352 HUME experimental method, Is in fact equivalent to the statement of the principle implied in Locke s Essay, that the problems of psychology and of theory of knowledge are identical. And this view is the char acteristic of what we may call the English school of philosophy. In order to make perfectly clear the full significance of the prin ciple which Hume applied to the solution of the chief philosophical questions, it is necessary to render somewhat more precise and com plete the statement of the psychological view which lies at the foundation of the empirical theory, and to distinguish from it the problem of the theory of knowledge upon which it was brought to bear. Without entering into details, which it is the less necessary to do because the subject has been recently discussed with great fulness in works readily accessible, it may be said that for Locke as for Hume the problem cf psychology was the exact description of the contents of the individual mind, and the determination of the conditions of the origin and development of conscious experience in the individual mind. And the answer to the problem which was furnished by Locke is in effect that with which Hume started. The conscious experience of the individual is the result of interaction between the individual mind and the universe of things. It is evi dent that this solution presupposes a peculiar conception of the general relation between the mind and things which in itself re quires justification, and which, so far at least as the empirical theory was developed by Locke and his successors, could not be obtained from psychological analysis. Either we have a right to the assump tion contained in the conception of the individual mind as standing in relation to tilings, in which case the grounds of the assumption must be sought elsewhere than in the results of this reciprocal re lation, or we have no right to the assumption, in which case refer ence to the reciprocal relation can hardly be accepted as yielding any solution of the psychological problem. But in any case, and, as we shall see, Hume endeavours so to state his psychological premises as to conceal the assumption made openly by Locke, it is apparent that this psychological solution does not contain the answer to the wider and radically distinct problem of the theory of knowledge. For here we have to consider how the individual in telligence comes to know any fact whatsoever, and what is meant by the cognition of a fact. With Locke, Hume professes to regard this problem as virtually covered or answered by the fundamental psychological theorem ; but the superior clearness of his reply enables ITS to mark with perfect precision the nature of the difficulty inher ent in the attempt to regard the two as identical. For purposes of psychological analysis the conscious experience of the individual mind is taken as given fact, to be known, i.e., observed, discrimin ated, classified, and explained in the same way in which any one special portion of experience is treated. Now if this mode of treat ment be accepted as the only possible method, and its results assumed to be conclusive as regards the problem of knowledge, the fundamental peculiarity of cognition is overlooked. In all cog nition, strictly so-called, there is involved a certain synthesis or relation of parts of a characteristic nature, and if we attempt to discuss this synthesis as though it were in itself but one of the facts forming the matter of knowledge, we are driven to regard this relation as being of the quite external kind discovered by observa tion among matters of knowledge. The difficulty of reconciling the two views is that which gives rise to much of the obscurity in Locke s treatment of the theory of knowledge; in Hume the effort to identify them, and to explain the synthesis which is essential to cognition as merely the accidental result of external relations among the elements of conscious experience, appears with the utmost clear ness, and gives the keynote of all his philosophical work. The final perplexity, concealed by various forms of expression, comes forward at the close of the Treatise as absolutely unsolved, and leads Hume, as will be pointed out, to a truly remarkable confession of the weak ness of his own system. While, then, the general idea of a theory of knowledge as based upon psychological analysis is the groundwork of the Treatise, it is a particular consequence of this idea that furnishes to Hume the characteristic criterion applied by him to all philosophical questions. If the relations involved in the fact of cognition are only those dis coverable by observation of any particular portion of known experi ence, then such relations are quite external and contingent. The only necessary relation which can be discovered in a given fact of experience is that of non-contradiction ; the thing must be what it is, and cannot be conceived as having qualities contradictory of its nature. The universal test, therefore, of any supposed philosophi cal principle, seeing that such principles are but expressions of relations among facts, is the possibility or impossibility of imagin ing its contrail ictory. All our knowledge is but the sum of our conscious experience, and is consequently material for imagination. Let us fix our attention out of ourselves as much as possible ; let, us chase our imagination to the heavens or to the utmost limits of the universe ; we never really advance a step beyond ourselves, nor can conceive any kind of existence, but those perceptions which have appeared in that narrow compass. This is the universe of the imagination, nor have we any idea but what is there produced " (Works, ed. of 1854, i. 93, cf. i. 107). The course of Hume s work follows immediately from his fun da- mental principle, and the several divisions of the treatise, so far as the theoretical portions are concerned, are but its logical con sequences. The first part of the first book contains a brief state ment of the contents of mind, a description of all that observation can discover in conscious experience. The second part deals with those judgments which rest upon the formal elements of experience, space and time. The third part discusses the principle of real con nexion among the elements of experience, the relation of cause and effect. The fourth part is virtually a consideration of the ultimate significance of this conscious experience, of the place it is supposed to occupy in the universe of existence, in other words, of the re lations between the conscious experience of an individual mind as disclosed to observation ind the supposed realities of self and ex ternal things. In the first part Hume gives his own statement of the psycho logical foundations of his theory. Viewing the contents of mind as matter of experience, he can discover among them only one distinc tion, a distinction expressed by the terms impressions and ideas. Ideas are secondary in nature, copies of data supplied we know not whence. All that appears in conscious experience as primary, as arising from some unknown cause, and therefore relatively as origi nal, Hume designates by the term iiityrcssion, and claims to imply by such term no theory whatsoever as to the origin of this portion of experience. There is simply the fact of conscious experience, arising we know not how. Moreover, if we remain faithful to the fundamental conception of the contents of mind as being merely matters of experience, it is evident in the first place that as impressions aie strictly individual, ideas also must be strictly parti cular, and in the second place that the faculties of combining, discri minating, abstracting, and judging, which Locke had admitted, are merely expressions for particular modes of having mental experience, i.e., are modifications of conceiving (cf. i. 128 n., 137, 192). Thus at a single stroke Hume removes all the philosophical discussions that had centred round the problem of abstract ideas and the nature of judgment. It is merely by accidental concomitance, which on the subjective side is custom, that one fact, a word, sign, symbol, or type comes to stand for a series of resembling facts, while the comparison of perceptions, with resulting consciousness of their re semblance or difference, is in itself a single, isolated perception (see i. 37, 38, 100). Such, in substance, is Hume s restatement of Locke s empirical view. Conscious experience consists of isolated states, each of which is as a fact and is related to others in a quite external fashion. It remains to be seen how knowledge can be explained from such a basis ; but, before proceeding to sketch Hume s answer to this ques tion, it is necessary to draw attention, first, to the peculiar device invariably 1 esorted to by him when any exception to his general principle that ideas are secondary copies of impressions presents itself, and, secondly, to the nature of the substitute offered by him for that perception of relations or synthesis which even in Locke s con fused statements had appeared as the essence of cognition. When ever Hume finds it impossible to recognize in an idea the mere copy of a particular impression, he introduces the phrase "manner of conceiving." Thus general or abstract ideas are merely copies of a particular impression conceived in a particular manner. The ideas of space and time, as will presently be pointed out, are copies of impressions conceived in a particular manner. The idea of neces sary connexion is merely the reproduction of an impression which the mind feels itself compelled to conceive in a particular manner. Such a fashion of disguising difficulties points, not only to an in consistency in Hume s theory as stated by himself, but to the initial error tipon which it proceeds ; for these perplexities are but the consequences of the doctrine that cognition is to be explained from what can be discovered by observation among the facts of experience, and observation can discover none but external relations. These external relations are, in fact, what Hume describes as the natural bonds of connexion among ideas, and, regarded subjectively as pi-inciples of association among facts of mental experience, they form the substitute he offers for the synthesis implied in knowledge. These principles of association determine the imagination to combine ideas in various modes, and by this mechanical combination Hume, for a time, endeavoured to explain what are otherwise called judgments of relation. It was impossible, however, for him to carry out this view consistently. The only combination which, even in appearance, could be explained satisfactorily by its means was the formation of a complex idea out of simpler parts, but it is absurd to describe the idea of a relation among facts as a complex idea ; and, as such relations have no basis in impressions, Hume is finally driven to a confession of the abso lute impossibility of explaining them. Such confession, however, is only reached after a vigorous effort had been made to render some account of knowledge by the experimental method. The psychological conception, then, on the basis of which Hume proceeds to discuss the theory of knowledge, is that of conscious experience as containing merely the succession of isolated impres
sions and their fainter copies, ideas, and as bound together by
