tions, as willing, imagining, remembering, about them. This perceiving, active, being," says Berkeley, "is what I call mind, spirit, soul, or myself." But spirit is not strictly speaking an object of knowledge. "Such is the nature of Spirit, or that which acts, that it cannot be of itself perceived, but only by the effects which it produceth." And the status thus assigned to spirit corresponds almost exactly to that possessed by matter in the traditional view which Berkeley had himself discredited; so that the same dilemma may be urged against it. Spirit, like matter, must either come within knowledge or fall outside it. If it comes within knowledge it coincides with some idea or group of ideas; if it falls outside of knowledge, as a mere "producer" of ideas, it is arbitrary and meaningless. Hume adopts the former alternative with reference to spirit, precisely as Berkeley had adopted it with reference to matter: with the result that both spirit and matter are reduced to one manifold of ideas.

The question now arises as to the propriety of the term 'idea,' as applied to these common elements to which both spirit and matter have been reduced. If spirit be defined as a group of ideas, then it is clear that ideas themselves cannot be defined in terms of spirit. They become simply elements or qualities. Hume felt the force of this consideration, and it led him to the tentative supposition that perceptions can exist apart from the mind. Had he adopted and fortified this view, he would have been the founder of a new realism, instead of a link in the development of idealism. He rejected the view, however, summarily and unequivocally. He attributed his rejection of it to "those experiments which convince us that our perceptions are not possessed of any independent existence," such as the displacement of the field of vision by pressure on the eyeball. These and kindred phenomena, such as colorblindness, are cited to prove that "all our perceptions are

1 Berkeley: Principles of Human Knowledge, Fraser's edition, Vol. I, pp. 258, 272.

२

dependent on our organs, and the disposition of our nerves and animal spirits." 1 But since Hume has led us to suppose that our bodies themselves along with the rest of physical nature are no more than perceptions, he cannot properly argue that perceptions in general are dependent on the body. If relativity is to be advanced as an argument for idealism, it must be, not a relativity of ideas to body, but of body to ideas. And this relativity must be proved, if it is to be proved at all, by Berkeleyan methods.

Hume was especially influenced, I think, by the error of 'exclusive particularity.' He agreed with Berkeley that the elements of physical nature are the same as those of mental states; and advanced beyond Berkeley in reaching the same conclusion concerning spiritual nature. He found, in short, that the traditional substances, material and spiritual, are made up from the same manifold of elements. But instead of recognizing their interchangeable character, he named these elements, following Berkeley, after one of the rôles in which they appear. Finding them in the succession of the individual's mental states, he identified them with this order, and regarded them as belonging to it essentially and exclusively. The result is Hume's radical phenomenalism, or psychologism. To be is to be a particular mental state; and a particular mental state has no being whatsoever, other than its momentary presence. To be perceived or thought, to occur in consciousness, is to come into being; and to lapse from consciousness is to cease to be. There can be no permanence, and no sameness, because each unit of existence belongs wholly and exclusively to the moment of its occurrence in consciousness. The world consists, in short, of the coexistence and succession of unique individuals which instantly arise and instantly perish.

Hume not improperly regarded this outcome as equiva

1 Hume: Treatise of Human Nature (1739), Selby-Bigge's edition, pp. 207, 210-211. Cf. below, pp. 306-307.

lent to scepticism. It is, he thinks, the only conclusion that can consistently be reached on strict theoretical grounds. Nevertheless it is impossible for the ordinary man, or even for the philosopher in his ordinary moods, to believe it. The difficulty, according to Hume, is essentially a practical one. In order to live, it is necessary to regard the environment as having sameness and permanence; it is necessary to assume that one may have dealings at different times with an identical object, and that the objects on which one acts persist in one's absence. Such suppositions concerning the external world provide the orientation which is necessary for action. But in Hume's opinion they cannot be justified theoretically. "Carelessness and inattention alone can afford us any remedy." 1

Kant to the

'Categories'

Unity'

3. Kant agreed with Hume that the situation just described was practically intolerable, but added that it was theoretically intolerable as well. And it is not only contradicted by the whole body of existing Rescue. The science, but it is also self-contradictory. For and 'Synthetic the flux implies an order at least a temporal successiveness - which cannot be contained within any merely momentary state. Furthermore, Hume's whole procedure implies that this general fluxcharacter of things can be known by various knowers at various times; so that this, at least, must possess sameness and permanence. In other words, without order, sameness and permanence, no knowledge whatever is possible not even knowledge enough to warrant scepticism.

Kant doubtless rendered a service to all subsequent thinkers in proving the necessity of the principles of order, sameness, and permanence. Any object or world whatsoever must possess, in some measure, the structure and determinateness which such 'categories' can alone supply. But the status which Kant assigned to them, is another and more doubtful matter.

Let us consider, first, the alternative which he neglected. 1 Hume: op. cit., p. 218.

Avoiding the error of Hume, he might have declined to identify the elements of experience with the experiencemanifold exclusively. Had he adopted this course, his deduction of the categories would have amounted to proving that the elements of experience do stand in other orders besides the order of their successive and transient appearance; and that there are principles of order, such as space, time, substance, and causality, which cannot be identified with any of the particular transient appearances that presuppose them. In this case, neither physical nature nor the categories would have been construed as in any sense mental.

But Kant did not adopt this course. With Hume and Berkeley, he regarded the terms of experience as essentially "phenomena" or "representations." "They form an object that is within us only, because a mere modification of our sensibility can never exist outside us." Then, recognizing that the merely psychological order of Hume presupposed a more fundamental physical order, he regarded this also as an order of phenomena or representations, and its principles as forms of consciousness. "The very idea that all these phenomena, and therefore all objects with which we have to deal, are altogether within me, or determinations of my own identical self, implies by itself the necessity of a permanent unity of them in one and the same apperception." The order, in other words, borrows a mental character from its terms. A unity of phenomena will be a unity of "apperception." The new order is not, it is true, mental in the psychological sense. But this leads not to the denial of its mental character altogether, but to the new conception of a non-psychological or logical mind.

A second reason for Kant's version of the categories is his theory that a priori or necessary knowledge can be possible only on the supposition that knowledge dictates to its objects. "If the objects with which our knowledge 1 For a criticism of this view, cf. below, p. 160.

has to deal were things by themselves, we could have no concepts a priori of them." To be able to know beyond the present experience, to be able to know universally, implies that knowledge shall be able to lay down the conditions which all experience shall fulfil. Logic, for example, can hold of all experiences whatsoever, only provided it be construed as determining what can be experienced. The categories thus appear as "the necessary conditions of a synthetical unity of the manifold of intuition in a possible experience."

But quite apart from these considerations, the quasimental status of Kant's categories could be accounted for by his adoption of the epistemological standpoint. He undertook "to determine the possibility, the principles, and the extent of all cognitions a priori." Then, when, in fulfilment of this task, he discovered the categories, he named them after their rôle in cognition. "I call all knowledge transcendental," he says, "which is occupied not so much with objects, as with our manner of knowing objects, so far as this is meant to be possible a priori." The "transcendental deduction" of the categories introduces them as the indispensable condition of a priori knowledge. They are the forms of a transcendental synthesis or "unity of apperception," which is the supreme intellectual function.1 Introduced to meet the exigencies of knowledge, the whole Kantian logic thus obtains at the outset a cognitive or mental status which it never loses, even among its most rigorously "logical" exponents."

Kant is to be credited with proving that if any knowledge is to be possible, then physical, mathematical, and logical knowledge must be possible. The knowledge of the momentary presence of a state, to which Hume had sought to reduce all knowledge, is not self-sufficient. It 1 Kant: Critique of Pure Reason, Max Müller's trans., second edition, pp. 105-106, 129, 2 (note), 9 (note), 100.

See below, p. 147.

'Cf. G. E. Moore: "The Nature of Judgment," Mind, N. S., Vol. VIII, 1899, pp. 190 sq.