This line of argument, however, only removes the difficulty one step further. Sensation is said always to reproduce faithfully the picture which affects the organs of sense, but the pictures do not always reproduce the object with equal faithfulness. How then can a faithful picture be known from one which is not faithful? To this question the Epicurean system can furnish no real answer. To say that the wise man knows how to distinguish a faithful from an unfaithful picture [1145] is to despair of an absolute standard at all, and to make the decision of truth or error depend upon the individual’s judgment. Such a statement reduces all our impressions of the properties of things to a relative level. If sensation does not show us things themselves, but only those impressions of them which happen to affect us, it does not supply us with a knowledge of things as they are, but as they happen to be related to us. It was, therefore, a legitimate inference from this theory of knowledge for Epicurus to deny that colour belongs to bodies in themselves, since some only see colour in the dark, whilst others do not. [1146] Like his predecessor, Democritus, he must have been brought to this view by his theory of atoms. Few of the properties belong to atoms which we perceive in things, and hence all other properties must be explained as not belonging to the essence, but only to the appearance of things. [1147] The taste for speculation was, however, too weak, and the need of a direct truth of the senses too strong in Epicurus for him to be able to turn his thoughts in this direction for long. Whilst allowing to certain properties of things only a relative value, he had no wish to doubt the reality of objects, nor to disparage the object-pictures which furnish us with sensations. [1148] CHAPTER XVII. THE EPICUREAN VIEWS ON NATURE. [A. General views on nature.] [(1) Object, value, and method of the study of nature.]