225. Hitherto the idea of relation which we have had specially in view has been that of relation in the way of resemblance, and the propositions have been such as represent the most obvious ‘facts of observation’--facts about this or that ‘body,’ man or horse or ball. We have seen that these already suppose the thought of an object qualified, not transitory as are feelings, but one to which feelings are referred on their occurrence as resemblances or differences between it and other objects; but that by an equivocation, which unexamined phraseology covers, between the thought of such an object and feeling proper--as if because we talk of seeing a man, therefore a man were a feeling of colour--Hume is able to represent them as mere data of sense, and thus to ignore the difference between related feelings and ideas of relation. Thus the first step has been taken towards transferring to the sensitive subject, as merely sensitive, the power of thought and significant speech. The next is to transfer to it ideas of those other relations [1] which Hume classifies as ‘relations of time and place, proportion in quantity or number, degrees in any quality’. [2] This done, it is sufficiently equipped for achieving its deliverance from metaphysics. An animal, capable of experiments concerning matter of fact, and of reasoning concerning quantity and number, would certainly have some excuse for throwing into the fire all books which sought to make it ashamed of its animality. [3] [1] The course which our examination of Hume should take was marked out, it will be remembered, by his enumeration of the ‘_natural_’ relations that regulate the association of ideas. It might seem a departure from this course to proceed, as in the text, from the relation of resemblance to ‘relations of time and place, proportion in quantity or number, and degrees of any quality,’ since these appear in Hume’s enumeration, not of ‘_natural_’ but of ‘_philosophical_’ relations. Such departure, however, is the consequence of Hume’s own procedure. Whether he considered these relations merely equivalent to the ‘natural ones’ of resemblance and contiguity, he does not expressly say; but his reduction of the principles of mathematics to data of sense implies that he did so. The treatment of degrees in quality and proportions in quantity as sensible implies that the difference between resemblance and measured resemblance, between contiguity and measured contiguity, is ignored. [2] p. 368 [Book I, part II., sec. V.] [3] If we take in our hand any volume of divinity or school-metaphysics, for instance, let us ask, _Does it contain any abstract reasoning for quantity or number?_ No. _Does it contain any experimental reasoning concerning matter of fact and existence?_ No. _Commit it then to the flames, for it can contain nothing but sophistry and illusion.’_--‘Inquiry concerning the Human Understanding,’ at the end.