About abstract names, or the names of attributes, see Mr. John Stuart Mill's 'System of Logic,' i. 2, 4, p. 30, edit. 5th. "When only one attribute, neither variable in degree nor in kind, is designated by the name--as visibleness, tangibleness, equality, &c.--though it denotes an attribute of many different objects, the attribute itself is always considered as _one_, not as _many_." Compare, also, on this point, p. 153, and a note added by Mr. Mill to the fifth edition, p. 203, in reply to Mr. Herbert Spencer. The _oneness_ of the attribute, in different subjects, is not conceded by every one. Mr. Spencer thinks that the same abstract word denotes one attribute in Subject A, and another attribute, though exactly like it, in Subject B (Principles of Psychology, p. 126 seq.) Mr. Mill's view appears the correct one; but the distinction (pointed out by Archbishop Whately) between _undistinguishable likeness_ and _positive identity_, becomes in these cases imperceptible or forgotten. Aristotle, however, in the beginning of the Categories ranks [Greek: ê( ti/s grammatikê\] as [Greek: a)/tomon kai\ _e(\n a)rithmô=|_] (pp. 1, 6, 8), which I do not understand; and it seems opposed to another passage, pp. 3, 6, 15. The argument between two such able thinkers as Mr. Mill and Mr. Spencer, illustrates forcibly the extreme nicety of this question respecting the One and the Many, under certain supposable circumstances. We cannot be surprised that it puzzled the dialecticians of the Platonic Aristotelian age, who fastened by preference on points of metaphysical difficulty.] [Footnote 40: See interesting remarks on the application of this logical distinction in Galen, De Methodo Medendi, Book iii. vol. x. p. 130 seq. Aristotle and Theophrastus both dwelt upon it.] [Side-note: Plato never expected to make his Ideas fit on to the facts of sense: Aristotle tried to do it and partly succeeded.]