This section, the longest of the three, comprises about half of the Foundation. It is the best known and most quoted section, containing those ideas which many consider the core of Kant’s ethical thought. Here we find an explicit description of the categorical imperative, the idea of man as a being of inherent value and dignity, the concept of a kingdom of ends, the autonomy of the will, and the raising of the critical question of the validity of the categorical imperative as a genuine law governing human conduct. It is a well-packed section, including a famous quartet of examples, three or more formulations of the categorical imperative, and criticisms of opposing theories.
Because of its length, Section II will be presented in four chapters. Chapter 5 will examine the first few pages, in which Kant recapitulates his discussion of Section I, indicates what is to come, and gives reasons why we must follow this procedure. In Chapter 6, Kant begins the main argument, investigating what it means to act from principle or from the idea of law itself. He will argue that men act from either of two kinds of principle: one based on self-interest, the other based on reason.
Chapter 7 is an interlude in the main argument that has led some to interpret Kant’s Foundation as primarily an ethical treatise,1 rather than as a search for the foundations of morality itself. The interlude is primarily ethical, suggesting certain rules of conduct, specifying examples of moral action, and posing ideals for man’s moral understanding.
In the fourth part of Section II, Chapter 8, Kant returns to the main argument, showing that the basic principle of morality—the categorical imperative—is only valid as a genuine law of human conduct if the will is self-legislating. All other moral foundations prove invalid. He thus arrives at the basic question of the critique of practical reason (the main issue of Section III) : how is it possible for the will to give law to itself?
Some readers may be puzzled by the title of this section. Section I consisted of a transition from ordinary rational knowledge. Section II, however, does not begin with this philosophical knowledge, but proposes a transition from popular moral philosophy to a metaphysics of morals. What is the popular moral philosophy to which Kant alludes? As we shall see, he thinks very little of some varieties of popular philosophy, calling them a “disgusting mishmash of examples strung together by crackpot principles,” lovingly produced by boneheads. Kant never explicitly describes the popular moral philosophy which he employs, nor does he explain why he takes this approach.
We can, however, discover his reasons for ourselves from his other writings. Likewise we will have to examine the rationale of his method. We should notice here, however, that there is a great difference between a popular moral philosophy and a philosophy derived from ordinary rational knowledge. The latter has an a priori structure, whereas the former may be based on any number of grounds, such as emotion, love, fear of God, benevolence, and so on. In each case, however, Kant’s method is the same: he takes what is already at hand and draws from it the a priori ground upon which such principles must be based if they are to have any validity for moral decision.
Kant begins this section with a review of the a priori method used in Section I: namely, that any examination of moral value must be divorced from empirical conditions.
Even though we discovered the idea of duty by analyzing how commonsense moral reasoning functions, we must not think that we derived this newfound idea from experience. On the contrary, when we examine our experiences of the way men do act, we find no instances which provide us with unquestionable examples of actions done purely for the sake of duty. Indeed, of the many actions which accord with duty, some persons suggest that since the motive is always in question, so the moral value of the action is doubtful.
For this reason, every age has its philosophers who absolutely deny that duty is a real motive in human choice. Without first finding out whether their own ideas of morality are correct, they turn all human motives into sophisticated varieties of self-interest. They express a profound regret that human nature, which nobly holds up to itself the respected ideal of duty, is yet too weak and corrupt to live by it. Reason, they say, ought to rule human activity, yet it merely serves the satisfaction of some single desire or (in better men) the harmonious satisfaction of all desires.
We have all met the amateur psychologist who will question our apparent motives for any action. If my father happens to make me angry, the cause is not so much what he does as it is my subconscious rivalry for my mother’s love. Or if I dislike a certain teenage vocalist, it is not his voice or bearing which displeases me, but a hidden homosexual attraction that causes my overt reaction of displeasure. Etc., etc. Behind such callow analyses lies the truth that no motive is unmixed. There may be hidden elements involved in every human choice; we can never say with certainty that any particular action was done solely and purely from duty or for the sake of moral righteousness. We cannot derive the pure concept of duty from any examination of mans actual performance, nor will such evidence explain what it means to act from duty.
David Hume believed that no idea by itself could ever cause us to act. Actions he said are passional, initiated by internal stimulus or desire. We must “feel for” something before we “act for” it. Reason’s role is merely to determine the best way to satisfy our desires or achieve our goals (which for Hume were the same things). Hume recognized that men have feelings in common; life is not psychological chaos. He explained man’s general preference for goodness and the peaceful life by postulating a “universal sentiment of approbation,”2 a kind of emotional conscience common to all men. All we must do to learn about this sentiment is to study the moral judgments of men.
To some extent, of course, Hume is correct. Kant’s basic disagreement with Hume on this point is that the a priori concept of an action done purely for the sake of duty could not originate from any examination of what men do, since there may never have been such a case.
As a matter of fact, it is absolutely impossible for experience to prove beyond question even one case in which the agent’s maxim for performing a duty was moral, that is, a case in which he acted purely from the idea of doing his duty. Certainly there are times when the most careful self-examination reveals nothing to us but the moral motive of duty as powerful enough to persuade us to do the good deed in question, something demanding great sacrifice. Yet it is still possible that what seems to be the idea of duty is in fact a hidden impulse of self-interest moving the will. We like to puff up our egos with the notion that we act from noble motives but, as it turns out, not even the most careful examination can discover the bottom of the well of our subconscious impulses. Our difficulty is that in moral matters we are dealing not with visible actions but with the unseen inner principles of action.
If we are to base our examination on experience, then how can we determine what goes on deep within the psyche, far removed from any possible introspective view? Hidden motives are just that—hidden. We can never, from an empirical examination, conclude absolutely that no hidden motive is involved in a particular case. We all like to believe that our motives are pure, especially in moral situations, but we can never be sure. Since the moral worth of an action is determined by the motive, then no experience of what an agent does will give us any sure knowledge of the moral value of the action. But, if we have no sure knowledge of actions done from duty, perhaps there are none.
Consider a further consequence of claiming that we derive the idea of duty from experience only. By this we would assure victory for those who belittle all morality as being only a figment of a vain, human imagination, stretching beyond its limits. (They already believe that all other ideas are drawn from experience.) Because I think well of my fellow man, I readily believe that most human actions conform to the norms of duty. However, if we carefully watch our intentions and goals, we generally find the ever-present and dearly beloved Self at their heart, not the stern command of frequently painful duty.
We cannot call someone an enemy of virtue simply because he doubts whether there has ever been an act of genuine virtue. He may be an objective observer who, as he grows older, is led by his added experience and sharper moral insight to see the difference between an ideal and reality. Since there may never have been such an act, the only way for us to insure our understanding of duty and secure a basic respect for law is a firm belief that reason itself, without depending on experience in any way, can tell us what we ought to do—even if no one has ever acted purely from reason.
We are concerned with those deeds which the world may never have seen, with deeds which the behaviorists believe to be impossible, namely, deeds which reason inexorably commands. For instance, we insist that a friend be completely sincere, and we continue to insist on it even though no such friend ever lived. The duty of perfectly sincere friendship comes under the idea of duty in general, so is independent of experience, and arises from reason guiding the will on a priori grounds.
The older and wiser we become, the more evident it is that, as Buttercup observes, “Things are seldom what they seem.” If we are to retain our conviction that an action done for duty’s sake is at least possible, we must appeal to reason alone, not to experience. Experience will tell us only what was, what is, or what may be—what may please, displease, cause pain or joy. Reason alone informs us of what we ought to choose. But reason cannot command the will under necessity if it derives its rule from experience.
Take friendship. We know what a perfect friend ought to be. But has anyone ever known such a friend? No, for a man would have to be more than human to be a perfect friend. We know that our friends, however loyal and true, are imperfect and might possibly fail us. Yet we still expect them to be perfect. We still insist that a friend ought to be perfect. But if we never saw or knew a perfect friend, how do we know what one should be like? Only by reason, says Kant. Our reason, understanding the nature of true friendship, determines what a true friend ought to be. This idea, arrived at independently of empirical evidence, is a priori.
Obviously experience cannot provide instances from which we could derive absolute laws. But even more, unless we want to reject the truth of morality and its applicability to anything at all, we must allow that moral law is so all-embracing that it rules not men only, but any being which is rational. The moral law must be absolute, not subject to conditions, provisions, or exceptions. How could we respect something less than absolute, something subject to the changing conditions of human existence? How could a law which applies to human beings also be a law which applies to all rational beings—including humans—unless that law were derived, not from experience, but a priori from pure practical reason?
Morality must be applicable to all rational beings, and not merely to human beings.3 We know human beings only through experience, but rational beings we can know by reason. If the principles of morality are independent of experience, then any rational being would be capable of discovering them and would be equally bound by them. We do not know, of course, whether there are rational beings other than ourselves. But if there be such beings (e.g., rational angels), they are morally responsible agents. They must obey the moral law as much as any of us humans.
This directly opposes those who claim that we can distinguish moral right from moral wrong by a study of mans nature (psychology), his primative cultures (anthropology), his present culture (sociology), his needs and projects (pragmatism), his spiritual needs (theology), or his physical needs (physiology). Morality, Kant insists, is based on man’s rationality and it is as a rational creature that man become a moral one.
No advice can be worse than to suggest that morality be derived from examples. Any example of morality must first be judged by some moral standard to find out whether it can serve as a good example to imitate. How then can an example itself serve as the ideal of morality? Even the Holy One of the Gospel, before we can call Him good, must be judged by an ideal of moral goodness. He says of Himself: “Why do you call me (whom you see) good? God alone (the ideal of goodness, whom you cannot see) is good.” But where else do we get the idea of God’s supreme goodness except from the a priori idea of moral perfection and its logical relation to the idea of free will?
Imitation is out of place in moral decision. Examples allow nothing more than encouragement, by proving that what the law commands can be done and by providing specific instances of what the law commands in general. But in seeking a moral ideal, we can never substitute examples for the true original in reason.
Once again, but from a different approach, Kant denies that morality can be based on what men actually do. Examples serve only to illustrate what has already been judged by some general principle. Nor can moral principles be derived from what most or all men do. We may discover in men’s actions a certain common principle and to that extent an examination may be useful. But it is one thing to learn that men follow a certain principle and quite another to justify that principle as one they ought to follow, i.e., as a valid moral principle. We can never justify our actions by arguing that “everybody is doing it.” That is lazy man’s morality and is more often than not an excuse of the crudest sort.
The remark about the “Holy One of the Gospel” raises an age-old question: is God called good because we recognize His goodness or is He called good simply because He is God? For Kant, the answer is simple. If we call God good, it is because God conforms to our a priori standard of what goodness is. If there were no rational beings, God would not be called good, since He would fulfill no conditions in any rational understanding. And yet God would be no less perfect than He is. We have certain conceptions of goodness, and we call good those beings which conform to them. Such requirements for goodness are rational, not empirical. For we call God, whom we cannot see, good. How could we call God good if our idea of goodness were based on what we saw only?
Morality cannot be based upon any foundation other than reason. Moral rules must be justifiable by a priori concepts, such as the ideas of duty and law. The deeds of humankind may show us that men do live by such moral rules, but their deeds cannot justify themselves. We may learn of the law through experience, but we recognize it—see it as law—by reason alone. Law contains within itself the necessity of compliance, and experience grants no such necessity.
Kant is well aware of the lack of popular demand for an a priori investigation. Were all men philosophically inclined, the desirability of such an examination would be obvious to everyone. Yet even in that case, few would be eager to undertake such a difficult search.
Suppose there were no genuine supreme principle of morality, no principle derived from pure reason and independent of any experience. Were this so, I could see no point to a search for a priori concepts and the principles associated with them. (We could not even distinguish ordinary knowledge from philosophical knowledge.) But the search for an a priori principle is especially relevant to our era. In a contest of preference between a pure rational knowledge that lacks any empirical ground (i.e., a metaphysics of morals) and a popular moral philosophy—well, we can easily guess which side would win.
No thoughtful person will deny the necessity of theoretical research, particularly in the scientific field. But though the need for research is less obvious in moral philosophy, it is just as great. Furthermore, investigation into the abstract principles of moral conduct must precede any popularization of the principles. Yet the opposite procedure is generally followed: moral standards in effect are taken as the source of a moral theory, rather than moral theory being used to judge the standards we follow. Kant waxes sarcastic about such a reversal of procedure.
We must grant the value of a popular version of moral philosophy, but only after we have gained a firm grasp of the principles of pure reason. Thus our first task must be to prove the metaphysical basis of morality, and only then to seek popular approval. But how thoughtless it would be to reverse this order, to seek popularity first, leaving the discovery of true principles to follow. To begin with, such a procedure would prevent any claim to genuine philosophical popularity, since no skill at all is needed to explain a simple-minded theory.
What is more, this so-called popular theory would consist of a disgusting mishmash of examples strung together by crackpot principles, the kind of nonsense so often heard in conversations with boneheads. But persons with any sense at all will be confused and discouraged without knowing why. Even philosophers, who easily see through the confusion, find few followers when they seek to persuade men that worthwhile popularity must wait upon a firm grasp of true principles.
Those who do feel some need for a priori principles, who are aware that these principles must be based on man’s primary claim to moral responsibility, his reason—even they do not undertake the search without at the same time trying to popularize it. Yet it should be a separate enquiry. Kant mentions below a few of these mixed theories, previewing his refutation (at the close of Section II) of the spurious theories of morality. But Kant follows his own rule: only after he has completed his a priori investigation of the genuine foundation of morality does he reject these other principles.
Just glance at the work of moral philosophers who try to be popular. We find principles based on man’s unique human nature (or sometimes his rational nature), the ideal of perfection or happiness, a moral sense, the fear of God—a pinch of this, a dash of that, making a delicious hash. But none of these writers ever wonders whether he ought or ought not seek principles in a knowledge of human nature, which he knows only by experience. Even if he raises this question—that is, even if his principles be purely a priori, based solely on ideas of reason and on nothing else at all in experience—he never wonders whether he should pursue moral philosophy as a distinct branch of pure philosophy, namely as a metaphysics* of morals (to use a term of low popularity). Thus these writers fail to complete their investigations. They speak out too soon, rather than requesting their readers, who clamor for a popular version, to wait until the work is properly finished.
* We can distinguish pure moral philosophy (metaphysics) from applied moral philosophy (applied to the human situation). We make the same kind of distinction between theoretical and applied mathematics and between abstract and applied logic. This distinction emphasizes that moral principles must be a priori and self-validating, rather than dependent upon the particularities of human existence. But, on the other hand, it also emphasizes that from the a priori principles we must be able to derive practical moral rules which do apply to our human circumstances.
To have a positive knowledge of our duties, we must first possess a metaphysics of morals which completely lacks any dependence on anthropology, theology, physics, and, especially, cultish doctrines such as theosophy. In fact, if we do possess such an a priori foundation for our moral principles, our compliance with the command of duty is much more likely. For when the pure idea of duty and the moral law join forces, without help from any sensuous inclinations, their combined effect on the human heart exceeds any other motive or desire known to us.* Aware of its own worth and coming to realize that it can act as its own master in human affairs, reason loathes the lesser motives and gradually enslaves them. But when reason tries to decide on the basis of a moral theory which mixes feelings and desires with a priori ideas, reason wavers between this motive and that, none of which can be derived from a valid moral principle. Such motives may accidentally lead to good results, but they may just as well produce the opposite.
* The eminent Herr Sulzer wrote me to ask why moral theories are so ineffective, even though much in them is quite reasonable. I have put off answering him until I could offer a complete explanation. The fault lies with teachers who fail to make their ideas clear. They look elsewhere for motives which will induce their charges to be moral, but by trying to find a compelling theory, they ruin the whole affair. For even the man on the street can see that an act of honesty, performed with a firm determination, without any hope of reward here or hereafter, and in the face of the strongest obstacles of human necessity or sensual temptation—that such an act exceeds in value and outshines any act performed for some non-moral reason. To imagine such actions inspires us to imitate them. Even fairly young children feel this, and we should never try to explain duty to them in any other way.
No motive appears to the will so powerful as that of pure obligation. Men will occasionally do what they feel they ought to do in the face of misfortune and death. We honor the true heroes of history who braved danger to do what they believed to be right. But when we see the motive of someone’s honorable action to be the allurement of gain, the applause of the crowd, or the thirst of desire, our picture of moral honor fades and the agent appears no more the true hero but merely mortal, like us. Leander performed a mighty deed in swimming the Hellespont each night. But his purpose was to spend the evening with his ladylove, Hero. When he drowned, he drowned in the pursuit of his desires, not in the performance of his duty. He is an inspiring example to separated lovers but not an example of moral righteousness. Kant saw many practical difficulties for those who try to follow a “mixed” theory of morality. We see such difficulties today in the quandry of many young people regarding sexual desires. On the one hand they live in a society which emphasizes sexuality in almost every area. Songs are filled with obvious references to physical intimacy: the atmosphere churns with ideas of individual standards and the testing of the older morality. On the other hand, they are told that pre-marital intimacy is sinful, that it leads to a lessening of genuine love, that early marriage is unwise. What is right, what wrong? Their minds “waver between this motive and that, none of which can be derived from a valid moral principle.” If their ardent urges are dangerous, why so much emphasis on satisfying them? If they follow where passion leads, why are they condemned for doing what was so alluringly suggested? The mixed theories provide no solution, and the average young man and woman are woefully confused.
Kant’s footnote offers valuable practical insight. One should never represent moral goodness to children in any guise but its own. If we can assume a parent’s basic knowledge of right and wrong, then we must see that it is an error for him to “motivate” a child into morality. Suppose a parent trains his child to follow the moral law because in that way he will achieve his own best interests. When the child grows to adulthood, he will then decide questions of morality on the basis of the same motive of self-interest. However, if the child is trained to do what is right because it is the moral thing to do whatever the apparent consequences to his own interests, then when the child is grown he will decide, not from self-interest, but from moral character. Suppose we see two men, both doing what we know to be morally wrong. One of the men is acting from self-interest, while the other is acting from an erroneous belief about what he ought to do. Who is more likely to be changed by an appeal to principle? Obviously the man in error. His fault is one of honest mistake, but the other is pursuing what he desires, whether it be moral or not. When we consider that most men do have a commonsense idea of what is right and wrong, we can see the immediate and long-range benefit of training a child always to do what he knows to be right because it is right.
Someone may ask: “Isn't it better to teach a child to do the right thing for some reason, rather than merely because it is right?” Such a question indicates a lack of understanding of the nature of moral goodness. Moral rules are not set up as means to some other end: they are established so that people may know or judge what is the right thing to do. That an action is the right thing to do is the best reason for doing it. Any other reason is a non-moral reason, annulling any possible moral credit for the action. If a person obeys a moral rule because he wishes to reach heaven, his motive is not a moral one but a desire for spiritual reward. He acts, not because the action is itself right, but because it is a means to something he wants. There is a subtle paradox about this. If a person does what is right because he believes it to be the right thing to do, he will always reach heaven. However, if he does what he believes will gain him heaven because he wants to gain heaven, he could conceivably fail to reach that celestial estate. It is a matter of motive. For Kant, the moral motive, acting for the sake of moral rightness (aus Pflicht), is the best of all possible motives and the only one which can guarantee salvation, if salvation is in any way a matter of human fortune.
Having argued his case for an a priori investigation, Kant briefly summarizes his argument.
- Moral concepts are derived from reason.
- They cannot be derived from experience.
- The more empirical the principle, the less it will be a valid moral law.
What conclusions can we draw at this point? First, every moral idea is a priori, having its foundation and source in reason alone. The purity of this source makes such ideas the basis of worthwhile and ultimate moral rules. Whether the ideas arise in ordinary moral thinking or on the highest theoretical plane, they clearly cannot be taken from experience since empirical knowledge lacks necessity. Whenever moral ideas become mixed with experience, they lose their validity and the actions they cause lack unconditional value.
- 2. Moral ideas are based on reason alone.
- Since we know human nature by experience, this knowledge cannot be the source of morality.
- Only from reason can moral principles acquire their universal validity.
Secondly, the ideas and laws of morality, both in theory and in practice, must be derived from pure reason alone and presented in a pure, unmixed setting. Furthermore, we must discover the entire scope of pure practical a priori knowledge—that is, the limits of pure practical reason. However, we must not seek moral principles in the particular nature of human reason—even though speculative philosophy must follow this very course. Moral laws apply to any being that can reason, and so we must derive moral rules from the abstract idea of a rational being as such.
- 3. Without an a priori moral philosophy:
- We cannot define duty in general.
- We cannot determine our duties in particular.
Any moral philosophy which applies to human beings must incorporate some empirical study of mankind (e.g., anthropology). But we must first develop moral philosophy as a pure metaphysics of morals, without involving experiences of mankind. (The philosophical method simplifies this procedure.) Clearly, if we lack an a priori metaphysics of morals, we would be foolish to look for the common moral element in all dutiful actions in order to have a standard for theoretical conclusions. Nor would we have valid principles of morality for everyday moral decisions or for moral instruction. Yet only by such valid principles can genuine moral habits be instilled in the human soul, prompting it to seek the highest possible good.
Paraphrased, Kant’s summary might read: “If we do not justify morality by basing it on man’s reason, we will not understand what it means to say that there are duties which we ought to perform. Furthermore, it will be impossible to point to a single action as an unquestionable instance of duty. It will be impossible to teach children moral principles except on grounds of expediency. And finally, we will have lost the surest and best possible appeal for good, the appeal to moral character.”
To anyone who reads contemporary ethical literature, especially works by those who deny an a priori foundation for morality, Kant’s appeal seems all too timely and his argument weighty. Almost without exception such writers are forced to a position of moral relativism or, even further, to a denial that morality is anything but emotional persuasion. Kant’s position is quite clear: If we have duties which are indeed duties, then their claim upon us must be based on reason. If we cannot discover the source of their claim, our belief that there are duties may be a delusion. So, in order to have an unshakable foundation for our moral beliefs, we surely must search for such a source. And we must search for it in reason alone, by reason alone.
So far in our undertaking we have examined ordinary moral judgments (which are certainly worthy of our respect) to find their philosophical foundation. Now we will investigate popular moral philosophy, which is derived principally from examples, as a first step towards a metaphysics of morals. Since metaphysics encloses the entire sphere of rational knowledge (including abstract ideas), experience and examples are little help. So in order to progress from popular morality to the metaphysics of morals, we must examine how reason derives the idea of duty from universally applicable laws.
Section I dealt with the concepts of the good will, duty, and law. By an analysis of these concepts as they are found in the principles of ordinary moral thinking, Kant determined that universality is the fundamental characteristic of the ultimate principle of morality.
However, we can do only so much with an analysis of concepts. To discover a principle of morality, we must examine principles, particularly principles found in moral theories, in order to learn what it means to act on principle or from the idea of law. Though universality is the a priori characteristic of morality, there remains the problem of relating universality to human action. By an a priori analysis of what it means to act on principle, Kant will deduce the supreme principle of morality as a general moral law, which he calls a categorical imperative. From this he will proceed to the idea of autonomy as the ultimate foundation of morality. Autonomy will finally be seen as an Idea of pure practical reason, of which experience offers no examples.