Critique of Pure Reason

by Immanuel Kant

  Instead of considering these subjects, the full development of which constitutes in reality the peculiar character and dignity of philosophy, we have to occupy ourselves at present with a task less brilliant, though not less useful, of building and strengthening the foundation of that majestic edifice of morality, which at present is undermined by all sorts of mole-tracks, the work of our reason, which thus vainly, but always with the same confidence, is searching for buried treasures. It is our duty at present to acquire an accurate knowledge of the transcendental use of the pure reason, its principles and ideas, in order to be able to determine and estimate correctly their influence and value. But before I leave this preliminary introduction, I beg those who really care for philosophy (which means more than is commonly supposed), if they are convinced by what I have said and shall still have to say, to take the term idea, in its original meaning, under their special protection, so that it should no longer be lost among other expressions, by which all sorts of representations are loosely designated, to the great detriment of philosophy. There is no lack of names adequate to express every kind of representation, without our having to encroach on the property of others. I shall give a graduated list of them. The whole class may be called representation (repraesentatio). Under it stands conscious representation, perception (perceptio). A perception referring to the subject only, as a modification of his state, is sensation (sensatio), while an objective sensation is called knowledge, cognition (cognitio). Cognition is either intuition or concept (intuitus vel conceptus). The former refers immediately to an object and is singular, the latter refers to it mediately, that is, by means of a characteristic mark that can be shared by several things in common. A concept is either empirical or pure, and the pure concept, so far as it has its origin in the understanding only (not in the pure image of sensibility) is called notion (notio). A concept formed of notions and transcending all possible experience is an idea, or a concept of reason. To any one who has once accustomed himself to these distinctions, it must be extremely irksome to hear the representation of red colour called an idea, though it could not even be rightly called a notion (a concept of the understanding).

  Transcendental Dialectic

  Book I

  Second Section

  Of Transcendental Ideas

  We had an instance in our transcendental Analytic, how the mere logical form of our knowledge could contain the origin of pure concepts a priori, which represent objects antecedently to all experience, or rather indicate a synthetical unity by which alone an empirical knowledge of objects becomes possible. The form of judgments (changed into a concept of the synthesis of intuitions) gave us the categories that guide and determine the use of the understanding in every experience. We may expect, therefore, that the form of the syllogisms, if referred to the synthetical unity of intuitions, according to the manner of the categories, will contain the origin of certain concepts a priori, to be called concepts of pure reason, or transcendental ideas, which ought to determine the use of the understanding within the whole realm of experience, according to principles.

  The function of reason in its syllogisms consists in the universality of cognition, according to concepts, and the syllogism itself is in reality a judgment, determined a priori in the whole extent of its condition. The proposition 'Caius is mortal,' might be taken from experience, by means of the understanding only. But what we want is a concept, containing the condition under which the predicate (assertion in general) of that judgment is given (here the concept of man), and after I have arranged it under this condition, taken in its whole extent (all men are mortal), I proceed to determine accordingly the knowledge of my object (Caius is mortal).

  What we are doing therefore in the conclusion of a syllogism is to restrict the predicate to a certain object, after we have used it first in the major, in its whole extent, under a certain condition. This completeness of its extent, in reference to such a condition, is called universality (universalitas); and to this corresponds, in the synthesis of intuitions, the totality (universitas) of conditions. The transcendental concept of reason is, therefore, nothing but the concept of the totality of the conditions of anything given as conditioned. As therefore the unconditioned alone renders a totality of conditions possible, and as conversely the totality of conditions must always be unconditioned, it follows that a pure concept of reason in general may be explained as a concept of the unconditioned, so far as it contains a basis for the synthesis of the conditioned.

  As many kinds of relations as there are, which the understanding represents to itself by means of the categories, so many pure concepts of the reason we shall find, that is, first, the unconditioned of the categorical synthesis in a subject; secondly, the unconditioned of the hypothetical synthesis of the members of a series; thirdly, the unconditioned of the disjunctive synthesis of the parts of a system.

  There are exactly as many kinds of syllogisms, each of which tries to advance by means of pro-syllogisms to the unconditioned: the first to the subject, which itself is no longer a predicate; the second to the presupposition, which presupposes nothing else; and the third to an aggregate of the members of a division, which requires nothing else, in order to render the division of the concept complete. Hence the pure concepts of reason implying totality in the synthesis of the conditions are necessary, at least as problems, in order to carry the unity of the understanding to the unconditioned, if that is possible, and they are founded in the nature of human reason, even though these transcendental concepts may be without any proper application in concrete, and thus have no utility beyond bringing the understanding into a direction where its application, being extended as far as possible, is brought throughout in harmony with itself.

  Whilst speaking here of the totality of conditions, and of the unconditioned, as the common title of all the concepts of reason, we again meet with a term which we cannot do without, but which, by long abuse, has become so equivocal that we cannot employ it with safety. The term absolute is one of those few words which, in their original meaning, were fitted to a concept, which afterwards could not be exactly fitted with any other word of the same language, and the loss of which, or what is the same, the loose employment of which, entails the loss of the concept itself, and that of a concept with which reason is constantly occupied, and cannot dispense with without real damage to all transcendental investigations. At present the term absolute is frequently used simply in order to indicate that something applies to an object, considered in itself, and thus as it were internally. In this way absolutely possible would mean that something is possible in itself (interné), which in reality is the least that could be said of it. It is sometimes used also to indicate that something is valid in all respects (without limitation), as people speak of absolute sovereignty. In this way absolutely possible would mean that which is possible in all respects, and this is again the utmost that could be said of the possibility of a thing. It is true that these two significations sometimes coincide, because something that is internally impossible is impossible also in every respect, and therefore absolutely impossible. But in most cases they are far apart, and I am by no means justified in concluding that, because something is possible in itself, it is possible also in every respect, that is, absolutely possible. Nay, with regard to absolute necessity, I shall be able to show hereafter that it by no means always depends on internal necessity, and that the two cannot therefore be considered synonymous. No doubt, if the opposite of a thing is intrinsically impossible, that opposite is also impossible in every respect, and the thing itself therefore absolutely necessary. But I cannot conclude conversely, that the opposite of what is absolutely necessary is internally impossible, or that the absolute necessity of things is the same as an internal necessity. For in certain cases that internal necessity is an entirely empty expression, with which we cannot connect the least concept, while that of the necessity of a thing in every respect (with regard to all that is possible) implies very peculiar determinations. As therefore the l
oss of a concept which has acted a great part in speculative philosophy can never be indifferent to philosophers, I hope they will also take some interest in the definition and careful preservation of the term with which that concept is connected.

  I shall therefore use the term absolute in this enlarged meaning only, in opposition to that which is valid relatively and in particular respects only, the latter being restricted to conditions, the former free from any restrictions whatsoever.

  It is then the absolute totality in the synthesis of conditions at which the transcendental concept of reason aims, nor does it rest satisfied till it has reached that which is unconditioned absolutely and in every respect. Pure reason leaves everything to the understanding, which has primarily to do with the objects of intuition, or rather their synthesis in imagination. It is only the absolute totality in the use of the concepts of the understanding, which reason reserves for itself, while trying to carry the synthetical unity, which is realised in the category, to the absolutely unconditioned. We might therefore call the latter the unity of the phenomena in reason, the former, which is expressed by the category, the unity in the understanding. Hence reason is only concerned with the use of the understanding, not so far as it contains the basis of possible experience (for the absolute totality of conditions is not a concept that can be used in experience, because no experience is unconditioned), but in order to impart to it a direction towards a certain unity of which the understanding knows nothing, and which is meant to comprehend all acts of the understanding, with regard to any object, into an absolute whole. On this account the objective use of the pure concepts of reason must always be transcendent: while that of the pure concepts of the understanding must always be immanent, being by its very nature restricted to possible experience.

  By idea I understand the necessary concept of reason, to which the senses can supply no corresponding object. The concepts of reason, therefore, of which we have been speaking, are transcendental ideas. They are concepts of pure reason, so far as they regard all empirical knowledge as determined by an absolute totality of conditions. They are not mere fancies, but supplied to us by the very nature of reason, and referring by necessity to the whole use of the understanding. They are, lastly, transcendent, as overstepping the limits of all experience which can never supply an object adequate to the transcendental idea. If we speak of an idea, we say a great deal with respect to the object (as the object of the pure understanding) but very little with respect to the subject, that is, with respect to its reality under empirical conditions, because an idea, being the concept of a maximum, can never be adequately given in concreto. As the latter is really the whole aim in the merely speculative use of reason, and as the mere approaching a concept, which in reality can never be reached, is the same as if the concept were missed altogether, people, when speaking of such a concept, are wont to say, it is an idea only. Thus one might say, that the absolute whole of all phenomena is an idea only, for as we can never form a representation of such a whole, it remains a problem without a solution. In the practical use of the understanding, on the contrary, where we are only concerned with practice, according to rules, the idea of practical reason can always be realised in concreto, although partially only; nay, it is the indispensable condition of all practical use of reason. The practical realisation of the idea is here always limited and deficient, but these limits cannot be defined, and it always remains under the influence of a concept, implying absolute completeness and perfection. The practical idea is therefore in this case truly fruitful, and, with regard to practical conduct, indispensable and necessary. In it pure reason becomes a cause and active power, capable of realising what is contained in its concept. Hence we cannot say of wisdom, as if contemptuously, that it is an idea only, but for the very reason that it contains the idea of the necessary unity of all possible aims, it must determine all practical acts, as an original and, at least, limitative condition.

  Although we must say that all transcendental concepts of reason are ideas only, they are not therefore to be considered as superfluous and useless. For although we cannot by them determine any object, they may nevertheless, even unobserved, supply the understanding with a canon or rule of its extended and consistent use, by which, though no object can be better known than it is according to its concepts, yet the understanding may be better guided onwards in its knowledge, not to mention that they may possibly render practicable a transition from physical to practical concepts, and thus impart to moral ideas a certain strength and connection with the speculative knowledge of reason. On all this more light will be thrown in the sequel.

  For our present purposes we are obliged to set aside a consideration of these practical ideas, and to treat of reason in its speculative, or rather, in a still more limited sense, its purely transcendental use. Here we must follow the same road which we took before in the deduction of the categories; that is, we must consider the logical form of all knowledge of reason, and see whether, perhaps, by this logical form, reason may become a source of concepts also, which enable us to regard objects in themselves, as determined synthetically a priori in relation to one or other of the functions of reason.

  Reason, if considered as a faculty of a certain logical form of knowledge, is the faculty of concluding, that is, of judging mediately, by bringing the condition of a possible under the condition of a given judgment. The given judgment is the general rule (major). Bringing the condition of another possible judgment under the condition of the rule, which may be called subsumption, is the minor, and the actual judgment, which contains the assertion of the rule in the subsumed case, is the conclusion. We know that the rule asserts something as general under a certain condition. The condition of the rule is then found to exist in a given case. Then that which, under that condition, was asserted as generally valid, has to be considered as valid in that given case also, which complies with that condition. It is easy to see therefore that reason arrives at knowledge by acts of the understanding, which constitute a series of conditions. If I arrive at the proposition that all bodies are changeable, only by starting from a more remote knowledge (which does not yet contain the concept of body, but a condition of such a concept only), namely, that all which is composite is changeable; and then proceed to something less remotely known, and depending on the former, namely, that bodies are composite; and, lastly, only advance to a third proposition, connecting the more remote knowledge (changeable) with the given case, and conclude that bodies therefore are changeable, we see that we have passed through a series of conditions (premisses) before we arrived at knowledge (conclusion). Every series, the exponent of which (whether of a categorical or hypothetical judgment) is given, can be continued, so that this procedure of reason leads to ratiocinatio polysyllogistica, a series of conclusions which, either on the side of the conditions (per prosyllogismos) or of the conditioned (per episyllogismos), may be continued indefinitely.

  It is soon perceived, however, that the chain or series of prosyllogisms, that is, of knowledge deduced on the side of reasons or conditions of a given knowledge, in other words, the ascending series of syllogisms, must stand in a very different relation to the faculty of reason from that of the descending series, that is, of the progress of reason on the side of the conditioned, by means of episyllogisms. For, as in the former case the knowledge embodied in the conclusion is given as conditioned only, it is impossible to arrive at it by means of reason in any other way except under the supposition at least that all the members of the series on the side of the conditions are given (totality in the series of premisses), because it is under that supposition only that the contemplated judgment a priori is possible; while on the side of the conditioned, or of the inferences, we can only think of a growing series, not of one presupposed as complete or given, that is, of a potential progression only. Hence, when our knowledge is considered as conditioned, reason is constrained to look upon the series of conditions in the ascending line as complete, and given in their totality. But if the s
ame knowledge is looked upon at the same time as a condition of other kinds of knowledge, which constitute among themselves a series of inferences in a descending line, it is indifferent to reason how far that progression may go a parte posteriori, or whether a totality of the series is possible at all, because such a series is not required for the conclusion in hand, which is sufficiently determined and secured on grounds a parte priori. Whether the series of premisses on the side of the conditions have a something that stands first as the highest condition, or whether it be without limits a parte priori, it must at all events contain a totality of conditions, even though we should never succeed in comprehending it; and the whole series must be unconditionally true, if the conditioned, which is considered as a consequence resulting from it, is to be accepted as true. This is a demand of reason which pronounces its knowledge as determined a priori and as necessary, either in itself, and in that case it requires no reasons, or, if derivative, as a member of a series of reasons, which itself is unconditionally true.