Ancient Greek philosophy was divided into three sciences: physics, ethics, and logic. This division is perfectly suitable to the nature of the thing. The division cannot be made better, except perhaps by adding in the principle by which the division is made. This addition would ensure the division's completeness and reveal the division's necessary subdivisions.
All rational knowledge is either material and has to do with some object, or it is formal and has to do with the form of the understanding, with the form of reason itself, and with the universal rules of thinking in general, no matter what objects the knowledge might be about. Formal philosophy is called logic. Material philosophy, though,
which has to do with specific objects and the laws that govern those objects, is again twofold. This twofold division occurs because these laws are either laws of nature or laws of freedom. The science of the laws of nature is called physics or the doctrine of nature. The science of the laws of freedom is called ethics or the doctrine of morals.
Logic can have no empirical part. That is, logic can have no part which would rest the universal and necessary laws of thinking on grounds based on experience. Logic cannot have such a part because, if the grounds were based on experience, logic would not be logic. Logic would then not be a canon for the understanding or for reason, that is, would not be a collection of strict and rigorous guidelines valid for all thinking and capable of demonstration. On the other hand, natural philosophy as well as moral philosophy can each have its empirical part. Natural philosophy can have its empirical part because nature is an object of experience, and natural philosophy must specify nature's laws according to which everything occurs. Moral philosophy can have its empirical part because the will of the human being is affected by nature, and moral philosophy must specify the laws of freedom
according to which everything ought to be done; but moral philosophy must also mention the conditions under which what human beings ought to do frequently does not get done.
All philosophy, so far as it is based on grounds of experience, can be called empirical. But philosophy, so far as it presents its teachings only on the basis of
a priori principles, can be called pure philosophy. But pure philosophy, if it is merely formal, is called logic. If pure philosophy is restricted to specific objects, then it is called metaphysics.
Because of these various conceptual subdivisions within philosophy, there arises the idea of a twofold metaphysics: a metaphysics of nature and a metaphysics of morals. So physics will have its empirical part, but also a rational part. Ethics, too, will have both kinds of parts. In the case of ethics, though, the empirical part especially could be called practical anthropology, while the rational part could properly be called moral.
All trades, crafts and arts, have gained through the division of labor.
The gain is due to the fact that in the division of labor no one makes everything. Instead, each person limits herself to certain work which, in how it needs to be handled, differs markedly from other work. This limiting makes it possible to perform the work with increasing perfection and with greater efficiency. Where labor is not distinguished and divided in this way, where everyone is a Jack-of-all-trades, trade remains woefully undeveloped. It would be worth asking the following questions. Does pure philosophy in all its parts require a person with special skills? Would the whole of the learned profession be better off if those, who promote themselves as "independent thinkers" while calling others "hair-splitters" who work only with the rational part of philosophy, were warned not to try to perform two tasks at the same time? Would it not be better if these so-called independent thinkers, who, accustomed to trying to satisfy the tastes of the public, mix the empirical with the rational in all kinds of proportions unknown even to themselves, were warned not to multi-task,
because multi-tasking by a single person produces only a mess when each individual task demands a special talent? But, although those are worthwhile questions, I here only ask whether the nature of science demands that the empirical part always be carefully separated from the rational part. I here also only ask whether the nature of science requires a metaphysics of nature to precede a proper (empirical) physics and requires a metaphysics of morals to precede a practical anthropology. In both cases, the metaphysics must be carefully cleansed of everything empirical in order to know how much pure reason could achieve and from what sources pure reason could create its own teaching
a priori. It is all the same to me whether the latter task is conducted by all moralists (whose name is legion) or only by those who feel a calling to take on the task.
Since my aim here is squarely directed at moral philosophy, I limit the above questions about metaphysics in general to this question about the metaphysics of morals in particular: whether it is of the greatest importance to work out once a pure moral philosophy which would be thoroughly cleansed of everything
which might be empirical and belong to anthropology. For that there must be such a pure moral philosophy is evident from the common idea of duty and of moral laws. Everyone must admit the following points: that a law, if it is to be moral, if, that is, it is to be a ground of an obligation, must carry with it absolute necessity; that the command, "thou shalt not lie," holds not just for human beings, as if other rational beings were not obliged to obey it, and the same goes for all other genuine moral laws; that, therefore, the ground of obligation for moral laws must be sought, not in the nature of the human being or in the circumstances of the world in which the human being lives, but rather must be sought
a priori only in concepts of pure reason; and that every other prescription based on principles of mere experience can never be called a moral law but at most only a practical rule, and even a prescription that might be universal in a certain way — perhaps only in its motive — can only be a practical rule and never a moral law if it is based in the least part on empirical grounds.
So moral laws, together with their principles, are essentially different from all other practical knowledge in which there is something empirical. But the scope is even wider: all moral philosophy, not just moral laws and their principles, rests wholly on its pure part. Moral philosophy, when applied to human beings, borrows nothing from the knowledge of human beings (anthropology), but rather gives the human being, as a rational being, laws
a priori. These laws still require a power of judgment that is sharpened through experience, partly to distinguish those cases to which the laws apply, partly to give the laws access to the will of the human being and energy for putting the laws into practice. This access to the will and energy for implementation are needed because human beings, though capable of the idea of a pure practical reason, are affected by so many inclinations that they find it difficult to make the idea concretely effective in the way they live their lives.
A metaphysics of morals is therefore indispensably necessary. It is indispensable not merely to satisfy deep-rooted curiosity about the source of the practical principles that are present
a priori in our reason.
It is also indispensable because morals themselves remain vulnerable to all kinds of corruption so long as that guiding thread and highest norm of correct moral judgment is lacking. For in the case of what is to be morally good, it is not enough that it is in conformity with the moral law, but rather it must also be done for the sake of the moral law. If it is not also done for the sake of the moral law, then that conformity is only very coincidental and precarious because, although the non-moral ground will now and then produce actions that are in conformity with the moral law, the non-moral ground will again and again produce actions that are not in conformity with the moral law. But, now, the moral law, in its purity and genuineness (which is what is most important in moral matters), is to be found no where else than in a pure philosophy. So this (metaphysics) must come first, and without it there can be no moral philosophy at all. That which mixes pure principles with empirical principles does not even deserve to be called a philosophy (for philosophy distinguishes itself from common rational knowledge by presenting as a separated science that which common rational knowledge comprehends only in a confused way).
Even less does it deserve to be called a moral philosophy because, through this confusion that it creates by mixing pure principles with empirical principles, it trashes the purity of morality itself and undermines its own ends.
You would be way off base to think that in the preparatory study to the famous Wolff's moral philosophy, specifically in what Wolff called universal practical philosophy, you already have what is here demanded and therefore that no new ground needs to be broken. It is just because Wolff's moral philosophy was to be a universal practical philosophy that it did not consider a will of any special kind. In particular, it did not look into the possibility of a will which would be fully motivated by
a priori principles. Such a will, animated without empirical motives, could be called a pure will. Instead, Wolff considered willing in general, with all actions and conditions that belong to willing in this general sense. Because it considers willing in general, Wolff's moral philosophy differs from a metaphysics of morals, just as general logic differs from transcendental philosophy.
General logic presents the operations and rules of thinking in general, but transcendental philosophy merely presents the special operations and rules of pure thinking, i.e., those operations and rules by which objects are cognized completely
a priori. For the metaphysics of moral is to investigate the idea and the principles of a possible pure will and not the actions and conditions of human willing in general, which for the most part are drawn from psychology. It is no objection to what I am saying that this universal practical philosophy also speaks (although without any warrant) of moral laws and duty. For the authors of that science remain true to their idea of it also in this: those authors do not distinguish the motives which, as such, are represented completely
a priori merely by reason and which are genuinely moral from those motives which are empirical and which the understanding raises to universal concepts merely by comparing experiences. These authors instead, without paying attention to the different
sources of motives, consider only the intensity of the motives (looking at them as all being of the same kind), and from this sole consideration they put together their concept of obligation. Their concept is, of course, anything but moral. But a concept so constructed is all that can be expected from a philosophy that makes no attempt to decide the origin of all possible practical concepts and that makes no attempt to decide whether the concepts occur
a priori or merely
Having the intention to publish someday a metaphysics of morals, I prepare the way for it with this groundlaying. Without a doubt, there is properly no other foundation for a metaphysics of morals than the critique of a pure practical reason, just as for metaphysics there is no other foundation than a critique of pure speculative reason, which I have already published. But, first of all, a critique of pure practical reason is not so extremely necessary as is a critique of pure speculative reason. A critique of pure practical reason is not as necessary because in moral matters human reason, even in cases of merely average intelligence, can easily be brought to a high level of correctness and completeness. In contrast, human reason in its theoretical but pure use is through and
through dialectical. In the second place, I require that a critique of pure practical reason, if it is to be complete, must at the same time be capable of presenting in a common principle practical reason's unity with speculative reason. Such a critique must be capable of presenting this unity because in the end there can be only one and the same reason which is distinguished only in its application. But in this groundlaying I was not yet able to pull off such a feat of completeness; doing so would have required that I drag in considerations of a quite different kind and confuse the reader. Because of this incompleteness, I have called this work a groundlaying toward the metaphysics of morals rather than a critique of pure practical reason.
But in the third place, because a metaphysics of morals, despite the scary title, is capable of a high degree of popularity and resonance with the thinking of ordinary folks, I find it useful to separate off this preparation of the foundation of the metaphysics of morals so that the subtleties that are unavoidable in this preparation
need not bog down the more comprehensible teachings of the metaphysics of morals which I will publish in the future.
The present groundlaying, however, is nothing more than the search for and establishment of the highest principle of morality. In its purpose, this task is by itself complete and to be kept separate from all other moral inquiry. There is no doubt that what I have to say about this main question, which is an important question but which has up to now been the subject of very unsatisfying discussion, would be made much clearer through the application of that highest principle to the whole system and that what I have to say would be strongly confirmed by the adequacy that the principle displays everywhere. But I had to forgo this advantage, which would have been more self-serving than generally useful anyway, because a principle's ease of use and apparent adequacy provide no sure proof at all of the correctness of the principle. Instead, a principle's ease of use and apparent adequacy awaken a certain bias against investigating and weighing the principle itself, apart from any consideration of consequences, in a strict way.
I have selected a method for this book which, I believe, will work out best if we proceed in the following way. First, we proceed analytically from common knowledge to the formulation of the highest principle. Then, second, we synthetically work our way back from the examination of this principle and its sources to common knowledge in which we find the principle applied. Using this method, the sections of the book turn out to be:
1. First Section: Transition from common moral rational knowledge to the philosophical.
2. Second Section: Transition from popular moral philosophy to the metaphysics of morals.
3. Third Section: Last step from the metaphysics of morals to the critique of pure practical reason.
There is nothing at all in the world, or even out of it, that could possibly be thought to be good without qualification except a good will. Intelligence, humor, power of judgment, and whatever else the talents of the mind may be called, are without doubt in many respects good and desirable. Likewise, courage, decisiveness, and perseverance in pursuit of goals, as qualities of temperament, are without doubt in many respects good and desirable. But these talents of the mind and qualities of temperament can also become extremely bad and harmful, if the will that is to make use of these natural gifts, and so a will whose distinctive quality is therefore called character, is not good. It is just the same with gifts of fortune. Power, wealth, reputation, even health and the whole well-being and satisfaction with your condition, which
goes by the name of happiness, produce courage; but these gifts of fortune frequently also produce arrogance as a by-product when there is no good will present to check their influence on the mind, no good will present to correct the whole principle of acting, and when there is no good will present to make these gifts of fortune and principle of acting conform to universal standards. And it goes without saying that a rational and impartial spectator, at the sight of the uninterrupted prosperity of someone who has no trace of a pure and good will, can never be satisfied, and so a good will appears to constitute the indispensable condition of even the worthiness to be happy.
Some qualities are even helpful to this good will itself and can make its work easier. But these qualities still have no inner unconditional worth. Instead, the qualities always presuppose a good will which limits the esteem which we otherwise justly have for them and which does not allow them to be considered absolutely good. Moderation in volatile emotions and passions, self-control and sober reflection are not only good for many purposes, but they even appear to constitute a part of the inner worth of a person. But there is much that these qualities lack that would be needed in order to declare them to be good without qualification (however much the ancients praised them unconditionally). For, without basic principles of a good will, these qualities can become very bad, and the cold blood of a scoundrel makes her
not only far more dangerous, but also in our eyes even more immediately abominable than she would be held to be without such cold-bloodedness.
The good will is good only through its willing, i.e., is in itself good. It is not good because of what it effects or accomplishes, nor is it good because of its suitability for achieving some proposed end. Considered in itself, the good will is, without comparison, of far higher value than anything that it could ever bring about in favor of some inclination or even in favor of the sum of all inclinations. Even if a good will wholly lacked the capacity to carry out its purposes, due to an especially unfavorable turn of fate or due to the scanty provision of a step-motherly nature, it would still shine for itself like a jewel, like something that has all its worth in itself. A good will would even shine like this if, despite its greatest efforts (not, of course, as a mere wish but rather as calling upon all means so far as they are in our power), it never could accomplish anything and remained only a good will. The good will's usefulness or fruitlessness can neither add something to that will's worth nor take anything away from that worth. Any such usefulness would, as it were, only be the setting that would make the will easier to handle in everyday activities or the setting that would attract the attention of people who do not yet know enough about the good will.
Such usefulness would not recommend a good will to those people who do know about the will and such usefulness would not play a role in ascertaining the worth of the good will.
There is, however, something very strange in the idea of the absolute worth of the mere will: in figuring the value of this will, no account is made of its usefulness. Because of this strangeness, and despite the agreement of even ordinary reason with the idea, a suspicion must nevertheless arise that perhaps mere high-flying fantasy is secretly the basis of the idea. The suspicion also arises that nature, in making reason the boss of our wills, may be misunderstood. So we will put this idea to the test from the point of view that sees reason as the commander of our wills.
In the natural makeup of an organized being, i.e., a being that is put together for living, we take it to be a basic principle that, for any organ with a specific job to do in the being, the organ will be the most appropriate for the job and the most suitable. Now if, for a being with reason and a will, its preservation, its well-being, in a nutshell, its happiness, were the end or goal of nature, then nature would have hit upon a very poor arrangement by putting reason in charge of the creature in order to achieve this end or goal. For all the actions that the creature has to carry out to achieve this end or goal of happiness
and the whole rule of its behavior would be prescribed to the creature much more precisely by instinct. The end or goal to obtain happiness, too, could have been much more certainly attained by instinct than it ever can be by reason. If reason had anyway been given to the favored creature, then reason would only have had to serve the creature by helping the creature meditate on the fortunate makeup of its nature, admire it, enjoy it, and be thankful for it. Reason would not have served to subject the creature's powers of desiring to reason's weak and deceitful guidance and to meddle in the purposes of nature. In short, nature would have ensured that reason did not try for practical use, that is, was not used for making decisions about what to do, and would have ensured that reason, with its weak insights, did not have the audacity to think out for itself the plan for the creature's happiness and the means to carry out that plan. Nature would have taken over for itself not only the choice of the ends or goals but also of the means and with wise foresight would have entrusted both ends and means only to instinct.
In fact, we also find that the more a cultivated reason occupies itself with the aim of obtaining happiness and of enjoying life the more the human being departs from true contentment. In pursuing this aim, in many people — and indeed those most experienced in the use of reason, if they are only honest enough to admit it —
there arises a certain degree of misology, i.e., hatred of reason. This misology arises because, after these people estimate all the advantages which they receive from not only the invention of all arts of common luxury but also even from the sciences (which appears to them at bottom also to be a luxury of the understanding), they still find that they have in fact created more trouble for themselves than they have gained in happiness. In the end, these people wind up envying rather than despising the more ordinary kind of human being who is closer to the guidance of mere natural instinct and who does not permit reason much influence on her conduct. Some people greatly moderate, or even reduce below zero, the boastful high praises of the advantages that reason is supposed to provide us in terms of happiness and satisfaction in life; we must admit that the judgment of these people is in no way bitter or unthankful for the goodness that exists in the way the world is governed. And so, instead, we must admit that these judgments secretly have as their basis the idea of a different and much worthier purpose for their existence. Reason is quite properly to be used for this worthier purpose and not for happiness. It is therefore to this worthier purpose, as the highest condition, that the private purposes of humans beings must in large part defer.
For since reason is not sufficiently able to guide the will reliably with regard to the will's objects
and with regard to the satisfaction of all of our needs (which reason in part even multiplies) — an end to which an implanted natural instinct would have led much more certainly — and since reason has nevertheless been given to us as a practical faculty, i.e., as a capacity that is to exercise an influence on the will, the true function of reason must be to produce, not at all a will that is good as a means to achieve some end, but rather a will good in itself. Because in all other circumstances nature has worked purposefully in distributing its capacities, reason was absolutely necessary in order to produce such a will that is good in itself. So, to be sure, this will may not be the only and the whole good, but it must still be the highest good and be the condition for all the other goods, even the condition for all longing for happiness. As such a condition, the good will is quite consistent with the wisdom of nature. You can appreciate this consistency even when you notice that the cultivation of reason, which is required for the first and unconditional end of producing a good will, in may ways limits, at least in this life, the attainment of the second and always conditional end of happiness. Indeed, the good will can even reduce happiness to something less than zero and still be consistent with the purposeful activity of nature. Even such an extreme reduction would be consistent with nature's purposes because reason, which acknowledges its highest practical function to be the production of a good will, is only capable of a satisfaction of its own kind — namely from the attainment of an end that again reason alone sets — when it produces such a good will.
Reason is even capable of this satisfaction in cases when producing such a good will is connected with many infringements on the ends of inclination.
The concept of a good will already dwells in the natural sound understanding and needs not so much to be taught as instead only to be clarified. This concept also always stands highest in the valuation of the whole worth of our actions and constitutes the condition of everything else. In order to dissect this concept of a good will, a will that is to be highly esteemed in itself and for no further purpose, we will lay bare the concept of duty, which contains the concept of a good will. Although the concept of duty contains the concept of a good will, it does so only under certain subjective limitations and restrictions. Far from hiding and disguising the concept of a good will, these subjective limitations and restrictions instead let the concept of a good will stand out by contrast and allow the concept to shine even more brightly.
I here pass over all actions that are already recognized as contrary to duty, even though the actions might be useful for this or that purpose; for in the case of these actions, the question does not even arise as to whether they are done from duty, since they even conflict with duty. I also put to the side actions that are actually in accordance with duty but are also actions to which human beings have no inclination that is direct or immediate but which human beings perform because they are driven to do so by another inclination. For
in these cases it is easy to tell whether the action conforming to duty is done from duty or from a self-serving purpose. It is much more difficult to notice this difference in cases where the action conforms to duty and the subject also has an immediate or direct inclination for the action. For example, a shopkeeper who does not overcharge his inexperienced customers is certainly acting in conformity with duty, and, where there are many transactions, the prudent shopkeeper does not overcharge. Instead, the prudent shopkeeper sets a fixed common price for everyone so that a child can shop at her store just as well as anyone else. So the public is honestly served. But this honest treatment of the customers is not nearly enough to be the basis for the belief that the shopkeeper acted from duty and principles of honesty. Her self-interest required it. But it cannot here be assumed that the shopkeeper also had an immediate or direct inclination to give the customers, out of love for them, so to speak, no preference of one over the other in terms of the price. So the action was done neither from duty nor from immediate or direct inclination, but instead the action was done merely for a self-interested purpose.
On the other hand, to preserve your life is a duty, and everyone also has an immediate inclination to do this. But, because of this inclination, the often anxious care that most of the human race has for life is an anxious care that still has no inner worth, and their maxim prescribing self-preservation has no moral content. Their action to preserve their lives definitely conforms to duty,
but it is not done from duty. By contrast, when adversities and hopeless sorrow have completely taken away the zest for living, when the unhappy person, strong of soul, angered over her fate more than faint-hearted or dejected, wishes for death and yet preserves her life without loving it, not from inclination or fear, but from duty, then her maxim has moral content.
To be beneficent where you can is a duty and there are also many souls so compassionately disposed that they find an inner satisfaction in spreading joy around them and can take delight in the satisfaction of others so far as it is their work. These compassionately attuned souls even experience this inner satisfaction without any motive of vanity or usefulness to themselves. But I maintain that in such cases an action of this kind, however much it may conform to duty, however kind it may be, nevertheless has no true moral worth. Instead, actions of this kind are on a par with other inclinations, for example, with the inclination to honor. This inclination to honor, when it is lucky enough to hit what is generally useful and in line with duty, and is therefore worthy of honor, deserves praise and encouragement, but not esteem. For the maxim lacks moral content, namely, to do such actions not from inclination, but rather from duty. Granted, then, that the mind of that friend of the human being were clouded by its own sorrow, which extinguishes all
compassion for the fate of others. Suppose she still had the power to benefit others who are suffering, but that strangers in need did not move her because she is sufficiently occupied with her own needs. And now she still rips — since no inclination prods her to it — herself out of this deadly insensitivity and does the action without any inclination, merely from duty. Then her action has for the first time its genuine moral worth. Suppose further still: if nature had put very little sympathy in the heart of this or that person, if she (after all an honest person) were of cold temperament and indifferent — perhaps, because she herself is equipped with the special gift of patience and enduring strength against her own suffering, she presumes or even demands the same in the case of every other person — toward the sufferings of others, if nature had not exactly formed such a person (who truly would not be nature's worst product) to be a friend of human beings, would she not still find in herself a source that would give herself a worth far higher than might be the worth of a good-natured temperament? Certainly! It is precisely here that the worth of character begins, a worth that is moral and above all comparison the highest. In particular, that worth begins in that she is beneficent, not from inclination, but from duty.
To secure your own happiness is a duty (at least an indirect duty), for the lack of satisfaction
with your condition, in a crowd of many worries and in the middle of unsatisfied needs, could easily become a great temptation to the transgression of duties. But, even without looking at duty here, all human beings already have of themselves the most powerful and most intimate inclination for happiness, because precisely in this idea of happiness all inclinations are united into a collection. But the prescription of happiness is for the most part constituted in such a way that the prescription greatly infringes on some inclinations, and yet the human being can formulate no definite and secure concept of the collective satisfaction of all inclinations, which goes by the name of happiness. It should come as no surprise, then, how a single inclination — which specifies what it promises and the time within which its satisfaction can be felt — might be able to outweigh a wavering idea. For example, a person suffering from gout might be able to choose to eat or drink what tastes good to her and to suffer the consequences because she, according to her way of calculating the costs and benefits in this case at least, does not miss out on a present enjoyment through a perhaps groundless expectation of a happiness that is supposed to be found in health. But even in this case, if the universal inclination to happiness does not control her will, if health for her at least is not so necessary in her calculations of costs and benefits, then there remains in this case, as in all other cases, a law, namely, to promote her happiness
not from inclination but from duty. And then her conduct, for the first time, has genuine moral worth.
No doubt, it is also in this way that we are to understand the scriptural passages in which we are commanded to love our neighbor and even to love our enemy. For love as an inclination cannot be commanded. But beneficence from duty itself, even if no inclination at all drives us to it — indeed, even if natural and invincible disinclination stands against us — is practical and not pathological love. This practical love lies in the will and not in tendency to feeling, lies in basic principles of action and not in melting compassion. This practical love alone can be commanded.
The second proposition is this: an action done from duty has its moral worth not in the purpose which is to be achieved by performing the action, but rather in the maxim according to which the action is decided upon. So the worth of such an action depends not on the actuality of the object of the action but only on the principle of willing according to which the action, regardless of any objects of the faculty of desire, is done. It is clear from what I have already said that the purposes which we may have in our actions, and the effects of our actions, as ends or goals and incentives of the will, can give no unconditional and moral worth to the actions. Where, then, can this worth be located, if it is not
to be found in the will, in the will's relation to the hoped-for effect of the actions? The worth can be located nowhere else than in the principle of the will, regardless of the ends that can be brought about by such action. For the will stands, so to speak, at a crossroads right in the middle between its principle
a priori, which is formal, and between its motive
a posteriori, which is material. Since the will must still be controlled by something, it must be guided by the formal principle of willing in general when an action is done from duty, because every material principle has been removed from the will.
I would express the third proposition, which is a consequence of the previous two, in this way: duty is the necessity of an action out of respect for the law. I can of course have an inclination for an object as an effect of my intended action, but I can never have respect for such an object precisely because the object is merely an effect and not the activity of a will. Likewise, I cannot have respect for inclination in general, whether it is my own inclination or someone else's. With an inclination of my own, I can at most approve of it; regarding others' inclinations, I can sometimes even love them, that is, view their inclinations as favorable to my own self-interest. But only something that is connected to my will merely as a ground, never as an effect, something that does not serve my inclination but instead outweighs it — something at least that wholly excludes inclination
from rough-and-ready decisions about what choices to make — and therefore only something that is the mere law itself, can be an object of respect and thus a command. Now an action from duty is to be detached completely from the influence of inclination and along with inclination from every object of the will. So nothing that could control the will remains except objectively the law and subjectively pure respect for this practical law. And so all that remains to guide the will is the maxim* of obeying such a law, even if this obedience involves dialing back all my inclinations.
So the moral worth of an action does not lie in the effect that is expected from the action; nor, therefore, is the moral worth of an action in some principle of action which has to get its motivating ground from this expected effect. For all these effects (pleasantness of your condition, and even the promotion of the happiness of others) can also be brought about by other causes, and so the will of a rational being is not needed, even though it is only in a rational being that the highest and unconditional good can be found. So nothing but the intellectual representation of the law in itself, which of
* A maxim is the subjective principle of willing; the objective principle is the practical law. (That is, the objective principle is the practical principle that would serve all rational beings as a subjective principle, too, if reason had full control over the faculty of desire.)
course can only be found in a rational being, so far as this representation or thought, and not the expected effect of the action, is the controlling motivational ground of the will, can constitute the pre-eminent good which we call moral. This pre-eminent moral good is already present in the person who acts according to the representation of the law in itself, and this moral good does not need to wait for the expected effect of the action in order to become good.*
* You could object that by using the word "respect" I am only seeking to escape in an obscure feeling instead of bringing clarity to the question through a concept of reason. But although respect is a feeling, it is not a feeling received by influence. Instead, respect is a feeling self-woven through a rational concept. The feeling of respect, therefore, is specifically different from all feelings of the kind received by influence, which reduce to inclination or fear. What I immediately cognize or intellectually apprehend as a law for myself, I cognize with respect, which just signifies the consciousness of the subordination of my will to a law, without the mediation of other influences on my sense. The immediate or direct determination of the will by the law and the consciousness of that subordination is called respect. So respect, this awareness of the will's being guided by the law, must be thought of as an effect of the law on a person and not as a cause of the law. Respect is actually the representation of a worth that does damage to my self-love. So respect is something that is considered neither to be an object of inclination nor an object of fear, although it has something analogous to both at the same time. The object of respect is therefore only the law and indeed that law which we ourselves impose on ourselves and yet which is necessary in itself. Considered as a law, we are subject to this object of respect without consulting self-love; as self-imposed, this object is nevertheless a consequence of our will. Viewing it in the first way, as a law, the object is analogous to fear; viewing it in the second way, as self-imposed, the object is analogous to inclination.
But what kind of law can that really be, the representation of which — without even taking into consideration the expected effect from the action — must guide the will so that the will can be called absolutely good without qualification? Since I have robbed the will of any impulse that could arise from the will by following any law, nothing remains except the universal conformity of actions to law in general; this universal conformity is to serve the will as a principle. That is, I ought never act except in this way: that I could also will that my maxim should become a universal law. Here now is the mere conformity to law in general (without making a law for specific actions a ground) that serves the will as its principle and even must serve it as its principle if duty is not to be everywhere an unfounded delusion and chimerical concept. In its judgments about what to do, ordinary human reason agrees completely with this principle and always has the principle in view.
All respect for a person is actually only respect for the law (of integrity, etc.) of which the person provides us with an example. Because we look at the development of our talents as a duty, we conceive of a person who has talents as, so to speak, an example of a law and that conception constitutes our respect. All so-called moral interest consists simply in respect for the law.
The question might be, for instance, the following. When I am in a tight spot, may I not make a promise with the intention of not keeping it? I easily make here the difference in meaning that the question can have: whether it is prudent, or whether it is in accord with duty, to make a false promise. There is no doubt that making a false promise can often be prudent. Indeed, I see very well that it is not enough that I extricate myself from a present embarrassment by means of this excuse. Instead, I must consider carefully whether from this lie far greater trouble than the trouble from which I now set myself free might not arise for me afterwards. And, since the consequences of all my supposed slyness are not so easy to predict and that a trust once lost could be far more disadvantageous to me than any evil that I now intend to avoid, I must also consider whether it might be more prudently handled to act in this matter according to a universal maxim and to make it a habit to promise nothing except with the intention of keeping the promise. But after considering these possibilities, it soon becomes clear to me that such a prudential maxim would only be based on the fear of consequences. Now it is certainly something quite different to be truthful from duty than to be truthful out of fear of disadvantageous consequences. For, in the case of being truthful from duty, the concept of the action in itself already contains a law for me. In the case of being truthful out of fear, I must first look around elsewhere for the effects on me which are likely
to be connected with the action. For, if I deviate from the principle of duty, then it is quite certainly bad. If, however, I desert my maxim of prudence, then that can sometimes be very advantageous to me, although it is of course safer to stay with the maxim of prudence. But, in order to inform myself, in the shortest and yet least deceptive way, of the answer to this problem of whether a lying promise conforms to duty, I ask myself the following. Would I be quite content that my maxim (to extricate myself from an embarrassment by means of an untruthful promise) should hold as a universal law (for me as well as for others) and would I be well able to say to myself that everyone may make an untruthful promise when she finds herself in an embarrassment from which she cannot escape in any other way? I soon become aware that I can indeed will the lie but that I definitely cannot will a universal law to lie. I cannot will a universal law to lie, for according to such a law there would actually be no promise at all. There would actually be no promise because it would be pointless to pass off my intentions regarding my future actions to others who would certainly not believe this pretence or who, if they did rashly believe it, would certainly pay me back in like coin. My maxim, therefore, as soon as it became a universal law, would have to destroy itself.
What I therefore have to do so that my willing is morally good requires no far-reaching
acuteness. Inexperienced as to how the world operates, incapable of preparing myself for any events that might occur in the world, I only ask myself: can you also will that your maxim become a universal law? If the maxim cannot become a universal law, then the maxim is objectionable. It is objectionable not because it presents an impending disadvantage to you or even to others; instead, the maxim is objectionable because it cannot fit as a principle into a possible universal lawgiving. Reason compels respect from me for this universal lawgiving. I certainly do not yet see on what the respect is based (a topic which the philosopher may investigate), but I at least understand this much: respect is the estimation of a worth that outweighs all the worth of anything that inclination praises, and the necessity of my actions from pure respect for the practical law is what constitutes duty, and every motivating ground must yield to duty because duty is the condition of a will good in itself and whose worth exceeds the worth of everything else.
We have, then, in the moral knowledge of common human reason, arrived at its principle. Common human reason of course does not abstractly think of this principle in such a universal form, but it does actually always have the principle before its eyes and uses the principle as the standard for its judgment. It would be easy to show here how
common human reason, with this compass in hand, very well knows in all cases that it encounters how to distinguish what is good, what is bad, what conforms to duty, or what is contrary to duty. If only we, as Socrates did, draw its attention to its own principle, common human reason can make these distinctions without our having to teach it anything new. So there is, in order to know what you have to do in order to be honest and good — or even to be wise and virtuous — no need for science and philosophy. It might even have been supposed well in advance that the knowledge that is incumbent on everyone — knowledge of what to do and therefore of what to know — would be the concern of everyone, even the concern of the most ordinary human being. It is at this point that you have to look with admiration at how the power of practical judgment has an advantage over the theoretical in ordinary human understanding. In theoretical matters, when ordinary reason dares to depart from the laws of experience and the perceptions of sense, it gets into nothing but incomprehensibilities and contradictions with itself. At the very least, when ordinary reason dares to make these departures, it gets into a chaos of uncertainty, obscurity, and instability. But in practical matters, it is just when ordinary understanding excludes all sensuous motives for practical laws that the power of judgment first begins to show itself to advantage. When ordinary understanding makes these exclusions it even becomes subtle, whether it be in quibbling with its conscience or with other claims in reference to what is to be called right or
whether it be in wanting correctly to determine the worth of actions for its own instruction. But what is most remarkable is that, in determining the worth of actions, ordinary understanding can have just as good a hope of getting it right as a philosopher herself can ever promise. In fact, ordinary understanding is almost more secure in determining the worth of actions than the philosopher because the philosopher can have no other principle than the principle that ordinary understanding has and because the philosopher's judgment can easily be confused by a crowd of extraneous considerations not pertinent to the matter at hand and can be diverted from the right direction. Would it not, accordingly, be more advisable in moral matters to rest content with ordinary rational judgment? Would it not be more advisable to bring in philosophy at most only in order to present the system of morals more completely and more comprehensibly? Would it not be more advisable to bring in philosophy only so that it can present the system's rules in a way more convenient for their use (especially in disputation)? And would it not be less advisable, for practical purposes, to allow philosophy to drag ordinary human understanding away from its happy simplicity and to put the understanding on a new path of investigation and instruction?
Innocence is a magnificent thing, but it is also very bad in that it cannot be easily preserved and can easily be misled. Because of these deficiencies, even wisdom — which otherwise perhaps consists more in doing and letting than in knowing — still requires science, not in order to learn from science, but rather
to gain accessibility and permanence for wisdom's prescriptions. The human being feels in itself a powerful counterweight to all commands of duty, commands which reason represents to the human being as so worthy of great respect. This counterweight is the needs and inclinations of the human being, and the whole satisfaction of its needs and inclinations is included under the name of happiness. Now reason's prescriptions are commanded without apology and without a promise of anything to the inclinations. Reason therefore commands, so to speak, dismissively and with no regard for those claims that are so impulsive and yet that appear so reasonable (and which can be willed away by no command). From this, however, a natural dialectic arises, that is, a tendency to rant about those strict laws of duty and to cast doubt on the validity — at least the purity and strictness — of those laws and, if possible, to make the laws more suitable to our wishes and inclinations. That is, a tendency arises that attempts to corrupt the laws at their foundations and to destroy their dignity. The result of this natural dialectic, then, is something that in the end even ordinary practical reason cannot call good.
Because of this destructive tendency of natural dialectic, ordinary human reason is driven to go out of its comfort zone and to take a step into the field of practical philosophy. Ordinary human reason is driven to this not by some intellectual need to theorize (a need which never afflicts it so long as it is satisfied with being merely sound reason), but instead it is driven to it for practical reasons. In the field of practical philosophy, ordinary reason hopes, regarding the source of its principle
and the correct determination of its principle, in contrast with the maxims or principles that rest on need and inclination, to receive information and clear instruction. Having received these, ordinary reason can perhaps escape the embarrassment resulting from the flip-flopping claims of dialectic and perhaps not run the risk of losing all genuine moral principles in the ambiguity into which ordinary reason easily slips. So there arises unnoticed a dialectic which requires reason to seek help in philosophy. This dialectic arises just as much in practical ordinary reason, when it is cultivated, as it does in the theoretical use of reason. Both uses of reason will therefore only find peace in a complete critique of our reason.
Even if we have drawn our previous concept of duty from the ordinary use of our practical reason, this is no reason to conclude that we have treated the concept of duty as a concept of experience. Rather, when we pay attention to the experience of the way human beings act and fail to act, we encounter frequent and, as we ourselves admit, justified complaints that no one can provide a sure example of the disposition to act from pure duty. There are also justified complaints that even though much of what duty commands may be done according to duty, it is always still doubtful whether what is done really is done from duty and so has moral worth. Because of complaints like these, there have always been philosophers who have absolutely denied the reality of this disposition in human actions and who have attributed everything to a more or less refined self-love. These philosophers nevertheless do not call into question the correctness of the concept of morality. Rather, with heartfelt regret for the frailty and impurity of human nature, these philosophers make mention of a human nature which, though definitely noble enough
to make an idea so worthy of respect into its prescription, is at the same time too weak to follow the prescription. So, instead of serving this human nature for lawgiving, reason only serves it in order to provide for the interest of inclinations, whether providing for the inclinations individually or at most for their greatest compatibility with each other.
In fact, it is absolutely impossible to find with certainty through experience a single case in which the maxim of an action that is otherwise in accord with duty has rested only on moral grounds and on the representation of a person's duty. For it is certainly sometimes the case that the most thorough self-examination does not turn up anything, except the moral ground of duty, that could have been strong enough to move us to do this or that good action and to move us to make such a great sacrifice. It cannot, however, be safely concluded from this unsuccessful self-examination that there really is no hidden impulse of self-love which, under the mere guise of that idea of duty, really was the determining cause of the will. Because of this self-love, masquerading as duty, we then gladly flatter ourselves with a nobler motive which we falsely claim for ourselves. But, in fact, we can never, even through the most strenuous examination, fully get behind the hidden incentives because, when the issue is about moral worth, what matters are not the actions that you see but rather the inner principles that you do not see.
There are some people who ridicule all morality as a mere mental fantasy of a human imagination super-sized through its own boasting. You cannot do a greater service for such people than to admit to them that the concepts of duty (just as you gladly convince yourself from convenience that the same applies to all other concepts) must be drawn only from experience; for by this admission you prepare for these people a guaranteed triumph. I am willing to admit out of a love of humankind that most of our actions are in accord with duty. But if you look at people's intentions and endeavors more closely, you will bump into the dear self everywhere; it is on this dear self, which is always popping out, that their intentions are based, not on the strict command of duty. You do not need to be an enemy of virtue in order to become (especially with increasing years and a power of judgment that through experience has been made partly shrewder and partly more observant) doubtful at certain moments whether any true virtue is really to be found in the world. To become doubtful about the reality of true virtue, you only need to be a cold-blooded observer who does not immediately take the liveliest wish for the good to be the actualization of that good. And now here nothing can protect us from falling completely away from our ideas of duty and preserve in our soul a well-grounded respect for duty's laws except the clear conviction that, even if there never have been actions
which arose from such pure sources, the question here is not whether this or that happens but rather whether reason by itself and independently of all appearances commands what ought to happen. Therefore, without letting up even a bit, reason still commands actions of which the world has perhaps never given an example and commands actions the feasibility of which might very much be doubted by someone who bases everything on experience. For example, pure honesty in friendship can no less be demanded of every human being, even if up to now there might never have been an honest friend, because this duty — as duty in general — lies before all experience in the idea of a reason that controls the will through
a priori grounds.
Unless you want to deny entirely to the concept of morality all truth and reference to a possible object, you must allow that the law of morality is of such widespread significance that it must hold not just for human beings but for all rational beings in general, not just under contingent conditions and with exceptions, but with absolute necessity. Given this widespread significance and necessity, it is clear that no experience can provide the occasion to infer even the possibility of such absolutely necessary laws. For with what right can we
turn something that perhaps is only valid under the contingent conditions of humanity into a universal prescription valid for every rational nature? In addition, how should laws for the determination of our will be taken to be laws for the determination of the will of a rational being in general? And, only as laws for rational beings in general, how can they be taken to be laws for us? These questions could not be answered if moral laws were merely empirical and did not have their origin completely
a priori in pure but practical reason.
You also could not advise morality more badly than by wanting to derive it from examples. For each example of morality that is presented to me must itself first be judged according to principles of morality in order to see whether the example is worthy to serve as an original example, that is, as a model. In no way, however, can the example provide the concept of morality at the highest level. Even the Holy One of the Gospel must first be compared with our ideal of moral perfection before you can recognize Him as the Holy One. Even he says of himself: why do you call me (whom you see) good when no one is good (the archetype of the good) except the one God (whom you do not see)? Where, though, do we get the concept of God as the highest good? We get it only from the idea that reason sketches
a priori of moral perfection and that reason inseparably connects with the concept of a free will. In moral matters, imitation has
no place at all, and examples only serve as encouragement; that is, they put beyond doubt the practicability of the commands of the moral law. Examples make intuitive what the practical rule expresses more generally. But examples can never justify setting aside their true original which lies in reason and can never justify us in letting ourselves be guided by examples.
If, then, there is no genuine highest basic principle of morality, which would not have to rest independently of all experience merely on pure reason, then I believe it would not even be necessary to ask whether it would be good to present these concepts in general (in the abstract). For these concepts, together with the principles that belong to them, are established
a priori, so that presenting them in general is unnecessary provided that the knowledge of the concepts and principles is to differ from common knowledge and is to be called philosophical. But in our times this presentation might well be necessary. For if you were to take a vote as to whether pure rational knowledge apart from anything empirical — and therefore metaphysics of morals — or popular practical philosophy were preferred, you can easily guess on which side the preponderance of votes would fall.
This descent into folk concepts is certainly commendable if the ascent to the principles of pure reason has already taken place and has been attained with complete satisfaction. A successful ascent would mean
grounding the doctrine of morals first on metaphysics and later, when it is established, providing the doctrine with accessibility by popularizing it. But it is extremely silly already to want to give in to this crowd-pleasing popularizing in the first investigation on which all the correctness of the basic principles depends. Not only can this process of popularization never lay claim to the most rare merit of a true philosophical popularity since it is no art at all to be understandable by the ordinary person if you, in the process, give up all fundamental insight; the process of popularization produces a disgusting hodge-podge of mashed up observations and crack-pot principles which airheads thoroughly enjoy because it is after all something quite useful for everyday blathering. In contrast to the airheads, those people with insight feel confused and, dissatisfied, they look away, unable to help themselves. Meanwhile, philosophers see quite well through the deception, but few people pay attention when the philosophers call for a suspension of the pretended popularizing for a short time so that the philosophers may become rightly popular only after first acquiring definite insight.
You only need to look at the attempts to write about morality in that style that is thought proper. If you do, you will sometimes find the special configuration of human nature (but sometimes also the idea of a rational nature in general), now perfection, now happiness,
here moral feeling, there the fear of God, something of this, something of that, in a wondrous mixture. All the while, it never occurs to anyone to ask whether the principles of morality are even to be looked for anywhere in the knowledge of human nature (which we can still only get from experience). It also occurs to no one to ask whether, if the principles are not to be found in human nature — if, instead, the principles are to be found fully
a priori, free from anything empirical, simply in pure rational concepts and nowhere else to even the slightest degree — it would be better to form a plan to separate off this investigation completely as pure practical philosophy or (if a name much decried may be used) as metaphysics* of morals. This separation would allow the investigation by itself alone to be brought to its full completeness and allow the public, which demands popularity, to be put off until the investigation is finished.
But a metaphysics of morals that is mixed with no anthropology, with no theology,
* You can, if you want, (just as pure mathematics is distinguished from applied mathematics, and pure logic is distinguished from applied logic, therefore) distinguish pure philosophy of morals (metaphysics) from applied (namely to human nature) philosophy of morals. By using this nomenclature, you are also reminded right away that moral principles must not be grounded on the peculiarities of human nature. Instead, moral principles must be
a priori and independent. But, though not grounded on human nature, the moral principles must still be of such a kind that it remains possible to derive from them practical rules for every rational nature and therefore for human nature.
with no physics, or hyperphysics, still less with occult qualities (which you could call hypophysical), is not only an indispensable substrate for all securely established theoretical knowledge of duties, but it is at the same time a metaphysics desired because of its great importance for the actual fulfillment of moral prescriptions. For the representation of duty, pure and unmixed with any foreign additions of empirical stimuli, and in general the representation of the moral law, has an influence on the human heart so much more powerful than any other incentive* that you might summon up from the empirical field. The representation has this influence on the heart by way of reason alone (and it is in this way that reason first becomes aware that it can by itself also be practical). This influence is so strong that reason, conscious of its dignity, despises empirical incentives and little by little can become their master. In place of this pure metaphysics of morals, a mixed doctrine of morals, which is put together from incentives of feelings and inclinations and at the same time from rational concepts,
* I have a letter from the late excellent Sulzer. In this letter, he asks me what the cause might be that would explain why the teachings of virtue, however much they have that is convincing to reason, nevertheless accomplish so little. My answer was delayed by my preparations to make it complete. But the answer is nothing other than that the teachers of virtue themselves have not brought their concepts into purity and have, in wanting to make the medicine good and strong, looked around everywhere for motives for moral goodness, only to wind up spoiling the medicine. For the most ordinary
must make the mind waver between motives that cannot be brought under any principle and that only coincidentally lead to the good and more often lead to the bad.
The following is evident from what has been said: that all moral concepts have their seat and origin fully
a priori in reason, and this is the case in the most ordinary human reason just as it is in the case of a reason that is intellectually curious to the highest degree; that moral concepts cannot be abstracted from any empirical cognition and therefore from any merely contingent cognition; that it is just in the purity of the origin of the moral concepts that their dignity to serve us as the highest practical principles lies; that, each time you add something empirical to the principles, you also subtract just as much from the genuine influence and unlimited worth of the actions done from those principles; that it is not only of the greatest necessity for theoretical purposes, when it is merely a matter of intellectual curiosity,
observation shows that, if you represent an action of integrity, showing how it, separated from any intention of any advantage in this or another world, was done with a steadfast soul even under the greatest temptation of need or of enticement and showing how it leaves far behind itself and eclipses every similar action that was affected in even the least way by a foreign incentive, then that representation of the action lifts the soul and arouses the wish to be able to act in such a way, too. Even fairly young children feel this uplifting impression, and you should never represent duties to them in any other way.
but it is also of the greatest practical importance to get practical reason's concepts and laws from pure reason, to present them pure and unmixed. Indeed, it is of the greatest practical importance to determine the extent of this whole practical or pure rational knowledge, that is, to determine the whole faculty of pure practical reason. In determining this, however, the principles are not to be made to depend on the special nature of human reason in the way that speculative philosophy does permit this dependence and sometimes even finds necessary. Instead, because moral laws are to be valid for every rational being in general, moral laws are to be derived from the universal concept of a rational being in general. By means of this derivation, all of morals, which requires anthropology for its application to human beings, is first presented completely independently of anthropology as pure philosophy, that is, presented first as metaphysics (which is quite possible in this kind of knowledge that is separated from anything empirical). Without possessing this presentation of pure philosophy, it would certainly be pointless to determine for judgments arising from intellectual curiosity what precisely the moral aspect of duty is in everything that conforms with duty. Not only would that determination be pointless, but without that metaphysical presentation it would be impossible to base morals on their genuine principles even for the merely ordinary and practical use of morals in, to give a particular example, moral instruction. As a result, without this derivation of all morals in a metaphysics of morals, it would be impossible to raise people to have pure moral dispositions and impossible to implant these dispositions in their minds for the highest good of the world.
By natural steps we have already progressed in this work from ordinary moral judgment (which is here very worthy of respect) to the philosophical. But additional natural steps are needed now in order to progress from a popular philosophy, which goes no further than it can get by groping about by means of examples, up to metaphysics (which does not let itself be held back further by anything empirical since it has to size up all the contents of rational knowledge of this kind, going in any case up to ideas, where even examples desert us). We must follow the practical rational faculty from its universal rules of determination up to the place where the concept of duty springs from that faculty and then we must clearly present that faculty.
Each thing in nature works according to laws. Only a rational being has the capacity to act according to the representation of law, that is, according to principles, or has a will. Since reason is required for the derivation of actions from laws, the will is nothing other than practical reason. If reason unfailingly controls the will, then the actions of such a being that are recognized as objectively necessary are also subjectively necessary actions. That is to say, the will is a faculty to choose only what reason, independently of inclination,
recognizes as practically necessary, that is, recognizes as good. But if reason by itself alone does not have sufficient control over the will, if the will is still a slave to subjective conditions (such as certain incentives) that do not always agree with the objective conditions, if, in short, the will in itself is not fully in conformity with reason (as is actually the case with human beings), then the actions that are objectively recognized as necessary are subjectively contingent. The determination or directing of such a will according to objective laws is necessitation; that is, the relation of objective laws to a will that is not thoroughly good is represented as the steering of the will of a rational being that listens to reason but that, according to the nature of its will, does not necessarily follow what it hears.
The representation of an objective principle, insofar as it is necessitating for a will, is called a command (of reason), and the formula of the command is called an imperative.
All imperatives are expressed through an ought. Through this ought, imperatives show the relation of an objective law of reason to a will that, according to its subjective makeup, is not necessarily determined or directed by the ought (a necessitation). These imperatives say that it would be good to do or not do something, but
they say it to a will that does not always do something just because it has been told that it is a good thing to do. Practical good, however, guides the will by means of representations of reason and therefore does not guide it by subjective causes but rather by objective causes, that is, by reasons that are valid for every rational being as such. Practical good is distinguished from the pleasant. They are different in that the pleasant exercises influence on the will only by means of sensation from mere subjective causes that hold only for the senses of this or that person, and the pleasant does not exercise influence on the will as a principle of reason that holds for everyone.*
* The dependence of the faculty of desire on sensations is called inclination, and so this always indicates a need. The dependence of the will, however, on principles of reason is called an interest. This, therefore, only occurs in the case of a dependent will that of itself is not always in conformity with reason; in the case of a divine will, you cannot think of an interest. But even the human will can take an interest in something without acting from interest. The first, taking an interest, signifies a practical interest in the action. The second, acting from interest, signifies a pathological interest in the object of the action. The first shows only dependence of the will on principles of reason in themselves. The second shows a dependence of the will on principles of reason that benefit inclination; in this second case, reason only furnishes a practical rule that shows how the needs of inclination might be satisfied. In the first case, the action interests me. In the second case, the object of the action interests me (insofar as I find that object pleasant). In the first section we saw the following: that, in the case of an action from duty, none of our attention must be given to the interest in the object of the action; instead, all our attention must be focused on interest in the action itself and on the action's principle in our reason (on the law).
So a completely good will would stand just as much under objective laws (of the good). But such a will would not, by standing under objective laws, be able to be represented as necessitated to actions that are in conformity with law. Such a will could not be represented as necessitated because such a will of itself, according to its subjective makeup, can only be controlled by the intellectual representation of the good. No imperatives, therefore, hold for the divine will and in general for a holy will; the ought is here out of place because the willing is already of itself in necessary agreement with the law. Imperatives are, therefore, only formulas that express the relation of objective laws of willing in general to the subjective imperfection of the will of this or that rational being, for example to the subjective imperfection of the human will.
Now, all imperatives command either hypothetically or categorically. The former, hypothetical imperatives, represent the practical necessity of a possible action as a means to get something else that you want (or that you might possibly want). The categorical imperative would be the imperative which represented an action as objectively necessary in itself, without reference to any other end.
Because each practical law represents a possible action as good and therefore, for a subject practically directed by reason, as necessary,
all imperatives are formulas for the specification of an action that is necessary according to the principle of a will that is good in some way. If now the action would be good merely as a means to something else, then the imperative is hypothetical. If the action is thought of as good in itself, and therefore as necessary in a will that is itself in conformity with reason, reason serving as the will's principle, then the imperative is categorical.
So the imperative says which action that is possible through me would be good. The imperative represents the practical rule in relation to a will that does not immediately do an action because the action is good. The will does not do it partly because the subject does not always know that the action would be good and partly because, even if the subject did know the action would be good, the subject's maxims could still be at odds with the objective principles of a practical reason.
So the hypothetical imperative only says that an action would be good for some possible or actual purpose. In the first case, about a possible purpose, the hypothetical imperative is a problematically practical principle. In the second case, about an actual purpose, the hypothetical imperative is an assertorically practical principle. The categorical imperative, which declares the action to be objectively necessary in itself without reference to any purpose, that is, even without any other end, holds as an absolutely necessary (practical) principle.
Something that is only possible through the powers of some rational being is something you can also think of as a possible purpose of some will. Therefore, there are in fact infinitely many principles of action, provided that the action is thought of as necessary in order to accomplish a possible purpose that the action works to bring about. All sciences have some practical part that consists of problems claiming that some end or goal is possible for us and that consists of imperatives specifying how that end or goal can be reached. These imperatives, therefore, can in general be called imperatives of skill. The question here is not at all about whether the end is rational and good, but instead about what you must do in order to reach the end. The prescriptions that the doctor uses in order to make her patient one hundred percent again are of equal worth with the prescriptions that a poisoner uses to bump off her victim insofar as each set of prescriptions serves perfectly to accomplish its purpose. Because you do not know when you are young what ends you may stumble across later in life, parents seek above all to have their children learn lots and lots of things and provide for skill in the use of means to all kinds of arbitrary ends. The parents cannot identify any of these optional ends as an end that in the future will become an actual goal of their child, but they are all still ends that it is possible that their child might one day have. The parents' concern is so great that they typically neglect to shape and to correct their children's judgments about the worth
of things that the children would perhaps like to make into ends.
There is, nevertheless, one end that you can presuppose as actual in the case of all rational beings (so far as imperatives apply to them, namely, as dependent beings). So there is a purpose that all rational beings not only merely can have but also a purpose which you can safely presuppose that all rational beings do have according to a natural necessity, and this is the purpose that all rational beings have with regard to pursuing happiness. The hypothetical imperative, which represents the practical necessity of action as a means to the advancement of happiness, is assertoric. You must not present this kind of imperative merely as necessary for an uncertain, merely possible purpose, but you must present the imperative as necessary for a purpose which you can safely and
a priori presuppose in the case of every human being; and you can safely so presuppose this because the purpose belongs to the nature of any human being. Now, you can call skill in the choice of means to your own greatest well-being prudence* in the narrowest sense of the word. Therefore,
* The word "prudence" has two senses. In one sense, it goes by the name "worldly prudence." In the second sense, the word bears the name "private prudence." The first sense, worldly prudence, is the skill of a human being to have influence on others in order to use them for the human being's own purposes. The second sense, private prudence, is the insight to unite all these purposes for the human being's own lasting advantage. The latter, private prudence, is properly the one to which even the worth of the former, worldly prudence, is traced back. Whoever is prudent in the first worldly sense but not in the second private sense is someone of whom you could more appropriately say: she is clever and cunning, but, on the whole, still not prudent.
the imperative which refers to the choice of means to your own happiness, that is, the prescription of prudence, is always hypothetical; the action is commanded not absolutely but only as a means to some other purpose.
Finally, there is an imperative which immediately commands certain conduct and which does not lay down as a condition for the imperative's basis some other purpose that is to be achieved by that conduct. This imperative is categorical. This imperative does not deal with the matter of action and the consequences of action. Instead, this imperative deals with the form and the principle from which the action follows, and the action's essential good consists in the disposition, whatever the consequences turn out to be. This imperative may be called the imperative of morality.
Willing according to these three kinds of principles is also clearly distinguished by the dissimilarity of the necessitation in the will. In order to make this stand out now, too, I think that you would classify these three kinds of principles most appropriately in their order if you said it in this way: the principles are either rules of skill, or counsels of prudence, or commands (laws) of morality. For only the law carries with it the concept of an unconditional necessity that is definitely objective and therefore universally valid. Furthermore, commands are laws
that must be obeyed, that is, must be obeyed even against inclination. Advice certainly contains necessity, but this necessity can hold only under a merely subjective contingent condition. This condition is whether this or that human being counts this or that as belonging to her happiness. In contrast, the categorical imperative is limited by no condition and, as absolutely necessary even though also practically necessary, can quite properly be called a command. You could also call the first kind of imperative technical (belonging to art), the second pragmatic* (belonging to well-being), the third moral (belonging to free conduct in general, that is, to morals).
The question now arises: how are all these imperatives possible? This question does not demand to know how we are to understand the performance of an action that the imperative commands. Instead, the question just demands to know how we are to understand the necessitation of the will, which the imperative expresses when it tells us what to do. How an imperative of skill is possible really requires no special discussion. Whoever wills the end, wills (to the extent that reason has
* It seems to me that the proper meaning of the word "pragmatic" can be defined most precisely in this way. For those sanctions are called pragmatic which flow, not out of the right of states as necessary laws, but which flow out of the provision for the general welfare. A history is pragmatic when it makes us prudent, that is, when it teaches the world how it can take better — or at least just as good — care of its advantage than the world did in previous eras.
decisive influence over her actions) also the indispensable means that are necessary to achieve the end and that are in her power to do. This proposition is, as concerns willing, analytic; for, in the willing of an object as my effect, my causality as an acting cause, that is, the use of means, is already thought, and the imperative already extracts the concept of actions necessary to achieve this end from a willing of this end. (To be sure, synthetic propositions are needed in order to figure out the means to achieve an intended purpose, but these synthetic propositions have to do with making the object of the action actual and not with grounding the act of will.) Mathematics, of course, teaches only through synthetic propositions that, in order to divide a line in accordance with a reliable principle into two equal parts, I must make two intersecting arcs from the endpoints of the line. But if I know that an intended effect can only occur by such an action, then the following proposition is analytic: if I fully will the effect, then I also will the action that is required to achieve the effect. This proposition is analytic because thinking of something as an effect that is possible for me to bring about in a certain way is exactly the same as thinking of myself as acting in the same bringing-about way with respect to that same something.
The imperatives of prudence would, if it were only as easy to give a well-defined concept of happiness,
agree completely with the imperatives of skill, and the imperatives of prudence would likewise be analytic. For the following could be said about imperatives of prudence just as well as it is said about imperatives of skill: who wills the end also wills (necessarily in accordance with reason) the sole means to the end that are in her power to do. But it is unfortunate that the concept of happiness is such an ill-defined concept that, although each human being wishes to achieve happiness, she can still never say in a definite and self-consistent way what she really wishes and wants. The cause of this wishy-washiness is this: that all the elements that belong to the concept of happiness are one and all empirical, that is, all the elements must be borrowed from experience; that, despite the empirical basis of the concept of happiness, the idea of happiness requires an absolute whole, a maximum of well-being, in my present and every future condition. Now, it is impossible that the most insightful and at the same time most capable, but still finite being, could make for itself a well-defined concept of what she here really wants. If she wants riches, how much worry, envy and intrigue might she bring down on her own head? If she wants lots of knowledge and insight, they might just make her eyes sharper so that she can see all the more dreadfully the evil that currently is hidden from her but that she cannot avoid; or they might just burden her eager desires, which already trouble her enough, with even more needs. If she wants a long life, then who can guarantee her
that it will not be a long misery? If she at least wants health, how often has discomfort of the body kept her from excess into which unlimited health would have let her fall, and so on? In short, she is not able to figure out with complete certainty according to any basic principle what will make her truly happy, for figuring this out would require omniscience. So you cannot act according to well-defined principles so as to be happy. You can only act according to empirical counsels, for example, counsels to diet, to be thrifty, to be courteous, to be reserved and so on. Experience teaches us that these counsels on the average do most to promote our well-being. From these considerations about happiness, the following can be concluded: that the imperatives of prudence, strictly speaking, do not command at all, that is, the imperatives of prudence cannot present actions objectively as practically necessary; that the imperatives of prudence are to be held to be counsels (
consilia) rather than to be commands (
praecepta) of reason; that the problem of determining reliably and universally which action will promote the happiness of a rational being is completely insoluble; that, therefore, no imperative with a view to happiness is possible which in the strict sense would command you to do what will make you happy, and such an imperative is not possible because happiness is not an ideal of reason but instead an ideal of imagination. This imagination rests merely on empirical grounds, and it is pointless for you to expect that these empirical grounds should specify an action by which a totality of an
in fact infinite series of consequences would be attained. This imperative of prudence would, nevertheless, if you assume that the means to happiness could be accurately specified, be an analytic practical proposition. For the imperative of prudence is distinguished from the imperative of skill only in this: in the case of the latter, the imperative of skill, the end is merely possible, while in the case of the former, the imperative of prudence, the end is given as actual. But, since both kinds of imperative merely command the means to something that you assume someone wants as an end, the imperative, which commands the willing of the means for someone who wants the end, is in both cases analytic. So there is also no difficulty with regard to the possibility of such an imperative of prudence.
On the other hand, the question of how the imperative of morality is possible is without doubt the only question in need of a solution. For the imperative of morality is not hypothetical at all and so the objectively represented necessity can be based on no presupposition, as in the case of the hypothetical imperatives. But when thinking about the imperative of morality it should always be kept in mind that whether there is any such imperative of morality is a claim that can be established by no example and that therefore cannot be established empirically. Instead of looking to examples, it should also always be kept in mind that care must be taken with anything that appears categorical because it might yet be hypothetical in a hidden way. For example, when it is said that you should not make deceitful promises, and you assume that the necessity of complying with this is definitely not merely advice to avoid
some other evil, what is said might in a hidden way be saying that you should not make lying promises so that you do not, when your deceitful behavior becomes public knowledge, ruin your reputation. An action of this kind, which appears to be based on a categorical imperative but might actually be based on a hypothetical imperative in hiding, must be considered to be bad in itself, and so the imperative prohibiting the action is categorical. So in no example can you prove with certainty that the will is controlled only by the law and not by any other incentive, even though it might appear as if only the law is controlling the will; for it is always possible that fear of embarrassment, perhaps also vague worries about other dangers, might secretly have an influence on the will. Who can prove through experience the nonexistence of a cause since experience teaches nothing further than that we do not perceive the cause? If there were such secret influences on the will, the so-called moral imperative, which, as moral, appears categorical and unconditional, would in fact only be a pragmatic prescription that makes us attentive to our advantage and merely teaches us to take care of this advantage.
So we will have to investigate the possibility of a categorical imperative completely
a priori since we do not here have the advantage that the actuality of the categorical imperative is given in experience. If we had that advantage, we would need the possibility of the categorical imperative not to establish it but merely to explain it. Though we lack that advantage, this much is provisionally evident: that the categorical imperative alone
reads as a practical law; all other imperatives can indeed be called principles of the will, but they cannot be called laws. The categorical imperative alone is a practical law, while all other imperatives are only principles of the will, because whatever is necessary to do in order merely to attain an arbitrary end is something that can itself be considered as contingent, and we can be free of the prescription if we give up the purpose; on the other hand, the unconditional command leaves the will no wiggle room with regard to the opposite, and therefore the unconditional command alone carries with it the necessity which we demand of the law.
Secondly, in the case of this categorical imperative or law of morality, the reason for the difficulty (of looking into the possibility of such an imperative or law) is also very great. A categorical imperative is a synthetic practical proposition*
a priori, and, since to look into the possibility of propositions of this kind is so difficult in theoretical knowledge, it is easy to see that it will be no less difficult to look into the possibility of synthetic propositions
a priori in practical knowledge.
* Without presupposing a condition from any inclination, I connect
a priori a deed with the will. Because the connection is
a priori, the connection is also necessary (although only objectively necessary, that is, the connection would hold up only under the idea of a reason that had full control over all subjective motives). So this is a practical proposition which does not derive the willing of an action analytically from another already presupposed willing of an action (for we have no such perfect will). Instead, the practical proposition immediately connects the willing of an action with the concept of the will of a rational being, the willing of the action being something that is not contained in the concept of the will of the rational being.
In tackling this problem of the possibility of a categorical imperative, we want first to see whether the mere concept of a categorical imperative might also provide the formula of a categorical imperative, the formula containing the proposition which alone can be a categorical imperative; for how such an absolute command is possible, even if we also know how the command reads, will still require special and difficult effort, which we, however, put off until the last section.
If I think of a hypothetical imperative in general, then I do not know in advance what the imperative will contain until the imperative's condition is given. If, however, I think of a categorical imperative, then I know at once what the imperative contains. For, since the imperative contains, besides the law, only the necessity of the maxim* to be in conformity with this law, and the law contains no condition to which was limited, nothing remains except the universality of law in general to which the maxim of the action is to conform,
* A maxim is the subjective principle of acting and must be distinguished from the objective principle, namely from the practical law. The former, a maxim or subjective principle, contains the practical rule which reason specifies in accordance with the conditions of the subject (often the subject's ignorance or also the subject's inclinations). So a maxim is the basic principle according to which the subject acts. The law, however, is the objective principle; it is valid for every rational being and is the basic principle according to which every rational being ought to act. That is, the objective principle, the practical law, is an imperative.
and it is this conformance alone which the imperative properly represents as necessary.
So there is only one categorical imperative and it is just this: act only according to that maxim through which you can at the same time will that it become a universal law.
Now, if all imperatives of duty can be derived, as from their principle, from this one imperative, then, even though we leave it unsettled whether or not in general what we call duty is an baseless concept, we will still at least be able to indicate what we think by the concept of duty and what this concept means.
Because the universality of the law according to which effects occur constitutes what is properly called nature in the most general sense (according to nature's form), that is, the existence of things so far as the existence is determined according to universal laws, the universal imperative of duty could also be expressed like this: so act as if the maxim of your action were to become through your will a universal law of nature.
Now we will list some duties according to the usual division of duties into duties
to oneself and to other human beings, and into perfect and imperfect duties*.
1) A person, who is disgusted with life because of a series of misfortunes that has grown into hopelessness, is still sufficiently in possession of her reason that she is able to ask herself whether it is not wholly contrary to duty to oneself for her to commit suicide. Now she tests whether her maxim of her action could indeed be a universal law of nature. But her maxim is: from self-love, I make it my principle to shorten my life when continuing to live threatens more misery than pleasantness. All that remains is the question whether this principle of self-love could be a universal law of nature. But you then soon see that a nature whose law it was, through the same feeling that is
* You must here be sure to note that I reserve the division of duties for a future metaphysics of morals. So this division only stands here as arbitrary (in order to order my examples). Moreover, by a perfect duty, I here understand a duty that allows of no exception that is to the advantage of inclination, and regarding such duties I have not merely outer but also inner perfect duties. This way of understanding perfect duty runs counter to the terminology used in the schools, but I do not intend to defend it here because for my purpose it is all the same whether you do or do not concede it to me.
to urge on the blossoming of life, to destroy life would contradict itself and would not endure as a nature. So that maxim could not possibly exist as a universal law of nature and consequently would wholly conflict with the highest principle of all duty.
2) Another person sees herself forced by need to borrow money. She very well knows that she will not be able to repay the money, but she also sees that nothing will be lent to her if she does not firmly promise to pay the money back at a specific time. She feels like making the promise; but she still has enough of a conscience to ask herself: is it not impermissible and contrary to duty to get out of difficulty in this way? Assuming that she still resolves to make the promise, then her maxim of action would read like this: when I believe myself to be in need of money, I will borrow money and promise to repay it even though I know that the money will never be repaid. Now, this principle of self-love or of one's own advantage is perhaps quite compatible with my whole future well-being, but the question now is whether the principle is right. So I change the unreasonable demand of self-love into a universal law and put the question like so: how would things then stand if my maxim were to become a universal law? Putting it this way, I now see at once that the maxim could never hold as a universal law of nature and be compatible with itself, but
must necessarily contradict itself. For the universality of a law, that everyone, accordingly as she believes herself to be in need, can promise whatever she pleases with the intention of not keeping the promise, would make the promise itself, and perhaps the end to be achieved by making the promise, impossible. The promise would be impossible because no one would believe that anything was promised to her; instead, such utterances of promising would be ridiculed as idle pretense.
3) A third person finds in herself a talent which by means of some cultivation could make her a human being useful for all kinds of purposes. But she sees herself in comfortable circumstances and prefers to indulge in pleasure rather than to strive to enlarge and improve her fortunate natural predispositions. But still she asks whether, besides agreeing in itself with her tendency to amusement, her maxim of neglecting her natural gifts also agrees with what is called duty. Upon asking this, she now sees for sure that a nature could always endure according to such a natural law even if the human being (like the South Sea Islanders) let her talents rust and was intent on devoting her life merely to idleness, amusement, casual sex — in a word, to enjoyment. But she cannot possibly will that this law become a universal law of nature or that such a natural law be put in us by natural instinct.
For as a rational being she necessarily wills that all capacities in her be developed because they, after all, are given to her and serve her for all kinds of purposes.
Yet a fourth, for whom things are going well, meanwhile sees that other people (whom she could also easily help) have to struggle with great difficulties. She thinks: what's that to me? May each person just be as happy as heaven allows or as happy as she can make herself. I will not take anything from her or even envy her. But I do not feel like contributing anything to her well-being or to come to her assistance in times of need! Now, of course, if such a way of thinking became a universal law of nature, the human race could quite well endure. Indeed, it could endure even better than it does when everyone blathers on nonstop about compassion and kindness and even occasionally tries to put these into practice but, on the other hand, also tries to cheat, sell the right of the human being, or otherwise violate that right. But, although it is possible that a universal law of nature could quite well endure according to that maxim, it is nevertheless impossible to will that such a principle hold everywhere as a universal law of nature. For a will that resolved to will according to that maxim would conflict with itself. Such a will would conflict with itself because many cases can arise in which a person needs the love and compassion of others and in which the person, through such a natural law that sprung from the person's own will,
would rob herself of all hope for the assistance that she wants.
These, then, are some of the many actual — or that we at least take to be actual — duties whose spinning off from the one principle cited above is clear. You must be able to will that a maxim of your action become a universal law; this is the canon for morally judging action in general. Some actions are constituted in such a way that their maxim cannot without contradiction even be thought as a universal law of nature. Even more implausible is that you could will that the maxim of such actions should become such a universal law of nature. In the case of other actions, that inner impossibility is definitely not present, but to will that the actions' maxim be elevated to the universality of a law of nature is still impossible because such a will would contradict itself. You can easily see that the first kind of actions, having maxims that are unthinkable as universal laws, conflict with strict or narrower (never slackening) duty and that the second kind of actions, having maxims that are unwillable as universal laws, conflict with wide (meritorious) duty. Consequently, you can also easily see that these examples thoroughly present all duties, as far as the kind of obligation (not the object of the dutiful action) is concerned, as dependent on the one principle.
If we now pay attention to ourselves whenever we transgress a duty, we find that we
actually do not will that our maxim should become a universal law, for that is impossible for us. Instead, the opposite of the maxim should rather remain a law generally. We only take the liberty for ourselves, or (even only for this one time) to the advantage of our inclination, to make an exception to the law. Consequently, if we were to weigh everything from one and the same point of view, namely that of reason, then we would encounter a contradiction in our own will. The contradiction would be that a certain principle should be objectively necessary as a universal law and yet subjectively should not hold universally but should permit exceptions. But since we at one time consider our action from the point of view of a will wholly in accord with reason, but then also consider the very same action from the point of view of a will affected by inclination, there is actually no contradiction here. Though there is no contradiction, there is an opposition of inclination to the prescription of reason (
antagonismus). Through this opposition, the universality of the principle (
universalitas) is changed into a mere generality (
generalitas). By means of this transformation, the practical principle of reason is to meet the maxim half way. Now, although this resolution of the opposition cannot be justified by our own judgment when our judgment is used impartially, the resolution still proves that we actually do acknowledge the validity of the categorical imperative and that we (with all respect for the imperative) only permit ourselves a few,
as it seems to us, exceptions that are minor and forced from us.
So we have at least shown as much as the following. We have shown that if duty is a concept that is to contain meaning and actual lawgiving for our actions, then this duty can only be expressed in categorical imperatives and can in no way be expressed in hypothetical imperatives. We have also clearly and distinctly set forth for every use, which is already to have done a great deal, the content of the categorical imperative, which must contain the principle of all duty (if there were to be such a principle at all). But, still, we are not so far along as to prove
a priori that there actually is an imperative of this kind, that there is a practical law which commands absolutely and by itself without any incentives, and that following this law is duty.
With the aim of obtaining this
a priori proof, it is of the utmost importance to be warned against your wanting to derive the reality of this principle from the special quality of human nature. For duty is to be the practical-unconditional necessity of action. So duty must hold for all rational beings (and only to such beings can an imperative apply at all) and only for this reason can duty be a law for all human wills. Whatever, on the other hand,
is derived from the special natural predispositions of the human being is something that can provide a maxim for us. Whatever is derived from certain feelings and propensities is something that can provide a maxim for us. Indeed, whatever is derived, where possible, from a special tendency peculiar to human reason and not necessarily valid for the will of every rational being is something that can definitely provide a maxim for us, but it is not something that can provide a law for us. All these predispositions, feelings, and tendencies can provide a subjective principle according to which we may act and may have a propensity and inclination to act, but they cannot provide an objective principle according to which we are directed to act even if all our propensity, inclination and natural makeup were against it. What is more, the fewer the subjective causes of a command and the more the subjective causes against it, the more the sublimity and inner dignity of the command in a duty is shown. This highlighting of sublimity and dignity occurs without these subjective causes weakening even in the least the necessity of the law or taking anything away from the validity of the law.
Here we now see philosophy put in fact in a precarious position. This position is to be firm even though it is neither suspended from anything in heaven nor supported by anything on earth. This is where philosophy is to prove her purity as caretaker of her own laws, not as the spokeswoman of what an implanted sense whispers to philosophy or as the spokeswoman of who knows what whispering tutelary nature. Though this whispering sense and whispering nature might always be better than nothing at all, they can still never provide basic principles which reason dictates and which must throughout have their origin fully
a priori and, along with this
a priori origin, at the same time have their commanding authority.
a priori basic principles expect nothing from the inclination of the human being. Instead, they expect everything from the supreme power of the law and from the respect owed to the law. If their expectation is not met, then the human being is condemned to self-contempt and inner abhorrence.
So everything that is empirical is not only wholly unsuitable as an addition to the principle of morality, but everything empirical is highly damaging to the purity of morals themselves. In this purity of morals is found the proper worth, raised above all price, of an absolutely good will. This purity of morals consists just in this: that the principle of action is free from all influences of contingent grounds which can only be provided by experience. You also cannot too frequently issue too many warnings against this carelessness and even base way of thinking which searches for the principle of morality among empirical motives and laws. These warnings cannot be too many or too frequent because human reason, in its weariness, gladly rests on this pillow of empirical mush, and, in a dream of sweet illusions (which, after all, allows reason to embrace a cloud instead of Juno), substitutes for morality a bastard patched up from limbs of completely different ancestry. This patched up bastard, masquerading as morality, looks like everything that you want to see in it, except like virtue for those who have once beheld virtue in her true form*.
* To behold virtue in her proper form is nothing other than to exhibit morality stripped of all admixture of sensuous
So the question is this: is it a necessary law for all rational beings that they judge their actions always according to maxims that they themselves can will as maxims that should serve as universal laws? If there is such a necessary law, then it must (completely
a priori) already be connected with the concept of the will of a rational being in general. But in order to discover this connection, you must, even though you would rather not, take a step out into metaphysics. In particular, you must take a step out into the metaphysics of morals, which covers an area of metaphysics that is different from the area covered by speculative philosophy. In a practical philosophy, it is not our concern to assume grounds for what happens but rather laws for what ought to happen even if it never does happen; that is, in a practical philosophy our concern is with objective-practical laws. In a practical philosophy, we have no need to undertake an investigation into the reasons why something pleases or displeases us, how the enjoyment of mere sensation differs from taste, and whether taste is different from a universal satisfaction of reason. We have no need to investigate what the feeling of pleasure and displeasure rests on, and how from this feeling eager desires and inclinations arise, and then how, through the cooperation of reason, from these desires and inclinations maxims
and all fake decorations of reward or of self-love. By means of the slightest exercise of one's reason, as long as that reason has not been completely ruined for all abstraction, everyone can easily become aware of how much virtue then eclipses everything else that appears enticing to inclinations.
arise. For all that belongs to an empirical doctrine of the soul, which would make up the second part of the doctrine of nature if you consider it as philosophy of nature as far as it is grounded on empirical laws. Here, however, we are talking about objective-practical laws and are therefore talking about the relation of a will to itself so far as the will controls itself merely through reason. When this happens, when the will controls itself merely through reason, everything that has reference to the empirical falls away by itself. Everything that is empirical falls away because if reason by itself alone controls behavior (and the possibility of this kind of control is exactly what we now want to investigate) then reason must necessarily execute this control in an
a priori way.
The will is thought as a capacity to direct itself to act according to the representation of certain laws. And such a capacity can only be found in rational beings. An end is what serves the will as an objective ground of the will's self-direction. This end or goal, if it is given only by reason, must hold equally for all rational beings. On the other hand, a means is what contains merely the ground of possibility of an action that has an end as its effect. The subjective ground of desiring is an incentive; the objective ground of willing is a motive; thus the difference between subjective ends, which rest on incentives, and objective ends, which depend on motives that
hold for every rational being. Practical principles are formal if they abstract from all subjective ends. But practical principles are material if they make subjective ends, and therefore certain incentives, their basis. The ends that a rational being arbitrarily aims at as effects of her action (material ends) are one and all only relative. For only the ends' mere relation to a particularly fashioned faculty of desire of the subject gives the ends their worth. This worth can therefore provide no valid and necessary universal principles, that is, practical laws, for all rational beings or for every case of willing. All these relative ends are therefore only the ground of hypothetical imperatives.
Suppose, however, that there were something whose existence in itself has an absolute worth, something which as an end in itself could be a ground of well-defined laws. If that were supposed, then the ground of a possible categorical imperative, that is, the ground of a practical law, would lie in that something and only in that something.
Now I say: the human being and in general every rational being exists as an end in itself, not merely as a means for the optional use of this or that will. Instead, the human being must in all its actions, whether the actions are directed at the human being performing the action or are directed at other rational beings,
always be considered at the same time as an end. All objects of inclinations have only a conditional worth; for, if the inclinations and needs grounded on them did not exist, then their object would be without worth. But inclinations themselves, as sources of need, are very far from having an absolute worth so that they would be wished for in themselves. Instead, it must be the universal wish of every rational being to be completely free of inclinations. So the worth of any objects to be attained through our action is always conditional. The beings whose existence rests not, to be sure, on our will but on nature still have, if they are beings without reason, only a relative worth as means and are therefore called things. On the other hand, rational beings are called persons because their nature already marks them out as ends in themselves, that is, as something that may not be used merely as a means, and therefore their nature as persons limits any choice about how to act (and is an object of respect). So persons are not merely subjective ends whose existence as an effect of our action has a worth for us. Instead, persons are objective ends, that is, things whose existence in itself is an end. In particular, their existence in itself is an end that cannot be replaced by some other end in such a way that their existence is to serve the substituted end merely as a means. Another end cannot be put in place of their existence as an end because, if the substitution could occur, no absolute worth at all would be found anywhere; but if all
worth were conditional and therefore contingent, then no highest practical principle for reason could be found anywhere.
So if there is to be a highest practical principle and, with regard to the human will, a categorical imperative, then it must be a principle that, from the thought or representation of what is necessarily an end for everyone because it is an end in itself, constitutes an objective principle of the will and so can serve as a universal practical law. The ground of this principle is: rational nature exists as an end in itself. The human being necessarily conceives of her own existence in this way. Limited to the individual in this way, the principle is thus a subjective principle of human actions. But every other rational being also conceives of its existence in this way on the very same rational ground that also holds for me*. Hence, the principle is at the same time an objective principle from which, as a highest practical ground, all laws of the will must be able to be derived. So the practical imperative will be the following: act in such a way that you treat humanity, in your own person, as well as in the person of every other, always at the same time as an end, never
* Here I set this proposition out as a postulate. In the last section you will find the reasons for the proposition.
merely as a means. We want to see if this principle can be worked out.
If we stay with the previous examples, then we will have the following.
Firstly, as regards the concept of necessary duty toward oneself, a person who has suicide in mind will ask herself whether her action can be compatible with the idea of humanity as an end in itself. If she, in order to escape from a troublesome situation, destroys herself, then she makes use of a person merely as a means for maintaining a tolerable situation until the end of life. But the human being is not a thing and therefore is not something that can be used merely as a means. Instead, the human being must in all her actions always be considered as an end in herself. So I can dispose of nothing about the human being in my person, cannot maim her, corrupt her, or kill her. (Although it would help to avoid any misunderstanding, I have to forego a more precise specification of this basic principle, for example, of how the principle would apply to the amputation of limbs in order to save myself, how it would apply to cases in which I expose my life to danger in order to preserve my life, and so on; this more precise specification of the principle belongs to morals proper.)
Secondly, as concerns the necessary or owed duty to others, someone who intends to make a lying promise to others will see at once that she wants to make use of another human being
merely as a means, without the other person at the same time having the same the end. For the person whom I want to use for my purposes by making such a promise cannot possibly agree with my way of proceeding against her, and she cannot therefore contain in herself the end of my action. This conflict with the principle of duty owed to other human beings more clearly catches the eye when you bring in examples of attacks on the freedom and property of others. For then it is evident that the transgressor of the rights of human beings intends to make use of the person of others merely as a means and intends to do this without taking into consideration that the others, as rational beings, ought always to be valued at the same time as ends, that is, ought always to be valued as beings who must also be able to have in themselves the end of the very same action*.
Thirdly, with regard to the contingent (meritorious) duty to oneself, it's not enough that the
* You should not think that here the trivial:
what you do not want done to you etc. can serve as a rule of thumb for conduct or as a guiding principle. For this trivial saying is, although with various limitations, only derived from the principle of duty owed to others; it cannot be a universal law, for it does not contain the ground of duties to oneself, does not contain the ground of duties of love to others (for many a person would gladly agree that others should not do anything to benefit her if only she may be excused from showing them any kindness). And, finally, this trivial saying does not contain the ground of duties owed to one another; for the criminal would use this deficiency to argue against the judges who are punishing her, and so on.
action not conflict with the humanity in our person as an end in itself; the action must also harmonize with that humanity in our person. Now, in humanity there are predispositions to greater perfection that belong to the end of nature with regard to humanity in our subject. To neglect these predispositions would be, at most, probably compatible with the preservation of humanity as an end in itself, but neglecting them would not be compatible with the promotion of this end.
Fourthly, with regard to meritorious duty to others, the natural end that all human beings have is their own happiness. Now, humanity would no doubt endure if no one contributed anything to the happiness of others but also, in so doing, intentionally withdrew nothing from that happiness. But, if everyone does not also try, as far as she can, to promote the ends of others, then this neutrality is still only a negative and not positive harmonization with humanity as an end in itself. For the ends of a subject which is an end in itself must, as far as possible, also be my ends, if that thought of an end in itself is to have full effect in me.
This principle of humanity and of each rational nature in general as an end in itself (which is the highest limiting condition on the freedom
of action of every human being) is not borrowed from experience. First, because of the principle's universality, applying as it does to all rational beings in general, and since no experience is sufficient to say anything definite about all rational beings in general, the principle is not borrowed from experience. Secondly, the principle also is not borrowed from experience because, in the principle, humanity is not represented or thought of as an end of human beings (subjectively); that is, humanity is not represented as an object which you by yourself actually make into an end; instead, humanity is represented as an objective end which, whatever ends we might happen to have, as a law is to constitute the highest limiting condition of all subjective ends. Therefore, the principle must arise from pure reason. In particular, the ground of all practical lawgiving resides objectively in the rule and in the form of universality. This universality (according to the first principle) makes the rule capable of being a law (possibly a natural law). Subjectively, however, the ground of practical lawgiving resides in the end. The subject of all ends, however, is each rational being as an end in itself (according to the second principle). From this the third practical principle of the will, as the highest condition of the harmony of the will with universal practical reason, now follows: the idea of the will of every rational being as a will giving universal law.
According to this third practical principle of the will, all maxims which are not consistent with the will's own universal lawgiving are rejected. So the will is not only subject to the law,
but the will is subject to the law in such a way that the will must also be seen as giving law to itself; and, just because the will does give law to itself, the will must be seen as first of all subject to the law (of which the will itself can consider itself the author).
Up to now, imperatives have been modelled according to two different ways of thinking of the imperatives. One way of thinking of imperatives is to represent them as expressing a conformity of actions to law, that conformity being generally similar to a natural order. A second way represents imperatives as expressing the universal priority of the end of rational beings. Both of these ways of representing imperatives definitely excluded from the imperatives' commanding authority all admixture of any interest as an incentive. All interest was excluded precisely because the imperatives were represented as categorical; they, however, were only assumed to be categorical because you had to assume that they were categorical if you wanted to explain the concept of duty. That there are, however, practical propositions that command categorically could not itself be proved. No more than before, that there are such propositions can also not yet be proved anywhere here in this section. But one thing could still have been done, namely: that in cases of willing from duty, the renunciation of any interest — that renunciation being the specific mark distinguishing categorical imperatives from hypothetical imperatives — would be jointly indicated in the imperative itself by some specific feature that the imperative contains. This joint indication of renunciation of interest and distinction between types of imperative occurs in the present third formula of the principle, namely, in the idea of the will of each rational being as a will giving universal law.
For if we think of such a will, then, although a will that stands under laws may still be connected to this law by an interest, it is impossible for a will which itself is highest in lawgiving to be dependent to such an extent on any interest; for such a dependent will would itself require still another law that would limit the interest of the will's self-love to the condition of the interest's validity as universal law.
So the principle of every human will as a will giving universal law through all its maxims* would be quite well-suited to be a categorical imperative, if the principle were quite correct in other ways. The principle would be well-suited to be a categorical imperative because the principle, just for the sake of the idea of universal lawgiving, rests on no interest and therefore alone among all possible imperatives can be unconditional. The reason for the well-suitedness of the principle can be stated even better if we turn the proposition around: if there is a categorical imperative (that is, a law for the will of every rational being), then the imperative can only command that the rational being always act from the maxim of the being's will regarded as a will that at the same time could have itself as giving universal law
* I can here be excused from citing examples to illustrate this principle, for those examples first used in this way to illustrate the categorical imperative and its formula can all serve just the same purpose here.
as an object. For only then is the practical principle and the imperative which the will obeys unconditioned because the imperative can have no interest at all as a ground.
It is now not surprising, when we look back on all previous efforts that have ever been undertaken to discover the principle of morality, why they had to fail in every case. You saw the human being bound by its duty to laws, but it never occurred to anyone that the human being is subject only to its own, but still universal, lawgiving and that the human being is only obligated to act according to its own will which, according to nature's end, however, is universally lawgiving. For, if you conceived of the human being only as subject to a law (whichever law it might be), then this law had to carry with itself some interest as an attraction or constraint. The law had to have this attracting or constraining interest because the law did not arise from the human being's will as a law; instead, the human being's will was necessitated to act in a certain way in conformity to law by something else. But by this entirely necessary consequence, all labor expended in trying to find a highest ground of duty was irretrievably lost. For you never got duty; instead, you only got necessity of action from a certain interest. Now, this interest might be your own or another's. But in either case the imperative always had to turn out conditioned
and could not at all be suited to be the moral command. So I want to call this basic principle the principle of the autonomy of the will, in opposition to every other principle which I therefore count as heteronomy.
The concept of any rational being which must consider itself as giving universal law through all of the maxims of its will, in order to judge itself and its action from this point of view, leads to a very fruitful concept. This latter, very fruitful concept hangs on the former concept of any rational being and is the concept of an empire of ends.
But, by an empire, I understand the systematic union of different rational beings through a common law. Now, because laws determine ends according to the laws' universal validity, an empire of ends can be thought which is possible according to the above principles. But the thought of this empire of ends becomes possible in this way only if you also abstract from the personal differences of rational beings and from all content of their private ends. If you abstract in this way, then the thought of a whole of all ends (not only a whole of rational beings as ends in themselves but also of individual ends which each rational being may set for herself) in a systematic bond is possible.
For rational beings all stand under the law that each rational being is to treat itself and all other rational beings
never merely as a means, but instead always at the same time as an end in itself. But from this law, and from the treatment the law prescribes, there arises a systematic union of rational beings through common objective laws. That is, an empire arises which, because these laws have as their aim just the relation of these beings to each other as ends and means, can be called an empire of ends (which is, admittedly, only an ideal).
A rational being, however, belongs to an empire of ends as a member, if the rational being is, of course, universally lawgiving in the empire but also is itself subject to these laws. A rational being belongs to an empire of ends as head, if the rational being as lawgiving is subject to the will of no other.
The rational being must always consider itself as lawgiving in an empire of ends possible through freedom of the will, whether it be as member or as head. A rational being cannot keep the seat of the latter, the head's seat, merely by the maxims of its will; instead, a rational being can keep the seat only when the rational being is a completely independent being without need and without limitation to its power that is adequate to its will.
So morality consists in the relation of all action to the lawgiving through which alone an empire of ends is possible. This lawgiving must, however,
be found in every rational being itself and must be able to arise from the rational being's will. The principle of the rational being's will is thus this: to do no action according to any maxim unless the maxim could be a universal law and thus to do an action only if the will could through its maxim consider itself at the same time as giving universal law. Now, if the maxims are not by their nature already necessarily in agreement with this objective principle of rational beings as giving universal law, then the necessity of action according to that principle is called practical necessitation, that is, duty. Duty does not apply to the head in the empire of ends, but duty surely does apply to each member and, to be sure, to each member in equal measure.
The practical necessity of acting according to this principle, that is, the duty, does not rest at all on feelings, impulses and inclinations. Instead, the practical necessity of acting according to this principle rests merely on the relation of rational beings to each other. In this relation, the will of a rational being must always at the same time be considered as giving law because otherwise the rational being could not think other rational beings as ends in themselves. So reason refers every maxim of the will as giving universal law to every other will and also to every action towards oneself. Reason definitely does not make these references to other wills and to self-directed actions for the sake of any other practical motive or for the sake of future advantage. Instead, reason makes these references from the idea of the
dignity of a rational being who obeys no law except a law that the rational being itself gives at the same time.
In the empire of ends everything has either a price or a dignity. What has a price is something in the place of which something else, as an equivalent, can also be placed. What, on the other hand, is elevated above all price, that has a dignity.
What refers to general human inclinations and needs has a market price. That which, even without presupposing a need, accords with a certain taste, that is, accords with a delight in the mere purposeless play of our powers of mind, has a fancy price. That, however, which constitutes the condition under which alone something can be an end in itself has not merely a relative worth, that is, a price, but instead has an inner worth, that is, dignity.
Now, morality is the condition under which alone a rational being can be an end in itself. Morality is the only condition because only through morality is it possible to be a lawgiving member in the empire of ends. So morality, and humanity insofar as it is capable of morality, is that which alone has dignity. Skill and diligence in work have a market price; wit,
lively imagination and humor have a fancy price. In contrast, sincerity in promising, kindness from basic principles (not from instinct), have an inner worth. Nature as well as art contain nothing which they, lacking sincerity and kindness, could put in place of sincerity and kindness; for the worth of sincerity and kindness consists not in the effects which arise from them, not from the advantage and profit which they provide. Instead, the worth of sincerity and kindness consists in the dispositions, that is, in the maxims of the will, that are ready to reveal themselves in this way in actions even if success does not favor them. These actions also require no recommendation from any subjective disposition or taste in order to be regarded with immediate favor and delight; they require no immediate tendency or feeling in order to be regarded in such a way. These actions of sincerity and kindness present the will that practices them as an object of an immediate respect. For this presentation of the will as a respected object, nothing but reason is required in order to impose the actions on the will. To coax the actions from the will, which in the case of duties would anyhow be a contradiction, is not required for the presentation of the will as a respected object. This valuation thus shows the worth of such a way of thinking as dignity and puts dignity infinitely far above all price. Dignity cannot be brought into calculation or comparison with price at all without, so to speak, assaulting dignity's holiness.
And now, then, what is it that justifies the morally good disposition or virtue in making such lofty claims?
What justifies it is nothing less than the share that the disposition provides to the rational being in universal lawgiving. By providing this share in universal lawgiving, the disposition makes the rational being fit to be a member in a possible empire of ends. The rational being was already destined by its own nature as an end in itself and therefore as a lawgiver in an empire of ends to be fit to be such a member and to be free with regard to all natural laws, obeying only those laws that the rational being itself gives and only those laws according to which the rational being's maxims can belong in a universal lawgiving (to which the rational being at the same time subjects itself). For nothing has a worth except that worth which the law determines for it. But lawgiving itself, which determines all worth, must for just that reason have a dignity, that is, have unconditional, incomparable worth. Only the word 'respect' provides the appropriate expression of the valuation that a rational being must assign to dignity. Autonomy is therefore the ground of the dignity of human nature and of all rational nature.
The three ways above, however, of representing the principle of morality are at bottom only so many formulas of the very same law, in which one by itself unites the other two in itself. Meanwhile, there is still a difference in them that is definitely subjectively practical rather than objectively practical, namely, so as to bring an idea of reason closer to intuition (according to a certain analogy)
and, by bringing the idea closer to intuition, bringing the idea closer to feeling. All maxims have, namely
1) a form, which consists in universality, and here the formula of the moral imperative is expressed in this way: that maxims must so be chosen as if they were to hold as universal laws of nature;
2) a matter, namely an end, and here the formula says: that the rational being, as an end according to its nature, therefore as an end in itself, must serve for every maxim as the limiting condition of all merely relative and optional ends;
3) a complete determination of all maxims through that formula, namely: that all maxims as individual lawgiving ought to harmonize with a possible empire of ends, as with an empire of nature*. The progression happens here as through the categories of unity of the form of the will (of the universality of the will), of plurality of the matter (of the objects, that is, of the ends), and of allness or totality of the system of ends. But you do better if in moral judgment you always
* Teleology considers nature as an empire of ends. Morals considers a possible empire of ends as an empire of nature. In the former, teleological, consideration, the empire of ends is a theoretical idea that explains what exists. In the latter, moral, consideration, the empire of ends is a practical idea for bringing into existence what does not exist but which can, in accordance of course with precisely this practical idea, become actual through our conduct.
proceed according to the strict method and make the universal formula of the categorical imperative the ground of judgment: act according to the maxim which can make itself at the same time into a universal law. If, however, you want at the same time to make the moral law more accessible, then it is very useful to lead one and the same action through the three named concepts of unity of form, plurality of matter, and allness of the system of ends and, by doing this, bring the three concepts, as far as possible, closer to intuition.
We can now end where we began, namely, with the concept of an unconditionally good will. That will is absolutely good which cannot be bad and therefore whose maxim, if the maxim is made into a universal law, can never conflict with itself. So this principle is also the will's highest law: act always according to that maxim whose universality as law you can at the same time will; this is the sole condition under which a will can never be in conflict with itself, and such an imperative is categorical. Because the validity of the will, as a universal law for possible actions, is analogous to the universal connection of the existence of things according to universal laws, which is what is formal in nature in general, the categorical imperative can also be expressed in this way: Act according to maxims which can have themselves, as universal laws of nature, at the same time as an object.
That, then, is the makeup of the formula of an absolutely good will.
Rational nature distinguishes itself from the others by setting an end for itself. This end would be the matter of any good will. Since, however, in the idea of an absolutely good will without a limiting condition (of the attainment of this or that end) complete abstraction must be made from any end to be produced (as this kind of end would make every will only relatively good), the end here must be thought not as one to be produced but rather as a self-sufficient end. So the end here must be thought only negatively, that is, as something never acted against, and therefore as something which must never be valued merely as a means but which must instead always at the same time in every act of willing be valued as an end. This end can be nothing other than the subject of all possible ends itself because this subject at the same time is the subject of a possible absolutely good will; for this will can, without contradiction, be subordinated to no other object. The principle: act in reference to each rational being (to yourself and others) in such a way that the rational being is considered in your maxim at the same time as an end in itself, is accordingly at bottom one and the same as the basic principle: act according to a maxim that contains in itself at the same time its own universal validity for every rational being. For, saying that I ought to limit my maxim, in the use
of means to every end, to the condition of the maxim's universal validity as a law for every subject, is the same as saying that the subject of ends must be made the ground of all maxims of actions. That is, it is the same as saying that the rational being itself must never be treated as a mere means but instead must be treated as the highest limiting condition in the use of all means, that is, must always be treated at the same time as an end.
From what has been said above, these points now follow incontestably. First, each rational being, as an end in itself, must, with reference to all laws to which the rational being may ever be subject, be able to look at itself at the same time as giving universal law. The rational being must be able to look at itself in this way because it is just this fitness of the rational being's maxims for universal lawgiving that mark out the rational being as an end in itself. Second, the dignity of the rational being (its prerogative) before all merely natural beings brings with it that the rational being's maxims must always be taken from the point of view of the rational being itself and also at the same time from the point of view of each other rational being as a lawgiving being (for which reason the other rational beings are also called persons). Now, in this way a world of rational beings (
mundus intelligibilis) as an empire of ends is possible, and indeed possible through the individual lawgiving of all persons as members. Accordingly, each rational being must act in such a way as if the rational being, through its maxims, always were a lawgiving member in the universal empire of ends. The formal principle of these maxims is:
act in such a way as if your maxim at the same time were to serve as the universal law (of all rational beings). So an empire of ends is only possible according to the analogy with an empire of nature. But, in thinking by means of this analogy, it must be kept in mind that the former, the empire of ends, operates only according to maxims, that is, to self-imposed rules, and that the latter, the empire of nature, operates only according to laws of externally necessitated efficient causes. Despite this difference in operation, we still call the whole of nature an empire of nature; we still give the whole of nature this name, even though the whole of nature is seen as a machine, insofar as the whole of nature has reference to rational beings as its ends. Now, such an empire of ends would actually come into existence through maxims whose rule the categorical imperative prescribes to all rational beings, if the maxims were universally followed. The following are things that the rational being cannot count on happening: first, that, even if the rational being itself were to follow this maxim to the letter, every other rational being would therefore faithfully follow the same maxim; second, that the empire of nature and its purposive order will harmonize with the rational being as with a fitting member of an empire of ends possible through the rational being itself — that is, that the empire of nature will favor the rational being's expectation of happiness. But, although the rational being cannot count on these things, that law still remains: act according to maxims of a member giving universal law to a merely possible empire of ends. That law remains in full force because it commands categorically. And it is just in this that the paradox lies: first, that merely the dignity of the human being, as rational
nature without any other end or advantage to be attained by this dignity, therefore with respect for a mere idea, is nevertheless to serve as the constant prescription of the will; and second, that it is just in this independence of the maxim from all such incentives that the sublimity of the maxim consists and in which the worthiness of any rational subject to be a lawgiving member in the empire of ends consists. For without this independence the rational subject would have to be thought of as subject only to the natural laws of its needs. Even if the natural empire as well as the empire of ends were thought as united under one head and through this unification the latter, the empire of ends, no longer remained a mere idea but instead received true reality, the idea would definitely gain a strong incentive, but through this unification the idea would never receive an increase in its inner worth. For, if this unification under one head did occur, even this sole unlimited lawgiver would still always have to be thought of as judging the worth of the rational being only according to the rational beings' disinterested conduct that the rational beings prescribe for themselves merely from that idea of an empire of ends. The essence of things does not change through their outer relations, and, without thinking of these outer relations, what alone constitutes the absolute worth of the human being has to be that according to which the human being must also be judged, no matter who the judge may be — even if the judge is the highest being. So morality is the relation of actions to the autonomy of the will, that is, to the possible universal
lawgiving through the will's maxims. An action that is compatible with the autonomy of the will is permitted. An action that is not compatible with the autonomy of the will is impermissible. The will whose maxims necessarily harmonize with the laws of autonomy is a holy, absolutely good will. The dependence of a will that is not absolutely good on the principle of autonomy (moral necessitation) is obligation. So obligation cannot apply to a holy being. The objective necessity of an action from obligation is called duty.
You can now easily explain from what has just been said how it comes about: that, although under the concept of duty we think a subjection under the law, in thinking this we still at the same time imagine a certain sublimity and dignity in that person who fulfills all of her duties. For there is definitely no sublimity in the person insofar as the person is subject to the moral law. More plausibly, however, there is sublimity in the person insofar as the person, with regard to the very same moral law, at the same time is lawgiving and only because of that lawgiving is subject to that law. We have also shown above how neither fear nor inclination but, instead, how only respect for the law is that incentive which can give an action a moral worth. Our own will, so far as it would act only under the condition of a universal lawgiving possible through the will's maxims,
is the proper object of respect. This will is possible for us in the idea of an empire of ends; and the dignity of the human being consists just in this capability to give universal law, although on the condition of being itself at the same time subject to just this lawgiving.
Autonomy of the will is the characteristic of the will by which the will is a law to itself (independently of any characteristic of the objects of willing). So the principle of autonomy is: not to choose otherwise than in such a way that the maxims of your choice are included as universal law at the same time in the same act of will. That this practical rule is an imperative, that is, that the will of every rational being is necessarily bound to the rule as a condition, cannot be proven by mere analysis of the concepts present in the principle because the principle is a synthetic proposition. To prove that this practical rule is an imperative, you would have to go out beyond the knowledge of objects and to a critique of the subject, that is, a critique of pure practical reason; and you would have to undertake such a critique because this synthetic proposition, which commands with absolute necessity, must be able to be known completely
a priori. This task of a critique, however, does not belong in the present
section. But that the aforesaid principle of autonomy is the sole principle of morals can quite well be shown by mere analysis of the concepts of morality. For by carrying out such an analysis, we find that the principle of morality must be a categorical imperative and that this imperative commands nothing more nor less than just this autonomy.
If the will seeks what is to guide it in anything else than in the suitability of the will's maxims to the will's own universal lawing, then heteronomy always results. If, that is, the will, in going out beyond itself, seeks the law that is to guide the will in the character of any of the will's objects, then heteronomy always results. In cases of heteronomy, the will does not give itself the law; but, instead, the object through its relation to the will gives the law to the will. This relation, whether it rests now on inclination or on representations of reason, only allows hypothetical imperatives to be possible: I ought to do something just because I want something else. In contrast, the moral imperative, and therefore the categorical imperative, says: I ought to act thus and so even if I wanted nothing else. For example, the former, hypothetical imperative, says: I ought not lie, if I want to retain my honorable reputation; but the latter,
moral or categorical imperative, says: I ought not lie even if it brought upon me not the least shame. So the latter, categorical imperative, must abstract from all objects to such an extent that the objects would have no influence at all on the will, so that practical reason (will) would not merely administer alien interest but instead would merely prove its own commanding authority as highest lawgiving. So I ought, for example, to seek to promote the happiness of others, not as if the existence of that happiness were any of my concern (whether it be through immediate inclination or some satisfaction provided indirectly through reason); instead, I ought to promote the existence of that happiness just because the maxim that excludes that happiness cannot be included in one and the same willing as a universal law.
Human reason has here, as everywhere in human reason's pure use so long as human reason lacks a critique, previously tried all possible incorrect ways before human reason succeeds in hitting upon the one correct way.
All principles that you might take from the point of view of human reason are either empirical or
rational. The first, from the principle of happiness, are built on physical or moral feeling. The second, from the principle of perfection, are built either on the rational concept of perfection as a possible effect or on the concept of a self-sufficient perfection (the will of God) as a controlling cause of our will.
Empirical principles are not at all fit to be the ground of moral laws. For the universality with which the laws are to hold for all rational beings without difference — the unconditional practical necessity that is imposed on rational beings by this universality of moral laws — falls away if the ground of the laws is taken from the particular arrangement of human nature or from the contingent circumstances in which that nature is placed. But the principle of personal happiness is most objectionable, not merely because it is false, and because experience contradicts the pretense that well-being always adjusts itself according to good conduct, and also not merely because the principle contributes nothing at all to the grounding of morality since it is something quite different to make a happy human being than to make a good human being and something quite different to make a human being prudent and alert to what might be to her advantage than to make her virtuous. To be sure, those flaws make the principle of personal happiness objectionable, but it is most objectionable because it puts incentives underneath morality, and these incentives, rather than supporting morality, instead undermine it and destroy its entire sublimity.
The incentives undermine morality because they put motives to virtue in the same class with motives to vice and because the incentives only teach us to calculate better what is to our personal advantage or disadvantage, thus thoroughly obliterating the specific difference between virtue and vice. On the other hand, moral feeling, this supposed special sense*, (however shallow the appeal to this sense is, in that those who cannot think even about what depends merely on universal law believe they can help themselves out through feeling, feelings, which according to their ranking by nature are infinitely different from each other, provide just as little a uniform standard of good and bad; you also cannot judge at all validly through your feeling for others), nevertheless remains closer to morality and its dignity for the following reasons. First, moral feeling remains closer because moral feeling does virtue the honor of ascribing immediately to virtue the delight and high esteem that we have for virtue. Second, moral feeling remains closer to morality and its dignity because moral feeling does not say to virtue, as if to her face, that it is not her beauty but instead only the advantage to us that ties us to her.
Among the rational grounds of morality or grounds based on reason, there is still the ontological concept of
* I classify the principle of moral feeling with the principle of happiness because any empirical interest promises a contribution to well-being through the agreeableness that something offers us, whether this agreeableness is immediate and without a view to advantages or whether the agreeableness occurs with regard to those advantages. Likewise, you must classify, with Hutcheson, the principle of compassion for the happiness of others with the same moral sense that he assumed.
perfection. (This concept is exceedingly unfounded, indeterminate, and therefore useless for discovering in the immense field of possible reality the greatest sum appropriate for us. The concept also has an unavoidable tendency, in specifically distinguishing reality, which is here under discussion, from every other, to turn around in a circle and cannot avoid secretly presuming the morality that the concept is to explain.) Despite the drawbacks of this concept of perfection, it is still better than the theological concept, still better than deriving morality from a divine all-perfect will. The concept of perfection is better not merely because we cannot of course see the divine will's perfection but instead can only derive that perfection from our concepts, chief among our concepts being that of morality. Rather, the concept of perfection is also better because, if we do not do this derivation (which, if we did do it, would amount to a crude circle in the explanation), the concept left to us of the divine will would have to be made the foundation for a system of morals; but that concept left to us would be made up of the attributes of eager desire for glory and dominion, combined with terrible thoughts of power and of thirst for vengeance, and a concept made up of such attributes would pit the concept directly against morality.
But if I had to choose between the concept of moral sense and that of perfection in general (both of which at least do no harm to morality, although they are not at all suited to support morality as its foundations), then I would decide for the latter.
I would choose the concept of perfection because the concept of perfection, since it at least transfers the decision of the question from sensibility to the court of pure reason, although here the concept also decides nothing, nevertheless preserves unfalsified the vague idea (of a will good in itself) for more precise specification.
Regarding the remaining rational grounds for morality, I believe I can be excused from a lengthy refutation of all these doctrines. It is so easy to refute these doctrines that even those whose job requires that they declare themselves for one of these theories (because listeners will not put up with a postponement of judgment) presumably see through the theories, so that refuting the theories here would only be superfluous labor. What interests us more, however, is to know the following: that these principles everywhere set up nothing but heteronomy of the will as the first ground of morality, and that for just this reason these principles must necessarily fail in their purpose.
In all cases in which an object of the will must be made the basis of action in order to prescribe to the will the rule that is to guide the will, the rule is nothing but heteronomy; the imperative is conditional, namely: if or because you want this object, you ought to act in such and such a way. Therefore, the imperative can never command morally, that is, categorically. Whether the object controls the will by means of inclination, as with the principle of your own happiness,
or controls the will by means of reason directed to objects of our possible willing in general, in the principle of perfection, the will never controls itself immediately by the thought of an action. Instead, the will controls itself only by the incentive which the anticipated effect of the action has on the will; I ought do something just because I want something else, and here yet another law must be put in my subject as a ground according to which I necessarily will this other thing that I want, and this other law again requires an imperative which would limit this maxim. The reason for this lack of direct self-control by the will is the following: the thought of an object that we can bring about through our own powers is to exert an impulse on the subject's will; this exertion occurs according to the natural constitution of the subject; so the impulse belongs to the nature of the subject; whether the impulse belongs to the nature of the subject's sensibility (of inclination and taste) or to the nature of the subject's understanding and reason, these features of the subject, according to the special arrangement of their nature, allow the subject to take delight in an object. In this way, it is, properly speaking, nature that would give the law. This law, as one given by nature, must be recognized and proved through experience, and so is contingent in itself. Because of this contingency, this law given by nature becomes unfit to be an absolutely necessary practical rule, which is the kind of practical rule that the moral rule must be. Not only is this law given by nature contingent and so unfit to be a moral law, but this law given by nature is always only heteronomy of the will; the will does not give the law to itself, but rather an alien impulse gives the law to the will by means of a
nature of the subject that is disposed to receive the law.
So the absolutely good will, whose principle must be a categorical imperative and whose choices are not controlled by any objects, will contain merely the form of willing in general. Indeed, the absolutely good will contains this form of willing in general as autonomy. That is to say, the suitability of the maxim of any good will to make itself into a universal law is itself the sole law that the will of any rational being imposes on itself, and the rational being imposes this law on itself without making any incentive or interest of the maxim the basis of the law.
How such a synthetic practical proposition
a priori is possible and why the proposition is necessary, is a problem whose solution no longer lies within the boundaries of the metaphysics of morals. We have also not asserted the proposition's truth, much less pretending to have within our power a proof of the truth of the proposition. We only showed by analyzing the generally accepted concept of morality that an autonomy of the will, in an unavoidable way, attaches to the will or, rather, is the ground of the will. So, whoever takes morality to be something and not to be a wildly fanciful idea without truth must at the same time admit morality's principle of autonomy that was cited above. So this
section was merely analytic, just like the first section. Now, that morality is not a phantom, which follows if the categorical imperative and with it the autonomy of the will is true and is absolutely necessary as a principle
a priori, requires a possible synthetic use of pure practical reason. But we may not venture on this use of pure practical reason without first giving a critique of this rational faculty itself. Sufficient for our purpose, we have to present the main features of such a critique in the last section.
The will is a kind of causality that living beings have insofar as they are rational. Freedom would be that property of this causality by which the causality can be effective independently of alien causes controlling the will as a causality. Similarly, natural necessity is the property of causality of all non-rational beings to be directed to activity by the influence of alien causes.
The above explanation of freedom is negative and is therefore unfruitful for seeing into the essence of freedom. But out of this negative explanation there flows a positive concept of freedom which is so much richer and more fruitful. The concept of a causality carries with it the concept of laws according to which, by something that we call a cause, something
else, namely the effect, must be assumed as a fact. Because the concepts of causality and law are related in this way, although freedom is not a property of the will according to natural laws, freedom is still not entirely lawless. Instead of operating according to natural laws, freedom must rather be a causality according to unchanging laws, but unchanging laws of a special kind; for a free will would be an impossibility if it did not operate according to some kind of law. Natural necessity was a heteronomy of efficient causes; for each effect was possible only according to the law that something else determined the efficient cause to become causally active. What, then, can freedom of the will possibly be other than autonomy, that is, the property of the will to be a law to itself? But the proposition that the will is in all actions itself a law signifies only the principle to act according to no other maxim except one that can also have itself as a universal law as an object. This principle, however, is just the formula of the categorical imperative and the principle of morality. So a free will and a will under moral laws are one and the same.
If, therefore, freedom of the will is presupposed, then morality together with morality's principle follow from that presupposition merely by analysis of the presupposition's concept. Nevertheless, the latter, morality's principle, is still always a synthetic proposition: an absolutely good will is a will whose maxim always can contain itself, considered as a universal law, in itself,
for through analysis of the concept of an absolutely good will that property of the maxim (i.e., the maxim's property to be able to contain itself as a universal law) cannot be found. Such synthetic propositions, however, are only possible by this: that both cognitions are bound to each other through the connection with a third in which both cognitions are to be found. The positive concept of freedom provides this third cognition. Unlike in cases dealing with physical causes, in this case this third cognition cannot be the nature of the world of sense (in which concept the concept of something as a cause in relation to something else as an effect come together). We cannot yet show here right now what this third cognition is to which freedom points us and of which we have an
a priori idea. We also cannot yet make the deduction of the concept of freedom from pure practical reason comprehensible and, along with this deduction, cannot yet make the possibility of a categorical imperative comprehensible. Still further preparation is required in order to identify the third cognition and in order to make the deduction and possibility comprehensible.
It is not enough that we ascribe, for whatever reason, freedom to our will. We also need to have sufficient reason to attribute the very same freedom of the will to all rational beings.
For, since morality serves as a law for us only because we are rational beings, morality must also hold for all rational beings; and, since morality must be derived merely from the property of freedom, freedom must also be proved as a property of the will of all rational beings. In addition, it is not enough to demonstrate freedom from certain alleged experiences of human nature (although this is also absolutely impossible and freedom can only be demonstrated
a priori); instead, you must prove freedom as belonging to the activity of rational beings in general endowed with a will. I say now: any being that cannot act other than under the idea of freedom, is, just for that reason, in a practical respect, actually free. That is to say, all laws that are inseparably bound up with freedom are laws that hold for such a being just as if the being's will also in itself and in theoretical philosophy would be validly declared to be free.* Now I maintain: that we
* I suggest that to assume this way of only taking the mere idea of freedom to be the basis for the actions of rational beings is sufficient for our purpose. I suggest this so that I may not also be bound to prove freedom in its theoretical aspect. For, even if this theoretical aspect of proving freedom is left undecided, the same laws that hold for a being that cannot act except under the idea of the being's own freedom are laws that still would hold for a being that was actually free. So we can here free ourselves from the burden that presses on the theory.
must also necessarily lend to each rational being that has a will the idea of freedom under which alone the being can act. For in such a being we conceive of a reason that is practical, that is, has a causality with respect to its objects. Now, you cannot possibly conceive of a reason that, with its own consciousness with regard to its judgments, receives direction from elsewhere, for then the subject would ascribe the control of the power of judgment not to the subject's reason but instead to an impulse in the subject. Reason must view itself as the authoress of its principles, independently of alien influences. Consequently, reason, as practical reason or as the will of a rational being, must be viewed by itself as free. That is to say, the will of a rational being can only be a will of its own under the idea of freedom and so such a will must, for practical purposes, be attributed to all rational beings.
We have at last traced the specific concept of morality back to the idea of freedom. We were not able, however, to prove this idea of freedom to be something actual, not even in ourselves and in human nature. We only saw that we must presuppose the idea if we
want to conceive of a being as rational and with consciousness of its causality with regard to actions, that is, as endowed with a will. And so we find that we must, for the very same reason, attribute this property, namely, the property of directing itself to action under the idea of its freedom, to each being endowed with reason and a will.
But from the presupposition of these ideas there also flowed the consciousness of a law of acting: that the subjective basic principles of actions, that is, maxims, must always be taken in such a way that they also hold objectively, that is, hold universally as basic principles, and therefore can serve for our own universal lawgiving. But why then ought I subject myself to this principle and indeed, as a rational being in general, subject therefore also all other rational beings endowed with a will to this principle? I am willing to admit that no interest impels me to this subjection; for that would give rise to no categorical imperative. But I must still necessarily take an interest in this subjection and look into how it comes about; for this ought is actually a want that holds for each rational being under the condition that in the case of each being reason would be practical without hindrances. For beings such as ourselves, who are still affected by sensibility, as incentives of a different kind, and for whom what reason for itself alone would do does not always happen,
that necessity of action is only called an ought and the subjective necessity is distinguished from the objective necessity.
So it appears as if we actually only presupposed the moral law, namely, the principle of autonomy of the will itself, in the idea of freedom and could not prove for itself the reality and objective necessity of the moral law. If that is indeed all that we have done, then we would still have gained something quite considerable in the process; we would at least have specified the genuine moral principle moral precisely than otherwise would have been done. But with regard to the validity of the moral principle and the practical necessity of subjecting ourselves to that principle, we would have gotten no farther along; for we could give no satisfactory answer to someone who asked the following questions. Why, then, must the universal validity of our maxim, as a law, be the limiting condition of our actions? On what do we base the worth that we attribute to this way of acting, a worth which is to be so great that there can be no higher interest anywhere? And how does it come to pass that the human being believes that she feels her personal worth to reside only in this subjection to moral law, a worth against which the worth of a pleasant or unpleasant condition is held to be nothing?
We surely do find that we can take an interest in a personal characteristic which
carries with itself no interest in any condition, if only the former characteristic makes us capable of sharing in the latter condition in case reason were to bring about the distribution of the condition. That is to say, the mere worthiness to be happy, even without the motive of sharing in this happiness, can itself be of interest to us. But this judgment of worthiness is in fact only the effect of the already presupposed importance of moral laws (when we separate ourselves from all empirical interest through the idea of freedom). But in this way we cannot yet see into the following: that we ought to separate ourselves from this empirical interest, that is, ought to consider ourselves to be free in acting and so ought nevertheless to hold ourselves to be subject to certain laws in order to find a worth merely in our person, a worth that can compensate us for the loss of everything that gives worth to our condition; how this separation is possible; and so from what source or on what basis the moral law binds us.
You must freely admit that there appears to be a circle here from which it seems there is no recovery. We take ourselves to be free in the order of efficient causes in order to think ourselves in the order of ends under moral laws, and we afterwards think ourselves as subject to these laws because we have attributed freedom of the will to ourselves, for freedom and individual lawgiving of the will are both
autonomy, and so they are reciprocal concepts. But, precisely because they are reciprocal concepts, one of them cannot be used to explain the other and to specify the ground of the other. At most, one concept can only be used for logical purposes to reduce different appearing representations of the very same object to a single concept (as different fractions of equal value are reduced to the simplest expression).
But one way out of the circle still remains open to us, namely, to try to find: whether we, when we think ourselves through freedom as
a priori efficient causes, do not take a different standpoint than we do when we represent ourselves according to our actions as effects that we see before our eyes.
No subtle reflection at all is required to post the following remark; indeed, you can assume that even the most common understanding may make the remark, although such an understanding makes the remark in its own way through an obscure distinction of the power of judgment which it calls feeling. The remark is this: all ideas that we receive involuntarily (like those ideas we receive through the sense organs) give us no knowledge of objects except as the objects affect us; what the objects may be in themselves remains unknown to us. So, as far as this involuntary kind of ideas is concerned, we can, even with the most strenuous
attentiveness and clarity that the understanding may ever add, still only arrive at knowledge of appearances, never at knowledge of the things in themselves. As soon as this distinction (perhaps merely through the noticed difference between the ideas that are given to us from somewhere else and with which we are passive and the ideas that we produce only from ourselves and with which we prove our activity) is made once, then it follows of itself that you must admit and assume that behind the appearances there is after all still something else that is not appearance, namely, the things in themselves. Although we admit and assume the existence of these things in themselves, we resign ourselves to the fact that, since they can never become known to us in themselves but always only by how they affect us, we cannot get closer to them and can never know what they are in themselves. This must provide a distinction, although crude, between a world of sense and the world of understanding. The first, the world of sense, according to difference of sensibility in various observers, can also be very diverse. Meanwhile, the second, the world of understanding, which is the basis for the world of sense, always remains the same. Even the human being herself cannot presume to know, by the knowledge she has of herself through inner sensation, what she is in herself. For since she after all does not, so to speak, create herself, and she gets her concept of herself not
a priori but instead empirically, it is natural that she also gets information about herself through the inner sense and
consequently only through the appearance of her nature and through the way in which her consciousness is affected. Meanwhile, she must still necessarily assume that beyond this constitution, put together from nothing but appearances, of her own subject there is something else that is the basis of her constitution. This basis of her natural makeup or constitution is her I or ego, in whatever way it may be constituted in itself. So, with regard to the mere perception and receptivity of sensations she must count herself as belonging to the world of sense; but, with regard to what may be pure activity in her (to what arrives in consciousness not by affecting the senses but instead to what arrives in consciousness immediately), she must count herself as belonging to the world of the intellect. She knows nothing further, however, about this latter, intellectual world.
A reflective human being must draw a conclusion of this kind from all things that may appear to her. Presumably, this conclusion is also to be found in the most common understanding which, as is well-known, is always very inclined to expect something invisible and active in itself behind the objects of the senses. But the common understanding again corrupts this invisible something by wanting to make the invisible something into a sensuous thing again, that is, by wanting to make the invisible something into an object of intuition. And so, by trying to make something invisible into something sensuous, the common understanding does not become even a little bit wiser.
Now, the human being actually finds in herself a capacity by which she distinguishes herself from all other things, and even from
herself so far as she is affected by objects; and this capacity is reason. This reason, as pure self-activity, is even in this self-activity still raised above the understanding in this way: that reason in self-activity is higher because, although the understanding is also self-activity and does not, as sense does, merely contain ideas that only arise when you are affected by things (and are therefore passive), the understanding nevertheless can produce from its activity no concepts other than those that serve merely to bring sensuous representations under rules and that, by bringing the representations under these rules, unite the representations in a single consciousness; without this use of sensibility, the understanding would think nothing at all. On the other hand, reason, under the name of ideas, shows such a pure spontaneity that the human being, by this spontaneity, goes out far beyond anything that sensibility only can provide to the human being and showcases reason's foremost occupations by distinguishing the world of sense from the world of understanding; in making this distinction, however, reason marks out the boundaries for the understanding itself.
Because of this distinction that reason makes, a rational being, as an intelligence (so not from the perspective of the rational being's lower powers), must look at itself as belonging not to the world of sense but instead as belonging to the world of the understanding. So the rational being has two standpoints from which it can consider itself and can recognize laws for the use of its powers and, consequently, can recognize laws governing all of its actions. First, as far as the rational being belongs to the world of sense,
the rational being can consider itself as under laws of nature (heteronomy). Secondly, as belonging to the intelligible world, the rational being can consider itself as under laws that are independent of nature and are not empirical; instead, these independent and non-empirical laws are grounded only in reason.
As a rational being, and therefore as a being belonging to the intelligible world, the human being can never think of the causality of its own will except as under the idea of freedom; for independence from the determinate causes of the world of sense (which is the kind of independence that reason must always attribute to itself) is freedom. Now, with the idea of freedom the concept of autonomy is inseparably connected, but the concept of autonomy is inseparably connected with the universal principle of morality; and the principle of morality underlies in the idea as a ground all actions of rational beings just as natural law, as an idea and ground, underlies all appearances.
The suspicion that we stirred up earlier has now been removed. The suspicion was that a hidden circle might have been contained in our inference from freedom to autonomy and then from autonomy to the moral law. In particular, the circle might have been that we perhaps made the idea of freedom a ground only for the sake of the moral law in order afterwards in turn to conclude the moral law from freedom. So, because of this hidden circle, we could provide no ground at all for the moral law; instead, we could only provide the moral law as a begging of a principle that friendly souls will probably gladly grant us, but which we
never could set up as a provable proposition. For we now see that, when we think ourselves as free, we transport ourselves as members into the world of understanding and recognize the autonomy of the will together with its consequence, morality. But when we think ourselves as obligated, then we consider ourselves as belonging to the world of sense and yet at the same time as belonging to the world of understanding.
The rational being, as an intelligence, counts itself as belonging to the world of understanding, and the rational being, merely as an efficient cause belonging to this world of understanding, calls its causality a will. But from a different point of view, the rational being is also conscious of itself as a piece of the world of sense in which the rational being's actions, as mere appearances of that causality, are found. But we cannot comprehend the possibility of these actions as effects of that causality with which we have no acquaintance; instead, in place of that comprehension, we must understand those actions as determined by other appearances, namely, by eager desires and inclinations, and as belonging to the world of sense. So, as only a member of the world of understanding, all my actions would be in perfect conformity with the principle of autonomy of the pure will; as only a piece of the world of sense, my actions would have to be taken as in complete conformity with the natural law of eager desires and inclinations, and therefore with the heteronomy of
nature. (The first actions, those of the world of understanding, would rest on the highest principle of morality; the second actions, those in the world of sense, would rest on the principle of happiness.) But the world of understanding contains the ground of the world of sense and therefore also the ground of the laws of the world of sense; thus, the world of understanding is immediately lawgiving with respect to my will (which belongs entirely to the world of understanding); so the world of understanding must also be thought as lawgiving; for these reasons, I will have to recognize that, although from another point of view I am a being belonging to the world of sense, I am nevertheless subject as an intelligence to the law of the first world, the world of understanding, that is, of reason. Reason contains the law of the world of understanding in reason's idea of freedom and so I will also have to recognize that I am subject as an intelligence to the autonomy of the will. Consequently, I will have to look at the laws of the world of understanding as imperatives for me and have to look at the actions that are in conformity with this principle as duties.
And it is in this way that categorical imperatives are possible. They are possible because the idea of freedom turns me into a member of an intelligible world by which, if I were only such a member, all my actions would always be in conformity with the autonomy of the will. But, since I at the same time intuit myself as a member of the world of sense, my actions ought always to conform with the autonomy of the will. This categorical ought represents a synthetic proposition
a priori because to my will that is affected by sensuous eager desires is added the idea of just the same will, but pure, in itself practical, and belonging to the world of understanding.
This pure will contains, according to reason, the highest condition of the first, the sensuously affected, will. This addition is approximately like the way in which concepts of the understanding, which in themselves signify nothing but lawful form in general, are added to the intuitions of the world of sense. By their addition to intuitions, these concepts of the understanding make synthetic propositions
a priori possible, and it is on such propositions that all knowledge of a nature rests.
The practical use of common human reason confirms the correctness of this deduction. There is no one, even the most vile miscreant as long as she is otherwise accustomed to using reason, who, when you present her with examples of honesty in intentions, of steadfastness in obeying good maxims, of compassion and of common kindness (and joined moreover with great sacrifices of advantages and convenience), does not wish that she might also be so disposed. But, only because of her inclinations and impulses, she cannot bring these examples fully about in herself; although she does not do well in realizing the examples in herself, she still wishes to be free of such inclinations that are burdensome to her. She proves by this wish, therefore, that she, with a will that is free from impulses of sensibility, transfers herself in thought into an order of things entirely different from that of her eager desires in the field of sensibility. This is proved because from that wish she expects no satisfaction of her eager desires and so expects for all of her actual or otherwise
imaginable inclinations no satisfying condition (for by this even the idea which coaxes the wish from her would lose its preeminence); instead, she can expect only a greater inner worth of her person. She believes herself to be this better person when she transfers herself into the standpoint of a member of the world of understanding. It is to this standpoint that she is involuntarily necessitated by the idea of freedom, that is, independence from the determining causes of the world of sense. And it is in this standpoint that she, according to her own admission, is conscious of a good will that constitutes the law for her bad will as a member of the world of sense. She is acquainted with the authority of this law whenever she transgresses the law. So the moral ought is one's necessary willing as a member of an intelligible world, and the moral ought is only thought by a member of an intelligible world as an ought insofar as she at the same time considers herself to be a member of the world of sense.
All human beings think of themselves as having a free will. It is from this thought that all judgments about actions, as actions that ought to have been done although they were not done, come. But this freedom is not a concept of experience, and also cannot be such a concept, because the concept of freedom always remains even though experience shows the opposite
of those demands that are represented as necessary under the presupposition of freedom. From a different point of view, it is just as necessary that everything that happens be determined without exception according to natural laws, and this natural necessity is also not a concept of experience precisely because the concept of natural necessity carries with it the concept of necessity and therefore of a cognition
a priori. But this concept of a nature is confirmed by experience and must itself be unavoidably presupposed if experience, that is, coherent cognition of objects of sense in accordance with universal laws, is to be possible. Freedom is therefore only an idea of reason, and the idea's objective reality is in itself doubtful. Nature, however, is a concept of the understanding, and this concept proves, and must necessarily prove, its reality in examples from experience.
A dialectic of reason now arises from this since, as regards the will, the freedom attributed to the will appears to stand in contradiction to natural necessity and since, with this parting of the ways, reason finds, for purposes of intellectual curiosity, the way of natural necessity much more traveled and usable than the way of freedom. Although this dialectic arises, the footpath of freedom is still, for practical purposes, the one path on which it is possible to make use of one's reason in our conduct. So it is just as impossible for the most subtle
philosophy as for the most common human reason to argue away freedom. So this philosophy must indeed presuppose the following: that no true contradiction will be found between freedom and natural necessity of the very same human actions, for philosophy can give up the concept of nature no more than it can give up the concept of freedom.
While we wait for no true contradiction to be found, this apparent contradiction must at least be dissolved in a convincing way, even if we could never understand how freedom is possible. For, if even the thought of freedom contradicts itself or contradicts the thought of nature, which is just as necessary, then freedom, as opposed to natural necessity, would have to be given up completely.
But it is impossible to escape this contradiction, if the subject who imagines itself free thought of itself in the same sense or in the same relation when it calls itself free as it did when it assumes itself subject to natural laws with respect to the same action. So it is an inescapable task of speculative philosophy to show at least the following things. First, speculative philosophy must show that philosophy's deception about the contradiction rests in our thinking the human being in a different sense and relation when we call the human being free than we do when we hold the human being to be a piece of nature
subject to nature's laws. Second, speculative philosophy must show that these two senses and relations can exist together not only quite well but must also be thought as necessarily united in the same subject; for, if not necessarily united in the same subject, no justification could be given why we should burden reason with an idea that, although the idea can be united without contradiction with a different sufficiently established idea, nevertheless ensnares us in a task that puts reason in its theoretical use in a bind. This duty, however, is incumbent only on speculative philosophy, so that speculative philosophy might prepare a clear path for practical philosophy. Thus it is not at the discretion of the philosopher to decide whether she wants to remove the apparent contradiction or wants to leave the apparent contradiction untouched; for, if left untouched, the theory about this is
bonum vacans and the fatalist can justifiably take possession of the property, driving all morals out of morals' alleged property which morals has no title to occupy.
Nevertheless, you can not yet say that the boundary of practical philosophy begins here. For that settlement of the controversy does not belong to practical philosophy; instead, practical philosophy demands only of speculative reason that speculative reason bring to an end the discord into which speculative philosophy involves itself in theoretical questions. If speculative reason can bring this discord to an end, then practical reason might have rest and security against external attacks that could make contentious the ground on which practical reason wants to establish itself.
But the rightful claim, even of common human reason, to freedom of the will is grounded on the consciousness and the granted presupposition of the independence of reason from merely subjectively determining causes. These causes together constitute what belongs merely to sensation and so what belongs under the general name of sensibility. The human being considers herself in such a way as an intelligence; by doing so, she puts herself in a different order of things and in a relation to determining grounds of a quite different kind when she thinks of herself as an intelligence endowed with a will and consequently as endowed with causality than she does when she perceives herself as a phenomenon in the world of sense (which she actually is, too) and subjects her causality, according to external determination, to natural laws. Now, she soon becomes aware that both ways of thinking of herself can, and indeed even must, take place at the same time. For the following does not contain the least contradiction: that a thing as an appearance (that belonging to the world of sense) is subject to certain laws while the very same as a thing or being in itself is independent of those laws. But that she must imagine and think herself in this twofold way rests on different kinds of awareness. First, as a thing as an appearance, her thinking rests on the consciousness of herself as an object affected by the senses. Second, as a thing in itself, her thinking rests on the consciousness of herself as an intelligence, that is, as independent of sensuous impressions in the use of reason (and therefore as belonging to the world of understanding).
So it happens that the human being claims for herself a will that does not let what belongs merely to her eager desires and inclinations enter into her accountability. On the contrary, she thinks of actions as possible — indeed even as necessary — through herself, actions that can be done only by disregarding all eager desires and sensuous impulses. The causality of these actions lies in her as an intelligence and in the laws of effects and actions according to principles of an intelligible world. She certainly knows nothing of this intelligible world except that in this intelligible world only reason — and, for sure, pure reason independent of sensibility — gives the law. Also, since in this intelligible world she is only as an intelligence her proper self (as a human being, in contrast, only an appearance of herself), those laws apply to her immediately and categorically. Because those laws apply to her directly and without exception, her inclinations and impulses (and so the whole nature of the world of sense), no matter what they prod her to do, cannot infringe the laws of willing as an intelligence. This insulation of those laws from infringement is so thorough that she does not answer for the inclinations and impulses and does not ascribe them to her proper self, that is, to her will. She does, however, ascribe to her will the indulgence that she would show the inclinations and impulses if she, to the disadvantage of the rational laws of the will, permitted the inclinations and impulses influence on her maxims.
By thinking itself into a world of understanding, practical reason does not overstep its bounds at all. But practical reason certainly would overstep its bounds if it wanted to look or feel itself into such a world. The former, merely thinking itself into a world of understanding, is only a negative
thought with regard to the world of sense. This negative thought is that the world of sense gives no laws to reason for controlling the will. The thought is positive only in this one point: that that freedom, as a negative determinant or controller, is combined at the same time with a (positive) capacity and even with a causality of reason, which we call a will; this capacity or causality of reason is a capacity to act in such a way that the principle of actions is in accordance with the essential character of a rational cause as a law, that is, with the condition of the universal validity of the maxim. But, if practical reason were still to fetch an object of the will, that is, a motive, from the world of understanding, then practical reason would overstep its bounds and presume to be acquainted with something which it knows nothing about. So the concept of a world of understanding is only a standpoint which reason sees itself necessitated to take outside of the appearances in order to think of itself as practical. Thinking of itself as practical would not be possible if the influences of sensibility had control of the human being. But thinking of itself as practical is still necessary if the consciousness of itself as an intelligence and therefore as a cause that is rational and active through reason, that is, is free acting, is not to be denied to the human being. This thought, of course, brings about the idea of a different order and lawgiving than the idea of a mechanism of nature which concerns the world of sense. This thought also makes the concept of an intelligible world (that is, the whole of rational beings as things in themselves)
necessary, but without the least presumption to think further here than merely in accordance with the formal condition of the intelligible world. That is to say, the concept of an intelligible world is made necessary just by thinking in conformance with the universality of the maxims of the will as laws and therefore with the autonomy of the will, that autonomy alone being able to coexist with the freedom of the will. While, on the other hand, all laws that are specified by an object give heteronomy which can only be found in natural laws and which also can only concern the world of sense.
But then reason would overstep its entire boundary if it attempted to explain how pure reason can be practical, which would be exactly the same as the problem of explaining how freedom is possible.
For we can explain nothing except what we can trace back to laws whose object can be given in some possible experience. But freedom is a mere idea whose objective reality can in no way be set forth according to natural laws and cannot, therefore, be set forth in any possible experience. So the idea's objective reality can never be comprehended or even glimpsed precisely because an example along the lines of an analogy may never be put underneath freedom itself. The idea of freedom holds only as a necessary presupposition of reason in a being that believes itself to be conscious of a will, that is, of a capacity still different from the mere faculty of desire. (This capacity is, in particular, the capacity to resolve to act as an intelligence and therefore according to laws of reason, independently of
natural instincts.) But where the determination of natural laws stops, all explanation stops, too, and nothing remains except defense, that is, repelling the objections of those who pretend to have seen deeper into the essence of things and, because of that alleged insight, audaciously declare freedom to be impossible. You can only point out to them that the contradiction that they supposedly have discovered in freedom lies nowhere else than in this: that they, in order to make the natural law hold with regard to human actions, had to consider the human being necessarily as an appearance; and now, since you demand of them that they should think of the human being as an intelligence also as a thing in itself, they go on considering the human being in this (i.e., as a thing in itself), too, as an appearance. Of course, in this case, where a thing in itself is confused with an appearance, the separation of the human being's causality (i.e., its will) from all natural laws of the world of sense in one and the same subject would give rise to a contradiction. But this contradiction would fall away if they wanted to reflect and, as is reasonable, to admit that behind the appearances there must still lie as a ground the things in themselves (although hidden). You cannot demand that the laws governing the working of the things in themselves should be the same as those laws under which the appearances of the things in themselves stand.
The subjective impossibility of explaining the freedom of the will is one and the same with the impossibility
of discovering and making understandable an interest* which the human being might take in moral laws. Though it is impossible to understand, the human being nevertheless actually does take an interest in moral laws, and moral feeling is what we call the foundation in us of this interest. This moral feeling has been falsely given by some people as the measuring stick for our moral judgment. Moral feeling is a false measuring stick for moral judgment since moral feeling must instead be seen as the subjective effect that the law exercises on the will, while reason alone provides the will with the objective grounds of the law.
In order to will what reason alone prescribes that the sensuously-affected rational being ought to do, a faculty of reason is of course required. This faculty must instill a feeling of pleasure or of satisfaction in the fulfillment of duty; so a causality
* Interest is that by which reason becomes practical, that is, becomes a cause determining or directing the will. For this reason, you can only say of a rational being that it takes an interest in something, creatures without reason feeling only sensuous impulses. Reason takes an immediate interest in an action only when the universal validity of the maxim of the action is a sufficient ground of determination of the will. Only such an interest is pure. But if reason can direct the will only by means of another object of desire or by means of a special feeling of the subject, then reason takes only a mediate interest in the action; and, since reason by itself alone, without experience, can discover neither objects of the will nor a special feeling underlying the will as the will's ground, the latter, mediate, interest would only be empirical and not a pure rational interest. The logical interest of reason (to advance its insights) is never immediate; instead, that logical interest presupposes purposes for its use.
to configure sensibility according to rational principles must belong to reason. It is, however, completely impossible to figure out, that is, to make
a priori understandable, how a mere thought that contains nothing sensuous in itself could produce a sensation of pleasure or displeasure. Such
a priori understanding is impossible because the production of a sensation from such a thought is a special kind of causality about which, as with all kinds of causality, we can specify nothing at all
a priori; instead, to say anything about such a production, we must consult experience alone. But since experience can provide no relation of cause to effect except between two objects of experience and since here pure reason is through mere ideas (which furnish no object at all for experience) to be the cause of an effect which admittedly lies in experience, it is completely impossible for us human beings to explain how and why the universality of a maxim as law, and therefore morality, interests us. Only this much is certain: it is not because the moral law interests us that the moral law is valid for us (for that is heteronomy and dependence of practical reason on sensibility, in particular, dependence on a feeling lying as the ground of practical reason, in which case practical reason could never be morally lawgiving); instead, it is because the moral law is valid for us as human beings that the moral law interests us, since the moral law arose from our will as an intelligence and therefore from our genuine self. But what belongs merely to appearance is necessarily subordinated by reason to the make-up of the thing in itself.
So the question of how a categorical imperative is possible can for sure be answered so far as you can provide the sole presupposition under which the imperative is possible. That sole presupposition is the idea of freedom. Also, the question can be answered so far as you can see into the necessity of this presupposition, which is sufficient for the practical use of reason, that is, for confidence in the validity of this imperative and so also for confidence in the moral law. But how this presupposition itself is possible is an insight that can never be grasped by any human reason. Under the presupposition of the freedom of the will of an intelligence, though, the will's autonomy, as the formal condition under which the will can alone be guided, is a necessary consequence. To presuppose this freedom of the will is also not only (without falling into contradiction with the principle of natural necessity in the connection of appearances of the world of sense) entirely possible (as speculative philosophy can show), but it is also practically necessary. That is to say, putting freedom, as an idea and as a condition of action, underneath all voluntary actions of a rational being is necessary without further condition for a rational being who is conscious of its causality through reason and therefore conscious of a will (which is distinct from eager desires). But now how pure reason, without other incentives that might be taken from somewhere else, can be practical by itself is beyond the capability of any human reason to comprehend. That is to say, how the mere principle of the universal
validity of all of the will's maxims as laws (which of course would be the form of a pure practical reason), without any matter (object) of the will, in which you may in advance take some interest, can by itself provide an incentive and produce an interest which would be called purely moral is beyond the capability of any human reason to explain. Or, in other words: all human reason is completely incapable of explaining how pure reason can be practical, and all effort and labor spent in searching for an explanation is wasted.
It is just the same as if I were trying to figure out how freedom itself is possible as causality of the will. For in such an attempt I leave the philosophical ground of explanation and have no other ground. Now, of course, I could bumble around in the intelligible world that remains to me, in the world of intelligences; but, although I have an idea of such a world and although the idea has its good ground, I still have not the least knowledge of that world and also can never arrive at this knowledge through any effort of my natural rational faculty. The idea only signifies a something that remains when I have excluded from the grounds directing my will everything that belongs to the world of sense; I exclude everything in the world of sense merely in order to limit the principle of motives from the field of sensibility, and I bring about this limitation by confining the field and by showing that the field does not contain everything in itself but rather that there is still more outside of the field. But I do know anything further about
this 'more' that is outside of the field. After separation of all matter, that is, cognition of objects, nothing remains to me of the pure reason which thinks this ideal except the following two items. First, the form, namely, the practical law of the universal validity of maxims, remains to me. Second, it also remains to me to think, in accordance with this practical law, of reason with reference to a pure world of understanding as a possible efficient cause, that is, as a cause determining the will. Here, in these two items that remain to me, the incentive must be completely absent. If the incentive were not absent, then this idea of an intelligible world itself would have to be the incentive or would have to be that in which reason originally took an interest; but to make understandable how the idea could be the incentive or how reason could originally take an interest in the idea is precisely the problem which we are not able to solve.
This, then, is where the highest boundary of all moral inquiry is. To specify this boundary, however, is also already of the greatest importance for these reasons: so that reason, on the one hand, does not hunt around in the world of sense, in a way detrimental to morals, for the highest motive and for an understandable but empirical interest; but, on the other hand, so that reason does not powerlessly, without moving from the place, flap it wings in a space of transcendent concepts, a space that is empty for reason and that goes by the name of the intelligible world; and so that reason does not lose itself among phantoms. Yet another reason for specifying the boundary is that the idea of a pure world of understanding as a whole of intelligences to which we ourselves belong as rational beings (although on the other side at the same time members of the world of sense) always remains a useful and permitted idea for the purpose of a
rational faith. This idea of a pure world of understanding remains useful and permitted, even though all knowledge ends at the boundary of the idea, in order to produce a lively interest in the moral law that is in us. The idea produces this interest through the magnificent ideal of a universal empire of ends in themselves (of rational beings), an empire to which we can belong only when we carefully conduct ourselves according to maxims of freedom, as if the maxims were laws of nature.
The speculative use of reason, with respect to nature, leads to the absolute necessity of some highest cause of the world; the practical use of reason, with regard to freedom, also leads to absolute necessity, but only to absolute necessity of laws of actions of a rational being as such. Now, it is an essential principle of all use of our reason to push reason's cognition up to the consciousness of a cognition's necessity (for without this necessity the cognition would not be a cognition of reason). But it is also an equally essential limitation of the very same reason that reason can see into neither the necessity of what exists, what happens, or of what ought to happen, unless a condition is made the ground under which what exists exists, what happens happens, or what ought to happen happens as it ought to happen. In this way, however, because of the constant inquiry after the
condition, the satisfaction of reason is only further and further postponed. So reason searches restlessly for the unconditioned-necessary and sees itself necessitated to assume the unconditioned-necessary without any means of making the unconditioned-necessary comprehensible to reason. Reason is lucky enough if it can just find the concept that is compatible with this presupposition of the unconditioned-necessary. So it is no shortcoming of our deduction of the highest principle of morality, but instead an objection that you would have to make against human reason in general, that reason cannot make comprehensible the absolute necessity of an unconditional practical law (which is the kind of law that the categorical imperative must be); for reason cannot be blamed for not wanting to make this absolute necessity comprehensible through a condition, namely, by means of an interest that is made the ground of the necessity. Reason cannot be blamed because, if the necessity of the practical law were based on an interest, then the law would not be a moral law, that is, the highest law of freedom. And so we certainly do not comprehend the practical unconditional necessity of the moral imperative; we do, though, at least comprehend the incomprehensibility of that necessity, and that is all that can fairly be demanded of a philosophy that strives to reach up to the boundary of human reason in principles.