You are here

Book II: The Categories

Chapter I: Nature of the Categories

Categories and Qualities.

Space-Time then is in Kantian language an infinite given whole, that is to say, it is experienced as such, where the term experience includes thought as well as sensible experience. Its elements are represented conceptually as point-instants or bare events ; and we have added the hypothesis that other empirical things or existents are groupings of such events, whirlpools within that ocean, or they are crystals in that matrix. Only whereas a crystal may be separated from its matrix, existents never can ; they remain swimming in the medium of Space-Time. Their very being is continuity ; they are themselves continuously connected groupings of motions, and they are connected through the circumambient Space-Time with other such groupings or complexes. In less metaphorical language, they are complexes of motion differentiated within the one all-containing and all-encompassing system of motion. Primarily, therefore, empirical existents are spatio-temporal and remain so to the end. But with certain groupings of motion, certain spatio-temporal complexes, there are correlated what we call qualities, such as materiality, life, colour, consciousness. What the exact relation is between the quality and its spatio-temporal basis is to be the subject matter of a part of the next Book. We shall have to ask there whether it is fitly to be described as mere correlation or is still more intimate. The brief description contained in the name correlation is sufficient for our present purposes. Finite existents so understood, with their correlated qualities, are the things and events of our ordinary experience, moving about or happening in Space-Time, and endowed with qualities the laws of which it is the office of the special sciences to discover and co-ordinate. So much by way of explanation of our hypothesis as to empirical existence.

Unlike the hypothesis of the Introduction, (that the world of things might be treated as existing in its own right and not dependent on the mind,) which is a hypothesis of method; it is a hypothesis as to the nature of things, or, in ordinary language, one of substance, not merely of method. In order to avoid the constant use of the long phrase empirical existents, I shall speak simply of existents. These include not only ordinary finites but also point-instants which are the limiting cases at which we arrive in infinite division, and infinites like infinite lines or numbers, which are the limiting cases in the other direction; and for this reason, in order to include these two classes of existents which involve the notion of infinitude, I speak of existents rather than of finites. But while there will be much to say of point-instants, I shall for the most part disregard infinites till a later stage, and then touch upon them only briefly.1

Now amongst the characters of empirical existents there is a clear distinction between those which are variable and those which are pervasive. Some things possess life, others not. Some things are red, others green or yellow; some are sweet, others sour. Some have colour but no taste. Matter has mass but is not conscious. These characters are what have been called above qualities, and because they vary from thing to thing they may be called empirical characters. But there are other characters which are pervasive and belong in some form to all existents whatever. Such are identity (numerical identity for example), substance, diversity, magnitude, even number. Moreover, not only are these characters of what we commonly call things, but they are characters of all existents whatever, that is to say of everything, where the word thing is equivalent to any finite object of experience. Thus not only is a living thing an extended substance of a certain magnitude and number of parts; but a life itself, if you consider it, or so far as you can consider it, without direct reference to its body whose life it is, is extended, a substance, and possessed of magnitude, and moreover it is spread out into a multiplicity of parts and therefore contains number. Even mind, now that we have satisfied ourselves of its extended character in its enjoyment of itself, possesses these characters.

It is true the pervasive characters also undergo variation according to the empirical circumstances. The wax is always extended, but its particular magnitude and shape change when it is melted. Still, it retains some extension and magnitude and shape in all its empirical transformations. An earthquake may last a long or short time, an illumination may be constant or intermittent. But they are never without temporal character. Such empirical variations of the pervasive characters of things may be called primary qualities in distinction from the secondary qualities, where the phrase covers not only the traditional secondary qualities of matter but qualities like life or consciousness. These qualities may be present in one thing and absent from another, and differ in this respect from the empirical variations of the pervasive characters.

The pervasive characters of existents are what are known from Kant's usage as the categories of experience, and I shall call them, in distinction from the empirical ones or qualities, categorial characters. They may also be called the a priori or non-empirical characters. But the contrast must be taken at its face value as a distinction within the characters of experienced things. It does not imply that a priori or categorial characters, because not empirical, are not experienced. On the contrary, they are the essential and universal constituents of whatever is experienced, and in the wider sense of that term are therefore empirical. It was in this wider sense that philosophy was described as the empirical (or experiential) study of the non-empirical. The word categorial is not so much exposed to misunderstanding as non-empirical or in consequence of its history a priori; and I shall most frequently employ it. At any rate the two classes of characters are distinguished within experience itself.

These categories then are the prerogative characters of things which run through all the rest as the warp on which the others are woven. Or, to vary the metaphor, they are the grey or neutral-coloured canvas on which the bright colours of the universe are embroidered. The primary 'qualities' are variations of them in empirical circumstance. The secondary qualities are correlated with complexities in the primary qualities themselves. Life is correlated with physical and chemical movements, themselves reducible to complexities of more elementary movements. Mind is correlated in turn with vital movements of a certain sort. Colour (whether it is partly dependent upon mind or not) corresponds, it is thought, to vibrations in a hypothetical medium, the ether, which hypothetically (and there is reason to think, superfluously) fills all Space. The categories are thus the groundwork of all empirical reality; what Plato called the highest kinds of beings (μεγιστα γενη των οντων). According to his latest interpreter, the interest of these highest kinds displaced in his latest writings that of the Forms of sensible things; and justly. For the Forms for all their eternal nature are, as compared with the categories, empirical—the form of dog in which individual dogs participate or which they imitate, but which trees do not; the form of tree, or the form of justice, and the like. These are empirical universals. But the categories are not only universals, but, though I do not know if Plato would have said so, are truly universal in the sense that all existents partake of them.

Why the categories are pervasive: not because they are due to mind;

The most remarkable feature of the categories which is disclosed to inspection is that they are common to mind and to physical and generally non-mental things. Consider mind as it is known by direct acquaintance, that is by enjoyment, without the addition of indirect knowledge from any source, whether from reflective experience about mind, or from speculative theory. It has identity, is a substance, exhibits causality, etc. Something has been said of this in the introductory chapter and need not be repeated. What is the meaning of this presence of the categories not only in the contemplated but the enjoyed?

One way of solving this problem is to say that the mind is aware of the categories in its experience of itself, and then imputes them to its objects. Whether this answer has ever been attempted on a thorough-going scale, I do not know. But it has often been attempted in respect of the categories of causality and substance in particular. We find these characters in ourselves, and we interpret things, it is said, in our own likeness and find that the interpretation is successful. Now it is certain that experience of our own minds and experience of external things play upon each other reciprocally, reinforce and elucidate each other. When we have learned in ourselves the continuity, of a decision with its motives, of the issue of a train of thought with its premisses, of the mere unfolding of an idea in its details with the vague and implicit apprehension of the same idea, and particularly the continuity of our performances with our intentions; we can then look to external things and events to see whether there is not such continuity also there, the same definite order of succession. Or, again, whether in things there is not the like permanence in change that we can so easily detect in our enjoyment of ourselves. We speak then of causality or substance in external things, of physical causality and physical substances; and having these conceptions we come back to our own minds and ask whether we ourselves are not subject to physical causation, or are not substances in the same sense as external things, and we may thus raise problems which seem to us of great difficulty. Out of this interplay of mind and things it follows that while, on the one hand, we speak of force or power in physical things in language borrowed from our own wills; on the other hand, psychological terminology, as in such terms as apprehension or comprehension or conception, is largely derived from experience of physical things or of the action of our bodies on physical things.

But the mutual interplay of our experience of mind and things, which is an indisputable fact, is very far from the imputation by the mind of its own characters to external things. One simple consideration is enough to show that we do not merely construe things on the analogy of ourselves. For there must be something in the things which makes the analogy valid, or which gives a handle to the alleged imputation. If all we observe in external events is uniform succession, to impute to one of them a power to produce the other is a fiction, the fiction which Hume set himself to discredit. It may be. serviceable anthropomorphism, but it is not science nor philosophy. If there is no power traceable in things, then there is none; if the number of things is due to our counting, then there is no number in the things. The world then becomes indebted for its pervasive and prerogative characters to mind. Such a result is only satisfactory if the process is carried further, and if every character in things is attributed to mind, otherwise we could not understand how things should offer a reason to us to construe them so. I do not say this result is not true merely because it disagrees with our hypothesis of method, that we may treat mind as merely one of the many things in the universe. Yet at any rate we are bound before accepting it to see whether an explanation is not possible consistent with that hypothesis.

But now if there is something in the things which gives colour to the imputation, if for instance there is something in external things which is identical with the causal or substantial continuity which we find in mind, when we do not take that experience to be more than it really is, the imputation is unnecessary. Things may be numbered because they already contain number, not because they can be counted. On the contrary, they can be counted because they are countable and numerical. All the profit then that we can derive from the interplay of mind and things in becoming aware of the categories is that we may more easily derive from the enjoyed than from the contemplated the nature of the categories; which categories they share in common. Of this liberty we shall avail ourselves.

but because they are fundamental properties of Space-Time.

Are we then to be content with the bare fact that the categories are unlike empirical characters in belonging to all things, and in particular in belonging to minds as well as to external things? Such a coincidence would be sufficiently remarkable, but it clamours for the discovery of a reason. The reason is that the categories prove upon examination to be fundamental properties or determinations of Space-Time itself, not taken as a whole, but in every portion of it. They belong to all existents because, if our hypothesis is sound, existents are in the end, and in their simplest terms, differentiations of Space-Time, the complexes of events generated within that matrix. If that hypothesis be sound we should expect to find the pervasive features of things in the characters of their ultimate foundation. Or to put the same thing in another way, when and if it is seen that the categorial characters of things are features of any bit of Space-Time as such, merely so far as it is spatio-temporal, we are forced to the further conclusion that the empirical characters of things, their qualities, are correlated with the empirical groupings in Space-Time, and that things with their qualities are, as our hypothesis supposes, complexes within Space-Time. The categories are, as it were, begotten by Time on Space. It will be our business to exhibit this proposition in some detail with respect to the various categories.

The gist of the formula will perhaps be understood best by meeting in advance a possible misunderstanding, to the danger of which I shall recur more than once as the inquiry proceeds. Spaces or times it will be said have, it is true, magnitude, have identity, have a universal character, have existence. The categories, or at least some of them, are indeed applicable to spaces and times or, if you will, bits of Space-Time. These are instances which fall under these various categories, just as trees and dogs and tables do. But since they are but instances of the categories, the source of these categories must be found elsewhere. Now the clue to the understanding of our thesis is that the categories are not applicable as it were ab extra to spaces and times, but that they are applicable to things (including minds) because they flow from the nature of the space-times which they occupy or which they are. Applicability to space-times has no meaning for the categories, which are the features or determinations of the space-times themselves. I do not wish to anticipate too much, but a single instance may suffice. My mind exists at this moment because it occupies a certain portion of Space-Time, and that bare occupation is existence. Moreover, it is so far universal, that I remain in broad outlines the same mind whether I am here in Glasgow or there in Florence. That transplantation does not affect my identity. Caelum non animum mutant qui trans mare currunt.

Kant's treatment of the Categories.

In making this inquiry into the categories I have the good fortune to be able to make use for my own purposes, first, of the great later dialogues of Plato and, next, of Kant's work in the ‘Schematism of the Categories’ and above all in the’ Principles of the Understanding,’ the most significant and fruitful chapter of the Critique of Pure Reason. But it would be at once tedious to the reader and an interruption to the argument to indicate in detail where I have been helped by Kant. Indeed it would seem at first sight as if little help were to be derived from him in this matter. For the drawbacks and deficiencies of Kant's doctrine of knowledge in general and of the categories in particular are obvious enough. The categories are referred, like the forms of Space and Time, to the mind, because it is thought that what Hutchison Stirling called the “empirical instruction” does not contain them already. They are universal and a priori and belong therefore to the understanding, and are sharply separated from sense and its forms. Nevertheless, Kant is far removed from the notion that we manufacture or work up objects of knowledge by means of the categories, still less that we impute these forms to objects. They are for him veritable elements in objective knowledge, though they are the contribution of objective mind and not of the empirical instruction. And of still more importance and value is his effort to supply what he calls a "proof" of the principles of the understanding. In essentials the “proof” is this,2 that objective external experience contains the categories in correspondence with the features which the experience of Time possesses as given in the inner sense,—such as that it has duration, determinate order, permanence, is fuller or less full3 and the like. Since the form of external experience is Space, it is not so far a cry from this reasoning to the present doctrine, founded not on any pretence of proof or reasoning but on empirical inspection, that the categories are begotten by Time on Space, or are fundamental features of any space-time. For Space and Time are for Kant also forms of the mind, though the categories belong to understanding and they to sense.

Unfortunately the separation of the forms of sense or intuition from those of the understanding, and of both from the empirical instruction, gives to Kant's analysis an air of artificiality and unresolved miracle, and perhaps it is not to be wondered at that those who have regarded his formal procedure rather than the spirit of it have represented the forms as if they were instruments used in working up knowledge, as planes or chisels are used in carpentering wood. The artificial separation does not arise for us. For the categories are for us expressions of the nature of Space-Time itself, and on the other hand the empirical instruction consists of nothing but complexes of this same space-time stuff. All the elements of things we know are ultimately of the same stuff. But in spite of these difficulties I cannot think that this part of Kant's doctrine is so innocently inadequate as is often believed. And I am making these remarks not in order to fortify myself by his authority, which I certainly could not invoke, but to record a grateful conviction that with or after Plato there is nothing comparable in importance upon this subject with what may be learned from him, even by one who believes that mind which is Kant's source of categories has nothing whatever to do with the matter, and that mind is only a name for minds which are empirical things like other empirical things, and like them possess categorial characters and for the same reason as other things possess them, that they are all alike empirical complexes of space-time stuff. Leave out from Kant the objective mind with all the dependencies of that conception; and what he teaches us is mainly sound. It is true that the omission produces a considerable transformation, so considerable that the result would hardly be recognised as related to his doctrine by any affiliation of descent. But it is to be remembered that for a man of Kant's age the only method open to a philosopher, whether it was Kant or Reid, of indicating that the world of experience contains pervasive features as well as variable ones, was to refer this part of experience to mind in its objective character. Be this as it may, it is not always those who teach us most truth from whom we learn most, but those who best point the way to truth.

There are one or two questions of a general character about the categories which, to avoid repetition, will best be deferred till we have reviewed the categories in detail. For instance, whether at all, and if so in what sense, Space and Time themselves are to be called categories. Categorial they plainly are, and equally plainly Space-Time itself, which is the infinite matrix of all finites, is not a category. Again, it is plain from our description of the relation of empirical quality to Space-Time (that it is correlated with a certain complexity within Space-Time) that if our account be correct quality is not a category, and is no more than a comprehensive name for all the empirical qualities, and does not follow from the characters of Space-Time as such. Even for Kant, who regarded quality as a category, it only anticipates experience in respect of the intensity of the quality. It is in fact only another name for the empirical element in things But to avoid repetition at a later stage or imperfect discussion now, I omit these matters for the present.

We proceed then to describe the categories in order. The reader will bear in mind that they enter as constituent factors or as constitutive characters into every existent, whatever its quality. He needs only, in order to help himself in the abstract (that is elementary) inquiry, to think of empirical things, divest them of their qualitative colouring, and single out the categorial foundations of what the colouring is correlated with. While he may, if he chooses, regard also the embroidery he will be pleased to think only of the canvas.

  • 1.

    Bk. IV. ch. i. Some remarks upon point-instants and infinites will be found in ch. ix. of Bk. II. pp. 324 ff.

  • 2.

    At least this is one of the lines of thought Kant pursues in his proof.

  • 3.

    To which corresponds the category of intensive quantity.