orrect reporters of the object with which they profess to
acquaint us. Food, which is the same as far as _sight_ and _touch_ are
concerned, _tastes_ differently to different individuals; fire, which is
the same to the _eye_, communicates a sensation of _pain_ at one time,
of _pleasure_ at another; the oar _appears_ crooked in the water, while
the _touch_ assures us it is as straight as before it was immersed.[158]
Again, in dreams, in intoxication, in madness, impressions are made upon
the mind, vivid enough to incite to reflection and action, yet utterly
at variance with those produced by the same objects when we are awake,
or sober, or in possession of our reason.[159]

It appears, then, that we cannot prove that our senses are _ever_
faithful to the things they profess to report about; but we do know they
_often_ produce erroneous impressions of them. Here then is room for
endless doubt; for why may they not deceive us in cases in which we
cannot detect the deception? It is certain they _often_ act irregularly;
is there any consistency _at all_ in their operations, any law to which
these varieties may be referred?

It is undeniable that an object often varies in the impression which it
makes upon the mind, while, on the other hand, the same impression may
arise from different objects. What limit is to be assigned to this
disorder? is there any sensation strong enough to _assure_ us of the
presence of the object which it seems to intimate, any such as to
preclude the possibility of deception? If, when we look into a mirror,
our minds are impressed with the appearance of trees, fields, and
houses, which are unreal, how can we ascertain beyond all doubt whether
the scene we directly look upon has any more substantial existence than
the former?[160]

From these reasonings the Academics taught that nothing was certain,
nothing was to be known ([Greek: katalepton]). For the Stoics
themselves, their most determined opponents, defined the [Greek:
kataleptike phantasia] (the phantasy or impression which involved
knowledge[160a]) to be one that was capable of being produced by no object
except that to which it really belonged.[161]

Since then we cannot arrive at knowledge, we must suspend our decision,
pronounce absolutely on nothing, nay, according to Arcesilas, never even
form an opinion.[162] In the conduct of life, however, probability[163]
must determine our choice of action; and this admits of different
degrees. The