FundamentalPrinciples of theMetaphysic of

Morals

Inmanuel Kant

Wor

k re

prod

uced

with

no

edito

rial

resp

onsi

bilit

y

Notice by Luarna Ediciones

This book is in the public domain becausethe copyrights have expired under Spanish law.

Luarna presents it here as a gift to its cus-tomers, while clarifying the following:

1) Because this edition has not been super-vised by our editorial deparment, wedisclaim responsibility for the fidelity ofits content.

2) Luarna has only adapted the work tomake it easily viewable on common six-inch readers.

3) To all effects, this book must not be con-sidered to have been published byLuarna.

www.luarna.com

FIRST SECTION

TRANSITION FROM THE COMMONRATIONAL KNOWLEDGE

OF MORALITY TO THE PHILOSOPHICAL

Nothing can possibly be conceived in theworld, or even out of it, which can be calledgood, without qualification, except a good will.Intelligence, wit, judgement, and the other tal-ents of the mind, however they may be named,or courage, resolution, perseverance, as quali-ties of temperament, are undoubtedly good anddesirable in many respects; but these gifts ofnature may also become extremely bad andmischievous if the will which is to make use ofthem, and which, therefore, constitutes what is

called character, is not good. It is the same withthe gifts of fortune. Power, riches, honour, evenhealth, and the general well-being and con-tentment with one's condition which is calledhappiness, inspire pride, and often presump-tion, if there is not a good will to correct theinfluence of these on the mind, and with thisalso to rectify the whole principle of acting andadapt it to its end. The sight of a being who isnot adorned with a single feature of a pure andgood will, enjoying unbroken prosperity, cannever give pleasure to an impartial rationalspectator. Thus a good will appears to consti-tute the indispensable condition even of beingworthy of happiness.

There are even some qualities which are of ser-vice to this good will itself and may facilitate itsaction, yet which have no intrinsic uncondi-tional value, but always presuppose a goodwill, and this qualifies the esteem that we justlyhave for them and does not permit us to regard

them as absolutely good. Moderation in theaffections and passions, self-control, and calmdeliberation are not only good in many re-spects, but even seem to constitute part of theintrinsic worth of the person; but they are farfrom deserving to be called good without quali-fication, although they have been so uncondi-tionally praised by the ancients. For withoutthe principles of a good will, they may becomeextremely bad, and the coolness of a villain notonly makes him far more dangerous, but alsodirectly makes him more abominable in oureyes than he would have been without it.

A good will is good not because of what it per-forms or effects, not by its aptness for the at-tainment of some proposed end, but simply byvirtue of the volition; that is, it is good in itself,and considered by itself is to be esteemed muchhigher than all that can be brought about by itin favour of any inclination, nay even of thesum total of all inclinations. Even if it should

happen that, owing to special disfavour of for-tune, or the niggardly provision of a step-motherly nature, this will should wholly lackpower to accomplish its purpose, if with itsgreatest efforts it should yet achieve nothing,and there should remain only the good will(not, to be sure, a mere wish, but the summon-ing of all means in our power), then, like ajewel, it would still shine by its own light, as athing which has its whole value in itself. Itsusefulness or fruitlessness can neither add nortake away anything from this value. It wouldbe, as it were, only the setting to enable us tohandle it the more conveniently in commoncommerce, or to attract to it the attention ofthose who are not yet connoisseurs, but not torecommend it to true connoisseurs, or to de-termine its value.

There is, however, something so strange in thisidea of the absolute value of the mere will, inwhich no account is taken of its utility, that

notwithstanding the thorough assent of evencommon reason to the idea, yet a suspicionmust arise that it may perhaps really be theproduct of mere high-flown fancy, and that wemay have misunderstood the purpose of naturein assigning reason as the governor of our will.Therefore we will examine this idea from thispoint of view.

In the physical constitution of an organizedbeing, that is, a being adapted suitably to thepurposes of life, we assume it as a fundamentalprinciple that no organ for any purpose will befound but what is also the fittest and best adap-ted for that purpose. Now in a being which hasreason and a will, if the proper object of naturewere its conservation, its welfare, in a word, itshappiness, then nature would have hit upon avery bad arrangement in selecting the reason ofthe creature to carry out this purpose. For allthe actions which the creature has to performwith a view to this purpose, and the whole rule

of its conduct, would be far more surely pre-scribed to it by instinct, and that end wouldhave been attained thereby much more cer-tainly than it ever can be by reason. Shouldreason have been communicated to this fa-voured creature over and above, it must onlyhave served it to contemplate the happy consti-tution of its nature, to admire it, to congratulateitself thereon, and to feel thankful for it to thebeneficent cause, but not that it should subjectits desires to that weak and delusive guidanceand meddle bunglingly with the purpose ofnature. In a word, nature would have takencare that reason should not break forth intopractical exercise, nor have the presumption,with its weak insight, to think out for itself theplan of happiness, and of the means of attain-ing it. Nature would not only have taken onherself the choice of the ends, but also of themeans, and with wise foresight would haveentrusted both to instinct.

And, in fact, we find that the more a cultivatedreason applies itself with deliberate purpose tothe enjoyment of life and happiness, so muchthe more does the man fail of true satisfaction.And from this circumstance there arises inmany, if they are candid enough to confess it, acertain degree of misology, that is, hatred ofreason, especially in the case of those who aremost experienced in the use of it, because aftercalculating all the advantages they derive, I donot say from the invention of all the arts ofcommon luxury, but even from the sciences(which seem to them to be after all only a lux-ury of the understanding), they find that theyhave, in fact, only brought more trouble ontheir shoulders, rather than gained in happi-ness; and they end by envying, rather than de-spising, the more common stamp of men whokeep closer to the guidance of mere instinct anddo not allow their reason much influence ontheir conduct. And this we must admit, that thejudgement of those who would very much

lower the lofty eulogies of the advantageswhich reason gives us in regard to the happi-ness and satisfaction of life, or who would evenreduce them below zero, is by no means mo-rose or ungrateful to the goodness with whichthe world is governed, but that there lies at theroot of these judgements the idea that our exis-tence has a different and far nobler end, forwhich, and not for happiness, reason is prop-erly intended, and which must, therefore, beregarded as the supreme condition to which theprivate ends of man must, for the most part, bepostponed.

For as reason is not competent to guide the willwith certainty in regard to its objects and thesatisfaction of all our wants (which it to someextent even multiplies), this being an end towhich an implanted instinct would have ledwith much greater certainty; and since, never-theless, reason is imparted to us as a practicalfaculty, i.e., as one which is to have influence

on the will, therefore, admitting that naturegenerally in the distribution of her capacitieshas adapted the means to the end, its true des-tination must be to produce a will, not merelygood as a means to something else, but good initself, for which reason was absolutely neces-sary. This will then, though not indeed the soleand complete good, must be the supreme goodand the condition of every other, even of thedesire of happiness. Under these circumstances,there is nothing inconsistent with the wisdomof nature in the fact that the cultivation of thereason, which is requisite for the first and un-conditional purpose, does in many ways inter-fere, at least in this life, with the attainment ofthe second, which is always conditional,namely, happiness. Nay, it may even reduce itto nothing, without nature thereby failing ofher purpose. For reason recognizes the estab-lishment of a good will as its highest practicaldestination, and in attaining this purpose iscapable only of a satisfaction of its own proper

kind, namely that from the attainment of anend, which end again is determined by reasononly, notwithstanding that this may involvemany a disappointment to the ends of inclina-tion.

We have then to develop the notion of a willwhich deserves to be highly esteemed for itselfand is good without a view to anything further,a notion which exists already in the soundnatural understanding, requiring rather to becleared up than to be taught, and which in es-timating the value of our actions always takesthe first place and constitutes the condition ofall the rest. In order to do this, we will take thenotion of duty, which includes that of a goodwill, although implying certain subjective re-strictions and hindrances. These, however, farfrom concealing it, or rendering it unrecogniz-able, rather bring it out by contrast and make itshine forth so much the brighter.

I omit here all actions which are already recog-nized as inconsistent with duty, although theymay be useful for this or that purpose, for withthese the question whether they are done fromduty cannot arise at all, since they even conflictwith it. I also set aside those actions which rea-lly conform to duty, but to which men have nodirect inclination, performing them becausethey are impelled thereto by some other incli-nation. For in this case we can readily distin-guish whether the action which agrees withduty is done from duty, or from a selfish view.It is much harder to make this distinction whenthe action accords with duty and the subjecthas besides a direct inclination to it. For exam-ple, it is always a matter of duty that a dealershould not over charge an inexperienced pur-chaser; and wherever there is much commercethe prudent tradesman does not overcharge,but keeps a fixed price for everyone, so that achild buys of him as well as any other. Men arethus honestly served; but this is not enough to

make us believe that the tradesman has soacted from duty and from principles of hon-esty: his own advantage required it; it is out ofthe question in this case to suppose that hemight besides have a direct inclination in fa-vour of the buyers, so that, as it were, from lovehe should give no advantage to one over an-other. Accordingly the action was done neitherfrom duty nor from direct inclination, butmerely with a selfish view.

On the other hand, it is a duty to maintain one'slife; and, in addition, everyone has also a directinclination to do so. But on this account the ofanxious care which most men take for it has nointrinsic worth, and their maxim has no moralimport. They preserve their life as duty re-quires, no doubt, but not because duty requires.On the other band, if adversity and hopelesssorrow have completely taken away the relishfor life; if the unfortunate one, strong in mind,indignant at his fate rather than desponding or

dejected, wishes for death, and yet preserveshis life without loving it- not from inclinationor fear, but from duty- then his maxim has amoral worth.

To be beneficent when we can is a duty; andbesides this, there are many minds so sympa-thetically constituted that, without any othermotive of vanity or self-interest, they find apleasure in spreading joy around them and cantake delight in the satisfaction of others so faras it is their own work. But I maintain that insuch a case an action of this kind, howeverproper, however amiable it may be, has never-theless no true moral worth, but is on a levelwith other inclinations, e.g., the inclination tohonour, which, if it is happily directed to thatwhich is in fact of public utility and accordantwith duty and consequently honourable, de-serves praise and encouragement, but not es-teem. For the maxim lacks the moral import,namely, that such actions be done from duty,

not from inclination. Put the case that the mindof that philanthropist were clouded by sorrowof his own, extinguishing all sympathy with thelot of others, and that, while he still has thepower to benefit others in distress, he is nottouched by their trouble because he is absorbedwith his own; and now suppose that he tearshimself out of this dead insensibility, and per-forms the action without any inclination to it,but simply from duty, then first has his actionits genuine moral worth. Further still; if naturehas put little sympathy in the heart of this orthat man; if he, supposed to be an upright man,is by temperament cold and indifferent to thesufferings of others, perhaps because in respectof his own he is provided with the special giftof patience and fortitude and supposes, or evenrequires, that others should have the same- andsuch a man would certainly not be the meanestproduct of nature- but if nature had not spe-cially framed him for a philanthropist, wouldhe not still find in himself a source from

whence to give himself a far higher worth thanthat of a good-natured temperament could be?Unquestionably. It is just in this that the moralworth of the character is brought out which isincomparably the highest of all, namely, that heis beneficent, not from inclination, but fromduty.

To secure one's own happiness is a duty, atleast indirectly; for discontent with one's condi-tion, under a pressure of many anxieties andamidst unsatisfied wants, might easily becomea great temptation to transgression of duty. Buthere again, without looking to duty, all menhave already the strongest and most intimateinclination to happiness, because it is just inthis idea that all inclinations are combined inone total. But the precept of happiness is oftenof such a sort that it greatly interferes withsome inclinations, and yet a man cannot formany definite and certain conception of the sumof satisfaction of all of them which is called

happiness. It is not then to be wondered at thata single inclination, definite both as to what itpromises and as to the time within which it canbe gratified, is often able to overcome such afluctuating idea, and that a gouty patient, forinstance, can choose to enjoy what he likes, andto suffer what he may, since, according to hiscalculation, on this occasion at least, he has notsacrificed the enjoyment of the present momentto a possibly mistaken expectation of a happi-ness which is supposed to be found in health.But even in this case, if the general desire forhappiness did not influence his will, and sup-posing that in his particular case health was nota necessary element in this calculation, thereyet remains in this, as in all other cases, thislaw, namely, that he should promote his hap-piness not from inclination but from duty, andby this would his conduct first acquire truemoral worth.

It is in this manner, undoubtedly, that we are tounderstand those passages of Scripture also inwhich we are commanded to love our neigh-bour, even our enemy. For love, as an affection,cannot be commanded, but beneficence for du-ty's sake may; even though we are not impelledto it by any inclination- nay, are even repelledby a natural and unconquerable aversion. Thisis practical love and not pathological- a lovewhich is seated in the will, and not in the pro-pensions of sense- in principles of action andnot of tender sympathy; and it is this love alonewhich can be commanded.

The second proposition is: That an action donefrom duty derives its moral worth, not from thepurpose which is to be attained by it, but fromthe maxim by which it is determined, andtherefore does not depend on the realization ofthe object of the action, but merely on the prin-ciple of volition by which the action has takenplace, without regard to any object of desire. It

is clear from what precedes that the purposeswhich we may have in view in our actions, ortheir effects regarded as ends and springs of thewill, cannot give to actions any unconditionalor moral worth. In what, then, can their worthlie, if it is not to consist in the will and in refer-ence to its expected effect? It cannot lie any-where but in the principle of the will withoutregard to the ends which can be attained by theaction. For the will stands between its a prioriprinciple, which is formal, and its a posteriorispring, which is material, as between tworoads, and as it must be determined by some-thing, it that it must be determined by the for-mal principle of volition when an action is donefrom duty, in which case every material princi-ple has been withdrawn from it.

The third proposition, which is a consequenceof the two preceding, I would express thusDuty is the necessity of acting from respect forthe law. I may have inclination for an object as

the effect of my proposed action, but I cannothave respect for it, just for this reason, that it isan effect and not an energy of will. Similarly Icannot have respect for inclination, whether myown or another's; I can at most, if my own, ap-prove it; if another's, sometimes even love it;i.e., look on it as favourable to my own interest.It is only what is connected with my will as aprinciple, by no means as an effect- what doesnot subserve my inclination, but overpowers it,or at least in case of choice excludes it from itscalculation- in other words, simply the law ofitself, which can be an object of respect, andhence a command. Now an action done fromduty must wholly exclude the influence of in-clination and with it every object of the will, sothat nothing remains which can determine thewill except objectively the law, and subjectivelypure respect for this practical law, and conse-quently the maxim * that I should follow thislaw even to the thwarting of all my inclinations.

* A maxim is the subjective principle of voli-tion. The objective principle (i.e., that whichwould also serve subjectively as a practicalprinciple to all rational beings if reason had fullpower over the faculty of desire) is the practicallaw.

Thus the moral worth of an action does not liein the effect expected from it, nor in any princi-ple of action which requires to borrow its mo-tive from this expected effect. For all these ef-fects- agreeableness of one's condition and eventhe promotion of the happiness of others- couldhave been also brought about by other causes,so that for this there would have been no needof the will of a rational being; whereas it is inthis alone that the supreme and unconditionalgood can be found. The pre-eminent goodwhich we call moral can therefore consist innothing else than the conception of law in itself,

which certainly is only possible in a rationalbeing, in so far as this conception, and not theexpected effect, determines the will. This is agood which is already present in the personwho acts accordingly, and we have not to waitfor it to appear first in the result. *

* It might be here objected to me that I take ref-uge behind the word respect in an obscure feel-ing, instead of giving a distinct solution of thequestion by a concept of the reason. But al-though respect is a feeling, it is not a feelingreceived through influence, but is self-wroughtby a rational concept, and, therefore, is specifi-cally distinct from all feelings of the formerkind, which may be referred either to inclina-tion or fear, What I recognise immediately as alaw for me, I recognise with respect. Thismerely signifies the consciousness that my willis subordinate to a law, without the interven-

tion of other influences on my sense. The im-mediate determination of the will by the law,and the consciousness of this, is called respect,so that this is regarded as an effect of the lawon the subject, and not as the cause of it. Re-spect is properly the conception of a worthwhich thwarts my self-love. Accordingly it issomething which is considered neither as anobject of inclination nor of fear, although it hassomething analogous to both. The object of re-spect is the law only, and that the law whichwe impose on ourselves and yet recognise asnecessary in itself. As a law, we are subjectedtoo it without consulting self-love; as imposedby us on ourselves, it is a result of our will. Inthe former aspect it has an analogy to fear, inthe latter to inclination. Respect for a person isproperly only respect for the law (of honesty,etc.) of which he gives us an example. Since wealso look on the improvement of our talents asa duty, we consider that we see in a person oftalents, as it were, the example of a law (viz., to

become like him in this by exercise), and thisconstitutes our respect. All so-called moral in-terest consists simply in respect for the law.

But what sort of law can that be, the conceptionof which must determine the will, even withoutpaying any regard to the effect expected fromit, in order that this will may be called goodabsolutely and without qualification? As I havedeprived the will of every impulse which couldarise to it from obedience to any law, there re-mains nothing but the universal conformity ofits actions to law in general, which alone is toserve the will as a principle, i.e., I am never toact otherwise than so that I could also will thatmy maxim should become a universal law.Here, now, it is the simple conformity to law ingeneral, without assuming any particular lawapplicable to certain actions, that serves the willas its principle and must so serve it, if duty is

not to be a vain delusion and a chimerical no-tion. The common reason of men in its practicaljudgements perfectly coincides with this andalways has in view the principle here sug-gested. Let the question be, for example: May Iwhen in distress make a promise with the in-tention not to keep it? I readily distinguish herebetween the two significations which the ques-tion may have: Whether it is prudent, orwhether it is right, to make a false promise?The former may undoubtedly of be the case. Isee clearly indeed that it is not enough to extri-cate myself from a present difficulty by meansof this subterfuge, but it must be well consid-ered whether there may not hereafter springfrom this lie much greater inconvenience thanthat from which I now free myself, and as, withall my supposed cunning, the consequencescannot be so easily foreseen but that credit oncelost may be much more injurious to me thanany mischief which I seek to avoid at present, itshould be considered whether it would not be

more prudent to act herein according to a uni-versal maxim and to make it a habit to promisenothing except with the intention of keeping it.But it is soon clear to me that such a maxim willstill only be based on the fear of consequences.Now it is a wholly different thing to be truthfulfrom duty and to be so from apprehension ofinjurious consequences. In the first case, thevery notion of the action already implies a lawfor me; in the second case, I must first lookabout elsewhere to see what results may becombined with it which would affect myself.For to deviate from the principle of duty is be-yond all doubt wicked; but to be unfaithful tomy maxim of prudence may often be very ad-vantageous to me, although to abide by it iscertainly safer. The shortest way, however, andan unerring one, to discover the answer to thisquestion whether a lying promise is consistentwith duty, is to ask myself, "Should I be contentthat my maxim (to extricate myself from diffi-culty by a false promise) should hold good as a

universal law, for myself as well as for others?"and should I be able to say to myself, "Everyone may make a deceitful promise when hefinds himself in a difficulty from which he can-not otherwise extricate himself?" Then I pres-ently become aware that while I can will the lie,I can by no means will that lying should be auniversal law. For with such a law there wouldbe no promises at all, since it would be in vainto allege my intention in regard to my futureactions to those who would not believe thisallegation, or if they over hastily did so wouldpay me back in my own coin. Hence mymaxim, as soon as it should be made a univer-sal law, would necessarily destroy itself.

I do not, therefore, need any far-reaching pene-tration to discern what I have to do in orderthat my will may be morally good. Inexperi-enced in the course of the world, incapable ofbeing prepared for all its contingencies, I onlyask myself: Canst thou also will that thy maxim

should be a universal law? If not, then it mustbe rejected, and that not because of a disadvan-tage accruing from it to myself or even to oth-ers, but because it cannot enter as a principleinto a possible universal legislation, and reasonextorts from me immediate respect for suchlegislation. I do not indeed as yet discern onwhat this respect is based (this the philosophermay inquire), but at least I understand this, thatit is an estimation of the worth which far out-weighs all worth of what is recommended byinclination, and that the necessity of actingfrom pure respect for the practical law is whatconstitutes duty, to which every other motivemust give place, because it is the condition of awill being good in itself, and the worth of sucha will is above everything.

Thus, then, without quitting the moral knowl-edge of common human reason, we have ar-rived at its principle. And although, no doubt,common men do not conceive it in such an ab-

stract and universal form, yet they always haveit really before their eyes and use it as the stan-dard of their decision. Here it would be easy toshow how, with this compass in hand, men arewell able to distinguish, in every case that oc-curs, what is good, what bad, conformably toduty or inconsistent with it, if, without in theleast teaching them anything new, we only, likeSocrates, direct their attention to the principlethey themselves employ; and that, therefore,we do not need science and philosophy toknow what we should do to be honest andgood, yea, even wise and virtuous. Indeed wemight well have conjectured beforehand thatthe knowledge of what every man is bound todo, and therefore also to know, would bewithin the reach of every man, even the com-monest. Here we cannot forbear admirationwhen we see how great an advantage the prac-tical judgement has over the theoretical in thecommon understanding of men. In the latter, ifcommon reason ventures to depart from the

laws of experience and from the perceptions ofthe senses, it falls into mere inconceivabilitiesand self-contradictions, at least into a chaos ofuncertainty, obscurity, and instability. But inthe practical sphere it is just when the commonunderstanding excludes all sensible springsfrom practical laws that its power of judgementbegins to show itself to advantage. It then be-comes even subtle, whether it be that it chi-canes with its own conscience or with otherclaims respecting what is to be called right, orwhether it desires for its own instruction todetermine honestly the worth of actions; and,in the latter case, it may even have as good ahope of hitting the mark as any philosopherwhatever can promise himself. Nay, it is almostmore sure of doing so, because the philosophercannot have any other principle, while he mayeasily perplex his judgement by a multitude ofconsiderations foreign to the matter, and soturn aside from the right way. Would it nottherefore be wiser in moral concerns to acqui-

esce in the judgement of common reason, or atmost only to call in philosophy for the purposeof rendering the system of morals more com-plete and intelligible, and its rules more con-venient for use (especially for disputation), butnot so as to draw off the common understand-ing from its happy simplicity, or to bring it bymeans of philosophy into a new path of inquiryand instruction?

Innocence is indeed a glorious thing; only, onthe other hand, it is very sad that it cannot wellmaintain itself and is easily seduced. On thisaccount even wisdom- which otherwise con-sists more in conduct than in knowledge- yethas need of science, not in order to learn fromit, but to secure for its precepts admission andpermanence. Against all the commands of dutywhich reason represents to man as so deservingof respect, he feels in himself a powerful coun-terpoise in his wants and inclinations, the entiresatisfaction of which he sums up under the

name of happiness. Now reason issues itscommands unyieldingly, without promisinganything to the inclinations, and, as it were,with disregard and contempt for these claims,which are so impetuous, and at the same timeso plausible, and which will not allow them-selves to be suppressed by any command.Hence there arises a natural dialectic, i.e., a dis-position, to argue against these strict laws ofduty and to question their validity, or at leasttheir purity and strictness; and, if possible, tomake them more accordant with our wishesand inclinations, that is to say, to corrupt themat their very source, and entirely to destroytheir worth- a thing which even common prac-tical reason cannot ultimately call good.

Thus is the common reason of man compelledto go out of its sphere, and to take a step intothe field of a practical philosophy, not to satisfyany speculative want (which never occurs to itas long as it is content to be mere sound rea-

son), but even on practical grounds, in order toattain in it information and clear instructionrespecting the source of its principle, and thecorrect determination of it in opposition to themaxims which are based on wants and inclina-tions, so that it may escape from the perplexityof opposite claims and not run the risk of losingall genuine moral principles through theequivocation into which it easily falls. Thus,when practical reason cultivates itself, thereinsensibly arises in it a dialetic which forces itto seek aid in philosophy, just as happens to itin its theoretic use; and in this case, therefore,as well as in the other, it will find rest nowherebut in a thorough critical examination of ourreason.

SEC_2

SECOND SECTION

TRANSITION FROM POPULAR MORALPHILOSOPHY

TO THE METAPHYSIC OF MORALS

If we have hitherto drawn our notion of dutyfrom the common use of our practical reason, itis by no means to be inferred that we have trea-ted it as an empirical notion. On the contrary, ifwe attend to the experience of men's conduct,we meet frequent and, as we ourselves allow,just complaints that one cannot find a singlecertain example of the disposition to act frompure duty. Although many things are done inconformity with what duty prescribes, it is ne-vertheless always doubtful whether they aredone strictly from duty, so as to have a moralworth. Hence there have at all times been phi-losophers who have altogether denied that thisdisposition actually exists at all in human ac-

tions, and have ascribed everything to a moreor less refined self-love. Not that they have onthat account questioned the soundness of theconception of morality; on the contrary, theyspoke with sincere regret of the frailty and cor-ruption of human nature, which, though nobleenough to take its rule an idea so worthy ofrespect, is yet weak to follow it and employsreason which ought to give it the law only forthe purpose of providing for the interest of theinclinations, whether singly or at the best in thegreatest possible harmony with one another.

In fact, it is absolutely impossible to make outby experience with complete certainty a singlecase in which the maxim of an action, howeverright in itself, rested simply on moral groundsand on the conception of duty. Sometimes ithappens that with the sharpest self-examination we can find nothing beside themoral principle of duty which could have beenpowerful enough to move us to this or that ac-

tion and to so great a sacrifice; yet we cannotfrom this infer with certainty that it was notreally some secret impulse of self-love, underthe false appearance of duty, that was the ac-tual determining cause of the will. We likethem to flatter ourselves by falsely taking creditfor a more noble motive; whereas in fact we cannever, even by the strictest examination, getcompletely behind the secret springs of action;since, when the question is of moral worth, it isnot with the actions which we see that we areconcerned, but with those inward principles ofthem which we do not see.

Moreover, we cannot better serve the wishes ofthose who ridicule all morality as a mere chi-mera of human imagination over stepping itselffrom vanity, than by conceding to them thatnotions of duty must be drawn only from ex-perience (as from indolence, people are readyto think is also the case with all other notions);for or is to prepare for them a certain triumph. I

am willing to admit out of love of humanitythat even most of our actions are correct, but ifwe look closer at them we everywhere comeupon the dear self which is always prominent,and it is this they have in view and not thestrict command of duty which would oftenrequire self-denial. Without being an enemy ofvirtue, a cool observer, one that does not mis-take the wish for good, however lively, for itsreality, may sometimes doubt whether truevirtue is actually found anywhere in the world,and this especially as years increase and thejudgement is partly made wiser by experienceand partly, also, more acute in observation.This being so, nothing can secure us from fal-ling away altogether from our ideas of duty, ormaintain in the soul a well-grounded respectfor its law, but the clear conviction that al-though there should never have been actionswhich really sprang from such pure sources,yet whether this or that takes place is not at allthe question; but that reason of itself, inde-

pendent on all experience, ordains what oughtto take place, that accordingly actions of whichperhaps the world has hitherto never given anexample, the feasibility even of which might bevery much doubted by one who founds every-thing on experience, are nevertheless inflexiblycommanded by reason; that, e.g., even thoughthere might never yet have been a sincerefriend, yet not a whit the less is pure sincerityin friendship required of every man, because,prior to all experience, this duty is involved asduty in the idea of a reason determining thewill by a priori principles.

When we add further that, unless we deny thatthe notion of morality has any truth or refer-ence to any possible object, we must admit thatits law must be valid, not merely for men butfor all rational creatures generally, not merelyunder certain contingent conditions or withexceptions but with absolute necessity, then itis clear that no experience could enable us to

infer even the possibility of such apodeicticlaws. For with what right could we bring intounbounded respect as a universal precept forevery rational nature that which perhaps holdsonly under the contingent conditions of hu-manity? Or how could laws of the determina-tion of our will be regarded as laws of the de-termination of the will of rational beings gener-ally, and for us only as such, if they weremerely empirical and did not take their originwholly a priori from pure but practical reason?

Nor could anything be more fatal to moralitythan that we should wish to derive it from ex-amples. For every example of it that is set be-fore me must be first itself tested by principlesof morality, whether it is worthy to serve as anoriginal example, i.e., as a pattern; but by nomeans can it authoritatively furnish the concep-tion of morality. Even the Holy One of the Gos-pels must first be compared with our ideal ofmoral perfection before we can recognise Him

as such; and so He says of Himself, "Why callye Me (whom you see) good; none is good (themodel of good) but God only (whom ye do notsee)?" But whence have we the conception ofGod as the supreme good? Simply from theidea of moral perfection, which reason frames apriori and connects inseparably with the notionof a free will. Imitation finds no place at all inmorality, and examples serve only for encour-agement, i.e., they put beyond doubt the feasi-bility of what the law commands, they makevisible that which the practical rule expressesmore generally, but they can never authorize usto set aside the true original which lies in rea-son and to guide ourselves by examples.

If then there is no genuine supreme principle ofmorality but what must rest simply on purereason, independent of all experience, I think itis not necessary even to put the question whet-her it is good to exhibit these concepts in theirgenerality (in abstracto) as they are established

a priori along with the principles belonging tothem, if our knowledge is to be distinguishedfrom the vulgar and to be called philosophical.

In our times indeed this might perhaps be nec-essary; for if we collected votes whether purerational knowledge separated from everythingempirical, that is to say, metaphysic of morals,or whether popular practical philosophy is tobe preferred, it is easy to guess which sidewould preponderate.

This descending to popular notions is certainlyvery commendable, if the ascent to the princi-ples of pure reason has first taken place andbeen satisfactorily accomplished. This impliesthat we first found ethics on metaphysics, andthen, when it is firmly established, procure ahearing for it by giving it a popular character.But it is quite absurd to try to be popular in thefirst inquiry, on which the soundness of theprinciples depends. It is not only that this pro-ceeding can never lay claim to the very rare

merit of a true philosophical popularity, sincethere is no art in being intelligible if one re-nounces all thoroughness of insight; but also itproduces a disgusting medley of compiled ob-servations and half-reasoned principles. Shal-low pates enjoy this because it can be used forevery-day chat, but the sagacious find in it onlyconfusion, and being unsatisfied and unable tohelp themselves, they turn away their eyes,while philosophers, who see quite well throughthis delusion, are little listened to when theycall men off for a time from this pretended po-pularity, in order that they might be rightfullypopular after they have attained a definite in-sight.

We need only look at the attempts of moralistsin that favourite fashion, and we shall find atone time the special constitution of human na-ture (including, however, the idea of a rationalnature generally), at one time perfection, atanother happiness, here moral sense, there fear

of God. a little of this, and a little of that, inmarvellous mixture, without its occurring tothem to ask whether the principles of moralityare to be sought in the knowledge of humannature at all (which we can have only from ex-perience); or, if this is not so, if these principlesare to be found altogether a priori, free fromeverything empirical, in pure rational conceptsonly and nowhere else, not even in the smallestdegree; then rather to adopt the method of ma-king this a separate inquiry, as pure practicalphilosophy, or (if one may use a name so de-cried) as metaphysic of morals, * to bring it byitself to completeness, and to require the public,which wishes for popular treatment, to awaitthe issue of this undertaking.

* Just as pure mathematics are distinguishedfrom applied, pure logic from applied, so if wechoose we may also distinguish pure philoso-

phy of morals (metaphysic) from applied (viz.,applied to human nature). By this designationwe are also at once reminded that moral prin-ciples are not based on properties of humannature, but must subsist a priori of themselves,while from such principles practical rules mustbe capable of being deduced for every rationalnature, and accordingly for that of man.

Such a metaphysic of morals, completely iso-lated, not mixed with any anthropology, theol-ogy, physics, or hyperphysics, and still lesswith occult qualities (which we might call hy-pophysical), is not only an indispensable sub-stratum of all sound theoretical knowledge ofduties, but is at the same time a desideratum ofthe highest importance to the actual fulfilmentof their precepts. For the pure conception ofduty, unmixed with any foreign addition ofempirical attractions, and, in a word, the con-

ception of the moral law, exercises on the hu-man heart, by way of reason alone (which firstbecomes aware with this that it can of itself bepractical), an influence so much more powerfulthan all other springs * which may be derivedfrom the field of experience, that, in the con-sciousness of its worth, it despises the latter,and can by degrees become their master;whereas a mixed ethics, compounded partly ofmotives drawn from feelings and inclinations,and partly also of conceptions of reason, mustmake the mind waver between motives whichcannot be brought under any principle, whichlead to good only by mere accident and veryoften also to evil.

* I have a letter from the late excellent Sulzer, inwhich he asks me what can be the reason thatmoral instruction, although containing muchthat is convincing for the reason, yet accom-

plishes so little? My answer was postponed inorder that I might make it complete. But it issimply this: that the teachers themselves havenot got their own notions clear, and when theyendeavour to make up for this by raking upmotives of moral goodness from every quarter,trying to make their physic right strong, theyspoil it. For the commonest understandingshows that if we imagine, on the one hand, anact of honesty done with steadfast mind, apartfrom every view to advantage of any kind inthis world or another, and even under the grea-test temptations of necessity or allurement, and,on the other hand, a similar act which was af-fected, in however low a degree, by a foreignmotive, the former leaves far behind and eclip-ses the second; it elevates the soul and inspiresthe wish to be able to act in like manner oneself.Even moderately young children feel this im-pression, ana one should never represent dutiesto them in any other light.

From what has been said, it is clear that all mo-ral conceptions have their seat and origin com-pletely a priori in the reason, and that, more-over, in the commonest reason just as truly asin that which is in the highest degree specula-tive; that they cannot be obtained by abstrac-tion from any empirical, and therefore merelycontingent, knowledge; that it is just this purityof their origin that makes them worthy to serveas our supreme practical principle, and that justin proportion as we add anything empirical, wedetract from their genuine influence and fromthe absolute value of actions; that it is not onlyof the greatest necessity, in a purely speculativepoint of view, but is also of the greatest practi-cal importance, to derive these notions andlaws from pure reason, to present them pureand unmixed, and even to determine the com-pass of this practical or pure rational knowl-edge, i.e., to determine the whole faculty ofpure practical reason; and, in doing so, wemust not make its principles dependent on the

particular nature of human reason, though inspeculative philosophy this may be permitted,or may even at times be necessary; but sincemoral laws ought to hold good for every ra-tional creature, we must derive them from thegeneral concept of a rational being. In this way,although for its application to man morality hasneed of anthropology, yet, in the first instance,we must treat it independently as pure phi-losophy, i.e., as metaphysic, complete in itself(a thing which in such distinct branches of sci-ence is easily done); knowing well that unlesswe are in possession of this, it would not onlybe vain to determine the moral element of dutyin right actions for purposes of speculativecriticism, but it would be impossible to basemorals on their genuine principles, even forcommon practical purposes, especially of moralinstruction, so as to produce pure moral dispo-sitions, and to engraft them on men's minds tothe promotion of the greatest possible good inthe world.

But in order that in this study we may not me-rely advance by the natural steps from the com-mon moral judgement (in this case very worthyof respect) to the philosophical, as has beenalready done, but also from a popular philoso-phy, which goes no further than it can reach bygroping with the help of examples, to meta-physic (which does allow itself to be checkedby anything empirical and, as it must measurethe whole extent of this kind of rational knowl-edge, goes as far as ideal conceptions, whereeven examples fail us), we must follow andclearly describe the practical faculty of reason,from the general rules of its determination tothe point where the notion of duty springs fromit.

Everything in nature works according to laws.Rational beings alone have the faculty of actingaccording to the conception of laws, that is ac-cording to principles, i.e., have a will. Since thededuction of actions from principles requires

reason, the will is nothing but practical reason.If reason infallibly determines the will, then theactions of such a being which are recognised asobjectively necessary are subjectively necessaryalso, i.e., the will is a faculty to choose that onlywhich reason independent of inclination recog-nises as practically necessary, i.e., as good. Butif reason of itself does not sufficiently deter-mine the will, if the latter is subject also to sub-jective conditions (particular impulses) whichdo not always coincide with the objective con-ditions; in a word, if the will does not in itselfcompletely accord with reason (which is actu-ally the case with men), then the actions whichobjectively are recognised as necessary are sub-jectively contingent, and the determination ofsuch a will according to objective laws is obli-gation, that is to say, the relation of the objec-tive laws to a will that is not thoroughly good isconceived as the determination of the will of arational being by principles of reason, but

which the will from its nature does not of ne-cessity follow.

The conception of an objective principle, in sofar as it is obligatory for a will, is called a com-mand (of reason), and the formula of the com-mand is called an imperative.

All imperatives are expressed by the wordought [or shall], and thereby indicate the rela-tion of an objective law of reason to a will,which from its subjective constitution is notnecessarily determined by it (an obligation).They say that something would be good to door to forbear, but they say it to a will whichdoes not always do a thing because it is con-ceived to be good to do it. That is practicallygood, however, which determines the will bymeans of the conceptions of reason, and conse-quently not from subjective causes, but objec-tively, that is on principles which are valid forevery rational being as such. It is distinguishedfrom the pleasant, as that which influences the

will only by means of sensation from merelysubjective causes, valid only for the sense ofthis or that one, and not as a principle of rea-son, which holds for every one. *

* The dependence of the desires on sensationsis called inclination, and this accordingly al-ways indicates a want. The dependence of acontingently determinable will on principles ofreason is called an interest. This therefore, isfound only in the case of a dependent willwhich does not always of itself conform to rea-son; in the Divine will we cannot conceive anyinterest. But the human will can also take aninterest in a thing without therefore acting frominterest. The former signifies the practical inter-est in the action, the latter the pathological inthe object of the action. The former indicatesonly dependence of the will on principles ofreason in themselves; the second, dependence

on principles of reason for the sake of inclina-tion, reason supplying only the practical ruleshow the requirement of the inclination may besatisfied. In the first case the action interestsme; in the second the object of the action (be-cause it is pleasant to me). We have seen in thefirst section that in an action done from dutywe must look not to the interest in the object,but only to that in the action itself, and in itsrational principle (viz., the law).

A perfectly good will would therefore be equa-lly subject to objective laws (viz., laws of good),but could not be conceived as obliged therebyto act lawfully, because of itself from its subjec-tive constitution it can only be determined bythe conception of good. Therefore no impera-tives hold for the Divine will, or in general for aholy will; ought is here out of place, becausethe volition is already of itself necessarily in

unison with the law. Therefore imperatives areonly formulae to express the relation of objec-tive laws of all volition to the subjective imper-fection of the will of this or that rational being,e.g., the human will.

Now all imperatives command either hypo-thetically or categorically. The former representthe practical necessity of a possible action asmeans to something else that is willed (or atleast which one might possibly will). The cate-gorical imperative would be that which repre-sented an action as necessary of itself withoutreference to another end, i.e., as objectively ne-cessary.

Since every practical law represents a possibleaction as good and, on this account, for a sub-ject who is practically determinable by reason,necessary, all imperatives are formulae deter-mining an action which is necessary accordingto the principle of a will good in some respects.If now the action is good only as a means to

something else, then the imperative is hypo-thetical; if it is conceived as good in itself andconsequently as being necessarily the principleof a will which of itself conforms to reason,then it is categorical.

Thus the imperative declares what action pos-sible by me would be good and presents thepractical rule in relation to a will which doesnot forthwith perform an action simply becauseit is good, whether because the subject does notalways know that it is good, or because, even ifit know this, yet its maxims might be opposedto the objective principles of practical reason.

Accordingly the hypothetical imperative onlysays that the action is good for some purpose,possible or actual. In the first case it is a prob-lematical, in the second an assertorial practicalprinciple. The categorical imperative whichdeclares an action to be objectively necessary initself without reference to any purpose, i.e.,

without any other end, is valid as an apodeictic(practical) principle.

Whatever is possible only by the power of so-me rational being may also be conceived as apossible purpose of some will; and thereforethe principles of action as regards the meansnecessary to attain some possible purpose arein fact infinitely numerous. All sciences have apractical part, consisting of problems express-ing that some end is possible for us and of im-peratives directing how it may be attained.These may, therefore, be called in general im-peratives of skill. Here there is no questionwhether the end is rational and good, but onlywhat one must do in order to attain it. The pre-cepts for the physician to make his patient tho-roughly healthy, and for a poisoner to ensurecertain death, are of equal value in this respect,that each serves to effect its purpose perfectly.Since in early youth it cannot be known whatends are likely to occur to us in the course of

life, parents seek to have their children taught agreat many things, and provide for their skill inthe use of means for all sorts of arbitrary ends,of none of which can they determine whether itmay not perhaps hereafter be an object to theirpupil, but which it is at all events possible thathe might aim at; and this anxiety is so greatthat they commonly neglect to form and correcttheir judgement on the value of the thingswhich may be chosen as ends.

There is one end, however, which may be as-sumed to be actually such to all rational beings(so far as imperatives apply to them, viz., asdependent beings), and, therefore, one purposewhich they not merely may have, but which wemay with certainty assume that they all actu-ally have by a natural necessity, and this is hap-piness. The hypothetical imperative which ex-presses the practical necessity of an action asmeans to the advancement of happiness is as-sertorial. We are not to present it as necessary

for an uncertain and merely possible purpose,but for a purpose which we may presupposewith certainty and a priori in every man, be-cause it belongs to his being. Now skill in thechoice of means to his own greatest well-beingmay be called prudence, * in the narrowest sen-se. And thus the imperative which refers to thechoice of means to one's own happiness, i.e.,the precept of prudence, is still always hypo-thetical; the action is not commanded abso-lutely, but only as means to another purpose.

* The word prudence is taken in two senses: inthe one it may bear the name of knowledge ofthe world, in the other that of private prudence.The former is a man's ability to influence othersso as to use them for his own purposes. Thelatter is the sagacity to combine all these pur-poses for his own lasting benefit. This latter isproperly that to which the value even of the

former is reduced, and when a man is prudentin the former sense, but not in the latter, wemight better say of him that he is clever andcunning, but, on the whole, imprudent.

Finally, there is an imperative which com-mands a certain conduct immediately, withouthaving as its condition any other purpose to beattained by it. This imperative is categorical. Itconcerns not the matter of the action, or its in-tended result, but its form and the principle ofwhich it is itself a result; and what is essentiallygood in it consists in the mental disposition, letthe consequence be what it may. This impera-tive may be called that of morality.

There is a marked distinction also between thevolitions on these three sorts of principles inthe dissimilarity of the obligation of the will. Inorder to mark this difference more clearly, I

think they would be most suitably named intheir order if we said they are either rules ofskill, or counsels of prudence, or commands(laws) of morality. For it is law only that in-volves the conception of an unconditional andobjective necessity, which is consequently uni-versally valid; and commands are laws whichmust be obeyed, that is, must be followed, evenin opposition to inclination. Counsels, indeed,involve necessity, but one which can only holdunder a contingent subjective condition, viz.,they depend on whether this or that man reck-ons this or that as part of his happiness; thecategorical imperative, on the contrary, is notlimited by any condition, and as being abso-lutely, although practically, necessary, may bequite properly called a command. We mightalso call the first kind of imperatives technical(belonging to art), the second pragmatic * (towelfare), the third moral (belonging to freeconduct generally, that is, to morals).

* It seems to me that the proper signification ofthe word pragmatic may be most accuratelydefined in this way. For sanctions are calledpragmatic which flow properly not from thelaw of the states as necessary enactments, butfrom precaution for the general welfare. A his-tory is composed pragmatically when it teachesprudence, i.e., instructs the world how it canprovide for its interests better, or at least aswell as, the men of former time.

Now arises the question, how are all these im-peratives possible? This question does not seekto know how we can conceive the accomplish-ment of the action which the imperative or-dains, but merely how we can conceive the ob-ligation of the will which the imperative ex-presses. No special explanation is needed toshow how an imperative of skill is possible.Whoever wills the end, wills also (so far as rea-

son decides his conduct) the means in his po-wer which are indispensably necessary thereto.This proposition is, as regards the volition, ana-lytical; for, in willing an object as my effect,there is already thought the causality of myselfas an acting cause, that is to say, the use of themeans; and the imperative educes from theconception of volition of an end the conceptionof actions necessary to this end. Syntheticalpropositions must no doubt be employed indefining the means to a proposed end; but theydo not concern the principle, the act of the will,but the object and its realization. E.g., that inorder to bisect a line on an unerring principle Imust draw from its extremities two intersectingarcs; this no doubt is taught by mathematicsonly in synthetical propositions; but if I knowthat it is only by this process that the intendedoperation can be performed, then to say that, ifI fully will the operation, I also will the actionrequired for it, is an analytical proposition; forit is one and the same thing to conceive some-

thing as an effect which I can produce in a cer-tain way, and to conceive myself as acting inthis way.

If it were only equally easy to give a definiteconception of happiness, the imperatives ofprudence would correspond exactly with thoseof skill, and would likewise be analytical. For inthis case as in that, it could be said: "Whoeverwills the end, wills also (according to the dic-tate of reason necessarily) the indispensablemeans thereto which are in his power." But,unfortunately, the notion of happiness is soindefinite that although every man wishes toattain it, yet he never can say definitely andconsistently what it is that he really wishes andwills. The reason of this is that all the elementswhich belong to the notion of happiness arealtogether empirical, i.e., they must be bor-rowed from experience, and nevertheless theidea of happiness requires an absolute whole, amaximum of welfare in my present and all fu-

ture circumstances. Now it is impossible thatthe most clear-sighted and at the same timemost powerful being (supposed finite) shouldframe to himself a definite conception of whathe really wills in this. Does he will riches, howmuch anxiety, envy, and snares might he notthereby draw upon his shoulders? Does he willknowledge and discernment, perhaps it mightprove to be only an eye so much the sharper toshow him so much the more fearfully the evilsthat are now concealed from him, and that can-not be avoided, or to impose more wants on hisdesires, which already give him concernenough. Would he have long life? who guaran-tees to him that it would not be a long misery?would he at least have health? how often hasuneasiness of the body restrained from excessesinto which perfect health would have allowedone to fall? and so on. In short, he is unable, onany principle, to determine with certainty whatwould make him truly happy; because to do sohe would need to be omniscient. We cannot

therefore act on any definite principles to se-cure happiness, but only on empirical counsels,e.g. of regimen, frugality, courtesy, reserve, etc.,which experience teaches do, on the average,most promote well-being. Hence it follows thatthe imperatives of prudence do not, strictlyspeaking, command at all, that is, they cannotpresent actions objectively as practically neces-sary; that they are rather to be regarded ascounsels (consilia) than precepts precepts ofreason, that the problem to determine certainlyand universally what action would promote thehappiness of a rational being is completely in-soluble, and consequently no imperative re-specting it is possible which should, in the strictsense, command to do what makes happy; be-cause happiness is not an ideal of reason but ofimagination, resting solely on empiricalgrounds, and it is vain to expect that theseshould define an action by which one couldattain the totality of a series of consequenceswhich is really endless. This imperative of pru-

dence would however be an analytical proposi-tion if we assume that the means to happinesscould be certainly assigned; for it is distin-guished from the imperative of skill only bythis, that in the latter the end is merely possible,in the former it is given; as however both onlyordain the means to that which we suppose tobe willed as an end, it follows that the impera-tive which ordains the willing of the means tohim who wills the end is in both cases analyti-cal. Thus there is no difficulty in regard to thepossibility of an imperative of this kind either.

On the other hand, the question how the im-perative of morality is possible, is undoubtedlyone, the only one, demanding a solution, as thisis not at all hypothetical, and the objective ne-cessity which it presents cannot rest on anyhypothesis, as is the case with the hypotheticalimperatives. Only here we must never leaveout of consideration that we cannot make outby any example, in other words empirically,

whether there is such an imperative at all, but itis rather to be feared that all those which seemto be categorical may yet be at bottom hypo-thetical. For instance, when the precept is:"Thou shalt not promise deceitfully"; and it isassumed that the necessity of this is not a merecounsel to avoid some other evil, so that itshould mean: "Thou shalt not make a lyingpromise, lest if it become known thou shouldstdestroy thy credit," but that an action of thiskind must be regarded as evil in itself, so thatthe imperative of the prohibition is categorical;then we cannot show with certainty in any ex-ample that the will was determined merely bythe law, without any other spring of action,although it may appear to be so. For it is alwayspossible that fear of disgrace, perhaps also ob-scure dread of other dangers, may have a secretinfluence on the will. Who can prove by experi-ence the non-existence of a cause when all thatexperience tells us is that we do not perceive it?But in such a case the so-called moral impera-

tive, which as such appears to be categoricaland unconditional, would in reality be only apragmatic precept, drawing our attention toour own interests and merely teaching us totake these into consideration.

We shall therefore have to investigate a priorithe possibility of a categorical imperative, as wehave not in this case the advantage of its realitybeing given in experience, so that [the elucida-tion of] its possibility should be requisite onlyfor its explanation, not for its establishment. Inthe meantime it may be discerned beforehandthat the categorical imperative alone has thepurport of a practical law; all the rest may in-deed be called principles of the will but notlaws, since whatever is only necessary for theattainment of some arbitrary purpose may beconsidered as in itself contingent, and we can atany time be free from the precept if we give upthe purpose; on the contrary, the unconditionalcommand leaves the will no liberty to choose

the opposite; consequently it alone carries withit that necessity which we require in a law.

Secondly, in the case of this categorical impera-tive or law of morality, the difficulty (of dis-cerning its possibility) is a very profound one.It is an a priori synthetical practicalproposition; * and as there is so much difficultyin discerning the possibility of speculativepropositions of this kind, it may readily besupposed that the difficulty will be no less withthe practical.

* I connect the act with the will without pre-supposing any condition resulting from anyinclination, but a priori, and therefore necessar-ily (though only objectively, i.e., assuming theidea of a reason possessing full power over allsubjective motives). This is accordingly a prac-tical proposition which does not deduce thewilling of an action by mere analysis from an-

other already presupposed (for we have notsuch a perfect will), but connects it immediatelywith the conception of the will of a rationalbeing, as something not contained in it.

In this problem we will first inquire whetherthe mere conception of a categorical imperativemay not perhaps supply us also with the for-mula of it, containing the proposition whichalone can be a categorical imperative; for evenif we know the tenor of such an absolute com-mand, yet how it is possible will require furtherspecial and laborious study, which we post-pone to the last section.

When I conceive a hypothetical imperative, ingeneral I do not know beforehand what it willcontain until I am given the condition. Butwhen I conceive a categorical imperative, Iknow at once what it contains. For as the im-

perative contains besides the law only the ne-cessity that the maxims * shall conform to thislaw, while the law contains no conditions re-stricting it, there remains nothing but the gen-eral statement that the maxim of the actionshould conform to a universal law, and it is thisconformity alone that the imperative properlyrepresents as necessary.

* A maxim is a subjective principle of action,and must be distinguished from the objectiveprinciple, namely, practical law. The formercontains the practical rule set by reason accord-ing to the conditions of the subject (often itsignorance or its inclinations), so that it is theprinciple on which the subject acts; but the lawis the objective principle valid for every rationalbeing, and is the principle on which it ought toact that is an imperative.

There is therefore but one categorical impera-tive, namely, this: Act only on that maxim whe-reby thou canst at the same time will that itshould become a universal law.

Now if all imperatives of duty can be deducedfrom this one imperative as from their princi-ple, then, although it should remain undecidedwhat is called duty is not merely a vain notion,yet at least we shall be able to show what weunderstand by it and what this notion means.

Since the universality of the law according towhich effects are produced constitutes what isproperly called nature in the most general sen-se (as to form), that is the existence of things sofar as it is determined by general laws, the im-perative of duty may be expressed thus: Act asif the maxim of thy action were to become bythy will a universal law of nature.

We will now enumerate a few duties, adoptingthe usual division of them into duties to our-

selves and ourselves and to others, and intoperfect and imperfect duties. *

* It must be noted here that I reserve the divi-sion of duties for a future metaphysic of mor-als; so that I give it here only as an arbitraryone (in order to arrange my examples). For therest, I understand by a perfect duty one thatadmits no exception in favour of inclinationand then I have not merely external but alsointernal perfect duties. This is contrary to theuse of the word adopted in the schools; but I donot intend to justify there, as it is all one for mypurpose whether it is admitted or not.

1. A man reduced to despair by a series of mis-fortunes feels wearied of life, but is still so farin possession of his reason that he can ask him-

self whether it would not be contrary to hisduty to himself to take his own life. Now heinquires whether the maxim of his action couldbecome a universal law of nature. His maximis: "From self-love I adopt it as a principle toshorten my life when its longer duration is like-ly to bring more evil than satisfaction." It is as-ked then simply whether this principle foun-ded on self-love can become a universal law ofnature. Now we see at once that a system ofnature of which it should be a law to destroylife by means of the very feeling whose specialnature it is to impel to the improvement of lifewould contradict itself and, therefore, could notexist as a system of nature; hence that maximcannot possibly exist as a universal law of na-ture and, consequently, would be wholly in-consistent with the supreme principle of allduty.

2. Another finds himself forced by necessity toborrow money. He knows that he will not be

able to repay it, but sees also that nothing willbe lent to him unless he promises stoutly torepay it in a definite time. He desires to makethis promise, but he has still so much con-science as to ask himself: "Is it not unlawful andinconsistent with duty to get out of a difficultyin this way?" Suppose however that he resolvesto do so: then the maxim of his action would beexpressed thus: "When I think myself in wantof money, I will borrow money and promise torepay it, although I know that I never can doso." Now this principle of self-love or of one'sown advantage may perhaps be consistent withmy whole future welfare; but the question nowis, "Is it right?" I change then the suggestion ofself-love into a universal law, and state thequestion thus: "How would it be if my maximwere a universal law?" Then I see at once that itcould never hold as a universal law of nature,but would necessarily contradict itself. For sup-posing it to be a universal law that everyonewhen he thinks himself in a difficulty should be

able to promise whatever he pleases, with thepurpose of not keeping his promise, the prom-ise itself would become impossible, as well asthe end that one might have in view in it, sinceno one would consider that anything was pro-mised to him, but would ridicule all such sta-tements as vain pretences.

3. A third finds in himself a talent which withthe help of some culture might make him a use-ful man in many respects. But he finds himselfin comfortable circumstances and prefers toindulge in pleasure rather than to take pains inenlarging and improving his happy naturalcapacities. He asks, however, whether hismaxim of neglect of his natural gifts, besidesagreeing with his inclination to indulgence,agrees also with what is called duty. He seesthen that a system of nature could indeed sub-sist with such a universal law although men(like the South Sea islanders) should let theirtalents rest and resolve to devote their lives

merely to idleness, amusement, and propaga-tion of their species- in a word, to enjoyment;but he cannot possibly will that this should be auniversal law of nature, or be implanted in usas such by a natural instinct. For, as a rationalbeing, he necessarily wills that his faculties bedeveloped, since they serve him and have beengiven him, for all sorts of possible purposes.

4. A fourth, who is in prosperity, while he seesthat others have to contend with great wretch-edness and that he could help them, thinks:"What concern is it of mine? Let everyone be ashappy as Heaven pleases, or as he can makehimself; I will take nothing from him nor evenenvy him, only I do not wish to contribute any-thing to his welfare or to his assistance in dis-tress!" Now no doubt if such a mode of think-ing were a universal law, the human race mightvery well subsist and doubtless even betterthan in a state in which everyone talks of sym-pathy and good-will, or even takes care occa-

sionally to put it into practice, but, on the otherside, also cheats when he can, betrays the rightsof men, or otherwise violates them. But al-though it is possible that a universal law of na-ture might exist in accordance with that maxim,it is impossible to will that such a principleshould have the universal validity of a law ofnature. For a will which resolved this wouldcontradict itself, inasmuch as many cases mightoccur in which one would have need of thelove and sympathy of others, and in which, bysuch a law of nature, sprung from his own will,he would deprive himself of all hope of the aidhe desires.

These are a few of the many actual duties, or atleast what we regard as such, which obviouslyfall into two classes on the one principle thatwe have laid down. We must be able to willthat a maxim of our action should be a univer-sal law. This is the canon of the moral apprecia-tion of the action generally. Some actions are of

such a character that their maxim cannot with-out contradiction be even conceived as a uni-versal law of nature, far from it being possiblethat we should will that it should be so. In oth-ers this intrinsic impossibility is not found, butstill it is impossible to will that their maximshould be raised to the universality of a law ofnature, since such a will would contradict itselfIt is easily seen that the former violate strict orrigorous (inflexible) duty; the latter only laxer(meritorious) duty. Thus it has been completelyshown how all duties depend as regards thenature of the obligation (not the object of theaction) on the same principle.

If now we attend to ourselves on occasion ofany transgression of duty, we shall find that wein fact do not will that our maxim should be auniversal law, for that is impossible for us; onthe contrary, we will that the opposite shouldremain a universal law, only we assume theliberty of making an exception in our own fa-

vour or (just for this time only) in favour of ourinclination. Consequently if we considered allcases from one and the same point of view,namely, that of reason, we should find a con-tradiction in our own will, namely, that a cer-tain principle should be objectively necessaryas a universal law, and yet subjectively shouldnot be universal, but admit of exceptions. Ashowever we at one moment regard our actionfrom the point of view of a will wholly con-formed to reason, and then again look at thesame action from the point of view of a willaffected by inclination, there is not really anycontradiction, but an antagonism of inclinationto the precept of reason, whereby the universal-ity of the principle is changed into a mere gen-erality, so that the practical principle of reasonshall meet the maxim half way. Now, althoughthis cannot be justified in our own impartialjudgement, yet it proves that we do really rec-ognise the validity of the categorical imperativeand (with all respect for it) only allow ourselves

a few exceptions, which we think unimportantand forced from us.

We have thus established at least this much,that if duty is a conception which is to have anyimport and real legislative authority for ouractions, it can only be expressed in categoricaland not at all in hypothetical imperatives. Wehave also, which is of great importance, exhib-ited clearly and definitely for every practicalapplication the content of the categorical im-perative, which must contain the principle ofall duty if there is such a thing at all. We havenot yet, however, advanced so far as to prove apriori that there actually is such an imperative,that there is a practical law which commandsabsolutely of itself and without any other im-pulse, and that the following of this law is duty.

With the view of attaining to this, it is of ex-treme importance to remember that we mustnot allow ourselves to think of deducing thereality of this principle from the particular at-

tributes of human nature. For duty is to be apractical, unconditional necessity of action; itmust therefore hold for all rational beings (towhom an imperative can apply at all), and forthis reason only be also a law for all humanwills. On the contrary, whatever is deducedfrom the particular natural characteristics ofhumanity, from certain feelings and propen-sions, nay, even, if possible, from any particulartendency proper to human reason, and whichneed not necessarily hold for the will of everyrational being; this may indeed supply us witha maxim, but not with a law; with a subjectiveprinciple on which we may have a propensionand inclination to act, but not with an objectiveprinciple on which we should be enjoined toact, even though all our propensions, inclina-tions, and natural dispositions were opposed toit. In fact, the sublimity and intrinsic dignity ofthe command in duty are so much the moreevident, the less the subjective impulses favourit and the more they oppose it, without being

able in the slightest degree to weaken the obli-gation of the law or to diminish its validity.

Here then we see philosophy brought to a criti-cal position, since it has to be firmly fixed, not-withstanding that it has nothing to support it inheaven or earth. Here it must show its purity asabsolute director of its own laws, not the heraldof those which are whispered to it by an im-planted sense or who knows what tutelary na-ture. Although these may be better than noth-ing, yet they can never afford principles dic-tated by reason, which must have their sourcewholly a priori and thence their commandingauthority, expecting everything from the su-premacy of the law and the due respect for it,nothing from inclination, or else condemningthe man to self-contempt and inward abhor-rence.

Thus every empirical element is not only quiteincapable of being an aid to the principle ofmorality, but is even highly prejudicial to the

purity of morals, for the proper and inestimableworth of an absolutely good will consists just inthis, that the principle of action is free from allinfluence of contingent grounds, which aloneexperience can furnish. We cannot too much ortoo often repeat our warning against this laxand even mean habit of thought which seeksfor its principle amongst empirical motives andlaws; for human reason in its weariness is gladto rest on this pillow, and in a dream of sweetillusions (in which, instead of Juno, it embracesa cloud) it substitutes for morality a bastardpatched up from limbs of various derivation,which looks like anything one chooses to see init, only not like virtue to one who has once be-held her in her true form. *

* To behold virtue in her proper form is noth-ing else but to contemplate morality stripped ofall admixture of sensible things and of every

spurious ornament of reward or self-love. Howmuch she then eclipses everything else thatappears charming to the affections, every onemay readily perceive with the least exertion ofhis reason, if it be not wholly spoiled for ab-straction.

The question then is this: "Is it a necessary lawfor all rational beings that they should alwaysjudge of their actions by maxims of which theycan themselves will that they should serve asuniversal laws?" If it is so, then it must be con-nected (altogether a priori) with the very con-ception of the will of a rational being generally.But in order to discover this connexion wemust, however reluctantly, take a step into me-taphysic, although into a domain of it which isdistinct from speculative philosophy, namely,the metaphysic of morals. In a practical phi-losophy, where it is not the reasons of what

happens that we have to ascertain, but the lawsof what ought to happen, even although itnever does, i.e., objective practical laws, there itis not necessary to inquire into the reasons whyanything pleases or displeases, how the pleas-ure of mere sensation differs from taste, andwhether the latter is distinct from a generalsatisfaction of reason; on what the feeling ofpleasure or pain rests, and how from it desiresand inclinations arise, and from these againmaxims by the co-operation of reason: for allthis belongs to an empirical psychology, whichwould constitute the second part of physics, ifwe regard physics as the philosophy of nature,so far as it is based on empirical laws. But herewe are concerned with objective practical lawsand, consequently, with the relation of the willto itself so far as it is determined by reasonalone, in which case whatever has reference toanything empirical is necessarily excluded;since if reason of itself alone determines theconduct (and it is the possibility of this that we

are now investigating), it must necessarily doso a priori.

The will is conceived as a faculty of determin-ing oneself to action in accordance with theconception of certain laws. And such a facultycan be found only in rational beings. Now thatwhich serves the will as the objective ground ofits self-determination is the end, and, if this isassigned by reason alone, it must hold for allrational beings. On the other hand, that whichmerely contains the ground of possibility of theaction of which the effect is the end, this is ca-lled the means. The subjective ground of thedesire is the spring, the objective ground of thevolition is the motive; hence the distinctionbetween subjective ends which rest on springs,and objective ends which depend on motivesvalid for every rational being. Practical princi-ples are formal when they abstract from all sub-jective ends; they are material when they as-sume these, and therefore particular springs of

action. The ends which a rational being pro-poses to himself at pleasure as effects of hisactions (material ends) are all only relative, forit is only their relation to the particular desiresof the subject that gives them their worth,which therefore cannot furnish principles uni-versal and necessary for all rational beings andfor every volition, that is to say practical laws.Hence all these relative ends can give rise onlyto hypothetical imperatives.

Supposing, however, that there were some-thing whose existence has in itself an absoluteworth, something which, being an end in itself,could be a source of definite laws; then in thisand this alone would lie the source of a possiblecategorical imperative, i.e., a practical law.

Now I say: man and generally any rational be-ing exists as an end in himself, not merely as ameans to be arbitrarily used by this or that will,but in all his actions, whether they concernhimself or other rational beings, must be al-

ways regarded at the same time as an end. Allobjects of the inclinations have only a condi-tional worth, for if the inclinations and thewants founded on them did not exist, then theirobject would be without value. But the inclina-tions, themselves being sources of want, are sofar from having an absolute worth for whichthey should be desired that on the contrary itmust be the universal wish of every rationalbeing to be wholly free from them. Thus theworth of any object which is to be acquired byour action is always conditional. Beings whoseexistence depends not on our will but on na-ture's, have nevertheless, if they are irrationalbeings, only a relative value as means, and aretherefore called things; rational beings, on thecontrary, are called persons, because their verynature points them out as ends in themselves,that is as something which must not be usedmerely as means, and so far therefore restrictsfreedom of action (and is an object of respect).These, therefore, are not merely subjective ends

whose existence has a worth for us as an effectof our action, but objective ends, that is, thingswhose existence is an end in itself; an endmoreover for which no other can be substi-tuted, which they should subserve merely asmeans, for otherwise nothing whatever wouldpossess absolute worth; but if all worth wereconditioned and therefore contingent, thenthere would be no supreme practical principleof reason whatever.

If then there is a supreme practical principle or,in respect of the human will, a categorical im-perative, it must be one which, being drawnfrom the conception of that which is necessarilyan end for everyone because it is an end in it-self, constitutes an objective principle of will,and can therefore serve as a universal practicallaw. The foundation of this principle is: rationalnature exists as an end in itself. Man necessarilyconceives his own existence as being so; so farthen this is a subjective principle of human ac-

tions. But every other rational being regards itsexistence similarly, just on the same rationalprinciple that holds for me: * so that it is at thesame time an objective principle, from which asa supreme practical law all laws of the willmust be capable of being deduced. Accordinglythe practical imperative will be as follows: Soact as to treat humanity, whether in thine ownperson or in that of any other, in every case asan end withal, never as means only. We willnow inquire whether this can be practicallycarried out.

* This proposition is here stated as a postulate.The ground of it will be found in the conclud-ing section.

To abide by the previous examples:

Firstly, under the head of necessary duty tooneself: He who contemplates suicide shouldask himself whether his action can be consistentwith the idea of humanity as an end in itself. Ifhe destroys himself in order to escape frompainful circumstances, he uses a person merelyas a mean to maintain a tolerable condition upto the end of life. But a man is not a thing, thatis to say, something which can be used merelyas means, but must in all his actions be alwaysconsidered as an end in himself. I cannot, there-fore, dispose in any way of a man in my ownperson so as to mutilate him, to damage or killhim. (It belongs to ethics proper to define thisprinciple more precisely, so as to avoid all mis-understanding, e. g., as to the amputation of thelimbs in order to preserve myself, as to expos-ing my life to danger with a view to preserve it,etc. This question is therefore omitted here.)

Secondly, as regards necessary duties, or thoseof strict obligation, towards others: He who is

thinking of making a lying promise to otherswill see at once that he would be using anotherman merely as a mean, without the latter con-taining at the same time the end in himself. Forhe whom I propose by such a promise to usefor my own purposes cannot possibly assent tomy mode of acting towards him and, therefore,cannot himself contain the end of this action.This violation of the principle of humanity inother men is more obvious if we take in exam-ples of attacks on the freedom and property ofothers. For then it is clear that he who trans-gresses the rights of men intends to use theperson of others merely as a means, withoutconsidering that as rational beings they oughtalways to be esteemed also as ends, that is, asbeings who must be capable of containing inthemselves the end of the very same action. *

* Let it not be thought that the common "quodtibi non vis fieri, etc." could serve here as therule or principle. For it is only a deduction fromthe former, though with several limitations; itcannot be a universal law, for it does not con-tain the principle of duties to oneself, nor of theduties of benevolence to others (for many a onewould gladly consent that others should notbenefit him, provided only that he might beexcused from showing benevolence to them),nor finally that of duties of strict obligation toone another, for on this principle the criminalmight argue against the judge who punisheshim, and so on.

Thirdly, as regards contingent (meritorious)duties to oneself: It is not enough that the ac-tion does not violate humanity in our own per-son as an end in itself, it must also harmonizewith it. Now there are in humanity capacities of

greater perfection, which belong to the end thatnature has in view in regard to humanity inourselves as the subject: to neglect these mightperhaps be consistent with the maintenance ofhumanity as an end in itself, but not with theadvancement of this end.

Fourthly, as regards meritorious duties to-wards others: The natural end which all menhave is their own happiness. Now humanitymight indeed subsist, although no one shouldcontribute anything to the happiness of others,provided he did not intentionally withdrawanything from it; but after all this would onlyharmonize negatively not positively with hu-manity as an end in itself, if every one does notalso endeavour, as far as in him lies, to forwardthe ends of others. For the ends of any subjectwhich is an end in himself ought as far as pos-sible to be my ends also, if that conception is tohave its full effect with me.

This principle, that humanity and generallyevery rational nature is an end in itself (whichis the supreme limiting condition of everyman's freedom of action), is not borrowed fromexperience, firstly, because it is universal, ap-plying as it does to all rational beings whatever,and experience is not capable of determininganything about them; secondly, because it doesnot present humanity as an end to men (subjec-tively), that is as an object which men do ofthemselves actually adopt as an end; but as anobjective end, which must as a law constitutethe supreme limiting condition of all our sub-jective ends, let them be what we will; it musttherefore spring from pure reason. In fact theobjective principle of all practical legislationlies (according to the first principle) in the ruleand its form of universality which makes itcapable of being a law (say, e. g., a law of na-ture); but the subjective principle is in the end;now by the second principle the subject of allends is each rational being, inasmuch as it is an

end in itself. Hence follows the third practicalprinciple of the will, which is the ultimate con-dition of its harmony with universal practicalreason, viz.: the idea of the will of every ra-tional being as a universally legislative will.

On this principle all maxims are rejected whichare inconsistent with the will being itself uni-versal legislator. Thus the will is not subjectsimply to the law, but so subject that it must beregarded as itself giving the law and, on thisground only, subject to the law (of which it canregard itself as the author).

In the previous imperatives, namely, that basedon the conception of the conformity of actionsto general laws, as in a physical system of na-ture, and that based on the universal preroga-tive of rational beings as ends in themselves-these imperatives, just because they were con-ceived as categorical, excluded from any sharein their authority all admixture of any interestas a spring of action; they were, however, only

assumed to be categorical, because such an as-sumption was necessary to explain the concep-tion of duty. But we could not prove independ-ently that there are practical propositionswhich command categorically, nor can it beproved in this section; one thing, however,could be done, namely, to indicate in the im-perative itself, by some determinate expression,that in the case of volition from duty all interestis renounced, which is the specific criterion ofcategorical as distinguished from hypotheticalimperatives. This is done in the present (third)formula of the principle, namely, in the idea ofthe will of every rational being as a universallylegislating will.

For although a will which is subject to lawsmay be attached to this law by means of aninterest, yet a will which is itself a supremelawgiver so far as it is such cannot possiblydepend on any interest, since a will so depend-ent would itself still need another law restrict-

ing the interest of its self-love by the conditionthat it should be valid as universal law.

Thus the principle that every human will is awill which in all its maxims gives universallaws, * provided it be otherwise justified,would be very well adapted to be the categori-cal imperative, in this respect, namely, that justbecause of the idea of universal legislation it isnot based on interest, and therefore it aloneamong all possible imperatives can be uncondi-tional. Or still better, converting the proposi-tion, if there is a categorical imperative (i.e., alaw for the will of every rational being), it canonly command that everything be done frommaxims of one's will regarded as a will whichcould at the same time will that it should itselfgive universal laws, for in that case only thepractical principle and the imperative which itobeys are unconditional, since they cannot bebased on any interest.

* I may be excused from adducing examples toelucidate this principle, as those which havealready been used to elucidate the categoricalimperative and its formula would all serve forthe like purpose here.

Looking back now on all previous attempts todiscover the principle of morality, we need notwonder why they all failed. It was seen thatman was bound to laws by duty, but it was notobserved that the laws to which he is subjectare only those of his own giving, though at thesame time they are universal, and that he isonly bound to act in conformity with his ownwill; a will, however, which is designed by na-ture to give universal laws. For when one hasconceived man only as subject to a law (no mat-ter what), then this law required some interest,either by way of attraction or constraint, since itdid not originate as a law from his own will,

but this will was according to a law obliged bysomething else to act in a certain manner. Nowby this necessary consequence all the labourspent in finding a supreme principle of dutywas irrevocably lost. For men never elicitedduty, but only a necessity of acting from a cer-tain interest. Whether this interest was privateor otherwise, in any case the imperative mustbe conditional and could not by any means becapable of being a moral command. I will there-fore call this the principle of autonomy of thewill, in contrast with every other which I ac-cordingly reckon as heteronomy.

The conception of the will of every rationalbeing as one which must consider itself as giv-ing in all the maxims of its will universal laws,so as to judge itself and its actions from thispoint of view- this conception leads to anotherwhich depends on it and is very fruitful, name-ly that of a kingdom of ends.

By a kingdom I understand the union of differ-ent rational beings in a system by commonlaws. Now since it is by laws that ends are de-termined as regards their universal validity,hence, if we abstract from the personal differ-ences of rational beings and likewise from allthe content of their private ends, we shall beable to conceive all ends combined in a system-atic whole (including both rational beings asends in themselves, and also the special endswhich each may propose to himself), that is tosay, we can conceive a kingdom of ends, whichon the preceding principles is possible.

For all rational beings come under the law thateach of them must treat itself and all othersnever merely as means, but in every case at thesame time as ends in themselves. Hence resultsa systematic union of rational being by com-mon objective laws, i.e., a kingdom which maybe called a kingdom of ends, since what theselaws have in view is just the relation of these

beings to one another as ends and means. It iscertainly only an ideal.

A rational being belongs as a member to thekingdom of ends when, although giving uni-versal laws in it, he is also himself subject tothese laws. He belongs to it as sovereign when,while giving laws, he is not subject to the willof any other.

A rational being must always regard himself asgiving laws either as member or as sovereign ina kingdom of ends which is rendered possibleby the freedom of will. He cannot, however,maintain the latter position merely by the max-ims of his will, but only in case he is a com-pletely independent being without wants andwith unrestricted power adequate to his will.

Morality consists then in the reference of allaction to the legislation which alone can rendera kingdom of ends possible. This legislationmust be capable of existing in every rational

being and of emanating from his will, so thatthe principle of this will is never to act on anymaxim which could not without contradictionbe also a universal law and, accordingly, al-ways so to act that the will could at the sametime regard itself as giving in its maxims uni-versal laws. If now the maxims of rational be-ings are not by their own nature coincidentwith this objective principle, then the necessityof acting on it is called practical necessitation,i.e., duty. Duty does not apply to the sovereignin the kingdom of ends, but it does to everymember of it and to all in the same degree.

The practical necessity of acting on this princi-ple, i.e., duty, does not rest at all on feelings,impulses, or inclinations, but solely on the rela-tion of rational beings to one another, a relationin which the will of a rational being must al-ways be regarded as legislative, since otherwiseit could not be conceived as an end in itself.Reason then refers every maxim of the will,

regarding it as legislating universally, to everyother will and also to every action towards one-self; and this not on account of any other prac-tical motive or any future advantage, but fromthe idea of the dignity of a rational being, obey-ing no law but that which he himself also gives.

In the kingdom of ends everything has eithervalue or dignity. Whatever has a value can bereplaced by something else which is equivalent;whatever, on the other hand, is above all value,and therefore admits of no equivalent, has adignity.

Whatever has reference to the general inclina-tions and wants of mankind has a market va-lue; whatever, without presupposing a want,corresponds to a certain taste, that is to a satis-faction in the mere purposeless play of our fac-ulties, has a fancy value; but that which consti-tutes the condition under which alone anythingcan be an end in itself, this has not merely a

relative worth, i.e., value, but an intrinsicworth, that is, dignity.

Now morality is the condition under whichalone a rational being can be an end in himself,since by this alone is it possible that he shouldbe a legislating member in the kingdom ofends. Thus morality, and humanity as capableof it, is that which alone has dignity. Skill anddiligence in labour have a market value; wit,lively imagination, and humour, have fancyvalue; on the other hand, fidelity to promises,benevolence from principle (not from instinct),have an intrinsic worth. Neither nature nor artcontains anything which in default of these itcould put in their place, for their worth consistsnot in the effects which spring from them, notin the use and advantage which they secure,but in the disposition of mind, that is, the max-ims of the will which are ready to manifestthemselves in such actions, even though theyshould not have the desired effect. These ac-

tions also need no recommendation from anysubjective taste or sentiment, that they may belooked on with immediate favour and satisfac-tion: they need no immediate propension orfeeling for them; they exhibit the will that per-forms them as an object of an immediate re-spect, and nothing but reason is required toimpose them on the will; not to flatter it intothem, which, in the case of duties, would be acontradiction. This estimation therefore showsthat the worth of such a disposition is dignity,and places it infinitely above all value, withwhich it cannot for a moment be brought intocomparison or competition without as it wereviolating its sanctity.

What then is it which justifies virtue or the mo-rally good disposition, in making such loftyclaims? It is nothing less than the privilege itsecures to the rational being of participating inthe giving of universal laws, by which it quali-fies him to be a member of a possible kingdom

of ends, a privilege to which he was alreadydestined by his own nature as being an end inhimself and, on that account, legislating in thekingdom of ends; free as regards all laws ofphysical nature, and obeying those only whichhe himself gives, and by which his maxims canbelong to a system of universal law, to which atthe same time he submits himself. For nothinghas any worth except what the law assigns it.Now the legislation itself which assigns theworth of everything must for that very reasonpossess dignity, that is an unconditional in-comparable worth; and the word respect alonesupplies a becoming expression for the esteemwhich a rational being must have for it. Auton-omy then is the basis of the dignity of humanand of every rational nature.

The three modes of presenting the principle ofmorality that have been adduced are at bottomonly so many formulae of the very same law,and each of itself involves the other two. There

is, however, a difference in them, but it is rathersubjectively than objectively practical, intendednamely to bring an idea of the reason nearer tointuition (by means of a certain analogy) andthereby nearer to feeling. All maxims, in fact,have:

1. A form, consisting in universality; and in thisview the formula of the moral imperative isexpressed thus, that the maxims must be sochosen as if they were to serve as universallaws of nature.

2. A matter, namely, an end, and here the for-mula says that the rational being, as it is an endby its own nature and therefore an end in itself,must in every maxim serve as the conditionlimiting all merely relative and arbitrary ends.

3. A complete characterization of all maxims bymeans of that formula, namely, that all maximsought by their own legislation to harmonizewith a possible kingdom of ends as with a king-

dom of nature. * There is a progress here in theorder of the categories of unity of the form ofthe will (its universality), plurality of the matter(the objects, i.e., the ends), and totality of thesystem of these. In forming our moral judge-ment of actions, it is better to proceed alwayson the strict method and start from the generalformula of the categorical imperative: Act ac-cording to a maxim which can at the same timemake itself a universal law. If, however, wewish to gain an entrance for the moral law, it isvery useful to bring one and the same actionunder the three specified conceptions, and the-reby as far as possible to bring it nearer to intui-tion.

* Teleology considers nature as a kingdom ofends; ethics regards a possible kingdom of endsas a kingdom nature. In the first case, the king-dom of ends is a theoretical idea, adopted to

explain what actually is. In the latter it is apractical idea, adopted to bring about thatwhich is not yet, but which can be realized byour conduct, namely, if it conforms to this idea.

We can now end where we started at the be-ginning, namely, with the conception of a willunconditionally good. That will is absolutelygood which cannot be evil- in other words,whose maxim, if made a universal law, couldnever contradict itself. This principle, then, isits supreme law: "Act always on such a maximas thou canst at the same time will to be a uni-versal law"; this is the sole condition underwhich a will can never contradict itself; andsuch an imperative is categorical. Since the va-lidity of the will as a universal law for possibleactions is analogous to the universal connexionof the existence of things by general laws,which is the formal notion of nature in general,

the categorical imperative can also be ex-pressed thus: Act on maxims which can at thesame time have for their object themselves asuniversal laws of nature. Such then is the for-mula of an absolutely good will.

Rational nature is distinguished from the rest ofnature by this, that it sets before itself an end.This end would be the matter of every goodwill. But since in the idea of a will that is abso-lutely good without being limited by any con-dition (of attaining this or that end) we mustabstract wholly from every end to be effected(since this would make every will only rela-tively good), it follows that in this case the endmust be conceived, not as an end to be effected,but as an independently existing end. Conse-quently it is conceived only negatively, i.e., asthat which we must never act against andwhich, therefore, must never be regarded mere-ly as means, but must in every volition be es-teemed as an end likewise. Now this end can be

nothing but the subject of all possible ends,since this is also the subject of a possible abso-lutely good will; for such a will cannot withoutcontradiction be postponed to any other object.The principle: "So act in regard to every ra-tional being (thyself and others), that he mayalways have place in thy maxim as an end inhimself," is accordingly essentially identicalwith this other: "Act upon a maxim which, atthe same time, involves its own universal valid-ity for every rational being." For that in usingmeans for every end I should limit my maximby the condition of its holding good as a law forevery subject, this comes to the same thing asthat the fundamental principle of all maxims ofaction must be that the subject of all ends, i.e.,the rational being himself, be never employedmerely as means, but as the supreme conditionrestricting the use of all means, that is in everycase as an end likewise.

It follows incontestably that, to whatever lawsany rational being may be subject, he being anend in himself must be able to regard himself asalso legislating universally in respect of thesesame laws, since it is just this fitness of his max-ims for universal legislation that distinguisheshim as an end in himself; also it follows thatthis implies his dignity (prerogative) above allmere physical beings, that he must always takehis maxims from the point of view which re-gards himself and, likewise, every other ra-tional being as law-giving beings (on whichaccount they are called persons). In this way aworld of rational beings (mundus intelligibilis)is possible as a kingdom of ends, and this byvirtue of the legislation proper to all persons asmembers. Therefore every rational being mustso act as if he were by his maxims in every casea legislating member in the universal kingdomof ends. The formal principle of these maximsis: "So act as if thy maxim were to serve like-wise as the universal law (of all rational be-

ings)." A kingdom of ends is thus only possibleon the analogy of a kingdom of nature, theformer however only by maxims, that is self-imposed rules, the latter only by the laws ofefficient causes acting under necessitation fromwithout. Nevertheless, although the system ofnature is looked upon as a machine, yet so faras it has reference to rational beings as its ends,it is given on this account the name of a king-dom of nature. Now such a kingdom of endswould be actually realized by means of maximsconforming to the canon which the categoricalimperative prescribes to all rational beings, ifthey were universally followed. But although arational being, even if he punctually followsthis maxim himself, cannot reckon upon allothers being therefore true to the same, norexpect that the kingdom of nature and its or-derly arrangements shall be in harmony withhim as a fitting member, so as to form a king-dom of ends to which he himself contributes,that is to say, that it shall favour his expectation

of happiness, still that law: "Act according tothe maxims of a member of a merely possiblekingdom of ends legislating in it universally,"remains in its full force, inasmuch as it com-mands categorically. And it is just in this thatthe paradox lies; that the mere dignity of manas a rational creature, without any other end oradvantage to be attained thereby, in otherwords, respect for a mere idea, should yet serveas an inflexible precept of the will, and that it isprecisely in this independence of the maxim onall such springs of action that its sublimity con-sists; and it is this that makes every rationalsubject worthy to be a legislative member in thekingdom of ends: for otherwise he would haveto be conceived only as subject to the physicallaw of his wants. And although we should sup-pose the kingdom of nature and the kingdomof ends to be united under one sovereign, sothat the latter kingdom thereby ceased to be amere idea and acquired true reality, then itwould no doubt gain the accession of a strong

spring, but by no means any increase of its in-trinsic worth. For this sole absolute lawgivermust, notwithstanding this, be always con-ceived as estimating the worth of rational be-ings only by their disinterested behaviour, asprescribed to themselves from that idea [thedignity of man] alone. The essence of things isnot altered by their external relations, and thatwhich, abstracting from these, alone constitutesthe absolute worth of man, is also that bywhich he must be judged, whoever the judgemay be, and even by the Supreme Being. Mo-rality, then, is the relation of actions to the rela-tion of actions will, that is, to the autonomy ofpotential universal legislation by its maxims.An action that is consistent with the autonomyof the will is permitted; one that does not agreetherewith is forbidden. A will whose maximsnecessarily coincide with the laws of autonomyis a holy will, good absolutely. The dependenceof a will not absolutely good on the principle ofautonomy (moral necessitation) is obligation.

This, then, cannot be applied to a holy being.The objective necessity of actions from obliga-tion is called duty.

From what has just been said, it is easy to seehow it happens that, although the conception ofduty implies subjection to the law, we yet as-cribe a certain dignity and sublimity to the per-son who fulfils all his duties. There is not, in-deed, any sublimity in him, so far as he is sub-ject to the moral law; but inasmuch as in regardto that very law he is likewise a legislator, andon that account alone subject to it, he has sub-limity. We have also shown above that neitherfear nor inclination, but simply respect for thelaw, is the spring which can give actions amoral worth. Our own will, so far as we sup-pose it to act only under the condition that itsmaxims are potentially universal laws, thisideal will which is possible to us is the properobject of respect; and the dignity of humanityconsists just in this capacity of being univer-

sally legislative, though with the condition thatit is itself subject to this same legislation.

The Autonomy of the Will as the Supreme Prin-ciple of Morality

Autonomy of the will is that property of it bywhich it is a law to itself (independently of anyproperty of the objects of volition). The princi-ple of autonomy then is: "Always so to choosethat the same volition shall comprehend themaxims of our choice as a universal law." Wecannot prove that this practical rule is an im-perative, i.e., that the will of every rational be-ing is necessarily bound to it as a condition, bya mere analysis of the conceptions which occurin it, since it is a synthetical proposition; wemust advance beyond the cognition of the ob-

jects to a critical examination of the subject, thatis, of the pure practical reason, for this syntheticproposition which commands apodeicticallymust be capable of being cognized wholly apriori. This matter, however, does not belong tothe present section. But that the principle ofautonomy in question is the sole principle ofmorals can be readily shown by mere analysisof the conceptions of morality. For by this ana-lysis we find that its principle must be a cate-gorical imperative and that what this com-mands is neither more nor less than this veryautonomy.

Heteronomy of the Will as the Source of allspurious Principles

of Morality

If the will seeks the law which is to determine itanywhere else than in the fitness of its maximsto be universal laws of its own dictation, conse-quently if it goes out of itself and seeks this lawin the character of any of its objects, there al-ways results heteronomy. The will in that casedoes not give itself the law, but it is given bythe object through its relation to the will. Thisrelation, whether it rests on inclination or onconceptions of reason, only admits of hypo-thetical imperatives: "I ought to do somethingbecause I wish for something else." On the con-trary, the moral, and therefore categorical, im-perative says: "I ought to do so and so, eventhough I should not wish for anything else."E.g., the former says: "I ought not to lie, if Iwould retain my reputation"; the latter says: "Iought not to lie, although it should not bringme the least discredit." The latter thereforemust so far abstract from all objects that theyshall have no influence on the will, in order thatpractical reason (will) may not be restricted to

administering an interest not belonging to it,but may simply show its own commandingauthority as the supreme legislation. Thus, e.g.,I ought to endeavour to promote the happinessof others, not as if its realization involved anyconcern of mine (whether by immediate incli-nation or by any satisfaction indirectly gainedthrough reason), but simply because a maximwhich excludes it cannot be comprehended as auniversal law in one and the same volition.

Classification of all Principles of Moralitywhich can be

founded on the Conception of Heteronomy

Here as elsewhere human reason in its pureuse, so long as it was not critically examined,

has first tried all possible wrong ways before itsucceeded in finding the one true way.

All principles which can be taken from thispoint of view are either empirical or rational.The former, drawn from the principle of happi-ness, are built on physical or moral feelings; thelatter, drawn from the principle of perfection,are built either on the rational conception ofperfection as a possible effect, or on that of anindependent perfection (the will of God) as thedetermining cause of our will.

Empirical principles are wholly incapable ofserving as a foundation for moral laws. For theuniversality with which these should hold forall rational beings without distinction, the un-conditional practical necessity which is therebyimposed on them, is lost when their foundationis taken from the particular constitution of hu-man nature, or the accidental circumstances inwhich it is placed. The principle of privatehappiness, however, is the most objectionable,

not merely because it is false, and experiencecontradicts the supposition that prosperity isalways proportioned to good conduct, nor yetmerely because it contributes nothing to theestablishment of morality- since it is quite adifferent thing to make a prosperous man and agood man, or to make one prudent and sharp-sighted for his own interests and to make himvirtuous- but because the springs it providesfor morality are such as rather undermine itand destroy its sublimity, since they put themotives to virtue and to vice in the same classand only teach us to make a better calculation,the specific difference between virtue and vicebeing entirely extinguished. On the other hand,as to moral feeling, this supposed special sense,* the appeal to it is indeed superficial whenthose who cannot think believe that feeling willhelp them out, even in what concerns generallaws: and besides, feelings, which naturallydiffer infinitely in degree, cannot furnish a uni-form standard of good and evil, nor has anyone

a right to form judgements for others by hisown feelings: nevertheless this moral feeling isnearer to morality and its dignity in this re-spect, that it pays virtue the honour of ascribingto her immediately the satisfaction and esteemwe have for her and does not, as it were, tellher to her face that we are not attached to herby her beauty but by profit.

* I class the principle of moral feeling underthat of happiness, because every empirical in-terest promises to contribute to our well-beingby the agreeableness that a thing affords, whet-her it be immediately and without a view toprofit, or whether profit be regarded. We mustlikewise, with Hutcheson, class the principle ofsympathy with the happiness of others underhis assumed moral sense.

Amongst the rational principles of morality, theontological conception of perfection, notwith-standing its defects, is better than the theologi-cal conception which derives morality from aDivine absolutely perfect will. The former is, nodoubt, empty and indefinite and consequentlyuseless for finding in the boundless field ofpossible reality the greatest amount suitable forus; moreover, in attempting to distinguish spe-cifically the reality of which we are now speak-ing from every other, it inevitably tends to turnin a circle and cannot avoid tacitly presuppos-ing the morality which it is to explain; it is nev-ertheless preferable to the theological view,first, because we have no intuition of the divineperfection and can only deduce it from our ownconceptions, the most important of which isthat of morality, and our explanation wouldthus be involved in a gross circle; and, in thenext place, if we avoid this, the only notion ofthe Divine will remaining to us is a conceptionmade up of the attributes of desire of glory and

dominion, combined with the awful concep-tions of might and vengeance, and any systemof morals erected on this foundation would bedirectly opposed to morality.

However, if I had to choose between the notionof the moral sense and that of perfection in ge-neral (two systems which at least do not wea-ken morality, although they are totally incapa-ble of serving as its foundation), then I shoulddecide for the latter, because it at least with-draws the decision of the question from thesensibility and brings it to the court of purereason; and although even here it decides noth-ing, it at all events preserves the indefinite idea(of a will good in itself free from corruption,until it shall be more precisely defined.

For the rest I think I may be excused here froma detailed refutation of all these doctrines; thatwould only be superfluous labour, since it is soeasy, and is probably so well seen even by tho-se whose office requires them to decide for one

of these theories (because their hearers wouldnot tolerate suspension of judgement). Butwhat interests us more here is to know that theprime foundation of morality laid down by allthese principles is nothing but heteronomy ofthe will, and for this reason they must necessar-ily miss their aim.

In every case where an object of the will has tobe supposed, in order that the rule may be pre-scribed which is to determine the will, there therule is simply heteronomy; the imperative isconditional, namely, if or because one wishesfor this object, one should act so and so: henceit can never command morally, that is, cate-gorically. Whether the object determines thewill by means of inclination, as in the principleof private happiness, or by means of reasondirected to objects of our possible volition gen-erally, as in the principle of perfection, in eithercase the will never determines itself immedi-ately by the conception of the action, but only

by the influence which the foreseen effect of theaction has on the will; I ought to do something,on this account, because I wish for somethingelse; and here there must be yet another lawassumed in me as its subject, by which I neces-sarily will this other thing, and this law againrequires an imperative to restrict this maxim.For the influence which the conception of anobject within the reach of our faculties can ex-ercise on the will of the subject, in consequenceof its natural properties, depends on the natureof the subject, either the sensibility (inclinationand taste), or the understanding and reason,the employment of which is by the peculiarconstitution of their nature attended with satis-faction. It follows that the law would be, prop-erly speaking, given by nature, and, as such, itmust be known and proved by experience andwould consequently be contingent and there-fore incapable of being an apodeictic practicalrule, such as the moral rule must be. Not onlyso, but it is inevitably only heteronomy; the will

does not give itself the law, but is given by aforeign impulse by means of a particular natu-ral constitution of the subject adapted to receiveit. An absolutely good will, then, the principleof which must be a categorical imperative, willbe indeterminate as regards all objects and willcontain merely the form of volition generally,and that as autonomy, that is to say, the capa-bility of the maxims of every good will to makethemselves a universal law, is itself the onlylaw which the will of every rational being im-poses on itself, without needing to assume anyspring or interest as a foundation.

How such a synthetical practical a priori pro-position is possible, and why it is necessary, is aproblem whose solution does not lie within thebounds of the metaphysic of morals; and wehave not here affirmed its truth, much less pro-fessed to have a proof of it in our power. Wesimply showed by the development of the uni-versally received notion of morality that an

autonomy of the will is inevitably connectedwith it, or rather is its foundation. Whoeverthen holds morality to be anything real, and nota chimerical idea without any truth, must like-wise admit the principle of it that is here as-signed. This section then, like the first, wasmerely analytical. Now to prove that moralityis no creation of the brain, which it cannot be ifthe categorical imperative and with it theautonomy of the will is true, and as an a prioriprinciple absolutely necessary, this supposesthe possibility of a synthetic use of pure practi-cal reason, which however we cannot ventureon without first giving a critical examination ofthis faculty of reason. In the concluding sectionwe shall give the principal outlines of this criti-cal examination as far as is sufficient for ourpurpose.

SEC_3

THIRD SECTION

TRANSITION FROM THE METAPHYSICOF MORALS TO THE

CRITIQUE OF PURE PRACTICAL REA-SON

The Concept of Freedom is the Key that ex-plains the Autonomy of the Will

The will is a kind of causality belonging to liv-ing beings in so far as they are rational, andfreedom would be this property of such causal-ity that it can be efficient, independently of for-eign causes determining it; just as physical ne-cessity is the property that the causality of allirrational beings has of being determined toactivity by the influence of foreign causes.

The preceding definition of freedom is negativeand therefore unfruitful for the discovery of itsessence, but it leads to a positive conceptionwhich is so much the more full and fruitful.

Since the conception of causality involves thatof laws, according to which, by something thatwe call cause, something else, namely the ef-fect, must be produced; hence, although free-dom is not a property of the will depending onphysical laws, yet it is not for that reason law-less; on the contrary it must be a causality act-ing according to immutable laws, but of a pecu-liar kind; otherwise a free will would be anabsurdity. Physical necessity is a heteronomy ofthe efficient causes, for every effect is possibleonly according to this law, that something elsedetermines the efficient cause to exert its cau-sality. What else then can freedom of the willbe but autonomy, that is, the property of thewill to be a law to itself? But the proposition:"The will is in every action a law to itself," only

expresses the principle: "To act on no othermaxim than that which can also have as an ob-ject itself as a universal law." Now this is pre-cisely the formula of the categorical imperativeand is the principle of morality, so that a freewill and a will subject to moral laws are oneand the same.

On the hypothesis, then, of freedom of the will,morality together with its principle followsfrom it by mere analysis of the conception.However, the latter is a synthetic proposition;viz., an absolutely good will is that whosemaxim can always include itself regarded as auniversal law; for this property of its maximcan never be discovered by analysing the con-ception of an absolutely good will. Now suchsynthetic propositions are only possible in thisway: that the two cognitions are connected to-gether by their union with a third in which theyare both to be found. The positive concept offreedom furnishes this third cognition, which

cannot, as with physical causes, be the nature ofthe sensible world (in the concept of which wefind conjoined the concept of something in rela-tion as cause to something else as effect). Wecannot now at once show what this third is towhich freedom points us and of which we havean idea a priori, nor can we make intelligiblehow the concept of freedom is shown to be le-gitimate from principles of pure practical rea-son and with it the possibility of a categoricalimperative; but some further preparation isrequired.

Freedom must be presupposed as a Property ofthe Will

of all Rational Beings

It is not enough to predicate freedom of ourown will, from Whatever reason, if we have notsufficient grounds for predicating the same ofall rational beings. For as morality serves as alaw for us only because we are rational beings,it must also hold for all rational beings; and asit must be deduced simply from the property offreedom, it must be shown that freedom also isa property of all rational beings. It is notenough, then, to prove it from certain supposedexperiences of human nature (which indeed isquite impossible, and it can only be shown apriori), but we must show that it belongs to theactivity of all rational beings endowed with awill. Now I say every being that cannot act ex-cept under the idea of freedom is just for thatreason in a practical point of view really free,that is to say, all laws which are inseparablyconnected with freedom have the same forcefor him as if his will had been shown to be freein itself by a proof theoretically conclusive. *Now I affirm that we must attribute to every

rational being which has a will that it has alsothe idea of freedom and acts entirely under thisidea. For in such a being we conceive a reasonthat is practical, that is, has causality in refer-ence to its objects. Now we cannot possiblyconceive a reason consciously receiving a biasfrom any other quarter with respect to itsjudgements, for then the subject would ascribethe determination of its judgement not to itsown reason, but to an impulse. It must regarditself as the author of its principles independentof foreign influences. Consequently as practicalreason or as the will of a rational being it mustregard itself as free, that is to say, the will ofsuch a being cannot be a will of its own exceptunder the idea of freedom. This idea must the-refore in a practical point of view be ascribed toevery rational being.

* I adopt this method of assuming freedom me-rely as an idea which rational beings supposein their actions, in order to avoid the necessityof proving it in its theoretical aspect also. Theformer is sufficient for my purpose; for eventhough the speculative proof should not bemade out, yet a being that cannot act exceptwith the idea of freedom is bound by the samelaws that would oblige a being who was actu-ally free. Thus we can escape here from theonus which presses on the theory.

Of the Interest attaching to the Ideas of Moral-ity

We have finally reduced the definite concep-tion of morality to the idea of freedom. Thislatter, however, we could not prove to be actu-

ally a property of ourselves or of human na-ture; only we saw that it must be presupposedif we would conceive a being as rational andconscious of its causality in respect of its ac-tions, i.e., as endowed with a will; and so wefind that on just the same grounds we mustascribe to every being endowed with reasonand will this attribute of determining itself toaction under the idea of its freedom.

Now it resulted also from the presupposition ofthese ideas that we became aware of a law thatthe subjective principles of action, i.e., maxims,must always be so assumed that they can alsohold as objective, that is, universal principles,and so serve as universal laws of our own dic-tation. But why then should I subject myself tothis principle and that simply as a rational be-ing, thus also subjecting to it all other beingendowed with reason? I will allow that no in-terest urges me to this, for that would not givea categorical imperative, but I must take an

interest in it and discern how this comes topass; for this properly an "I ought" is properlyan "I would," valid for every rational being,provided only that reason determined his ac-tions without any hindrance. But for beingsthat are in addition affected as we are bysprings of a different kind, namely, sensibility,and in whose case that is not always donewhich reason alone would do, for these thatnecessity is expressed only as an "ought," andthe subjective necessity is different from theobjective.

It seems then as if the moral law, that is, theprinciple of autonomy of the will, were prop-erly speaking only presupposed in the idea offreedom, and as if we could not prove its real-ity and objective necessity independently. Inthat case we should still have gained somethingconsiderable by at least determining the trueprinciple more exactly than had previouslybeen done; but as regards its validity and the

practical necessity of subjecting oneself to it, weshould not have advanced a step. For if we we-re asked why the universal validity of ourmaxim as a law must be the condition restrict-ing our actions, and on what we ground theworth which we assign to this manner of act-ing- a worth so great that there cannot be anyhigher interest; and if we were asked furtherhow it happens that it is by this alone a manbelieves he feels his own personal worth, incomparison with which that of an agreeable ordisagreeable condition is to be regarded as not-hing, to these questions we could give no satis-factory answer.

We find indeed sometimes that we can take aninterest in a personal quality which does notinvolve any interest of external condition, pro-vided this quality makes us capable of partici-pating in the condition in case reason were toeffect the allotment; that is to say, the mere be-ing worthy of happiness can interest of itself

even without the motive of participating in thishappiness. This judgement, however, is in factonly the effect of the importance of the morallaw which we before presupposed (when bythe idea of freedom we detach ourselves fromevery empirical interest); but that we ought todetach ourselves from these interests, i.e., toconsider ourselves as free in action and yet assubject to certain laws, so as to find a worthsimply in our own person which can compen-sate us for the loss of everything that givesworth to our condition; this we are not yet ableto discern in this way, nor do we see how it ispossible so to act- in other words, whence themoral law derives its obligation.

It must be freely admitted that there is a sort ofcircle here from which it seems impossible toescape. In the order of efficient causes we as-sume ourselves free, in order that in the orderof ends we may conceive ourselves as subject tomoral laws: and we afterwards conceive our-

selves as subject to these laws, because we haveattributed to ourselves freedom of will: for free-dom and self-legislation of will are both auton-omy and, therefore, are reciprocal conceptions,and for this very reason one must not be usedto explain the other or give the reason of it, butat most only logical purposes to reduce appar-ently different notions of the same object to onesingle concept (as we reduce different fractionsof the same value to the lowest terms).

One resource remains to us, namely, to inquirewhether we do not occupy different points ofview when by means of freedom we think our-selves as causes efficient a priori, and when weform our conception of ourselves from our ac-tions as effects which we see before our eyes.

It is a remark which needs no subtle reflectionto make, but which we may assume that eventhe commonest understanding can make, al-though it be after its fashion by an obscure dis-cernment of judgement which it calls feeling,

that all the "ideas" that come to us involuntarily(as those of the senses) do not enable us toknow objects otherwise than as they affect us;so that what they may be in themselves re-mains unknown to us, and consequently that asregards "ideas" of this kind even with the clos-est attention and clearness that the understand-ing can apply to them, we can by them onlyattain to the knowledge of appearances, neverto that of things in themselves. As soon as thisdistinction has once been made (perhaps mere-ly in consequence of the difference observedbetween the ideas given us from without, andin which we are passive, and those that weproduce simply from ourselves, and in whichwe show our own activity), then it follows ofitself that we must admit and assume behindthe appearance something else that is not anappearance, namely, the things in themselves;although we must admit that as they can neverbe known to us except as they affect us, we cancome no nearer to them, nor can we ever know

what they are in themselves. This must furnisha distinction, however crude, between a worldof sense and the world of understanding, ofwhich the former may be different according tothe difference of the sensuous impressions invarious observers, while the second which is itsbasis always remains the same, Even as to him-self, a man cannot pretend to know what he isin himself from the knowledge he has by inter-nal sensation. For as he does not as it were cre-ate himself, and does not come by the concep-tion of himself a priori but empirically, it natu-rally follows that he can obtain his knowledgeeven of himself only by the inner sense and,consequently, only through the appearances ofhis nature and the way in which his conscious-ness is affected. At the same time beyond thesecharacteristics of his own subject, made up ofmere appearances, he must necessarily supposesomething else as their basis, namely, his ego,whatever its characteristics in itself may be.Thus in respect to mere perception and recep-

tivity of sensations he must reckon himself asbelonging to the world of sense; but in respectof whatever there may be of pure activity inhim (that which reaches consciousness imme-diately and not through affecting the senses),he must reckon himself as belonging to the in-tellectual world, of which, however, he has nofurther knowledge. To such a conclusion thereflecting man must come with respect to allthe things which can be presented to him: it isprobably to be met with even in persons of thecommonest understanding, who, as is wellknown, are very much inclined to suppose be-hind the objects of the senses something elseinvisible and acting of itself. They spoil it, how-ever, by presently sensualizing this invisibleagain; that is to say, wanting to make it an ob-ject of intuition, so that they do not become awhit the wiser.

Now man really finds in himself a faculty bywhich he distinguishes himself from everything

else, even from himself as affected by objects,and that is reason. This being pure spontaneityis even elevated above the understanding. Foralthough the latter is a spontaneity and doesnot, like sense, merely contain intuitions thatarise when we are affected by things (and aretherefore passive), yet it cannot produce fromits activity any other conceptions than thosewhich merely serve to bring the intuitions ofsense under rules and, thereby, to unite them inone consciousness, and without this use of thesensibility it could not think at all; whereas, onthe contrary, reason shows so pure a spontane-ity in the case of what I call ideas [ideal concep-tions] that it thereby far transcends everythingthat the sensibility can give it, and exhibits itsmost important function in distinguishing theworld of sense from that of understanding, andthereby prescribing the limits of the under-standing itself.

For this reason a rational being must regardhimself qua intelligence (not from the side ofhis lower faculties) as belonging not to theworld of sense, but to that of understanding;hence he has two points of view from which hecan regard himself, and recognise laws of theexercise of his faculties, and consequently of allhis actions: first, so far as he belongs to theworld of sense, he finds himself subject to lawsof nature (heteronomy); secondly, as belongingto the intelligible world, under laws which be-ing independent of nature have their founda-tion not in experience but in reason alone.

As a rational being, and consequently belong-ing to the intelligible world, man can neverconceive the causality of his own will otherwisethan on condition of the idea of freedom, forindependence of the determinate causes of thesensible world (an independence which reasonmust always ascribe to itself) is freedom. Nowthe idea of freedom is inseparably connected

with the conception of autonomy, and thisagain with the universal principle of moralitywhich is ideally the foundation of all actions ofrational beings, just as the law of nature is of allphenomena.

Now the suspicion is removed which we raisedabove, that there was a latent circle involved inour reasoning from freedom to autonomy, andfrom this to the moral law, viz.: that we laiddown the idea of freedom because of the morallaw only that we might afterwards in turn inferthe latter from freedom, and that consequentlywe could assign no reason at all for this law,but could only [present] it as a petitio principiiwhich well disposed minds would gladly con-cede to us, but which we could never put for-ward as a provable proposition. For now wesee that, when we conceive ourselves as free,we transfer ourselves into the world of under-standing as members of it and recognise theautonomy of the will with its consequence, mo-

rality; whereas, if we conceive ourselves as un-der obligation, we consider ourselves as be-longing to the world of sense and at the sametime to the world of understanding.

How is a Categorical Imperative Possible?

Every rational being reckons himself qua intel-ligence as belonging to the world of under-standing, and it is simply as an efficient causebelonging to that world that he calls his causal-ity a will. On the other side he is also consciousof himself as a part of the world of sense inwhich his actions, which are mere appearances[phenomena] of that causality, are displayed;we cannot, however, discern how they are pos-sible from this causality which we do not know;but instead of that, these actions as belonging

to the sensible world must be viewed as deter-mined by other phenomena, namely, desiresand inclinations. If therefore I were only amember of the world of understanding, then allmy actions would perfectly conform to theprinciple of autonomy of the pure will; if I wereonly a part of the world of sense, they wouldnecessarily be assumed to conform wholly tothe natural law of desires and inclinations, inother words, to the heteronomy of nature. (Theformer would rest on morality as the supremeprinciple, the latter on happiness.) Since, how-ever, the world of understanding contains thefoundation of the world of sense, and conse-quently of its laws also, and accordingly givesthe law to my will (which belongs wholly to theworld of understanding) directly, and must beconceived as doing so, it follows that, althoughon the one side I must regard myself as a beingbelonging to the world of sense, yet on theother side I must recognize myself as subject asan intelligence to the law of the world of un-

derstanding, i.e., to reason, which contains thislaw in the idea of freedom, and therefore assubject to the autonomy of the will: conse-quently I must regard the laws of the world ofunderstanding as imperatives for me and theactions which conform to them as duties.

And thus what makes categorical imperativespossible is this, that the idea of freedom makesme a member of an intelligible world, in conse-quence of which, if I were nothing else, all myactions would always conform to the autonomyof the will; but as I at the same time intuite my-self as a member of the world of sense, theyought so to conform, and this categorical"ought" implies a synthetic a priori proposition,inasmuch as besides my will as affected by sen-sible desires there is added further the idea ofthe same will but as belonging to the world ofthe understanding, pure and practical of itself,which contains the supreme condition accord-ing to reason of the former will; precisely as to

the intuitions of sense there are added conceptsof the understanding which of themselves sig-nify nothing but regular form in general and inthis way synthetic a priori propositions becomepossible, on which all knowledge of physicalnature rests.

The practical use of common human reasonconfirms this reasoning. There is no one, noteven the most consummate villain, providedonly that he is otherwise accustomed to the useof reason, who, when we set before him exam-ples of honesty of purpose, of steadfastness infollowing good maxims, of sympathy and gen-eral benevolence (even combined with greatsacrifices of advantages and comfort), does notwish that he might also possess these qualities.Only on account of his inclinations and im-pulses he cannot attain this in himself, but atthe same time he wishes to be free from suchinclinations which are burdensome to himself.He proves by this that he transfers himself in

thought with a will free from the impulses ofthe sensibility into an order of things whollydifferent from that of his desires in the field ofthe sensibility; since he cannot expect to obtainby that wish any gratification of his desires, norany position which would satisfy any of hisactual or supposable inclinations (for thiswould destroy the pre-eminence of the veryidea which wrests that wish from him): he canonly expect a greater intrinsic worth of his ownperson. This better person, however, he imag-ines himself to be when be transfers himself tothe point of view of a member of the world ofthe understanding, to which he is involuntarilyforced by the idea of freedom, i.e., of independ-ence on determining causes of the world ofsense; and from this point of view he is con-scious of a good will, which by his own confes-sion constitutes the law for the bad will that hepossesses as a member of the world of sense- alaw whose authority he recognizes while trans-gressing it. What he morally "ought" is then

what he necessarily "would," as a member ofthe world of the understanding, and is con-ceived by him as an "ought" only inasmuch ashe likewise considers himself as a member ofthe world of sense.

Of the Extreme Limits of all Practical Philoso-phy.

All men attribute to themselves freedom ofwill. Hence come all judgements upon actionsas being such as ought to have been done, al-though they have not been done. However, thisfreedom is not a conception of experience, norcan it be so, since it still remains, even thoughexperience shows the contrary of what on sup-position of freedom are conceived as its neces-sary consequences. On the other side it is equa-

lly necessary that everything that takes placeshould be fixedly determined according to lawsof nature. This necessity of nature is likewisenot an empirical conception, just for this rea-son, that it involves the motion of necessity andconsequently of a priori cognition. But this con-ception of a system of nature is confirmed byexperience; and it must even be inevitably pre-supposed if experience itself is to be possible,that is, a connected knowledge of the objects ofsense resting on general laws. Therefore free-dom is only an idea of reason, and its objectivereality in itself is doubtful; while nature is aconcept of the understanding which proves,and must necessarily prove, its reality in exam-ples of experience.

There arises from this a dialectic of reason, sin-ce the freedom attributed to the will appears tocontradict the necessity of nature, and placedbetween these two ways reason for speculativepurposes finds the road of physical necessity

much more beaten and more appropriate thanthat of freedom; yet for practical purposes thenarrow footpath of freedom is the only one onwhich it is possible to make use of reason inour conduct; hence it is just as impossible forthe subtlest philosophy as for the commonestreason of men to argue away freedom. Philoso-phy must then assume that no real contradic-tion will be found between freedom and physi-cal necessity of the same human actions, for itcannot give up the conception of nature anymore than that of freedom.

Nevertheless, even though we should never beable to comprehend how freedom is possible,we must at least remove this apparent contra-diction in a convincing manner. For if thethought of freedom contradicts either itself ornature, which is equally necessary, it must incompetition with physical necessity be entirelygiven up.

It would, however, be impossible to escape thiscontradiction if the thinking subject, whichseems to itself free, conceived itself in the samesense or in the very same relation when it callsitself free as when in respect of the same actionit assumes itself to be subject to the law of na-ture. Hence it is an indispensable problem ofspeculative philosophy to show that its illusionrespecting the contradiction rests on this, thatwe think of man in a different sense and rela-tion when we call him free and when we regardhim as subject to the laws of nature as beingpart and parcel of nature. It must thereforeshow that not only can both these very well co-exist, but that both must be thought as neces-sarily united in the same subject, since other-wise no reason could be given why we shouldburden reason with an idea which, though itmay possibly without contradiction be recon-ciled with another that is sufficiently estab-lished, yet entangles us in a perplexity whichsorely embarrasses reason in its theoretic em-

ployment. This duty, however, belongs only tospeculative philosophy. The philosopher thenhas no option whether he will remove the ap-parent contradiction or leave it untouched; forin the latter case the theory respecting thiswould be bonum vacans, into the possession ofwhich the fatalist would have a right to enterand chase all morality out of its supposed do-main as occupying it without title.

We cannot however as yet say that we are tou-ching the bounds of practical philosophy. Forthe settlement of that controversy does not be-long to it; it only demands from speculativereason that it should put an end to the discordin which it entangles itself in theoretical ques-tions, so that practical reason may have restand security from external attacks which mightmake the ground debatable on which it desiresto build.

The claims to freedom of will made even bycommon reason are founded on the conscious-

ness and the admitted supposition that reasonis independent of merely subjectively deter-mined causes which together constitute whatbelongs to sensation only and which conse-quently come under the general designation ofsensibility. Man considering himself in this wayas an intelligence places himself thereby in adifferent order of things and in a relation todetermining grounds of a wholly different kindwhen on the one hand he thinks of himself asan intelligence endowed with a will, and con-sequently with causality, and when on the ot-her he perceives himself as a phenomenon inthe world of sense (as he really is also), andaffirms that his causality is subject to externaldetermination according to laws of nature.Now he soon becomes aware that both can holdgood, nay, must hold good at the same time.For there is not the smallest contradiction insaying that a thing in appearance (belonging tothe world of sense) is subject to certain laws, ofwhich the very same as a thing or being in itself

is independent, and that he must conceive andthink of himself in this twofold way, rests as tothe first on the consciousness of himself as anobject affected through the senses, and as to thesecond on the consciousness of himself as anintelligence, i.e., as independent on sensibleimpressions in the employment of his reason(in other words as belonging to the world ofunderstanding).

Hence it comes to pass that man claims the pos-session of a will which takes no account of any-thing that comes under the head of desires andinclinations and, on the contrary, conceivesactions as possible to him, nay, even as neces-sary which can only be done by disregardingall desires and sensible inclinations. The causal-ity of such actions lies in him as an intelligenceand in the laws of effects and actions [whichdepend] on the principles of an intelligibleworld, of which indeed he knows nothing morethan that in it pure reason alone independent of

sensibility gives the law; moreover since it isonly in that world, as an intelligence, that he ishis proper self (being as man only the appear-ance of himself), those laws apply to him di-rectly and categorically, so that the incitementsof inclinations and appetites (in other wordsthe whole nature of the world of sense) cannotimpair the laws of his volition as an intelli-gence. Nay, he does not even hold himself re-sponsible for the former or ascribe them to hisproper self, i.e., his will: he only ascribes to hiswill any indulgence which he might yield themif he allowed them to influence his maxims tothe prejudice of the rational laws of the will.

When practical reason thinks itself into a worldof understanding, it does not thereby transcendits own limits, as it would if it tried to enter itby intuition or sensation. The former is only anegative thought in respect of the world of sen-se, which does not give any laws to reason indetermining the will and is positive only in this

single point that this freedom as a negative cha-racteristic is at the same time conjoined with a(positive) faculty and even with a causality ofreason, which we designate a will, namely afaculty of so acting that the principle of the ac-tions shall conform to the essential character ofa rational motive, i.e., the condition that themaxim have universal validity as a law. Butwere it to borrow an object of will, that is, amotive, from the world of understanding, thenit would overstep its bounds and pretend to beacquainted with something of which it knowsnothing. The conception of a world of the un-derstanding is then only a point of view whichreason finds itself compelled to take outside theappearances in order to conceive itself as prac-tical, which would not be possible if the influ-ences of the sensibility had a determiningpower on man, but which is necessary unlesshe is to be denied the consciousness of himselfas an intelligence and, consequently, as a ra-tional cause, energizing by reason, that is, op-

erating freely. This thought certainly involvesthe idea of an order and a system of laws dif-ferent from that of the mechanism of naturewhich belongs to the sensible world; and itmakes the conception of an intelligible worldnecessary (that is to say, the whole system ofrational beings as things in themselves). But itdoes not in the least authorize us to think of itfurther than as to its formal condition only, thatis, the universality of the maxims of the will aslaws, and consequently the autonomy of thelatter, which alone is consistent with its free-dom; whereas, on the contrary, all laws thatrefer to a definite object give heteronomy,which only belongs to laws of nature and canonly apply to the sensible world.

But reason would overstep all its bounds if itundertook to explain how pure reason can bepractical, which would be exactly the same pro-blem as to explain how freedom is possible.

For we can explain nothing but that which wecan reduce to laws, the object of which can begiven in some possible experience. But freedomis a mere idea, the objective reality of which canin no wise be shown according to laws of na-ture, and consequently not in any possible ex-perience; and for this reason it can never becomprehended or understood, because we can-not support it by any sort of example or anal-ogy. It holds good only as a necessary hypothe-sis of reason in a being that believes itself con-scious of a will, that is, of a faculty distinct frommere desire (namely, a faculty of determiningitself to action as an intelligence, in otherwords, by laws of reason independently onnatural instincts). Now where determinationaccording to laws of nature ceases, there allexplanation ceases also, and nothing remainsbut defence, i.e., the removal of the objectionsof those who pretend to have seen deeper intothe nature of things, and thereupon boldly de-clare freedom impossible. We can only point

out to them that the supposed contradictionthat they have discovered in it arises only fromthis, that in order to be able to apply the law ofnature to human actions, they must necessarilyconsider man as an appearance: then when wedemand of them that they should also think ofhim qua intelligence as a thing in itself, theystill persist in considering him in this respectalso as an appearance. In this view it would nodoubt be a contradiction to suppose the causal-ity of the same subject (that is, his will) to bewithdrawn from all the natural laws of the sen-sible world. But this contradiction disappears,if they would only bethink themselves and ad-mit, as is reasonable, that behind the appear-ances there must also lie at their root (althoughhidden) the things in themselves, and that wecannot expect the laws of these to be the sameas those that govern their appearances.

The subjective impossibility of explaining thefreedom of the will is identical with the impos-

sibility of discovering and explaining an inter-est * which man can take in the moral law. Ne-vertheless he does actually take an interest in it,the basis of which in us we call the moral feel-ing, which some have falsely assigned as thestandard of our moral judgement, whereas itmust rather be viewed as the subjective effectthat the law exercises on the will, the objectiveprinciple of which is furnished by reason alone.

* Interest is that by which reason becomes prac-tical, i.e., a cause determining the will. Hencewe say of rational beings only that they take aninterest in a thing; irrational beings only feelsensual appetites. Reason takes a direct interestin action then only when the universal validityof its maxims is alone sufficient to determinethe will. Such an interest alone is pure. But if itcan determine the will only by means of an-other object of desire or on the suggestion of a

particular feeling of the subject, then reasontakes only an indirect interest in the action,and, as reason by itself without experience can-not discover either objects of the will or a spe-cial feeling actuating it, this latter interestwould only be empirical and not a pure ra-tional interest. The logical interest of reason(namely, to extend its insight) is never direct,but presupposes purposes for which reason isemployed.

In order indeed that a rational being who isalso affected through the senses should willwhat reason alone directs such beings that theyought to will, it is no doubt requisite that rea-son should have a power to infuse a feeling ofpleasure or satisfaction in the fulfilment of du-ty, that is to say, that it should have a causalityby which it determines the sensibility accordingto its own principles. But it is quite impossible

to discern, i.e., to make it intelligible a priori,how a mere thought, which itself contains noth-ing sensible, can itself produce a sensation ofpleasure or pain; for this is a particular kind ofcausality of which as of every other causalitywe can determine nothing whatever a priori;we must only consult experience about it. Butas this cannot supply us with any relation ofcause and effect except between two objects ofexperience, whereas in this case, although in-deed the effect produced lies within experience,yet the cause is supposed to be pure reasonacting through mere ideas which offer no objectto experience, it follows that for us men it isquite impossible to explain how and why theuniversality of the maxim as a law, that is, mo-rality, interests. This only is certain, that it isnot because it interests us that it has validity forus (for that would be heteronomy and depend-ence of practical reason on sensibility, namely,on a feeling as its principle, in which case itcould never give moral laws), but that it inter-

ests us because it is valid for us as men, inas-much as it had its source in our will as intelli-gences, in other words, in our proper self, andwhat belongs to mere appearance is necessarilysubordinated by reason to the nature of thething in itself.

The question then, "How a categorical impera-tive is possible," can be answered to this extent,that we can assign the only hypothesis onwhich it is possible, namely, the idea of free-dom; and we can also discern the necessity ofthis hypothesis, and this is sufficient for thepractical exercise of reason, that is, for the con-viction of the validity of this imperative, andhence of the moral law; but how this hypothe-sis itself is possible can never be discerned byany human reason. On the hypothesis, how-ever, that the will of an intelligence is free, itsautonomy, as the essential formal condition ofits determination, is a necessary consequence.Moreover, this freedom of will is not merely

quite possible as a hypothesis (not involvingany contradiction to the principle of physicalnecessity in the connexion of the phenomena ofthe sensible world) as speculative philosophycan show: but further, a rational being who isconscious of causality through reason, that is tosay, of a will (distinct from desires), must ofnecessity make it practically, that is, in idea, thecondition of all his voluntary actions. But toexplain how pure reason can be of itself practi-cal without the aid of any spring of action thatcould be derived from any other source, i.e.,how the mere principle of the universal validityof all its maxims as laws (which would cer-tainly be the form of a pure practical reason)can of itself supply a spring, without any mat-ter (object) of the will in which one could ante-cedently take any interest; and how it can pro-duce an interest which would be called purelymoral; or in other words, how pure reason canbe practical- to explain this is beyond the

power of human reason, and all the labour andpains of seeking an explanation of it are lost.

It is just the same as if I sought to find out howfreedom itself is possible as the causality of awill. For then I quit the ground of philosophicalexplanation, and I have no other to go upon. Imight indeed revel in the world of intelligenceswhich still remains to me, but although I havean idea of it which is well founded, yet I havenot the least knowledge of it, nor an I ever at-tain to such knowledge with all the efforts ofmy natural faculty of reason. It signifies only asomething that remains over when I haveeliminated everything belonging to the worldof sense from the actuating principles of mywill, serving merely to keep in bounds theprinciple of motives taken from the field of sen-sibility; fixing its limits and showing that itdoes not contain all in all within itself, but thatthere is more beyond it; but this somethingmore I know no further. Of pure reason which

frames this ideal, there remains after the ab-straction of all matter, i.e., knowledge of ob-jects, nothing but the form, namely, the practi-cal law of the universality of the maxims, andin conformity with this conception of reason inreference to a pure world of understanding as apossible efficient cause, that is a cause deter-mining the will. There must here be a total ab-sence of springs; unless this idea of an intelligi-ble world is itself the spring, or that in whichreason primarily takes an interest; but to makethis intelligible is precisely the problem that wecannot solve.

Here now is the extreme limit of all moral in-quiry, and it is of great importance to deter-mine it even on this account, in order that rea-son may not on the one band, to the prejudiceof morals, seek about in the world of sense forthe supreme motive and an interest compre-hensible but empirical; and on the other hand,that it may not impotently flap its wings with-

out being able to move in the (for it) emptyspace of transcendent concepts which we callthe intelligible world, and so lose itself amidstchimeras. For the rest, the idea of a pure worldof understanding as a system of all intelligen-ces, and to which we ourselves as rational be-ings belong (although we are likewise on theother side members of the sensible world), thisremains always a useful and legitimate idea forthe purposes of rational belief, although allknowledge stops at its threshold, useful,namely, to produce in us a lively interest in themoral law by means of the noble ideal of a uni-versal kingdom of ends in themselves (rationalbeings), to which we can belong as membersthen only when we carefully conduct ourselvesaccording to the maxims of freedom as if theywere laws of nature.

Concluding Remark

The speculative employment of reason withrespect to nature leads to the absolute necessityof some supreme cause of the world: the practi-cal employment of reason with a view to free-dom leads also to absolute necessity, but onlyof the laws of the actions of a rational being assuch. Now it is an essential principle of reason,however employed, to push its knowledge to aconsciousness of its necessity (without which itwould not be rational knowledge). It is, how-ever, an equally essential restriction of the samereason that it can neither discern the necessityof what is or what happens, nor of what oughtto happen, unless a condition is supposed onwhich it is or happens or ought to happen. Inthis way, however, by the constant inquiry forthe condition, the satisfaction of reason is onlyfurther and further postponed. Hence it unceas-ingly seeks the unconditionally necessary andfinds itself forced to assume it, although with-out any means of making it comprehensible toitself, happy enough if only it can discover a

conception which agrees with this assumption.It is therefore no fault in our deduction of thesupreme principle of morality, but an objectionthat should be made to human reason in gen-eral, that it cannot enable us to conceive theabsolute necessity of an unconditional practicallaw (such as the categorical imperative mustbe). It cannot be blamed for refusing to explainthis necessity by a condition, that is to say, bymeans of some interest assumed as a basis, sin-ce the law would then cease to be a supremelaw of reason. And thus while we do not com-prehend the practical unconditional necessityof the moral imperative, we yet comprehend itsincomprehensibility, and this is all that can befairly demanded of a philosophy which strivesto carry its principles up to the very limit ofhuman reason.