Page:The works of Plato, A new and literal version, (vol 1) (Cary, 1854).djvu/374

362 object seen nor in the eye itself, but results from the application of the eye to the object, and so is the intermediate production of both. Again if you compare six with four they appear to be half as many again, but if with twelve, only the half, whence it appears that the same number is at one time great, at another small, which would not he the case if numbers had a fixed and determined magnitude. The principle then on which all things depend is this, That the universe is nothing but motion, of which there are two species, the one active, the other passive, by the union of which that which is perceivable and perception itself consist. Thus when the eye and a corresponding object, meeting together, produce whiteness and its connate perception, the eye sees, and becomes not vision, but a seeing eye, and the object itself becomes not whiteness but white: so that nothing is essentially one, but is always being produced by something else, and therefore the word "being" must be entirely done away with. But here it may be objected that the perceptions produced in persons who dream, or are diseased or mad, are utterly false, and so far are the things that appear to them from existing, that none of them have any real existence at all; how then can it be said that perception is science, and that things which appear to every one are to that person what they appear to be? The answer is, that the things which appear are most certainly true to the Percipient; Just as if wine appears bitter to a sick person, to him it is certainly bitter; and again with regard to dreams, there is no certain way of distinguishing a state of being awake from dreaming; and as the object perceived and the percipient exist or are produced by relation to each other, neither exists or is produced of itself, but the object perceived does exist in relation to the percipient and to him is true, so that he has a scientific knowledge of what he perceives.

Socrates then proposes to examine the correctness of Prota-