Utility and Rules of Morality: Kant, Mill and Hare

Tyrus Fisher

*First Place Winner of the Prestigious Bassen Prize*

"Come now, if you had to choose between boiling one baby and letting some frightful disaster befall a thousand people—or a million people, if a thousand is not enough—what would you do?" (Anscombe, 64). Elizabeth Anscombe famously wrote these words in her 1956 essay and pamphlet "Mr. Truman's Degree". Anscombe holds that some actions, regardless of their consequences are always wrong. This question of whether an agent should consider the consequences of a deed or follow inviolable rules as a guide to right action is central to contemporary ethics. In this paper I will examine the ethical systems proposed by Immanuel Kant, John Stuart Mill, and Richard Mervyn Hare. I will subject each of these theories to hypothetical situations in order to illustrate them further. Ethical dilemmas will emerge and Hare's system will show how these dilemmas might be resolved through combining elements of Kant and Mill's thought. Kant's opinion not withstanding, I mean to show that that the idea of universal maxim and utilitarian considerations are not mutually exclusive and I will present a brief reply to a common objection to universal prescriptivism.

According to Kant there are absolute truths of morality. One can arrive at these truths through rational thought. Kant gives us various formulations of his rule of the categorical imperative which is, Kant believes, wholly derived from rational thought. In contrast to Kantian ethics, the nature of morality in Mill's formulation of utilitarianism does not require an appeal to absolute ethical truths separate from situational applications. For Mill, morality is established through consideration of the utility to humankind of any given action. In R. M. Hare's universal prescriptivism morality is a human construct that is subject to the rules of logical consistency. Under universal prescriptivism, if a person would still prescribe an action were they to find themselves utterly in the situation of any of the affected agents said person is morally justified in prescribing the action.

For Kant, morality is ontologically independent of our thoughts about it. That is to say, regardless of anyone's opinions or feelings about the world there is always an objectively correct course of action in morally relevant affairs. In the following footnote from the Grundlegung Kant relates this idea in a thankfully straightforward manner:

Pure philosophy of morals (metaphysics) may be distinguished from the

applied (viz. applied to human nature) just as pure mathematics is distinguished from applied mathematics and pure logic from applied logic. By this designation one is almost immediately reminded that moral principles are not grounded on the peculiarities of human nature but must subsist a priori themselves, and that from such principles practical rules must be derivable for every rational nature, and accordingly for human nature (Kant, 22).

Kant is saying here that ethics can be divorced from particular affairs in the same way that math and logic can be separated from application to the world and still remain formally coherent. In performing this separation a person should see that morality is subject to formal rules of rationality in the same way as math and logic (which are separate for Kant). Further implied here is the idea that because rules of morality remain articulated without application to worldly affairs what we are left with must be knowledge that does not require derivation from experience and is therefore a priori. Now, because we are dealing with truths independent of experience we should not consider the empirical world in deriving further truths because this can only confuse our judgments, we need only deduce these truths from what necessarily follows from consideration of certain conjoined a priori truths. It is worth pointing out that because the "truths" of morality we are discovering are not implicit in any single concept they are not what Kant would call analytic. Because these a priori truths require the conjoining of other a priori truths to arrive at conclusions not implicit in any single concept, Kant believes moral truths are what he calls synthetic a priori knowledge. It is this type of reasoning that allows us to derive new truths from pure reasoning. The important thing to take from this citation is that, for Kant, we can arrive at moral truths absent any worldly considerations and the existence of a single rationally deducible standard of morality is, for him, certain.

According to Kant, the categorical imperative is deduced from a priori truths and so adherence to it is necessary for proper conduct, it is presented in five formulations each designed to help the moral agent commit actions in accord with what is demanded of the will by rationality. The categorical imperative is a way to judge whether or not a proposed action is rational. Bearing in mind that it is a widely held notion that any formulation of the categorical imperative is reducible to any other, I will limit my illustration of the categorical imperative to the "universal law formulation". Kant writes, "Hence there is only one categorical imperative and it is this: Act only according to that maxim whereby you can at the same time will that it should become a universal law" (Kant, 30). A "universal law" is a regulation that is inescapable, the assertion that "all matter is subject to the law of gravity" is an example of a universal law. Thus, what Kant means is that a rational agent is morally obligated to act only in that manner that he or she could rationally will all agents adhere to in a situation that shares the relevant moral concerns. Logical consistency is an obvious requirement of rationality, so to follow the categorical imperative we must consider whether we can will a possible act to be a universal law of nature without introducing rational inconsistency. If we find that we cannot do this, i.e. it is formally impossible for our hypothetical universal law to exist, or we wish to act contrary to what we would will for others, our action would be irrational and therefore immoral. For example, if stealing were a universal law the concept of personal property would dissolve and the concept of stealing would be meaningless, so stealing could not logically be a universal law, or if I will that no one steal from me, yet I steal from others my will is not universal and so is logically inconsistent. In both cases reasoning seems to make it apparent that stealing is irrational and therefore immoral. The categorical imperative is the point where Kant's meta-ethical theory touches the ground, that is, where it becomes relevant to practical application.

John Stuart Mill believes that practical considerations are all an ethical theory need appeal to. He considers it futile to attempt to justify an ethic a priori, as a result, morality for Mill does not rest on an ideal that is justified by independent principles. For Mill, morality is constructed around "the good". "The good" is that which is desired for its own sake. Mill tells us that this thing is happiness, because happiness is the only thing desirable for itself and the thing that all of our actions are aimed at. Mill, at least in part, has Kant in mind when he states the following: "Questions of ultimate ends are not amenable to direct proof. Whatever can be proved to be good must be so by being shown to be a means to something admitted to be a good without proof" (Mill, 4).

This is both a strength and a weakness for Mill's system, on the one hand we can appeal to practical considerations to justify the notion of happiness as "the good", because it is true that we often desire states of affairs that bring happiness; and we do label these states of affairs "good". Unfortunately this means Mill's utilitarianism already presupposes an ethic and so embraces a circularity as a foundation. Nevertheless, for Mill, if an ideal is intrinsically good it is impossible to prove its goodness by pointing to extrinsic notions because this would indicate that the ideal is good only as far as it is a means for achieving a further end. Mill wishes us to observe that by nature, an ultimate end may not appeal to a further end.

In Mill's utilitarianism an action's moral worth is determined by the consequences of that action. In particular, an action is good insofar as it produces the greatest amount of happiness for the greatest number of people. Mills presents this idea in the following:

The creed that accepts as the foundations of morals ‘utility' or the ‘greatest happiness principle' holds that actions are right in proportion as they tend to promote happiness; wrong as they tend to produce the reverse of happiness. By happiness is intended pleasure and the absence of pain; by unhappiness, pain and the privation of pleasure (Mill, 7).

Mill tells us here that utilitarians are those people who subscribe to the view that deeds are right to the degree that they increase the net happiness in the world and wrong as they do the reverse. Implicit in this statement is the idea that actions themselves are morally relevant, not an agent's motive. Mill also defines happiness for us as pleasure and the absence of pain and unhappiness as pain and the lack of pleasure. I would further note that, since the "net happiness" is the ultimate judge of an action's worth, for an act to be good it need not avoid causing any pain whatsoever it must merely cause more pleasure than pain. Having attended in brief to Kant and Mill's conception of morality I will move on to universal prescriptivism and then explore some hypothetical applications of these systems.

Because Richard Mervyn Hare's universal prescriptivism combines elements of Kantian maxim and classical utilitarianism, the system can be seen as an attempt to resolve the difficulties unique to each while combining their strengths. For Hare it is up to humankind to create and define right and wrong but the demands of logical consistency will prevent a proper moral theory from collapsing into relativism (Hare, Essays, 95). To use universal prescriptivism a person creates universalized maxims for a specific situation and takes into account the consequences of an action with regard to the preferences of the affected parties. Right actions for Hare are actions that a person would universally will any agent prescribe after considering the preferences of all involved. Hare expresses this in the following, "It is in the endeavour to find lines of conduct which we can prescribe universally in a given situation that we find ourselves bound to give equal weight to the desires of all parties . . . and this, in turn leads to such views as that we should seek to maximise satisfactions (Hare, Freedom, 123). So, according to Hare, in seeking the "ought to do" of a situation we are looking for those rules of operation that we would will universally. Further, it is this goal that leads us to preference utilitarianism as a guide to right conduct. What Hare has given us is a theory of preference utilitarianism in the spirit of Kantian maxim. Under universal prescriptivism we first universalize a proposition, that is, we consider the preferences of all involved in all situations relevant to the act and then consider the consequences of our hypothetical prescription. The hypothetical action that produces the best results, yet is acceptable to any rational agent involved is the best course of action. I will now attend to further illustration of the systems of Kant, Mill, and Hare by subjecting them to a hypothetical situation and its emerging dilemmas.

Let us suppose, as all philosophy students do, that my life is in immanent danger. I am running through the woods miles from help with a would-be axe murderer quickly gaining on me when I come to a cabin. No one is home in this cabin but I come across an automobile with the keys in the ignition. The problem as I perceive it, while I stare at the keys and wait for the blood thirsty lumberjack to catch up, is whether or not it is morally acceptable to steal this car to ensure my flight from mortal danger. Good Kantian that I am, I recall the categorical imperative, "act only according to that maxim whereby you can at the same time will that it should become a universal law". I can't bring myself to will that stealing become a universal law because I know I don't want my stuff stolen so my proposed maxim introduces a contradiction of the will, contradictions are necessarily irrational and so my proposed solution is wrong according to Kant. Bound to moral duty, I turn away from the car just in time to for an axe blade to issue a glancing blow to my head. I retain consciousness just long enough to notice a car approaching. As I sleep peacefully it occurs to me that I no longer want to abide by the categorical imperative. The categorical imperative just seems too rigid and inappropriate for many situations. I recall the "classical utilitarians" and it occurs to me that, were I to have considered the principle of utility the dilemma of whether or not to drive the stolen car to safety would have evaporated. For if I considered the net gain in happiness, as defined by Mill, it would have been clear that the privation of the great pain I will be in for some time to come, would have outweighed the disappointment of the killer and since I could have returned the car to its rightful owner the preservation of my well-being would clearly outweigh all the negatives. The next day I wake up in a hospital bed where I am informed that the owner of the cabin pulled up just as I was being attacked, scaring the would-be assassin away.

I am in pain but able to think clearly enough to resolve to be an act utilitarian. Because Kantian maxim failed me I have decided that I will judge the moral worth of every action based on the utility of the action. My smug state of mind is short-lived however because almost immediately following these thoughts my doctor enters the room and informs me that while I was unconscious they performed a CT scan of my brain and found a large tumor that surely would have killed me had they not removed it on the spot. Happy as I am, and grateful to the doctor for his action that has saved my life, it occurs to me that as an act utilitarian I should consider the action of the axe wielder morally correct. After all, it was the injury inflicted by him that allowed my life threatening tumor to be discovered. I decide I must reject utilitarianism because judging an act morally right wholly based on effects regardless of the motivation behind the act is unacceptably opposed to my notion of right and wrong. However, if I revert to Kantian deontology my most recent dilemma will dissolve but the previous difficulty will emerge again. So it seems, I cannot embrace Kantian deontology because it demands too much and doesn't allow consideration of consequences. And I must likewise reject utilitarianism because I feel the will of the agent affects the worth of the action. Both systems appear unacceptably counter to my moral intuitions. I go to sleep in my hospital bed still troubled by my inability to resolve these ethical questions.

As we have seen, serious objections to Kantian maxim and utilitarianism can be raised. Universal prescriptivism attempts to fuse the strengths of these systems while eliminating their problems. I will now apply universal prescriptivism to the previous thought experiment and consider what it yields. When I find myself gazing at the automobile that could allow my escape from harm, universal prescriptivism requires that I formulate a maxim, reflect on the position of all those involved, and then consider whether I would still prescribe my proposed action. But what does "universal maxim" mean? In Hare's words, "universality must not be confused with generality… My moral principles do not have to be as general as ‘never tell lies' they can be more specific, like never tell lies except when it is necessary to save an innocent life, and except when …, and except when…(Hare, "Universal", 457). So for Hare, universality applies to the affected moral agents not the situation. Accordingly, step one for considering whether I should take the car to escape would be formulation of a rule. This rule may be as specific as necessary as long as it accurately represents the situation, consider the following rule, "When Tyrus is being chased by Joe the killer and Tyrus has the opportunity to steal Nancy the forest dweller's car, Tyrus should steal the car to secure his escape from bodily harm". Step two will be to universalize the rule, "When one is being chased by another who has the intent to kill him or her and the fleeing person has the opportunity to steal another's car to secure his or her escape from immanent bodily harm one should do so". Step 3 is to consider the preferences of the fleeing person and the owner of the car. I should not ignore consideration of the axe maniac but using universal prescriptivism it is a relatively simple matter to determine that escaping from a murderer in this situation is a morally acceptable action, so I need not consider the killer's preference presently, further, if the would be killer succeeds in catching me his punishment may be such that it is in his best interest not to succeed. Now, if I consider whether I would allow someone to take my car if they were being chased by an axe murderer and I judge in the affirmative, my proposed action has passed the first test. Next, I consider whether I find the idea of stealing the car to secure my own safety acceptable. From both perspectives I cannot imagine being opposed to my proposed action and so I judge stealing a car in the situation outlined above is not immoral. I have come to this conclusion by formulating a universal maxim in some ways similar to what Kant would have us do but my conclusion is certainly opposed to anything we could derive under Kant's system.

But what of our later dilemma regarding how to judge the morality of the axe wielder's action that led to the discovery and removal of a brain tumor in my head? For this quandary a possible universal maxim could be, "Whenever one attempts to kill another out of anger it is morally right to do so as long as they recover completely and the attempt on the person's life somehow prevents untimely death of the victim." Again, an act utilitarian would have to agree with this statement. And indeed, if this maxim could somehow be applied to real life the outcome would be positive in each case. But universal prescriptivism will quickly reject the potential maxim because prescriptivism allows consideration of motive and I maintain that it is always wrong for someone to try to kill me out of anger. Also, because the desire of the killer is to harm someone out of anger, and if an axe wielder knew that his or her attempt on someone's life would have a positive outcome, the motivation would no longer be anger, but instead the other person's welfare. The maxim must also be rejected because of its paradoxical nature. We can conclude therefore, that the action was morally unacceptable. This is a judgment that, although utility has been considered, gives us a result conflicting with utilitarianism. So we have two instances in which, following the tenets of universal prescriptivism, the principle of utility and ethical maxims have been employed together to arrive at conclusions in accord with commonsense morality.

Universal Prescriptivism is not immune to criticisms however. For instance, a person may object that the preferences of those potentially affected by a maxim may be contrary to one another. For instance, if one's girlfriend is sad and kissing her will cheer her up. I may come up with the universalized maxim: "When someone is sad one should kiss that someone to cheer that someone up". However, if Frank is sad and Bertha kisses him Frank may feel worse. The problem here is not with universalization though; it is the failure to acknowledge that "different desires make the situation different" (Hare, "Universal", 456). Universal Prescriptivism acknowledges that if the situation is different so must be the maxim, in this case with regard to the preferences of the agents involved. Another objection, arguably the most fundamental and damaging, to universal prescriptivism is revealed if one questions why logical consistency, with regard to consideration of the preference of others, should be required of a proper moral theory. More concretely, just because I value my own preferences why should I value yours? Maybe your preferences are morally absurd and to consider them when attempting to arrive at an "ought to do" is wrong? Consider the would-be axe-murderer again and imagine that that this axe murderer is a fanatical neo-Nazi who believes he is justified, even obligated, to harm members of certain ethnic groups. Now, if the Nazi's fanaticism is so pervasive that he would wish himself harmed were he to discover that he was a member of any of a number of ethnic groups, universal prescriptivism can not deem his actions morally unacceptable (Hare, Freedom, 192).

I will attempt to reply to this objection succinctly but it is a serious objection so I will treat it in two parts. The assertions my reply rests on are as follows: First, logical consistency, on some level, is required of any theory that attempts to guide action. Any theory of applied ethics must rely on logical consistency or the theory simply cannot guide actions and is nonsensical. For instance if I arbitrarily decree that my actions have moral worth only if I am wearing green socks. It necessarily follows that if I have red socks on my actions are not moral. If someone accepts either of the preceding tenets, the only way it can be meaningful is if any contrary action is judged immoral, or at least amoral. So we see, all ethical theories require consistency to be meaningful. Secondly, the neo-Nazi's belief that two people deserve different treatment based on their ethnicity is itself the result of a logical inconsistency, were this inconsistency resolved the present dilemma would evaporate. Recall that universal prescriptivism is a theory that attempts to prescribe what we ought to do given the facts relevant to a situation. The hypothetical prescriptivist Nazi is ignoring, or is ignorant of, a situationally relevant fact about people. This fact is, in and of itself, extra-moral and so is not explicitly laid out by Hare's prescriptivism; but in any situation that a fact is relevant, universal prescriptivism requires it be considered. The fact I am alluding to is that the bulk of biological and anthropological information about people shows there are no differences between any two ethnic groups that warrant differing moral standards for interacting with moral agents of other ethnicities. Because of the relevant physical facts, it is clear that it is logically inconsistent to confer worth or value on one individual yet deny it to another based on ethnicity because the empirical evidence shows ethnicity to be irrelevant. And so the fanatic Nazi dilemma dissolves. Because of these considerations universal prescriptivism is justified in demanding logical consistency with regard to preferences and the Nazi dilemma is disposed of.

I have roughly sketched the systems of Kantian deontology, utilitarianism and universal prescriptivism. I have briefly considered the character of morality and what constitutes a moral action for each of these methods. In considering possible applications of utilitarianism and the categorical imperative practical difficulties were shown to emerge. Finally, I have presented R.M Hare's universal prescriptivism as a practically superior construct that plainly draws on both the categorical imperative and the principle of utility. In showing how universal prescriptivism might successfully resolve some ethical quandaries I have illustrated one way that the ideas Mill and Kant work well together.

Works Cited

  • Anscombe, G.E. M. "Mr. Truman's Degree". Ethics Religion and Politics: Collected Philosophical Papers, vol. III. Minneapolis: University of Minnesota Press, 1981.
  • Hare, Richard Mervyn. Essays on Political Philosophy. New York: Oxford University Press, 1998.
  • Hare, Richard Mervyn. Freedom and Reason. New York: Oxford University Press, 1977.
  • Hare, Richard Mervyn. "Universal Prescriptivism". A Companion to Ethics. Ed. Peter Singer. Boston: Blackwell Publishers, 1991, 451-463.
  • Kant, Immanuel. Grounding for the Metaphysics of Morals and On a Supposed Right to lie because of Philanthropic Concerns. Translated by James W. Ellington. Indianapolis: Hackett Publishing Company, 1993.
  • Mill, John Stuart. Utilitarianism and the 1868 Speech on Capital Punishment. 2nd ed. Ed. George Sherr. Indianapolis:
  • Hackett Publishing Company, 2001.