224 H. A. PRICHARD : (2) That this is a mis-analysis and that the distinction really understood leads to the contrary conclusion. We can easily trace the origin of the view that we only know things as they appear to us, by taking any case where we dis- tinguish between ' appearance ' and ' reality '. Take, for example, Plato's instance of a straight stick partly submerged in water, which, as we say, looks bent, though in reality it is straight. If some one knows nothing about refraction and confines himself to a single perception, he will assert that the stick is bent ; and if he is asked why, he will answer, ' because I see it to be so '. But if afterwards he sees the stick under other conditions and has learned about refraction, he will say that the stick only looks bent and is really straight. Thus having just identified what the stick is and what it looks, he afterwards draws a distinction between them. And this distinction presents an obvious difficulty to knowledge in general. If a thing is not necessarily what it looks, how are we to learn what it really is ? A thing in the sense in question is an individual ; to get at it therefore we must perceive it. But the perception, to be of value, must give us the thing as it is. And that it does so is implied by the original assertion 'the stick is bent, because it is seen to be so '. Seeing is believing. The sub- sequent assertion, however, denies this, ' the stick only looks bent, but it is not what it looks'. Perception, that is to say, does not give us the thing as it really is. The conclusion at once seems to follow that we only know things as they look or appear to us and not as they are. And from this it is but a short step to the second view that we only know ' appearances ' or ' phenomena '. To put the difficulty shortly. Access to things implies percep- tion. Yet if perception only gives us things as they look and not as they are, access to things as they are is impossible. But per- ception does in fact only give us things as they look, for this is presupposed by the distinction we actually draw between what they look and what they are. This conclusion can only be avoided by maintaining that the reality of the distinction is still compatible with the position that perception at least gives us things as they are in some qualified way ; that after all there is some identity between what things look and what they are. And a closer analysis of the distinction vindicates this identity. That the analysis may be as concrete as possible, it will be well to bear in mind three prominent types of case. (1 ) That of refraction already mentioned. (2) That due to the nature of perspective. E.g. (1) Eailway lines may be said though really parallel to look convergent. (2) A horizontal building may be said to look as though it were lower in its more distant parts. (3) That due to distance from the observer. E.g., the moon may be said to look as large as the sun. These types suggest that the distinction between appearance and
