As also does the ‘observation concerning identity,’ which the comparison involves.
301. Relation in the way of identity is treated by Hume in the third part of the Treatise [1] pretty much as he treats contiguity and distance. He admits that it does not depend on the nature of any ideas so related—in other words, that it is not constituted by feelings as they would be for a merely feeling consciousness—yet he denies that the mind ‘in any observations we may make concerning it’ can go beyond what is immediately present to the senses. Directly afterwards, however, we find that there is a judgment of identity which involves a ‘conclusion beyond the impressions of our senses’—the judgment, namely, that an object of which the perception is interrupted continues individually the same notwithstanding the interruption. Such a judgment, we are told, is a supposition founded only on the connection of cause and effect. How any ‘observation concerning identity’ can be made without it is not there explained, and, pending such explanation, observations concerning identity are freely taken for granted as elements given by sense in the experience from which the judgment of cause and effect is derived. In the second chapter of Part IV., however, where ‘belief in an external world’ first comes to be explicitly discussed by Hume, we find that ‘propensities to feign’ are as necessary to account for the judgment of identity as for that of necessary connection. If that chapter had preceded, instead of following, the theory of cause and effect as given in Part III., the latter would have seemed much less plain sailing than to most readers it has done. It is probably because nothing corresponding to it appears in that later redaction of his theory by which Hume sought popular acceptance, that the true suggestiveness of his speculation was ignored, and the scepticism, which awakened Kant, reduced to the commonplaces of inductive logic. To examine its purport is the next step to be taken in the process of testing the possibility of a ‘natural history’ of knowledge. Its bearing on the doctrine of cause will appear as we proceed.
[1] P. 376. [Book I, part III., sec. II.]
Identity of objects an unavoidable crux for Hume. His account of it.
302. The problem of identity necessarily arises from the fusion of reality and feeling. We must once again recall the propositions in which Hume represents this fusion—that ‘everything which enters the mind is both in reality and appearance as the perception;’ that ‘so far as the senses are judges, all perceptions are the same in the manner of their existence;’ that ‘perceptions’ are either impressions, or ideas which are ‘fainter impressions;’ and ‘impressions are internal and perishing existences, and appear as such.’ If these propositions are true—and the ‘new way of ideas’ inevitably leads to them—how is it that we believe in ‘a continued existence of objects even when they are not present to the senses,’ and an existence ‘distinct from the mind and perception’? They are the same questions from which Berkeley derived his demonstration of an eternal mind—a demonstration premature because, till the doctrine of ‘ideas,’ and of mind as their subject, had been definitely altered in a way that Berkeley did not attempt, it was explaining a belief difficult to account for by one wholly unaccountable. Before Theism could be exhibited with the necessity which Locke claimed for it, it was requisite to try what could be done with association of ideas and ‘propensities to feign’ in the way of accounting for the world of knowledge, in order that upon their failure another point of departure than Locke’s might be found necessary. The experiment was made by Hume. He has the merit, to begin with, of stating the nature of identity with a precision which we found wanting in Locke. ‘In that proposition, an object is the same with itself, if the idea expressed by the word object were no ways distinguished from that meant by itself, we really should mean nothing.’ ‘On the other hand, a multiplicity of objects can never convey the idea of identity, however resembling they may be supposed. … Since then both number and unity are incompatible with the relation of identity, it must lie in something that is neither of them. But at first sight this seems impossible.’ The explanation is that when ‘we say that an object is the same with itself, we mean that the object existent at one time is the same with itself existent at another. By this means we make a difference betwixt the idea meant by the word object and that meant by itself without going the length of number, and at the same time without restraining ourselves to a strict and absolute unity.’ In other words, identity means the unity of a thing through a multiplicity of times; or, as Hume puts it, ‘the invariableness and uninterruptedness of any object through a supposed variation of time’. [1]
[1] Pp. 489, 490. [Book I, part IV., sec. II.]
Properly with him it is a fiction, in the sense that we have no such idea. Yet he implies that we have such idea, in saying that we mistake something else for it.
303. Now that ‘an object exists’ can with Hume mean no more than that an ‘impression’ is felt, and without succession of feelings according to him there is no time. [1] It follows that unity in the existence of the object, being incompatible with succession of feelings, is incompatible also with existence in time. Either then the unity of the object or its existence at manifold times—both being involved in the conception of identity—must be a fiction; and since ‘all impressions are perishing existences,’ perishing with a turn of the head or the eyes, it cannot be doubted which it is that is the fiction. That the existence of an object, which we call the same with itself, is broken by as many intervals of time as there are successive and different, however resembling, ‘perceptions,’ must be the fact; that it should yet be one throughout the intervals is a fiction to be accounted for, Hume accounts for it by supposing that when the separate ‘perceptions’ have a strong ‘natural relation’ to each other in the way of resemblance, the transition from one to the other is so ‘smooth and easy’ that we are apt to take it for the ‘same disposition of mind with which we consider one constant and uninterrupted perception;’ and that, as a consequence of this mistake, we make the further one of taking the successive resembling perceptions for an identical, i.e. uninterrupted as well as invariable object. [2] But we cannot mistake one object for another unless we have an idea of that other object. If then we ‘mistake the succession of our interrupted perceptions for an identical object,’ it follows that we have an idea of such an object—of a thing one with itself throughout the succession of impressions—an idea which can be a copy neither of any one of the impressions nor, even if successive impressions could put themselves together, of all so put together. Such an idea being according to Hume’s principles impossible, the appearance of our having it was the fiction he had to account for; and he accounts for it, as we find, by a ‘habit of mind’ which already presupposes it. His procedure here is just the same as in dealing with the idea of vacuum. In that case, as we saw, having to account for the appearance of there being the impossible idea of pure space, he does so by showing, that having ‘an idea of distance not filled with any coloured or tangible object,’ we mistake this for an idea of extension, and hence suppose that the latter may be invisible and intangible. He thus admits an idea, virtually the same with the one excluded, as the source of the ‘tendency to suppose’ which is to replace the excluded idea. So in his account of identity. Either the habit, in virtue of which we convert resembling perceptions into an identical object, is what Hume admits to be a contradiction, ‘a habit acquired by what was never present to the mind’; [3] or the idea of identity must be present to the mind in order to render the habit possible.
[1] ‘Wherever we have no successive perceptions, we have no notion of time.’ (p. 342) [Book I, part II., sec. III.].
[2] P. 492. [Book I, part IV., sec. II.]