(You can find an even briefer summary of Kant’s ethics here.)

Kant’s Deontological Ethics

1. Kant and Hume

The German philosopher Immanuel Kant (1724-1804), called by many the greatest of modern philosophers, was the preeminent defender of deontological (duty) ethics. He lived such an austere and regimented life that the people of his town were reported to have set their clocks by the punctuality of his walks. He rose at 4 a.m., studied, taught, read, and wrote the rest of the day. He was an accomplished astronomer, mathematician, metaphysician, one of the most celebrated epistemologists and ethicists of all time, and, in many ways, the crowning figure of the Enlightenment. During the Enlightenment, European civilization celebrated the idea that human reason was sufficient to understand, interpret, and restructure the world. Perhaps the greatest rationalist ever, Kant defended this view in both his epistemology and ethics. His motto was “dare to think.”

To understand Kant, we might briefly consider his immediate predecessor David Hume (1711-1776). Hume had awakened Kant “from his dogmatic slumber,” forcing him to reconsider all of his former beliefs.  Hume’s skepticism had challenged everything for which the Enlightenment stood, and he was, perhaps, the greatest and most consistent skeptic the Western world had yet produced. He argued that Christianity was nonsense, that science was uncertain, that the source of sense experience was unknown, and that ethics was purely subjective.

Hume believed that moral judgments express our sentiments or feelings and that morality was based upon an innate sympathy we have for our fellow human beings. If humans possess the proper sentiments, they were moral; if they lack such sympathies, they were immoral. Thus, Hume continued the attack on authority and tradition—an attack characteristic of the Enlightenment—but without the Enlightenment’s faith in reason. In particular, he criticized the view that morality was based on reason which, according to Hume, can tell us about facts, but never tell us about values. In short, reason is practical; it determines the means to some end. But ends come from desires and sentiments, not from reason. Hume made these points in a few famous passages: “Reason is, and ought only to be the slave of the passions…[Thus]…Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger.”

Hume’s skepticism stunned Kant. What of the Enlightenment’s faith in reason? If desire preceded reason and desires aren’t rational, then Enlightenment rationalism was dead. How can we reestablish faith in reason? How can we show that some passions and inclinations are rational? In his monumental work, The Critique of Pure Reason, Kant attempted to elucidate the rational foundations of both the natural and mathematical sciences, defending reason against Hume’s onslaught. He then turned his attention to establishing a foundation for ethics in The Critique of Practical Reason. If morality were subjective, as Hume thought, then the concept of an objective moral law was a myth. In essence, Kant needed to answer Hume’s subjectivism and irrationalism by demonstrating the rational foundations of the moral law.

2. Freedom and Rationality 

Kantian philosophy is enormously complex and obscure. Yet, Kant’s basic ideas are surprisingly simple. His most basic presupposition was his belief in human freedom. While the natural world operates according to laws of cause and effect, he argued, the moral world operates according to self-imposed “laws of freedom.” We may reconstruct one of his arguments for freedom as follows:

1. Without freedom, morality is not possible.
2. Morality exists, thus
3. Freedom exists.

The first premise follows if we consider how determinism undermines morality. The second premise Kant took as self-evident, and the conclusion follows logically from the premises. But where does human freedom come from? Kant believed that freedom came from rationality, and he advanced roughly the following argument to support this claim:

1. Without reason, we would be slaves to our passions
2. If we were slaves to our passions, we would not be free; thus
3. Without reason, we would not be free.

Together, we now have the basis upon which to cement the connection between reason and morality.

1. Without reason, there is no freedom
2. Without freedom, there is no morality, thus
3. Without reason, there is no morality.

Kant believed moral obligation derived from our free, rational nature. But how should we exercise our freedom? What should we choose to do? 

3. Intention, Duty, and Consequences 

Kant began his most famous work in moral philosophy with these immortal lines: “Nothing in the world—indeed nothing even beyond the world—can possibly be conceived which could be called good without qualification except a good will.” For Kant, a good will freely conformed itself and its desires to the moral law. That is its duty! Nevertheless, the moral law does not force itself upon us, we must freely choose to obey it. For Kant, the intention to conform our free will to the moral law, and thereby do our duty, is the essence of morality.

The emphasis on the agent’s intention brings to light another salient issue in Kant’s ethics. So long as the intention of an action is to abide by the moral law, then the consequences are irrelevant. For instance, if you try valiantly to save someone from a burning building but are unsuccessful, no one holds you responsible for your failure. Why? Because your intention was good. The reverse is also true. If I intend to harm you, but inadvertently help you, I am still morally culpable. Kant gave his own example to dramatize the role intention played in morality. Imagine shopkeepers who would cheat their customers given the opportunity, but who do not only because it is bad for business. In other words, the shopkeepers do the right thing only because the consequences are good. If they could cheat their customers without any repercussions, they would do so. According to Kant, these shopkeepers are not moral. On the other hand, shopkeepers who gave the correct change out of a sense of duty are moral.

The emphasis on the agent’s intention captures another important idea in deontology, the emphasis on the right over good. Right actions are done in accordance with duty; they do not promote values like happiness or the common good. Kant makes it clear that dutiful conduct does not necessarily make us happy. In fact, it often makes us unhappy! We should do the right thing because it is our duty, not because it makes us happy. If we want to be happy, he says, we should follow our instincts, since instinct is a better guide to happiness than reason.

But morality cannot rest upon passions. If it did, morality would be both subjective and relative. For ethics to be objective, absolute, and precise—to be like the sciences—it needs to be based upon reason. Only the appeal to the objectivity of reason allows us to escape the subjectivity of the passions. In summary, a good will intends to do its duty and follows the moral law without consideration of the consequences.

4. Hypothetical Imperatives

But what exactly does reason command? We have already seen how reason commands actions given antecedent desires. If we want a new car, then reason tells us the various means to achieve this end. We can save or borrow the money, pray, enter a raffle, call our mother, or steal a car. But whatever we do, reason only tells us how to pursue the end; it does not tell us which ends are worth pursuing. Commands or imperatives of this sort, Kant called hypothetical imperatives, since they depend upon some desires or interests that we happen, hypothetically, to have.

Kant distinguished between two types of hypothetical imperatives. The type we have been discussing so far, what he called “rules of skill,” demand a definite means to a contingent (dependent) end.  There are also what  Kant called “counsels of prudence,” which are contingent means to a definite end. Kant recognized that happiness was a common end or universal goal for all individuals, but that the means to this end was uncertain. For example, we may think that getting a new car or losing weight will make us happy, but when we get the new car or figure we may still be unhappy. Even though the end is definite, the means to the end are not. Thus, there are no universal hypothetical imperatives because either the ends are contingent or the means to the end are uncertain.

5. The Categorical Imperative

Does reason command anything absolutely? In other words, does reason issue any imperatives which do not depend upon contingent ends or un-certain means? Hume had claimed that reason did not command in this way and that any rational commands depend upon our passions. But if absolute commands exist—commands independent of personal taste—then the essence of the moral law is revealed.

If we think about any law—say temporal relativity—we recognize immediately that law is characterized by its universal applicability. So that, if relativity theory is true, then time is relative to motion everywhere through-out the universe. Similarly, the distributive law of mathematics applies no matter what numbers we insert into it or what planet we are on. Mundane physical laws are similar. Suppose we are asked about the post-operative effects of aspirin. We do not know about the anti-clotting effects of aspirin and believe it should be given after operations. In this example, it seems clear that the truth of the matter does not depend upon us; it depends upon laws governing how human bodies respond to aspirin. Kant believed that the moral law was like this. If there really is a reason why killing innocent people is wrong, then the reason applies universally. It doesn’t matter that we want, desire, or like to kill innocent persons; we violate the moral law by doing so.

Of course, we can say that killing innocent people does not violate the moral law just as we can say that time is not relative to motion, that the distributive law works only on Monday, or that aspirin should only be given after operations. But our statements do not affect these laws; rather, the laws determine the truth of our statements. Kant held that a universally applicable moral law governs human behavior and can be discovered by human reason.

Kant had seized upon the idea of universalization as the key to the moral law. He called the first and most famous formulation of the moral law the categorical imperative: “Act only according to that maxim by which you can at the same time will that it should become a universal law.” A maxim is a subjective principle of action which reveals our intention. To universalize a maxim is simply to ask, “what if everybody did this?”  We should act according to a principle which we can universalize with consistency or without inconsistency.  By testing the principle of our actions in this way, we determine if they are moral. If we can universalize our actions without any inconsistency, then they are moral; if we cannot do so, they are immoral. Ponder these simple examples. There is no logical inconsistency in universalizing the maxim, whenever we need a car we will work hard to earn the money. However, there is something inconsistent about universalizing the maxim, whenever we need a car we will steal it.

Kant advanced five formulations of the same imperative. Another famous formulation was: “Act so that you treat humanity, whether in your own person or in that of another, always as an end and never merely as a means.” This formulation introduces us to the idea of respect for persons. Individuals are not a means to an end; we should not use people. Instead, they are ends in themselves with their own goals and purposes. Whether we use ourself or others, we violate the imperative if we treat any human being without dignity and respect. Certainly, it is true that we all use people to an extent. We use physicians, teachers, nurses, and auto mechanics to get what we want. But there is a difference between paying persons for services and using them merely as a means to your end. In the latter case, we disregard their inherent worth.

6. Perfect and Imperfect Duties

The categorical imperative commands actions in two different ways. It specifically forbids or requires certain actions, and it commands that certain general goals be pursued. The former are called perfect duties, the latter imperfect duties. Perfect duties include: do not lie, do not kill innocent persons, and do not use people. We should never perform these actions! Imperfect duties include: helping others, developing our talents, and treating others with respect. These duties are absolute, but the way we satisfy them varies. There is flexibility in how we help others, treat them with respect, or develop our talents. When we universalize a maxim that violates a perfect duty, we will an inconsistent world. When we universalize a maxim that violates an imperfect duty, we will an unpleasant world.

7. Kant’s Examples

Kant provided four examples—making false promises, committing suicide, developing our talents, and helping others—to demonstrate how the categorical imperative governs human conduct. Consider Kant’s first example, making a false promise. Can we consistently will the principle, “whenever in need of money make a false promise to get it?” We cannot, since a world where everyone acts according to this maxim would be inconsistent. This is easy to demonstrate. In such a world: 1) false promises would be useful because there would be persons to believe them; and 2) false promises would not be useful because, in a short time, nobody would believe them. Such a world is not even possible. On the one hand, it would contain the necessary preconditions for false promises to be successful—people to believe our lies—and, on the other hand, the normal and predictable result of universal false promising would be that no lies would be believed. So it is not just that this world is unpleasant; it is logically impossible!

Consider Kant’s second example. Imagine that we are depressed and contemplate suicide. Our principle of action is “whenever we are depressed we will commit suicide.” Now try to universalize a world in which everyone does this. What would it be like? In such a world: 1) people would exist to commit suicide; and 2) people would not exist to commit the suicides they intend. Such a world is not logically possible. On the one hand, it would contain the necessary preconditions of suicide—live people to commit the act—and, on the other hand, the normal and predictable result of universal suicide would be that everyone would be dead. It is easy to think of other examples. Worlds, where everyone were killers or bank robbers, would be logically impossible in the same way. Kant had demonstrated, at least to his own satisfaction, that these actions were both immoral and irrational!

If we consider the same two actions—making false promises and suicide—in terms of the second formulation of the categorical imperative, we discover that they violate it as well. If we make a false promise to someone, then we use that person as a means to our end. Analogously, if we commit suicide, then we use ourself to achieve some end. When universalization of a maxim is inconsistent or when we use ourself or others, we violate perfect duties. Kant believed that telling the truth and not committing suicide exemplify perfect duties. There are no exceptions to them.

Kant believed we have a moral obligation to develop our talents, which was his third example. Suppose we are comfortable and prefer to indulge ourselves rather than develop our talents. We act according to this maxim: “since we are reasonably well-off, we won’t develop our talents.” Upon reflection, we recognize that failure to develop our talents violates a duty and could not be universalized consistently. For if everyone failed to develop their natural talents, they would not fulfill the purpose for which those talents exist.

Furthermore, he might have added, nothing useful would be accomplished in human society without the development of talent. Yet, Kant never claimed such a world was impossible, unimaginable, or logically inconsistent. Rather, rational persons cannot will this maxim to be a universal law without disastrous and unpleasant results.

Similarly, we have a moral obligation to help others, Kant’s fourth example. Suppose we are prosperous and care little for others. We violate a duty by not helping others, and we cannot universalize the maxim. For we may need the benefit of others in the future. Again, Kant did not say this world was impossible, but he did not think any rational person desired such a world.

If we consider the same two actions—developing our talents and helping others—in terms of the second formulation of the categorical imperative, we discover similar difficulties. When universalization of a maxim has disastrous results or when we fail to treat ourselves and others as ends, we violate imperfect duties. Therefore, developing our talents and helping others are imperfect duties. They are absolute duties, but the specific means by which we satisfy these duties are open.

We may say that the categorical imperative is the formal representation of the moral law to the human mind. It commands human conduct independent of context. Compare the categorical imperative, as an abstract formulation of the moral law, to the distributive law in mathematics. This law states: a(b+c)=ab+ac. As stated, the principle is merely formal and without content. We give it content by putting numbers into the equation. The categorical imperative functions similarly in the moral domain. There, we place the maxim that operates in the moral context (situation) into the formulation to determine what to do. When we want to steal a library book or trash the sidewalk we ask, “what if everybody did this?” Recognizing the negative implications of our maxim, we see how it violates the categorical imperative. Theoretically, we may place any principle into the formulation to determine its morality. Those who do not test their maxim in this manner, turn away from the moral law.

8. Contemporary Applications

Let us consider a contemporary application of deontology to medical ethics. The emphasis on truth-telling precludes lying by health-care professionals to their patients or research subjects. Imperfect duties such as beneficence are straightforward, but how we help others is vague. The permissibility of euthanasia is also problematic. On the one hand, we may be able to universalize some forms of euthanasia or physician-assisted suicide, but, on the other hand, suicide is unequivocally forbidden. Thus, the permissibility of euthanasia depends in large part on how suicide is interpreted. The respect for persons notion is equally vague since it is not clear what it entails. Again, we are prohibited from treating ourselves or others as means, yet we should respect our’s and others’ autonomy.

9. Problems with Universalization

Despite its initial plausibility, universalization is problematic. For one thing, it is easy to universalize immoral maxims. Suppose we act according to the maxim, “Catholics should be exterminated.” There is no problem universalizing this maxim, in fact, we hope it does become universal if we really hate Catholics. The maxim “always kill Catholics,” just like the maxim “never kill Catholics,” can be universalized without contradiction by consistent Catholic-haters. Therefore, the test for universalization cannot discriminate between the two actions. We can also universalize a non-moral action like, “whenever we are alone, we sing.” We may universalize this without contradiction, but that does not mean it is moral.

It is also easy enough to think of non-moral and supposedly moral maxims which cannot be universalized. We cannot universalize maxims like, “whenever hungry, go to Sue’s diner,” or “whenever we want to go to school, go to our school.” It is not possible for everyone to go Sue’s diner or our school. More significantly, many moral actions cannot be universalize. We cannot universalize the maxim, “sell all you have and follow the Lord.” If everyone is selling, no one is buying!  We cannot even universalize a simple maxim like, “put other people first,” since everyone cannot be last! (The so-called altruist’s dilemma.)   So the test for universalization does not seem to adequately distinguish moral from immoral actions.

This brings to light a related difficulty. What maxim must we test for universalization? Maxims vary according to their generality or specificity. Kant tested very general maxims for universalization. “We cannot lie to achieve an end.” Suppose we made the maxim more specific. “We cannot lie except to save innocent people from murder.” This maxim is universalizable and spares us from telling the truth to inquiring murderers who ask the whereabouts of their intended victims. We could make the maxim even more specific. “We cannot lie except to save innocent people from murder and to spare people’s feelings.”

The problem is that as maxims become more specific, more questionable maxims become capable of consistent universalization. Eventually, we would be testing very specific maxims. Suppose a bald, bearded, philosopher professor, Horatio Rumpelstiltskin, was about to steal a book from the college library on Thursday at noon. He would discover, upon careful examination, that he could universalize a world where all so named and described individuals stole books at precisely that time. If maxims become too specific, universalization has no meaning. Thus, maxims must have some generality to be properly tested.

Now suppose I test the following maxim. “We cannot lie except to achieve our ends.” This maxim is sufficiently general to be universalized, but not sufficiently specific to rule out immoral actions. And the problem is not ameliorated by turning to the second formulation of the imperative. Does respect for persons tell us anything about whether we should universalize general or specific maxims? Should I always respect persons or always respect them except in certain situations? It appears that universalization is not as simple as it initially appeared.

10. General Difficulties

Kant claimed that duties are absolute. If duties are absolute, then what about conflicts between duties? Kant states that perfect duties supersede imperfect ones, and thus the duty not to lie precedes the duty to help others. If this is so, it follows that we must tell the truth to inquiring murderers. But this presented great difficulties for Kant. Surely duties have exceptions and perfect duties are not sacrosanct. Kant might have avoided this difficulty, as we have seen, by advocating that we universalize maxims with exceptions. A maxim like, “never lie except to inquiring murderers,” is not problematic.

Along these lines, the twentieth-century philosopher W.D. Ross argued that no duties were absolute. Ross, who taught at Oxford for nearly fifty years and was one of the world’s great Aristotelian and Kantian scholars, tried to modify Kant’s theory to account for conflict of duty cases. according to Ross, we have prima facie—at first glance—duties, but they are conditional. Our actual duties—at second glance, you might say—depend upon the situation. In conflict of duty cases, we carefully weigh our duties and then proceed to do the best we can. The problem is whether Ross’ conception of duties is too subjective and situational since individuals decide which duties apply in given situations. The main problem with Ross’ version of deontology is its emphasis on subjects and situations, an emphasis Kant wanted to avoid.

Another problem with Kant’s system is that it is so formal and abstract it hardly motivates us. Even if Kant could prove that ethics were completely rational, wouldn’t this take something away from the importance of moral choice? Isn’t ethics too messy and imprecise for the formality, precision, and logic of Kant’s system? Aristotle said that ethics could never be so precise. Maybe Kant demanded too much precision from his ethics?

Another general difficulty is Kant’s rejection of the importance of consequences. According to Kant, if we do our duty we are absolved of all responsibility for the consequences of our action. He defends this view in part because he believes we can never know the consequences with certainty. This is true to an extent, but this view rests upon very pessimistic assumptions about our knowledge of the consequences of our actions. If for no apparent reason we tell our friend she looks positively awful and disgusting, we can be pretty sure she will feel bad about this. We are hardly absolved by our claim that we were not sure she would feel bad. Sometimes we can be reasonably sure of the consequences, in which case duty may not be important. Much trouble has been caused by people who were simply “doing their duty.”

11. Kant’s Fundamental Idea

Despite the nuances connected with the idea of universalization, there is a core idea at the heart of Kant’s theory which is his lasting legacy. We have all been reprimanded by someone saying “how would you like someone to do that to you?” This is Kant’s fundamental idea. If there is a reason why you don’t want people to do something to you, then that same reason applies to what you want to do to others. It gives you a reason not to treat others in a way that you do not want to be treated. And, if you ignore that reason, you are acting irrationally. This is the kind of rational constraint Kant believed imposed itself upon our conduct. Of course, we have all experienced people who believe that the rules that apply to us do not apply to them, and, if they are bigger or more powerful than we are there is not much we can do. They might say to us, “You help me move on Saturday, but I won’t help you move next week.” We feel that they are doing something unfair and inconsistent, whether or not they recognize it. That is Kant’s fundamental idea. A reason for one is a reason for all.

A purely rational morality is a fascinating idea. We saw in an earlier chapter how moral judgments might be truths of reason. Whether this is true depends upon our understanding of concepts like rationality, interests, and individuality. In the strong conception of rationality, others’ interests give us a reason to act. In the weak conception, others’ interests do not give us a reason. This issue also relates to the earlier discussion of egoism between Kalin and Medlin. If we think other people should respect our interests, so the argument goes, then we should respect theirs. But when we say others should respect our interests does that mean: 1) we want them to respect our interests; or 2) they have a reason to respect our interests. Kant and his contemporary followers argue for “2,” while other philosophers argue for “1.” Clearly, we want others to act in our interest, but it is not clear our interests give others a reason to act.

A conception of individualism is also relevant. If we have a strong conception of individuality—one in which individuals are radically separate—it is hard to see how the other’s desires/interests/wants give us a reason to do anything. On the other hand, if we have a weak conception of individuality —one in which all individuals are intimately connected—it is easy to see how the other’s interests give us reason to act. Maybe the rise of individualism lessens our sense of obligation toward others, or maybe communalism lessens our sense of obligation toward ourselves. Whatever our conclusions, the conceptions of rationality, interests, and individuality play a significant role in determining whether Kant’s primary idea is convincing for us.

Kant’s basic idea is that morality is grounded in reason. Essentially, if there really is a reason why we should not commit immoral acts, then that reason applies to all of us. If there really is a reason to treat people with dignity and respect, or not to lie or cheat them, then this reason applies to all of us whether we want it to or not. To say there are universal moral reasons ultimately confirms our belief in the intelligibility of reality. And, if the moral universe is unintelligible, nothing matters.

12. Conclusion

Despite all the positive contributions of Kant’s moral thought, one final difficulty plagues the theory. Kant argued that the good life is a life of duty and that other lives are not worthwhile. But there have been many decent and happy lives that were not motivated by duty. Consider also persons who live from a sense of duty, but who are miserable and unhappy. They live without love, compassion, pleasure, beauty, or intellectual stimulation. Are such individuals moral exemplars? True, many live decadent lives in exclusive pursuit of pleasure or happiness while dismissing moral virtue. But Kant’s ethics suffer from its emphasis on duty and virtue while neglecting happiness and pleasure. And if a philosophy stresses duty over happiness, then why should we do our duty? Duty may be part of morality, but so is happiness. We now turn to a moral theory which emphasizes the good over the right, happiness over duty. That theory is utilitarianism.

(You can find an even briefer summary of Kant’s ethics here.)