consciousness, but as a difference in one of their elements between two complexes or wholes. That is to say, the two purely visual sensations cannot be brought into clear consciousness and recognised as compared with each other alone, but only come out clearly as combined with certain other elements in complexes or wholes; it is the presence of two or more such wholes, which we wish to compare, that primarily conditions the narrowing of the attention to the particular similar or dissimilar elements; and it is from the presence of two or more such wholes, of which we are conscious as compared, that we are led to infer the presence of single psychic elements in a dim unanalytic consciousness, as a necessary condition to the possibility of all ordinary comparison and classification.
When I form the concept of length, by comparing two objects in length and affirming agreement, and then recognising as a distinct element that in which they agree, I certainly do not compare the objects simply as wholes, but compare the lengths; and I must certainly have had these elements in mind in some way in which I had not the other elements which go to make up the object. Whether I can call into clear consciousness the psychic elements present during the operation or not, it does not much matter; I evidently have specialised, selected one element from among others, and compared length with length as element with element. The name which we give to such resemblances is the name representing a general or abstract idea.
Hume warmly applauds the position taken by Berkeley with reference to the abstract idea, calling it "one of the greatest and most valuable discoveries that have been made of late years in the republic of letters," and he undertakes to confirm it with proofs that he hopes will put it "beyond all doubt and controversy". For the same purpose for which I have quoted the two extracts from Berkeley, I will quote the last part of the section which he devotes to the establishment of this position: