One proposition may
tend to explain or illustrate another; but every proposition, correctly
so termed, relates only to itself.

The other mode by which we reason, is on abstract or general terms,
which are not the representatives of individual substances, or the
objects of our perc