the double sense of the word ‘likeness,’ which seems to authorize this use of likeness for sameness. Likeness may mean my specific experience of resemblance—and that of course itself is not identity—or it may mean the real partial sameness in character of two things whether to me they resemble or not. Thus ‘exact likeness’ can be used for the identical character which makes the point of likeness, and it need not mean the mere likeness which can be opposed to identity. And where exact likeness does not mean the identical character, bankruptcy at once is patent.[1]
We are warned, “You must not say that two notes are the same note, or that two peas have the same colour, for that is to prove yourself incompetent to draw an elementary distinction; or rather you may say this with us, if with us you are clear that you do not mean it, but mean with us mere resemblance.” And when we ask, Are the notes and colours then really different? we hear that ‘the likeness is exact.’ But with this I myself am not able to be satisfied. I want to know whether within the character of the sounds and within the character of the colours there is asserted any difference or none. And here, as I understand it, the ways divide. If you mean to deny identity, your one consistent course is surely to reply, “Of course there is a difference. I know what words mean, and when I said that it was not the same but only alike, I meant to assert an internal diversity, though I do not know exactly what that is. Plainly for me to have said in one breath, The character has no difference and yet it is not the same character, would have been suicidal.” And this position, I admit, is so far self-consistent; but it ends on all sides in intellectual ruin. But the other way, so far as I understand it, is to admit and to assert that in exact likeness there is really no difference, to admit and to assert that it involves a point of resemblance in which internally no diversity is taken to exist, and which we use logically on the understanding that divergence of character is excluded—and then, on the other side, to insist that here we still have no
- ↑ I may perhaps be allowed to illustrate the above by an imaginary dialogue. “Is that piece of work the same?” “Well, it’s exactly like.” “You’re sure?” “Oh yes, it’s identical, it’s a fac-simile.” “H’m, it looks exactly like, but, as I’ve examined the other, I’d rather take that, though I dare say there’s really no difference.” The “looking exactly like,” the producing the same impression, implies of course a real identity in the two things, but as I do not know what that is, I do not know if it is what I want. It is this ambiguity of ‘likeness’ which gave its plausibility to J. S. Mill’s doctrine of reasoning from particular to particular, and it is this again which has enabled Mr. Hobhouse (pp. 280-5) to represent that Mill’s doctrine, once held to be original and revolutionary, consists really in the view that you never do proceed direct from particular to particular, but always through a universal. The task that still awaits Mr. Hobhouse is the proof that, when Mill talked of Association by Similarity, he always meant nothing whatever but Redintegration through identity. But I am not persuaded after all that Mill must have been a prophet because he has at last found a disciple to build his sepulchre.
