Sextus Empiricus: Logic 349
this criterion is apparent or non-evident. And it is certainly not non-
evident; for just as now the appearances alone are assumed to be true.
- But if it is apparent, since the question of which appearances are
true and which false is what is being investigated, the apparent [criterion]
taken up for the judgement of appearances will again be in need of
another apparent criterion, and that will be in need of yet another, [and
so on] to infinity. But it is impossible to make an infinite number of
decisions. So, it is impossible to grasp whether the true things are
only appearances.
90. Similarly, one who says that only non-evident things are true will
not say that they are all true (for he will not say that 'the number of
stars is even' is true and 'the number of stars is odd' is true). If, however,
only some are true, by means of what shall we judge that these non-
evident things are true and those false? Certainly not by an apparent
[criterion]. But if it is a non-evident one, since we are investigating which
ones are true and which false, the non-evident criterion will need another
non-evident criterion for deciding this, and that one will need another,
[and so on] to infinity. Therefore, it is not the case that the true things
are only non-evident.
- The remaining possibility is that of true things some are apparent
and some non-evident. But this too is absurd. For either all appearances
and all non-evident things are true, or some appearances and some non-
evident things. If, then, it is all, again the argument is overturned, since
'there is nothing true' will be granted to be true and it will be said that
'the number of stars is even' and the 'the number of stars is odd' are
true. 92. If, however, only some of the appearances and some of the non-
evident things are true, how shall we decide that these appearances are
true and those not? If it is by means of an apparent [criterion] the
argument will be forced into an infinite regress. If it is by means of a
non-evident criterion, since the non-evident things are in need of judge-
ment, again by means of what will this non-evident [criterion] be judged?
If it is by means of an apparent [criterion], the mode based on circular
reasoning is adduced; and if it is by means of a non-evident [criterion],
the mode forcing an infinite regress is adduced. 93. And regarding non-
evident things, similar things should be said. For he who attempts to
judge things by something non-evident is forced into an infinite regress,
whereas he who attempts to judge things by something apparent is either
always piling on another apparent criterion [to judge the first], [and so
on] to infinity, or by shifting to a non-evident [criterion], falls into circular
reasoning. It is therefore false to say that of true things some are apparent
and some non-evident.
- If, therefore, neither the appearances nor the non-evident things
