Mill’s Utilitarianism: Collected Works vol. X

Source: Editor's introduction to The Collected Works of John Stuart Mill, Volume X - Essays on Ethics, Religion, and Society, ed. John M. Robson, Introduction by F.E.L. Priestley (Toronto: University of Toronto Press, London: Routledge and Kegan Paul, 1985).

Copyright Statement: The online edition of the Collected Works is published under licence from the copyright holder, The University of Toronto Press. ©2006 The University of Toronto Press. All rights reserved. No part of this material may be reproduced in any form or medium without the permission of The University of Toronto Press.

Mill’s Utilitarianism by F.E.L. Priestley

the majority of serious students of ethics today are utilitarians, and those who are not see utilitarianism as the chief position in need of amendment. John Stuart Mill’s writings on ethics, and especially on utilitarianism, are thus of vital contemporary interest and importance. More than any other thinker, Mill is responsible for laying down the principal directions ethics has taken since his day. He did not, however, embody his full views in any single volume or one set of writings, and the main lines in ethics which he sketched were worked out in detail only after his death by Henry Sidgwick. A generation later, G. E. Moore sought to refine upon Sidgwick’s results, and subsequent ethical theory has taken Moore’s work as its starting point.

The most complete guide to undertaking a detailed examination of Mill’s ethical views is his Utilitarianism, and so I have used it as the basis of this introductory essay. His other essays on ethics are valuable as supplements to the opinions he puts forward in this work, and they are referred to where appropriate. Five main topics have been selected for detailed treatment in the discussion that follows. The first section sorts out some of Mill’s more important principles. Section II examines his dictum that the sole evidence that anything is desirable is that people desire it. In the third, consideration is given to what Mill holds that this evidence discloses. Section IV deals with Mill’s analysis of moral concepts. The discussion concludes with an examination of his views on the use of the principle of utility.

I.

THE PRINCIPLE OF UTILITY

Mill writes, “happiness is the sole end of human action, and the promotion of it the test by which to judge of all human conduct. . . .”1 He also makes it clear that the test is its promotion of happiness “to the greatest extent possible” (214). By such conduct Mill does not mean that which would promote happiness to the greatest extent conceivable, but that which would promote it to a greater extent than would any alternative. Mill also makes it clear that when he speaks of the promotion of happiness as “the test by which to judge of all human conduct,” the aspect of conduct of which he means that it is a test is whether it should be done.2 He thus holds that the test of whether something should be done is whether it would promote more happiness than would any alternative to it. Mill implies that if an action would satisfy this test, it should be done, and that if it would not, it is not one that should be done. Accordingly, the main principle which Mill maintains is that something should be done if and only if it would cause more happiness than would any alternative, and that something should not be done if and only if it would fail to cause as much happiness as would some alternative.

The chief support Mill offers for this principle is that “happiness is desirable, and the only thing desirable, as an end. . .” (234). He distinguishes things desirable as a means and things desirable for their own sake. What is desirable for its own sake he speaks of as desirable as an end. He argues that it is because happiness is the only thing desirable for its own sake that the test of conduct generally is its promotion of happiness. The principle he employs in taking this step is that if there is one sort of thing which is alone desirable for its own sake, then the promotion of it is the test of all human conduct. By test of human conduct he means test of what should be done. An action is then one that should be done if and only if it satisfies this test. Mill thus takes it for granted that something should be done if and only if its consequences would be more desirable than would those of any alternative to it.

From his main principle in turn Mill draws a conclusion about what it would be right to do and what it would be wrong to do. The question of whether it would be right or wrong to do a certain action is a question about its morality. Mill writes, “the morality of an individual action is . . . a question . . . of the application of a law to an individual case” (206). He thus holds that it would be wrong to do a certain action only if it would be at variance with a certain rule. If we ask what sort of rule he is referring to, Mill makes it clear that he means a rule that should generally be observed. By his main principle Mill has already given a general answer as to what should be done. In accordance with it he holds that a certain rule is one that should generally be observed if and only if its general observance would cause more happiness than would any alternative to its general observance.3 Mill thus maintains that it would be wrong to do a certain action only if it would be at variance with such a rule.4

Some prolixity is required to clarify what Mill understands by an action that would cause more happiness than any alternative to it.5 The only respect in which an action is thereby compared to its alternatives is its consequences, and the only consequences by which it is compared are those consisting of happiness and unhappiness. Mill writes, “By happiness is intended pleasure, and the absence of pain; by unhappiness, pain, and the privation of pleasure” (210). He states: “Of . . . philosophers who have taught that happiness is the end of life . . . [the] happiness which they meant was not a life of rapture; but moments of such, in an existence made up of few and transitory pains, many and various pleasures. . .” (215). Hence the only consequences of an action that are relevant are pleasures and pains. All the pleasures and pains among the consequences of an action are relevant, whether remote or near, whether experienced by humans or by other sentient creatures.6

If Mill held that the only relevant difference among pleasures and pains was whether one was greater than another, there would be only six possibilities for the total effects of an action. They would contain (1) an excess of pleasure over pain, (2) an excess of pain over pleasure, (3) an excess of neither, (4) pleasure and no pain, (5) pain and no pleasure, (6) neither pleasure nor pain. Mill argues, however, that pleasures and pains differ in a further respect which is relevant—some are more desirable than others.7 Accordingly, eight possibilities may be distinguished with regard to the total effects of an action:

(1) They contain some pleasures and no pains.

(2) They contain both pleasures and pains, and regardless of whether there is an excess of pleasure over pain, the pleasures are on the whole more desirable than the pains are undesirable.

(3) They contain both pleasures and pains; neither the pleasures nor pains are of sorts such that the pleasures on the whole are more desirable than the pains are undesirable or such that the pains on the whole are more undesirable than the pleasures are desirable; but there is an excess of pleasure over pain.

(4) They contain some pains and no pleasures.

(5) They contain both pleasures and pains, and regardless of whether there is an excess of pain over pleasure, the pains are on the whole more undesirable than the pleasures are desirable.

(6) They contain both pleasures and pains; neither the pleasures nor pains are of sorts such that the pleasures on the whole are more desirable than the pains are undesirable or such that the pains on the whole are more undesirable than the pleasures are desirable; but there is an excess of pain over pleasure.

(7) They contain no pleasures or pains.

(8) They contain both pleasures and pains, and regardless of whether there is an excess of pleasure over pain, of pain over pleasure, or an excess of neither, the pleasures and pains they contain are of sorts such that the pleasures on the whole are not more desirable than the pains are undesirable and such that the pains on the whole are not more undesirable than the pleasures are desirable.

If (1) or (2) or (3) holds of a certain action, Mill would classify it as one that would cause an excess of happiness over unhappiness. If (4) or (5) or (6) holds, he would classify it as one that would cause an excess of unhappiness over happiness. If one of the other alternatives holds, he would classify an action as one that would cause an excess of neither.

Having distinguished the possibilities for any action, taken by itself, we may notice how any two actions taken at random may stand to one another in these respects. Since there are three possibilities for each, there are nine possible combinations. Call one action A and the other B. (1) Both A and B would cause an excess of happiness. (2) A would cause an excess of happiness but B would cause an excess of neither. (3) A would cause an excess of happiness but B would cause an excess of unhappiness. (4) A would cause an excess of neither but B would cause an excess of happiness. (5) Both would cause an excess of neither. (6) A would cause an excess of neither but B would cause an excess of unhappiness. (7) A would cause an excess of unhappiness but B would cause an excess of happiness. (8) A would cause an excess of unhappiness but B would cause an excess of neither. (9) Both would cause an excess of unhappiness. Within (9) three possibilities are to be distinguished: (9.1) B would cause a greater excess of unhappiness. (9.2) Neither would cause a greater excess of unhappiness. (9.3) A would cause a greater excess of unhappiness. Also, within (1), that is, where both A and B would cause an excess of happiness, three possibilities are to be distinguished: (1.1) A would cause a greater excess of happiness. (1.2) Neither would cause a greater excess of happiness. (1.3) B would cause a greater excess of happiness. There are thus thirteen ways in which any two actions may stand to one another. These thirteen ways may be grouped into three. If (1.1), (2), (3), (6) or (9.1) obtains, Mill would say that A would cause more happiness than B or that B would cause less than A. If (1.3), (4), (7), (8) or (9.3) obtains, he would say that B would cause more happiness than A or that A would cause less than B. If any of the three remaining combinations obtains, he would say that either would cause as much happiness as the other.

We have noticed three ways in which Mill would hold that any two actions taken at random could stand to one another. If any set of two or more actions is considered, we may notice three ways in which one of the actions of the set might stand to the others: (1) it would cause more happiness than any of the others, (2) it would cause less happiness than some of the others, (3) it would cause as much happiness as any of the others. The only sort of set of two or more actions to which Mill directs attention is that made up of a certain action and of the alternatives to it. This set includes whatever an agent would succeed in doing upon a given occasion if he tried hard enough, and excludes whatever he would not succeed in doing no matter how hard he tried. Accordingly, Mill would distinguish three ways in which an action may stand to the alternatives to it: (1) it would cause more happiness than any alternative, (2) it would cause less happiness than some alternative, (3) it would cause as much happiness as any alternative.

So far attention has been paid to one set of features of which Mill’s main principle makes mention, apart from their role in it. There is a second set of features of actions which this principle mentions—whether it is one that should be done or one that should not. What Mill’s main principle asserts is a relation between features of the first set and features of the second. It asserts that something should be done if and only if it would cause more happiness than any alternative; that something should not be done if and only if it would cause less happiness than some alternative; and that a certain action is not one that should not be done if and only if it would cause as much happiness as any alternative.

By his main principle Mill thus declares that a certain feature is a universal and peculiar feature of actions that should be done, and that a certain other feature is a universal and peculiar feature of actions that should not be done. It implies that whenever anyone judges that a certain action should be done, this is a condition that must be fulfilled for the judgment to be true. This is the case whether the judgment is about a past or future action, an actual or possible action, something done by oneself or another, or something done by an individual, a nation, or any group. Mill’s principle does not, however, imply that the only way by which anyone can know whether a certain action should be done is by seeking to make out whether it would cause more happiness than any alternative. Although Mill speaks of it as the “sole criterion,” his principle is quite compatible with using many other tests. It is compatible with using now one test and now another. Nor does Mill’s principle imply that it affords the only universal test by which to judge what should be done. All that it does imply is that whatever other test be used, it must yield results compatible with this principle. Mill’s principle does not supply the only test; it only lays down a condition to which any test must comply.

Although Mill’s principle sets forth a universal and peculiar feature of actions that should be done, there is nothing about it which implies that this is the only universal and peculiar feature of such actions. It would be compatible with it to maintain, for instance, that something should be done if and only if it is commanded by God. Mill’s principle provides nothing that rules this out. Indeed, it is conceivable that there are ten thousand other universal and peculiar features of actions that should be done. One consequence which Mill draws from his principle is that it would be wrong to do a certain action only if it would violate a rule the general observance of which would cause more happiness. Many would agree with Mill in this. They would agree that whenever anyone does what is wrong, he is violating a rule the general observance of which would in fact cause more happiness. But they would not hold that this is the reason it would be wrong to do it. They would hold that the reason it is wrong to do any action is that it violates God’s law. They would urge that God wants his creatures to be happy and that because of this whoever disobeys God’s laws violates a law the general observance of which would cause more happiness. They would agree with Mill that by doing what is wrong someone violates a rule the general observance of which would cause more happiness. But they would say that it is not because of this that someone is doing wrong; it is rather because he breaks a rule laid down by God.

There is nothing in this view incompatible with what we have so far seen of Mill’s main principle. When we notice how Mill deals with such a view, we find that he takes a further step. He holds not merely that someone does what is wrong only if he breaks a rule the general observance of which would cause more happiness, but also that what he does is wrong because it violates such a rule. Mill maintains not merely that those rules which should generally be observed would in fact cause more happiness, but also that it is because their general observance would cause more happiness that they should be observed. He does not thereby deny that by violating rules that should generally be observed, someone is disobeying God’s will. But he holds that the reason why a rule should be generally observed is not because it is prescribed by God but because its observance would cause more happiness.8

There is a further implication differentiating Mill from the view we have been considering. Those who maintain that the reason why a certain action is wrong is that it violates a rule laid down by God are committed to holding that if God should will something other than the happiness of his creatures, then an action would be wrong even though it would not violate a rule whose general observance would cause more happiness. Anyone who holds that an action is wrong because it violates a rule laid down by God is committed to holding that if there is no god or if he lays down no rules for men, then there is nothing which it would be wrong to do or wrong not to do. Mill not only holds that an action is wrong if it violates a rule the general observance of which would cause more happiness, he also contends that it is because it violates such a rule that an action is wrong. He thereby implies that even if God should will something other than the happiness of his creatures, or even if there is no god, an action would be wrong if it were to violate a rule the general observance of which would cause more happiness.9 In the first step, Mill asserts that a certain feature is a universal and peculiar feature of actions that should be done. In the second step, he states that it is because they have this feature that actions should be done.

There is nothing incompatible between Mill’s principle and the view that something should be done if and only if it would bring about a greater realization of men’s capacities than would any alternative. But his principle is incompatible with the view that something should be done because it would have this result. Similarly, Mill’s principle is not incompatible with the view that something should be done if and only if it would bring about a greater fulfilment of human wants than would any alternative. But it is incompatible with the view that something should be done because it would have this result. One alternative to Mill’s principle is the view that something should be done because it would maximize human happiness. Another alternative to it is that something should be done because it would maximize the agent’s happiness. The former is the humanistic variant to Mill’s principle; the latter the egoistic variant to it. In contrast to both, Mill’s principle is the universalistic variant. Many other alternatives to Mill’s principle are conceivable. One view already noted is that which maintains that something should be done because it would maximize fulfilment of human wants. The universalistic variant to this view is that something should be done because it would maximize fulfilment of wants generally. The egoistic variant is that something should be done because it would maximize fulfilment of the agent’s wants. The theistic variant to this is that something should be done because it would maximize fulfilment of God’s wants. Still another alternative is the view that something should be done because it would maximize the fulfilment of human capacities. Two further conceivable views are the egoistic and universalistic variants of this.

All such views differ from Mill’s principle in but one respect. They all agree that there is some feature which not only holds of every action that should be done and only of such, but which also constitutes the reason why it should be done. They all agree that this feature consists in a respect in which an action compares with its alternatives. They are also all agreed that this feature consists in how an action’s consequences would compare with those of its alternatives. These several views differ from each other and from Mill’s principle only in the sorts of consequences which they specify and the sorts of beings to whom they accrue.

The chief support that Mill offers for his main principle, to vindicate it against such other views, is that happiness is the only thing desirable for its own sake. From this contention it does indeed follow that an action would have more desirable consequences than any alternative if and only if it would cause more happiness. But this contention does not by itself support his main principle. It does so only if a further premise is added, namely, that something should be done if and only if it would have more desirable consequences than any alternative. Mill does not explicitly avow this further premise. Yet, since he holds that the contention which he offers in support of his main principle does in fact support it, he may be presumed to take this premise for granted as not requiring any attention or defence. It then looks as if Mill contends that something should be done because it would cause more happiness, but that it is not only because of this that it should be done; that the reason in turn why what would cause more happiness should be done is that happiness is the only thing desirable for its own sake.

One can at most speculate as to how Mill would meet this challenge. He might retort that the fact that an action would have more desirable consequences than any alternative could not be the ultimate reason why it should be done, since the ultimate reason why something should be done must consist in some other fact about it than the fact that it should be done, but in saying that an action would have more desirable consequences than any alternative, nothing more nor less is then said than that it should be done. Although it is not transparently evident that these are but two ways of saying the same thing, it is far from implausible to urge that by analysis they amount to the same. Two steps are involved in the analysis: (1) something should be done if and only if it would on the whole be more desirable for it to be done than any alternative; (2) it would on the whole be more desirable for something to be done than any alternative to it if and only if what would come of its being done would be more desirable than what would come of any alternative to it. If each of these is analytically true, nothing further is required.

In behalf of the first step, the following may be urged. Whenever it is said that something should be done it is implied that it is capable of being done. It is also implied that it is capable of not being done, that is, that some alternatives are capable of being done in its stead. When it is said that something should be done, it is not only implied that it is one of a number of alternatives; it is also implied that it stands in a certain relation to the others. When it is said that something should be done, it is not implied that it would be more desirable for some alternative to it to be done; nor is it implied that it would be as desirable for some alternative to be done in its stead. What is rather implied is the denial of both these implications. When it is said that something should be done, it is thus implied that it would on the whole be more desirable for it rather than any alternative to be done.10 In behalf of the second step the following may be urged. It cannot be denied that an action may have consequences, and that whether it would be desirable for it to be done is affected by what would come of its being done. Nor can it be denied that the desirability of some alternative being done is affected by the desirability of what would come of it. It is then more desirable on the whole that one alternative rather than another be done if and only if what would come of the first would be more desirable than what would come of the other. Hence it would on the whole be more desirable for something to be done rather than any alternative if and only if what would come of it would be more desirable than what would come of any alternative.

The chief premise that Mill offers in support of his main principle is that happiness is the only thing desirable for its own sake. This premise affords support only in conjunction with the added premise, that something should be done if and only if it would have more desirable consequences than any alternative. Consequently, Mill’s contention that happiness is the only thing desirable for its own sake cannot support his main principle against any sort of ethical theory which rejects the second premise. Against any such theory he seeks to vindicate his main principle by clearing up the relation of the conception of a wrong action and of an action which there is an obligation not to do to that of an action that should not be done.11 On the other hand, any sort of ethical theory that rejects Mill’s main principle but which holds that whether something should be done turns on how its consequences would compare with those of any alternative to it need not be incompatible with the second premise. To vindicate his main principle against any theory of that sort, it is sufficient for Mill to make good his contention that happiness is the only thing desirable for its own sake.12

Before we go on to examine how Mill seeks to make good this contention, certain implications of it may be noted. It implies that if A are the consequences of one action, X, and B the consequences of another action, Y, A would be more desirable for their own sake than B if and only if they would contain more happiness. It implies that if A should be the consequences of some other action than X, they would still be more desirable for their own sake than B. It thus implies that whether the consequences of an action are more desirable for their own sake than those of another does not depend on what action they are the consequences of. Mill’s contention also implies that if A are the consequences of a natural occurrence and B the consequences of another natural occurrence, A would still be more desirable for their own sake than B. This thus means that whether one set of consequences is more desirable for its own sake than another does not depend on what caused them. It does not depend on A or B being a set of consequences. Mill’s contention that happiness is the only thing desirable for its own sake has therefore a wider scope than his main principle. It implies that any state of affairs is more desirable for its own sake than another if and only if it contains more happiness than the other.

When Mill is described as speaking of one state of affairs as “containing more happiness” than another, it must be borne in mind that this expression is used in the same sense as that in which he understands the consequences of one action as related to those of another when he regards one action as “causing more happiness” than the other. Accordingly, Mill’s contention that happiness is the only thing desirable for its own sake may be stated more fully as signifying that something is desirable for its own sake if and only if it is a state of affairs of one of three sorts: (1) a state containing some pleasure and no pain; (2) a state containing both pleasure and pain, but in which, whether or not there is an excess of pleasure over pain, the pleasures on the whole are more desirable than the pains are undesirable; (3) a state containing both pleasure and pain, and in which, although neither the pleasures nor pains are of sorts such that the pleasures on the whole are more desirable than the pains are undesirable or such that the pains on the whole are more undesirable than the pleasures are desirable, there is an excess of pleasure over pain. Mill likewise holds that something is undesirable for its own sake if and only if it is a state of affairs the opposite of one of these three.

Mill’s contention implies that no inanimate thing or state of affairs made up only of inanimate things is desirable or undesirable for its own sake. It implies that no human being or human disposition is desirable or undesirable for its own sake. According to it, the only sort of matter that is desirable or undesirable for its own sake is a state of affairs comprising sentient beings. It implies that neither justice nor liberty nor peace is desirable for its own sake. It implies, moreover, that there is nothing desirable for its own sake save where there is life; and that there is nothing undesirable for its own sake save where there is life. Although Mill’s contention affirms a certain universal and peculiar feature of whatever is desirable for its own sake, it does not also state any such feature of whatever is desirable. While it implies that an inanimate thing, a human being, or justice or liberty or peace or life is not desirable for its own sake, it does not imply that none of these can be desirable for what will come of it. Mill’s contention implies that although a certain state of affairs is desirable for its own sake, it may still be undesirable; and even though a certain state of affairs is undesirable for its own sake, it may still be desirable, for what comes of it. Mill’s main principle implies that even if it would be undesirable for a certain action to be done, it would not follow that it should not be done. It implies that even if a certain action would have desirable effects, it should not be done, if some alternative to it would have more desirable effects. Mill’s principle implies that even though the consequences of a certain action would on the whole be undesirable for their own sake, it may still be the case that it should be done. This would be the case if the consequences of any alternative to it would be more undesirable for their own sake.

II.

THE EVIDENCE OF WHAT IS DESIRABLE

mill’s argument to support his contention that happiness is the only thing desirable for its own sake contains two steps. In the first step he seeks to show that happiness is desirable; in the second, he seeks to show that it is the only thing desirable for its own sake. He writes, in the first step:

In the second step Mill acknowledges that men actually “do desire things which, in common language, are decidedly distinguished from happiness” (235). But he endeavours to show that “Whatever is desired otherwise than as a means to some end beyond itself, and ultimately to happiness, is desired as itself a part of happiness. . .” (237). Central to both steps in his argument is Mill’s contention that the sole evidence that anything is desirable is that it is desired.

G. E. Moore urges that by asserting that the fact that something is desired is evidence that it is desirable, Mill is holding that if anything is desired it is desirable; and that by affirming that this is the sole evidence, Mill is holding that nothing is desirable unless desired. Moore also interprets Mill as inferring from this that “desirable” means “desired.” He points out, moreover, that Mill uses the words “good” and “desirable” interchangeably. Hence Moore contends that Mill is claiming that “good” means “desired.”13 Moore urges two objections against Mill: first, that “desirable” does not mean “desired,” and secondly, that even if something is desirable if and only if it is desired, it is fallacious to infer that “desirable” means “desired.” Both objections fail to apply to Mill; Mill does not draw the inference Moore attributes to him, nor does he maintain that “desirable” means “desired.” Mill also does not hold that “visible” means “seen.” Instead he asserts that the proof that something is visible is that it is seen. Similarly, what he affirms is that the sole evidence that anything is desirable is that it is desired.

To this Moore urges two further objections, independent of the foregoing. The fact that something is desired would be evidence that it is desirable if and only if it is the case that from the mere fact that anything is desired it follows that it is also desirable. But from the mere fact that something is desired Moore objects that it does not follow that it is desirable. Moore does not question Mill’s contention that the fact that something is seen is proof that it is visible, for by “visible” is meant “capable of being seen.” He contends, however, that Mill is wholly unwarranted in arguing that “in like manner” the fact that a thing is desired is evidence that it is desirable, for he points out that by “desirable” is not meant “capable of being desired.” Just as “detestable” means not “capable of being detested” but “worthy of being detested,” so similarly, Moore urges, when something is said to be desirable, what is meant is that it ought to be desired, that it is worthy of being desired. From the fact that something is actually desired it does not follow that it ought to be desired.

Moore urges a second objection against anyone who would try to save Mill’s dictum by holding that Mill uses “desirable” in it to mean “capable of being desired.” He points out that Mill puts forth this dictum to establish the conclusion that the general happiness is the only thing desirable for its own sake. If Mill is construed as using “desirable” in the sense of “capable of being desired” in his premise, Moore contends that his argument then becomes fallacious, since Mill does not use “desirable” in this sense in the conclusion. If anyone should still try to save Mill’s argument against this objection by urging that in the conclusion as well Mill means by “desirable,” “capable of being desired,” Moore contends that this will not do. He points out that in saying that happiness alone is desirable for its own sake, Mill makes it clear that he means that it alone is good for its own sake. Moore also points out that in saying that the general happiness alone is desirable for its own sake, Mill does not mean that it alone is capable of being desired for its own sake. Since Mill himself mentions that each person desires his own happiness, he acknowledges that men are capable of desiring something other than the general happiness for its own sake. Moore calls attention to another connection in which Mill makes this point. Mill remarks that it is a mistake to “confound the rule of action with the motive of it,” and continues, “ninety-nine hundredths of all our actions are done from other motives. . .” (219). Here too Mill makes it clear that, in saying that the general happiness is the only thing desirable for its own sake, he in no way holds that the only desire from which men can act or the only desire of which they are capable is desire for the general happiness.14

D. Raphael and E. W. Hall seek to defend Mill against these objections urged by Moore.15 They contend that Moore’s objections are beside the point, since they criticize Mill for doing something which he does not profess to do. They urge that Mill does not claim to prove that happiness is desirable because it is desired. They direct attention to what Mill has to say upon this matter. Mill writes, “The medical art is proved to be good, by its conducing to health. . . .” He generalizes, “Whatever can be proved to be good, must be so by being shown to be a means to something admitted to be good without proof” (207-208). Here Mill is saying two things: first, that whatever can be proved to be good can be so proved only by being shown conducive to something else that is good; second, that since something cannot be proved to be desirable for its own sake by being shown to be desirable as a means to something else, no proof can be given of what is desirable for its own sake. This conclusion Mill at once qualifies: “Questions of ultimate ends are not amenable to direct proof.” Mill still concedes that such questions are not amenable to what is “commonly understood by proof,” but he contends that they are amenable to a “larger meaning of the word proof. . . . Considerations may be presented capable of determining the intellect either to give or withhold assent.” Moore recognizes that Mill does not claim to give a proof of what things are desirable for their own sake in terms of what is commonly understood by proof. He agrees with Mill that no such proof can be given of what things are desirable for their own sake. Moore also agrees with Mill that considerations may be presented in favour of thinking that certain things and not others are desirable for their own sake. Raphael and Hall err in accusing Moore of taking Mill to be offering a proof in the “commonly understood” sense. Moore’s objection is rather that one consideration which Mill presents “to determine the intellect to give assent” to what is desirable is invalid. Because something is desired it does not follow that it is desirable. Hence the fact that something is desired does not constitute evidence that it is desirable.

To make good his defence of Mill, Raphael must show that this consideration which Mill presents is not open to Moore’s objection. Raphael points out that in his Logic Mill maintains that whoever says that something should be done is recommending that it be done. Such a person, Mill writes, “speaks in rules, or precepts.”16 Mill continues, such “propositions . . . enjoin or recommend that something should be. They are . . . expressed by the words ought or should be.”17 Second, Raphael contends that Mill holds that “all rules or precepts are aimed at the promotion of ends.” He is referring to Mill’s remark, “All action is for the sake of some end, and rules of action, it seems natural to suppose, must take their whole character and colour from the end to which they are subservient” (206). Third, Raphael takes Mill as holding that “an ultimate end is that by reference to which we prove the propriety of adopting subordinate ends or particular rules.” He thereby construes Mill as maintaining that whenever men recommend something as desirable, their recommendations must ultimately have reference to an ultimate end. Finally, Raphael ascribes to Mill the view that “the ultimate end or criterion of human action is what human beings desire.”18 Accordingly, Raphael maintains that what Mill means by his dictum that “the sole evidence . . . that anything is desirable, is that people do actually desire it” is that when “we recommend . . . as ‘desirable’ . . . our recommendations must ultimately have reference to actual desires.”19

Raphael’s interpretation of Mill’s dictum fails to free it of the objection urged by Moore. For Moore urges that even if someone aims at a certain thing as an ultimate end, that is, as an end for its own sake, it still makes sense to ask whether that at which he aims is desirable for its own sake. From the fact that it is aimed at for its own sake, it does not follow that it is desirable for its own sake. Raphael also misrepresents Mill’s dictum, in construing it as maintaining that when anything is recommended as desirable, the recommendation must ultimately have reference to men’s desires. He construes it in this way by ascribing to Mill the view that when anything is recommended as desirable, it can be recommended only by reference to an ultimate end. Mill, however, does not hold that something can be shown to be desirable only by being shown to be a means to an ultimate end. He is instead concerned with how it is possible to make out what is desirable for its own sake. It is just in this connection that he puts forth his dictum.

Mill not only speaks of what is desired and what is desirable. Again and again he speaks of ends. In doing so, he makes many statements reminiscent of Aristotle. Aristotle writes, “Every action and pursuit is thought to aim at some good; and for this reason the good has rightly been declared to be that at which all things aim. . . . Will not the knowledge of it, then, have a great influence on life?”20 In a similar vein, as we have seen, Mill says, “All action is for the sake of some end, and rules of action, it seems natural to suppose, must take their whole character and colour from the end to which they are subservient. When we engage in a pursuit, a clear and precise conception of what we are pursuing would seem to be the first thing we need. . . .” (206.) Mill also asserts, “Questions about ends are . . . questions what things are desirable.” The “sole evidence it is possible to produce that anything is desirable, is that people do actually desire it” (234). Aristotle writes, “If, then, there is some end of the things we do, which we desire for its own sake (everything else being desired for the sake of this) . . . this must be the good and the chief good.”21 In virtue of such similarities, the objection Moore urges against Mill is equally applicable to Aristotle’s arguments. Moore would contend that because there is that which is desired for its own sake, and all else that is desired is desired for the sake of it, it does not follow that it is desirable for its own sake, or that it alone is desirable for its own sake.

Mill also writes, “happiness is desirable, and the only thing desirable, as end. . . .” Each of virtue, pleasure, money, power, and fame, “once desired as an instrument for the attainment of happiness, has come to be desired for its own sake. In being desired for its own sake it is, however, desired as part of happiness. The person is made, or thinks he would be made, happy by its mere possession. . . . Whatever is desired otherwise than as a means . . . to happiness, is desired as itself a part of happiness. . . .” (236-7.) Aristotle similarly writes,

Here Aristotle distinguishes what is worthy of pursuit from what is pursued and what is desirable from what is desired. Yet in the third sentence he again exposes himself to Moore’s objection: because something is chosen for its own sake and never for the sake of something else, it does not follow that it is “worthy of pursuit,” “desirable in itself,” or “always desirable in itself.” In his Logic, Mill asserts: “Every art . . . enunciates the object aimed at, and affirms it to be a desirable object. The builder’s art assumes that it is desirable to have buildings. . . . The hygienic and medical arts assume, the one that the preservation of health, the other that the cure of disease, are fitting and desirable ends.”23 Aristotle similarly writes, “In different actions and arts . . . the good of each [is] that for whose sake everything else is done . . . the end.”24 To this Moore’s objection again applies. Undeniably that for the sake of which everything that is done in a certain sphere of activity is often something good, something desirable. But because there is that in a certain art or sphere of activity for the sake of which everything within it is done, it does not follow that it is desirable. Here it is to be noted that Mill is in complete accord with Moore’s objection. He follows up the last passage by writing, “To this art [the Art of Life] . . . all other arts are subordinate; since its principles are those which determine whether the special aim of any particular art is worthy and desirable.” Here Mill clearly recognizes that the fact that something is the aim of a certain pursuit in no way implies that that aim is desirable. Elsewhere Mill makes it quite clear that he holds that whether a certain pursuit should be engaged in depends not on what its aim is but on whether the consequences of engaging in it would be more desirable.

The core of Moore’s objection to Mill’s dictum, on the evidence for what is desirable, is that from the fact that something is aimed at, it does not follow that it ought to be aimed at; and that from the fact that something is desired, it does not follow that it ought to be desired. Mill is in complete accord with Moore on the general point of which these are instances. He devotes his entire essay, “Nature,” to refuting the notion that nature, that which is, determines that which ought to be.25 Mill is also in full agreement with the specific point Moore urges in objection to him. Neither nature generally nor man’s own nature can determine what ought to be. Many a propensity is to be extirpated.26 Because men have a propensity or desire for something, it in no way follows that it ought to be desired. A further look may then be taken at Mill’s argument to see if it is free of Moore’s objection.

In support of the conclusion that only happiness is desirable for its own sake, Mill urges that only happiness is desired for its own sake. Moore contends that in speaking of what is desirable for its own sake, Mill is speaking of what ought to be desired for its own sake. Moore objects that from the fact that something is desired it does not follow that it ought to be desired. We may then inquire what can be inferred from the premise that only happiness is desired for its sake. If something is incapable of being done, it cannot be the case that it ought to be done. Accordingly,

(a) Only that which is capable of being desired for its own sake ought to be desired for its own sake.

Moore does not question that whatever is desired for its own sake is capable of being desired for its own sake. Similarly, it seems that

(b) Only that which is desired for its own sake is capable of being desired for its own sake.

Completing the argument,

(c) Only that which is desired for its own sake ought to be desired for its own sake.

(d) Only happiness is desired for its own sake.

(e) Hence, only happiness is desirable for its own sake.

The question at issue is not whether (d) is correct, but whether (c) is. Statement (c) follows from (a) and (b), so what calls for scrutiny is (b). If something is desired for its own sake it follows that it is capable of being desired for its own sake. It does not in like manner hold, nor can it be inferred from this, that if something is alone desired for its own sake it alone is capable of being desired for its own sake. Mill, however, does not include (b) in his argument. He does not hold that whatever is visible is seen; he contends rather that only that which is seen is that for which there is evidence that it is capable of being seen. Mill is similarly concerned to determine whether there is evidence that anything other than happiness is capable of being desired for its own sake. He urges that the only evidence that is offered is that virtue, money, power, and fame are desired for their own sake. Mill does not reject this evidence. Instead, he seeks to show that when any of these is desired for its own sake, it is desired only as a part of happiness. Instead of (b), Mill would aver

(b′) Only that which is desired for its own sake is that for which there is evidence that it is capable of being desired for its own sake.

From (a) follows

(a′) Only that for which there is evidence that it is capable of being desired for its own sake is that for which there is evidence that it ought to be desired for its own sake.

From (b′) and (a′) follows

(c′) Only that which is desired for its own sake is that for which there is evidence that it ought to be desired for its own sake.

Moore objects to Mill’s dictum, on the evidence for what is desirable, by construing it as affirming that from the fact that something is desired it follows that it ought to be desired. Mill, however, does not hold that from the fact that something is desired, it follows that it ought to be desired. He does not maintain that whatever is desired ought to be desired; he speaks rather of the only evidence that something is desirable. Moore says that by “desirable” Mill means “ought to be desired,” and it is only on this interpretation that he raises his objection against Mill’s dictum. If Moore is correct in this, then what Mill’s dictum maintains is (c′). Moore’s objection against Mill’s dictum carries no weight against it; there is nothing incompatible in affirming (c′) and denying that whatever is desired ought to be desired.

Moore is correct in pointing out that when Mill argues that happiness is the only thing desirable for its own sake, he means by “desirable” not “capable of being desired” but “good,” and that by “desirable for its own sake” he means “good in itself,” “intrinsically good.” Moore also contends that by “desirable” Mill, or anyone else, means “ought to be desired” or that which it would be good to desire. There is a fatal objection to this contention, at least in regard to Mill. Since Mill holds by his main principle that something ought to be done only if it would cause more happiness, he holds that something ought to be desired only if desiring it would cause more happiness. Hence if Mill is construed as meaning by “desirable,” “ought to be desired,” he would then be maintaining that the consequences of an action would be desirable only if desiring them would cause more happiness. But this is clearly not what Mill contends; for him the consequences of an action would be more desirable only if that action would cause more happiness.

There is a further objection to contending that “desirable” means “ought to be desired,” which applies to Mill, or to anyone who agrees with him that something should be done only if its consequences would be more desirable. For he then holds that something ought to be desired only if the effects of desiring it would be more desirable. But if “desirable” is construed as “ought to be desired,” Mill would then have to say that the consequences of an action would be desirable only if desiring these consequences would have more desirable consequences. He would similarly have to say that the consequences of desiring the consequences of a certain action would be desirable only if desiring them in turn would have more desirable effects. And so on. But Mill clearly does not think that the desirability of the consequences of an action is affected by what would be the consequences of desiring these consequences, or by what would be the consequences of desiring the consequences of desiring the consequences of the action. He maintains that the consequences of an action would be more desirable only if it would cause more happiness.

Moore overlooks certain differences between the conception of that which ought to be desired and the conception of that which is desirable. When it is said that something ought to be done, it is implied that there is some respect in which it stands in contrast to anything capable of being done instead of it. “Ought” is a superlative, as is also the conception of that which ought to be desired, but the adjective, “desirable,” is a positive term, which takes the comparative “more desirable” and the superlative “most desirable.” In accord with Mill’s assumption that something ought to be done only if what would come of it would be more desirable for its own sake, something ought to be desired for its own sake only if what would come of so desiring it would be more desirable for its own sake. Hence if something ought to be desired for its own sake, it does not follow that it would be desirable for its own sake; and because something would be desirable for its own sake, it does not follow that it ought to be desired for its own sake. Since Mill’s dictum on the evidence for what is desirable cannot be taken as a dictum on the evidence for what ought to be desired, it must be given some other interpretation than that set forth in the preceding paragraph.

It is doubtful whether anyone sincerely believes that a certain thing should be done without feeling on the whole in favour of its being done. It is similarly extremely doubtful that anyone believes that something would be undesirable without feeling some displeasure at the thought of it, or that anyone is genuinely convinced that something would be desirable without to some measure feeling pleased at the thought of it. Someone may, indeed, believe that something would be desirable in a certain respect, and yet on the whole not be in favour of it, through thinking it undesirable in other respects. Nonetheless, Mill points out that no one feels pleased to some measure at the thought of a certain state of affairs, without feeling some desire for its occurrence (237). Someone does not therefore manage to convince another that something would be desirable unless he induces him to feel some desire for it. This suggests that what Mill may be maintaining by his dictum is that no one has evidence for believing something desirable unless he has some desire for it. If it is interpreted in this way, it may be objected that people often believe that others desire something, and desire it for its own sake, without thinking that it would be desirable for its own sake. It may also be objected that on occasion a man is well aware that he desires something for its own sake, but still does not think that it would be desirable. These objections merely show that someone may believe that something is desired without believing that it would be desirable. They do not show that anyone is ever convinced that something would be desirable without having some desire for it. There is a further objection to Mill’s dictum, if it is interpreted in this way. Someone has a desire for something whenever he believes it would be desirable. He has some desire for it, whether he is correct or mistaken in believing that it would be desirable. Consequently the fact that he has a desire for something cannot serve as evidence that what he believes would be desirable would really be such. What is rather the case is that the fact that someone believes that something would be desirable is evidence that he has some desire for it.

Although the fact of something’s being desired cannot serve as evidence for the correctness of all judgments of what is desirable, it may still be the case that there are some such judgments for which it alone can serve as evidence. It is important to note the limitations which Mill himself places on the dictum that the only evidence that something is desirable is that it is desired. He does not hold that this is the evidence for all sorts of judgments of what is desirable. Nor does he claim that all desires are qualified to serve as evidence. Mill does not state that the only evidence that something is desirable as a means is that it is desired. He maintains that something is good as a means, desirable as a means, if and only if it would bring about something else that is desirable (207-8). He would contend that there is no evidence that it is desirable as a means unless there is evidence that it would have a certain effect. If something is desired in the belief that something desirable would come of it, Mill does not hold that such a desire is evidence that something desirable would come of it. He maintains that whether something is desired or not, it is desirable as a means just so long as it would have some desirable effects. He thus does not claim that the fact something is desired is either the sole evidence or even a part of the evidence to support a judgment that it is desirable as a means.

Mill also does not hold that the fact that something is desired is the sole evidence to support a judgment that it is intrinsically desirable, that is, desirable for its own sake. On this point he writes, “No reason can be given why the general happiness is desirable, except that each person, so far as he believes it to be attainable, desires his own happiness. This, however, being a fact, we have . . . all the proof . . . that each person’s happiness is a good to that person, and the general happiness, therefore, a good to the aggregate of all persons.” (234). Countless critics have urged that it is fallacious for Mill to infer that since each desires his own happiness therefore everyone desires the general happiness. Mill, however, does not here infer that the general happiness is desired. What he argues, rather, is that it is desirable. In this passage he certainly claims that the fact that each desires his own happiness is evidence that the happiness of each is desirable. But he does not base his claim that the general happiness is desirable on the evidence that it is desired. In a letter he explains, “when I said that the general happiness is a good . . . I merely meant . . . to argue that since A’s happiness is a good, B’s a good, C’s a good, &c., the sum of all these goods must be a good.”27 Mill is holding that if the happiness of A is intrinsically desirable and the happiness of B is intrinsically desirable and the happiness of C is intrinsically desirable, then the “sum” of the happiness of A and the happiness of B and the happiness of C is intrinsically desirable. Put generally, what Mill is arguing is that a whole is intrinsically desirable if it is made up of components which are intrinsically desirable and which exceed intrinsically undesirable components. Mill does not hold that the fact that a whole is desired for its own sake is either necessary or sufficient evidence that it is made up of an excess of intrinsically desirable components. Thus a second sort of judgment to which Mill does not apply his dictum is a judgment that something is intrinsically desirable because it is a whole containing an excess of intrinsically desirable components.

Moore and C. I. Lewis distinguish two further sorts of judgments of what is desirable to which Mill, for similar reasons, would not regard his dictum as applicable.28 It does not apply to a judgment that something is desirable because it is a component of something intrinsically desirable. For example, when he considers it by itself, a mountain climber may well regard the toil he undergoes in reaching a mountain peak as undesirable in itself. Yet he would regard it as desirable because it enhances the desirability of the experience of reaching the mountain top, making the venture far more desirable than it would have been had he reached the peak by helicopter. In considering his toil as desirable for this reason, the climber is making a judgment which in one respect resembles judging that something is desirable as a means. He regards it as desirable because of its relation to something else. In another respect it differs. When something is judged desirable as a means, it is merely claimed that it would bring about something else desirable, whereas the climber regards one component of an experience as desirable because its experienced quality enhances the desirability of the whole experience of which it is a part. Although Mill also distinguishes that which is desirable because a part of happiness from that which is desirable because a means to happiness, he fails to mention that the fact that someone desires something because he “thinks he would be made” happy by its mere possession, supplies no evidence that it would actually enhance his happiness (236). A fourth sort of judgment is exemplified by the lover of mountain scenery who regards a certain mountain as desirable because of the delight to be had in beholding it. He is not regarding the mere existence of the mountain as desirable for its own sake. He regards the mountain as desirable because the experience of beholding it is desirable.

We have noticed four distinct ways in which something may be judged to be desirable: (1) as a means, (2) because it enhances the intrinsic desirability of something of which it is a part, (3) because it is an object of an intrinsically desirable experience, (4) intrinsically, because made up of an excess of intrinsically desirable components. By the fourth sort of judgment something is judged intrinsically desirable; by the other three, extrinsically desirable. All four sorts make the claim that something is desirable because it stands in a certain relation to something else. The evidence required for each is evidence that the relation obtains. Consequently no judgment of one of these sorts is one in which a desire for what is judged desirable is evidence of the correctness of the judgment. A fifth sort of judgment, fundamentally distinct from these four, is that something is intrinsically desirable independent of its relation to something else. For brevity, we may refer to such a judgment as a judgment of what is “desirable of itself.” Judgments of the other four sorts are logically dependent on judgments of this sort, for what they affirm to be desirable they imply is related in a certain way, directly or indirectly, to something desirable of itself. The fifth sort of judgment is logically independent of the other four.

For someone to be assured whether he is correct in judging that something is desirable of itself, one preliminary is that he avoid confusing this judgment with the other four. For a judgment of this sort, it would be out of place to adduce the kind of evidence distinctively relevant to one of the four other sorts of judgments. When Mill speaks of desires as evidence of what is desirable, he would regard this dictum as holding only for judgments of the fifth sort. The same is true when he speaks of a preference for one sort of matter over another as evidence that the one is more desirable than the other. Even for judgments of the fifth sort Mill does not claim that every sort of desire or preference can serve as evidence. He does not hold that a desire for something qualifies as evidence if it rests on the belief that it would have desirable effects, or upon the beliefs on which judgments of the other three sorts rest. He contends that someone’s preference for one sort of matter over another does not qualify as evidence unless he has had experiences of matters of both sorts and his preference is based on such experiences (211). He would not hold that his preference is based on such experiences unless they led him to it. Mill would hold that a preference by someone who has had such experiences would not qualify as evidence unless he was gladder at the one than the other. He therefore maintains that a preference for one sort of matter over another qualifies as evidence so long as it rests on nothing but having had experiences of matters of both sorts and having been gladder at the one than the other.

It may be presumed that Mill likewise holds that someone’s desire for something of a certain sort does not qualify as evidence unless it rests on experience of matters of that sort. When someone desires something, he prefers its existence to its non-existence. Since he argues that a preference for one matter over another does not qualify as evidence unless it rests on experience of matters of both sorts, Mill may be presumed to hold that someone’s desire for a certain thing does not qualify as evidence unless he has had experience of something of its sort, as well as some experience from which such a thing was absent. Mill would also hold that a desire by someone who had had such an experience would not qualify as evidence unless he was glad at what he experienced. He then holds that someone’s desire for something qualifies as evidence so long as it rests on nothing but having had experience of something of that sort and having been glad at it. It would serve as evidence for someone else, as well as for him who had the desire. Mill would certainly admit that if someone was glad at what he experienced because he expected that something desirable would come of it, or if his gladness was mediated by another of the four sorts of judgments distinguished above, such gladness would not count as evidence. For the same reason, if he was glad at it because of the kind of person he is, that is, because he desired things of that sort, his gladness would not count as evidence, if his desire in turn was mediated by any of the four other sorts of judgments. Someone’s gladness at what he experienced counts as evidence only if he was glad at it on its own account, only, that is, if his gladness was unaffected by any beliefs he has about its relation to other things. If this is a correct interpretation of Mill’s dictum, he then holds that someone’s preference or desire for something qualifies only secondarily as evidence, and that the primary evidence anyone has of what is desirable of itself is having experienced it and being glad at it.

III.

WHAT IS DESIRABLE FOR ITS OWN SAKE

having fixed on what Mill holds is the only ultimate evidence of what is desirable, we may now turn to what he maintains such evidence discloses. Mill urges that no one is ever glad on its own account at some state which his experience has disclosed to him unless some pleasure occurred in it, and therefore that no one is led by such experiences to desire like states to come about unless he expects that they will be pleasant. He also urges that no one is ever sorry on its own account about some state with which his experience has acquainted him unless there was something painful in it. Accordingly, the first thing which Mill argues that the relevant evidence discloses is that nothing is desirable of itself unless it is a state in which some pleasure is experienced and nothing is undesirable of itself unless it is a state in which some pain is felt.

Moore attacks Mill for maintaining that only pleasure is desired.29 He concedes that in instances of many desires, pleasure is one feature of that of which someone is desirous. But he urges that on such occasions, what someone looks forward to and is desirous of is a pleasant walk or a pleasant conversation with a certain person, a pleasant party with certain companions or a pleasant smoke. To this some retort that while sometimes a walk, sometimes a smoke, sometimes a party is desired, each is desired only for the sake of the pleasure it will afford, so that it is pleasure alone which is desired for its own sake. Against this others urge that while the pleasure is one element of what someone looks forward to when he desires a walk, a smoke, or a party, the walk or the smoke or the party is also a component of what he is desirous of. Aristotle points out that when someone desires a certain walk but is denied it and is provided something else that affords him pleasure, his desire for the walk remains unfulfilled.30 If pleasure alone were desired for its own sake, any pleasure would serve to fulfil a desire. Yet when someone desires a certain pleasant thing, his desire is fulfilled only by it, not by any pleasure at random. Secondly, Moore urges that on many occasions there is no expectation of pleasure characterizing that which someone is desirous of. Often someone desires to eat when hungry. While he feels pleasure at the prospect of eating, the prospect before his mind is simply that of eating certain things. A spectator watching a football game wants his team to score a goal. That of which he is thinking and of which he is desirous is its scoring. He has no thought of pleasure. When someone is struggling with a certain problem he desires a solution. No thought of pleasure is before his mind. Thirdly, Moore urges that although occasions are conceivable on which someone desires nothing but pleasure, if any occur, they are very rare; for what generally seems to be found is that someone is desirous of a pleasure of a certain sort, that is, a state characterized not only by pleasure but by other features as well.

There is nothing in these objections put by Moore which Mill does not agree with or which is incompatible with the evidence he adduces for what is desirable. Mill does not hold that only pleasure is desired. He agrees that there are many occasions on which that of which someone is desirous includes no thought of pleasure. Mill points out that many things are desired as a means to a certain end, and he notices that when something is desired as a means, there is very often no thought of it as pleasant. Mill also does not maintain that whenever something is desired without thought of what will come of it, it may be described as being desired for its own sake. He points out that men often desire something simply because they are in the habit of pursuing it, and have no thought of what it will lead to. He adds, “any . . . person whose purposes are fixed, carries out his purposes without any thought of the pleasure he has in contemplating them, or expects to derive from their fulfilment . . .” (238). He does indeed contend that nothing is desired for its own sake unless it is expected that it will be a state of affairs in which some pleasure will be experienced. But he does not claim that there are any occasions upon which pleasure alone characterizes what is desired. Although he contends that only what is desired for its own sake is evidence of what is desirable of itself, he does not think that whenever someone desires something for its own sake this counts as evidence; he holds that a man’s desire of this sort counts as evidence only if it is based on experience of similar matters and he was glad on its own account at what he experienced.

According to Mill, the evidence whether one matter is more desirable of itself than another is of the same kind. He urges that no one who has ever actually had experience of two occasions in which only pleasure of the same sort was felt is gladder on its own account about one than the other unless it was more pleasant. Nor, he argues, is anyone, who has had experience of two occasions in which only pain of the same sort was felt, sorrier on its own account about one than the other unless more pain was felt in it; and no one is led by such experiences to prefer one to another of that sort unless he expects it would be less painful. No one with experience of toothaches prefers of itself a more severe to a less severe toothache. Accordingly, Mill argues, the relevant evidence further shows that as between two states in which only pleasure of the same sort is felt, one is more desirable of itself than the other only if it is more pleasant; and as between two states in which only pain of the same sort is felt, one is more undesirable of itself than the other only if it is more painful.

What evidence has someone in judging between states in which different sorts of pleasure or pain are felt? The same kind of evidence, Mill maintains. Someone has ultimate evidence for thinking the one more desirable of itself than the other only if he experienced both and was gladder at one than the other. Even though a toothache was more painful than a grief, someone has evidence for concluding that the grief was more undesirable of itself than the toothache if he experienced both and was sorrier at the grief. As between two painful states of different sorts Mill holds that the ultimate evidence that one was more undesirable of itself than the other is that someone who experienced both is sorrier at the one than the other. As between two pleasant states of different sorts he holds that regardless of whether one was more pleasant than the other, the ultimate evidence that it was more desirable of itself is that someone who experienced both was gladder at the one and is led by this to prefer, in the future, experiences like the one to experiences like the other. From this Mill ventures also to generalize what sorts of experiences are more desirable of themselves, independent of whether they are more pleasant: “the pleasures of the intellect, of the feelings and imagination, and of the moral sentiments [have] a much higher value as pleasures than . . . those of mere sensation,” than “bodily pleasures” (211). For such a generalization that compares sorts of pleasures, the experiences of many are clearly relevant: “Of two pleasures, if there be one to which all or almost all who have experience of both give a decided preference, irrespective of any feeling of moral obligation to prefer it, that is the more desirable pleasure” (211). In making the generalization that pleasant experiences of one sort are more desirable of themselves than those of another sort, Mill does not deny that a certain experience of a less desirable sort may be so much more pleasant than one of a more desirable sort as to be more desirable than it, or that certain painful experiences of a less undesirable sort may be so much more painful as to be more undesirable. He does not deny that a certain bodily agony may be more undesirable of itself than a certain grief. Here too Mill holds that the ultimate evidence someone has that the bodily agony was of such an intensity that it was more undesirable than the grief, is his having experienced both and being sorrier about the agony (213).

Moore charges that Mill’s contention that some experiences are more desirable of themselves than others, even though not more pleasant, is inconsistent with his contention that nothing is desirable of itself unless it is a state in which some pleasure is experienced. Raphael seeks to free Mill of this charge of inconsistency by urging that Mill does not hold that it is possible “that a pleasure of higher quality may contain a lesser or no greater quantity of pleasure than a pleasure of lower quality.”31 Raphael continues, “Mill’s criterion is preference, and I think he would say that to prefer one pleasure to another is to desire it the more strongly. And since he says later, in Chapter IV, that to desire a thing is the same as to think it pleasant, it follows, on this view, that to prefer a thing is to think it more pleasant.” Mill certainly holds that whoever prefers one thing to another desires it more. He also writes, “desiring a thing and finding it pleasant . . . are . . . inseparable” (237). But he does not claim that no one desires one thing more than another unless he expects that it will be more pleasant. Instead he writes,

In pursuing the point further, Raphael gives up his contention that “the distinction of quality is, at bottom, the same as the distinction of quantity.” He no longer interprets Mill as holding that it is impossible for one experience to be more desirable of itself unless it is more pleasant. Instead, he takes Mill to mean that one experience is never in fact more desirable of itself unless it is more pleasant. In support of this, Raphael points out that when Mill remarks that it is “better to be Socrates dissatisfied than a fool satisfied,” Mill also denies that “this preference takes place at a sacrifice of happiness” (212). Raphael urges that although Socrates is dissatisfied and the fool not, it is consistent for Mill to maintain that Socrates is happier than the fool and his happiness more desirable, in so far as Socrates “enjoys a greater balance of pleasure over pain, than the fool.” Raphael does indeed show that in maintaining that it is better to be Socrates dissatisfied than a fool satisfied, it would be consistent for Mill to hold that no experience is more desirable of itself than another unless it is more pleasant. In making this point, however, Raphael fails to show that Mill does in fact maintain that one experience is more desirable of itself than another only if it is more pleasant. Mill speaks, instead, of “what makes one pleasure more valuable than another . . . except its being greater in amount”; he writes of being “justified in ascribing . . . a superiority in quality, so far outweighing quantity . . .” (211). In a journal of 1854, Mill remarks, “Quality as well as quantity of happiness is to be considered; less of a higher kind is preferable to more of a lower.”32 Raphael himself acknowledges that Mill would “say that a superior pleasure may be less intense than an inferior.”

Moore, however, charges that “Mill’s judgment of preference, so far from establishing the principle that pleasure alone is good, is obviously inconsistent with it. . . . If one pleasure can differ from another in quality, that means, that a pleasure is something complex, something composed, in fact, of pleasure in addition to that which produces pleasure.”33 Mill is involved in no difficulty here. When he holds that only a pleasure is desirable of itself he is not holding that only the pleasantness of a pleasant experience is desirable. By a pleasure he understands a pleasant experience. He maintains that only a complex, that is, an experience having pleasantness as one of its features, is desirable of itself. The inconsistency with which Moore charges Mill is this:

This charge is easily rebutted. In holding that some experiences are more desirable of themselves than others, although not more pleasant, Mill certainly admits that the intrinsic desirability of an experience may be enhanced by other components of it than the pleasure enjoyed in it. He would therefore agree with Moore that such components “may be good . . . independently of the pleasure which accompanies them.” But Mill would hold that such components are desirable as contributing to the intrinsic desirability of the experience. He does not maintain that any experience is desirable of itself if it has such other components but is not also pleasant. Consequently, when Mill argues that the relevant evidence shows that nothing is desirable in itself unless it is a state in which some pleasure is enjoyed, it is not inconsistent for him to argue that the relevant evidence also shows that some pleasant experiences are more desirable of themselves although not more pleasant.

One writer contends that Mill means by “ ‘pleasure,’ whatever is made the object of desire.”35 Mill, however, does not hold that whenever anyone desires something as a means—say, having a tooth extracted—it is to be described as a pleasure. Mill also mentions that men often desire something simply because they are in the habit of pursuing it and that “any . . . person whose purposes are fixed, carries out his purposes without any thought of . . . pleasure.” Another writer contends that Mill uses “pleasure” as “a technical term for whatever anyone desires for its own sake.”36 Mill, however, does not regard an enjoyable experience as any less a pleasure when it comes to a man without having been desired. He maintains that some experiences are desired for their own sake more than others although not more pleasant. Moreover, while he holds that there is no happiness without pleasure, he does not think that when someone desires happiness for its own sake, what he desires is to be described as a pleasure.

Mill does not hold that the only things that are desirable of themselves are transient experiences in which pleasure alone is felt or that the only things that are undesirable of themselves are transient experiences in which pain alone is felt. He does not question that even if it involves both pleasure and pain, the whole of a man’s life, or some prolonged portion of it, may be desirable or undesirable of itself. We might expect Mill to hold that one portion of a man’s life is more desirable than another if the pleasant experiences comprising it are more pleasant and more numerous and the painful less painful and less numerous, provided these component experiences are not of more desirable sorts than others; and that in so far as some of the component experiences are of more desirable sorts than others, one portion of a man’s life is more desirable if its components are more desirable and its more desirable components are more numerous. In one passage Mill speaks as if one portion of a man’s life is happier and more desirable so long as these conditions alone are fulfilled. He writes, “. . . Greatest Happiness . . . is an existence exempt as far as possible from pain, and as rich as possible in enjoyments, both in point of quantity and quality. . . .” He immediately adds, “the test of quality, and the rule for measuring it against quantity, being the preference felt by those who, in their opportunities of experience . . . are best furnished with the means of comparison” (214). In this passage Mill speaks as if he regards the intrinsic desirability of a portion of a man’s life, taken on the whole, as dependent only on the intensity and intrinsic desirability of each of the several component pleasant and painful experiences and upon the proportion among them.

Elsewhere, however, Mill does not maintain that there is immediate evidence for the intrinsic desirability only of momentary experiences. For he urges that evidence that a man’s happiness is desirable is furnished by the fact that he desires it (234). From what Mill says regarding preferences, it is clear that he would not hold that a man’s desire for happiness supplied evidence unless he had experience of the matters comprising happiness and was glad at them. Mill also urges that the evidence that one sort of life is more desirable of itself than another is preference (211). But he does not hold that a man’s preference is evidence that one “mode of existence” is on the whole more desirable of itself than another, unless his experience has acquainted him with both and he was gladder at one sort than the other. He then holds that the ultimate evidence that one portion of a man’s life was intrinsically more desirable than another is that he who had experience of both was gladder on the whole at it. Mill would not hold that someone’s being gladder at one portion of life is evidence that it was more desirable on the whole, unless he was acquainted with the many experiences comprising each. In what way would he take account of the component experiences? In looking back over a portion of his life, someone will look upon some experiences that were quite desirable of themselves as detracting from the desirability of the whole and will see others of little desirability in themselves as appreciably enhancing the desirability of the whole. In assessing the intrinsic desirability, on the whole, of a portion of a man’s life, the desirability of each component experience to be reckoned with is not the desirability it has of itself but its desirability as contributing to the intrinsic desirability of that portion of life on the whole. In desiring his own happiness henceforth, moreover, it is then reasonable for a man to rate any experience that may befall him not in terms of its intrinsic desirability but in terms of its desirability as enhancing the desirability of his life on the whole.

When Mill speaks of the most desirable life for a man as an “existence exempt as far as possible from pain, and as rich as possible in enjoyments, both in quantity and quality,” he holds that a man’s life is intrinsically more desirable the greater the preponderance of intrinsically desirable experiences comprising it. It is indeed logically possible that the greater the preponderance of intrinsically desirable experiences comprising a man’s life the more it would also be made up of component experiences which enhanced its desirability on the whole. Yet it seems doubtful that this often in fact would be the case. Mill hardly faces this issue. At all events, he would hold that the reason why any experience is desirable as a component of happiness is not that it is desirable of itself but that it enhances the desirability of the life of which it is a part. It is doubtless not because he regards active pleasures as more pleasant or as of an intrinsically more desirable sort, but because he regards them as enhancing the desirability of life, that Mill speaks of a man’s happiness as greater if it includes “many and various pleasures, with a decided predominance of the active over the passive . . .” (215).

Although Mill neglects to distinguish the desirability of a pleasant experience as a part of happiness from its desirability of itself, he uses this distinction with regard to other matters. Mill acknowledges that men desire for their own sake “things which, in common language, are decidedly distinguished from happiness” (235). He cites virtue, money, power, fame. In order to show that desires for these do not supply evidence that other things than happiness are intrinsically desirable, Mill seeks to argue that when any of these comes to be desired no longer as a means, it is desired only as a part of happiness. Moore urges three objections against this.37 He contends that “these admissions are . . . in . . . glaring contradiction with his argument that pleasure . . . is the only thing desired.” He reproaches Mill for holding that “ ‘money,’ these actual coins . . . are . . . a part of my pleasant feelings.” He condemns Mill for holding that “what is only a means to an end, is the same thing as a part of that end.” When Mill speaks of things desired as a part of happiness, he is not speaking of them as a part of pleasant feelings but as a part of “an existence made up of few and transitory pains, many and various pleasures. . .” (215). Mill’s contention that nothing is desired for its own sake save that which involves some pleasant experience is not contradicted by his contention that objects of desire are characterized by other features as well. Mill also does not claim that whatever is desired as a means to happiness is desired as a part of happiness. He claims rather that certain things desired as a means to happiness come through that association to be desired no longer as a means, and that when this has occurred, they are desired as a part of happiness. C. D. Broad attacks Mill for contending that originally human beings desire things because they expect them to be pleasant and later come to desire other things as well by association. He urges that “it is unlikely that” humans in early infancy “have the experience of desiring . . . for a reason at all.”38 But Mill does not hold that infants originally desire things only because they expect them to be pleasant. He points out that it is not the case that whatever even adults desire “they have the experience of desiring . . . for a reason.”39

Mill seeks to show that only happiness is intrinsically desirable by arguing that when anything else—virtue, fame, power, money—once desired as a means to happiness, comes to be desired for its own sake, it is desired only as a part of happiness. Even if that which is desired is in fact a part of “an existence made up of few and transitory pains, many and various pleasures,” this does not show that it is desired only as a part of happiness. Mill argues: “What was once desired as an instrument for the attainment of happiness, has come to be desired for its own sake. In being desired for its own sake it is, however, desired as part of happiness. The person is made, or thinks he would be made, happy by its mere possession. . . .” (236.) This argument is open to more serious objections. If Mill can succeed in showing that virtue, or fame, or power, or money comes to be desired only as a part of happiness, he can no longer hold that it is desired for its own sake. He then removes his ground for arguing that the “ingredients of happiness are very various, and each of them is desirable in itself. . .” (235). And even if he is successful in showing that each comes to be desired only as a part of happiness, this in no way establishes that each is desirable as a part of happiness. The fact that a certain individual desires money because “it has come to be itself a principal ingredient of the individual’s conception of happiness” or because he “thinks he would be made happy by its mere possession” does not show that his happiness would in fact be enhanced thereby.

Mill is particularly concerned about virtue. He notices that a man is not virtuous unless he enjoys acting virtuously (239). Virtuous conduct is therefore not only desirable of itself; it is also a pleasant activity which is desirable because it enhances a man’s happiness. Mill also notices that men cannot be virtuous without acting disinterestedly (235). He urges that it is desirable that they be virtuous, for they then have dispositions leading them to do what is desirable (235). Mill hereby acknowledges that in this respect virtue is desirable as instrumental to happiness, not desirable as a component of happiness which enhances it. Although it is desirable that men be virtuous as a means to happiness, Mill notices that a man cannot be virtuous if he desires to be virtuous or to do what is virtuous as a means to happiness. A man cannot be virtuous unless he desires to do what is virtuous for its own sake. What appears to trouble Mill is how to acknowledge the disinterestedness of virtue without acknowledging that it is something other than happiness desired for its own sake, and therefore desirable for its own sake. The solution Mill adopts is that when a man desires virtue for its own sake, he desires it only as a part of happiness, that is, in the belief that it will enhance his happiness. This solution will not do. If a man desires to be virtuous because it will enhance his happiness, he falls short of being genuinely virtuous just as when he desires to be virtuous as a means to happiness. When a man desires to be virtuous he also hopes for happiness, but he does not desire to be virtuous out of the hope that it will yield him happiness. Mill overlooks another solution which his own line of reasoning affords. No one who considers the matter dispassionately regards it as desirable of itself that the virtuous suffer and the evil be meted out happiness.

Mill maintains that only happiness or what includes happiness is intrinsically desirable. He also commonly speaks of only a life or an extended portion of a life as happy or unhappy. This raises a further issue. If Mill thinks that only a period of life comprising several component experiences can be happy or unhappy, he must then deny that any momentary pleasant experience is intrinsically desirable and that any transient painful experience is intrinsically undesirable. On the other hand, Mill holds that if someone who has had first-hand acquaintance with an experience is glad on its own account that it occurred, this is conclusive evidence that it was intrinsically desirable. He would hold that this evidence would not be upset if later someone would be gladder if that particular pleasant experience had not occurred because it detracted from the happiness of a period of life of which it was a part. Mill also seems to maintain that if, among its many consequences, the only effect that an action has on a certain man is to cause him some brief pleasure, it then causes him happiness. Even if the brief pleasure it caused him was such that it detracted from his happiness, Mill would have to admit that it was intrinsically desirable. He can then not continue to adhere to the contention that the relevant evidence shows that only happiness is intrinsically desirable. Mill would not be troubled by this qualification, for he can still maintain that happiness or what includes happiness is invariably intrinsically more desirable than that which does not.

If Mill held that happiness is the only thing intrinsically desirable, he could not claim that the effects of one action are intrinsically more desirable than those of another if and only if it causes more happiness. But he can maintain this because he contends that the effects of one action are intrinsically more desirable than those of another if the one set of effects contains more happiness than does the other. Mill does not support this contention by direct appeal to the ultimate evidence for what is desirable but by inference from what it discloses. His inference is that since it is intrinsically desirable for A to be happy and intrinsically desirable for B to be happy and intrinsically desirable for C to be happy, it is intrinsically desirable for A and B and C each to be happy.40 While he would hold that the reason why any experience is desirable as a component of a certain man’s happiness is not that it is intrinsically desirable but that it enhances the desirability of his life on the whole, a like consideration does not apply regarding the “general happiness.” Mill contends that a state of affairs comprising the happiness and unhappiness of many beings is intrinsically more desirable the more happiness it comprises and the greater the preponderance of happiness over unhappiness within it. Some critics charge him with introducing an extraneous consideration when he adds Bentham’s dictum, “everybody to count for one, nobody for more than one.” Mill, however, points out that when this dictum is understood as asserting that “equal amounts of happiness are equally desirable, whether felt by the same or by different persons,” and that “one person’s happiness, supposed equal in degree (with the proper allowance made for kind), is counted for exactly as much as another’s,” all that is spelled out by it is that the preponderance of happiness over unhappiness be “both in point of quantity and quality” and in nothing else (257, 214).

IV.

ANALYSIS OF MORAL CONCEPTS

we have now to consider a further set of objections urged against Mill’s utilitarianism. It is urged that if it is correct, whenever someone could more effectively promote the general happiness by taking another’s automobile and continuing to use it without his consent, it would be quite right for him to do so. Whenever someone could make better use of another’s house or clothing or other possession, there would be nothing wrong in his stealing it. The fact that it belonged to another would be irrelevant. It is contended that utilitarianism rides roughshod over all rights, not only rights of property. If a wife and children are burdened with a cantankerous husband and father, it would be right for her to drown him secretly and replace him with another husband, if everybody affected would be happier in consequence. Since utilitarianism reckons only with consequences, it is also urged that it can find no place for what is fair or just, or for men being rewarded as they deserve. Because it is unfair of a father to provide for some of his children while neglecting the others, or for some to cheat on their income tax while deriving the advantages from those who make full returns, or for many to toil long hours with little returns while the idle and lazy enjoy an abundance of good things, or for one to receive the credit for what another has accomplished—all this is irrelevant, so long as the resultant enjoyment is maximized. If the happiness of a country is best realized by slavery, it is claimed that any appeal to the injustice of slavery or to men’s right to freedom are considerations of which utilitarianism can take no account.41

Utilitarianism is also criticized for holding that men have but one duty, to maximize enjoyment. This is not a duty to any specific persons. Humans and other animals are looked upon as only so many “dumping grounds” on which to bestow enjoyment. It is not denied that people have duties to promote the happiness of others, but what is urged is that they have duties to provide different sorts of happiness to different persons and other duties to certain persons than to promote their happiness.42 A man has a duty to afford his wife certain enjoyments which he does not have a duty to furnish other women. He has duties to his children which he does not owe to other children. The happiness which he owes his children is different from that which he owes his wife. When someone has hired a man to paint his house, he thinks that it is right to pay him because he has promised to. He does not reckon whether some alternative use of his money would more effectively promote the general happiness. It is urged that utilitarianism takes account only of consequences but that duties such as these arise from an antecedent relationship in which someone stands to certain persons. It is pointed out that besides these duties, men have duties which they owe to all men—to tell the truth, for instance. Granted that this duty may be outweighed on occasion by a more stringent obligation, it is argued that it does not cease whenever the general happiness would be more effectively promoted by neglecting it. It is not denied that by doing what is right a man very often does what will in fact promote the general happiness, but it is urged that utilitarianism is guilty of gross oversimplification, disregarding the diversity of considerations determining what is the right thing to do. In virtue of these it is contended that it is very often morally incumbent on a man to do a certain thing whether or not it would maximize the general happiness.

Most of these objections are not to Mill’s contention that happiness is the only thing intrinsically desirable; they rather criticize Mill for contending that questions of right and wrong are questions of what would have the most desirable consequences. Mill seeks to cope with objections such as these by elucidating what is implied when it is asserted that it would be right or wrong or unjust to do a certain thing, and by analyzing what is meant when someone is said to have a right to something or to have an obligation to do a certain thing.

Mill notices that very often when people say that a certain thing ought not to be done they would not also be prepared to say that it would be wrong to do it. He writes, “the morality of an individual action is not a question of direct perception, but of the application of a law to an individual case” (206). He maintains that whenever it is asserted that it would be wrong to do a certain action, it is claimed that there is some “rule of morality” against it. He also writes, “it would be unworthy of an intelligent agent not to be consciously aware that the action is of a class which, if practised generally, would be generally injurious. . .” (220). From this it might be thought that Mill holds that all that is contained in the claim that there is a rule of morality against a certain action is that it is an action of a kind which generally ought not to be done. If this is Mill’s view, there is a fatal objection to it. Glancing at his fuel gauge, a motorist thinks he ought to get more gasoline. He thinks that he ought to do so in the belief that a motorist in general ought to replenish his supply of fuel when it is almost exhausted. Yet he would not think that he would be doing something wrong if he were not to get more gasoline. Mill does not maintain that to claim that there is a rule of morality against a certain action is simply to claim that it is an action of a kind which generally ought not to be done.

In his essay, On Liberty, Mill distinguishes two sorts of rules of conduct. He holds that a rule of conduct is not part of the law of the land unless infractions of it incur punishment by the government. To laws he contrasts rules sanctioned by general condemnation.43 Since he also speaks of these as sanctioned by “moral coercion,”44 it might be thought that he holds that when anyone claims that there is a rule of morality against a certain action, all that he is claiming is that it is an action of a kind which incurs general condemnation. Mill, however, does not deny that men often believe that it would be wrong to do a certain action although welll aware that it is not of a sort that is generally condemned.45 He does not maintain that the fact that an action is of a kind that incurs general condemnation entails that there is some rule of morality against it.46 Instead, he writes, “We do not call anything wrong, unless we mean to imply that a person ought to be punished in some way or other for doing it; if not by law, by the opinion of his fellow creatures. . .” (246). Mill thus urges that when it is said that it would be wrong to do a certain action it is implied not that it is an action of a kind which is in fact generally condemned but rather that it is of a kind which ought in general to be condemned by others. In the same passage, he continues, “This seems the real turning point of the distinction between morality and simple expediency.” He contends that when it is said that it would be wrong to do a certain action, it is implied what others ought to do about it, by way of condemnation. For this, if for no other reason, the distinction between the notion of “wrong” and the notion of “ought not” cannot be erased.

Mill distinguishes something further implied in the claim that there is a rule of morality. He urges that no one claims that there is a rule of morality against a certain action without implying that it is a rule which ought in general to be observed.47 The claim that a certain action is contrary to a rule which ought in general to be observed implies that it is an action of a kind which in general ought not to be done. The former is a stronger claim than the latter. When someone has in mind actions of a certain description and believes that such actions in general ought not to be done, his belief implies that actions of that description are in general capable of being avoided. But his belief does not also imply that men in general are capable of understanding the description of action which he has in mind or that they are capable of avoiding such actions through having such a description in mind. On the other hand, whoever claims that a certain rule ought in general to be observed implies that men in general are capable of observing it. He therefore implies that actions of the kind covered by the rule are of a description which is intelligible to men generally, and that it is a description simple enough and precise enough so that men generally are capable of making out whether some action they are considering would accord with the rule. Consequently someone may be correct in claiming that actions of a certain sort ought not to be done, but not correct in claiming that a rule against them ought in general to be observed.

Mill maintains that two claims are made when it is asserted that it would be wrong to do a certain action: it is not only implied that it is an action of a kind which ought in general to be condemned; it is also implied that it would be contrary to a rule which ought in general to be observed. If such an assertion carried only these two implications, it would not be inconsistent for someone to hold that it would be wrong for him to do a certain thing but deny that he ought not to do it. Although Mill does not speak clearly on this matter, something he says in discussing the concept of justice is applicable. He points out that even though a man believed that a certain action was of a sort which in general would be unjust, he would not regard that particular action as unjust if he believed that it would not be wrong to do it (259). It may be presumed that Mill similarly holds that even if someone believed that a certain action was of a kind which in general would be wrong, he would still not think that it would be wrong to do it if he did not think that it ought not to be done. He then acknowledges that when it is asserted that it would be wrong to do a certain action it is implied that it ought not to be done.

Some thinkers hold that “ought” is ambiguous. They contend that when it is said that someone ought to do something, sometimes all that is asserted is that he has an obligation to do it, while at other times this is not implied.48 Mill notices that an action is spoken of as one that ought to be done both in contexts in which it is said that there is an obligation to do it and in contexts in which this would not also be said.49 But he does not accept the view that “ought” is ambiguous on this account. He maintains that what differentiates a context in which it is said that someone has an obligation is that something more is then asserted. Mill holds that when it is asserted that someone has an obligation to do a certain thing it is implied that this is so in virtue of the sort of action it is. He contends that this assertion also carries an implication as to adverse responses by others for failure to act: “We do not call anything wrong, unless we mean to imply that a person ought to be punished in some way or other for doing it. . . . It is a part of the notion of Duty in every one of its forms. . . .” (246.) From this passage it might seem that Mill regards the claim that there is an obligation to do a certain thing as equivalent to the claim that it would be wrong not to do it. Mill, however, mentions two respects in which these claims differ. He notices that one obligation may be overruled by another. When it is, it would not be wrong to fulfil it (259). Consequently the claim that someone has an obligation to do a certain thing implies rather that there is a presumption that it would be wrong for him not to do it. For the same reason, it implies not that he ought to do it but that there is a presumption that he ought to.

In the passage cited Mill continues: “It is a part of the notion of Duty in every one of its forms, that a person may rightfully be compelled to fulfil it. Duty is a thing which may be exacted from a person. . .” (246.) Exacted by whom? Mill distinguishes a perfect from an imperfect obligation according to whether there is some assignable person to whom a man is under an obligation (247). H. L. A. Hart contends that when it is asserted that one person, A, has an obligation to some assignable person, B, to do X, it is implied that it would be morally legitimate for B to compel A to do X, but not that it would be morally legitimate for others to compel A to do X.50 Mill does not agree that when it is asserted, for example, that a wife has certain obligations to her husband, it is implied that it would not be wrong for him to force her to fulfil them. He holds rather that when it is asserted that A has an obligation to some assignable person, B, to do X, it is implied that it would in general not be wrong for others to compel A to do X, but he does not hold that it is implied that there are certain assignable persons for whom it would not be wrong to exercise such compulsion. As an example of an imperfect obligation Mill mentions the obligation to be generous. Although Mill writes, “It is a part of the notion of Duty in every one of its forms, that a person may rightfully be compelled to fulfil it,” he later abandons this contention, and takes it rather as a distinguishing mark of a perfect obligation. For when it is said that someone has an obligation to be generous, Mill points out that it is not implied that it would not be wrong for others to force him to be generous. All that is implied is that there is a presumption that it would be wrong for him not to be generous.

Hand in hand with the question of what is claimed when someone is said to have an obligation is the question of what is claimed when someone is said to have a certain right. Some urge that sometimes when it is asserted that a man has a right to do something, all that is meant is that it would be right, that is, not wrong, for him to do it.51 Mill does not acknowledge that this assertion ever bears this sense, for when it is asserted that a man has a certain right, it is implied that his right is capable of being violated by others. What more is implied? One suggestion is that to assert that a man has a right to something is equivalent to saying that others ought not to deprive him of it. Mill does not accept this view. Someone may hold that motorists who are running out of gasoline ought to stop at the nearest service station and yet deny that the operators of service stations have a right to their patronage. Mill would hold that by denying that they have a right to such patronage, one is denying that such motorists ought to be compelled to give the nearest service station their patronage. He maintains that when it is claimed that a man has a right to a certain thing, it is implied that in general others ought to prevent anyone from depriving him of it (250). Mill also contends that it is not claimed that a man has a right to a certain thing unless it is implied that others have an obligation not to deprive him of it. But he does not hold that this claim implies that it would invariably be wrong for anyone to deprive him of it, for the obligation not to deprive him of it may be overruled by another obligation. Mill therefore holds that the claim that a man has a right to a certain thing implies rather that there is a presumption, that is, that in general, it would be wrong for anyone to deprive him of it.

Mill rejects the view that no one can have an obligation without another person having a right. He points out that when it is said that someone has an obligation to be generous, it is not implied that others have a right to his generosity. Mill certainly holds that the claim that a man has a right to a certain thing implies that others have an obligation not to deprive him of it. This he classifies as a perfect obligation: “duties of perfect obligation are those duties in virtue of which a correlative right resides in some person or persons. . .” (247). From this it may be thought that Mill holds that no one can have an obligation to an assignable person without the latter having a right. The ascription to Mill of such a view is not borne out by his own analysis, for he holds that the assertion that someone has an obligation not to deprive A of X implies that it would in general not be wrong for others to prevent him from depriving A of X. But Mill contends that the assertion that A has a right to X carries a stronger implication, namely, that others in general ought to prevent anyone from depriving A of X. If it would in general not be wrong for others to prevent anyone from depriving A of X, it does not follow that they also ought to.

Mill’s analysis of the concept of justice can readily be shown in relation to his analyses of the concepts that have just been considered. Here as hitherto the question is not what actions or sorts of actions Mill maintains are unjust, but what he holds is being said about an action when it is asserted that it would be unjust to do it. Mill makes five main points. First, he writes, “Justice implies something which it is . . . wrong not to do. . .” (247). Here Mill is maintaining that when it is asserted that it would be unjust for someone to do a certain thing, it is implied that it would be wrong for him to do it. It therefore implies whatever the latter implies. Accordingly, he states, “the idea of penal sanction . . . enters not only into the conception of injustice, but into that of any kind of wrong. We do not call anything wrong, unless we mean to imply that a person ought to be punished in some way or other for doing it; if not by law, by the opinion of his fellow creatures. . . .” (246.) Mill’s second point is that “Justice implies something which it is not only . . . wrong not to do, but which some individual person can claim from us as his moral right” (247). He here notices that not all actions regarded as wrong are also classified as unjust. Mill’s third point is that when it is asserted that it would be unjust for someone to do a certain thing, it is implied that if he were to do it he would be violating an obligation that he has to some other assignable person (247). Mill’s fourth point is that when it is asserted that it would be unjust for someone to do a certain thing, it is implied that he would thereby be depriving another person of something to which he has a right. Speaking of “this distinction . . . which exists between justice and the other obligations,” he writes, “justice, the term, . . . involve[s] the idea of a personal right . . . injustice . . . implies two things—a wrong done, and some assignable person who is wronged” (247). Here Mill urges that someone is not described as having done anything unjust if the wrong that he did was to other animals or to himself. He is not described as having done something unjust unless he is regarded as having done something wrong to another human being. When we think that it would be unjust for someone to do a certain thing, we imply that it would not in general be wrong for others to compel him not to do it. This implication is contained in Mill’s third point. He brings out a further implication of his fourth point when he writes: “When we think that a person is bound in justice to do a thing, it is an ordinary form of language to say, that he ought to be compelled to do it” (245).

Mill’s fifth and last point is contained in the statement, “Wherever there is a right, the case is one of justice. . .” (247). If Mill means by this that whenever someone is said to have a right to something it is implied that it would be unjust to deprive him of it, then this fifth point is not compatible with what he says elsewhere. Mill holds that when it is claimed that a man has a right to a certain thing, it is implied that in general it would be wrong for anyone to deprive him of it. But he acknowledges that the obligation not to deprive him of it may be overruled by other considerations: “to save a life, it may not only be allowable, but a duty, to steal, or to take by force, the necessary food or medicine, or to kidnap, and compel to officiate, the only qualified medical practitioner. In such cases . . . we usually say, not that justice must give way to some other moral principle, but that what is just in ordinary cases is, by reason of that other principle, not just in the particular case.” (259.) Mill hereby points out that the claim that a man has a right to a certain thing does not imply that it would invariably be wrong for anyone to deprive him of it. In this passage he also writes, “justice is a name for certain moral requirements . . . of more paramount obligation, than any others.” Mill would certainly agree that the obligation to do what is just is absolutely paramount over all other considerations. But he also points out in this passage that the respect in which it is paramount is that when someone believes that a certain particular action would be of a sort which in general is unjust, but also believes that it would not be wrong to do it, he would not say that it would be unjust but not wrong to do it. He would instead say that since it would not be wrong to do it, it would not be unjust to do it. Mill points out that no one regards a certain action as unjust unless he also regards it as wrong.

Having focussed on Mill’s analyses of four chief concepts—right and wrong, obligation, a right, justice—we have now to notice certain bearings of these analyses. Mill holds that when it is asserted that it would be wrong for a man to do a certain action, it is not only implied that he ought not to do it, it is also implied that it is contrary to a rule which ought in general to be observed and that it is an action of a kind which ought in general to be condemned. He contends that this is all that is implied. Asserting that it would be wrong to do a certain action is then a short-hand way of making three distinct ought statements in regard to it. The adjectives “right” and “wrong” could then be eliminated from language. It is useful to retain them as a short-hand way of making these three distinct ought statements at once. A similar point applies to the other three concepts. Mill maintains that when it is asserted that a man has an obligation to do a certain action, all that is implied is that it is an action of a kind which in general it would be wrong not to do; and that when it is also understood that he is under an obligation to some assignable person to do it, all that is implied in addition is that it would in general not be wrong for others to compel him to do it. Since these implications in turn are equivalent to a number of ought statements, the assertion that someone has an obligation to do a certain action is also a short-hand way of making several ought statements in regard to it. Mill also holds that when it is asserted that a man has a right to a certain thing, it is not only implied that others in general ought to prevent anyone from depriving him of it; it is also implied that it would in general be wrong for anyone to deprive him of it. The noun, “a right,” could then be eliminated from language by replacing it with the several ought statements to which it is equivalent. Finally, Mill maintains that when it is asserted that it would be unjust for a certain man to do a certain action, it is not only implied that it would be wrong for him to do it; all that is implied in addition is that if he were to do it he would be violating someone’s right. Since each of these implications in turn is equivalent to a number of ought statements, the adjectives “just” and “unjust” could be eliminated from language, but are useful to retain as short-hand devices for asserting a cluster of ought statements.

Mill errs in two respects in his analysis of the concept of justice. When it is claimed that a man has a right to worship in accord with the dictates of his own conscience, it is not implied that if someone were to prevent him from worshipping in this manner, he would be doing something unjust. Similarly, a man who tortures or murders another is not described as doing something unjust, even though it is held that he is doing another wrong and is doing something that others in general ought to prevent anyone from doing. Consequently, Mill is not correct in maintaining that a man is described as doing something unjust whenever he is regarded as doing something wrong and as violating another’s right. Sidgwick points out that Mill is also not correct in maintaining that whenever it is asserted that it would be unjust for someone to do a certain thing it is implied that others ought to compel him not to do it.52 When it is claimed that a father is unjust to one of his children, it is not implied that others ought to use compulsion to prevent him. Mill can hardly be blamed for falling short in analysis of the concept of justice where others generally have failed. Although he is mistaken as to the specific set of ought statements which he holds is implied by the claim that it would be unjust to do a certain thing, his mistake in this does not show that there is not some set of ought statements to which this claim is equivalent.

What emerges from Mill’s analyses is that there is a common element to assertions using the terms right and wrong, obligation, a right, just and unjust. He does not maintain that all these are but different ways of saying that a certain action ought or ought not to be done, or that they are not different from each other. He holds that each implies nothing but a number of ought statements. The correctness of each cannot be made out without making out whether what it implies is correct. Hence each can be made out to be correct if there is some answer in general as to what ought to be done and what ought not. To make out whether it would be wrong for a certain person to do a certain thing, it is not sufficient to make out that he ought not to do it. Mill holds that it also has to be made out that it would be contrary to a moral principle for him to do it. Mill contends that it would be contrary to a moral principle only if it would be contrary to a rule which ought in general to be observed. If there is a general answer as to what ought to be done, it can then be made out what rules ought generally to be observed and what sorts of actions ought in general to be condemned. Since the question whether it would be wrong for a certain action to be done is a question in part whether it would be contrary to a moral principle, the question whether it would be right or wrong for a certain action to be done is a question about the morality of it. Mill holds that since questions of whether it would be unjust for a man to do a certain thing, or of whether he has a certain obligation or a certain right, also carry implications about what it would be wrong to do, they also are moral questions. If there is a general answer as to what ought to be done, the answers to moral questions can be made out. Mill holds that if there is such a general answer, it will apply not only to moral questions, but also wherever the question of what ought to be done arises and where moral considerations do not.53

Mill seeks to bring out how it can be determined whether it would be wrong to do a certain action by analyzing what is implied when it is asserted that it would be wrong to do it. We might similarly expect him to grapple with the question of how the correctness of any ought statement can be determined by inquiring what is implied by any such statement. Instead of taking this course, he inquires if there is a test in general for what ought to be done. We have seen that Mill maintains that something ought to be done if and only if it would maximize happiness. In putting forth this principle, Mill does not claim that when it is asserted that something ought to be done, it is implied that it would maximize happiness; he claims rather that it provides a test. Mill’s analyses of assertions employing the concepts of wrong, obligation, a right and justice are logically independent of his claim as to what is the supreme test of what ought to be done. He holds that it is the supreme test of the correctness of such assertions because they are equivalent to sets of ought statements and it is the supreme test of ought statements generally: “if . . . happiness is the sole end of human action, and the promotion of it the test by which to judge of all human conduct . . . it necessarily follows that it must be the criterion of morality, since a part is included in the whole.”54

Two parts may be distinguished in Mill’s contention as to what provides a supreme test. Although he does not say it, it may be presumed that he holds, in the first place, that when it is asserted that something ought to be done, it is implied that its consequences would be intrinsically more desirable than those of any alternative. The second step is his contention that the test of whether the consequences of something would be intrinsically more desirable than those of any alternative is afforded by whether it would cause more happiness. This he derives from the more general contention that the supreme test of whether one state of affairs is intrinsically more desirable than another is whether it contains more happiness. We may notice the bearing of each step in turn on moral judgments. In accord with the first step, Mill holds not only that a man ought not to do a certain action if and only if some alternative would have more desirable consequences, but also that a certain rule ought in general to be observed if and only if the observance of it would in general have more desirable consequences than would failure to observe it. The first step implies also that actions of a certain sort ought in general to be condemned if and only if the condemnation of such actions would in general have more desirable consequences than the absence of such general condemnation. Accordingly, Mill maintains that it would in fact be wrong for a man to do a certain action if and only if three conditions are fulfilled: (1) some alternative would have more desirable consequences, (2) it would be contrary to a rule the observance of which would in general have more desirable consequences than would failure to observe it, and (3) it is an action of a kind the condemnation of which would in general have more desirable consequences than the absence of such general condemnation.

By virtue of the second step, Mill contends that the supreme test of whether some alternative to a certain action would have more desirable consequences is whether it would cause more happiness; and that the supreme test of whether the observance of a certain rule would have more desirable consequences is whether the observance of it would cause more happiness.55 He therefore maintains that it would be correct to claim that it would be wrong for a certain man to do a certain action if and only if three conditions are fulfilled: (1) some alternative to it would cause more happiness, (2) it would be contrary to a rule the observance of which would in general cause more happiness than would failure to observe it, and (3) it is an action of a kind the condemnation of which would in general cause more happiness than would the absence of such general condemnation. In like fashion the conditions can be spelled out which Mill implies must be fulfilled for anyone to be correct in claiming that a certain man has an obligation to do a certain thing, that he has a right to a certain thing, or that it would be unjust for him to do a certain thing.

V.

THE USE OF THE PRINCIPLE OF UTILITY

in maintaining that the supreme test of whether something ought to be done is whether it would maximize happiness, Mill does not hold that this is the only test which is used, or can be used, or ought to be used. He recognizes that many other tests are used and holds that others are often more suitable. He suggests, for example, that as a test of conduct, it is often helpful for a man to ask himself whether a morally perfect being would approve of it.56 In speaking of other tests as often more suitable, Mill claims that other ways are available for making out whether something ought to be done than by considering directly all the happiness and unhappiness it would cause and comparing this with all the happiness and unhappiness that would be caused by each alternative to it. Mill speaks of a “subordinate,” “intermediate,” or “secondary” principle as being employed when it is determined that something ought to be done not by reckoning with these considerations but by reckoning with some other feature of it.57 He contends that it is not even possible to make out the morality of a certain action without taking account of whether it accords with a rule of morality. But even when moral considerations do not arise, Mill recognizes that men usually make out what ought to be done, and he urges that it is usually suitable for them to make out what ought to be done not by means of the supreme principle but by some intermediate principle. He holds that some intermediate principle is also often more suitable for making out whether a certain rule ought in general to be observed and is such that infractions of it ought in general to be condemned. In what he maintains is the supreme test, Mill is making three claims: (1) that something ought to be done if and only if it would maximize happiness, (2) that the ultimate reason why something ought to be done is because it would maximize happiness, and (3) that other tests are sound or suitable only if they would yield results compatible with it. In speaking of intermediate principles as “corollaries” of the supreme principle, he means that they are sound only if they yield results compatible with it.

There are many theories of morality which Mill rejects. He rejects the theory that what is meant by calling an action wrong or that the reason why an action is wrong is that it is the breaking of a divine commandment. He rejects such a theory even when it is united with a form of utilitarianism, as in Paley and Austin. He rejects the doctrine that any information about nature suffices to tell men what is right or wrong.58 He objects to Comte for contending that anything is wrong if done from some other motive than desire for the greatest happiness of humanity.59 He criticizes Bentham for not allowing that some experiences are more desirable than others independently of how pleasant they are. A further theory which Mill is particularly concerned to reject is what he calls the intuitive theory of morality.60 By it he understands the theory that it is intuitively self-evident what kinds of actions are wrong and what kinds are obligatory, and that all that is required to make out that some particular action would be wrong, or another obligatory, is to make out that it would be an action of some such kind. Mill does not deny that there is an intuitive character to the manner in which many moral judgments are made. Quite often someone thinks a particular action would be wrong because it is of a kind which he believes to be wrong. Not questioning the belief he is employing, a certain kind of action presents itself to his mind as wrong in itself (227). Mill would also agree with W. D. Ross’s remark, “When a plain man fulfils a promise . . . what makes him think it right to act in a certain way is the fact that he has promised to do so—that and, usually, nothing more. That his act will produce the best possible consequences is not his reason for calling it right.”61 But Mill would object that because the only thing that makes a man think that it would be wrong to do a certain action is the kind of action it is, it does not follow that the only reason why it would be wrong for him to do it is that it is an action of that kind. Because men often act upon a belief that actions of a certain kind are wrong, without reasoning further about it, he holds that it is not correct to infer that no reasons are to be given in behalf of such a belief and that certain kinds of actions are simply wrong in themselves.

Mill agrees with the intuitive theory that no one can make out by the principle of utility alone whether a certain action would be wrong or another obligatory. He would also point out that the principle of utility does not entail that men have but one obligation to others—to do what will cause most happiness. Because a certain action would cause most happiness it does not follow that it would not be wrong for others to compel it to be done. Moreover, the principle of utility does not entail that there is but one rule determining what is right or wrong. It does not imply that it would be wrong to do something if and only if it would cause less happiness than some alternative. If someone does something that will cause less happiness than would some alternative, it does not follow that he ought to be condemned by others for having done it. Far from maintaining that there is but one kind of action that is wrong, Mill holds that there are as many different kinds of wrong actions as there are rules which ought to be observed and ought to be enforced by moral sanctions. For determining in particular what is right or wrong, Mill contends that such rules are indispensable as subordinate principles. He also holds that such rules are often sufficient, no appeal to the principle of utility being called for.

When does Mill think that it is in place to appeal to the principle of utility to determine what it is right or wrong to do? He urges that someone is not warranted in believing that it would not be wrong to do a certain action simply because he is warranted in thinking that, considered by itself, it would cause more happiness. He writes, “though the consequences in the particular case might be beneficial—it would be unworthy of an intelligent agent not to be consciously aware that the action is of a class which, if practised generally, would be generally injurious. . . .”62 Here Mill speaks of appeal to the principle of utility to determine whether it would be wrong to do a particular action. Yet the appeal that is made is not to determine whether the particular action would have undesirable consequences but whether performance of actions of its kind would in general have undesirable consequences. Mill also holds that to be assured that it would be wrong to do a particular action, it is often sufficient for someone to think that it would be contrary to some rule which he believes ought generally to be observed, without testing on each occasion the correctness of the rule on which he is relying.63

A second sort of occasion on which Mill speaks of appeal to the principle of utility being called for is one in which someone is subject to conflicting rules. He writes, “only in . . . cases of conflict between secondary principles is it requisite that first principles should be appealed to” (226). Here appeal to utility is made to determine what particular action it would not be wrong to do. But Mill does not hold that when someone is faced with conflicting obligations, he can determine what it would not be wrong for him to do by disregarding his conflicting obligations and using the principle of utility to ascertain which action would have more desirable consequences. For whenever there is a question of whether it would be wrong to do a certain thing, the question of whether it would violate some rule remains. He holds rather that appeal to the principle of utility is called for to determine which obligation takes precedence. Yet Mill does not maintain that whenever there is a conflict of obligations such appeal is called for. He does not deny that such occasions recur and that men encounter them with their minds made up as to what kinds of obligation take precedence over others. They believe, for instance, that the obligation not to lie takes precedence in general over the obligation not to injure another, that the obligation not to injure another is more stringent than the obligation to help another, and that the obligation to help another who has helped one is greater than the obligation to benefit another who has not. Beliefs in rules of precedence such as these are second-order moral beliefs. Although he holds that it is often sufficient for men to resolve a conflict by means of such a belief, without appealing to the principle of utility, Mill urges that men cannot in the end be assured that they are correct in believing that one kind of obligation takes precedence in general over another without reckoning whether neglect of it would in general be more detrimental to human happiness than neglect of the other. Even where someone is correct in believing that one kind of obligation takes precedence in general over another, Mill urges that such a belief will not always suffice to enable him to resolve a conflict of obligations.

A third sort of occasion on which he speaks of appeal to the principle of utility being called for is one that presents an exception to a rule of precedence. He writes: “justice is a name for certain moral requirements . . . of more paramount obligation, than any others. . . . [P]articular cases may occur in which some other social duty is so important, as to overrule any one of the general maxims of justice. Thus, to save a life, it may not only be allowable, but a duty, to steal, or take by force, the necessary food or medicine, or to kidnap, and compel to officiate, the only qualified medical practitioner.” (259.)

Mill urges that all moralists recognize that every rule of morality admits of exceptions, and that there are occasions on which it would not be wrong to do a certain action even though it would violate a rule of morality. They thereby acknowledge that for it to be wrong to do a particular action, it is not sufficient that it be contrary to a rule of morality. Some further condition must be met. Mill points out that all moralists recognize that it would not be wrong for someone to do a certain action unless he also ought not to do it. Consequently if a certain action would violate a rule of morality, but it is not the case that it ought not to be done, it would then not be wrong to do it. Mill urges that where other moralists are at a loss is to state when this further condition is met. He not only affirms the principle that a certain action ought not to be done only if it would cause less happiness; he also speaks of appeal to this principle as called for to determine when to make an exception to a primary rule of morality, to determine when, for instance, it would not be wrong to steal, to lie, or to betray a solemn trust.

As an example of when it would not be wrong for someone to tell a certain lie, Mill cites an occasion in which “the withholding of some fact (as of information from a malefactor, or of bad news from a person dangerously ill) would preserve some one (especially a person other than oneself) from great and unmerited evil, and when the withholding can only be effected by denial” (223). Mill also points out that to be assured that it would not be wrong for a man to tell a certain lie, it is not sufficient to reckon with the “great evil” it would spare some person; against this must be weighed counter considerations.64 Account must be taken of the damage the lie may do in “weakening the trustworthiness of human assertion” and in undermining the benefits dependent upon it. Secondly, account must be taken of the damage the man’s lie may do in “weakening reliance” others will place on his veracity on future occasions. Third, account must be taken of the degree to which his readiness to lie upon one occasion may “enfeeble” his “sensitive feeling on the subject of veracity,” thereby making him less reluctant to lie on other occasions and further damaging his trustworthiness.

We have noticed four sorts of occasions which Mill speaks of as calling for appeal to the principle of utility on a moral question. In the first it is appealed to to determine whether a rule ought generally to be observed; in the second, to determine whether one kind of obligation takes precedence over another; in the third, to determine when to make an exception to such a rule of precedence; in the fourth, to determine when to make an exception to a primary rule of morality. The example Mill gives of the last is determining when it would not be wrong to tell a lie. Mill does not mention whether someone need ever reckon whether to violate a rule whose general observance and enforcement would cause more happiness, but which is not also generally observed and enforced by moral sanctions. Nor does he mention whether someone need reckon whether his action would conform to such a rule. Mill speaks of using the principle of utility to determine when to make an exception to a rule only if it is not merely a rule whose general observance and enforcement would cause more happiness, but is also a rule which is generally observed and enforced. The only considerations he mentions as to be taken into account against someone’s telling a certain lie are undermining reliance on his word, undermining his character, and impairing trust in men’s assertions generally. These considerations are relevant only in so far as the rule in question is one that is generally observed.

Of the four sorts of occasions for which Mill speaks of appeal to the principle of utility, he gives examples only of the third and fourth. These examples indicate how he expects such an appeal to be carried out. By the principle of utility, something ought to be done if and only if its consequences would be intrinsically more desirable than those of any alternative; and they would be intrinsically more desirable if and only if it would cause more happiness. A full use of this principle as a test therefore requires reckoning with all the alternatives, and with all the intrinsically desirable and undesirable consequences of each. An exclusive use of this principle as a test requires reckoning with nothing else. In the two examples Mill gives of appeal to the principle of utility, he mentions reckoning with but two alternatives—in one that of saving a certain person’s life or not saving it, in the other that of telling a certain lie or not telling it. A full use of the principle of utility requires reckoning with all intrinsically desirable and undesirable consequences to all sentient beings. In his two examples Mill does not speak of reckoning with consequences to other animals or to all human beings. Elsewhere he writes that in most cases in which someone appeals to the principle of utility “the interest or happiness of some few persons, is all he has to attend to” (220). Use of the principle of utility as a test requires reckoning only with intrinsically desirable and undesirable consequences—only happiness and unhappiness. Mill, however, mentions saving a person’s life as the only consequence to be reckoned with in the example he gives of breaking a rule of precedence. The only consequences he mentions to be reckoned with against someone’s telling a certain lie are undermining reliance on his word, undermining his character, and impairing trust in men’s assertions generally. He also reproaches Bentham for not including among consequences to be reckoned with effects of what a man does on his character.65 Yet Mill does not hold that the preservation of a man’s life is intrinsically desirable or that there is anything intrinsically undesirable about undermining character, about undermining reliance on a man’s word, or about impairing general trust in men’s assertions. He regards consequences such as these as undesirable only because they in turn would make for less happiness and he speaks of “weighing these conflicting utilities against one another” (223). Instead of a full use of the principle of utility, Mill would agree that reckoning with but a few alternatives and with but a few intrinsically desirable and undesirable consequences of each would be warranted if it would yield a result compatible with full use of the principle. It is not only this that Mill understands by appeal to utility. His examples show that he regards an appeal to utility as being made where what are reckoned with are other desirable and undesirable consequences than happiness and unhappiness.

Still greater latitude is to be observed in the argument which Mill holds is to be given for the desirability of men being compelled generally to observe certain rules. Among such rules he mentions those “which protect every individual from being harmed by others, either directly or by being hindered in his freedom of pursuing his own good,” which prevent anyone from “wrongfully withholding from” another “something which is his due,” or from depriving him “of some good which he had reasonable ground . . . for counting upon” (256). Although he speaks of such rules as grounded in “general utility” and as “more vital to human well-being than any” others, Mill does not feel called upon to show that use of compulsion to enforce them generally would make for more happiness than would absence of enforcement. He urges instead that men generally have such an intense interest in their enforcement that “if obedience to them were not the rule, and disobedience the exception, every one would see in every one else a probable enemy, against whom he must be perpetually guarding himself.” “It is their observance which alone preserves peace among human beings. . . .”66 He here argues that the enforcement of such rules is desirable because it is necessary to maintaining relationships among men which in turn are desirable because they are a necessary condition of men achieving to any degree anything desirable. Mill’s argument for the desirability of enforcing such rules is thus independent of any view as to what is intrinsically desirable, and therefore of his principle that happiness is the only thing intrinsically desirable.

Since he maintains that the principle of utility is the supreme test of conduct generally, Mill holds that it has application wherever anyone is pondering what to do, even though considerations of right and wrong are not or may not be involved.67 A man wonders, Should I change my job? Should I go to the mountains for my holiday? Should I invite the Jones for the evening? Should I put on a blue tie this morning? A business considers whether to reduce a certain line of investment. A plumber considers whether he should use copper piping. A municipality hesitates whether to resurface certain roads. Citizens discuss whether their country should reduce certain import tariffs, withdraw its troops from a troubled region, or increase its aid to another country. Mill holds that the answer to any such question is correct if and only if the course of action would maximize happiness. Although Mill is concerned to show that the principle of utility is the supreme test of conduct generally, in Utilitarianism he is largely occupied with its role in coping with moral problems. He has far less in general to say about its use in regard to other practical problems. Mill does not maintain that the only motive from which men act is interest in maximizing happiness or that the only principle by which they should test whether something should be done is by whether it would maximize happiness. A man may think that he should do something because he would enjoy doing it, because it is to his interest, because it would afford another enjoyment, because it would be impolite or unconventional not to. Mill does not deny the diversity of considerations employed in determining what an individual or a group should do. Sometimes a political policy is recommended to promote material prosperity, sometimes to promote progress, or freedom or enlightenment, or to relieve certain needs. Although the principle of utility is the supreme test, Mill urges that men cannot avoid using various subordinate principles for determining what should be done, even where questions of right and wrong are not involved. He writes, “all rational creatures go out upon the sea of life with their minds made up on the common questions of right and wrong, as well as on many of the far more difficult questions of wise and foolish. . . . Whatever . . . the fundamental principle . . . we require subordinate principles to apply it by. . . .” (225.)

Mill urges that the happiness of all is more effectively promoted by each pursuing his own happiness, subject to rules required by the good of others, than by each making the good of others his object.68 He also urges that each can more effectively promote his own happiness not by seeking it but by the active pursuit of ends beyond himself.69 Whether individuals or groups are engaged in farming, banking, teaching, medicine, or any other distinctive pursuit, Mill urges that it is usually sufficient for them to determine what they should do by reckoning only with what would most effectively promote the end of the pursuit. They are then called upon to consider only “that certain consequences follow from certain causes.”70 The conclusion that a certain thing should be done rests also, of course, on the assumption that the end is desirable. But “in various subordinate arts . . . there is seldom any visible necessity for justifying the end, since in general its desirableness is denied by nobody.” Mill mentions two errors to which the adoption of universal practical maxims in any pursuit is subject. One error is that of overlooking that the prescribed mode of action is effective only under certain circumstances. Quite another error is that of overlooking that though it is effective, its “success itself may conflict with some other end, which may possibly chance to be more desirable.”71 Where conflicting desirable ends are affected, Mill speaks of appeal to the principle of utility as called for. Yet for such appeal to be made, Mill does not require that only intrinsically desirable and undesirable consequences be reckoned with. Here too he regards an appeal to utility as being made where what are reckoned with are other desirable and undesirable consequences than happiness and unhappiness.

• University of Toronto

D. P. D.

[1 ]Utilitarianism, 237. Subsequent references are to the present edition of Utilitarianism, and are given in parenthesis.

[2 ]A System of Logic, 8th ed. (London: Longmans, Green, Reader, and Dyer, 1872), II, 552-4 (VI, xii, 6).

[3 ]“[H]appiness is the sole end of human action, and the promotion of it the test by which to judge of all human conduct; from whence it necessarily follows that it must be the criterion of morality, since a part is included in the whole.” Morality consists of “the rules . . . by the observance of which . . . [happiness] might be, to the greatest extent possible, secured. . . .” (Utilitarianism, 237, 214.)

[4 ]“[A]ctions are right in proportion as they tend to promote happiness, wrong as they tend to produce the reverse of happiness” (ibid., 210).

[5 ]Cf. G. E. Moore, Ethics (London: Oxford University Press, 1911), Chapter 1.

[6 ]See Utilitarianism, 214; “Sedgwick,” 69.

[7 ]“According to the Greatest Happiness Principle . . . the ultimate end . . . is an existence exempt as far as possible from pain, and as rich as possible in enjoyments, both in point of quantity and quality . . . ” (Utilitarianism, 214).

[8 ]See “Sedgwick,” 53; “Blakey,” 27; Utilitarianism, 222.

[9 ]See “The Utility of Religion,” 417.

[10 ]Mill comes closest to this when he writes: “The hygienic and medical arts assume, the one that the preservation of health, the other that the cure of disease, are fitting and desirable ends. These . . . propositions . . . do not assert that anything is, but enjoin or recommend that something should be. They are . . . expressed by the words ought or should be. . . .” (Logic, II, 552-3; VI, xii, 6.)

[11 ]See §§ IV and V, xcv-cxiii below.

[12 ]See §§ II and III, lxxiii-xcv below.

[13 ]G. E. Moore, Principia Ethica (Cambridge: Cambridge University Press, 1903), §40.

[14 ]“Bentham,” 96; “Remarks on Bentham’s Philosophy,” 112.

[15 ]D. D. Raphael, “Fallacies in and about Mill’s Utilitarianism,” Philosophy, 30 (1955), 344-57; E. W. Hall, “The ‘Proof’ of Utility in Bentham and Mill,” Ethics, 60 (1949), 1-18.

[16 ]Logic, II, 546 (VI, xii, 1).

[17 ]Ibid., 553 (VI, xii, 6)

[18 ]Raphael, 346.

[19 ]Raphael, 348.

[20 ]Aristotle, Nicomachean Ethics, trans. W. D. Ross (London: Oxford University Press, 1928), 1094a1.

[21 ]Ibid., 1094a17.

[22 ]Ibid., 1097a27.

[23 ]Logic, II, 552 (VI, xii, 6).

[24 ]Nicomachean Ethics, 1097a16.

[25 ]See “Nature,” 377.

[26 ]Ibid., 398.

[27 ]Letters of John Stuart Mill, ed. Hugh S. R. Elliot (London: Longmans, Green, 1910), II, 116 (to Henry Jones, 13/6/68).

[28 ]See More, Ethics, 167, 250, Principia Ethica, §121; C. I. Lewis, An Analysis of Knowledge and Valuation (La Salle, Illinois: Open Court, 1946), 486, 432.

[29 ]Principia Ethica, §42.

[30 ]Nicomachean Ethics, 1175b1.

[31 ]Raphael, 352.

[32 ]“Diary,” in Letters, ed. Elliot, II, 381 (23/3/54).

[33 ]Principia Ethica, §48.

[34 ]Ibid., §48.

[35 ]Anon., “Utilitarianism,” in J. O. Urmson, ed., The Concise Encyclopaedia of Western Philosophy and Philosophers (London: Hutchinson, 1960), 384.

[36 ]Anon., “John Stuart Mill,” ibid., 268.

[37 ]Principia Ethica, §43.

[38 ]Five Types of Ethical Theory (London: Kegan Paul, 1930), 190.

[39 ]Utilitarianism, 238; “Sedgwick,” 59.

[40 ]See lxxxii above.

[41 ]J. Rawls, “Justice as Fairness,” Philosophical Review, 67 (1958), 164-94.

[42 ]E. F. Carritt, Theory of Morals (London: Oxford University Press, 1928), 40.

[43 ]On Liberty (London: Parker, 1859), 14-15.

[44 ]Ibid., 21.

[45 ]See “The Utility of Religion,” 410.

[46 ]See “Whewell,” 184.

[47 ]Morality may be defined as “the rules and precepts for human conduct, by the observance of which an existence such as has been described might be, to the greatest extent possible, secured to all mankind. . .” (Utilitarianism, 214).

[48 ]H. A. Prichard, Moral Obligation (New York: Oxford University Press, 1950), 67.

[49 ]See Logic, II, 552-4 (VI, xii, 6). Cf. Letters, ed. Elliot, I, 229-31 (to W. G. Ward, 28/11/59).

[50 ]“Are There Any Natural Rights?” Philosophical Review, 54 (1955), 178, 183, 184.

[51 ]Ibid., 179.

[52 ]H. Sidgwick, The Methods of Ethics (London: Macmillan, 7th ed., 1907), 265.

[53 ]See Logic, II, 552-6 (VI, xii, 6-7).

[54 ]Utilitarianism, 237; cf. “Whewell,” 189.

[55 ]See Utilitarianism, 214; “Whewell,” 172; “Remarks on Bentham’s Philosophy,” 8.

[56 ]See “Theism,” 486.

[57 ]See Utilitarianism, 224-5; “Whewell,” 173; “Bentham,” 110; “Blakey,” 29 below.

[58 ]See “Nature,” 378.

[59 ]See Auguste Comte and Positivism, 335.

[60 ]See Utilitarianism, 206; “Whewell,” 170; “Sedgwick,” 51.

[61 ]The Right and the Good (Oxford: The Clarendon Press, 1930), 17.

[62 ]Utilitarianism, 220; cf. “Whewell,” 180.

[63 ]See Utilitarianism, 225; “Bentham,” 111.

[64 ]See Utilitarianism, 223; “Whewell,” 182.

[65 ]See “Bentham,” 98; “Remarks on Bentham’s Philosophy,” 8; “Sedgwick,” 56.

[66 ]Utilitarianism, 255; cf. “Whewell,” 192.

[67 ]See Logic, II, 552-6 (VI, xii, 6-7).

[68 ]Auguste Comte, 337.

[69 ]See Autobiography (New York: Columbia University Press, 1924), 100.

[70 ]Logic, II, 554 (VI, xii, 6).

[71 ]Ibid., 550 (VI, xii, 4).