Furthermore, among simple ideas of sensation, Locke made the distinction between ideas of primary and ideas of secondary qualities. The former are resemblances or copies of the actual properties of things ; the ideas we have of the extension, figure, texture and motion of the parts of things accurately represent the nature of the things that give rise to them. The latter are not resemblances of any characteristics possessed by things them selves, but are effects produced, through the operation upon our senses, of powers which things possess by reason of their primary qualities.
But in the light of Locke's investigation, he was convinced that he could push his way further. Locke had maintained that the range of our knowledge did not extend beyond the range of our "ideas"; and had, at the same time, not only detected in knowl edge a reference to "things" other than "ideas" but had intro duced into his catalogue of ideas those called abstract, which were confessedly devoid of the characteristics of "ideas" as supplied by sense or reflection. Indeed, in both the ordinary and the scientific notion of external things, Berkeley discerned a striking illustration of the result of attempting to carry out the illusory process of abstraction. For these external things were supposed to have somehow a place in the world of perceived reality, and yet to be of such a nature as to preclude the possibility of their being perceived. There is first withdrawn from them every characteristic necessary to constitute them possible objects of experience, and then they are still spoken of as though they did form part of our experience. Locke would have repudiated the doctrine of univer salia ante rem, but what he actually meant by such terms as "matter," "substance," and the like were just universals of this description, i.e., hypostatised abstractions. Berkeley's polemic against "abstract ideas" was intended, therefore, to prepare the way for his own construction. Although discarding Locke's doc trine of abstraction, he recognises that generalisation is indis pensable for knowledge and for communication of knowledge.
Locke had not ignored, but had certainly not done justice to, the function that accrues to an "idea" in virtue of its being representative of other "ideas"; and it is this function upon which Berkeley fastens in the present connection. "An idea which, con sidered in itself is particular, becomes general by being made to represent or stand for all other particular ideas of the same sort." It must be confessed, however, that important though the relation of sign and that which is signified is in the development of knowledge, it will not bear the weight that Berkeley here reposes upon it, for the "sort" or species must in some way have been already determined before any particular idea could "stand for" or "represent" it.