Immanuel Kant's Critique of Pure Reason, translated by Norman Kemp Smith





IT is humiliating to human reason that it achieves nothing in its pure employment, and indeed stands in need of a discipline to check its extravagances, and to guard it against the deceptions which arise therefrom. But, on the other hand, reason is reassured and gains self-confidence, on finding that it itself can and must apply this discipline, and that it is not called upon to submit to any outside censorship; and, moreover, that the limits which it is compelled to set to its speculative employment likewise limit the pseudo-rational pretensions of all its opponents, and that it can secure against all attacks whatever may remain over from its former exaggerated claims. The greatest and perhaps the sole use of all philosophy of pure reason is therefore only negative; since it serves not as an organon for the extension but as a discipline for the limitation of pure reason, and, instead of discovering truth, has only the modest merit of guarding against error.

There must, however, be some source of positive modes of knowledge which belong to the domain of pure reason, and which, it may be, give occasion to error solely owing to misunderstanding, while yet in actual fact they form the goal towards which reason is directing its efforts. How else can we account for our inextinguishable desire to find firm footing somewhere beyond the limits of experience? Reason has a presentiment of objects which possess a great interest for it. But when it follows the path of pure speculation, in order to approach them, they fly before it. Presumably it may look for better fortune in the only other path which still remains open to it, that of its practical employment. [630]

I understand by a canon the sum-total of the a priori principles of the correct employment of certain faculties of knowledge. Thus general logic, in its analytic portion, is a canon for understanding and reason in general; but only in regard to their form; it abstracts from all content. The transcendental analytic has similarly been shown to be the canon of the pure understanding; for understanding alone is capable of true synthetic modes of knowledge a priori. But when no correct employment of a faculty of knowledge is possible there is no canon. Now all synthetic knowledge through pure reason in its speculative employment is, as has been shown by the proofs given, completely impossible. There is therefore no canon of its speculative employment; such employment is entirely dialectical. All transcendental logic is, in this respect, simply a discipline. Consequently, if there be any correct employment of pure reason, in which case there must be a canon of its employment, the canon will deal not with the speculative but with the practical employment of reason. This practical employment of reason we shall now proceed to investigate.


Section 1


Reason is impelled by a tendency of its nature to go out beyond the field of its empirical employment, and to venture in a pure employment, by means of ideas alone, to the utmost limits of all knowledge, and not to be satisfied save through the completion of its course in [the apprehension of] a self-subsistent systematic whole. Is this endeavour the outcome merely of the speculative interests of reason? Must we not rather regard it as having its source exclusively in the practical interests of reason? I shall, for the moment, leave aside all question as to the success which attends pure reason in its speculative exercise, and enquire only as to the problems the solution of which [631] constitutes its ultimate aim, whether reached or not, and in respect of which all other aims are to be regarded only as means. These highest aims must, from the nature of reason, have a certain unity, in order that they may, as thus unified, further that interest of humanity which is subordinate to no higher interest.

The ultimate aim to which the speculation of reason in its transcendental employment is directed concerns three objects: the freedom of the will, the immortality of the soul, and the existence of God. In respect of all three the merely speculative interest of reason is very small; and for its sake alone we should hardly have undertaken the labor of transcendental investigation -- a labor so fatiguing in its endless wrestling with insuperable difficulties -- since whatever discoveries might be made in regard to these matters, we should not be able to make use of them in any helpful manner in concreto, that is, in the study or nature. If the will be free, this can have a bearing only on the intelligible cause of our volition. For as regards the phenomena of its outward expressions, that is, of our actions, we must account for them -- in accordance with a maxim which is inviolable, and which is so fundamental that without it we should not be able to employ reason in any empirical manner whatsoever -- in the same manner as all other appearances of nature, namely, in conformity with unchangeable laws. If, again, we should be able to obtain insight into the spiritual nature of the soul, and therewith of its immortality, we could make no use of such insight in explaining either the appearances of this present life or the specific nature of a future state. For our concept of an incorporeal nature is merely negative, and does not in the least extend our knowledge, yielding no sufficient material for inferences, save only such as are merely fictitious and cannot be sanctioned by philosophy. If, thirdly, the existence of a supreme intelligence be proved, by its means we might indeed render what is purposive in the constitution and ordering of the world comprehensible in a general sort or way, but we should not be in the least warranted in deriving from it any particular arrangement or disposition, or in boldly inferring any such, where it is not perceived. For it is a necessary rule of the speculative employment of reason, not to pass over natural causes, and, abandoning [632] that in regard to which we can be instructed by experience, to deduce something which we know from something which entirely transcends all our [possible] knowledge. In short, these three propositions are for speculative reason always transcendent, and allow of no immanent employment -- that is, employment in reference to objects of experience, and so in some manner really of service to us -- but are in themselves, notwithstanding the very heavy labors which they impose upon our reason, entirely useless.

If, then, these three cardinal propositions are not in any way necessary for knowledge, and are yet strongly recommended by our reason, their importance, properly regarded, must concern only the practical.

By 'the practical' I mean everything that is possible through freedom. When, however, the conditions of the exercise of our free will are empirical, reason can have no other than a regulative employment in regard to it, and can serve only to effect unity in its empirical laws. Thus, for instance, in the precepts of prudence, the whole business of reason consists in uniting all the ends which are prescribed to us by our desires in the one single end, happiness, and in coordinating the means for attaining it. In this field, therefore, reason can supply none but pragmatic laws of free action, for the attainment of those ends which are commended to us by the senses; it cannot yield us laws that are pure and determined completely a priori. Laws of this latter type, pure practical laws, whose end is given through reason completely a priori, and which are prescribed to us not in an empirically conditioned but in an absolute manner, would be products of pure reason. Such are the moral laws; and these alone, therefore, belong to the practical employment of reason, and allow of a canon.

The whole equipment of reason, in the discipline which may be entitled pure philosophy, is in fact determined with a view to the three above-mentioned problems. These, however, themselves in turn refer us yet further, namely, to the problem what we ought to do, if the will is free, if there is a God and a future world. As this concerns our attitude to the supreme end, it is evident that the ultimate intention of nature in her wise provision for us has indeed, in the [633] constitution of our reason, been directed to moral interests alone.

But we must be careful, in turning our attention to an object which is foreign{1} to transcendental philosophy, that we do not indulge in digressions to the detriment of the unity of the system, nor on the other hand, by saying too little on this new topic, fail in producing conviction through lack of clearness. I hope to avoid both dangers, by keeping as close as possible to the transcendental, and by leaving entirely aside any psychological, that is, empirical, factors that may perchance accompany it.

I must first remark that for the present I shall employ the concept of freedom in this practical sense only, leaving aside that other transcendental meaning which cannot be empirically made use of in explanation of appearances, but is itself a problem for reason, as has been already shown. A will is purely animal (arbitrium brutum), which cannot be determined save through sensuous impulses, that is, pathologically. A will which can be determined independently of sensuous impulses, and therefore through motives which are represented only by reason, is entitled free will (arbitrium liberum), and everything which is bound up with this will, whether as ground or as consequence, is entitled practical. [The fact of] practical freedom can be proved through experience. For the human will is not determined by that alone which stimulates, that is, immediately affects the senses; we have the power to overcome the impressions on our faculty of sensuous desire, by calling up representations of what, in a more indirect manner, is useful or injurious. But these considerations, as to what is desirable in respect of our whole state, that is, as to what is good and useful, are based on reason. Reason therefore provides [634] laws which are imperatives, that is, objective laws of freedom, which tell us what ought to happen -- although perhaps it never does happen -- therein differing from laws of nature, which relate only to that which happens. These laws are therefore to be entitled practical laws.

Whether reason is not, in the actions through which it prescribes laws, itself again determined by other influences, and whether that which, in relation to sensuous impulses, is entitled freedom, may not, in relation to higher and more remote operating causes, be nature again, is a question which in the practical field does not concern us, since we are demanding of reason nothing but the rule of conduct; it is a merely speculative question, which we can leave aside so long as we are considering what ought or ought not to be done. While we thus through experience know practical freedom to be one of the causes in nature, namely, to be a causality of reason in the determination of the will, transcendental freedom demands the independence of this reason -- in respect of its causality, in beginning a series of appearances -- from all determining causes of the sensible world. Transcendental freedom is thus, as it would seem, contrary to the law of nature, and therefore to all possible experience; and so remains a problem. But this problem does not come within the province of reason in its practical employment; and we have therefore in a canon of pure reason to deal with only two questions, which relate to the practical interest of pure reason, and in regard to which a canon of its employment must be possible -- Is there a God? and, Is there a future life? The question of transcendental freedom is a matter for speculative knowledge only, and when we are dealing with the practical we can leave it aside as being an issue with which we have no concern. Moreover, a quite sufficient discussion of it is to be found in the antinomy of pure reason. [635] (1) I am well aware that in the terminology of the Schools the title discipline is commonly used as synonymous with instruction. However, there are so many other cases where discipline in the sense of training by constraint is carefully distinguished from instruction in the sense of teaching, and the very nature of things itself makes it so imperative that we should preserve the only expressions suitable for this distinction, that it is desirable that the former term should never be used in any but the negative sense. (2) With the concept of cause I do really go beyond the empirical concept of an event (something happening), yet I do not pass to the intuition which exhibits the concept of cause in concreto, but to the time-conditions in general, which in experience may be found to be in accord with this concept. I therefore proceed merely with concepts; I cannot proceed by means of the construction of concepts, since the concept is a rule of the synthesis of perceptions, and the latter are not pure intuitions, and so do not permit of being given a priori. (3) Completeness means clearness and sufficiency of characteristics; by limits is meant the precision shown in there not being more of these characteristics than belong to the complete concept; by original is meant that this determination of these limits is not derived from anything else, and therefore does not require any proof; for if it did, that would disqualify the supposed explanation from standing at the head of all the judgments regarding its object. (4) Philosophy is full of faulty definitions, especially of definitions which, while indeed containing some of the elements required, are yet not complete. If we could make no use of a concept till we had defined it, all philosophy would be in a pitiable plight. But since a good and safe use can still be made of the elements obtained by analysis so far as they go, defective definitions, that is, propositions which are properly not definitions, but are yet true, and are therefore approximations to definitions, can be employed with great advantage. In mathematics definition belongs ad esse, in philosophy ad melius esse. It is desirable to attain an adequate definition, but often very difficult. The jurists are still without a definition of their concept of right. (5) All practical concepts relate to objects of satisfaction or dissatisfaction, that is, of pleasure and pain, and therefore, at least indirectly, to the objects of our feelings. But as feeling is not a faculty whereby we represent things, but lies outside our whole faculty of knowledge, the elements of our judgments so far as they relate to pleasure or pain, that is, the elements of practical judgments, do not belong to transcendental philosophy, which is exclusively concerned with pure a priori modes of knowledge.


Section 2


Reason, in its speculative employment, conducted us through the field of experience, and since it could not find complete satisfaction there, from thence to speculative ideas, which, however, in the end brought us back to experience. In so doing the ideas fulfilled their purpose, but in a manner which, though useful, is not in accordance with our expectation. One other line of enquiry still remains open to us: namely, whether pure reason may not also be met with in the practical sphere, and whether it may not there conduct us to ideas which reach to those highest ends of pure reason that we have just stated, and whether, therefore, reason may not be able to supply to us from the standpoint of its practical interest what it altogether refuses to supply in respect of its speculative interest.

All the interests of my reason, speculative as well as practical, combine in the three following questions:
1. What can I know? 2. What ought I to do? 3. What may I hope?

The first question is merely speculative. We have, as I flatter myself, exhausted all the possible answers to it, and at last have found the answer with which reason must perforce content itself, and with which, so long as it takes no account of the practical, it has also good cause to be satisfied. But from the two great ends to which the whole endeavour of pure reason was really directed, we have remained just as far removed as if through love of ease we had declined this labor of enquiry at the very outset. So far, then, as knowledge is concerned, this much, at least, is certain and definitively established, that in respect of these two latter problems, knowledge is unattainable by us.

The second question is purely practical. As such it can [636] indeed come within the scope of pure reason, but even so is not transcendental but moral, and cannot, therefore, in and by itself, form a proper subject for treatment in this Critique.

The third question -- If I do what I ought to do, what may I then hope? -- is at once practical and theoretical, in such fashion that the practical serves only as a clue that leads us to the answer to the theoretical question, and when this is followed out, to the speculative question. For all hoping is directed to happiness, and stands in the same relation to the practical and the law of morality as knowing and the law of nature to the theoretical knowledge of things. The former arrives finally at the conclusion that something is (which determines the ultimate possible end) because something ought to happen; the latter, that something is (which operates as the supreme cause) because something happens.

Happiness is the satisfaction of all our desires, extensively, in respect of their manifoldness, intensively, in respect of their degree, and protensively, in respect of their duration. The practical law, derived from the motive of happiness, I term pragmatic (rule of prudence), and that law, if there is such a law, which has no other motive than worthiness of being happy, I term moral (law of morality). The former advises us what we have to do if we wish to achieve happiness; the latter dictates to us how we must behave in order to deserve happiness. The former is based on empirical principles; for only by means of experience can I know what desires there are which call for satisfaction; or what those natural causes are which are capable of satisfying them. The latter takes no account of desires, and the natural means of satisfying them, and considers only the freedom of a rational being in general, and the necessary conditions under which alone this freedom can harmonize with a distribution of happiness that is made in accordance with principles. This latter law can therefore be based on mere ideas of pure reason, and known a priori.

I assume that there really are pure moral laws which determine completely a priori (without regard to empirical motives, that is, to happiness) what is and is not to be done, that is, which determine the employment of the freedom of a rational being in general; and that these laws command in an absolute manner (not merely hypothetically, on the supposition [637] of other empirical ends), and are therefore in every respect necessary. I am justified in making this assumption, in that I can appeal not only to the proofs employed by the most enlightened moralists, but to the moral judgment of every man, in so far as he makes the effort to think such a law clearly.

Pure reason, then, contains, not indeed in its speculative employment, but in that practical employment which is also moral, principles of the possibility of experience, namely, of such actions as, in accordance with moral precepts, might be met with in the history of mankind. For since reason commands that such actions should take place, it must be possible for them to take place. Consequently, a special kind of systematic unity, namely the moral, must likewise be possible. We have indeed found that the systematic unity of nature cannot be proved in accordance with speculative principles of reason. For although reason does indeed have causality in respect of freedom in general, it does not have causality in respect of nature as a whole; and although moral principles of reason can indeed give rise to free actions, they cannot give rise to laws of nature. Accordingly it is in their practical, meaning thereby their moral, employment, that the principles of pure reason have objective reality.

I entitle the world a moral world, in so far as it may be in accordance with all moral laws; and this is what by means of the freedom of the rational being it can be, and what according to the necessary laws of morality it ought to be. Owing to our here leaving out of account all conditions (ends) and even all the special difficulties to which morality is exposed (weakness or depravity of human nature), this world is so far thought as an intelligible world only. To this extent, therefore, it is a mere idea, though at the same time a practical idea, which really can have, as it also ought to have, an influence upon the sensible world, to bring that world, so far as may be possible, into conformity with the idea. The idea of a moral world has, therefore, objective reality, not as referring to an object of an intelligible intuition (we are quite unable to think any such object), but as referring to the sensible world, viewed, however, as being an object of pure reason in its practical employment, that is, as a corpus mysticum of the rational beings in it, so far as the free will of each being is, under moral laws, in [638] complete systematic unity with itself and with the freedom of every other.

This is the answer to the first of the two questions of pure reason that concern its practical interest: -- Do that through which thou becomest worthy to be happy. The second question is: -- If I so behave as not to be unworthy of happiness, may I hope thereby to obtain happiness? In answering this question we have to consider whether the principles of pure reason, which prescribe the law a priori, likewise connect this hope necessarily with it.

I maintain that just as the moral principles are necessary according to reason in its practical employment, it is in the view of reason, in the field of its theoretical employment, no less necessary to assume that everyone has ground to hope for happiness in the measure in which he has rendered himself by his conduct worthy of it, and that the system of morality is therefore inseparably -- though only in the idea of pure reason -- bound up with that of happiness.

Now in an intelligible world, that is, in the moral world, in the concept of which we leave out of account all the hindrances to morality (the desires), such a system, in which happiness is bound up with and proportioned to morality, can be conceived as necessary, inasmuch as freedom, partly inspired and partly restricted by moral laws, would itself be the cause of general happiness, since rational beings, under the guidance of such principles, would themselves be the authors both of their own enduring well-being and of that of others. But such a system of self-rewarding morality is only an idea, the carrying out of which rests on the condition that everyone does what he ought, that is, that all the actions of rational beings take place just as if they had proceeded from a supreme will that comprehends in itself, or under itself, all private wills. But since the moral law remains binding for every one in the use of his freedom, even although others do not act in conformity with the law, neither the nature of the things of the world nor the causality of the actions themselves and their relation to morality determine how the consequences of these actions will be related to happiness. The alleged necessary connection of the hope of happiness with the necessary endeavour to render the self worthy of happiness cannot there [639] fore be known through reason. It can be counted upon only if a Supreme Reason, that governs according to moral rules, be likewise posited as underlying nature as its cause.

The idea of such an intelligence in which the most perfect moral will, united with supreme blessedness, is the cause of all happiness in the world -- so far as happiness stands in exact relation with morality, that is, with worthiness to be happy -- I entitle the ideal of the supreme good. It is, therefore, only in the ideal of the supreme original good that pure reason can find the ground of this connection, which is necessary from the practical point of view, between the two elements of the supreme derivative good -- the ground, namely, of an intelligible, that is, moral world. Now since we are necessarily constrained by reason to represent ourselves as belonging to such a world, while the senses present to us nothing but a world of appearances, we must assume that moral world to be a consequence of our conduct in the world of sense (in which no such connection between worthiness and happiness is exhibited), and therefore to be for us a future world. Thus God and a future life are two postulates which, according to the principles of pure reason, are inseparable from the obligation which that same reason imposes upon us.

Morality, by itself, constitutes a system. Happiness, however, does not do so, save in so far as it is distributed in exact proportion to morality. But this is possible only in the intelligible world, under a wise Author and Ruler. Such a Ruler, together with life in such a world, which we must regard as a future world, reason finds itself constrained to assume; otherwise it would have to regard the moral laws as empty figments of the brain, since without this postulate the necessary consequence which it itself connects with these laws could not follow. Hence also everyone regards the moral laws as commands; and this the moral laws could not be if they did not connect a priori suitable consequences with their rules, and thus carry with them promises and threats. But this again they could not do, if they did not reside in a necessary being, as the supreme good, which alone can make such a purposive unit possible.

Leibniz entitled the world, in so far as we take account only of the rational beings in it, and of their connection according [640] to moral laws under the government of the supreme good, the kingdom of grace, distinguishing it from the kingdom of nature, in which these rational beings do indeed stand under moral laws, but expect no other consequences from their actions than such as follow in accordance with the course of nature in our world of sense. To view ourselves, therefore, as in the world of grace, where all happiness awaits us, except in so far as we ourselves limit our share in it through being unworthy of happiness, is, from the practical standpoint, a necessary idea of reason.

Practical laws, in so far as they are subjective grounds of actions, that is, subjective principles, are entitled maxims. The estimation of morality, in regard to its purity and consequences, is effected in accordance with ideas, the observance of its laws in accordance with maxims.

It is necessary that the whole course of our life be subject to moral maxims; but it is impossible that this should happen unless reason connects with the moral law, which is a mere idea, an operative cause which determines for such conduct as is in accordance with the moral law an outcome, either in this or in another life, that is in exact conformity with our supreme ends. Thus without a God and without a world invisible to us now but hoped for, the glorious ideas of morality are indeed objects of approval and admiration, but not springs of purpose and action. For they do not fulfil in its completeness that end which is natural to every rational being and which is determined a priori, and rendered necessary, by that same pure reason.

Happiness, taken by itself, is, for our reason, far from being the complete good. Reason does not approve happiness (however inclination may desire it) except in so far as it is united with worthiness to be happy, that is, with moral conduct. Morality, taken by itself, and with it, the mere worthiness to be happy, is also far from being the complete good. To make the good complete, he who behaves in such a manner as not to be unworthy of happiness must be able to hope that he will participate in happiness. Even the reason that is free from all private purposes, should it put itself in the place of a being that had to distribute all happiness to others, cannot judge otherwise; for in the practical idea both elements are essentially [641] connected, though in such a manner that it is the moral disposition which conditions and makes possible the participation in happiness, and not conversely the prospect of happiness that makes possible the moral disposition. For in the latter case the disposition would not be moral, and therefore would not be worthy of complete happiness -- happiness which in the view of reason allows of no limitation save that which arises from our own immoral conduct.

Happiness, therefore, in exact proportion with the morality of the rational beings who are thereby rendered worthy of it, alone constitutes the supreme good of that world wherein, in accordance with the commands of a pure but practical reason, we are under obligation to place ourselves. This world is indeed an intelligible world only, since the sensible world holds out no promise that any such systematic unity of ends can arise from the nature of things. Nor is the reality of this unity based on anything else than the postulate of a supreme original good. In a supreme good, thus conceived, self-subsistent reason, equipped with all the sufficiency of a supreme cause, establishes, maintains, and completes the universal order of things, according to the most perfect design -- an order which in the world of sense is in large part concealed from us.

This moral theology has the peculiar advantage over speculative theology that it inevitably leads to the concept of a sole, all- perfect, and rational primordial being, to which speculative theology does not, on objective grounds, even so much as point the way, and as to the existence of which it is still less capable of yielding any conviction. For neither in transcendental nor in natural theology, however far reason may carry us, do we find any considerable ground for assuming only some one single being which we should be justified in placing prior to all natural causes, and upon which we might make them in all respects dependent. On the other hand, if we consider from the point of view of moral unity, as a necessary law of the world, what the cause must be that can alone give to this law its appropriate effect, and so for us obligatory force, we conclude that there must be one sole supreme will, which comprehends all these laws in itself. For how, under different wills, should we find complete [642] unity of ends. This Divine Being must be omnipotent, in order that the whole of nature and its relation to morality in the world may be subject to his will; omniscient, that He may know our innermost sentiments and their moral worth; omnipresent, that He may be immediately at hand for the satisfying of every need which the highest good demands; eternal, that this harmony of nature and freedom may never fail, etc.

But this systematic unity of ends in this world of intelligences - - a world which is indeed, as mere nature, a sensible world only, but which, as a system of freedom, can be entitled an intelligible, that is, a moral world (regnum gratiae) -- leads inevitably also to the purposive unity of all things, which constitute this great whole, in accordance with universal laws of nature (just as the former unity is in accordance with universal and necessary laws of morality), and thus unites the practical with the speculative reason. The world must be represented as having originated from an idea if it is to be in harmony with that employment of reason without which we should indeed hold ourselves to be unworthy of reason, namely, with the moral employment -- which is founded entirely on the idea of the supreme good. In this way all investigation of nature tends to take the form of a system of ends, and in its widest extension becomes a physico-theology. But this, as it has its source in the moral order, as a unity grounded in freedom's own essential nature, and not accidentally instituted through external commands, connects the purposiveness of nature with grounds which must be inseparably connected a priori with the inner possibility of things, and so leads to a transcendental theology -- a theology which takes the ideal of supreme ontological perfection as a principle of systematic unity. And since all things have their origin in the absolute necessity of the one primordial being, that principle connects them in accordance with universal and necessary laws of nature.

What use can we make of our understanding, even in respect of experience, if we do not propose ends to ourselves? But the highest ends are those of morality, and these we can know only as they are given us by pure reason. But though provided with these, and employing them as a clue, we cannot make use of the knowledge of nature in any serviceable manner [643] in the building up of knowledge, unless nature has itself shown unity of design. For without this unity we should ourselves have no reason, inasmuch as there would be no school for reason, and no fertilization through objects such as might afford materials for the necessary concepts. But the former purposive unity is necessary, and founded on the will's own essential nature, and this latter unity [of design in nature] which contains the condition of its application in concreto, must be so likewise. And thus the transcendental enlargement of our knowledge, as secured through reason, is not to be regarded as the cause, but merely as the effect of the practical purposiveness which pure reason imposes upon us.

Accordingly we find, in the history of human reason, that until the moral concepts were sufficiently purified and determined, and until the systematic unity of their ends was understood in accordance with these concepts and from necessary principles, the knowledge of nature, and even a quite considerable development of reason in many other sciences, could give rise only to crude and incoherent concepts of the Deity, or as sometimes happened resulted in an astonishing indifference in regard to all such matters. A greater preoccupation with moral ideas, which was rendered necessary by the extraordinarily pure moral law of our religion, made reason more acutely aware of its object, through the interest which it was compelled to take in it. And this came about, independently of any influence exercised by more extended views of nature or by correct and reliable transcendental insight (for that has always been lacking). It was the moral ideas that gave rise to that concept of the Divine Being which we now hold to be correct - - and we so regard it not because speculative reason convinces us of its correctness, but because it completely harmonizes with the moral principles of reason. Thus it is always only to pure reason, though only in its practical employment, that we must finally ascribe the merit of having connected with our highest interest a knowledge which reason can think only, and cannot establish, and of having thereby shown it to be, not indeed a demonstrated dogma, but a postulate which is absolutely necessary in view of what are reason's own most essential ends. [644]

But when practical reason has reached this goal, namely, the concept of a sole primordial being as the supreme good, it must not presume to think that it has raised itself above all empirical conditions of its application, and has attained to an immediate knowledge of new objects, and can therefore start from this concept, and can deduce from it the moral laws themselves. For it is these very laws that have led us, in virtue of their inner practical necessity, to the postulate of a self- sufficient cause, or of a wise Ruler of the world, in order that through such agency effect may be given to them. We may not, therefore, in reversal of such procedure, regard them as accidental and as derived from the mere will of the Ruler, especially as we have no conception of such a will, except as formed in accordance with these laws. So far, then, as practical reason has the right to serve as our guide, we shall not look upon actions as obligatory because they are the commands of God, but shall regard them as divine commands because we have an inward obligation to them. We shall study freedom according to the purposive unity that is determined in accordance with the principles of reason, and shall believe ourselves to be acting in conformity with the divine will in so far only as we hold sacred the moral law which reason teaches us from the nature of the actions themselves; and we shall believe that we can serve that will only by furthering what is best in the world, alike in ourselves and in others. Moral theology is thus of immanent use only. It enables us to fulfil our vocation in this present world by showing us how to adapt ourselves to the system of all ends, and by warning us against the fanaticism, and indeed the impiety, of abandoning the guidance of a morally legislative reason in the right conduct of our lives, in order to derive guidance directly from the idea of the Supreme Being. For we should then be making a transcendent employment of moral theology; and that, like a transcendent use of pure speculation, must pervert and frustrate the ultimate ends of reason.


Section 3


The holding of a thing to be true is an occurrence in our understanding which, though it may rest on objective grounds, also requires subjective causes in the mind of the individual who makes the judgment. If the judgment is valid for everyone, provided only he is in possession of reason, its ground is objectively sufficient, and the holding of it to be true is entitled conviction. If it has its ground only in the special character of the subject, it is entitled persuasion.

Persuasion is a mere illusion, because the ground of the judgment, which lies solely in the subject, is regarded as objective. Such a judgment has only private validity, and the holding of it to be true does not allow of being communicated. But truth depends upon agreement with the object, and in respect of it the judgments of each and every understanding must therefore be in agreement with each other (consentientia uni tertio, consentiunt inter se). The touchstone whereby we decide whether our holding a thing to be true is conviction or mere persuasion is therefore external, namely, the possibility of communicating it and of finding it to be valid for all human reason. For there is then at least a presumption that the ground of the agreement of all judgments with each other, notwithstanding the differing characters of individuals, rests upon the common ground, namely, upon the object, and that it is for this reason that they are all in agreement with the object -- the truth of the judgment being thereby proved.

So long, therefore, as the subject views the judgment merely as an appearance of his mind, persuasion cannot be subjectively distinguished from conviction. The experiment, however, whereby we test upon the understanding of others whether those grounds of the judgment which are valid for us have the same effect on the reason of others as on our own, is a means, although only a subjective means, not indeed of producing conviction, but of detecting any merely private validity [646] in the judgment, that is, anything in it which is mere persuasion.

If, in addition, we can specify the subjective causes of the judgment, which we have taken as being its objective grounds, and can thus explain the deceptive judgment as an event in our mind, and can do so without having to take account of the character of the object, we expose the illusion and are no longer deceived by it, although always still in some degree liable to come under its influence, in so far as the subjective cause of the illusion is inherent in our nature.

I cannot assert anything, that is, declare it to be a judgment necessarily valid for everyone, save as it gives rise to conviction. Persuasion I can hold to on my own account, if it so pleases me, but I cannot, and ought not, to profess to impose it as binding on anyone but myself.

The holding of a thing to be true, or the subjective validity of the judgment, in its relation to conviction (which is at the same time objectively valid), has the following three degrees: opining, believing, and knowing. Opining is such holding of a judgment as is consciously insufficient, not only objectively, but also subjectively. If our holding of the judgment be only subjectively sufficient, and is at the same time taken as being objectively insufficient, we have what is termed believing. Lastly, when the holding of a thing to be true is sufficient both subjectively and objectively, it is knowledge. The subjective sufficiency is termed conviction (for myself), the objective sufficiency is termed certainty (for everyone). There is no call for me to spend further time on the explanation of such easily understood terms.

I must never presume to opine, without knowing at least something by means of which the judgment, in itself merely problematic, secures connection with truth, a connection which, although not complete, is yet more than arbitrary fiction. Moreover, the law of such a connection must be certain. For if, in respect of this law also, I have nothing but opinion, it is all merely a play of the imagination, without the least relation to truth. Again, opining is not in any way permissible in judging by means of pure reason. For since such judging is not based on grounds of experience, but being in [647] every case necessary has all to be arrived at a priori, the principle of the connection requires universality and necessity, and therefore complete certainty; otherwise we should have no guidance as to truth. Hence it is absurd to have an opinion in pure mathematics; either we must know, or we must abstain from all acts of judgment. It is so likewise in the case of the principles of morality, since we must not venture upon an action on the mere opinion that it is allowed, but must know it to be so.

In the transcendental employment of reason, on the other hand, while opining is doubtless too weak a term to be applicable, the term knowing is too strong. In the merely speculative sphere we cannot therefore make any judgments whatsoever. For the subjective grounds upon which we may hold something to be true, such as those which are able to produce belief, are not permissible in speculative questions, inasmuch as they do not hold independently of all empirical support, and do not allow of being communicated in equal measure to others.

But it is only from a practical point of view that the theoretically insufficient holding of a thing to be true can be termed believing. This practical point of view is either in reference to skill or in reference to morality, the former being concerned with optional and contingent ends, the latter with ends that are absolutely necessary.

Once an end is accepted, the conditions of its attainment are hypothetically necessary. This necessity is subjectively, but still only comparatively, sufficient, if I know of no other conditions under which the end can be attained. On the other hand, it is sufficient, absolutely and for everyone, if I know with certainty that no one can have knowledge of any other conditions which lead to the proposed end. In the former case my assumption and the holding of certain conditions to be true is a merely contingent belief; in the latter case it is a necessary belief. The physician must do something for a patient in danger, but does not know the nature of his illness. He observes the symptoms, and if he can find no more likely alternative, judges it to be a case of phthisis. Now even in his own estimation his belief is contingent only; another observer [648] might perhaps come to a sounder conclusion. Such contingent belief, which yet forms the ground for the actual employment of means to certain actions, I entitle pragmatic belief.

The usual touchstone, whether that which someone asserts is merely his persuasion -- or at least his subjective conviction, that is, his firm belief -- is betting. It often happens that someone propounds his views with such positive and uncompromising assurance that he seems to have entirely set aside all thought of possible error. A bet disconcerts him. Sometimes it turns out that he has a conviction which can be estimated at a value of one ducat, but not of ten. For he is very willing to venture one ducat, but when it is a question of ten he becomes aware, as he had not previously been, that it may very well be that he is in error. If, in a given case, we represent ourselves as staking the happiness of our whole life, the triumphant tone of our judgment is greatly abated; we become extremely diffident, and discover for the first time that our belief does not reach so far. Thus pragmatic belief always exists in some specific degree, which, according to differences in the interests at stake, may be large or may be small.

But in many cases, when we are dealing with an object about which nothing can be done by us, and in regard to which our judgment is therefore purely theoretical, we can conceive and picture to ourselves an attitude for which we regard ourselves as having sufficient grounds, while yet there is no existing means of arriving at certainty in the matter. Thus even in purely theoretical judgments there is an analogon of practical judgments, to the mental entertaining of which the term 'belief' is appropriate, and which we may entitle doctrinal belief. I should be ready to stake my all on the contention -- were it possible by means of any experience to settle the question -- that at least one of the planets which we see is inhabited. Hence I say that it is not merely opinion, but a strong belief, on the correctness of which I should be prepared to run great risks, that other worlds are inhabited.

Now we must admit that the doctrine of the existence of God belongs to doctrinal belief. For as regards theoretical knowledge of the world, I can cite nothing which necessarily presupposes this thought as the condition of my explanations [649] of the appearances exhibited by the world, but rather am bound so to employ my reason as if everything were mere nature. Purposive unity is, however, so important a condition of the application of reason to nature that I cannot ignore it, especially as experience supplies me so richly with examples of it. But I know no other condition under which this unity can supply me with guidance in the investigation of nature, save only the postulate that a supreme intelligence has ordered all things in accordance with the wisest ends. Consequently, as a condition of what is indeed a contingent, but still not unimportant purpose, namely, to have guidance in the investigation of nature, we must postulate a wise Author of the world. Moreover, the outcome of my attempts [in explanation of nature] so frequently confirms the usefulness of this postulate, while nothing decisive can be cited against it, that I am saying much too little if I proceed to declare that I hold it merely as an opinion. Even in this theoretical relation it can be said that I firmly believe in God. This belief is not, therefore, strictly speaking, practical; it must be entitled a doctrinal belief to which the theology of nature (physico- theology) must always necessarily give rise. In view of the magnificent equipment of our human nature, and the shortness of life so ill-suited to the full exercise of our powers, we can find in this same divine wisdom a no less sufficient ground for a doctrinal belief in the future life of the human soul.

In such cases the expression of belief is, from the objective point of view, an expression of modesty, and yet at the same time, from the subjective point of view, an expression of the firmness of our confidence. Were I even to go the length of describing the merely theoretical holding of the belief as an hypothesis which I am justified in assuming, I should thereby be pledging myself to have a more adequate concept of the character of a cause of the world and of the character of another world than I am really in a position to supply. For if I assume anything, even merely as an hypothesis, I must at least know so much of its properties that I require to assume, not its concept, but only its existence. The term 'belief' refers only to the guidance which an idea gives me, and to its subjective influence in that furthering of the activities of my reason which confirms me in the idea, and which [650] yet does so without my being in a position to give a speculative account of it.

But the merely doctrinal belief is somewhat lacking in stability; we often lose hold of it, owing to the speculative difficulties which we encounter, although in the end we always inevitably return to it.

It is quite otherwise with moral belief. For here it is absolutely necessary that something must happen, namely, that I must in all points conform to the moral law. The end is here irrefragably established, and according to such insight as I can have, there is only one possible condition under which this end can connect with all other ends, and thereby have practical validity, namely, that there be a God and a future world. I also know with complete certainty that no one can be acquainted with any other conditions which lead to the same unity of ends under the moral law. Since, therefore, the moral precept is at the same time my maxim (reason prescribing that it should be so), I inevitably believe in the existence of God and in a future life, and I am certain that nothing can shake this belief, since my moral principles would thereby be themselves overthrown, and I cannot disclaim them without becoming abhorrent in my own eyes.

Thus even after reason has failed in all its ambitious attempts to pass beyond the limits of all experience, there is still enough left to satisfy us, so far as our practical standpoint is concerned. No one, indeed, will be able to boast that he knows that there is a God, and a future life; if he knows this, he is the very man for whom I have long [and vainly] sought. All knowledge, if it concerns an object of mere reason, can be communicated; and I might therefore hope that under his instruction my own knowledge would be extended in this wonderful fashion. No, my conviction is not logical, but moral certainty; and since it rests on subjective grounds (of the moral sentiment), I must not even say, 'It is morally certain that there is a God, etc. ', but 'I am morally certain, etc. ' In other words, belief in a God and in another world is so interwoven with my moral sentiment that as there is little danger of my losing the latter, there is equally little cause for fear that the former can ever be taken from me. [651]

The only point that may seem questionable is the basing of this rational belief on the assumption of moral sentiments. If we leave these aside, and take a man who is completely indifferent with regard to moral laws, the question propounded by reason then becomes merely a problem for speculation, and can, indeed, be supported by strong grounds of analogy, but not by such as must compel the most stubborn skepticism to give way.{2} But in these questions no man is free from all interest. For although, through lack of good sentiments, he may be cut off from moral interest, still even in this case enough remains to make him fear the existence of a God and a future life. Nothing more is required for this than that he at least cannot pretend that there is any certainty that there is no such being and no such life. Since that would have to be proved by mere reason, and therefore apodeictically, he would have to prove the impossibility of both, which assuredly no one can reasonably undertake to do. This may therefore serve as negative belief, which may not, indeed, give rise to morality and good sentiments, but may still give rise to an analogon of these, namely, a powerful check upon the outbreak of evil sentiments.

But, it will be said, is this all that pure reason achieves in opening up prospects beyond the limits of experience? Nothing more than two articles of belief? Surely the common understanding could have achieved as much, without appealing to philosophers for counsel in the matter.

I shall not here dwell upon the service which philosophy has done to human reason through the laborious efforts of its criticism, granting even that in the end it should turn out to be merely negative; something more will be said on this point in the next section. [651] But I may at once reply: Do you really require that a mode of knowledge which concerns all men [652] should transcend the common understanding, and should only be revealed to you by philosophers? Precisely what you find fault with is the best confirmation of the correctness of the above assertions. For we have thereby revealed to us, what could not at the start have been foreseen, namely, that in matters which concern all men without distinction nature is not guilty of any partial distribution of her gifts, and that in regard to the essential ends of human nature the highest philosophy cannot advance further than is possible under the guidance which nature has bestowed even upon the most ordinary understanding.




BY an architectonic I understand the art of constructing systems. As systematic unity is what first raises ordinary knowledge to the rank of science, that is, makes a system out of a mere aggregate of knowledge, architectonic is the doctrine of the scientific in our knowledge, and therefore necessarily forms part of the doctrine of method. In accordance with reason's legislative prescriptions, our diverse modes of knowledge must not be permitted to be a mere rhapsody, but must form a system. Only so can they further the essential ends of reason. By a system I understand the unity of the manifold modes of knowledge under one idea.

This idea is the concept provided by reason -- of the form of a whole -- in so far as the concept determines a priori not only the scope of its manifold content, but also the positions which the parts occupy relatively to one another. The scientific concept of reason contains, therefore, the end and the form of that whole which is congruent with this requirement. The unity of the end to which all the parts relate and in the idea of which they all stand in relation to one another, makes it possible for us to determine from our knowledge of the other parts whether any part be missing, and to prevent any arbitrary addition, or in respect of its completeness any indeterminateness that does not conform to the limits which are thus determined a priori. The whole is thus an organized unity (articulatio), and not an aggregate (coacervatio). It may grow from within (per intussusceptionem), but not by external addition (per appositionem). It is thus like an animal body, the growth of which is not by [654] the addition of a new member, but by the rendering of each member, without change of proportion, stronger and more effective for its purposes.

The idea requires for its realization a schema, that is, a constituent manifold and an order of its parts, both of which must be determined a priori from the principle defined by its end. The schema, which is not devised in accordance with an idea, that is, in terms of the ultimate aim of reason, but empirically in accordance with purposes that are contingently occasioned (the number of which cannot be foreseen) yields technical unity; whereas the schema which originates from an idea (in which reason propounds the ends a priori, and does not wait for them to be empirically given) serves as the basis of architectonic unity. Now that which we call science, the schema of which must contain the outline (monogramma) and the division of the whole into parts, in conformity with the idea, that is, a priori, and in so doing must distinguish it with certainty and according to principles from all other wholes, is not formed in technical fashion, in view of the similarity of its manifold constituents or of the contingent use of our knowledge in concreto for all sorts of optional external ends, but in architectonic fashion, in view of the affinity of its parts and of their derivation from a single supreme and inner end, through which the whole is first made possible.

No one attempts to establish a science unless he has an idea upon which to base it. But in the working out of the science the schema, nay even the definition which, at the start, he first gave of the science, is very seldom adequate to his idea. For this idea lies hidden in reason, like a germ in which the parts are still undeveloped and barely recognizable even under microscopic observation. Consequently, since sciences are devised from the point of view of a certain universal interest, we must not explain and determine them according to the description which their founder gives of them, but in conformity with the idea which, out of the natural unity of the parts that we have assembled, we find to be grounded in reason itself. For we shall then find that its founder, and often even his latest successors, are groping for an idea which they have never succeeded in making clear to themselves, and that [655] consequently they have not been in a position to determine the proper content, the articulation (systematic unity), and limits of the science.

It is unfortunate that only after we have spent much time in the collection of materials in somewhat random fashion at the suggestion of an idea lying hidden in our minds, and after we have, indeed, over a long period assembled the materials in a merely technical manner, does it first become possible for us to discern the idea in a clearer light, and to devise a whole architectonically in accordance with the ends of reason. Systems seem to be formed in the manner of lowly organisms through a generatio aequivoca from the mere confluence of assembled concepts, at first imperfect, and only gradually attaining to completeness, although they one and all have had their schema, as the original germ, in the sheer self-development of reason. Hence, not only is each system articulated in accordance with an idea, but they are one and all organically united in a system of human knowledge, as members of one whole, and so as admitting of an architectonic of all human knowledge, which, at the present time, in view of the great amount of material that has been collected, or which can be obtained from the ruins of ancient systems, is not only possible, but would not indeed be difficult. We shall content ourselves here with the completion of our task, namely, merely to outline the architectonic of all knowledge arising from pure reason; and in doing so we shall begin from the point at which the common root of our faculty of knowledge divides and throws out two stems, one of which is reason. By reason I here understand the whole higher faculty of knowledge, and am therefore contrasting the rational with the empirical.

If I abstract from all the content of knowledge, objectively regarded, then all knowledge, subjectively regarded, is either historical or rational. Historical knowledge is cognitio ex datis; rational knowledge is cognitio ex principiis. However a mode of knowledge may originally be given, it is still, in relation to the individual who possesses it, simply historical, if he knows only so much of it as has been given to him from outside (and this in the form in which it has been given to him), whether through immediate experience or narration, or (as in the case [656] of general knowledge) through instruction. Anyone, therefore, who has learnt (in the strict sense of that term) a system of philosophy, such as that of Wolff, although he may have all its principles, explanations, and proofs, together with the formal divisions of the whole body of doctrine, in his head, and, so to speak, at his fingers' ends, has no more than a complete historical knowledge of the Wolffian philosophy. He knows and judges only what has been given him. If we dispute a definition, he does not know whence to obtain another. He has formed his mind on another's, and the imitative faculty is not itself productive. In other words, his knowledge has not in him arisen out of reason, and although, objectively considered, it is indeed knowledge due to reason, it is yet, in its subjective character, merely historical. He has grasped and kept; that is, he has learnt well, and is merely a plaster-cast of a living man. Modes of rational knowledge which are rational objectively (that is, which can have their first origin solely in human reason) can be so entitled subjectively also, only when they have been derived from universal sources of reason, that is, from principles --the sources from which there can also arise criticism, nay, even the rejection of what has been learnt.

All knowledge arising out of reason is derived either from concepts or from the construction of concepts. The former is called philosophical, the latter mathematical. I have already treated of the fundamental difference between these two modes of knowledge in the first chapter [of this Transcendental Doctrine of Method]. Knowledge [as we have just noted] can be objectively philosophical, and yet subjectively historical, as is the case with most novices, and with all those who have never looked beyond their School, and who remain novices all their lives. But it is noteworthy that mathematical knowledge, in its subjective character, and precisely as it has been learned, can also be regarded as knowledge arising out of reason, and that there is therefore in regard to mathematical knowledge no such distinction as we have drawn in the case of philosophical knowledge. This is due to the fact that the sources of knowledge, from which alone the teacher can derive his knowledge, lie nowhere but in the essential and genuine principles of reason, and consequently cannot be acquired by the novice from any other [657] source, and cannot be disputed; and this, in turn, is owing to the fact that the employment of reason is here in concreto only, although likewise a priori, namely, in intuition which is pure, and which precisely on that account is infallible, excluding all illusion and error. Mathematics, therefore, alone of all the sciences (a priori) arising from reason, can be learned; philosophy can never be learned, save only in historical fashion; as regards what concerns reason, we can at most learn to philosophize.

Philosophy is the system of all philosophical knowledge. If we are to understand by it the archetype for the estimation of all attempts at philosophizing, and if this archetype is to serve for the estimation of each subjective philosophy, the structure of which is often so diverse and liable to alteration, it must be taken objectively. Thus regarded, philosophy is a mere idea of a possible science which nowhere exists in concreto, but to which, by many different paths, we endeavour to approximate, until the one true path, overgrown by the products of sensibility, has at last been discovered, and the image, hitherto so abortive, has achieved likeness to the archetype, so far as this is granted to [mortal] man. Till then we cannot learn philosophy; for where is it, who is in possession of it, and how shall we recognize it? We can only learn to philosophize, that is, to exercise the talent of reason, in accordance with its universal principles, on certain actually existing attempts at philosophy, always, however, reserving the right of reason to investigate, to confirm, or to reject these principles in their very sources.

Hitherto the concept of philosophy has been a merely scholastic concept -- a concept of a system of knowledge which is sought solely in its character as a science, and which has therefore in view only the systematic unity appropriate to science, and consequently no more than the logical perfection of knowledge. But there is likewise another concept of philosophy, a conceptus cosmicus, which has always formed the real basis of the term 'philosophy', especially when it has been as it were personified and its archetype represented in the ideal philosopher. On this view, philosophy is the science of the relation of all knowledge to the essential ends of human reason [658] (teleologia rationis humanae), and the philosopher is not an artificer in the field of reason, but himself the lawgiver of human reason. In this sense of the term it would be very vainglorious to entitle oneself a philosopher, and to pretend to have equalled the pattern which exists in the idea alone.

The mathematician, the natural philosopher, and the logician, however successful the two former may have been in their advances in the field of rational knowledge, and the two latter more especially in philosophical knowledge, are yet only artificers in the field of reason. There is a teacher, [conceived] in the ideal, who sets them their tasks, and employs them as instruments, to further the essential ends of human reason. Him alone we must call philosopher; but as he nowhere exists, while the idea of his legislation is to be found in that reason with which every human being is endowed, we shall keep entirely to the latter, determining more precisely what philosophy prescribes as regards systematic unity, in accordance with this cosmical concept,{3} from the standpoint of its essential ends.

Essential ends are not as such the highest ends; in view of the demand of reason for complete systematic unity, only one of them can be so described. Essential ends are therefore either the ultimate end or subordinate ends which are necessarily connected with the former as means. The former is no other than the whole vocation of man, and the philosophy which deals with it is entitled moral philosophy. On account of this superiority which moral philosophy has over all other occupations of reason, the ancients in their use of the term 'philosopher' always meant, more especially, the moralist; and even at the present day we are led by a certain analogy to entitle anyone a philosopher who appears to exhibit self-control under the guidance of reason, however limited his knowledge may be.

The legislation of human reason (philosophy) has two objects, nature and freedom, and therefore contains not only [659] the law of nature, but also the moral law, presenting them at first in two distinct systems, but ultimately in one single philosophical system. The philosophy of nature deals with all that is, the philosophy of morals with that which ought to be.

All philosophy is either knowledge arising out of pure reason, or knowledge obtained by reason from empirical principles. The former is termed pure, the latter empirical philosophy.

The philosophy of pure reason is either a propaedeutic (preparation), which investigates the faculty of reason in respect of all its pure a priori knowledge, and is entitled the science which exhibits in systematic connection the whole body (true as well as illusory) of philosophical knowledge arising out of pure reason, and which is entitled metaphysics. The title 'metaphysics' may also, however, be given to the whole of pure philosophy, inclusive of criticism, and so as comprehending the investigation of all that can ever be known a priori as well as the exposition of that which constitutes a system of the pure philosophical modes of knowledge of this type -- in distinction, therefore, from all empirical and from all mathematical employment of reason.

Metaphysics is divided into that of the speculative and that of the practical employment of pure reason, and is therefore either metaphysics of nature or metaphysics of morals. The former contains all the principles of pure reason that are derived from mere concepts (therefore excluding mathematics), and employed in the theoretical knowledge of all things; the latter, the principles which in a priori fashion determine and make necessary all our actions. Now morality is the only code of laws applying to our actions which can be derived completely a priori from principles. Accordingly, the metaphysics of morals is really pure moral philosophy, with no underlying basis of anthropology or of other empirical conditions. The term 'metaphysics', in its strict sense, is commonly reserved for the metaphysics of speculative reason. But as pure moral philosophy really forms part of this special [660] branch of human and philosophical knowledge derived from pure reason, we shall retain for it the title 'metaphysics'. We are not, however, at present concerned with it, and may therefore leave it aside.

It is of the utmost importance to isolate the various modes of knowledge according as they differ in kind and in origin, and to secure that they be not confounded owing to the fact that usually, in our employment of them, they are combined. What the chemist does in the analysis of substances, and the mathematician in his special disciplines, is in still greater degree incumbent upon the philosopher, that he may be able to determine with certainty the part which belongs to each special kind of knowledge in the diversified employment of the understanding and its special value and influence. Human reason, since it first began to think, or rather to reflect, has never been able to dispense with a metaphysics; but also has never been able to obtain it in a form sufficiently free from all foreign elements. The idea of such a science is as old as speculative human reason; and what rational being does not speculate, either in scholastic or in popular fashion? It must be admitted, however, that the two elements of our knowledge -- that which is in our power completely a priori, and that which is obtainable only a posteriori from experience -- have never been very clearly distinguished, not even by professional thinkers, and that they have therefore failed to bring about the delimitation of a special kind of knowledge, and thereby the true idea of the science which has preoccupied human reason so long and so greatly. When metaphysics was declared to be the science of the first principles of human knowledge, the intention was not to mark out a quite special kind of knowledge, but only a certain precedence in respect of generality, which was not sufficient to distinguish such knowledge from the empirical. For among empirical principles we can distinguish some that are more general, and so higher in rank than others; but where in such a series of subordinated members -- a series in which we do not distinguish what is completely a priori from what is known only a posteriori -- are we to draw the line which distinguishes the highest or first members from the lower subordinate members? What should we say, if in the [661] reckoning of time we could distinguish the epochs of the world only by dividing them into the first centuries and those that follow? We should ask: Does the fifth, the tenth century, etc. , belong with the first centuries? So in like manner I ask: Does the concept of the extended belong to metaphysics? You answer, Yes. Then, that of body too? Yes. And that of fluid body? You now become perplexed; for at this rate everything will belong to metaphysics. It is evident, therefore, that the mere degree of subordination (of the particular under the general) cannot determine the limits of a science; in the case under consideration, only complete difference of kind and of origin will suffice. But the fundamental idea of metaphysics was obscured on yet another side, owing to its exhibiting, as a priori knowledge, a certain similarity to mathematics. Certainly they are related, in so far as they both have an a priori origin; but when we bear in mind the difference between philosophical and mathematical knowledge, namely, that the one is derived from concepts, whereas in the other we arrive at a priori judgments only through the construction which has indeed always been in a manner felt but could never be defined by means of any clear criteria. Thus it has come about that since philosophers failed in the task of developing even the idea of their science, they could have no determinate end or secure guidance in the elaboration of it, and, accordingly, in this arbitrarily conceived enterprise, ignorant as they were of the path to be taken, they have always been at odds with one another as regards the discoveries which each claimed to have made on his own separate path, with the result that their science has been brought into contempt, first among outsiders, and finally even among themselves.

All pure a priori knowledge, owing to the special faculty of knowledge in which alone it can originate, has in itself a peculiar unity; and metaphysics is the philosophy which has as its task the statement of that knowledge in this systematic unity. Its speculative part, which has especially appropriated this name, namely, what we entitle metaphysics of nature, and which considers everything in so far as it is (not that which [662] ought to be) by means of a priori concepts, is divided in the following manner.

Metaphysics, in the narrower meaning of the term, consists of transcendental philosophy and physiology of pure reason. The former treats only of the understanding and of reason, in a system of concepts and principles which relate to objects in general but take no account of objects that may be given (Ontologia); the latter treats of nature, that is, of the sum of given objects (whether given to the senses, or, if we will, to some other kind of intuition) and is therefore physiology -although only rationalis. The employment of reason in this rational study of nature is either physical or hyperphysical, or, in more adequate terms, is either immanent or transcendent. The former is concerned with such knowledge of nature as can be applied in experience (in concreto), the latter with that connection of objects of experience which transcends all experience. This transcendent physiology has as its object either an inner connection or an outer connection, both, however, transcending possible experience. As dealing with an inner connection it is the physiology of nature as a whole, that is, the transcendental knowledge of the world; as dealing with an outer connection, it is the physiology of the relation of nature as a whole to a being above nature, that is to say, it is the transcendental knowledge of God.

Immanent physiology, on the other hand, views nature as the sum of all objects of the senses, and therefore just as it is given us, but solely in accordance with a priori conditions, under which alone it can ever be given us. There are only two kinds of such objects. Those of the outer senses, and so their sum, corporeal nature. The object of inner sense, the soul, and in accordance with our fundamental concepts of it, thinking nature. The metaphysics of corporeal nature is entitled physics; and as it must contain only the principles of an a priori knowledge of it, rational physics. The metaphysics of thinking nature is entitled psychology, and on the same ground is to be understood as being only the rational knowledge of it.

The whole system of metaphysics thus consists of four main parts: (1) ontology; (2) rational physiology; (3) rational cosmology; (4) rational theology. The second part, namely, [663] the doctrine of nature as developed by pure reason, contains two divisions physica rationalis{4} and psychologia rationalis.

The originative idea of a philosophy of pure reason itself prescribes this division, which is therefore architectonic, in accordance with the essential ends of reason, and not merely technical, in accordance with accidentally observed similarities, and so instituted as it were at haphazard. Accordingly the division is also unchangeable and of legislative authority. There are, however, some points which may well seem doubtful, and may weaken our conviction as to the legitimacy of its claims.

First of all, how can I expect to have knowledge a priori (and therefore a metaphysics) of objects in so far as they are given to our senses, that is, given in an a posteriori manner? And how is it possible to know the nature of things and to arrive at a rational physiology according to principles a priori? The answer is this: we take nothing more from experience than is required to give us an object of outer or of inner sense. The object of outer sense we obtain through the mere concept of matter (impenetrable, lifeless extension), the object of inner sense through the concept of a thinking being (in the empirical inner representation, 'I think'). As to the rest, in the whole metaphysical treatment of these objects, we must entirely dispense with all empirical principles which profess to add to these concepts any other more special experience, with a view to our passing further judgments upon the objects. [663]

Secondly, how are we to regard empirical psychology, [664] which has always claimed its place in metaphysics, and from which in our times such great things have been expected for the advancement of metaphysics, the hope of succeeding by a priori methods having been abandoned. I answer that it belongs where the proper (empirical) doctrine of nature belongs, namely, by the side of applied philosophy, the a priori principles of which are contained in pure philosophy; it is therefore so far connected with applied philosophy, though not to be confounded with it. Empirical psychology is thus completely banished from the domain of metaphysics; it is indeed already completely excluded by the very idea of the latter science. In conformity, however, with scholastic usage we must allow it some sort of a place (although as an episode only) in metaphysics and this from economical motives, because it is not yet so rich as to be able to form a subject of study by itself, and yet is too important to be entirely excluded and forced to settle elsewhere, in a neighborhood that might well prove much less congenial than that of metaphysics. Though it is but a stranger it has long been accepted as a member of the household, and we allow it to stay for some time longer, until it is in a position to set up an establishment of its own in a complete anthropology, the pendant to the empirical doctrine of nature.

Such, then, in general, is the idea of metaphysics. At first more was expected from metaphysics than could reasonably be demanded, and for some time it diverted itself with pleasant anticipations. But these hopes having proved deceptive, it has now fallen into general disrepute. The argument of our Critique, taken as a whole, must have sufficiently convinced the reader that although metaphysics cannot be the foundation of religion, it must always continue to be a bulwark of it, and that human reason, being by its very nature dialectical, can never dispense with such a science, which curbs it, and by a scientific and completely convincing self- knowledge, prevents the devastations of which a lawless speculative reason would otherwise quite inevitably be guilty in the field of morals as well as in that of religion. We can therefore be sure that however cold or contemptuously critical may be the attitude of those who judge a science not by its nature but by its accidental effects, we shall always return to metaphysics as to a beloved one with whom we have had a quarrel. For here we are [665] concerned with essential ends -- ends with which metaphysics must ceaselessly occupy itself, either in striving for genuine insight into them, or in refuting those who profess already to have attained it.

Metaphysics, alike of nature and of morals, and especially that criticism of our adventurous and self-reliant reason which serves as an introduction or propaedeutic to metaphysics, alone properly constitutes what may be entitled philosophy, in the strict sense of the term. Its sole preoccupation is wisdom; and it seeks it by the path of science, which, once it has been trodden, can never be overgrown, and permits of no wandering. Mathematics, natural science, even our empirical knowledge, have a high value as means, for the most part, to contingent ends, but also, in the ultimate outcome, to ends that are necessary and essential to humanity. This latter service, however, they can discharge only as they are aided by a knowledge through reason from pure concepts, which, however we may choose to entitle it, is really nothing but metaphysics. For the same reason metaphysics is also the full and complete development of human reason. Quite apart from its influence, as science, in connection with certain specific ends it is an indispensable discipline. For in dealing with reason it treats of those elements and highest maxims which must form the basis of the very possibility of some sciences, and of the use of all. That, as mere speculation, it serves rather to prevent errors than to extend knowledge, does not detract from its value. On the contrary this gives it dignity and authority, through that censorship which secures general order and harmony, and indeed the well-being of the scientific commonwealth, preventing those who labor courageously and fruitfully on its behalf from losing sight of the supreme end, the happiness of all mankind.





THIS title stands here only in order to indicate one remaining division of the system, which future workers must complete. I content myself with casting a cursory glance, from a purely transcendental point of view, namely, that of the nature of pure reason, on the works of those who have labored in this field - - a glance which reveals [many stately] structures, but in ruins only.

It is a very notable fact, although it could not have been otherwise, that in the infancy of philosophy men began where we should incline to end, namely, with the knowledge of God, occupying themselves with the hope, or rather indeed with the specific nature, of another world. However gross the religious concepts generated by the ancient practices which still persisted in each community from an earlier more barbarous state, this did not prevent the more enlightened members from devoting themselves to free investigation of these matters; and they easily discerned that there could be no better ground or more dependable way of pleasing the invisible power that governs the world, and so of being happy in another world at least, than by living the good life. Accordingly theology and morals were the two motives, or rather the two points of reference, in all those abstract enquiries of reason to which men came to devote themselves. It was chiefly, however, the former that step by step committed the purely speculative reason to those labors which afterwards became so renowned under the name of metaphysics. [667]

I shall not here attempt to distinguish the periods of history in which this or that change in metaphysics came about, but shall only give a cursory sketch of the various ideas which gave rise to the chief revolutions [in metaphysical theory]. And here I find that there are three issues in regard to which the most noteworthy changes have taken place in the course of the resulting controversies.

1. In respect of the object of all our 'knowledge through reason', some have been mere sensualists, others mere intellectualists. Epicurus may be regarded as the outstanding philosopher among the former, and Plato among the latter. The distinction between the two schools, subtle as it is, dates from the earliest times; and the two positions have ever since been maintained in unbroken continuity. Those of the former school maintained that reality is to be found solely in the objects of the senses, and that all else is fiction; those of the latter school, on the other hand, declared that in the senses there is nothing but illusion, and that only the understanding knows what is true. The former did not indeed deny reality to the concepts of the understanding; but this reality was for them merely logical, whereas for the others it was mystical. The former conceded intellectual concepts, but admitted sensible objects only. The latter required that true objects should be purely intelligible, and maintained that by means of the pure understanding we have an intuition that is unaccompanied by the senses -- the senses, in their view, serving only to confuse the understanding.

2. In respect of the origin of the modes of 'knowledge through pure reason', the question is as to whether they are derived from experience, or whether in independence of experience they have their origin in reason. Aristotle may be regarded as the chief of the empiricists, and Plato as the chief of the noologists. Locke, who in modern times followed Aristotle, and Leibniz, who followed Plato (although in considerable disagreement with his mystical system), have not been able to bring this conflict to any definitive conclusion. However we may regard Epicurus, he was at least much more consistent in this sensual system than Aristotle and Locke, inasmuch as he never sought to pass by inference beyond the limits of experience. This is especially true as regards Locke, [668] who, after having derived all concepts and principles from experience, goes so far in the use of them as to assert that we can prove the existence of God and the immortality of the soul with the same conclusiveness as any mathematical proposition -- though both lie entirely outside the limits of possible experience.

3. In respect of method. -- If anything is to receive the title of method, it must be a procedure in accordance with principles. We may divide the methods now prevailing in this field of enquiry into the naturalistic and the scientific. The naturalist of pure reason adopts as his principle that through common reason, without science, that is, through what he calls sound reason, he is able, in regard to those most sublime questions which form the problem of metaphysics, to achieve more than is possible through speculation. Thus he is virtually asserting that we can determine the size and distance of the moon with greater certainty by the naked eye than by mathematical devices. This is mere misology, reduced to principles; and what is most absurd of all, the neglect of all artificial means is eulogized as a special method of extending our knowledge. For as regards those who are naturalists from lack of more insight, they cannot rightly be blamed. They follow common reason, without boasting of their ignorance as a method which contains the secret how we are to fetch truth from the deep well of Democritus. Quod sapio, satis est mihi, non ego curo, esse quod Arcesilas aerumnosique Solones{5} is the motto with which they may lead a cheerful and praiseworthy life, not troubling themselves about science, nor by their interference bringing it into confusion.

As regards those who adopt a scientific method, they have the choice of proceeding either dogmatically or skeptically; but in any case they are under obligation to proceed systematically. I may cite the celebrated Wolff as a representative of the former mode of procedure, and David Hume as a representative of the latter, and may then, conformably with my present purpose, leave all others unnamed. The critical path alone is still open. If the reader has had the courtesy and patience to accompany me along this path, he may now judge for himself whether, if he cares to lend his aid in making this path into a high-road, it may not be possible to achieve before the end of the present century what many centuries have not been able to accomplish; namely, to secure for human reason complete satisfaction in regard to that with which it has all along so eagerly occupied itself, though hitherto in vain.


[Table of Contents]


{1} All practical concepts relate to objects of satisfaction or dissatisfaction, that is, of pleasure and pain, and therefore, at least indirectly, to the objects of our feelings. But as feeling is not a faculty whereby we represent things, but lies outside our whole faculty of knowledge, the elements of our judgments so far as they relate to pleasure or pain, that is, the elements of practical judgments, do not belong to transcendental philosophy, which is exclusively concerned with pure a priori modes of knowledge. [Back]

{2} The human mind (as, I likewise believe, must necessarily be the case with every rational being) takes a natural interest in morality, although this interest is not undivided and practically preponderant. If we confirm and increase this interest, we shall find reason very teachable and in itself more enlightened as regards the uniting of the speculative with the practical interest. But if we do not take care that we first make men good, at least in some measure good, we shall never make honest believers of them.[Back]

{3} By 'cosmical concept' [Weltbegriff] is here meant the concept which relates to that in which everyone necessarily has an interest; and accordingly if a science is to be regarded merely as one of the disciplines designed in view of certain optionally chosen ends, I must determine it in conformity with scholastic concepts. [Back]

{4} I must not be taken as meaning thereby what is commonly called physica generalis; the latter is rather mathematics than philosophy of nature. The metaphysics of nature is quite distinct from mathematics. It is very far from enlarging our knowledge in the fruitful manner of mathematics, but still is very important as yielding a criticism of the pure knowledge of understanding in its application to nature. For lack of it, even mathematicians, holding to certain common concepts, which though common are yet in fact metaphysical, have unconsciously encumbered their doctrine of nature with hypotheses which vanish upon criticism of the principles involved, without, however, doing the least injury to the employment of mathematics -- employment which is quite indispensable in this field.[Back]

{5} Persius [Sat. iii 78-79].[Back]


[Table of Contents]