Chapter II: Part 2

The limitations of our perception have, for Kant, a double aspect, which determines his division of the first part of the _Critique_ into two parts--the _AEsthetic_ and the _Analytic_. In the first place, our direct knowledge of space at any one time is always knowledge only of a part of space; our direct knowledge of time, whether in present consciousness or in memory of our own experience, is knowledge of only a part of time; and the things in the space we directly perceive, or in the time we experience, are what they are by their relation to space outside the space we see, and time beyond the time we experience, and that limited space and time we treat, therefore, as parts of one all-embracing space and one all-embracing time, and in the conception of an indefinitely extended space and time we can think of the space in which all things exist, and the time in which all things occur, of which we only see and experience a small part. The science of astronomy obviously talks of space and time far beyond anything we could ever perceive, but we go beyond such direct perception in such simple expressions as "forty miles from here" or "three days hence." And, when Kant says that space and time are only phenomenal, he does not mean that they are mental, but that we only know them through perception, and that we get at absolute space and time not by going from what we perceive to what we think, but by thinking of what we perceive indefinitely extended. All definite statements about space must come back in the end to "so far from _here_," all about time to "so long from _now_," and the fact that all our knowledge of space and time is got by adding to or extending in thought the space and time we directly perceive does, according to Kant, solve some obstinate puzzles about the nature of space and time.

In the second place, if we consider our knowledge of objects, we realise that, as we said, at any one moment we only perceive them in part or from one position. What we directly perceive of them is fragmentary and discontinuous, one aspect seen now, and another aspect seen at another time. But we do not think of the things as existing in that discontinuous way; we think of them as having a nature of their own. That does not contradict, but is something very much more than, what we perceive, and our knowledge of any object is got by piecing together the aspects we directly perceive; but that piecing together, or synthesis, is not haphazard. It is governed by rules--rules partly derived from the nature of the particular thing we are concerned with, and partly more general rules, which come from the relation of this work of piecing together to the framework of space and time by help of which it is done.

Kant's conception of knowledge, then, is something like this. Each of us is in direct contact with reality, but we perceive directly only a small part of it, and, as our consciousness moves on in time, and as we change our position in space, we are directly conscious of different small portions of reality. A part of the whole is illumined by direct perception, but the whole stretches beyond that indefinitely in space and time. In the part we directly perceive there is a temporal order and a spatial order. Things are given to us arranged in space and ordered in time, and these arrangements or orders in the space and time that is directly given to us in perception have certain rules, and we think of these principles of arrangement as extending indefinitely beyond the space and time given to us in perception. When we make judgments about reality beyond our perception, we think of things as so arranged in the space and time beyond our perception as we should see them arranged were the range of our perception sufficiently wide. Further, it is most important to remember that we do not remain in one place and at one time and make guesses of what may happen in the darkness beyond. Though our perception at any one moment is limited, we can connect what we see at one time with what we see at another. We can, by means of language and writing, use the perception of others to fill out our experience, until gradually our scientific judgments, our knowledge of what we should perceive under all sorts of possible experience, seems to bulk much more largely than could our individual perceptions. But we are still, Kant would say, getting at our knowledge of what is beyond by piecing together what we and other people have perceived, and the whole is always much more than that.

What, then, is meant by the contention that we can know things in themselves which Kant is earnest to refute? It might mean that we do in perception attain to a complete knowledge, but that would be obviously untrue. As Kant understood the claim, it meant rather something like this: In thought we are obviously not limited by our perception. We are always assuming certain principles, such as the laws of space or the principle of causation, to hold of all reality, both what we do and what we do not directly perceive. May we not say, then, that these principles hold of all reality, and argue from that fact to what the nature of the whole must be? If everything that we know is caused, _e.g._, may we not apply the principle of causation to all reality and say that it must have a cause?

When we come to consider the _Dialectic_, the second main division of the first _Critique_, we shall notice Kant's detailed analysis of these arguments, and how he points out that you can in this way get contradictory results. In the meantime it must be observed that in these arguments we start from principles applied to what we perceive and expressing connections between the different things we perceive, and then apply them beyond everything we do or could perceive. That means that we imagine that we can take these principles out of relation not only to this or that detail of perception, but out of relation to any perception at all, and thus apprehend reality by thought independently of perception.

Kant's answer is that thought cannot directly apprehend the nature of the whole, and these universal principles, such as the principle of causation, are only principles by which we connect one perception with another to amend the discontinuous and fragmentary nature of our perception; they are rules for the synthesis of what we perceive. By so synthesising our perceptions we come to a less imperfect knowledge of the whole, but apart from perceptions the principles have no meaning at all.

Kant's idealism, _i.e._ his insistence that we know only phenomena, not things in themselves, is relevant to his problem, because it implies the denial of the view that thought has objects apprehended independently of perception, and because it insists that we can only know directly what we perceive, or things as they appear to us, that in our process from perception to knowledge we start with what is present to our perception and end with what is or with what might be present to our perception, and that this process is possible by reason of our continued consciousness in time. The process, Kant holds, is governed by certain principles. These depend upon the part played by space and time in all our perception, and the manner in which we employ space and time in piecing together our discontinuous perceptions.

Now, obviously it is quite possible to hold this position without having thought out what is implied in being present to the mind in perception. This is what Kant did. He describes perception in different and inconsistent ways. The reason for this inconsistency is that Kant is not concerned with the nature of perception, but with the relation of what is immediately perceived to what is not but may be immediately perceived, and he therefore never worked out any consistent account of perception. He sometimes talks of perception reaching objects directly, and refutes the view that we perceive only what is in our mind. (This, indeed, is implied in his distinction of space and time as forms of _external_ and _internal_ sense respectively.) But usually he takes the ordinary idealist view that we do not perceive things, but affections produced in us by things. Owing to this inconsistency Kant constantly seems to be stating very much more than he has any right to. This is especially true in all that he says about knowledge being confined to phenomena and not extending to things in themselves. When he talks of our knowing only phenomena, he sometimes seems to mean that we know objects, things in themselves, only in part, in so far as they appear to us. That would make the distinction between the phenomenon and the thing in itself a distinction between the same thing imperfectly and perfectly understood. He sometimes, and this is his more usual view, seems to mean that we are aware of appearances, entities separate and distinguishable from the objects which produce them in our minds. But if we work out in any of Kant's arguments the point of his appeal to the fact that knowledge is only of phenomena, we shall find that in every case the difference between a subjective idealist and a realist view of perception, of what "being present to the mind" means, is irrelevant, and that his argument holds on either theory.

We must now turn to Kant's account of space and time which is given in the _AEsthetic_, the first part of the _Critique_. He begins by showing the impossibility of the two views of the nature of space and time which then held the field, the views of Newton and Leibniz. Newton had thought of space and time as realities, things in themselves existing along with other things. But obviously we cannot think of space as a separate thing existing by itself; for space without things would have no determination or possibility of determination, and would be to us just nothing, whereas, as it is, it is something to us. The same holds of time. The Newtonian doctrine, Kant says, "forces us to assume two eternal, infinite, and self-subsisting non-realities, which are there, without any reality in them, only that they may comprehend all reality." Just because things are in space and time, space and time are not themselves things. But if this makes us say that space and time are only relations between or qualities of things, we find ourselves in difficulties as obvious. We do not come to apprehend space and time by comparing things and seeing that they have a common quality of being "spatial" or "temporal," as we come to apprehend redness, _e.g._, by seeing red things. The perception of space and time is implied in each and every perception of things. We cannot, therefore, derive them from our study of things; we must begin with them. Further, Kant notices, as against Leibniz, that space and time are not ordinary concepts because they have no instances. Different men are instances of man, but different spaces or times are only parts or determinations of the one space and the one time. As against the view, then, that would make space and time only relations, derived from our comparison of things which are not temporal or spatial, Kant insists that space and time are a priori. We cannot see things without seeing them outside one another--_i.e._ in space--or experience succession or change without experiencing it in time. Space and time, then, have a certain independence of things in space and time. The qualitative differences of things in space or events in time do not affect the nature of space and time, and we can and do study and discuss spatial and temporal relations quite independently of such differences.

Space and time, then, can be abstracted from things in space and time. Yet, on the other hand, we cannot think that space and time exist independently of things. They do not exist in abstraction; for, though the specific differences of things in space and time are irrelevant to the nature of space and time, if there were no things, or if there were no differences, there could be no space and time as we know them. "The empirical perception," says Kant, "is not compounded of phenomena and space, of the sensation and the empty perception." Space and time, therefore, Kant says, are not things in themselves.

What, then, are they? Kant's answer is that they are forms of our perception. Space is the form of external perception, and time is the form of internal perception, and Kant holds that by this answer we can understand both how our knowledge of space and time may be a priori, how spatial and temporal distinctions may be abstracted from the differences of things, and how we may avoid the difficulties consequent on regarding time and space as independent things.

What, then, does Kant mean by _form_? He seems to mean two things, which he does not clearly distinguish. The first meaning is best described in his own words: "In the phenomenon I call what corresponds to the sensation the matter of the phenomenon, and that which causes that the manifold of the phenomenon is perceived as arranged in specific relations I call the form of the phenomenon." We are here face to face with the ultimate difference of form and matter, or order and that which is ordered. When Kant calls time and space the form of our perception he is simply calling attention to the fact that in all that we perceive we find this distinction. It is something found, given, not made by us. By the word "form" Kant does not mean anything specially subjective as contrasted with matter or content, for he carefully distinguishes between space and time, and such qualities as colour, which get their nature in part, he thinks, from the specific nature of the sense organ. Compared with such qualities space and time are objective. The phrase "forms of our perception," then, does not really explain anything about space and time; it only emphasises the fact that the distinction between space and time and objects in them is found in what we perceive, and that there is no meaning in discussing either side of the distinction as though it were quite independent of what we perceive.

But form has also another meaning which justifies Kant in calling space and time _only_ forms of our perception, and hence subjective. For, while these forms are _found_ in what we perceive, the distinctive part which they play in our knowledge is due to the fact that we use space and time as a framework by which to connect our scattered experiences. We come to think of the space and time we perceive as parts of an absolute space and an absolute time. We perceive parts of space and time, but absolute space and absolute time we do not perceive. They are the form we perceive imagined indefinitely extended. We order the particular parts of space and time which we do perceive in reference to absolute space and time. Yet absolute space and time are only known through the finite parts of space and time which we actually experience. Hence absolute space and time are not perceived realities or perceived orders, but ways in which we organise and arrange what we perceive. Now, the qualities of space and time which are hard to think of as the qualities of a thing that exists, _i.e._ their infinite divisibility and infinite extension, are qualities of absolute space and time. When we say that space is infinitely divisible, we do not mean that any existing thing is made up of an infinite number of parts. The divisibility of space and the divisibility of matter are quite different. An inch as a spatial determination is infinitely divisible, but the divisibility of the actual stuff which any inch may measure is a matter of empirical investigation, and ought to admit of a definite answer. That means that, while we use determinations of space which we consider infinitely divisible and infinitely extensible to measure things in space, we do not consider that these determinations, fractions, or multiples of inches or centimetres, have anything to do with the constitution of the thing they measure. It was not put together in fractions of inches. Thus we must distinguish between space as the form of what we perceive, the next-each-otherness of things, and the use we make of that form to construct by means of measurement order in all different perception. The first is obviously the form only of what we perceive, and gives rise to no transcendental questions. But the second, infinite space, though it seems to transcend our perception, has still only meaning in reference to perception, is only a way of ordering our perceptions. The same holds good of time.

We can see now what Kant means by saying that time and space are empirically real and transcendentally ideal. Kant does not maintain that space and time are illusions. They are a constant element of what is given us in perception. It is only when we try and go beyond our perceptions, and take space and time as things existing independently of what we perceive, thus trying to transcend the limits of possible perception, that we fall into illusion. Space and time have meaning only as elements in what we perceive, or in connecting what we perceive now with what we may perceive.

*CHAPTER IV*

*THE CATEGORIES AND THE PRINCIPLES OF PURE UNDERSTANDING*

Kant makes the distinction between perception and understanding depend upon the distinction between the receptivity and the spontaneity of the mind. In the _AEsthetic_ he has been concerned with time and space as elements in what seems to be given to the mind. Before we begin to ask the questions of science, before we analyse, describe, or classify, before we have to think, we perceive. Time and space are not got at by thinking or generalisation. For before we can say anything about any part of our experience, it is _given_ us in a certain spatial and temporal order. If we open our eyes at any moment, we are, without any conscious effort of thought on our part, confronted with an elaborate content. It seems simple to distinguish this receptive attitude of the mind in perception from its activity in thinking.

The distinction is not really so simple as it appears. For we all know that what we perceive depends, at least to some extent, on the mind's activity. We are familiar with the reflection that men see what they want to see or what they are looking for. This is clearly shown in the case of hearing by the difference in what we hear when we are listening to a language we understand and when we are listening to an unfamiliar language, or in the common experience when, after failing to hear what someone has said, we think what it must have been, and then seem to recall the sound, not as we heard it, but as we should have heard it if we had heard it rightly. Anyone who reflects on the process of fast reading will realise that we do not perceive or notice all the letters on a page; we fill in from our imagination, as we discover when we read words that are not on the page. It is a very hard thing, giving up all interpretation and inference, to describe faithfully just what is there to see.

Passive perception, then, does not exist, and our thought affects our perception. Yet, at the same time, the distinction between thought and perception, although not simple, is real. For although our previous thought affects our perception and we see things already classified, see books, and tables, and chairs, not merely coloured surfaces, yet we can distinguish between simple immediate perception and the process of thought which begins when we ask, What is that? _i.e._ when we begin to make judgments.

The characteristic of thought, according to Kant, is synthesis, or putting together, and all synthesis is the work of the mind. When we begin to describe and classify the contents of our perception, we pick out separate qualities from the continuous whole we perceive, and group them together. This grouping is, of course, determined by the likenesses and differences which we perceive everywhere, but we do not, in judging, confine ourselves to noticing likeness and difference. For any content of our perception has some point of resemblance, and some of difference with any other. We are concerned with likenesses that go with or are the signs of other likenesses. On the basis of perceived likeness we erect the notion of things and qualities of a certain kind. In doing this we go beyond what we see, and unite and arrange the contents of our perception through concepts. That is what we are doing when we say that is a so-and-so. For example, if I say that rock is like a dog, I am simply expressing a likeness I perceive. I do not imply that the rock is therefore alive or will bark; I am not going beyond how the rock looks; but if I say that object is a dog, I assert that all that is implied in being a dog will hold of that object, _i.e._ that it will have a certain appearance and behaviour, which is known. I can anticipate, therefore, how it will behave, look, and sound under certain circumstances. All these phenomena, the appearance, the barking, and running, though I may perceive them at different times and places, are grouped together in the judgment, "That is a dog." This is what Kant means by saying, "Concepts depend on functions. By function I mean the unity of the act of arranging different representations under one common representation." Concepts, therefore, always refer to perception, and it is by means of concepts that we are enabled to introduce such order into what we perceive, that we can anticipate from what we perceive what we shall perceive. "Perceptions without concepts are blind." Without concepts what we perceive would not lead us in any way beyond what is immediately given. "Thoughts without contents are empty." Concepts are nothing, and have no meaning apart from the contents of perception which they unify.

Most of these concepts are what is called empirical. We get at them by observing likenesses and differences in what we perceive, and observing which are significant and important, and which are what we call accidental. Science, in its discovery of laws, is only carrying further this process which is implied in all simple judgments. By observing likenesses and differences, their uniformities and variations, and discovering those which are a key to the rest, we improve our concepts, and thereby have more knowledge of what we call natural laws, and can more and more anticipate experience. With these empirical concepts and their development Kant is not concerned. But there are certain concepts of which Hume had observed that they are not obtained in the ordinary way from an examination of the contents of experience. The two with which he chiefly concerned himself were substance and cause. These concepts seems to play an especially important part in the ordering and arranging of the concepts of experience. For the work of science, in moving from a simple observation of likenesses and differences to a knowledge of empirical laws, depends upon certain assumptions or principles, like the principle of causation or the principle of the conservation of energy. These principles imply concepts not derived, like the others, from generalisation from experience; they are the synthetic a priori judgments which, as we have seen, constituted a special problem for Kant.

Kant is first concerned to ask where these a priori concepts come from, and how many of them there are. This inquiry he calls the metaphysical deduction of the categories. Having answered that question, he then goes on to ask by what right we assume these principles in our dealing with experience. This, the most important and difficult section of the _Critique_, he calls the transcendental deduction of the categories.

Most concepts, as we saw, are empirical. We take certain likenesses and differences we observe as the mark of a real unity in the things. The different natures of different things we do not fully know, but we distinguish them by the different uniformities we observe, and in order to explain our experience we assume the unity underlying these perceived likenesses. Iron, dog, fire, are names for the natures of things which we see manifested in our experience. The concept, then, is got from what we perceive, though it stands for something more than we perceive. How, then, can there be any concepts which are not got from the empirical differences of things we perceive? Let us take such a concept as substance, and see whether we can discover where it comes from. Locke had been puzzled by discovering that he could not, in any object, find anything which was its substantiality. Calling anything a substance is not like saying that it is hard, or green, or heavy; we are not concerned with specific differences in things, but we are not therefore saying what is meaningless. There is something, namely substance, which we can distinguish from the hardness, or colour, or weight that we perceive. That something we do not perceive; we assume it whenever we talk of a thing being hard, and green, and heavy. A thing's substantiality is just the unity of its perceivable qualities. But such a unity is implied in the concept of any object. Substance, then, is a name for one of the general principles implied in our assuming that what we perceive are real objects.

Kant generalises the result of this inquiry into particular concepts of this kind. He holds that a priori concepts or categories (_i.e._ the concepts which we do not get from empirical differences of things) stand for principles implied in thinking of things as objects or in judging. If we want, therefore, to find out the number of the categories, we must ask how many different kinds of unity are implied in judgment, or what are the conditions of judging any object. Kant does not here help, but rather misleads us in this inquiry. For he unfortunately thought that the different kinds of judgment could be discovered without further ado by taking the list given in formal logic. He therefore first makes a list of categories, based on the logical forms of judgment, and then tries to show the connection between these categories and the principles which were, as he had discovered, assumed in the mathematical sciences.

The actual movement of his thought is, I think, different. He asks if there are any general conditions implied in all judgment. His answer is that all judgments, all statements, that is, which claim to be true, imply determination of time and space. From that determination certain principles can be deduced. If time and space are implied in all judging, then these principles will equally be implied, and will hold of all things which can be objects for us.

It will be easier to understand Kant's arguments if we invert the order of the _Critique_ and begin with examining the nature of the principles of the understanding or of one of them.

The categories which are of importance in Kant's argument are quantity, quality, substance, causation, and reciprocity, and necessity, possibility, and actuality. The last three are less important than the others, and we shall not deal with them.

To the first five of these categories correspond the following principles:

(1) Quantity. "All phenomena are, with reference to their perception, extensive quantities."

(2) Quality. "In all phenomena the real, which is the object of a sensation, has intensive quantity, that is, a degree."

The last three are classed under a general heading of _Analogies of Experience_, whose principle is: "Experience is possible only through the representation of a necessary connection of perceptions." They are

(3) _The principle of the permanence of substance_. "In all changes of phenomena the substance is permanent, and its quantum is neither increased nor diminished in nature."

(4) _Principle of the succession of time, according to the Law of Causality_. "All changes take place according to the law of connection between cause and effect."

(5) _Principle of co-existence, according to the law of reciprocity or community_. "All substances, so far as they can be perceived as co-existent in space, are always affecting each other reciprocally."

These principles, Kant points out, are assumed in the sciences of applied mathematics. The application of geometry to the world we experience assumes that all phenomena are extensive quantities; physics assumes that quantitative expression can be given to the qualities of objects other than their size, their weight, _e.g._, and all scientific determination of change assumes the three principles which Kant calls analogies of experience: the permanence or conservation of amount in changes, the necessary connection of things in time, and the reciprocal interdependence of things which exist at the same time. These principles are not proved by science; their validity is assumed in all scientific investigation. On what, then, does it rest?

We shall follow Kant's argument more easily if we take his account of one of these principles--the principle of causation. For what is said of that will hold, with necessary changes, of the others, and, as we have noticed, it was Hume's criticism of causation which first led Kant to formulate the critical problem. Hume had pointed out that we had never such insight into causal connection as to be able, from mere inspection of a cause, to foretell the effect without any reference to experience. He declared, on the contrary, that there was no difference between observed succession and causation so far as concerned the objects observed. In each case we see first one thing and then another. The difference, then, between mere succession and causal connection can only be in us, in the way we come to feel about certain successions we observe. In technical language, the necessity of causation is subjective.

How does Kant answer this position? He begins, as is usual with him, by taking the problem a little further back. Causation is a connection we predicate between what we see at one time, and what we see at another. Now if we take into account only the fact that we see one thing at one time and another thing at another, there is no difference between what we see when we successively see two things which we judge to co-exist, and when we see two things one of which we judge to have succeeded the other in time. Hume, therefore, proved too much. His argument would show that we have no grounds for distinguishing between apprehension of succession and succession in apprehending, but such a distinction is the basis of our apprehension and understanding of change or movement. If, then, we examine how we distinguish between apprehension of succession and succession in apprehending, we may see on what the principle of causation is based.

An instance will help to make this point clear. Suppose that I am sitting in a room, and look first at the door and then turn round and look at the window. There are two successive acts of apprehending; the content of the first is the door, of the second, the window, but the succession, I say, is in my apprehending. The door and the window have co-existed all the time. Suppose, again, that I look out of the window and see a cab in front of the house opposite, come back into the room, and then look out again and see the cab in front of a house further down. Here, again, are two successive acts of apprehending, the content of the first houses with cab in front of one house, of the second houses with cab in front of another. This time I say the houses have gone on co-existing, but the cab has moved. The difference in what I see this time is due not to me, but to the cab. The succession is in the thing apprehended. If we just think of the contents apprehended, we have first A, then B, and say A and B co-exist in the first instance, and have CD and CE, and say D and E have been successive in the other. Why in the second case do we not say when we look out of the window the second time: Here is another row of houses, which, though they look exactly the same as the ones I saw last time, have got the cab in a different place? That is the land of thing one does say in a dream. Why would it be inadmissible in waking life?

Let us first ask how we ever come to make the distinction between change in the content of our perception, which is due to change in us, and change in the things we perceive. Look out of a window into a busy street. As we look certain things remain the same, the houses opposite, the lamp-posts, and so on, but other things change. The permanence of part of the contents guarantees us that the change we perceive is not due to us: if it were, these would change also. Therefore it must be in certain of the things. Change is perceived against a background that is permanent and does not change. But any such particular perception is, of course, very limited. We do not see all the world at once, and we only come to know a larger extent of reality by means of memory, which enables us to put together what we see at one time with what we see at another. We have got to try and understand how it is that we make this distinction, which is clear to us in small isolated bits of experience, hold of all experience. Now if reality did not change, and we were conscious of our own movements, we could go from one point to another of reality and back again, and could be aware that the changes in our perception were all due, not to change in reality, but to us--were our history. We should know that the different things we saw were co-existing all the time, and we should, in describing them, try to describe them, as in a map, as we should perceive them if we saw them all at once. The succession would be subjective, the co-existence objective. If we perceived nothing but change, we should be incapable of distinguishing between our changes and the change outside us, for all succession in our experiencing would be experience of what was successive, and there could be no distinction between psychology and science. Our experience of reality is not like either of these suppositions, but like both of them combined. Some succession of our experiencing is experience of the co-existing, some experience of succession.

Reality stretches out beyond us in space, some of it changing and some of it permanent; we cannot tell simply from the difference in what we perceive whether the difference comes from change in us or change in the thing. We can tell that only on the assumption that we are having fragmentary views of a whole that is continuous. The only continuity we know is the continuity of our own experience made possible by memory, and we try to interpret that experience in the light of the larger continuity of the world which our experience breaks up. As we go from one place to another, notice now this thing, now that, we can test interpretations made on this assumption. Wrong interpretations are those which make our experience inconsistent. If we thought that what happened at one time had no relation with what happened at another, that anything might happen any time, our experience and our own life would be the merest jumble. Our experience attains consistency only as we learn more and more to disentangle the differences in experience which come from our changes, from the changes and the variety which are part of the whole connected system of reality, of which we see now one fragment, then another. The distinction between succession in our apprehending and apprehension of succession, which is the basis of all experience of change, implies the recognition of change as not arbitrary but part of a connected system of reality. As Kant puts it in his formulation of the principle of the analogies of experience, "Experience is only possible by means of the representation of a necessary connection of perceptions."

But if our perception of reality is fragmentary, how can we think of reality as other than fragmentary, how can we fill up the gaps? Only by thinking of the whole as a connected system in space and time. For it is the nature of space and time that they can be thought of independently of the specific nature of the things in space and time, and that the space and time we perceive in any one experience must be thought of as parts of an all-embracing space and an all-embracing time. We cannot follow the whole history of a change from A to B, we can only say that, if our experience is to have any consistency, we must think that the fact that we first saw A and then saw B implies in this case that the change from A to B is part of the continuous system of change in time, that it is determined in time. But to think of an event as determined in time is not to think of it as determined by time, for time in itself could not produce one thing more than another. It is to think of it as determined by the nature of what precedes it in time. We therefore conclude that like causes have like effects; for, if anything could cause anything, we should never know that change in what we observed was due to change in us--in the position of our bodies, _e.g._--and the experience of objective change would be impossible.

The principle does not tell us of itself what causes what. That can only be discovered by empirical investigation. That is necessary because we do not, as we seem to have assumed above, simply see one thing becoming another. We see parts of all kinds of changes. Hence succession may be objective but not causal. Science has, by observation and experiment, to disentangle and isolate different changes, but it could not do this without assuming the principle of causality.

Causation, then, and the other assumptions of the physical sciences, are shown by Kant to be "grounds of the possibility of experience." We cannot deny them without denying elementary distinctions in our experience, without which life would be a chaos, and which are assumed and justified every moment. While Kant thus demonstrates the validity of such principles, he is also insistent on the limitations of their application. They are principles which give consistency to experience, but must not be applied save in reference to what we experience. They apply, in his words, "only to phenomena." The purport of this limitation can, again, be most easily seen by examining the principle of causality. By means of that principle we connect one event with another, but the reality is not two different but connected events, but a continuous process. The continuous process escapes us, because our perception of it is fragmentary and discontinuous. Inasmuch as a judgment of causal connection asserts that the events we separately notice are connected, it is true, but it is false if taken to imply that reality consists of a series of discontinuous events or stages which are yet connected. Such an assumption would mean, in Kant's words, that causation is applied not to phenomena (things as they appear to us), but to things in themselves (things considered apart from the manner in which they appear to us). If we realise its falsehood, we can, he thinks, evade the contradictions which he examines in the _Dialectic_.

*CHAPTER V*

*THE ANTINOMIES AND CRITICISM OF THE PROOFS OF
THE EXISTENCE OF GOD*

So far we have been considering the positive side of Kant's argument, his attempt to confirm the validity of the principles of science. We must now notice the negative side, his attempt to limit the application of these principles, and his denial of the possibility of knowledge in certain spheres.

We saw that Kant in his _Prolegomena_ summed up the argument of the three chief divisions of the _Critique_ as an answer to the questions: How is mathematics possible? How is pure science of nature possible? and, How is metaphysics possible? He qualified the last question by adding "as a natural disposition of the mind." The argument of the _Dialectic_ is that metaphysics, in the sense of inquiry into objects which transcend the bounds of experience, is not possible as a science, but that metaphysical questions arise naturally from the nature of human reason. They cannot be answered. All we can do is to see why we cannot answer them.

Kant thought of knowledge as a process of extending the bounds of perception, of piecing together the fragmentary glimpses we get of the world, stretching them out in spatial and temporal determinations that go beyond what we have actually experienced, connecting and linking up the events which we perceive discontinuously. As science extends, the range of our knowledge widens, but the process of extension never reaches its completion. There are always more facts to be discovered and explained. Science, therefore, can never rest content with its achievements, but must always demand that the investigation of conditions should be pushed further back and on. From this sense of the incompleteness of all actual knowledge, and of all there is that might be but is not known, arises what Kant calls an ideal of reason, a demand that, in all investigation into the conditioned, we should go on till we come to the totality of conditions. This ideal he holds to be serviceable and necessary. It has, however, a natural tendency to pass from an ideal to an idea, and in so doing it gives rise to the contradictions with which the _Dialectic_ is concerned. If all our investigation is governed by the thought that it must go on until it reaches completion, we naturally speculate on the fulfilment of that ideal, and try to form an idea of that totality of conditions, of how we should think the world if we knew it in its completeness. Herein we hypostatize the ideal or make it an idea, and we fall into contradiction; for we cannot really know the whole without knowing all its parts. If we give up the slow and never-completed process of knowing one part after another, and try to jump to the idea of the whole, we reach quite contrary results, as we apply to the conception of the whole one or other of two assumptions implied in our investigation of the parts.

Kant sharply distinguishes between the principles of the pure understanding and the ideas of reason. The former are implied in all our knowledge, and the fact that experience is not chaotic confirms them at every moment. The second are ideals which guide knowledge, but are never realised. He calls them ideas of reason, because it is the special task of reason to lay down rules for the proper and complete working of the understanding. This task, he thinks, is exemplified in the logical nature of the syllogism which brings into unity the judgments of the understanding. As he used the forms of judgment as a guiding thread to discover a complete list of categories of the understanding, so he uses the forms of syllogism to discover a complete list of the ideas of reason. In both cases Kant's reference to logical forms is far-fetched. Actually the list in the _Dialectic_ seems to be influenced by a number of considerations not always consistent.

There are three main divisions of the _Dialectic_. (The first Kant calls the paralogisms of rational psychology.) All knowing and experience imply the unity of the self which knows. In actual experience that unity is qualified by the nature of what it unites, but we may try to think of it apart from and independent of this. This leads to an attempt to know the self by asking what must be its nature if it has the unity implied in knowing, and to argue that the soul is a substance and simple, not affected by the changes in the matter which it knows and therefore immortal.

The second division arises from the fact that in knowledge we are concerned with series--a series of addings together and a series of divisions, as of parts of space and time; a series of things arising one from the other, as in causation; and a series of things in dependence one upon the other. The _ideas_ of reason come from the thought of these series completed, and produce what Kant calls antinomies. For if we start with the thought that what we are trying to apprehend must be a whole, we get one series of results; if with the thought that we can only apprehend the whole by going from condition to condition indefinitely, we get another. Kant distinguishes four antinomies, each with thesis and antithesis. The thesis of the first is, "The world has a beginning in time, and is limited also in regard to space"; the antithesis, "The world has no beginning and no limits in space, but is infinite, in respect both to time and space." The thesis of the second is, "Every compound substance in the world consists of simple parts, and nothing exists anywhere but the simple or what is composed of it"; the antithesis is the contrary of this. The thesis of the third is, "Causality, according to the laws of nature, is not the only causality from which all the phenomena of the world can be deduced. In order to account for these phenomena it is necessary also to admit another causality, that of freedom"; the antithesis, "There is no freedom, but everything in the world takes place entirely according to the laws of nature." The thesis of the fourth is, "There exists an absolutely necessary Being belonging to the world, either as a part or as a cause of it"; the antithesis is a denial of this.

The problems of the third division of the _Dialectic_ arises from an attempt to think of a whole which shall include both the known world and the mind that knows. This attempt, which Kant calls the ideal of pure reason, leads to proofs of the existence of God.

As the _Dialectic_ proceeds, it becomes clear that Kant has another list to hand. He enumerates, as the three great objects of metaphysical inquiry, God, Freedom, and immortality, and in his discussion of the ideas of reason he treats them principally as attempts to give definite and dogmatic answers to the problems suggested by these three topics.

Immortality is the subject of what Kant calls the paralogisms of rational psychology. He argues that all attempts to prove the immortality of the soul by a priori arguments involve an argument of this kind: they begin by noting that death is always dissolution of some kind, that, therefore, what is not made up of parts and cannot be dissolved, cannot die. Then they urge that the soul is not made up of parts, and therefore cannot die. The fallacy in this argument is that it treats the unity of the self as though it were an object of knowledge. We can show that knowledge is only possible if the self has a unity other than that of a spatial whole, but we cannot therefore argue that it must be exactly like a spatial whole, in the sense that death in it can only be brought about by dissolution, but unlike a spatial whole in that in it there is nothing to be dissolved. The real nature of the unity of the self, Kant argues, cannot be known. All we can do is to reject a priori arguments either for or against its immortality.

Freedom is treated in the third antinomy of pure reason, and to that Kant devotes most attention, but others of the antinomies are concerned with the difficulties arising from the application of spatial and temporal determinations to reality as a whole, and to the category of necessity. Kant makes a distinction between the first two and the second two antinomies. It is the first two that express the inadequacy of temporal or spatial determination to reality as a whole. All such determination implies measurement, and measurement is always a relation of part to part. The antitheses of both antinomies express the inadequacy of any number to the expression of the nature of the whole, the thesis the inadequacy of regarding reality as an aggregate or addition of any kind. Each is strong in what it denies, and Kant's solution is that both thesis and antithesis are false, because you cannot apply spatial or temporal determination to the world as a whole.

In contrast the solution of the other antinomies is that both thesis and antithesis are true, and that is possible because they are concerned with different things. The third antinomy arises from the difficulty of applying the category of causation to the world as a whole. The assumption underlying the thesis is not, as is sometimes asserted, merely that the notion of infinity in itself implies a contradiction, but that a determinate result must have a determinate cause. If we think of what actually exists now as having been caused by what has preceded it, we must think of that which has had a determinate result being itself determinate. It is the familiar argument for a first cause. In causation we seem to be relating one event to another event, and are really only putting the question of origination further back. Yet, if we say that therefore we must suppose an absolute origination of change, a beginning of the series, we have to answer the question, How is it possible to think of the originating number of the series? For to think that something can arise from nothing is to contradict the principle of causation.

Kant's solution to this difficulty is important, for it had great influence upon his ethical theory. The category of causation applies only to phenomena. If we think of things as phenomena we must recognise that they are subject to the principle of causation; if we think of them as things in themselves, the category of causation does not apply to them, and their action may be free. The same action may therefore on its phenomenal side be determined, and on its nominal side, as the action of a thing in itself, be free. This may seem to be solving one contradiction by propounding another, till we remember that in causation we do not explain the relation of cause to effect. The relation we discover is between one instance of cause and effect and another. Like causes have like effects. The principle applies, then, in so far as things are like one another. It applies to changes which are aggregates or complexes of simpler changes which are like other changes. If and in so far as there are things which are more than aggregates of their elements, and are therefore unique, there are things to whose changes no laws of cause and effect are adequate. The point may be illustrated by the way we think about character. If we think of a man's character as his characteristics, his being this or that _kind_ of person, we must think of his action as so far determined, but that does not prevent us from thinking of his individuality as something more than any sum or combination of characteristics, as something essentially alive, which escapes all attempts to bind it by rules. It is the difference in Kant's words between man regarded "from the point of view of anthropology," and man regarded as a responsible moral being. We shall see in the next chapter that this distinction is the basis of Kant's moral theory. Here it must be noted that he does not claim that his solution of the third antinomy proves the fact of freedom. That, he held, no merely intellectual argument could prove. It only defends the possibility of freedom.