really be perceived;" and the second is--"Of those
perceptions between which there is no difference, it is impossible that
some should be of such a character that they can be perceived, and others
of such a character that they cannot."

But their other propositions they defend by numerous and varied arguments,
and they likewise are two in number. One is--"Of those things which appear,
some are true and others false;" the other is--"Every perception which
originates in the truth, is of such a character as it might be of, though
originating in what is false." And these two propositions they do not pass
by, but they expand in such a manner as to show no slight degree of care
and diligence. For they divide them into parts, and those also large
parts; first of all into the senses, then into those things which are
derived from the senses, and from universal custom, the authority of which
they wish to invalidate. Then they come to the point of laying it down
that nothing can be perceived even by reason and conjecture. And these
universal propositions they cut up into more minute parts. For as in our
yesterday's discussion you saw that they acted with respect to the senses,
so do they also act with respect to everything else. And in each separate
thing which they divide into the most minute parts, they wish to make out
that all these true perceptions have often false ones added to them, which
are in no respect different from the true ones; and that, as they are of
such a character, nothing can be comprehended.

XIV. Now all this subtlety I consider indeed thoroughly worthy of
philosophy, but still wholly unconnected with the case which they advocate
who argue thus. For definitions, and divisions, and a discourse which
employs these ornaments, and also similarities and dissimilarities, and
the subtle and fine-drawn distinctions between them, belong to men who are
confident that those arguments which they are upholding are true, and
firm, and certain; and not to men who assert loudly that those things are
no more true than false. For what would they do if, after they had defined
anything, some one were to ask them whether that definition could be
transferred to something else? If they said it could, then what reason
could they give why it should be a true definition? If they said no,--then
it must be confessed, since that definition of what is true cannot be
transferred to what is false, that that which is explained by that
defin