A Simplified Account of Kant's Ethics

Onora O'Neill

From Matters of Life and Death, ed. Tom Regan
Copyright 1986, McGraw-Hill Publishing Company.
Excerpted in Contemporary Moral Problems, ed. James E. White
Copyright 1994, West Publishing Company

Kant's moral theory has acquired the reputation of being forbiddingly difficult to understand and, once understood, excessively demanding in its requirements. I don't believe that this reputation has been wholly earned, and I am going to try to undermine it.... I shall try to reduce some of the difficulties.... Finally, I shall compare Kantian and utilitarian approaches and assess their strengths and weaknesses.

The main method by which I propose to avoid some of the difficulties of Kant's moral theory is by explaining only one part of the theory. This does not seem to me to be an irresponsible approach in this case. One of the things that makes Kant's moral theory hard to understand is that he gives a number of different versions of the principle that he calls the Supreme Principle of Morality, and these different versions don't look at all like one another. They also don't look at all like the utilitarians' Greatest Happiness Principle. But the Kantian principle is supposed to play a similar role in arguments about what to do.

Kant calls his Supreme Principle the Categorical Imperative; its various versions also have sonorous names. One is called the Formula of Universal Law; another is the Formula of the Kingdom of Ends. The one on which I shall concentrate is known as the Formula of the End in Itself. To understand why Kant thinks that these picturesquely named principles are equivalent to one another takes quite a lot of close and detailed analysis of Kant's philosophy. I shall avoid this and concentrate on showing the implications of this version of the Categorical Imperative.

Kant states the Formula of the End in Itself as follows:

Act in such a way that you always treat humanity, whether in your own person or in the person of any other, never simply as a means but always at the same time as an end.

To understand this we need to know what it is to treat a person as a means or as an end. According to Kant, each of our acts reflects one or more maxims. The maxim of the act is the principle on which one sees oneself as acting. A maxim expresses a person's policy, or if he or she has no settled policy, the principle underlying the particular intention or decision on which he or she acts. Thus, a person who decides "This year I'll give 10 percent of my income to famine relief' has as a maxim the principle of tithing his or her income for famine relief. In practice, the difference between intentions and maxims is of little importance, for given any intention, we can formulate the corresponding maxim by deleting references to particular times, places, and persons. In what follows I shall take the terms 'maxim' and 'intention' as equivalent.

Whenever we act intentionally, we have at least one maxim and can, if we reflect, state what it is. (There is of course room for self-deception here--"I'm only keeping the wolf from the door" we may claim as we wolf down enough to keep ourselves overweight, or, more to the point, enough to feed someone else who hasn't enough food.)

When we want to work out whether an act we propose to do is right or wrong, according to Kant, we should look at our maxims and not at how much misery or happiness the act is likely to produce, and whether it does better at increasing happiness than other available acts. We just have to check that the act we have in mind will not use anyone as a mere means, and, if possible, that it will treat other persons as ends in themselves.

To use someone as a mere means is to involve them in a scheme of action to which they could not in principle consent. Kant does not say that there is anything wrong about using someone as a means. Evidently we have to do so in any cooperative scheme of action. If I cash a check I use the teller as a means, without whom I could not lay my hands on the cash; the teller in turn uses me as a means to earn his or her living. But in this case, each party consents to her or his part in the transaction. Kant would say that though they use one another as means, they do not use one another as mere means. Each person assumes that the other has maxims of his or her own and is not just a thing or a prop to be manipulated.

But there are other situations where one person uses another in a way to which the other could not in principle consent. For example, one person may make a promise to another with every intention of breaking it. If the promise is accepted, then the person to whom it was given must be ignorant of what the promisor's intention (maxim) really is. If one knew that the promisor did not intend to do what he or she was promising, one would, after all, not accept or rely on the promise. It would be as though there had been no promise made. Successful false promising depends on deceiving the person to whom the promise is made about what one's real maxim is. And since the person who is deceived doesn't know that real maxim, he or she can't in principle consent to his or her part in the proposed scheme of action. The person who is deceived is, as it were, a prop or a tool-a mere means--in the false promisor's scheme. A person who promises falsely treats the acceptor of the promise as a prop or a thing and not as a person. In Kant's view, it is this that makes false promising wrong.

One standard way of using others as mere means is by deceiving them. By getting someone involved in a business scheme or a criminal activity on false pretenses, or by giving a misleading account of what one is about, or by making a false promise or a fraudulent contract, one involves another in something to which he or she in principle cannot consent, since the scheme requires that he or she doesn't know what is going on. Another standard way of using others as mere means is by coercing them. If a rich or powerful person threatens a debtor with bankruptcy unless he or she joins in some scheme, then the creditor's intention is to coerce; and the debtor, if coerced, cannot consent to his or her part in the creditor's scheme. To make the example more specific: If a moneylender in an Indian village threatens not to renew a vital loan unless he is given the debtor's land, then he uses the debtor as a mere means. He coerces the debtor, who cannot truly consent to this "offer he can't refuse." (Of course the outward form of such transactions may look like ordinary commercial dealings, but we know very well that some offers and demands couched in that form are coercive.)

In Kant's view, acts that are done on maxims that require deception or coercion of others, and so cannot have the consent of those others (for consent precludes both deception and coercion), are wrong. When we act on such maxims, we treat others as mere means, as things rather than as ends in themselves. If we act on such maxims, our acts are not only wrong but unjust: such acts wrong the particular others who are deceived or coerced.

Duties of justice are, in Kant's view (as in many others'), the most important of our duties. When we fail in these duties, we have used some other or others as mere means. But there are also cases where, though we do not use others as mere means, still we fail to use them as ends in themselves in the fullest possible way. To treat someone as an end in him or herself requires in the first place that one not use him or her as mere means, that one respect each as a rational person with his or her own maxims. But beyond that, one may also seek to foster others' plans and maxims by sharing some of their ends. To act beneficently is to seek others' happiness, therefore to intend to achieve some of the things that those others aim at with their maxims. If I want to make others happy, I will adopt maxims that not merely do not manipulate them but that foster some of their plans and activities. Beneficent acts try to achieve what others want. However, we cannot seek everything that others want; their wants are too numerous and diverse, and, of course, sometimes incompatible. It follows that beneficence has to be selective.

There is then quite a sharp distinction between the requirements of justice and of beneficence in Kantian ethics. Justice requires that we act on no maxims that use others as mere means. Beneficence requires that we act on some maxims that foster others' ends, though it is a matter for judgment and discretion which of their ends we foster. Some maxims no doubt ought not to be fostered because it would be unjust to do so. Kantians are not committed to working interminably through a list of happiness-producing and misery-reducing acts; but there are some acts whose obligatoriness utilitarians may need to debate as they try to compare total outcomes of different choices, to which Kantians are stringently bound. Kantians will claim that they have done nothing wrong if none of their acts is unjust, and that their duty is complete if in addition their life plans have in the circumstances been reasonably beneficent.

In making sure that they meet all the demands of justice, Kantians do not try to compare all available acts and see which has the best effects. They consider only the proposals for action that occur to them and check that these proposals use no other as mere means. If they do not, the act is permissible; if omitting the act would use another as mere means, the act is obligatory. Kant's theory has less scope than utilitarianism. Kantians do not claim to discover whether acts whose maxims they don't know fully are just. They may be reluctant to judge others' acts or policies that cannot be regarded as the maxim of any person or institution. They cannot rank acts in order of merit. Yet, the theory offers more precision than utilitarianism when data are scarce. One can usually tell whether one's act would use others as mere means, even when its impact on human happiness is thoroughly obscure….

Kantian ethics differs from utilitarian ethics both in its scope and in the precision with which it guides action. Every action, whether of a person or of an agency, can be assessed by utilitarian methods, provided only that information is available about all the consequences of the act. The theory has unlimited scope, but owing to lack of data, often lacks precision. Kantian ethics has a more restricted scope. Since it assesses actions by looking at the maxims of agents, it can only assess intentional acts. This means that it is most at home in assessing individuals' acts; but it can be extended to assess acts of agencies that (like corporations and governments and student unions) have decision-making procedures. It can do nothing to assess patterns of action that reflect no intention or policy, hence it cannot assess the acts of groups lacking decision-making procedures, such as the student movement, the women's movement, or the consumer movement.

It may seem a great limitation of Kantian ethics that it concentrates on intentions to the neglect of results. It might seem that all conscientious Kantians have to do is to make sure that they never intend to use others as mere means, and that they sometimes intend to foster other's ends. And, as we all know, good intentions sometimes lead to bad results and correspondingly, bad intentions sometimes do no harm, or even produce good. If Hardin is right, the good intentions of those who feed the starving lead to dreadful results in the long run. If some traditional arguments in favor of capitalism are right, the greed and selfishness of the profit motive have produced unparalleled prosperity for many.

But such discrepancies between intentions and results are the exception and not the rule. For we cannot just claim that our intentions are good and do what we will. Our intentions reflect what we expect the immediate results of our action to be. Nobody credits the "intentions" of a couple who practice neither celibacy nor contraception but still insist "we never meant to have (more) children." Conception is likely (and known to be likely) in such cases. Where people's expressed intentions ignore the normal and predictable results of what they do, we infer that (if they are not amazingly ignorant) their words do not express their true intentions. The Formula of the End in Itself applies to the intentions on which one acts--not to some prettified version that one may avow. Provided this intention--the agent's real intention--uses no other as mere means, he or she does nothing unjust. If some of his or her intentions foster others' ends, then he or she is sometimes beneficent. It is therefore possible for people to test their proposals by Kantian arguments even when they lack the comprehensive causal knowledge that utilitarianism requires. Conscientious Kantians can work out whether they will be doing wrong by some act even though they know that their foresight is limited and that they may cause some harm or fail to cause some benefit. But they will not cause harms that they can foresee without this being reflected in their intentions.

From the differing implications that Kantian and utilitarian moral theories have for our actions towards those who do or may suffer famine, we can discover two sharply contrasting views of the value of human life. Utilitarians value happiness and the absence or reduction of misery. As a utilitarian one ought (if conscientious) to devote one's life to achieving the best possible balance of happiness over misery. If one's life plan remains in doubt, this will be because the means to this end are often unclear. But whenever the causal tendency of acts is clear, utilitarians will be able to discern the acts they should successively do in order to improve the world's balance of happiness over unhappiness.

This task is not one for the faint-hearted. First, it is dauntingly long, indeed interminable. Second, it may at times require the sacrifice of happiness, and even of lives, for the sake of a greater happiness. Such sacrifice may be morally required not only when the person whose happiness or even whose life is at stake volunteers to make the sacrifice. It may be necessary to sacrifice some lives for the sake of others. As our control over the means of ending and preserving human life has increased, analogous dilemmas have arisen in many areas for utilitarians. Should life be preserved at the cost of pain when modern medicine makes this possible? Should life be preserved without hope of consciousness? Should triage policies, because they may maximize the number of survivors, be used to determine who should be left to starve? Should population growth be fostered wherever it will increase the total of human happiness-or on some views so long as average happiness is not reduced? All these questions can be fitted into utilitarian frameworks and answered if we have the relevant information. And sometimes the answer will be that human happiness demands the sacrifice of lives, including the sacrifice of unwilling lives. Further, for most utilitarians, it makes no difference if the unwilling sacrifices involve acts of injustice to those whose lives are to be lost. It might, for example, prove necessary for maximal happiness that some persons have their allotted rations, or their hard-earned income, diverted for others' benefit. Or it might turn out that some generations must sacrifice comforts or liberties and even lives to rear "the fabric of felicity" for their successors. Utilitarians do not deny these possibilities, though the imprecision of our knowledge of consequences often blurs the implications of the theory. If we peer through the blur, we see that the utilitarian view is that lives may indeed be sacrificed for the sake of a greater good even when the persons are not willing. There is nothing wrong with using another as a mere means provided that the end for which the person is so used is a happier result than could have been achieved any other way, taking into account the misery the means have caused. In utilitarian thought persons are not ends in themselves. Their special moral status derives from their being means to the production of happiness. Human life has therefore a high though derivative value, and one life may be taken for the sake of greater happiness in other lives, or for ending of misery in that life. Nor is there any deep difference between ending a life for the sake of others' happiness by not helping (e.g., by triaging) and doing so by harming. Because the distinction between justice and beneficence is not sharply made within utilitarianism, it is not possible to say that triaging is a matter of not benefiting, while other interventions are a matter of injustice.

Utilitarian moral theory has then a rather paradoxical view of the value of human life. Living, conscious humans are (along with other sentient beings) necessary for the existence of everything utilitarians value. But it is not their being alive but the state of their consciousness that is of value. Hence, the best results may require certain lives to be lost--by whatever means--for the sake of the total happiness and absence of misery that can be produced.

Kantians reach different conclusions about human life. Human life is valuable because humans (and conceivably other beings, e.g., angels or apes) are the bearers of rational life. Humans are able to choose and to plan. This capacity and its exercise are of such value that they ought not to be sacrificed for anything of lesser value. Therefore, no one rational or autonomous creature should be treated as mere means for the enjoyment or even the happiness of another. We may in Kant's view justifiably--even nobly--risk or sacrifice our lives for others. For in doing so we follow our own maxim and nobody uses us as mere means. But no others may use either our lives or our bodies for a scheme that they have either coerced or deceived us into joining. For in doing so they would fail to treat us as rational beings; they would use us as mere means and not as ends in ourselves.

It is conceivable that a society of Kantians, all of whom took pains to use no other as mere means, would end up with less happiness or with fewer persons alive than would some societies of complying utilitarians. For since the Kantians would be strictly bound only to justice, they might without wrongdoing be quite selective in their beneficence and fail to maximize either survival rates or happiness, or even to achieve as much of either as a strenuous group of utilitarians, who somehow make the right calculations. On the other hand, nobody will have been made an instrument of others' survival or happiness in the society of complying Kantians.