1. The Absolutism Problem
Mill’s On Liberty is often said to be infected with an “absolutism problem” in that his utilitarianism is held to be incompatible with the absolutist status which he assigns to his liberty principle (LP). There is no doubt that standard utilitarian maximizing reasoning demands exceptions to any social rule that prescribes absolute liberty of self-regarding conduct, that is, conduct that does not directly cause any non-consensual harm to others. Total homogeneous utility such as pleasure or preference-satisfaction can be increased significantly by preventing competent people from choosing dangerous self-regarding actions that are very likely to cause severe self-harm, for example, or by forcing them to refrain from self-regarding actions that cause a great amount of dislike or displeasure but no perceptible damage to others.
It should be remarked that the absolutism problem occurs in the context of civilized societies, to which Mill confines the application of LP. He does not deny that there are exceptions to LP as a social rule. It does not apply in uncivilized or barbarian contexts: “Liberty, as a principle, has no application to any state of things anterior to the time when mankind have become capable of being improved by free and equal discussion” (On Liberty, CW. xviii., p. 224).Footnote 1 Even in civil societies, LP does not apply to incompetent individuals such as young children or delusional adults who are incapable of understanding the probable consequences of their own actions. So there is no conflict between the absolutist status he gives to LP in civil society and his repeated insistence that social rules always admit of exceptions. “It is not the fault of any ethical creed, but of the complicated nature of human affairs,” he says, “that rules of conduct cannot be so framed as to require no exceptions” (Utilitarianism, CW. x., p. 225).
But the absolutism problem must be addressed in the civil context, which I henceforth take for granted. Piers Norris Turner proposes perhaps the most plausible resolution so far. He defends a “competence view, according to which Mill’s absolute anti-paternalism is an upshot of his claim that, in any domain, decision-making authority in each and every case should go to those most competent or likely to make the best utilitarian calculation in that domain” (Reference Turner2013, p. 330, emphasis original). Mill allegedly sees LP as absolute because any ordinary competent individual, in his self-regarding domain, is always more competent than others are to decide what is best for his own good. The individual has far more information than anyone else does about his own circumstances and he is far more interested than society and government are in promoting his own happiness. This does not imply that the individual is always correct about what his own good consists in, Turner says, or about which self-regarding actions are best for promoting it: “the most competent party’s decision need not always be right (i.e. expedient) – in fact, it is very unlikely that it would be – but it will always be rightful” (Reference Turner2013, p. 331, emphasis original). “The rational adult has rightful authority, even though he may go wrong” (Reference Turner2013, p. 333). “The ex ante irrationality, in each and every case, of substituting the lesser expertise for the greater makes the individual’s decisional authority over his own good exceptionless or absolute” (Reference Turner2013, p. 333).
The question arises as to what sort of utilitarianism Turner attributes to Mill. Standard maximizing utilitarians cannot reasonably be expected to grant absolute decisional authority to the ordinary competent individual given that he is observed to make imprudent self-regarding decisions about his own good. Turner himself becomes ambivalent about the competence solution when he concedes that “it should nevertheless have been clear to [Mill] that ordinary rational adults are not always the most competent judges about their own good” (Reference Turner2013, p. 334). He apparently excuses Mill’s surprising blindness on grounds that he was not aware as we are of the disturbing evidence from behavioral psychology and economics that even competent people often delude themselves about what is best for their own good.
Yet Mill is clear that competent individuals do make foolish and even contemptible self-regarding choices. He does not need the evidence of behavioral psychology to recognize that fact: “Though doing no wrong to any one, a person may so act as to compel us to judge him, and feel to him, as a fool, or as a being of an inferior order” (Liberty, CW. xviii., p. 278). Even so, he insists that “neither one person, nor any number of persons, is warranted in saying to another human creature of ripe years that he shall not do with his life for his own benefit what he chooses to do” (Liberty, CW. xviii., p. 277).
Turner is correct to suggest that Mill opts for agent-relative evaluation instead of impartial social evaluation of outcomes within any competent individual’s self-regarding sphere. As Mill stresses, the individual “is the person most interested in his own well-being … while with respect to his own feelings and circumstances, the most ordinary man or woman has means of knowledge immeasurably surpassing those that can be possessed by any one else” (Liberty, CW. xviii., p. 277). But Turner goes astray when he claims that Mill favors giving absolute decisional authority to the agent because that individual is invariably in the best position ex ante to always or nearly always make more competent decisions about his own good than anyone else is. Mill is under no illusions about the extensive “folly” and “depravation of taste” exhibited by ordinary humans. He even lists the “self-regarding faults” which lead others to judge and feel that the agent is a fool or a pitiful person lacking in dignity and self-respect (Liberty, CW. xviii., pp. 278–279). So he does not doubt that other people are at times, perhaps often, more competent and would make better choices than the ordinary agent would about his own good.
Mill is apparently proposing an agent-relative permission, that is, a permission to choose outcomes that do not maximize total homogeneous utility. In other words, the individual right-holder is not obligated to choose self-regarding actions that produce outcomes that standard utilitarians consider best. Instead, within his self-regarding domain, the agent has permission to make sub-optimal decisions about his own good which then must have a corresponding impact on total collective good. As Mill puts it: “If a person possesses any tolerable amount of common sense and experience, his own mode of laying out his existence is the best, not because it is the best in itself, but because it is his own mode” (Liberty, CW. xviii., p. 270, emphasis added).
Moreover, because they have correlative duties not to coercively interfere with any right-holder’s self-regarding choices, others have agent-relative restrictions, that is, they are restricted from maximizing standard homogeneous utility by forcing the right-holder to make what they consider, and may correctly consider, better or optimal self-regarding choices about his own good.
It might be objected that a departure from standard maximizing utilitarianism is not necessary if we assume that competent people are prudent most of the time or that most of them rarely act as fools. In that case, collective happiness may reasonably be assumed to increase when competent people exercise equal rights of self-regarding liberty, despite numerical majorities being occasionally imprudent and minorities often so: the misery suffered from imprudence is arguably outweighed by the happiness enjoyed by most individuals most of the time. Even if the assumption is accepted, however, the objection cuts no ice against my interpretation of Mill. It overlooks the crucial distinction which he draws in his System of Logic between the terms “general happiness” and “collective happiness,” a distinction missed by so many of his critics such as G.E. Moore and the hordes in his train.
For general happiness to be maximized, Mill makes clear, the personal happiness of each and every competent individual member of the “indefinite multitude” comprising the general society must also be maximized, with the caveat that different individuals may have different capacities for the enjoyment of pleasure. But this requirement is certainly not satisfied under standard act utilitarianism (AU) or an extensionally equivalent indirect utilitarianism such as rule utilitarianism (RU), given that some competent people at least occasionally make foolish and degrading self-regarding choices that detract from happiness both personal and collective. If RU is conceived as distinct from AU and held to incorporate a binding social rule that distributes and sanctions equal rights to self-regarding liberty such that deviations from the rule are prohibited, however, then we have “rule worship,” which implies that the sum total of homogeneous utility is not maximized: people must obey the rules even when deviating from them would promote more happiness than following them does. There is no convincing support for any of the standard readings in his writings. He explicitly repudiates as too demanding any requirement that individuals are obligated to behave like saints by sacrificing their vital interests to promote the happiness of others; he rejects unwavering obedience to fixed rules because rules cannot anticipate all changes of circumstances and unexpected events; and he never says a word in favor of any non-utilitarian form of indirect consequentialism that seeks to achieve some goal other than general utility-maximization.
To appreciate Mill’s non-standard version of utilitarianism, commentators must carefully attend to his admonition that “[i}t is necessary to distinguish general from collective names [or terms]” (Logic, CW. vii., p. 28, emphasis original). As he explains: “A general name is one which can be predicated of each individual of a multitude; a collective name cannot be predicated of each separately, but only of all taken together.” He then illustrates the distinction:
“The 76th regiment of foot in the British army,” which is a collective name, is not a general but an individual name; for though it can be predicated of a multitude of individual soldiers taken jointly, it cannot be predicated of them severally. We may say Jones is a soldier, and Thompson is a soldier, and Smith is a soldier, but we cannot say Jones is the 76th regiment, and Smith is the 76th regiment. We can only say, Jones, and Thompson, and Smith, and Brown, and so forth (enumerating all the soldiers), are the 76th regiment … “The 76th regiment” is a collective name, but not a general one: “a regiment” is both a collective and a general name. General with respect to all individual regiments, of each of which separately it can be affirmed: collective with respect to the individual soldiers of whom any regiment is composed. (Logic, CW. vii., pp. 28–29)
So general happiness is a term that refers to the happiness of each person in the multitude making up the general community. It can be said to define the class of individuals each of whose personal happiness is of concern: “A class is the indefinite multitude of individuals denoted by a general name” (Logic, CW. vii., p. 28).Footnote 2 General happiness does not mean collective happiness, which is a term that only refers to aggregate happiness taken as a lump sum and does not apply to each person’s happiness taken separately.
A general name is also the name of every individual member of the class it denotes and “is said by logicians to be distributed, or taken distributively” (Logic, CW.vii., p. 86, emphasis original). In the proposition, all people are happy, the subject is a general name which is distributed because happiness is an attribute predicated of each individual person. For Mill, general happiness is maximized if and only if personal happiness is maximized for each individual relative to the individual’s capacity for enjoyment. Happiness means feelings of pleasure including relief from pain. These feelings are inferred to be intrinsically similar for all who experience them, although different persons may experience different quantities and qualities of pleasure and obtain them from different sources.
It is fair to ask what it even means to maximize general happiness as Mill understands the term. The answer turns on the fact that general happiness is a distributive concept that refers to nothing but the separate personal happinesses in the general society. There is no notion of a collective happiness which is not experienced by any one determinate individual, as is the case under standard utilitarianism. As a result, it is not necessary to add up the personal happinesses to calculate a sum total which is independent of the personal happinesses.Footnote 3 Moreover, when general happiness is maximized at a given time, there is no implication that each person’s happiness is maximized at the same level or even that each person is enjoying the same kinds and qualities of pleasures. But each person achieves the greatest happiness of which he is capable at that time in point of quantity and quality of pleasure. Much more remains to be said about each person’s sources of pleasure, as will emerge during the course of my argument.Footnote 4
The implications of the Millian distinction between general and collective happiness are far-reaching. But space limitations confine me to pointing out only a couple of the important implications. Given what he means by general happiness, for example, Mill’s unusual utilitarianism is immune from the familiar objection pressed by Rawls, Nozick, Dworkin, and others that utilitarianism does not take seriously the separateness of persons. While standard utilitarianism may be vulnerable, Mill’s non-standard version is not because it holds that general utility increases if and only if at least one person’s own happiness increases and no person’s own happiness decreases.Footnote 5
Another example relates to Sidgwick’s purported dualism of practical reason, according to which the universalistic moral reasoning in hedonistic utilitarianism comes into unresolvable conflict with the self-interested moral reasoning of egoistic hedonism. Again, Mill’s utilitarianism denies that such a fundamental cleavage exists in practical reason and holds instead that general happiness can be maximized in harmony with each person’s own happiness. But further discussion of these matters must await another occasion.
My present focus is on a plausible resolution of Mill’s alleged absolutism problem, and I have suggested that such a resolution is connected to his implicit endorsement of agent-relative permissions and restrictions. In light of this endorsement, he rejects standard maximizing utilitarianism, according to which the rightness or wrongness of actions depends only on the value of their downstream consequences. For him, certain actions and the liberty to take them are intrinsically valuable and right even if their consequences do not maximize standard homogeneous collective utility, and other actions and decisional processes are wrong even if their consequences do maximize standard collective utility.Footnote 6
2. Consequentializing
Mill’s doctrine can be clarified by turning to what some philosophers have come to call “consequentializing,” which involves, among other things, expanding the idea of an outcome.Footnote 7 The expanded outcome includes not only the downstream consequences that purist consequentialists like standard utilitarians focus on exclusively to make moral judgments of right and wrong. Also included are the actions or decisional processes that produce those consequences and are intuitively judged as intrinsically right or wrong regardless of their consequences by purist deontologists like the god Krishna in the Bhagavadgita and also Kant as often (though perhaps mistakenly) interpreted. So a distinction is drawn between narrow outcomes and broad outcomes or, in other words, between what Sen calls “culmination outcomes” and “comprehensive outcomes” (2009, pp. 208–221, esp. at pp. 215–217). While there are different understandings of what consequentializing is trying to accomplish, and as Schroeder (Reference Schroeder2017) has shown its proponents offer conflicting arguments in defense of it, the core purpose as I (and apparently Sen) understand it is to provide more reasonable and compelling evaluations of outcomes by working with comprehensive outcomes and integrating consequence-sensitive reasoning with insights about the intrinsic values of actions and decisional processes. The purpose, in short, is to transcend the fruitless opposition between purist consequentialists who insist on deriving the moral properties of actions from culmination outcomes alone, and purist deontologists who rely exclusively on intuitions about the intrinsic moral properties of actions independently of culmination outcomes.Footnote 8
Some may immediately object that consequentializing is incompatible with Mill’s philosophy because his utilitarianism cannot recognize a central element of the consequentializing approach, namely, the prominence it gives to evaluator relativism, according to which an individual’s ranking of outcomes depends on his own criteria of evaluation and need not reflect what is judged impartially best for him or best overall. This objection would make sense if Mill’s utilitarianism were properly interpreted as AU or an extensionally equivalent RU or even a non-utilitarian version of indirect consequentialism that does not seek to maximize standard collective utility. But it must be kept in mind that I am rejecting such interpretations, despite their prominence in the literature, precisely because they ignore Mill’s apparent commitment to agent-relative permissions and restrictions.
Given the meaning he assigns to general happiness, Mill’s extraordinary utilitarianism can accommodate evaluator relativity in the self-regarding domain. Each individual has the moral right to evaluate and choose among his self-regarding outcomes as he sees fit, based on his own criteria of evaluation. Notice that the meaning of liberty and that of a right are distinct. For Mill, liberty means “doing as we like,” “doing what one desires,” “act[ing] according to [one’s] own inclination and judgment,” and, he might have added, choosing a self-regarding outcome in accord with one’s own criteria of evaluation (Liberty, CW. xviii., pp. 226, 260, 294). By contrast, a right is a claim on others, who have a correlative duty to respect the substance of the claim (Utilitarianism, CW. x., p. 250). In the case at hand, the substance of the claim is self-regarding liberty and so others have a correlative duty not to coercively interfere with an individual’s self-regarding choices.
Self-regarding conduct itself is strictly not morally right action. But it can never be morally wrong because it does not directly cause any non-consensual harm to other people, keeping in mind that their mere dislike or disapproval is not harm to them, and thus it cannot violate any duties owed to them: “Self-regarding faults … are not properly immoralities” (Liberty, CW. xviii., p. 279). Nor can it violate any moral duty to self because, Mill insists, no such duties exist. So the liberty of choosing self-regarding actions in accord with one’s own criteria of evaluation is protected by absolute moral right, and thus implicates the moral sentiment of justice. This powerful sentiment, Mill says, grows up around the idea of justice as a social code that distributes and enforces equal rights for some group of social peers. Historically, the idea was initially constrained to apply to an aristocratic elite but it has gradually expanded toward a democratic ideal of equal rights for all, though the ideal remains distant and requires ongoing struggle for its achievement.Footnote 9
True, the individual cannot legitimately have sole authority to pick and choose which rights he will recognize for himself alone and deny to others. But no reasonable commentator suggests that evaluator relativity properly extends so far: all must be given equal basic rights, including in Mill’s view rights of self-regarding liberty. The meaning and enforcement of justice and right is a shared social concern and not something an individual can legitimately decide is his own exclusive concern. Individuals have a moral right to protest against social choices about justice, however, in an attempt to get them reformed.
I reiterate that Mill is not a standard utilitarian who aims to maximize a sum total of homogeneous utility and claims that an action, rule, or disposition is right if and only if it brings about a culmination outcome which achieves that aim. Rather, his utilitarianism has an extraordinary structure which can be clarified with the help of Sen’s notion of comprehensive outcomes, which is another central element of the consequentializing lens. Given that a comprehensive outcome includes the self-regarding actions that produce the downstream consequences which are the exclusive focus of a culmination outcome, social or political assessments of comprehensive outcomes can incorporate the intrinsic moral value of the liberty given by moral right to agents to choose self-regarding actions as they wish in accord with their own evaluation criteria. Such a move is not possible within standard maximizing utilitarianism and is not welcome in any case because the implicit agent-relative permissions and restrictions defeat achievement of the ultimate goal of maximizing collective utility. But the move facilitates discovery of how Mill’s non-standard maximizing utilitarianism achieves its distinct ultimate goal of maximizing general happiness. Consequentializing enables us to see that an absolute moral right of self-regarding liberty is compatible with consequence-sensitive reasoning. As Sen argues, “if consequential reasoning is used without the additional limitations imposed by the quite different requirements of [standard utilitarianism, namely] welfarism, position independence, and the overlooking of possible intrinsic value of instrumentally important variables [such as actions or the processes of choosing them], then the consequential approach can provide a sensitive as well as a robust structure for prescriptive thinking on such matters as rights and freedom” (Reference Sen1987, pp. 77–78).
Sen himself employs consequentializing (a term he does not use) to develop a complex non-utilitarian but consequence-sensitive pluralistic theory of justice and right in which, depending on situation and on who is making decisions, standard utility information plays a role but so does non-utility information about freedom and equality, both of which have plural aspects. Freedom, for example, includes rights and duties of non-interference, positive capabilities and functionings, and agent-relative processes of self-direction in personal matters. Reason, and ultimately democratic public reason, is said to be able in principle to assign relative weights to these diverse and “non-commensurable” sources of value, weights that may vary across different situations. Reasonable judgments about how to mitigate injustice and wrong in this or that situation do not presuppose discovery of some morally ideal state of affairs at the head of a complete social ordering of possible culmination outcomes, Sen insists, as pursued by standard utilitarians. His theory can rest content with partial (that is, incomplete) orderings of comprehensive outcomes such that society has no best outcome but instead must choose between plural “maximal” outcomes, where a maximal outcome is unbeaten by any other outcome but does not itself beat other maximal outcomes. All maximal outcomes are unranked against one another.Footnote 10
But my pointing to Sen is not to critically assess his consequentialized pluralistic theory of justice, so much as to show that consequentializing is a flexible lens that can be taken in various different directions.Footnote 11 I am concerned to use it to clarify the extraordinary structure and operation of the non-standard utilitarian theory which I attribute to Mill. He clearly argues that justice requires any competent individual to have absolute control over her self-regarding sphere so that she can live her own life as she wishes without coercive interference by others. He claims that every civil society’s rules of justice must in principle include for all competent adults an equal absolute right of self-regarding liberty. Consistently with this, the individual has an agent-relative permission to choose among self-regarding actions as she pleases even if her morally protected choices result in culmination outcomes that are sub-optimal and harmful in terms of her personal happiness and the collective happiness as conceived by standard maximizing utilitarians.
But how does consequentializing enable us to see that Mill’s utilitarianism achieves its maximizing goal, given that its agent-relative permissions and restrictions are incompatible with standard utilitarian reasoning? By working with comprehensive outcomes, Mill can recognize the intrinsic moral value of a competent individual’s liberty of choosing any self-regarding action she wishes according to her own evaluation criteria, and take account of that intrinsic value in addition to the value of the downstream effects directly caused by the self-regarding action. This is where his doctrine of higher pleasures is crucial: he maintains that the pleasant quality of the moral sentiment of justice and right is higher than that of any conflicting kinds of pleasure, where a difference in quality as I interpret it means an infinite difference so that no amount of lower pleasure can ever be equal in value as pleasure to even a bit of higher pleasure.
My claim that a higher pleasure is infinitely more valuable than a lower pleasure does not imply that the higher pleasure itself has infinite value as pleasure, or that any one of two particular higher pleasures has infinite value in relation to the other. The claim says only that the higher kind is absolutely more valuable as pleasure when compared to the lower kind; it takes for granted (as Mill insists) that any two distinct higher pleasures have finite values relative to one another in proportion to their relative quantities just as any two distinct lower pleasures do, where quantity is typically measured as the product of intensity and duration. But quantities of the different qualities of pleasure are not commensurable: any quantity of higher pleasure is infinitely more valuable as pleasure than any quantity of lower pleasure.
The claim is a comparative claim which establishes a hierarchy of different kinds or species of pleasant feelings within the family or genus of pleasant feelings. It does not assert that a higher pleasure is infinitely more valuable than anything outside the family of pleasures. It does not presume that everyone endorses hedonism. Anti-hedonists may not be impressed by the superior quality of higher pleasures as compared to lower ones. The success or failure of hedonism, whether as a psychological theory or as an ethical theory, is left open, to be established by other arguments.
According to Mill, infinite is not a limited magnitude and so is not a quantity or a number. It indicates an unlimited, ever-expanding, never-ending process. Humans can only comprehend, produce, consume, or use some finite amount of anything, including any kind of pleasure or value.Footnote 12
If X is infinitely or absolutely more valuable than Y, then X is strictly preferable to, or better than, Y. Indeed, X and Y are properly said to be incommensurable (as Mill in fact uses this term). Technically, this means that X and Y cannot be ranked vis-a-vis one another on a common scale of rational numbers, although it does not mean that they cannot be ranked at all. Instead, given that an unlimited difference is the source of the incommensurability, X and Y belong on different scales that are separated by an infinite gulf. They are in different dimensions of value. When he says in Utilitarianism (CW. x., p. 207) that the lower pleasure is “of small account” in comparison to the higher, Mill is best read as arguing that the lower pleasure is of infinitesimal value as pleasure when compared to the higher.Footnote 13
Anyone who protests that Mill never says that a higher pleasure is infinitely more valuable as pleasure than a lower kind merely betrays that they have not studied his Logic, Book 1, chapter 7. He is clear that my interpretation is what he has in mind:
Now, these classes [called natural kinds], distinguished by unknown multitudes of properties, and not solely by a few determinate ones – which are parted off from one another by an unfathomable chasm, instead of a mere ordinary ditch with a visible bottom – are the only classes which, by the Aristotelian logicians, were considered as genera or species. Differences which extended only to a certain property or properties, and there terminated, they considered as differences only in the accidents of things; but where any class differed from other things by an infinite series of things, known and unknown, they considered the distinction as one of kind, and spoke of it as being an essential difference, which is also one of the current meanings of that vague expression at the present day. (Logic, CW. vii., p. 123, emphasis original)
He endorses the view of the Aristotelian logicians (i.e., the medieval schoolmen) and adopts their terminology. It follows that a higher kind of pleasant feeling is infinitely, and thus absolutely and always, more valuable as pleasure than a lower kind, regardless of the quantities of the two kinds: no quantity of lower pleasure, however large, can ever be equal in value as pleasure to even a bit of higher pleasure.
Although he does not use the terminology of lexical rankings, Mill is in effect speaking of a lexical ranking of different kinds and qualities of pleasant feelings. Higher pleasures of the same quality are ranked vis-a-vis one another in terms of their relative quantities on a common scale that is discontinuous with – separated by an infinite gulf from – the common scale used to rank pleasures of a lower quality in terms of their amounts (intensities times durations). The different kinds of pleasures are defined over different sources.Footnote 14
More must be said to appreciate Mill’s concept of higher pleasures. For instance, there is no way that a simple sensation of pleasure can be transformed into a complex mental pleasure of higher quality merely through a process of increasing intensification and duration. Increases of quantity alone can never produce a change of quality. Instead, Mill’s idea is that a complex mental pleasure such as the moral sentiment of justice differs from the simple sensation by an infinite series of ingredients or properties, known and unknown, which include the sensations themselves, present, remembered, and/or anticipated, since they are needed to make the complex sentiment feel pleasant, including in that term relief from pain and suffering. We do not know, and cannot reasonably expect ever to know, all of the properties that distinguish the physical nervous system, which receives and conveys sensations into our consciousness independently of our higher mental faculties and will, from the higher faculties and will, which together construct the complex mental pleasures.
The pleasant quality of the moral sentiment is higher than that of the simple sensations of pleasure among its ingredients, Mill argues, because something akin to a chemical reaction occurs when the myriad ingredients interact to produce the whole new sentiment with its own emergent properties, analogous to the way hydrogen and oxygen interact to produce water under certain conditions. He initially illustrates that mental chemistry generates new sensations of white from interactions among sensations or impressions of the seven prismatic colors:
When many impressions or ideas are operating in the mind together, there sometimes takes place a process of a similar kind to chemical combination. When impressions have been so often experienced in conjunction, that each of them calls up readily and instantaneously the ideas of the whole group, those ideas sometimes melt and coalesce into one another, and appear not several ideas, but one; in the same manner as, when the seven prismatic colours are presented to the eye in rapid succession, the sensation produced is that of white. (Logic, CW. viii., p. 853)
This is not to say that white is a more valuable color than any prismatic color. It merely shows that mental chemistry is a phenomenon we experience. In fact, he acknowledges that recognition of such chemistry can be traced back to the writings of his father and Hartley, both of whom opined that all complex feelings, including sensations, ideas, desires, emotions, and volitions, are chemical combinations of simple elements. But no one before Mill hypothesized that higher qualities of pleasure are generated by such chemical reactions.
The higher quality of pleasure which is inseparably associated with the moral sentiment of justice is an emergent property of that complex sentiment (or emotion) as generated by the chemical reaction among its infinite series of ingredients, known and unknown. Mill discusses some of the main known ingredients of the moral sentiment in Utilitarianism (CW. x., pp. 240–259), including an “animal instinct” to retaliate against anyone who threatens to harm one’s kin which is “moralized” by reason and enlarged sympathy for others so that it is transformed into due social punishment directed solely against wrongdoers who violate others’ basic rights.Footnote 15 This moral kind of pleasure is also higher in quality than non-moral merely expedient mental pleasures. Although the latter are higher in quality than simple sensations of pleasure, the pleasant moral sentiments are distinguished by an inexhaustible series of attributes from mental pleasures of ordinary expediency. The chemical reaction that produces the higher moral pleasures involves infinitely more ingredients than the chemical reaction that produces a non-moral mental pleasure.Footnote 16
Mill’s hypothesis that higher pleasures are generated by such chemical reactions is wildly original. He admits that it remains a hypothesis and cannot yet be considered a sound induction from experience (Logic, CW. viii., pp. 854–856). It was rejected out of hand by Sidgwick and Moore who have been given de facto support by the silence of other ethical thinkers in their wake. But there is no doubt that it is Mill’s hypothesis. Moreover, no proof of its falsity has ever been provided and, as he illustrates in the Logic, the laws of chemistry show that such mental chemistry is at least conceivable. As far as I am aware, the controversial intuitions of his critics are the only rationale given for its alleged implausibility, and these certainly do not constitute proof.
Although I cannot go into much detail here, the metaethical implications of this mental chemistry are of signal importance. Among the known components of the moral sentiment of justice as Mill conceives it are beliefs and desires or conations. There is the belief that humans have shared vital interests such as life, self-regarding liberty, individuality, ownership, and promises that are essential to well-being. There is the desire for society to protect these interests. A related belief is that protection ought to be provided by the distribution and enforcement of rights and duties, ideally equal for all. Another desire is to punish wrongdoers who violate the rights of others, a desire which, Mill claims, is rooted in the animal instinct to retaliate against those who threaten to harm oneself or one’s kin and close associates. Given that these beliefs and desires are fused in the moral sentiment, this moral emotion has a cognitive element insofar as the beliefs can be judged true or false, as well as a non-cognitive or expressive element which is not truth-evaluable per se but provides prima facie (and thus defeasible) reason to act in accord with moral judgments assessed as true.Footnote 17
For Mill, then, a moral agent, who is competently acquainted with the higher pleasure of the moral sentiments, makes truth-evaluable moral judgments of right and wrong and necessarily has motivation to act as required by his all-things-considered judgment that an action is a moral obligation. But people who are not moral agents, who have not developed their higher faculties and will sufficiently to appreciate the beliefs and desires which are ingredients of the higher pleasure of justice, cannot be expected to fulfill moral obligations, even if aware of them.Footnote 18
As interpreted, the higher pleasures doctrine is invulnerable to the oft-repeated objection that some value other than pleasure must be used to distinguish between different natural kinds and qualities of pleasure. Moreover, Sidgwick is plainly incorrect to imply that such qualitative distinctions render hedonistic utilitarianism incoherent “since it is hard to see in what sense a man who of two alternative pleasures chooses the less pleasant on the ground of its superiority in quality can be affirmed to take ‘greatest happiness’ or pleasure as his standard of preference” (Reference Sidgwick1988, p. 247). For any quantity of higher pleasure, however small, is a greater pleasure in comparison to any quantity of lower pleasure, however large. The pleasure of superior quality is infinitely more valuable as pleasure than the pleasure of inferior quality: the difference between them is unlimited.
Given his view that the pleasant quality of the moral sentiment of justice is superior to that of any conflicting kind of pleasant feeling, Mill is saying that the loss of higher pleasure from even a single violation of an individual’s moral rights cannot be compensated by any amount of lower pleasure of ordinary expediency. Moreover, the right of self-regarding liberty is absolute and cannot be overridden even by other rights. Not only can this right never be outweighed by considerations of mere general expediency but also the self-regarding conduct which the right protects can never be legitimately interfered with by society or government. Interference with the agent’s self-regarding actions, which she is entitled to choose as she pleases in accord with her own agent-relative criteria of evaluation, is absolutely prohibited by justice. No matter how large the quantity of lower pleasure gained by others from coercively interfering with her self-regarding conduct, that lower pleasure is of infinitesimal value as pleasure in comparison to the higher pleasure. The pleasure of the moral sentiment of justice always outweighs any amount of lower pleasure gained from immorality.
I cannot here discuss all the nuances of my consequentialized interpretation of Mill’s liberal utilitarianism. Enough has been said to indicate what is so distinctive about it. It bypasses all of the ongoing debates over standard consequentialist readings of Mill. It is also solidly grounded in his texts, whereas the standard consequentialist readings (including standard indirect utilitarian and non-utilitarian ones) most certainly are not. The latter approaches do not recognize his crucial distinction between general happiness and collective happiness. So they fail to see that his extraordinary utilitarianism accommodates agent-relative permissions and restrictions in its distribution of absolute rights of self-regarding liberty, which entitle the individual agent to spontaneously make imprudent self-regarding choices in accord with his own evaluation criteria and prohibit other people from interfering. The standard consequentialist readings also ignore his distinction between higher and lower qualities of pleasure. So they fail to appreciate the importance of his claim that liberty of self-regarding choices is protected by absolute moral right. This moral protection of self-regarding liberty implicates the higher pleasure of the moral sentiment of justice.
Readers can begin to see the plausibility of my interpretation if they carefully study the passage in which he describes the ultimate end of utilitarianism as he conceives it: “The ultimate end, with reference to and for the sake of which all other things are desirable (whether we are considering our own good or that of other people), 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, CW. x., pp. 214, emphasis added). The phrases I’ve emphasized confirm that, for him, utilitarianism involves the maximization of each individual’s happiness as well as of the general happiness, and maximization of both quantity and quality of pleasure, including relief from pain.Footnote 19
3. Misunderstandings
Misunderstandings of my interpretation of Mill’s non-standard liberal utilitarianism are inevitable so it is advisable to try to head off a couple of the more likely ones. Space limitations preclude dealing with others here.
3.1. Security
Some will insist that Mill never says that the moral sentiment of justice is a pleasure, let alone a higher quality pleasure. He says instead that security, the interest involved in justice and the defense of rights, is “to every one’s feelings the most vital of all interests” (Utilitarianism, CW. x., p. 251). But he does not call this vital interest a pleasure. So, the story goes, security is a precondition of our enjoying any good but not a good in itself.
But Mill explicitly tells us that “interest” is just another name for happiness or utility, which for him (as a hedonist) means pleasure including relief from pain: “the happiness [of an individual],” he says, “may be called … speaking practically [the person’s] interest” (Utilitarianism, CW. x., p. 218). Consistently with this, the general interest is synonymous in practice with general happiness. Moreover, when explaining the peculiar feeling of the strength of the obligations of justice, he points to the “animal element, the thirst for retaliation” which, along with reason and sympathy for others, is an ingredient of the moral sentiment: “this thirst derives its intensity, as well as its moral justification, from the extraordinarily important and impressive kind of utility [pleasure] which is concerned” (Utilitarianism, CW. x., pp. 250–251, emphasis added).
There is no doubt that security itself is a higher pleasure. Mill says as much when he tells us that the higher pleasures include “pleasures of the intellect, of the feelings and imagination, and of the moral sentiments” (Utilitarianism, CW. x., p. 211, emphasis added). Indeed, he implies that this kind of pleasure is higher in quality than all other kinds of pleasures which conflict with it, such as those of ordinary expediency and naked physical sensations. This follows from his insistence that security is so vital for everyone that it takes on that “character of absoluteness, that apparent infinity, and incommensurability with all other considerations” (Utilitarianism, CW. x., p. 251, emphasis added).
This absolute priority of the moral pleasure, which includes relief from suffering caused by wrongdoers and justifies their due punishment, is repeatedly confirmed by Mill in his writings.Footnote 20 In his view, nothing is more important than fulfilling one’s moral duties if general happiness is to be maximized. Moral individuals, who must be competently acquainted with the different qualities of pleasure to have become moral, are not motivated to pursue lower pleasures that conflict with the higher pleasure of justice. The moral agent, who is not a narrowly selfish person but instead has developed a noble character in which the conscientious disposition to do right is prominent, prefers the higher pleasure of doing his duty, and suffers at even the thought of not doing it. It is hardly surprising that Mill gives priority to justice and moral right over all competing considerations. For instance, Rawls among many others does so as well. What is so remarkable is that Mill justifies this from within his hedonistic utilitarianism.
It is important not to confuse the higher moral pleasure of security with the non-moral kinds of pleasure we receive from our self-regarding actions which security makes possible. There is no moral pleasure involved in ordering or drinking a cup of coffee, for instance. The higher pleasure is the feeling of security which is produced by living in a civil society with a working legal system and associated political and social institutions which together recognize and enforce our basic rights, ideally for all of us. Mill insists that basic rights include the absolute right of self-regarding liberty. Given the protection of this right, we can spontaneously engage in all sorts of self-regarding activities like having a coffee or engaging in the fine arts. Self-regarding actions do not harm others without their consent and so cannot conflict with security by violating any of their rights. Any kinds of pleasures we get from these actions, including aesthetic and merely expedient pleasures, can consistently supplement the pleasure of security as elements of happiness.Footnote 21
I am focusing on self-regarding conduct but similar remarks apply to the non-moral kinds of pleasures received from social (or other-regarding) actions that directly cause non-consensual harm to others but do not violate their rights. For example, winners in a fair and open competition over a job cause the losers to suffer harm in the form of wasted exertions but such harm does not violate their rights (Liberty, CW. xviii., pp. 292–293). Praiseworthy supererogatory actions which save the lives of some but are unable to rescue others bring aesthetic pleasure; those who are not rescued suffer harm yet they have no right to be rescued if the rescue is not feasible.
Security is a variable. Its quantity depends on the content of the assigned rights and the extent of their assignment to the members of society. The quantity is maximized if and only if all of the vital human interests shared by the members are covered by the rights, and equal rights are assigned to all. Security is also a permanent pleasure as long as the institutional machinery required to maintain it is “kept in active play.” Keep in mind that general security is maximized only if each person’s own security is maximized.
At the same time, Mill argues that we may come as a result of habit to no longer actually feel the higher pleasure of security. In other words, the continual repetition of all sorts of actions in peace (i.e., without violation of our rights) may lead us to form habits which cause us to take security for granted and displace the feeling of pleasure involved in it. These habits may even cause moral agents not only to forget but also to deny that pleasure is ever involved: security comes to be seen as an end-in-itself, at which time Mill says it has become a concrete part of our (otherwise abstract) notion of happiness as pleasure.
Even if the institutional machinery that creates security is initially seen as a means or precondition to pleasure, Mill holds that what is originally a means to happiness can come to be viewed as a concrete part of happiness itself (Utilitarianism, CW. x., pp. 235–239). He illustrates the point with respect to virtue. Virtuous actions and dispositions, he says, which are initially endorsed as means to pleasure, eventually become inseparably associated with pleasure such that the element of pleasure is forgotten, so much so that a moral agent views virtue as an end-in-itself, a concrete part of his conception of good, and denies that virtue is originally a means to pleasure or that pleasure would be felt but for habit in displays of virtue.
3.2. Agent-relative restrictions
Scheffler (Reference Scheffler1994, pp. 80–81) and Heumer (Reference Huemer2011) argue that agent-relative restrictions are problematic because it is plausible to permit or even require an unjust action if that action can prevent even more serious injustices from happening. So it may seem that absolute rights of self-regarding liberty must be rejected because it is sometimes possible to produce more general happiness, including security, by violating them and interfering with self-regarding liberty.
The objection to agent-centered restrictions is reasonable with respect to what Mill calls the domain of social conduct that (unlike self-regarding conduct) directly causes non-consensual harm to others such as violations of their rights. Any right to engage in social conduct arguably must admit of exceptions when the social conduct causes, or even credibly risks causing, multiple violations of others’ similar rights or even a single violation of another’s more important rights. A right of private ownership cannot give the owner absolute control over his boat, for example, if the boat is needed to save the life of a swimmer who will otherwise drown. Even the right to life might permissibly be violated if the murder of an innocent bystander is necessary to prevent a wicked bomber from murdering thousands or to block an evil trolley driver from deliberately steering into a crowd on the track.
But the absolute right of self-regarding liberty is immune from criticisms of agent-relative restrictions. There is no paradox of deontology in the context of self-regarding conduct. Given that person A’s self-regarding conduct does not cause any non-consensual harm to others, interfering with A’s self-regarding liberty cannot protect any other person B from suffering harm that he otherwise would have suffered from A’s self-regarding conduct. Nor can such interference obstruct B (whether a tyrant or not) from violating the rights of others. B is not harmed by A’s self-regarding conduct anyway so interfering with A’s self-regarding liberty does not prevent or discourage B from violating the rights of others, including their rights of self-regarding liberty. B may say that he is displeased by A’s self-regarding conduct because it is disgusting or offensive to him. But such displeasure alone is not harm. If it is taken to justify interfering with A’s self-regarding liberty, then B has achieved his unjust goal with respect to A and nothing prevents B from going on to similarly violate the self-regarding liberty of C, D, E, and so forth. In short, B has successfully interfered with A’s self-regarding liberty even though nobody is harmed by A’s conduct and there cannot be, as a causal consequence, any possibility that B or anyone else will not similarly interfere with the liberty of multiple others.
Consider a case in which violation of A’s right of self-regarding liberty is recommended to prevent B, who dislikes but is not harmed by A’s self-regarding action of reading a book that defends atheism, from interfering with the similar self-regarding actions of multiple other persons. Is it not obvious that interfering with A’s liberty to read the book merely serves B’s unjust purpose and gives B an incentive to demand permission to engage in such interference with the self-regarding liberty of others? Unless society enforces A’s absolute right to read the book as he wishes, B and other wrongdoers like him have every incentive to meddle so as to censor atheistic books merely because the wrongdoers dislike or disapprove of such harmless activities. Instead of caving in to the wrongdoers and relaxing agent-relative restrictions in the self-regarding context, a civil society justifiably inflicts due punishment on the wrongdoers for even credibly threatening to violate any individual’s absolute right of self-regarding liberty.
Equal absolute rights of self-regarding liberty for all are what Steiner (Reference Steiner1994) calls a “compossible” set. Everyone can exercise these rights in harmony, without harming one another or violating any other rights which each person is assigned. Moreover, such rights to liberty are prescribed by Mill as a permanent fixture in any civil society’s code of justice. As bears repeating, he insists: “No society in which these [self-regarding] liberties are not, on the whole, respected, is free, whatever may be its form of government; and none is completely free in which they do not exist absolute and unqualified” (Liberty, CW. xviii., p. 226).
4. Resolution
Mill’s extraordinary liberal utilitarianism seeks to maximize personal happiness and general happiness simultaneously. There is far more to say about its structure and operation. But even at this stage its unusual nature is evident.
I have argued that Turner’s attribution to Mill of a “competence solution” to resolve the supposed absolutism problem is unconvincing. Turner argues that in Mill’s view the ordinary individual is in a privileged position to make the most competent decisions about her own happiness. But even Turner apparently concedes that the competence solution fails. The failure occurs because Mill was supposedly not aware of the advances in modern behavioral psychology which show how deluded even competent people can be when judging what is best for their own good. The implication is that the absolutism problem remains unresolved.
But Mill certainly did realize that even competent people make foolish and degrading choices about how to pursue their own happiness. He did not need the lessons of behavioral psychology and economics to see that. Yet he nevertheless defends absolute moral rights of self-regarding liberty, which are incompatible with the goal of standard maximizing utilitarianism.
My novel interpretation, which is rooted solidly in Mill’s texts, provides a plausible explanation of why there is no absolutism problem once we appreciate the remarkable structure and operation of his non-standard liberal utilitarianism. The consequentializing lens allows us to see that he views the intrinsic moral value of self-regarding liberty, which is protected by absolute moral right, as infinitely more valuable than the non-moral merely expedient value of the downstream consequences of the self-regarding actions chosen by ordinary individuals in accord with their own agent-relative evaluation criteria. The fact that ordinary individuals may make foolish and degrading self-regarding choices is absolutely outweighed by the individual’s right to use her own evaluation criteria to seek her personal happiness as she sees fit.
It is liberty that Mill values above all else in the self-regarding context, and this requires acceptance of the fact that a competent individual may use her liberty imprudently in her self-regarding sphere. The individual’s right to do so is correlative with the duty of others not to interfere with her self-regarding choices. My interpretation accounts for Mill’s reasoning, without imposing on him some standard form of consequentialism which has no basis in his writings.
Acknowledgments
The author is grateful to the journal’s Editorial Board and especially to the Responsible Editor for helpful advice and critical comments on earlier versions of this essay. The constructive remarks of two anonymous reviewers must also be gratefully acknowledged.
Open access
