On Some Current Conceptions of the term 'Self'
It is the aim of this paper to analyse certain conceptions involved in the terms Self and Self-consciousness as currently used. No attempt will be made to judge of the values of the ideas themselves. Indeed, there is such confusion in the use of the conceptions that an independent analyse of them would seem to be a necessary preliminary to any decision upon their validity. Whether or not philosophy is exhausted in the clearing-up of connections, it is certain that without an occasional clearing-up philosophy will get so entangled in the impedimenta of its own notions as to be hindered in its onward march. Unless this analysis is confined to ideas having or claiming to have some community of meaning, it will include ideas wholly incomparable with one another, and thus end in a mere account of the way in which various writers use the same word. A study of the terminology of philosophy is, no doubt, helpful; but, as that is not intended in this paper, I shall confine my analysis to the conception of the transcendental self' -- to the idea of self which has affiliations with the movement set going by Kant, however divergent its various developments.
For a starting point, and to a certain extent for a basis, Prof. Seth's recent work, Hegeliansim and Personality, presents itself as convenient, occupied as it so largely is, with just this notion of the self. In that work, three separate conceptions -- used, however interchangeably -- may be discriminated. In the first place, we have it laid down that "the self is the world, and the world is the self . The self and the world are only two sides of the same reality ; they are the same intelligible world looked at from two opposite points of view.... The mind and the world, subject and object, are convertible terms; we may talk indifferently of the one or the other: the content of our notions remains the same in both cases" (pp 12-20). This result is based upon an examination of Kant's transcendental inquiry and method which is, so far as quoted above, accepted, to all appearances, by Prof. Seth. The
(59) meaning of this view of the self may stand out more plainly if we call attention to another feature of it. That is that the "ultimate fact of knowledge is neither pure subject not pure object" (p 13). These are both abstractions: to separate them, to make independent existences of them is to "substantiate abstractions." In truth, the self is a synthetic unity. "It binds together, as related member of one whole, what would otherwise fall apart as unrelated particulars ; and, moreover, it is only through this synthesis that the unity of the Self or Ego exists. It is the unity of the synthesis, and apart from its synthetic activity, would no more be real than the particulars of sense would be real without its action." It cannot be identified, in other words, with the mere act of uniting ; It includes within itself what is united, just as, on the other hand, what is united has no existence outside of its being united. Because this is so -- because as Prof. Seth expresses it (p. 19), "the form is the form of the matter, and the matter is, as it were, simply the exhibition of the form" the self and the world are correlative, and have the same content.
This, then, is the first notion conveyed by the term self -- the self is the correlative of the intelligible world. Its content is that of the intelligible world. It even is the intelligible world in one of its aspects. And since Prof. Seth has expounded wit great force the notion that the intelligible world is the only real world, that the unknowable to intelligence is "nonsense" (Scottish Philosophy, p. 162), we may say that, according to this notion, the self is one with the real world, when this is considered in its ultimate unity. This view is clear and self-consistent; with its truth we have nothing to do. But we find that the question as to the nature of the transcendental self has not be sufficiently answered. The question is again raised: What is the transcendental self (top of p. 22). And the question is answered in a way which seems to me the exact opposite of the answer just given. It now turns out that the transcendental Ego represents merely the formal unity of the universe (p. 27). Although the self was show to be a single self. Its singularity is simply that which belongs to every abstract notion -- a logical identity of type (p. 29). It is the "notions of knowledge in general" (p. 30). And, finally, Kant's characterisations of it are quoted. It is "a merely logical qualitative unity of self-consciousness in thought generally" (p.35). It is, finally, the "mere form of self-consciousness in general" (p. 230).
I confess that, to me, this second position, that the self if merely the formal unity of thought, appears to be the contrary of the first position taken by Prof. Seth. There the self was not formal; the form was an abstraction apart from matter. Kant was then rebuked for making the self formal. The necessity of correlating matter and form was the fundamental feature of the transcendental method. So far was the self from being merely formal that it was the world. Instead of being merely logical, the self was the unified universe; it was a synthetic unity which had no existence apart from the particulars unified in the synthesis. But in this second and revised view, Kant is praised for his superior consistency in holding that the self arrived at by his investigation is an abstract condition and not a metaphysical reality or concrete fact (p. 28). The subject which "exists only as the unity of the manifold whose central principle of connexion it is" (p. 17) becomes transformed in ten short pates into a "focus imaginarius into which the multiple relations which constitute the intelligible world return" -- a "principle of unity". To cut short this comparison of contradictory statements, the language first used regarding the self conveys, as clearly as language can convey anything, that the self is objective and real, is ontological; while the second view taken is that the self and the real cannot be separated without "substantiating abstractions"; the second view is that to unity them is to "hypostatise an abstraction" (p. 30).
But as we advance further, it appears that the outcome of the transcendental view of the self is not in reality either that the self is the real world, or that the self is a mere logical form or abstract unity of though. The view which finally emerges is that self is the "ultimate category of thought" (p. 98). So far as the varying expressions permitted us to judge, this is Prof. Seth's real thought on the matter. It is, at least, the view which is unambiguously reiterated in his Discussion' in MIND xiv. 117. It is stated once in connexion with passages which have been quoted as belong to the first interpretation ; "The transcendental self, as an implicate of all experience, is for a theory of knowledge simply the necessary point of view from which the universe can be unified, that is, from which it becomes a universe" (p.20). It is elsewhere stated that the transcendental theory of knowledge resolves itself into an immanent criticism of categories, or the conceptions by which we express and unify our experience. This criticism shows that self-consciousness is the highest category -- the most adequate to
(61) determine existence. We are thus "justified in using the conception of self-consciousness as our best key to the ultimate nature of existences as a whole" (p.89). In fine, "self-consciousness is the ultimate category of thought -- that through which we think everything else, and through which alone the universe is intelligible to us."
I cannot persuade myself that this third conception of self-consciousness is identical with either of the other two. It means less than the first, which identifies the self with the world; it means more than the second, which makes self-consciousness a merely formal or abstract unity of thought. For it must be remembered that Kant would no more have accepted self-consciousness as the ultimate category of experience, or as a category of experience at all, than he would have accepted it as identical with the real world. In fact, the various expressions which Prof. Seth has quoted with approval from Kant are. directed as much against making self-consciousness a category of experience as against making it a real self-existent being. How can the "poorest of all our ideas" be the richest and most comprehensive principle of philosophic explanation? The very reason for holding that the self is merely a logical unity of thought is that the self cannot be employed to determine experience at all. But perhaps it may be said that it was just the result of the Hegelian development of the Kantian method and presuppositions to demonstrate that the self, instead of being the emptiest of categories, a conception the sole use of which is to show that all our thoughts are accompanied by consciousness, is the organic system, the reality of all categories. I am not in the least concerned to deny such a contention. But this contention only shows the inadequacy of defining the self as a "merely logical qualitative unity of self-consciousness in thought generally," and not that it is consistent to unite such a view with a view that the self is our ultimate principle of verifying and explaining experience. Indeed, the purpose of Kant in calling the self merely logical was to oppose it to experience; but, when it is said from the point of view of the Hegelian development of Kant that the self is the highest logical category, the idea conveyed is that the complete correlativity of thought in general, and this thought in particular, to experience. When Kant speaks of a logical unity of thought he means that thought is formal, not real; Hegel in speaking of a logical unity means that thought is real and not formal. The relation between thought and knowledge is not at all the
(62) same in the two cases. With Hegel, to say that self is the highest type of thought is to say that self-consciousness is the ultimate principle of knowledge. The object of Kant is to show that the self, since merely a principle of thought, is not a principle of knowledge at all. While both therefore might call the self "the logical exposition of thought in general," the phrase would have absolutely opposed meaning in the case of the two writers.
No relation of opposition exists between the transcendental self as equal to the real world and as equal to the ultimate category -- between, that is, the first interpretation and the third which Prof. Seth gives. To pass directly from the one to the other would be to hypostatise an abstraction. The transition may be justifiable, but it cannot, of course, be assumed without justification. The transcendental self may be the highest thought of the world, but it cannot be said to be the correlative of the world, unless the content of the world can be shown to be exhausted in thinking it -- or unless the transcendental self is more than a principle of thought. Because thought is objective, it does not follow that it is all there is to objectivity. The world as thought -- and thus brought under the principle of self-consciousness -- may be real as far as it goes, and yet not be identical with the world as known -- with the whole meaning of the real world. The known world may be, for example, a world thought and felt, and not thought alone. Thus while self-consciousness -- if it equalled only the ultimate category of thinking -- would be an adequate determination of the world as thought, it would, after all, be only a partial determination of the whole as it really exists, and could not thus be called, as Prof. Seth at first calls it, a term convertible with the world and having the same content.
These may appear distinction so notorious that it is trifling to spend so much time upon them; but the fact that so experienced a writer as Prof. Seth has presented all three interpretations as explications of the meaning of the "transcendental self" is my excuse for dwelling upon them. There is a certain kinship, indeed, between the three interpretations which would render it easy to pass unwittingly from one to another. The idea of the self as the ultimate category of philosophic explanation stands between the other two. Its content is logical, or thought ; and thus when one is arguing against a writer who seems to transform this category into an existence by itself, it is easy to go
(63) to the extent of saying that it is merely logical, and approve an author who held to the view that it was wholly abstract, even though that author meant by that expression that self was not a category of explanation at all. But, on the other hand, having in mind the fact that self-consciousness is a notion for explaining the world in a sense in which mere being' or quantity' or mechanism' is not, -- that it exhausts the meaning of the universe as an object of thought, -- it is easy to go to the other extreme, and hold that self-consciousness is the intelligible world seen from one of its sides. But non the less the conception of self as merely formal and abstract contradicts the other tow conceptions; and these other two, while not mutually incompatible, are so far from being identical with each other that to pass from one to the other without more ado is to "erect an abstraction into a concrete existence."
As the object of this paper is not to convict Prof. Seth of either verbal or real inconsistencies, but to help to clear up certain ambiguities in the current use of the conception of transcendental self' (these ambiguities finding an unusually clear expression, as it were, in Prof. Seth's book), I wish now to pass to the historical origin of these various meanings, chiefly as found in Kant, incidentally in Hegel as related to Kant.
Kant's theory, is brought out in his "Transcendental Deduction". This is so familiar that it may be given summarily. Its gist, in the second edition of the K.d.r.V., is the proof that the identity of self-consciousness involves the synthesis of the manifold of feelings through rules or principles which render this manifold objective, and that, therefore, the analytic identity of self-consciousness involves an objective synthetic unity of consciousness. That self-consciousness is identical is, in itself, a merely analytic proposition. It means nothing more than that I am I -- that what I am conscious of is in my consciousness, and that what belongs to your consciousness I am not conscious of. It finds its empirical application in the fact that, unless the consciousness which has ideas to-day is identical with that which was conscious yesterday or a year ago, it can no more now be conscious of what it was conscious of then than it can now be conscious of what is in your consciousness. But this does not prove the existence of any real self or substantial mind. It is still an analytic proposition and means that the same consciousness is the same consciousness. But if we
(64) ask how we know this sameness or identity of consciousness, the barren principle becomes wonderfully fruitful. For we do not know this sameness through the various successive ideas ; they are not the same, but ex hypothesi various. And, furthermore, instead of knowledge of the identity of self depending upon them, I should not know them even as various, unless they were already mine. The identity of self-consciousness cannot be derived from knowledge of them, for this knowledge presupposes that identity. But perhaps we may go behind the apparent variety and disparateness of our ideas, and say that one consciousness accompanies all these different ideas, and that knowledge of this common element is the knowledge we are in search of. This does not suffice. The mere fact that consciousness accompanies every idea gives no identity unless these ideas are already conceived as mine -- unless identity is presupposed. Otherwise, I should "have as various and many-coloured a self as I have different ideas". If we say that the common element gives us that knowledge of the identity of self which we are in search for, we doubly beg the question. A common element means an identity present in the midst of difference, and thus presupposes the sameness of consciousness through different ideas; and knowledge of this common element could be attained only if it were possible to compare many and various ideas in one consciousness, and thus see that they had a common element. These methods of knowing the sameness of consciousness thus presupposes what they would account for.
The sole way of accounting for this analytic identity of consciousness is through the activity of consciousness in connecting of "putting together" the manifold of sense. Since this putting together occurs according to fixed rules and principles, it is an objective synthesis. Knowledge of the identity of self presupposes, therefore, a self which acts synthetically, regularly so, upon sense-material. "The original and necessary consciousness of the identity of one's self is, at the same time, a consciousness of the equally necessary unity of the synthesis of all phenomena according to conceptions .... The mind would never conceive the identity of itself in the manifoldness of its ideas, if it did not perceive the identity of the action by which it subjects this manifoldness to unity."
The Deduction' in the first edition, instead of beginning with consciousness of self-identity, begins with the consciousness of objects, and asks what is involved in that. The answer is the same. Consciousness of objectivity
(65) means unity of self-consciousness, and this not a formal or analytic activity, but one which connects the manifold of sense according to rules or conceptions. Whether, then, we inquire what is involved in mere sameness of consciousness, or what is involved in an objective world, we get the same answer; a consciousness which is not formal or analytic, but which is synthetic of sense, and which acts universally (according to principles) in this synthesis.
Apparently we have here a conception of the transcendental self like the first one laid down by Prof. Seth. This self, since its existence is its synthetic activity upon the particular manifold of sense, is thoroughly objective. It has precisely the same content as the real world. And the objective world, since it turns out to be the synthesis of particulars of sense through the action of self according to conceptions, is subjective ; it has the same content as the transcendental self. It is the transcendental self looked at as there,' as a product, instead of as an activity or process.
The next step in the analysis is to see why Kant, after having attained to the conception of an objective self, should shift his ground. Kant, in reaching this result, or in his transcendental deduction , has proceeded as if the synthetic action of self and the manifold of sense where wholly constituted through their mutual relations to each other -- as if each had no existence excepting as a factor in the self, or in the world, determined by the other. The conceptions exist only as synthetic activity upon the manifold of send; the manifold of sense exists only as connected by these conceptions. But while Kant has chosen in the deduction to consider them as mutually related to each other, they have a meaning entirely apart from this mutual qualification, which, having been abstracted from in the transcendental deduction, must now be brought in that we may see how it affects the result.
The final meaning of the manifold of sense is found, not in its relation to the synthetic notions of the understanding, but in its relation to a thing-in-itself which produces it. In order to be known by us, this manifold must, indeed, be subjected to synthesis, and enter into relation to the self. But it has its own being entirely apart form such qualification. And, on the other hand, the conceptions of the understanding are not exhaustively determined by their synthetic action upon sense. They have a nature of their own, entirely independent of this synthetic action. The transcendental deduction does not give us, therefore, an analysis of the self, or of knowledge, or of the world as
(66) such; but simply of the conditions under which a manifold of sense (having a nature outside its relations to self) is knowable by us, or of the conditions under which conceptions of the understanding become categories of experiences, these conceptions having their real and essential meaning, all the while, in a purely logical character which belongs to them apart form knowledge or experience. The transcendental self is thus a name for the incident under which or knowledge occurs, instead of giving the analysis of knowledge itself. It cannot be identified, therefore, as at first it seemed it might be, with either the real object (the thing-in-itself) of with the real subject. Just as the synthetic principles of experience are in themselves logical forms of analytic thought, so the self, in its own nature, is known only as the bare unity of these logical forms, the simple I think' that must accompany all thought. The introduction of the thing-in-itself, therefore, leads Kant to that view of the self which finally gets expression in the quotations which were made in connexion with Prof. Seth's second idea of the self. For it must be remembered that the introduction of the thing-in-itself into Kant's philosophy affects all the factors which enter into his account of knowledge -- the nature of thought as well as the nature of sensation. It is not an excrescence which can be lopped off without reconstruction of the whole theory of knowledge. Do away with the thing-in-itself, and the conceptions, instead of being merely logical, are also real, for their whole existence and meaning will then be found in their synthetic relation to the sense-manifold. And the transcendental self, instead of denoting a "logical exposition of thought in general," marks the synthetic union of the logical with the manifold of sense through regular principles of activity -- marks, therefore, the objective character of the self. Fir if we reconstruct the Kantian theory of knowledge upon its own basis and method of analysis, doing away with the thing-in-itself, the result is to show that the merely logical, equally with the merely ontological, is an impossible abstraction. The merely logical is not at all; the logical is only as the thought-factor in the entire determination of experience, requiring another factor in order to constitute the self. That Kant's position of the merely formal abstract character of the self is superior in consistency to some Neo-Kantians is, therefore, not so evident as is the inconsistency of the restatement of such a position by one who denies the whole notion of the thing-in-itself.
But even if we correct Kant's analysis by doing away with
(67) the thing-in-itself, retaining all features not inconsistent with it, can the result of the transcendental deduction stand without further interpretation? Admitting that the removal of the thing-in-itself would show the transcendental self not as a logical abstraction, but real as experience itself -- more real, indeed, in the sense that the reality of experience is show by analysis to involve the reality of this self, behind which we cannot go -- would this removal give a self whose content was the same as the content of the known world? The answer must be in the negative. The known world is constituted by the manifold of sensation, as connected by the self through its principles of synthesis. The content of the world, as known, will not be equivalent to the whole significance of the self, therefore, unless sensation is capable of being connected by principles of synthesis which manifest the entire nature of the self. But the position of Kant (a position entirely independent of any notion of Ding-an-sich) is, that sensation is incapable of being so determined as to equal self-consciousness ; or, if we put it from the other side, that self-consciousness, even as a real activity of synthesis, can never exhaust all its synthetic capacities upon a material of sense. Sense is, as it were, inadequate to the relations which constitute self-consciousness, and thus there must also remain a surplusage in the self, not entering into the make-up of the known world. The reason for this is, that all the manifold of sense must be determined by certain forms of perception, space and time, before being determinable by the categories of thinking. Perhaps it would be more in accordance with the Kantian spirit to say that sensation, since it is in relation to space and time, must always present itself to the synthetic action of self as a manifold of mutually external particulars. The conceptions are thus not capable of determining sensation independently, but only as sensation is already subject to time- and space cadres. Every category, therefore, must receive its value from its application to sensations already a manifold of external particulars, and the result can be only the system of objects in time and space. No category of experience can be found, accordingly, higher than that which determines most exhaustively the relations of objects and events in time and space, viz., reciprocity. And, correspondingly, no object can be known which is not an object in space and time. Hence the impossibility of making the self an object, since it is the condition of all objects, through its synthetic action upon sense. Stated in more Kantian language, the result would be that self-consciousness is the unconditioned, while experi-
(68)-ence, owing to the necessary relation of the synthetic activity of self to a material already determined as externally limiting and limited, can never present an unconditioned.  There thus remains a distinction between self and experience, due not now to the shadow thrown on knowledge by the thing-in-itself, but by the incompatibility of sensation, as rendered a manifold of external particulars in space and time, to the unconditioned content of self-consciousness. Experience can never be complete enough to have a content equal to that of self-consciousness, for experience can never escape its limitation through space and time. Self-consciousness is real, and not merely logical ; it is the ground of the reality of experience ; it is wider than experience, and yet is unknown except so far as it is reflected through its own determinations in experience, -- this is the result of our analysis of Kant, the Ding-an-sich being eliminated but the Kantian method of all presuppositions not involved in the notion of Ding-an-sich being retained. The resulting conception of the self is, evidently, not equivalent to either of Prof. Seth's two first definitions of the self. It is not a mere abstract and formal logical unity, for it involves the action of thought upon sense, and is thus synthetic and objective ; and yet it is not one side of the world of experience. The world of experience is constituted by it, but the world of experience does not exhaust it.
We have next to consider the relation of this revised Kantian conception of the self to the third notion of self stated in Prof. Seth's book -- the idea of self-consciousness as the highest category of thought and of explanation. So far we have dealt only with the general idea of thinking as synthesis of sense according to principles. The different forms of synthesis, or the categories, we have not dealt with. Kant, as is well known, had twelve of them, which he derived without
(69) further examination from certain notions which he found to be involved in the formally logical theory of judgment. It was the work of Hegel, first, to give an independent derivation of them, as contrasted with Kant's taking them for granted ; secondly, to give an organic derivation of them, in placing them in relation to one another, as contrasted with the simple juxtaposition of them which is found in Kant ; and, thirdly, to show the category of self-consciousness as their basis and system, instead of stopping short with reciprocity, and placing the categories in opposition to self-consciousness. Now, accepting Hegel's work so far as it thus relates to the categories, and accepting his criticisms upon the Kantian procedure in reference to them, let us again revise the Kantian results in view of Hegel's position. Will this give us the self as the supreme category of experience? The answer must be in the negative. In one way the Kantian conception will include more than the Hegelian ; in another way, less. It includes more, because what Kant offers is not primarily the self as a category of explanation at all, but the self as the real ground (not, however, to be confused with cause) of experience.  It includes less, because, however ready Kant might be to admit the Hegelian criticism and derivation of the categories as superior to his own, he could not admit that self-consciousness may be used as a category of experience. Self-consciousness would still have the function of the Idea for Kant. It would be an ideal regulative of experience, not a category constitutive of it.
Considering first this latter point, we may say that, admitting Kant's derivation of the categories from the forms of syllogistic logic to be insufficient and artificial, granting that it is impossible to stop short with the category of reciprocity, it does not follow that the category of self-consciousness is a category of experience. The distinction between conceptions of thought and conceptions of knowledge still remains. It is the peculiar relation of the categories to sense as qualified by the forms of space and time. While, therefore, we might have the thought of self-consciousness, and while as a thought it would not be empty but would be, in another sense that in which Kant actually uses the term, the vehicle of all notions of thought -- their organism, it would be impossible to use this category so as to determine sense by it. For it is impossible
(70) as long as we retain Kant's fundamental presupposition -- the idea of the partial determination of sensation by relation to perception, apart from its relation to conception -- to employ self-consciousness as a principle of explaining any fact of experience. Every fact of experience is capable of adequate explanation without any such category ; or, conversely put, experience can never convey anything adequate to the notion of self. Self-consciousness would thus be an ideal category -- that is to say, it would suggest the notion of a possible experience, unlike anything that we can possibly experience. It would be a notion which should regulate the successive organisation of our present experience by pointing to a goal that yet we never could reach, and which should also point out the limitation of our present experience. 
The reconstruction of the Kantian theory of categories in the light of the Hegelian logic would give the following points. First, it would derive the conceptions from a common root and place them in some organic connexion with one another. Secondly, it would place the Notion of the understanding and the Idea of reason in some connexion with each other The reason, with its Ideas, would not then appear, as it does now, an accidental afterthought of Kant, or an arbitrary derivation from the theory of the syllogism. The conception included under the Idea would follow by immanent development and criticism from what are now called Notions of the understanding, and would follow as their basis in thought. The distinction between them would be between conceptions that may be used to connect sensations subject to space and time-forms and those that may not be so used. Thirdly, the ideas of organism and teleology, which also now appear to be unconnected with the rest of the Kantian philosophy, sprung upon us without intrinsic necessity, would form part of the content of the Idea as distinguished from the Notion. And, finally, the distinction Kant now makes between theoretical and practical reason, between the fact which is and the ideal which out to be, would get an organic connexion with the rest of the philosophy. This gives the outline of a reconstruction of his ethics ; for it would appear that it is just the business of moral experience to overcome that distinction between experience and self-consciousness which theoretical know-
(71)-ledge cannot remove. All this we can get, if we read Kant with the eyes of Hegel ; but self-consciousness as an actual category of our scientific experience we cannot get unless we simply substitute Hegel for Kant.
But it is time to turn to the other point ; that the transcendental self of Kant is more than self-consciousness as a supreme category of explanation. It is more, because of the self of Kant (the self as it would be with the Ding-an-sich eliminated) is more that any category ; it is a real activity or being. And it cannot be said to be more than a category only because he has hypostatized a category -- that if he had understood himself he would have seen that it was just a category. There is a fundamental distinction between the Kantian critique of pure reason and the Hegelian theory of categories which makes their results disparate. Kant's object is not the examination of thought, but the examination of knowledge ; and his method is not a consideration of the significance, placing, relative adequacy and inadequacy of the conceptions or aspects of thought with a view to discovering the entire meaning of thought ; his method is a n analysis of the actual factors which actually constitute knowledge. One of these factors is thought and therefore, the complete carrying out of the method would undoubtedly involve an examination of thought as specified into its various conceptions. But because the Hegelian Logic is the development of one factor in Kant, it will hardly do to say that the purpose of the Kantian Critique is exhausted in the purpose of Hegel's Logic. At least, if we do say it, it should be with the distinct consciousness that we are not completing Kant, but are abandoning the characteristic feature of his undertaking and of his method. This is, I repeat, not an immanent "criticism of categories " but an analysis of experience into its aspects and really constituent elements. And in the course of this analysis Kant comes upon a self which through various principles of synthesis puts together the manifold of sense and, thereby, constitutes experience. This, indeed, is not a theory of creation; it is not an attempt to tell how a self set to work, or by necessity would set to work, to make a universe. But because it is not a theory of creation, it does not follow that it is only a criticism of categories. The assumption that there is not middle ground between a theory of creation and a mere analysis of forms of objective thought is, to say the least, a curious one. Kant's method is the analysis of the known universe or of experience ; and as a result it discovers a self acting through thought upon sensation. Thought as synthetic is action
(72) upon sense, and sense is through the synthetic action of thought. If we call them factors of experience it must be with the recognition of their intrinsic unity with each other. The self constitutes this unit ; it is the activity which is the source of the correlative synthesis of thought and sense. That analysis of reality should give anything but reality would be a strange result. And the reality is a self which through thought is synthetic sense determined to be a manifold of limiting particulars by relation to space and time.
There are two strains in Kant ; one is inquiry into the necessary thought or logical conditions of experience ; the other is the inquiry into the actual nature of experience. The Logic of Hegel undoubtedly works out the former to its consistent results. The latter it does not come in contact with. The former inquiry asks what are the forms or principles by which we must think of the world ; or, from the other side, what the world must be, as thought. The answer is that to think the world in its completeness is to think it as self-consciousness. Now this proposition is, as I attempted to show in the earlier portion of this article, not convertible with the proposition that the world is self-consciousness, unless it is also shown that the world is only and just as it is for thought. But the result of Kant's inquiry into the actual nature of experience it to show (to his satisfaction, I mean, the truth of the results not being under examination ) that it includes another element besides thought, namely, feeling, and that on account of this element -- or at least on account of its peculiar relation to forms of perception -- the world as experienced can never equal the world as thought. That is, while to think the world completely is to think it as self-consciousness, it is the very characteristic of experience or knowledge that it cannot be complete -- and hence cannot give self-consciousness.
We have thus another conception of self-consciousness to put beside the three obtained from the analysis of Prof. Seth. This is the conception which we reach in reconstructing Kant by means of the elimination of the Ding-an-sich, and by the more complete working-out of the logical side of his analysis of experience which was made by Hegel. This is the self as the activity of synthesis upon sense. Starting from this notion the other three notions may be at once placed with reference to it. The self as the merely logical or abstract unity of thought falls away entirely. Self-consciousness as a category of experience becomes changed
(73) into an ideal which serves at once to organize and to reveal the incompleteness of experience. Where (as in ethics) the ideal is the reality, self-consciousness is again a real category of experience -- but of practical experience, not of theoretical. The self which could use the category of self as a category of both practical and theoretical experience would be a self whose content was the same as that of the world. "The self and the world are only two sides of the same reality" in this case. While from the standpoint of Hegel's Logic (I am not speaking of the rest of his philosophy) such a result could be reached only by substantiating a category, from the standpoint of Kant's Critique it would be reached as an analysis of the reality of experience -- if it were reached at all. But it can be reached only as an ideal which serves by contrast to manifest the incompleteness of experience as it presents itself to us.
It is evident that we are now upon the verge of another difficulty. As long as sensation was regarded as given by a thin-in-itself, it was possible to form a conception of the self which did not identify it with the world. But when sense is regarded as having meaning only because it is there' as determined by though, just as thought is there' only as determining sense, it would seem either that the self is just their synthetic unity (thus equalling the world) or that it must be thrust back of experience, and become a thing-in-itself. The activity of the self can hardly be a third something distinct from thought and from sense, and it cannot be their synthetic union. What, then, is it? This is, I take it, the problem which finally emerges, when Kant is made self-consistent by the elimination of the thing-in-itself, and when the logical or thought-factor of his philosophy is developed in the Hegelian manner. It is precisely, as it seems to me, the difficulty which comes to the front in Green's reconstruction of Kant. It is to meet this difficulty that he frames the idea of a completely realised self making an animal organism the vehicle of its own reproduction in time. The conditions of the problem are : a denial of the Ding-an-sich ; the analysis of knowledge into thought, and feeling which is etepov " to thought ; the recognition that this feeling, after all, exists only as determined by though ; and the belief that feeling enters into our knowledge only under conditions of space and time, although space and time, in themselves, are feeling determined by thought. No space remains to consider how far Green's conception of an eternal self communicating itself gradually through physical conditions, and thereby constituting a human self, meets
(74) the demands of the problem. But it is evident that, when the problem is conceived as just stated, the self cannot be thought of as equivalent, on the one hand, to the world because this world, as knowable by us, is always subject to certain forms, namely, space and time, which condition sense; nor, on the other hand, as equivalent to the highest category of thought, because the self is more than thought, more than a category, namely, the activity of synthesis of sense through thought. It is, I think, this twofold character of time and space, as at once forms of knowledge conditioned by the self, and yet conditioning self as it works in us, that is the genesis of Green's notion. The truth of the conditions upon which it rests -- that is, Kant read in the light of Hegel so far as is necessary to make Kant consistent -- is not under examination here; but if we grant it, the theory of Green is a genuine attempt to meet a genuine problem, and not a mere hypostasis of an abstraction.