Chapter 15: Ends and Means

1. How Means Become Ends

All men act. They act purposefully. They employ means to achieve ends. This may seem elementary. Yet there has been no more fertile source of confusion in ethical philosophy than that concerning means and ends.

"Ends" may be "pluralistic," as many moral philosophers insist, but only if we recognize that this refers to subordinate or intermediate ends. Ends are never irreducibly pluralistic. In choosing between subordinate ends, as we constantly find ourselves obliged to do, we are necessarily guided by a preference of one over the other. And this preference is based on our judgment that one of these "ends" is more nearly an ultimate end for us, or at least a better means of realizing a more ultimate end, than the other.

Thus intermediate ends are at once means and ends. I am tempted to coin a new word, "means-ends," to emphasize this dual nature.

Our immediate end may always be described as a satisfaction or the removal of a dissatisfaction. Even our ultimate end may be described as the attainment of a state of affairs that suits us better than the alternatives. (1) But in achieving any end we have to use means that in turn we may come to think of as ends. A man and his wife, living in New York, may decide to take a trip to the Greek islands. They think of this as their end, though it could also be thought of as merely the means to achieve the enjoyment they expect to get out of the trip. But as they have never been abroad before, they decide that on the way they will visit London, Paris and Rome. Each of these visits then in turn becomes an end. They decide to go by boat; but this means of crossing the ocean is then also regarded as an end in itself. The man's wife, say, regards the ocean voyage as "the most enjoyable part of the trip," in which case, to her, what was originally merely a means to a more ultimate end becomes an end valued higher than the original end.

And this transformation of means into ends is illustrated in a whole life. A man not only wishes to protect himself against hunger and cold; he wishes to have a comfortable and attractive home, to marry and raise children, to send them to college. To achieve these more ulterior ends, he needs money. "Making money" then becomes both a means and a secondary end. To make money he must get a job. Getting a job is both a means and a tertiary end. Thus action and life are like a flight of steps in which each step is an end in relation to the preceding one and a means in relation to the next one.

The wise man tends to see his work, recreations, and ambitions in this dual way. He does not live wholly in the present moment. That would be to make no prudent provision for the future. He does not live wholly in the future. That would be never to enjoy the present moment. He lives in both the present and the future. He enjoys himself as he goes along, savoring life; but he also sets himself a goal or goals towards which he tries to make further progress.

The ideal balance is not easy to achieve. Our temperament or habits may lead us to err on one side or the other. One error is to think of everything merely as a means to something else; to become lost in work or duty; to be driven on, without ever savoring the fruits of past success, by a restless ambition that is never satisfied; to be, as Emerson put it, "always getting ready to live, but never living." Another error is to forget that something is primarily a means and to treat it only as an end in itself. A typical example of this perversion is the miser, constantly piling up money and working for still more, but never spending it.

2. Dewey, Kant, and Mill

The same confusions regarding means and ends that people fall into in practical life exist also in the theories of moral philosophers. An outstanding example of the tendency to blur entirely the distinction between means and ends -- to reduce all means to ends and all ends to means, to insist that nothing, even ideals, can be regarded as constant or permanent, to demand that everything must be always moving, changing, forward-gazing -- is found in John Dewey: "The end is no longer a terminus or limit to be reached. It is the active process of transforming the existent situation .... Growth itself is the only moral end." (2)

The question that immediately occurs to one is, Growth toward what? Should men grow twelve feet tall, and keep growing? Should population, overcrowding, noise, traffic jams, government power, delinquency, crime, filth, cancer, keep growing? If growth itself is the only moral end, then the growth of pain and misery is as much a moral end as the growth of happiness, and the growth of evil as much a moral end as the growth of good. The glorification of growth for growth's sake, change for change's sake, movement for movement's sake, reminds one of an old popular song, which went: "I don't know where I'm going, but I'm on my way!" Ethical values and ideals, as well as the distinction between means and ends, are dissolved and vaporized in such a philosophy.

But the opposite error, of regarding means as ultimate ends or ultimate ideals, is perhaps more frequent among traditional moral philosophers. This error is most conspicuous in a writer like Kant, whose concept of duty for duty's sake will be examined in our next chapter. But it is also found, in somewhat milder form, even in modern writers who call themselves "Ideal Utilitarians," such as Hastings Rashdall. (3) "The view that we have arrived at is that the morality of our actions is to be determined ultimately by its tendency to promote a universal end, which end consists of many ends, and in particular two -- Morality and pleasure." (4) In other places Rashdall substitutes the words Virtue and Happiness as if they were synonymous with these, and implies that "the Good" consists of these two elements.

Now if the Ultimate End consists of both Virtue and Happiness, it becomes impossible to resolve either into terms of the other. They then become not only incommensurable, but incomparable. So when we are confronted with the problem of which of two courses to adopt, one of which is conducive to more Virtue but to less Happiness, and the other of which is conducive to more Happiness and to less Virtue, or one of which will tend to increase Virtue more than Happiness and the other to increase Happiness more than Virtue, how can we decide which course to take?

Ends need not necessarily be commensurable, but they must be comparable; (5) otherwise there is no way to choose or decide between them. This is another way of saying that we cannot have "pluralistic" or heterogeneous ultimate ends. When we are confronted by two or more alleged ultimate ends, or two or more alleged "parts" of an ultimate end, neither or none of which can be reduced to the other or expressed in terms of the other, we shall do well to suspect that we are dealing merely with a confusion of thought, and that one of the two "ultimate" ends is really a means to the other.

Let us examine the confusion as it occurs in Kant. Kant is usually, and rightly, regarded as the arch antihedonist and antiutilitarian; but in one remarkable passage he assigns so important a role to happiness that he seems to teeter on the verge of eudaemonism:

"Virtue (as worthiness to be happy) is the supreme condition of all that can appear to us desirable, and consequently of all our pursuit of happiness, and is therefore the supreme good. But it does not follow that it is the whole and perfect good as the object of the desires of rational beings; this requires happiness also .... Now inasmuch as virtue and happiness together constitute the possession of the summum bonum in a person, and the distribution of happiness in exact proportion to morality (which is the worth of the person, and his worthiness to be happy) constitutes the summum bonum of a possible world; hence this summum bonum expresses the whole, the perfect good, in which, however, virtue as the condition is always the supreme good, since it has no condition above it; whereas happiness, while it is pleasant to the possessor of it, is not of itself absolutely and in all respects good, but always presupposes morally right behavior as its condition." (6)

Kant's subsequent discussion of the relationship of Virtue and Happiness is so confused that it seems unprofitable to follow it further. He concludes, among other things, that "happiness and morality are two specifically distinct elements of the summum bonum, and therefore their combination cannot be analytically cognized." (7) In the course of his argument he states but rejects the answer of "The Epicurean": "The Epicurean maintained that happiness was the whole summum bonum, and virtue only the form of the maxim for its pursuit, viz. the rational use of the means for attaining it." (8)

Yet if we interpret happiness in this context as referring not merely to the short-run happiness of the agent but to the general long-run happiness of the community, then this "Epicurean" view is obviously the correct solution. The ultimate end is happiness. Virtue is a necessary long-run means to that end.

Bertrand Russell has put the point clearly and simply: "What is called good conduct is conduct which is a means to other things which are good on their own account."

Some people will be shocked at this, because they will interpret it as a downgrading of virtue or morality to a mere means. But a necessary means to a great end is seldom regarded by us as a mere means; it becomes an (intermediate or penultimate) end in itself; it even becomes in our minds an indispensable part or ingredient of the ultimate end.

All this was clearly recognized by John Stuart Mill in his Utilitarianism:

Whatever may be the opinion of utilitarian moralists as to the original conditions by which virtue is made virtue; however they may believe (as they do) that actions and dispositions are only virtuous because they promote another end than virtue; ... they not only place virtue at the very head of the things which are good as means to the ultimate end, but they also recognize as a psychological fact the possibility of its being, to the individual, a good in itself, without looking to any end beyond it; ... as a thing desirable in itself, even although, in the individual instance, it should not produce those other desirable consequences which it tends to produce, and on account of which it is held to be virtue.

G. E. Moore later had great sport with Mill for the whole passage of which this is a part, accusing him of "glaring contradiction," and of having "broken down the distinction between means and ends." "We shall hear next," Moore went on, "that this table is really and truly the same thing as this room." Mill did lapse into some contradictions, but his discussion of the relation of means and ends was psychologically correct. There is a distinction between means and ends, indispensable for the intelligent conduct of life. But it is not an objective distinction, like that between a table and a room. The distinction between means and ends is subjective. Means and ends have meaning only in relation to human purposes and human satisfactions, and, for each individual, in relation to his purposes and his satisfactions. An object cannot be now a table and now a room, but it may very well be now a means and now an end. It can even be simultaneously a means and end, both a means and an end, an intermediate end, if we so treat it and regard it in achieving our purposes and deriving our satisfactions.

3. Virtue Is Instrumental

In short, we agree to call Virtue and Morality precisely those actions, dispositions, and rules of action that tend in the long run to promote Happiness. Actions and dispositions that tend in the long run not to promote Happiness, or to promote only pain or misery, we agree to call Vice or Immorality.

Hence when a satirical writer like Mandeville writes The Fable of the Bees, or Private Vices Made Public Benefits (1705), and argues that it is really the "vices" (i.e., the self-regarding actions of men) that, through luxurious living and extravagance, stimulate all invention, action and progress by circulating money and capital, what he is really saying is that what we call the vices we should call the virtues, and what we call the virtues we should call the vices. Mancleville was not wrong in principle (i.e., so far as the principle of the relationship of means and ends is concerned); he was wrong in his conclusion only because his economics were wrong. (Like his later disciple Keynes, he assumed that saving led only to economic stagnation and that only extravagance in consumption stimulated industry and trade.)

Whenever we are trying to discover which is means and which end, or which of two ends is ulterior, the test is simple. We have merely to ask ourselves two main questions, such as: Would it be better to have more Virtue (or Morality) in the world at the cost of less Happiness? Or would it be better to have more Happiness at the cost of less Virtue? The moment such questions are posed, it becomes obvious that, as between these two, Happiness is the ulterior end and Virtue or Morality the means.

Clarity on this point is so important that it is worth risking excessive repetition to achieve it. To recognize that some

thing is primarily a means -- in this case Virtue -- is not to deny that it has a high value also in itself. It is merely to deny that it has a value completely independent of its utility or necessity as a means. We may make the relation clear by an analogy from the world of economic value. Capital goods derive their value from the consumer goods they help to produce. The value of a plow or a tractor is derived from the value of the crops that it helps to create. The value of a shoe factory and its equipment is derived from the value of the shoes it helps to produce. If the crops or the shoes ceased to be needed, or ceased to be valued, the means that helped to produce them would also lose their value. What we call morality has tremendous value because it is an indispensable means of achieving human happiness. (Some readers may object that the phrase I have frequently been using to describe the ultimate end, "Happiness and Well-Being," really describes two ends, and that a test similar to the one I applied as between Happiness and Virtue should be applied as between Happiness and Well-Being to resolve the dualism and clarify the relationship. But when we ask: "Would it be better to have more [human] Happiness at the cost of less [human] Well-Being?"; or, "Would it be better to have more Well-Being at the cost of less Happiness?" we immediately perceive that the question cannot be meaningfully answered because we are simply dealing with synonyms that describe precisely the same thing. I have frequently been using the full phrase because this performs a double function. It emphasizes that I am using the word happiness in the broadest sense possible, to indicate not mere sensual or superficial pleasure no matter how prolonged, but to mean "everything that seems to us worth aiming at." And the full phrase emphasizes also that when I use the words "happiness" and "well-being" I am talking of precisely the same thing, and not of two different things, as Rashdall and other "Ideal Utilitarians" imagine they are). (12)

I have frequently spoken in this chapter of "ultimate ends,"by which I have meant simply ends pursued solely for their own sake and not also as means to something further. I have even occasionally spoken, as above, of "the ultimate end," using this merely as a synonym for "long-run happiness and well-being."

But in the interests of psychological realism I am perfectly willing to accept the qualification suggested by C. L. Stevenson:

"If [a writer on normative ethics] is sensitive to the plurality of ends that people habitually have in view, he will scarcely seek to exalt some one factor as the end, reducing everything to the exclusive status of means.... If he wishes general, unifying principles, he must attend not to 'the end,' and not even to 'ends,' exclusively, but rather to focal aims.... A focal aim is something valued partly as an end, perhaps, but largely as the indispensable means to a multitude of other ends. It may play a unifying role .n normative ethics; for once it is established, the value of a great many other things, being a means to it, can probably be established in their turn." (13)

That is why, though in the ethical system I am here proposing "the ultimate end" is Human Happiness, I have thought it preferable to put my emphasis on the "focal aim" -- Social Cooperation.

4. Does the End Justify the Means?

We come now to a further problem concerning the relationship of means and ends. Does the end justify the means?

Now we can answer this question affirmatively or negatively, depending upon how we interpret the terms of the question itself. Let us begin with the negative answer, because it is the one most frequently made by moral philosophers. I cannot do better than quote Aldous Huxley:

Good ends ... can be achieved only by the employment of appropriate means. The end cannot justify the means, for the simple and obvious reason that the means employed determine the nature of the ends produced.... (14)

Our personal experience and the study of history make it abundantly clear that the means whereby we try to achieve something are at least as important as the end we wish to attain. Indeed, they are even more important. For the means employed inevitably determine the nature of the result achieved; whereas, however good the end aimed at may be, its goodness is powerless to counteract the effects of the bad means we use to reach it. (15)

These quotations make clear that what people mean when they say that "the end does not justify the means" is simply that evil means cannot be justified on the argument that they are being pursued in order to achieve a "good" end. But the reason most of us accept this adage is that we do not believe that really evil means are ever necessary or that they can in fact lead to a really good end.

Let us look at the argument as it is stated by A. C. Ewing:

It is still often felt that ideal utilitarianism is not ethically satisfactory. One reason for this is because it seems to lead to the principle that "the end justifies the means," a principle commonly rejected as immoral. If the end is the greatest good possible and the means necessary to attain it include great moral evils such as deceit, injustice, gross violation of individual rights or even murder, the utilitarian will have to say that these things are morally justified, provided only their moral evil is exceeded by the goodness of the results, and this seems a downright immoral doctrine, and certainly a very dangerous one (as is shown by its applications in recent times in politics.) (16)

Ewing seems to me here to be (no doubt unconsciously) misrepresenting the position of the utilitarian, and certainly that of the rule-utilitist. The rule-utilitist would say that ordinarily "immoral" means could in a specific situation be justified, not only provided "their moral evil is exceeded by the goodness of the results," but provided these means were the only possible way to attain these good results, and provided also that these means led on net balance to greater long-run good than any other means.

This is, in fact, the answer of a rule-utilitist like John Hospers:

Sometimes the end justifies the means and sometimes it doesn't .... Even when the means involves agonizing sacrifice, the end may justify it if it can be achieved in no other way and if the end is worth it.

But when is the end worth the means? If the end is removal of war from the face of the earth and the means is the death of a few thousand human beings now, the utilitarian would say that the end is so supremely worthwhile that it justifies the means, provided that the means really involves no more evil than the statement indicates (often the evils involved in the means lead to other evils so that in the final analysis the means contains far more evil than the end does good), and provided that the end really will be achieved once this means is taken (there must be no slip), and provided that the end can be achieved by no other means that involves less evil than this one. In actual practice, the end doesn't justify the means as often as one might think because these conditions are not met. (17)

We must be very slow, in brief, to adopt means that involve evil even to secure the most desirable ends. We must tolerate, for example, even major injustices and suppressions of liberty before we resort to the certain evils of armed rebellion or revolution or civil war. And especially in today's world we must tolerate national insults and serious aggressions before we let loose the appalling disaster of nuclear war.

But the exact amount of injustice or suppression or aggression it is wise to tolerate before we resort to rebellion or war is a question that abstract ethical principles alone cannot answer. We are compelled to weigh alternatives and probabilities and to fall back upon our practical judgment in a specific situation.

It is not always a question, unfortunately, of whether "evil" means can ever lead to "good"; it is too often a question, in the actual world in which we live, of whether means generally and rightly regarded as evil may not sometimes be unavoidable to terminate or prevent a still greater evil.

We may illustrate this by answering a question raised by Ewing. "Might not a lie be justified to save an invalid from death or prevent a war?" (18) Any sensible person must admit (as against Kant, for example) that there are times, however rare, when a lie can be justified. If so, a lie in such circumstances is relatively "right." The supreme example of the folly of sanctifying the means while forgetting the end is probably found in Fichte's declaration: "I would not tell a lie to save the universe from destruction." We may continue to say (as Kant and Fichte do) that Iying is always an evil; but we may add that in some circumstances it may be necessary to avert a still greater evil. And we may say the same of resort to armed rebellion or to war. This principle is also the only possible justification for capital punishment.

In brief, our choice is sometimes forced. When we are reduced to a choice of evils, we must choose the lesser.

To sum up the central theme of this chapter: The logical distinction between ends and means is basic. To admit that men act purposively is to admit that they drive toward ends They must necessarily employ means to achieve them. Yet certain objects or activities can become ends in themselves as well as means to other ends. A man may work at a certain job not only for the money, but also because he enjoys the work itself. The primary purpose of his work is to earn money. This may therefore be said to be his "end." But he regards the money itself chiefly as a means to other ends. (19)

Thus we strive for intermediate ends that in turn become means toward still further ends. It is therefore not always possible to say precisely how much we value something "instrumentally" and how much "intrinsically." But it is always possible to be clear-headed about the distinction. Morality must be valued primarily as a means to human happiness. Because it is an indispensable means, it must be valued very highly. But its value is primarily "instrumental" or derivative, and it is only confusion of thought to hold that its value is something wholly apart from, and independent of, any contribution it may make to human happiness.

Notes

1. See Ludwig von Mises, Epistemological Problems of Economics (Van Nostrand, 1960), pp. 31-33; Theory and History (Yale University Press, 1957), p. 12 and passim.

2. Quoted by Alban G. Widgery in his additional chapter to Henry Sidgwick's Outlines of the History of Ethics (London: Macmillan, 1949), p. 327.

3. Rashdall actually coined this term to describe his own position. G. E. Moore also used it. See Rashdall's Theory of Good and Evil (Oxford University Press, 1907), I, Chap. VII, p. 217.

4. Ibid., p. 219.

5. An elaboration of this distinction will be found in Chapter 18, pp. 171-175.

6. Kant's Critique of Practical Reason, and Other Works on the Theory of Ethics, translated by T. K. Abbott (6th ea.; Longmans, Green, 1909), Book 11, Chap. 11, pp. 206-207.

7. Ibid., p. 209.

8. Ibid., p. 208.

9. Readings in Ethical Theory, selected and edited by Wilfred Sellars and John Hospers (New York: Appleton-Century-Crofts, 1952), p. 2. From a 1910 essay by Bertrand Russell.

10. Everyman's Edition, p. 44. The reader will notice the similarity of this reasoning to that of Hume regarding Justice.

11. Principia Ethica (Cambridge University Press, 1903, 1959), pp. 71-72.

12. E.g.: "Morality consists in the promotion of true human good, but a good of which pleasure is only an element."-Hastings Rashdall, The Theory of Good and Evil (Oxford University Press, 1907), 1, 217. Such a conclusion is possible only when "pleasure" is conceived in the sensual or superficial sense of the word. The whole case of the Ideal Utilitarians rests on this narrow definition.

13. Charles L. Stevenson, Ethics and Language (Yale University Press, 1944), pp. 329-330.

14. Ends and Means (Harper, 1937), p. 10.

15. Ibid., pp. 59-60.

16. Ethics (New York: Macmillan, ]953), p. 74.

17. Human Conduct (New York: Harcourt, Brace & World, 1961), p. 213.

18. Ethics, p. 74.

19. Cf. Murray N. Rothbard, Man, Economy, and State, p. 66.

This e-text is made available by the The Henry Hazlitt Foundation in cooperation with The Foundation for Economic Education. The Hazlitt Foundation is a member-supported 501(c)(3) non-profit corporation whose mission is to make the ideas of freedom more accessible. Please visit our flagship Internet service Free-Market.Net: The world's most comprehensive source for information on liberty.

© 1964 Henry Hazlitt. For permissions information, contact The Foundation for Economic Education, 30 South Broadway, Irvington-on-Hudson, NY 10533.