Mill: A Bibliographical Essay by John Gray

The Traditional View of an Incoherent Mill ... A "Two Mills" Thesis?

The traditional interpretation pictures John Stuart Mill (1806-1873) as one of history's paradigmatic transitional thinkers. Situated uncertainly in a no-man's land between the rival intellectual traditions of nineteenth-century England, Mill in his writings displays no settled or coherent doctrine on social and political questions. In Mill's work, the received view contends, competing sympathies and commitments are the subject matter of an ultimately unsuccessful eclectic method. This alleged hodgepodge produces a brittle conceptual framework which quickly disintegrates under any sustained critical pressure. Thus, Mill's utilitarianism seems at odds with his values of self-development and individuality; his democratic loyalties are in a tug-of-war with his elitist dread of majority tyranny; and his allegiance to laissez-faire principles is compromised by his concessions toward the socialist currents of his day. Some exponents of this traditional view have gone so far as to claim to discern in Mill's writings an intellectual schizophrenia: the lineaments of "two Mills," each with a distinctive expression and a coherent message.

There is, unfortunately, little agreement in identifying and describing these "two Mills," so that the vast secondary literature on the younger Mill contains a bewildering variety of pictures of him as at once a radical libertarian and a cautious, conservative, Whig trimmer; a moral totalitarian and a questing, open-minded skeptic; an unreconstructed empiricist and a free-wheeling epistemological pluralist. Whether they detect two (or more) Mills in John Mill's writings, or deny the presence of any integrated personality in his work, advocates of the received view all share the assumption that the promise of unity was not, and perhaps could never have been fulfilled in Mill's philosophy. A distinguished statement of the received view is that of John Plamenatz when he says of Mill's Utilitarianism, (1861, 1863), his Liberty (1859), and his Considerations on Representative Government (1861) that "These three essays written by a sick man in his premature old age, exhibit all his defects as a thinker, his lack of clarity, his inconsistency, and his inability either to accept whole-heartedly or to reject the principles inherited from his father and from Bentham."1 Even Isaiah Berlin, one of Mill's more sympathetic interpreters, speaks of the "outdated psychology and lack of logical cogency" of On Liberty, and concludes that "Rigour in argument is not among Mill's accomplishments."2

The Revisionist View of Mill as a Consistent Thinker

It must be admitted at once that there is much in Mill's work and in his life that supports the standard interpretation. Mill's notorious ambivalence to the utilitarian intellectual tradition he inherited from his father and Jeremy Bentham; his receptive response to some aspects of a German Idealist conception of the mind which the conservative Coleridge transmitted to the English world; his many shifts of position and emphasis on the great issues of socialism, democracy, and private property; together with the still intensely controversial question of how important for the development of his thought was his relationship with Harriet Taylor - all these vacillations conspire to suggest the image of a man inwardly divided. Mill seems a man at once acutely sensitive to the limitations of the utilitarian world view (whose official exponent he remained) but yet unable to abandon it decisively.

In recent years, however, a wave of revisionist scholarship and interpretation has emerged, whose theme is that the judgment of J. S. Mill as a hopelessly muddled thinker may yet be ill-founded and certainly remains premature. This post-war revisionism argues that our assessment of Mill is distorted by an earlier generation of intellectual historians who caricatured the aims and doctrines of nineteenth-century English utilitarianism. Furthermore, our view of Mill has been badly obscured by the hasty and presumptuous judgment of Mill's substantive argument by the philosophers and social theorists of the late nineteenth and early twentieth centuries. If the revisionist scholars are on the right track, the work of the younger Mill may be a natural development of his utilitarian predecessors' achievements. Mill's writings may contain a subtle and complex body of doctrine which may not be internally inconsistent.

Let us look, then, at the dialogue between traditional and revisionist interpretations of Mill. How convincingly does each interpretation deal with Mill on liberty, utility and morality, on private property, socialism and democracy, and on the scope and prospects of a science of society?

The Traditional View of J.S. Mill on Liberty, Morality, and Utility

The traditional accounts of Mill's doctrine of the limits of state interference interpret his enterprise in On Liberty (l859) as the impossible but perennially attractive one of squaring the circle: that of grounding a theory of the priority of liberty (itself part of a more comprehensive theory of justice and moral rights) in a utilitarian ethic. Mill, indeed, is clearly aware that some of his readers will see his enterprise as wholly misconceived. Thus, in the essay on Utilitarianism (1861, 1863) discussing the utilitarian foundation of his theory of moral rights he concedes: "To have a right, then, is I conceive, to have something which society ought to defend me in the possession of. If the objector goes on to ask, why it ought? I can give him no other reason than general utility."3 But the traditional view insists that liberal utilitarianism is itself a weak, incoherent "reason," since it is an unstable compound of two incompatible elements: (1) a teleological or maximizing element, in which the only duty any man or any government ever has is to promote the greatest good, and (2) a deontological or "side-constraint" element in which individuals are recognized as possessing inviolable moral rights against unjust treatment by state or society. What if achieving the greatest social good seems to require sacrificing some individual? The incompatible elements in utilitarianism itself create this dilemma.

Stephen's Analysis of Utility as Antiliberty and Proauthority

By far the most formidable of Mill's nineteenth-century English critics, the jurist James Fitzjames Stephen, criticizes Mill precisely because in On Liberty he illegitimately attempts to derive liberal conclusions supporting individual rights and liberty from a utilitarian outlook. Stephen, himself an avowed utilitarian, saw utilitarianism as having a natural antiliberal, authoritarian implication. In his great book, Liberty, Equality and Fraternity (1873), surely one of the world's masterpieces of conservative political thought, Stephen argues against Mill: if the only thing that has intrinsic value for utilitarians is happiness, and we are bound to promote happiness by the most efficacious means, then a consistent utilitarian policy of social betterment will not be especially tender toward individual liberty. In its political agenda utilitarianism will grant no priority to the protection of the classical liberal freedoms. Mill's utilitarian ancestors, such as Hume and Bentham, agree with Stephen in ranking liberty as, in fact, only one (and not always the most important) among the means necessary to security and good government in promoting happiness. Stephen's most forceful objection to Mill at this point of his critique is that, if Mill is truly a utilitarian, then liberty can have no intrinsic or inviolable value whatever: its value or disvalue will depend wholly on its contingent consequences which, given the variety of human circumstances, will be complex. As Stephen puts it:

if the word 'liberty' has any definite sense attached to it, and if it is consistently used in that sense, it is almost impossible to make any true general assertion whatever about it, and quite impossible to regard it either as a good thing or a bad one. If, on the other hand, the word is used merely in a general popular way without attaching any distinct signification to it, it is easy to make almost any general assertion you please about it; but these assertions will be incapable of either proof or disproof as they will have no definite meaning. Thus the word is either a misleading appeal to passion, or else it embodies or rather hints at an exceedingly complicated assertion, the truth of which can be proved only by elaborate historical investigations."4

Traditional Critique of Mill's Utilitarianism: Its Unwarranted Optimism About Human Nature and Failure to Support the Priority of Liberty

It is Stephen's charge that, given a less charitable historical view of human nature than the one Mill endorses, utilitarian principles in many circumstances might very well dictate supporting the stability of a traditional society of hierarchy and authority. In other circumstances they would sanction even more regimented schemes, such as Jeremy Bentham's Panopticon prison (or, a later objector might add, the Webbs' admiration of Soviet Five-Year Plans and Stalin's collectivization program), which seek to promote aggregate social welfare by the morally monstrous expedient of inflicting great hardship on some or many members of a society. In modern terms, Stephen's argument might be reformulated in the following question: How can Mill as a utilitarian consistently object to the kind of authoritarian society depicted in Huxley's Brave New World or B.F. Skinner's Walden Two? Stephen's own intention was not, indeed, to give a utilitarian defense of the dystopian schemes of Bentham and his associates, but simply to affirm that nothing in the utilitarian tradition gave liberty any special importance, while much in human experience testified to the greater importance of security, order, and discipline as conditions of a happy life.

The traditional criticism of Mill's enterprise in On Liberty really has two prongs: (1) On the one hand, how can Mill possibly hope to defend what he calls "one very simple principle"5 of giving liberty a privileged place among political values by invoking considerations of utility alone? Several of the critics discussed in J.C. Rees's classic study of Mill and his Early Critics (1956) highlight the incongruity in Mill's libertarian enterprise of defending this utilitarian principle "as entitled to govern absolutely" restrictions of liberty by society or state. However, as an avowed utilitarian, Mill is already committed to utility as yielding an absolute principle for determining the limits of state interference. (2) On the other hand, Mill's critics insist that, even supposing a successful utilitarian proof for liberty's priority over other political goods, its validity would hinge entirely on the accuracy of our conjectures about the effects on man and society of a regime of liberty. Such a utilitarian argument for liberty, in other words, is permanently defeasible and reversible. It yields antilibertarian results whenever particular predictions of the utility of liberty (or the picture of human nature on which such predictions depend) can be undermined by empirical investigation and argument.

Fitzjames Stephen, like many of Mill's Victorian critics, asserts vehemently that the utilitarian proof will work only on the basis of a wildly optimistic assessment of the prudence and virtue of the average sensual man and of his real moral psychology. Mill's account of human psychology, Mill's critics insist, is excessively and narrowly intellectualist, neglecting the central role of passion, prejudice, and sheer moral perversity in human life. As the writer in the London Review (1859) observes, ". . . the truth is, that intellectual independence, however theoretically desirable, is practically unattainable in the vast majority of cases."6 Given this more somber view of human psychology, can free men be trusted to promote social utility?

Mill was defended against Stephen by disciples such as John Morley, Viscount of Blackburn, Liberal statesman, and editor of the Fortnightly Review (1861-1882), and by writers such as the positivist Frederic Harrison. But the general reaction to On Liberty was by no means so generally favorable as much secondhand intellectual history has led generations of students to suppose. Principled argumentative defense of the doctrine of On Liberty was, in fact, a minority position throughout most of nineteenth-century English thought and letters. Probably the best available study of the whole period, apart from Rees's book, is John Roach's essay, "Liberalism and the Victorian Intelligentsia." B.E. Lippincott's broader study of conservative and liberal thought in Victorian times, Victorian Critics of Democracy (1938), should also be consulted for its chapter on J.F. Stephen and its sensible treatment of the antidemocratic liberal and conservative reaction. F.W. Knickerbocker's Free Minds-John Morley and his Friends (1943) is also useful as a source for information on such Liberals as Frederic Harrison.

The Revisionist View of Mill on Liberty, Morality, and Utility

Much of the best recent work on Mill's liberalism asserts that critics have misconstrued both of Mill's central principles of utility and of liberty. Mill's views on utility and liberty can be properly stated only with terms and distinctions taken from his own general theory of human nature and of practical reasoning.

As stated in the crucially important writings of Ryan 7, Brown 8, Dryer 9, and Lyons 10, the revisionist position begins by clarifying Mill's utility principle. It is neither a classical aggregative (i.e., average utility) principle, or a substantive moral principle. Whatever their differences in other areas of Mill scholarship, the revisionists agree that Mill saw the principle of utility as a very abstract principle, specifying that happiness alone was valuable for its own sake. Happiness governed not just morality but all the areas of practice identified in the theory of the "Art of Life" expounded in Mill's System of Logic (1843).

The Utility Principle, the Art of Life, and the Theory of Morality

In his System of Logic, Mill speaks of the three departments of the Art of Life as being "Morality, Prudence or Policy, and Aesthetics; the Right, the Expedient, and the Beautiful or Noble, in human conduct as works." The doctrine of the Art of Life (now widely seen as incorporating one of Mill's most valuable, original, and neglected insights) distinguishes between judgments of a properly moral character and judgments which appraise actions (or human characters) in terms of their prudence or of their nobility. As Alan Ryan intimates in his path-breaking explorations of these aspects of Mill's thought, the arguments of Utilitarianism and of On Liberty presuppose an understanding of the Art of Life defended in the System of Logic (1843). The plausibility of the substantive doctrines defended in these two essays thus depends in part upon the cogency of the conceptual analysis in Mill's Logic. It is the argument of Utilitarianism that the principle of utility does not allow judgments about men's moral obligation or rights to be derived in any very direct way. Indeed, the subject matter of utility is not the moral rightness or wrongness of actions at all. Rather as an axiological principle specifying happiness as the only desirable end, quite distinct from any substantive moral principle, Mill's utility principle is conceived as "the test of all conduct." As the revisionists understand it, the utility principle does not impose on anyone a moral obligation to maximize utility, and it does not condemn as a moral wrong any failure to do so. It follows from this that a utilitarian is not necessarily inconsistent if he knowingly sacrifices some utility for the sake of an equitable distribution of the utility that remains.

If the utility principle does not condemn as a moral wrong any discussion to maximize utility, what claims does it make on action, and how is it related to morality?

First, in specifying happiness as the only intrinsic value, the utility principle entails that all reasons for or against any act, policy, or practice must relate to and weigh its contribution to happiness. The principle of utility actually entails another principle, invoked by Mill but not named by him, which (following Brown and Lyons) I shall call the "Principle of Expediency." An act (for example) is expedient if it brings about a net utility benefit, and it is maximally expedient if it brings about greater utility than any available alternative. An avowed utilitarian violates consistency if he knowingly acts inexpediently, but Mill's theory of morality and of moral obligation insists that the man who acts inexpediently need not thereby commit any moral wrong. As Mill puts it in Utilitarianism:

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, then by the opinion of his fellow-creatures; if not by opinion, by the reproaches of his own conscience. This seems the real turning point of the distinction between morality and simple expediency. It is part of the notion of Duty in every one of its forms, that a person may rightfully be compelled to fulfill it. Duty is a thing that may be exacted from a person, as one exacts a debt. Unless we think that it may be exacted from him, we do not call it his duty.11

Mill's theory of morality and of moral obligation has here two levels - one conceptual, the other substantive. At the conceptual level, Mill proposes that we judge something morally right or wrong, only if its performance can be enforced, and its omission punished. There is a necessary conceptual connection, according to Mill, between the idea of a moral judgment and the legitimacy of its enforcement. Contrary to countless interpreters and historians, then, Mill believes in the Enforcement of Morality. But the morality in question is not necessarily the popular or positive morality of prejudice and tradition, but rather the utilitarianly-sanctioned "critical" morality which is the subject matter of Utilitarianism and On Liberty.

How, then, can we know the area of morality and of moral obligation? First of all, by applying the Principle of Expediency to the question of enforcement and punishability. An act is morally right, not if it is maximally expedient that it be done, but only if it is maximally expedient that its performance be enforced by penalties for noncompliance. It is worth noting that this aspect of Mill's theory of morality shows that his theory is not a species of act-utilitarianism.12 Mill cannot be an act-utilitarian, since his theory explicitly denies that an act's being maximally expedient generates any moral reason to do it. Nor, contrary to an influential current of interpretation begun by Urmson,13 can Mill be regarded as any sort of rule-utilitarian. Firstly, Mill's principle of utility, like the principle of expediency which it entails, does not mention either acts or rules, and, in fact, applies to things apart from acts and rules. Also, an act may be morally wrong, provided it is maximally expedient for the agent to suffer the penalties of conscience from it (regardless of whether any rule exists or might exist whose violation would be similarly wrong). Mill's moral theory, in short, is not accurately described in the traditional terms of act- and rule-utilitarianism. It remains recognizably utilitarian, nonetheless, in virtue of its clearly teleological orientation.

What Constitutes Moral Wrong: Mill's Harm Principle or Principle of Liberty

We have seen that for Mill moral wrongs are to be distinguished from merely inexpedient actions, and that a necessary condition of something being morally wrong is that punishing it would be maximally expedient. We have yet to discover what, according to Mill, is in fact morally wrong, and we can do this only by looking at the relationship between liberty and morality developed in the essay On Liberty. For it is there that Mill states his famous principle of liberty, sometimes called the self-protection or noninterference principle. This principle of liberty stipulates

that the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of action of any of their number is self-protection. That the only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant.

We need to clarify several important points about Mill's statement of this principle of liberty. Mill clearly means that unless "harm to others" can be prevented, there is no reason at all for any limitation of liberty. As Brown puts it, "By giving this necessary condition for the existence of a reason for restriction, it rules out as irrelevant absolutely everything but the prevention of harm to others. This sharp and unequivocal denial," as Brown rightly continues, "is the cutting edge of Mill's essay."14

Among the reasons which Mill's liberty principle rules out as irrelevant to justifying liberty-limiting acts, policies, and institutions are: (1) paternalist considerations, reasons having to do with preventing a person from harming himself, or with forcing him to benefit himself, and (2) moralist reasons, reasons to do with the enforcement of the positive or popular moral sentiments of a person's community. Mill also dismisses as legitimate reasons for limiting liberty: (3) welfarist considerations, reasons that favor restricting a person's liberty for the benefit of others.

Mill's liberty principle is, at first sight at least, a very stringent test of the legitimacy of state interference, one which should appeal strongly to economic and civil libertarians. For it condemns as illegitimate any restriction of liberty by state or society which is not designed to prevent men from harming one another. And, further, taken together with Mill's principles about enforcing morality, it yields a substantive criterion or moral wrongness. An act (or whatever) is morally wrong, if and only if punishing it both would prevent harm to others and would be maximally expedient. (We must always remember here that "punishment," for Mill, includes the sanctions of public opinion and the goads of conscience as well as legal penalties.) Mill's doctrine of liberty claims that the requirements of morality will be maximally expedient if they are themselves minimalist: we maximize utility if we restrict morality to questions of harm-prevention.

Traditional Objections to Mill's Harm Principle

At this point in stating Mill's doctrine, however, we may profitably raise a number of traditional objections. What, after all, are we to understand by the expression "harm to others"? Judgments about harm are often controversial (think of recent debates about the harmful effects of hallucinogenic drugs): how can we resolve such controversies? Does "harm" designate damage only to a person or property, or is there a class of moral harms, or harms to character, which may legitimately affect the liberty principle? Again, does the liberty principle license us to restrict liberty only where the conduct affected causes or threatens harm to others? Or does the harm principle sanction restrictions of liberty in all cases where harm to others can thereby be prevented? Further, is there really a category of actions which harm only the agent himself but not others? Is there in fact a class of self-regarding acts, whose primary effects are on the agent himself? If not, if all acts affect others through their effects on the agent, then the class of acts protected by the liberty principle would seem to be empty. Finally, even supposing these difficulties are solvable, it is far from obvious that Mill's liberty principle in fact expands liberty in its operations. Making "harm to others" the only good reason for interference, far from curtailing the legitimate powers of the state, might (because we all harm each other all the time in so many ways) indefinitely augment them.

The Revisionists' Clarification of Mill's Harm or Liberty Principle

All these questions have much exercised Mill's traditional critics, and to deal with these difficulties the revisionists have advanced a range of more or less persuasive answers. By far the most common accusation against the doctrine of On Liberty has always been that Mill's principle of self-protection presupposes a distinction that we cannot intelligibly make between acts which are "self-regarding" (in that they affect only or primarily the agent himself), and acts which are "other-regarding." As Fitzjames Stephen puts it, with characteristic bluntness and clarity:

I think that the attempt to distinguish between self-regarding acts and acts which regard others, is like an attempt to distinguish between acts which happen in time and acts which happen in space. Every act happens at some time and in some place, and in like manner every act that we do either does or may affect both ourselves and others. I think, therefore, that the distinction (which, by the way, is not at all a common one) is altogether fallacious and unfounded. 15

One of Mill's early critics, Joseph Parker, in his John Stuart Mill on Liberty, A Critique (1865) makes a similar point about determining the range of application of the self-protection principle, when he asks how far Mill is prepared to stretch the concept of harm. If, as Mill thought, the state is justified in imposing compulsory education, and this is warranted in that it prevents "harm to others," what policy could not similarly be justified? In the same vein, Leslie Stephen, James Fitzjames Stephen's brother and biographer, makes substantially the same objection, when in the third volume of his great work, The English Utilitarians (1900), he declares that "It is ... the acceptance of this antithesis, put absolutely, the 'individual', as something natural on one side, and law, on the other side, as a bond imposed upon the society, which at every step hampers Mill's statement of any vital truths."16

Rees's Defense of Mill's Harm Principle: Interests and Rights

How do the revisionists try to rebut these objections? By far the most powerful and influential attempt to clarify self- and other-regarding acts is made by J.C. Rees in his well-known 1960 paper, "A Re-reading of Mill on Liberty."17 Rees distinguishes between actions that merely affect others and actions that affect others' interests, and gives massive textual support for the claim that Mill's working conception of harm in On Liberty is that of harm to interests. The crucial difficulty for this interpretation, however, is how Mill (or anyone else) is to know what are a man's interests. Might not a committed puritan claim that he had an 'interest' in the moral environment in which he and his children live, and hold the state might restrict the liberty of those libertines and deviants who threaten to harm or damage the moral environment? Rees's interpretation is clearly open to such an objection, since he emphasizes that interests "depend for their existence on social recognition and are closely connected with prevailing standards about the sort of behavior a man can legitimately expect from others."18 It is, in fact, in order to distinguish human interests from "arbitrary wishes, fleeting fancies or capricious demands,"19 that Rees stresses their dependence on norms and values which enjoy social recognition.

But, as Professor Richard Wollheim recognized, in Rees's interpretation Mill's liberty or self-protection principle becomes relativistic and conservative in character, and this cannot possibly accord with Mill's intentions. For on Rees's interpretation the boundaries of the self-regarding area will be relativistically determined by the currently dominant conception of interests, and the liberty principle will expand freedom only insofar as legal and social limitations on liberty lag behind changing, more restrictive conceptions of human interests.

D.G. Brown has argued persuasively 20 that we can avoid this relativization of Mill's liberty principle only if we construe Mill as understanding "interests" in a strictly naturalistic and prudential fashion. Rees himself considers this question further in a subsequent "Postscript"21 to his paper, where he emphasizes the relevance to On Liberty of certain passages in Utilitarianism. Brown's interpretation is further supported by the independent work of D.G. Long. In his highly relevant book Bentham on Liberty (1977), Long emphasizes that several of the crucial distinctions at work in On Liberty are variants of distinctions made by Bentham. And this is most obviously the case with Mill's distinction between self-regarding and other-regarding actions.

What complicates Brown's revisionist interpretation is that in On Liberty, as in Utilitarianism, Mill recognizes that some, but not all, interests are crucially relevant in determining the self-regarding area and thus in applying principles about liberty of action. When Mill in On Liberty demarcates the area of life in which we may be held accountable to society, he speaks not of determining what are a man's interests but of ascertaining his rights. "This conduct," he says, "consists in not injuring the interests of one another; or rather certain interests which, either by express legal provision or by tacit understanding, ought to be considered as rights."22 Here the test is not whether a man's interests have been damaged by other men, but whether his interests ought to be protected as rights. Mill does not think, then, that if a man has an interest, he thereby has any kind of right. His reference to "certain interests" suggests that only some interests can be grounds for rights, but which ones?

Utility and the Permanent Interests of Man as A Progressive Being

In the introductory chapter of On Liberty Mill relinquishes any support for his argument derivable from ideas of abstract right. Furthermore, he insists that he regards utility "as the ultimate appeal on all ethical questions; but" he goes on at once, "it must be utility in the largest sense, grounded on the permanent interests of a man as a progressive being."23 The difficulty here is pushed one stage further back, in that we now need a criterion for distinguishing between those interests of man that are transitory and those that are permanently his in virtue of his character as a progressive being. What is there in Mill's doctrine of liberty that answers this need for a criterion?

Much of the secondary literature surrounding On Liberty might lead a student of Mill's thought to suppose that his use of terms like "harm" and "interests" is hopelessly vague. Given the apparent deficiency in Mill's argument, his principle of self-protection might also seem practically useless. As I have already observed, the force of that principle disqualifies anything but harm-prevention as a test for restricting liberty. Paternalist, welfarist, and moralistic interventions, therefore, all fall under the general ban.

Human Interests, Moralism, and Paternalism

But are we always able to differentiate paternalistic reasons for interference from moralistic ones? Is there, indeed, any determinate area in which paternalism is at all an issue? The controversies surrounding "moral offenses" suggest that judgments both about what is in a man's interests and about the general interest, have an inescapably controversial aspect. Professor Basil Mitchell shows this inherently debatable meaning of "interest" (while accurately reporting on the famous controversy between Lord Devlin and Professor H.L.A. Hart).24 The ambiguity of "interest" is evident in Hart's argument that much existing legislation that restricts liberty may be justified as protecting men's own interests by paternalistic, rather than moralistic reasonings. This argument, in other words, assumes that we can assess a man's interest without presupposing any evaluation of the worthiness or excellence of his way of life.

To put this logical situation in a later terminology,25 Hart (like Mill before him) can resist Devlin's and Fitzjames Stephen's argument that individual immorality is itself harmful to others by contending that "interests" designate a purely want-regarding concept, and by claiming that state interference can never rightly be ideal-regarding. Then the central claim of liberalism in Hart, as in Mill, is that the state in its liberty-restricting activities should be neutral between necessarily controversial competing ideals of human excellence. Mill's argument, indeed, is that since assessments of a man's excellence or nobility are not authentically moral evaluations at all, the liberal thesis that the state may properly enforce the requirements of critical or rational morality, and those alone, itself entails that the state may never coercively support one ideal of human excellence against its competitors. Mill differs from Hart, and lines up with later libertarians like Thomas Szasz,26 in his uncompromising opposition, not only to legal moralism, but also to state paternalism.

Man as a Progressive Being: Choice, Liberty, and the Psychology of Self-Realization

What is the nub of the revisionist interpretation? We can concede that these may well be "hard cases for the harm principle,"27 that is to say, cases where Mill's self-protection principle gives, at best, ambiguous guidance to action. But revisionists hold that Mill's theory of happiness and human nature is rich and dense enough to clarify how to apply the principle of liberty across a very wide area. The crucial point to recognize is that Mill's Aristotelian and Humboldtian conception of happiness had moved far enough away from old-fashioned psychological hedonism to allow considerations of individuality and self-realization to enter as constitutive ingredients into the idea of human happiness. It is the theory of the higher pleasures, as elaborated in Utilitarianism, that the exercise of the human capacities of choice, reflective thought, and active imagination is not just a means to human happiness, but a vital ingredient of it.

Mill further embeds this abstract and open-ended view of happiness in his characteristic theory of human nature as permanently capable of self-alteration and unpredictable self-transformation. Mill embraces this view in On Liberty following such German writers as Schiller and Novalis (who were in close touch with Wilhelm von Humboldt when he was writing his libertarian classic Limits of State Action). This is the same view which Mill elaborates more explicitly in the seminal articles on Bentham (1838) and Coleridge (1840): it is a mistake to regard man as a natural object with fixed qualities and predetermined possibilities. Rather, man is to be conceived as a reflective and self-critical agent, actively engaged in the open-ended venture of exploring his own powers and the world that he has created for himself. What distinguishes man from the inhabitants of the animal kingdom, and gives him a special relationship with nature, is only his capacity for reflective thought and deliberate choice; but this is of capital importance. For, unlike that of an animal, the shape of a man's life is not ordained in advance by a repertoire of unalterable instincts, but is never less than the permanently revisable product of his own past thought and action. Man, unlike the animals, is a progressive being. But Mill never unreservedly took this to mean that moral improvement or social progress are inevitable features of the human prospect. Being a progressive being means that man's life is not bound by any fixed, unalterable natural endowment, but is rather the unforseeable product of men's choices and experiments upon themselves.

The Interests of a Progressive Being: Autonomy, Security, and Liberty

We are now in a better position to understand what Mill means, when he speaks of "the permanent interests of man as a progressive being." The permanent interests of any person are those that concern him or her as a chooser, a creature who fashions his or her life by provisionally endorsing but forever criticizing principles and policies. We can turn to the essay on Utilitarianism for further illumination on Mill's notion of interests. We find there that Mill regards security as man's least dispensable interest, the precondition of any valuable form of life. We may suppose that Mill understands by security, security of person and property. The theory of the higher pleasures,28 in turn, assures us that Mill believed that what was in a person's interests was a choice-environment undistorted by invasive social and legal controls. This freedom of choice is an indispensable condition of the kind of happy life that is distinctive of a person. It is clear that we can secure free choice only by the social and legal protection of an area of individual liberty.

The permanent or vital interests of persons, accordingly, are the interests they have in security and in liberty. These interests thus ground their moral rights. Damaging these interests constitutes, not just harm, but injustice. Mill's doctrine of liberty and utility, we may repeat, judges that morality is maximally expedient (and utility is itself maximized) when we maximize personal choice or liberty and minimize moral requirements. As a general rule, these moral requirements should be restricted to a prohibition of aggression and of injury to individual security and liberty. Mill believes we have no moral duty to benefit others, except in special circumstances as when a person freely chooses contractual obligations. This is surely a conclusion which should be welcome to all radical libertarians. One objection may be that the conclusion is somehow suspect because it depends on contingent assumptions about man and society. But this objection surely begs the questions whether any social philosophy can avoid such assumptions.

Mill on the Priority of Liberty as Autonomy: Laissez-faire, Private Property, and Socialism

I have argued, from the revisionist viewpoint, that Mill could consistently attach a priority to individual liberty in political and social life. Allowing liberty to be preeminent whenever background conditions of security and an acceptable level of culture were established,29 Mill could yet remain faithful to his overriding utilitarian commitment.

But what does Mill's commitment to liberty's priority mean in the intensely controversial areas of his view on laissez-faire, socialism, and private property? As a start to answering this difficult tangle of questions, we need to challenge the traditional view that Mill's working conception of liberty was a negative one.30 For, first, several of the fairly explicit definitions he gives of liberty commit him to a strongly positive libertarian standpoint.31 Secondly, although On Liberty indeed discusses the classical-liberal grounds and limits of justified coercion, that essay makes clear that Mill would regard any society which lacks conflicting modes of thought and life as failing to fit the ideal type of a society of free persons. Central to the argument of On Liberty, then, is the notion of the free person as having available to him a wide range of alternative lifestyles and modes of thought. Mill sees the free person as liberated from the yoke of custom and convention, from the conformist pressures of peer-groups as well as the legal penalties of law, in areas where harm to others is not an issue. This positive notion of freedom as autonomy informs all of Mill's writings on socialism and private property: It is related to the idea of the autonomous man defined in David Riesman's well-known sociological study of the nonautonomous or "other directed" person in modern society, The Lonely Crowd. The intellectual pedigree of freedom as autonomy extends back at least as far as de Tocqueville's writings on American democracy.

Mill on the Role of Government: Neither Laissez-faire nor Socialism

It is evident that the argument of On Liberty (1859) is a natural development of Mill's discussion of the proper province of government in his immensely influential Principles of Political Economy (1848). Mill never unreservedly endorsed the standard slogans of laissez-faire, and much of the time, indeed, he has engaged in criticizing them, sometimes misguidedly. We would, however, fundamentally misconceive of Mill's intellectual development imagining (as is sometimes still done) that Mill was intellectually seduced by Harriet Taylor from an orthodox laissez-faire position to something more akin to Fabian socialism. Mill's criticisms of the capitalist political economy of his day, though often misconceived, fundamentally differ from those of the socialists of his time and ours.

Before we can demarcate Mill's critique of capitalism from that of the socialist orthodoxies, we need to be clear about Mill's relations to the doctrine of laissez-faire by making a number of distinctions. In the Principles of Political Economy (1848), Mill distinguished between 'necessary' and 'optional' state functions, and divides 'optional' into two types: 'authoritative' and 'nonauthoritative'. Mill differed from stringent laissez-faire noninterventionists, and argued that it was completely inadequate to restrict state activity merely to the prevention of force and fraud. He concluded pragmatically that the range of necessary government functions, though certainly broader than supposed by many exponents of laissez-faire, could not be identified by any universal rule, save the simple and vague one: that we should permit governmental intervention only when the case of expediency is strong.32

Against interventionists, however, Mill makes a crucial distinction between the two mentioned types of 'optional' government interference, the 'authoritative' and the 'nonauthoritative'. Since the 'authoritative' comprehends interventions by sanction and legal prohibition, there is a strong presumption against it deriving from utility in the larger sense. There is, however, no such presumption against the 'nonauthoritative' interference which merely supplements and does not replace successful private initiative. Unlike nonauthoritative interference, which avoids all coercion beyond that involved in the exercise of the state's taxing power, authoritative interference involves the state as order-giver and tends to stultify the spirit of independence.

Thus Mill believed that the larger utilitarian considerations on the one hand supported noninterference, but on the other hand allowed the state a wide range of functions, when it is clear that private institutions cannot adequately supply certain desirable things (public goods, as we should call them today). In this way, the state might properly assume a share of responsibility for such items as poor relief, colonization, scientific research, and the financing of education. Mill's overall view, in fact, was that the preservation of individuality in the modern world could not be achieved by sticking to any very fixed rule, but demanded great centralization of information in the state, together with great diffusion of power and initiative throughout society.

Mill's Conception of Distributive Justice vs. Orthodox Socialism

If Mill's criticism of orthodox laissez-faire went so far, how did his "new political economy" differ from contemporary and later socialist orthodoxy? Pedro Schwartz shows in his important book, The New Political Economy of J.S. Mill (1972) that the major targets of Mill's critique are the maldistribution of property and an oppressive system of industrial organization. One of the main causes of the maldistribution of property, according to Mill, was the concentrations of fortunes facilitated by uninterrupted accumulation of wealth across the generations. Mill's remedy for this maldistribution, which he proposed in the first edition (1848) of the Principles, was the institution, not of an estates duty, but of what we would nowadays call an accessions or inheritance tax, to be levied on the recipient and not on the donor of capital. For Mill, the merit of such a tax was that, unlike other arrangements, it need not transfer wealth from private individuals to the state, since it was easily avoidable by the desirable expedient of dispersing one's wealth widely. Importantly, Mill favored a steeply progressive inheritance tax. This tax, though it would allow the transfer of a "modest competence," would destroy all great fortunes in a couple of generations.

Mill's support of progression in inheritance taxation contrasts sharply with his opposition to it in the taxation of income. A progressive income tax, he argued, was tantamount to "hanging a weight upon the swift to diminish the distance between them and the slow"; it was to impose a penalty on people for having worked harder and saved more than their neighbors, which is the same as "relieving the prodigal at the expense of the prudent." One explanation for this disparity in kinds of taxes lies in Mill's constant preoccupation with saving and his lifelong distaste for conspicuous consumption. These motives led him to express his support in principle for an expenditure tax before the Select Committee on Income and Property Tax of 1861.

Mill's Lockean Distributive Theory vs. Egalitarianism

Another deeper reason for his contrasting attitudes to income and inheritance taxes, one which I shall need to expand upon, is that Mill's conception of distributive justice was by origin a Lockean one. Although this Lockean position tended to make him favor a redistribution of property and of incomes, it had no specifically egalitarian complexion. Mill clearly avows the Lockean pedigree of his doctrine of property and distributive justice, when he gives a quasicanonical statement of the grounds and limits of property rights:

The institution of property, when limited to its essential elements, consists in the recognition, in each person, of a right to the exclusive disposal of what he or she have produced by their own exertions or received either by gift or fair agreement, without force or fraud, from those who produced it. The foundation of the whole is, the right of producers to what they themselves have produced.33

Statements such as this (which could easily be multiplied) open up a gulf between Mill's doctrine of property and that elaborated in the tradition of Hume, Bentham, and the elder Mill. The gap develops because Mill absorbed a Lockean, Ricardian labor theory of value, which he used to ground a theory of justice in property titles based on notions of desert. This labor theory of the acquisition of property rights explains why Mill always treated the ownership of land as a special case, in which the existence of permanent bequeathable property rights is least justifiable. Similarly, the labor theory of property accounts for his sustained interest in schemes for peasant proprietorship and his unremitting hostility to landlords. Again, it is a Lockean conviction that the marginal productivity of a man's labor is one good measure of his worth and one that should be encouraged. This conviction accounts for Mill's uncompromising defense of labor competition and his unrepentant support for the incentives of piece-work in increasing individual productivity. Mill's redistributionist proposals about inheritance also owe their rationale to another Lockean belief. In the market economy of his day, Mill lamented that "reward instead of being proportioned to labour and the abstinence of individuals, is almost in inverse ratio to it." The Lockean background for Mill's conception of distributive justice is recognized in Lawrence C. Becker's recent study, Property Rights (1977), which expands and criticizes Locke's own theory.

However, Mill's distributionism, that is to say his desire to distribute property on the basis of individual desert, has another source. This is the ill-judged and fatal methodological dichotomy he sought to make between laws of production and laws of distribution. As he famously puts it:

The laws and conditions of the production of wealth, partake of the character of physical truths. There is nothing optional, or arbitrary in them... this is not so with the distribution of wealth. That is a matter of human institution solely. The things once there, mankind, individually or collectively, can do with them as they like.34

This split between production versus distribution may be restated as follows: Somehow persons produce wealth through rather mechanical procedures without any options or choices on their part, and we need not examine their motivations or incentives in doing so. The only question that seems relevant to this approach is how society should choose to distribute the wealth that mysteriously appears. However, this approach gives little thought to the effects on producers of social schemes to redistribute the wealth they create. In fact such redistribution may discourage producers from producing their product.

In this disastrous dissociation of production and distribution, with its implicit "manna from heaven" view of how goods and commodities are produced and with its failure to treat capitalism as a unified system of both production and distribution, Mill propounds the central heresy of modern Social Democracy. For this misleading dichotomy of production and distribution sanctions the belief that productive and distributive arrangements of different sorts may promiscuously be mixed so as to realize some ideal or preferred pattern of distribution. This is a delusion that is justly assaulted both by Marxians and by such neo-Austrian economists as F.A. Hayek and Murray Rothbard. In this belief, Mill fostered a harmful tradition of social criticism of capitalism. We are only lately recovering from this belief's ill-effects in social theory and political practice. At the same time, all who are not exponents of natural rights theory will commend Mill for arguing that property rights are not things settled once and for all, deducible from some supposed axioms of ethics. Mill viewed property rights, no less than political institutions, as creatures of "time, place and circumstance," to be assessed and altered to harmonize with "the permanent interests of man as a progressive being."

Mill's Syndicalism vs. Authoritarian Socialism

Mill thus advanced contemporary Social Democracy with his erroneous notions about what constituted justice in distribution. But we should not suppose that his form of anticapitalism had much in common with that of the Fabian socialists who came after him. (Nor is there any strong evidence to support the received view that Mill's approach to socialism and private property, or to any other major issue, was substantially modified by the influence of Harriet Taylor. 35) It is true, however, that Mill was a lifelong opponent of one mode of capitalist industrial organization. He opposed those enterprises which are owned and managed by owners of capital who stand in an authoritarian relationship with wage-earners. He thought this became worse rather than better with the growth of joint-stock companies. He opposed it because, in the first place, he thought it institutionalized a permanent conflict of interests between capital-owners and wage-earners, and he doubted if any productive system which rested on such a basis could be either stable or efficient. Again, he supposed that the separations between wage-earners and owner-managers deprived workers of any real opportunity for personal initiative and precluded their becoming anything like the self-reliant individuals celebrated in On Liberty. Such objections to the capitalist system of his day led Mill to take a continuing interest in schemes for profit sharing, industrial partnership, and producer's cooperation. But his utopian views went far beyond such proposals and (as Lionel Robbins has suggested 36) can best be characterized as a form of nonrevolutionary competitive syndicalism. As Mill put it himself in his Principles:

The form of association ... which if mankind continue to improve must be expected in the end to predominate is not that which can exist between a capitalist as Chief, and work people without a voice in the management, but the association of labourers themselves on forms of equality, collectively owning the capital with which they carry on their operations and working under managers elected and removable by themselves.37

It is worth emphasizing that, while there are many objectionable aspects of Mill's syndicalist or non-state socialist utopias, it has no affinities whatever with the paternalist State celebrated in the Fabian socialist tradition. In Mill's posthumous Autobiography (1873), he certainly envisaged an economic order which was no longer recognizably that of nineteenth-century England, but it differs at least as much from our own interventionist economy. If Mill is in any sense a socialist then his was decidedly a "market socialism." He nowhere fatally compromises the core capitalist institutions of private property in the instruments of production and commodity production for competitive markets. Further, in considering the relations between Mill's position and the various socialist orthodoxies, we should note that, despite his iconoclastic sympathies with trade unionism, he envisaged no place for trade unions in the society of the future. He looked forward to a time when the harmony of interests between all partners in production, facilitated by workers' ownership and self-management, would allow "the true euthanasia of trades unionism."38

Mill on the Limits of Economic Growth: Its Harm to Individual Character and Social Values

Finally, Mill's thought significantly contrasts with his socialist posterity in his opposition to productivist conceptions of the good life. Like the other classical economists, Mill accepted that economic growth could only be temporary in a world of scarce natural resources in which population constantly pressed on land and food reserves. In contrast with all other economists in the classical tradition and in its socialist aftermath, however, Mill did not fear the arrival of a stationary economy, but rather welcomed it as an opportunity for a large-scale transformation in social values. Doubtless, a part of Mill's concern that society be re-ordered to allow for a peaceful transition to a no-growth economy derives from his neo-Malthusian insistence on the finitude of the world's resources and the ever-present danger of overpopulation. Yet Mill's advocacy of a stationary-state economy is largely concerned, not with considerations of resource-depletion, but with the damaging effects on human character of the unremitting pursuit of possessions and with the alleged destructive consequences for the natural environment of open-ended economic growth. In Mill's own emphatic words, in the chapter on "The Stationary State" in the Principles:

"I confess I am not charmed with the ideal of life held out by those who think that the normal state of human beings is that of struggling to get on; that the trampling, crushing, elbowing and treading on each other's heels, which form the existing type of social life, are the most desirable lot of mankind, or anything but the disagreeable symptoms of one of the phases of industrial progress."39

In words which show him to have moved altogether outside the Benthamite utilitarian tradition, Mill goes on to illustrate the harmful consequences for human character and development of an overcrowded world: "It is not good for man to be kept perforce at all times in the presence of his species. A world from which solitude is extirpated is a very poor ideal ... Nor is there much satisfaction in contemplating the world with nothing left to the spontaneous activity of nature." Concluding the chapter in his Principles with the search that "a stationary condition of capital and population implies no stationary state of human improvement," Mill effectively confirms his distance from the productivist central stream of classical economic thought and of its socialist aftermath. Clearly John Stuart Mill, at least among the great liberals, owed little - too little perhaps - to any culture of possessive individualism.40

We may well question the practical cogency of Mill's vision of a society of fraternal but competitive workers' cooperatives. No one who now reads the Principles can help reflecting that it became the standard economics textbook at a time when Britain was still only semi-industrialized. At this time the statification of the economy by interventionism was minimal and the joint-stock revolution had only recently got under way. It was an era when it was unthinkable that multinational corporations should arise possessing a discretionary authority often exceeding that of sovereign states. Further, we now know something of the problems of labor-managed economies (such as postwar Yugoslavia) resembling Mill's syndicalist utopia. What we know suggests their liability to debilitating influences, including especially an ineradicable disposition to an irrational allocation of labor. And, as both F.A. Hayek and Milton Friedman have had occasion to observe,41 Mill's distributionism, combined with his belittling of the achievements of technology, caused him to support the bizarre view that no further economic growth was needed in mid-nineteenth century England, but only a radical redistribution of its products. As Hayek has put it,42 Mill "appears to have been unaware that an attempt to cure even extensive poverty by redistribution would in his time have led to the destruction of what he regarded as cultured life without achieving its object."

Mill and Socialism

These defects in Mill's positive doctrine of a post-capitalist society are widely admitted in the relevant secondary literatures.43 It remains unquestionably the case, however, that a deep gulf separates Mill's idiosyncratic synthesis of laissez-faire with socialism from any subsequent socialist orthodoxy. If today, we have little to learn from Mill's political economy, still we ought, in intellectual honesty, to distinguish his errors from the even worse ones of his socialist rivals and heirs. Indeed, many socialists today might still benefit from reading Mill's posthumously published Chapters on Socialism, in which he prophetically exposed the dangers to individuality posed by a socialist economy.

John Stuart Mill's Idea of a Science of Society: One or Two Mills?

We now move on to one of the oldest, and most persuasive, traditionalist objections to the unity and coherence of Mill's social philosophy, and one which has furthered a number of "two Mills" theses. This objection focuses on a tension between Mill's view of mind and action, the tension between his theory of human nature (presupposed by Mill's liberalism) and that to which he explicitly commits himself in his "official" philosophical canon. Broadly speaking, traditional critics point to a tension between the empirical, more deterministic, and passive conception of human nature (defended, with several changes of emphasis, in Mill's 1843 System of Logic and in his 1865 Examination of Hamilton's Philosophy), and the view of the mind as free, active, and creatively ordering the raw data of experience. This second view seems presupposed by the argument of On Liberty, and Mill gestures towards it in such occasional pieces as his essay on "Two Varieties of Poetry." This traditional criticism of Mill is powerfully made by a nineteenth-century writer, Charles Douglas, in his John Stuart Mill (1895):

Because all improvement depends upon ideas, it must come from individuals; and the most real and secure improvement - that of men themselves - consists in their adoption of new and better ways of thinking.

Personality is thus, for Mill, at the very centre of human affairs. Human progress depends, not only upon rational conditions, but still more upon choice, and thought, and character and qualities of personal life. If Mill is committed by his presuppositions to another way of conceiving men's relation to the world, yet his assertion of the fundamental importance of personality forces itself through his empiricism, and modifies the strictness of the theory (pp. 177-178).

A very similar argument, contending that Mill's ideal of a free man commits him to a view of the mind as creative and ungoverned by causal laws, has been elaborated much more recently by J.W.N. Watkins in a lecture to the Royal Institute of Philosophy.44 Mill's views of the sovereign autonomous individual thus seems to impute to man a contracausal freedom of action which Mill's official empiricist philosophy denies.

Mill as a Transitional Thinker on Human Nature: Between Mechanism and Idealism

This, however, is only one aspect of the claim that two views of human nature compete and conflict in Mill's thought. At the most general level, such arguments raise the question of how Mill's moral and political philosophy is related to his theory of knowledge, and especially to his account of the scope and methods of a science of society.

The traditional interpreters are on firm ground when they claim that Mill's theory of human nature is a halfway house between the avowedly mechanistic account that Bentham and Mill's father developed, and the Idealist view defended by such later liberal thinkers as Bernard Bosanquet and T.H. Green. Several recent writers acknowledge that Mill strongly inclined to endorse the view (intimated in On Liberty and expounded in the writings of Wilhelm von Humboldt and Coleridge) that emphatically denied the constancy of human nature and constantly emphasized its liability to unpredictable metamorphosis. Richard Wollheim has declared that "Mill denied the uniformity of human nature. In doing so he rejected a belief that, implicitly or explicitly, has been central to the thought of the European Enlightenment, and thus by descent to classical Utilitarianism." R.J. Halliday, in his recent important book on Mill, sympathetically airs many revisionist claims and states: "Mill felt himself emancipated from simple psychological beliefs. Psychological hedonism, in particular, implied too neat and too narrow an account of motivation, there was no permanent human nature, to be explained by universal and invariant laws. ... Mankind were not alike in all times and places."45 Given Mill's methodological eclecticism, we must regard such claims as only a little less extravagant than Karl Popper's account of Mill which castigates him as an exponent of psychologism.46 The real situation is more complex, and suggests that the traditionalists are right in affirming that Mill never enunciated a coherent philosophy of human nature.

Mill's Science of Human Nature: Uniform Laws vs. Constancy

The key point to make here is twofold: (1) Mill largely did free himself from any belief in the constancy of human nature as always and everywhere moved by a small, tight-knit family of motives; but (2) he never decisively relinquished the empiricist project of a science of society, which must presuppose that human conduct is sufficiently uniform to be brought under law-like statements having both explanatory and predictive value. Thus, though Mill did indeed respond to Macaulay's famous attack on his father's Essay on Government 47 by repudiating the apriorism of the classical utilitarian approach, he never gave up the empiricist assumption that the way to render human conduct intelligible was to subsume its episodes under laws akin to those we formulate in the natural sciences. Some evidence may suggest that Mill believed the methods of inquiry appropriate to the study of human social life may qualitatively differ from those appropriate to the study of nature. But in his official philosophical corpus, Mill always adhered to a doctrine of methodological monism, to a thoroughly reductionist account of man and society. Though at times Mill's intellectual integrity and open-minded candor admitted bewilderment at the difficulties arising from the empiricist projects of a science of society, he never abandoned that project.

In order to critically evaluate the various traditional and revisionist accounts of Mill's project of a science of human nature and society, it is necessary to consider just how far Mill endorsed the classic empiricist aspiration to formulate a theory of human nature using principles and methods no different from those employed by natural scientists. To succeed, such aspiration presupposes that human behavior is subject to universal regularities which are culturally and historically invariant. This aspiration also assumes that in the human or moral sciences, as in the physical sciences, explanation and understanding consist in fitting observed behavior under a general formula or natural law. It was, after all, that most skeptical of British empiricists, David Hume, who wrote that "mankind is much the same in all times and places."48 Before Hume, Machiavelli had expressed in the Discourses a similar conviction of the constancy of human nature: "In all cities and in all peoples there are the same desires and the same passions as there always were. ... Everything that happens in the world at any time has a genuine resemblance to what happened in ancient times. This is because the agents who bring such things about are men, and men have, and always have had, the same passions from which it necessarily comes about that the same effects are produced."49

Now it is true that, in his philosophical writings, such as the System of Logic, Mill did occasionally insist that there are such things as laws of human nature, determinate and ascertainable: "the laws of the phenomena of society are, and can be, nothing but the actions and passion of human beings," he says, namely "the laws of individual human nature." Mill goes on to insist that men are not "when brought together converted into another kind of substance, with different properties." Similarly, he declares that "Human beings in society have no properties but those which are derived from, and may be resolved into, the laws of nature of the individual." Thus far, Mill does indeed seem to be endorsing a historical, psychologistic empiricism about the study of human conduct.

Human Nature: the Laws of Individual Psychology and the Cultural, Historical Context

Such an impression of Mill's "official" theory of human nature is seriously misleading, however, unless we severely qualify it. For Mill himself qualifies his assertion of the primacy of psychology among the social sciences with a reminder that it is necessary to grasp the historical context of human behavior if one is to understand it adequately: "as society proceeds in its development" he says "its phenomena are determined more and more, not by the simple tendencies of human nature, but by the accumulated influence of past generations over the present."50 Mill's effort in his System of Logic to develop an account of the nature and scope of social explanation can be seen to embody an unresolved (and, very probably, insoluble) contradiction between the psychologistic methodological individualism (or "science of human nature") he had inherited from the empiricist tradition, and the Comtean, historicist belief that "the fundamental problem of the social sciences [is to discover] the laws according to which any state of society produces the state which succeeds it and which takes its place."51 It is widely recognized, even by the most sympathetic among Mill's interpreters, that his attempt to synthesize a form of methodological individualism which was no longer narrowly psychologistic with an emphasis on the cultural and historical contexts in which human behavior occurs was not, and could never have been successful.

Perhaps the most powerful statement of the philosophical inadequacy of Mill's conception of explanation and understanding in human studies has been given by the Wittgensteinian philosopher, Peter Winch, in his extremely influential and controversial book, The Idea of a Social Science. Winch identifies the main weakness in Mill's philosophy not as its psychologistic tendencies, but more fundamentally, as its commitment to methodological individualism. This commitment is to a version of the "resolutive-compositive method" for which Newtonian mechanics (rather than the "geometrical" and "classical" methods he ascribed to his father and Macaulay respectively) was in Mill's view the appropriate model. Winch's argument against Mill is, no doubt, part of a polemical argument against empiricism and against methodological individualism in the social sciences generally. As such, it is very powerful. However, the inadequacy of Mill's "official" philosophy of human nature is not sufficient to establish his philosophy as inconsistent.

Mill's Compatibilism: Free Will and Determinism

Traditionally, the latter objection of internal inconsistency chiefly addresses Mill's reflections on the questions of free will and determinism. Mill's account, which renews an ancient compatibilist tradition, seeks to reconcile freedom and determinism, to show that any threat to the reality of choice posed by causal determinism of human actions is fraudulent. Mill contends that the consistency of determinism with freedom is, in the last resort, a pseudo-problem generated by a conflation of causal necessity with coercion.

Mill directs the main force of his argument against the Owenite, necessitarian or modified fatalist view (which he had found so oppressive during the period of his mental collapse). This modified fatalism asserts human actions are the unavoidable results of human character. The very features of human character are themselves necessitated by circumstances which each man inherits from nature, history, and society. Mill's rebuttal of the Owenite view is straightforward enough, consisting of the assertion (unexceptionable so far as it goes) that a man can alter his own character if only he wishes to do so by (for example) placing himself under the influence of circumstances other than those which gave it its current attributes. The objection to this argument is equally straightforward, namely, that the impulse to change one's character must itself in any coherent determinism be determined by one's constitution, history, and circumstances.

Such objections are, however, far from conclusive. Any attempt to show that Mill's philosophy, and his moral and political theory, flounders on the problem of free will, involves a program of substantive philosophical argument against compatibilism. This is an area of philosophy in which nothing like a consensus has yet been reached (and in which one is not yet visible on the philosophical horizon). On this issue, at any rate, the charge that Mill's philosophy lacks internal consistency must be given the Scottish verdict of "not proven."

A more problematic issue is that of the compatibility of the strongly fallibilistic theory of knowledge intimated in On Liberty with the inductivism defended in the System of Logic and throughout Mill's writings on epistemological writings. Paul Feyerabend has gone so far as to base one version of a "two Mills" thesis 52 on this tension, claiming that in On Liberty Mill embraces a form of epistemological pluralism, stronger than Popper's falsificationism, in which human knowledge grows simply by the proliferation of conjectures and world views. Such a theory of knowledge would certainly conflict with the more straight-forwardly accumulationist, inductive account offered elsewhere in Mill's writings. But it is also contradicted by much of what Mill says in On Liberty.

Mill as a Transitional Thinker: Between the Objectivist and Skeptical Traditions of Liberalism

Quite apart from the question of Feyerabend's fidelity to evidence about Mill's intentions in On Liberty - a question treated authoritatively by J.C. Rees 53 - there is an overwhelming plausibility about the claim, recently advanced by Professor Basil Mitchell, that Mill stood between two traditions in liberal thought. According to Mitchell,54 the two kinds of liberalism are distinguished chiefly by their account of the value of freedom. The "old" liberalism valued freedom because only in a free society could men have the chance to discover the truth about basic questions in morality and metaphysics. The new liberalism valued freedom precisely because there are no objective truths (at any rate in respect of evaluative and metaphysical questions). Again, according to Mitchell, the new liberalism is represented by Strawson, who in a well-known paper 55 justifies the freedom of individuals to realize a diversity of competing ideals of life within a framework of shared morality and law by arguing that no one of these ideals can be shown to be uniquely rational or even to be rationally preferable to other, well-formulated ideals.

Interestingly, though Mitchell follows most interpreters in claiming that Mill belongs to the tradition of "old liberalism" he goes on to acknowledge that "the seeds of the new liberalism" are to be found in Mill's defense of individuality. In this respect, at any rate, the traditional interpretation seems irresistible: throughout his adult life, Mill was poised in unstable equilibrium between a dogmatic, objectivist posture towards truth and validity in the areas of morality, metaphysics, and science - a posture he inherited from his father - and a skeptical outlook in all of these areas. Part of the fascination of Mill's liberalism derives from the spectacle of his agonizingly self-conscious attempts to reconcile these irresolvably antagonistic outlooks.

Traditional and Revisionist Interpretations - A Provisional Evaluation

Speaking of his period of mental crisis, and of the change in his opinions which it wrought, Mill declared: "If I am asked what system of political philosophy I substituted for that which, as a philosophy, I had abandoned, I answer, no system; only a conviction that the true system was something much more comprehensive than I had previously had any idea of."56 There can be little doubt that it is this self-critical and open-minded eclecticism of Mill's thought which has led many commentators, exasperated by the systematic elusiveness of his standpoint on the great philosophical and social issues of his time, to despair of finding any coherent view in his writings. Certainly, these are good grounds for the traditional interpretation in Mill's own many-sided intellectual development. It must even be conceded that, in all probability, the traditionalists are right in their contention that Mill never succeeded in welding the diverse intellectual traditions by which he was influenced into an integrated system. To this extent, the traditional interpretation must be upheld.

Several considerations emerge from the preceding discussion, however, which should cause us to moderate the severity of tone with which the traditional interpretation has often been accompanied.

(1) In the first place, while Mill's eclectic aspiration to synthesize the claims of utility and justice, laissez-faire and socialism, empiricism and a creative view of the mind may ultimately fail, his argument has been shown by the recent revisionist wave of Mill scholarship to be far more complex and subtle, far more acutely aware of obvious counterarguments, than exponents of the traditional view habitually allow. In some areas, indeed, it would be hasty and premature to suppose that Mill's reconciling purpose had been decisively defeated by progress in philosophical inquiry.

(2) Secondly, though we must not suppose that Mill is always clear-headed and consistent in argument, the power of his argument about the relations of utility, liberty, and moral rights should at least give pause to those who think intellectual traditions can be identified schematically by reference to some small group of dominating principles. Even if Mill's attempt to make peace between utilitarianism and justice does not in the end come off (and the issue must be regarded as still an open one) it does not do so because the idea of a utilitarian theory of moral rights is self-evidently absurd. Indeed, this is one area where Mill's eclectic method produces hopeful results.

(3) Finally, it should be recognized that the construction of an integrated and comprehensive philosophy was not one of Mill's major aspirations. The revisionist literature on Mill will have done us all a service if, in encouraging us to look with respect on Mill's work, it encourages us also to emulate that tolerance of uncertainty, and reverence for diversity, which is the distinctive feature of Mill's intellectual personality.

Endnotes

Full citations for works listed in the Endnotes may be found in the following Bibliography.

[1] J.P. Plamenatz, The English Utilitarians, p. 123.

[2] Isaiah Berlin, "John Stuart Mill and the Ends of Life," in Four Essays on Liberty, pp. 174 and 189.

[3] J.S. Mill, "Utilitarianism," in Utilitarianism, On Liberty and Considerations on Representative Government, p. 50.

[4] James Fitzjames Stephen, Liberty, Equality, Fraternity, p. 176.

[5] On Liberty (Everyman edition) p. 72.

[6] I owe this quotation to John C. Rees's admirable book, Mill and his Early Critics, p. 31. The most comprehensive and reliable general bibliography of writings on John Stuart Mill is that published in the Mill News Letter. One of the best nineteenth-century criticisms of Mill on liberty is to be found in the Norton Critical Edition of On Liberty, edited by David Spitz, pp. 123-142, reproduced from an anonymous paper in the National Review 8 (1859).

[7] Alan Ryan's main contributions are to be found in "Mr. McCloskey on Mill's Liberalism," Philosophical Quarterly 14 (1964); "John Stuart Mill's Art of Living," The Listener, October 21, 1965; The Philosophy of John Stuart Mill; John Stuart Mill; "John Stuart Mill and the Open Society," The Listener, May 17, 1973.

[8] Donald G. Brown, "Mill on Liberty and Morality," Philosophical Review 81 (1972): 133-158. Iam indebted also to Brown's papers on "What is Mill's Principle of Utility?" Canadian Journal of Philosophy 3 (1973): 1-12; "Mill's Act-Utilitarianism," Philosophical Quarterly 24 (1974): 67-68; "John Mill: John Rawls," Dialogue 12, 3 (1973): 1-3.

[9] J.P. Dryer's contribution, on which Brown draws in part, entitled "Mill's Utilitarianism" may be found in Essays on Ethics, Religion and Society, J.M. Robson, ed., Collected Works of John Stuart Mill, 1969.

[10] Lyon's principal contributions are: "J.S. Mill's Theory of Morality," Nous 10 (May 1976); "Human Rights and the General Welfare," Philosophy and Public Affairs 6 (Winter 1977); his books, Forms and Limits of Utilitarianism and especially his In the Interest of the Governed, a revisionist interpretation of Bentham's legal and political thought, are relevant to the interpretation of On Liberty.

[11] Utilitarianism (Everyman edition) p. 45.

[12] This point is made in a perceptive paper by Professor David Copp of Simon Fraser University, entitled "The Iterated-Utilitarianism of John Stuart Mill," and delivered to the Canadian Philosophical Association Congress, June 1978.

[13] For the argument that Mill is a rule-utilitarian, see J.O. Urmson, "The Interpretation of the Moral Philosophy of J.S. Mill," Philosophical Quarterly 3 (1953).

[14] Brown, "Mill on Liberty and Morality," p. 136.

[15] Fitzjames Stephen, Liberty, Equality, Fraternity, p. 28.

[16] Leslie Stephen, The English Utilitarians, Vol. III, p. 296.

[17] John C. Rees, "A Re-reading of Mill on Liberty," Political Studies 8 (1960), reprinted with an important "Postscript" (1966) in Limits of Liberty, Peter Radcliff, ed. Rees's papers "A Phase in the Development of Mill's Ideas on Liberty," Political Studies 6 (1958); "Was Mill for Liberty?" Political Studies 14 (1966); and, "The Thesis of the 'Two Mills,"' Political Studies 25 (1977), should also be consulted.

[18] Rees, in Radcliff, ed. Limits of Liberty, pp. 101-102.

[19] Rees, in Radcliff, pp. 101-102.

[20] See Brown, "Mill on Harm to Others' Interests," Political Studies.

[21] See Radcliff, Limits, pp. 106-107.

[22] On Liberty (Everyman edition) p. 132.

[23] On Liberty, p. 74.

[24] See Herbert L.A. Hart, Law, Liberty and Morality and Patrick Devlin, The Enforcement of Morals.

[25] The terminology of 'want-regarding' and 'ideal-regarding' principles derives from Brian Barry's book Political Argument.

[26] See Szasz's many publications on involuntary hospitalization as an infringement of human rights.

[27] I owe the expression, "hard cases for the harm principle," to Joel Feinberg, who uses it as the title for Chapter 3 of his excellent Social Philosophy where these matters are discussed.

[28] The theory of the higher pleasures is expounded by Mill in Utilitarianism, Chapter 2.

[29] Mill makes his qualifications to the range of application of his principles on p. 73 of On Liberty (Everyman edition).

[30] The terminology of 'positive' and 'negative' liberty is owed to Isaiah Berlin, who develops its sense in his "Two Concepts of Liberty," in Four Essays on Liberty.

[31] Thus, Mill observes in the System of Logic (London, 1974 edition, p. 841) "it is said with truth, that none but a person of confirmed truth is completely free." In An Examination of Sir William Hamilton's Philosophy (London: Longman's, 1865, p. 510) he speaks of "that normal preponderance of love of right, which the best moralists and theologians consider to constitute the true definition of freedom."

[32] See Vol. III Principles of Political Economy in the Collected Works of John Stuart Mill, edited by J.M. Robson and V.W. Bladen, Toronto, 1965, p. 804.

[33] See Mill's Principles (Toronto edition) pp. 754-755.

[34] See Principles, Bk. II, Chapter I: "Of Property," first paragraph, in Collected Works of John Stuart Mill.

[35] The claim that Harriet Taylor's influence decisively affected the development of Mill's thought is still alive in much recent work. It is discussed critically by H.O. Pappe in his valuable monograph John Stuart Mill and the Harriet Taylor Myth, which is reviewed by John C. Rees in Political Studies 10 (1962): pp. 198-202.

[36] See Lord Robbin's Introduction to Vol. IV Essays on Economics and Society in the Collected Works of John Stuart Mill, edited by J.M. Robson and Lord Robbins, Toronto, 1967, p. xi.

[37] See Principles, (Penguin edition) p. 133.

[38] This point is discussed in Pedro Schwartz's The New Political Economy of J.S. Mill, p. 103.

[39] See Principles (Penguin edition), p. 113 ff.

[40] My reference to possessive individualism is, of course, intended to designate C.B. Macpherson's ambitious ideological interpretation of liberalism in his The Political Theory of Possessive Individualism. He gives a more balanced view of liberalism, and an occasionally perceptive account of Mill, in his The Life and Times of Liberal Democracy.

[41] See Friedman's Capitalism and Freedom, p. 170; and Friedrich A. Hayek's The Constitution of Liberty, p. 430.

[42] Hayek, The Constitution of Liberty, p. 430.

[43] See for example, Alan Ryan, John Stuart Mill, Chapter 6, for a critical discussion of some of the difficulties in Mill's account of property and distribution.

[44] See J.W.N. Watkins, "Three Views Concerning Human Freedom," in Nature and Conduct, Royal Institute of Philosophy Lectures Vol. 8, London, 1974.

[45] The preceding quotation from Wollheim occurs in his Introduction to the World's Classics Edition of On Liberty, Representative Government and the Subjection of Women, London, 1975, p. xi. The quotation from Halliday comes from his John Stuart Mill, pp. 55-56.

[46] Popper's attack on Mill's "psychologism" occurs in Vol. 2 of his The Open Society and Its Enemies, Chapter 14.

[47] The controversy surrounding Macaulay's attack on James Mill's Essay on Government has been marvelously presented in Utilitarian Logic and Politics by Jack Lively and John C. Rees.

[48] See Hume's Treatise of Human Nature, Bk. I, Part 4, Section 6, for a development of this claim.

[49] This passage from Machiavelli is quoted by Stuart Hampshire in a paper relevant to Mill Studies, "Uncertainty in Politics," Encounter (January 1957).

[50] See System of Logic, Bk. VI, Chapter VIII, p. 583 (new edition, London, 1930) for this and the preceding quotation.

[51] See Mill's Against Comte and Positivism for a development of this claim.

[52] See especially the original version of P.K. Feyerabend, "Against Method" in Vol. 4 of Minnesota Studies in Philosophy of Science: Analyses of Theories and Methods of Physics and Psychology, p. 112.

[53] See Rees, "The Thesis of the 'Two Mills,'" Political Studies (1977).

[54] Mitchell, Law, Morality, and Religion.

[55] P.F. Strawson, "Social Morality and Individual Ideal," Philosophy (1961).

[56] J.S. Mill's Autobiography, p. 97.

Bibliography

August, Eugene. John Stuart Mill: A Mind at Large. New York: Charles Scribner's Sons, 1975.

Barry, Brian. Political Argument. London: Routledge, Kegan, and Paul, 1965.

Becker, Lawrence C. Property Rights: Philosophical Foundations. London: Routledge, Kegan, and Paul, 1977.

Berlin, Isaiah. Four Essays on Liberty. Oxford: Oxford University Press, 1969.

Brown, Donald G. "John Mill: John Rawls." Dialogue 12 (1973): 1-3.

_______. "Mill on Harm to Others' Interests." Political Studies 26 (1978): 395-399.

_______. "Mill on Liberty and Morality." Philosophical Review 81 (1972): 133-158.

_______. "Mill's Act-Utilitarianism." Philosophical Quarterly 24 (1974): 67-68.

_______. "What is Mill's Principle of Utility?" Canadian Journal of Philosophy 3 (1973): 1-12.

Devlin, Patrick. The Enforcement of Morals. Oxford: Oxford University Press, 1965.

Douglas, Charles. John Stuart Mill: A Study of his Philosophy. Edinburgh: Blackwood, 1895.

Dryer, J.P. "Mill's Utilitarianism." In Essays on Ethics, Religion and Society. Edited by J.M. Robson. Vol. X of Collected Works. Toronto: University of Toronto Press, 1969.

Feinberg, Joel. Social Philosophy. Englewood Cliffs, N.J.: Prentice-Hall, 1973.

Feyerabend, Paul K. Against Method. London: New Left Books, 1975.

Friedman, Milton. Capitalism and Freedom. Chicago: University of Chicago Press, 1974.

Halliday, R.J. John Stuart Mill. London: Allen & Unwin, Ltd., 1976.

Hampshire, Stuart. "Uncertainty in Politics." Encounter 8 (January 1957): 34-37.

Hart, Herbert L.A. Law, Liberty and Morality. Stanford, Calif: Stanford University Press, 1963.

Hascall, Dudley L., and Robson, John M. "Bibliography of Writings on Mill." The Mill News Letter I (Fall 1965) through V (Spring 1970). Toronto: University of Toronto Press.

Hayek, Friedrich A. The Constitution of Liberty. Chicago: Henry Regnery Company, Gateway Edition, 1970.

Knickerbocker, F.W. Free Minds: John Morley and His Friends. Cambridge, Massachusetts: Harvard University Press, 1943.

Lippincott, B.E. Victorian Critics of Democracy. Minneapolis: University of Minnesota Press, 1938.

Lively, J. and Rees, J.C. Utilitarian Logic and Politics. Oxford: Oxford University Press, 1978.

Long, D.G. Bentham on Liberty. Toronto: University of Toronto Press, 1977.

Lyons, David. Forms and Limits of Utilitarianism. Oxford: Oxford University Press, 1965.

_______. "Human Rights and the General Welfare." Philosophy and Public Affairs 6 (Winter 1977): 113-129.

_______. In the Interest of the Governed. Oxford: Oxford University Press, 1973.

_______. "J.S. Mill's Theory of Morality." Nous 10 (May 1976): 101-119.

MacMinn, Ney; Hainds, J.R.; and McCrimmon, James McNab, eds. Bibliography of the Published Writings of John Stuart Mill. 1945; Reprint. New York: AMS Press, 1970.

Macpherson, C.B. The Life and Times of Liberal Democracy. Oxford: Oxford University Press, 1977.

_______. The Political Theory of Possessive Individualism. Oxford: Oxford University Press, 1962.

Mazlish, Bruce. James and John Stuart Mill. New York: Basic Books, 1975.

Mill, John Stuart. Autobiography. Oxford: Oxford University Press, 1971.

_______. An Examination of Sir William Hamilton's Philosophy. London: Longman's, 1865.

_______. On Liberty. Edited by H.B. Acton. London: Everyman edition, 1972.

_______. Principles of Political Economy, Collected Works. Vol III. Edited by J.M. Robson and V.W. Bladen. Toronto: University of Toronto Press, 1965.

_______. Principles of Political Economy. London: Penguin, 1970.

_______. System of Logic. London: Longman's, 1874.

_______. Utilitarianism, On Liberty and Considerations on Representative Government. London: Everyman edition, 1972.

Mitchell, Basil. Law, Morality, and Religion. Oxford: Oxford University Press, 1967.

Parker, Joseph. John Stuart Mill on Liberty, A Critique. London, 1865.

Pappe, H.O. John Stuart Mill and the Harriet Taylor Myth. Melbourne: Melbourne University Press, 1962.

Plamenatz, John Petrov. The English Utilitarians. Oxford: Basil Blackwell, 1958.

Popper, Karl R. The Open Society and Its Enemies. London: Routledge, Kegan, and Paul, 1978.

Radcliff, Peter, ed. Limits of Liberty. Belmont, Calif.: Wadsworth, 1966.

Rees, John C. "A Phase in the Development of Mill's Ideas on Liberty." Political Studies 6 (1958): 33-44.

_______. "A Re-Reading of Mill on Liberty." Political Studies 8 (1960): 113-129.

_______. "H.O. Pappe's John Stuart Mill and the Harriet Taylor Myth." Political Studies 10 (1962):198-202.

_______. Mill and His Early Critics. Leicester: Leicester University College, 1956.

_______. "The Thesis of the 'Two Mills."' Political Studies 25 (1977): 369-382.

_______. "Was Mill for Liberty?" Political Studies 14 (1966): 72-77.

Roach, John. "Liberalism and the Victorian Intelligentsia." The Cambridge Historical Journal 13 (1957): 58-81.

Robson, John M. "John Stuart Mill." Victorian Prose: A Guide to Research. Edited by David J. DeLaura. New York: The Modern Language Association of America, 1973.

_______. The Improvement of Mankind: The Social and Political Thought of John Stuart Mill. Toronto: University of Toronto Press, 1968.

Robson, John M., gen. ed. Collected Works of John Stuart Mill. Toronto: University of Toronto Press, 1963-. Authoritative critical editions of all of J.S. Mill's writings, still in the process of being published.

Robson, John M. and Laine, Michael, eds. The Mill News Letter. (Department of English, Victoria College, University of Toronto, Toronto, Canada.) Updated bibliographies of Mill published twice yearly since the fall of 1965.

Ryan, Alan. John Stuart Mill. London: Routledge, Kegan, and Paul, 1974.

_______. "John Stuart Mill and the Open Society." The Listener (May 17, 1973): 633-635.

_______. "John Stuart Mill's Art of Living." The Listener (Oct. 21, 1965): 60-62.

_______. "Mr. McCloskey on Mill's Liberalism." Philosophical Quarterly 14 (1964): 253-260.

_______. The Philosophy of John Stuart Mill. London: Macmillan & Co., 1970.

Schwartz, Pedro. The New Political Economy of J.S. Mill. Translated by B. Leblanc. Durham, N.C.: Duke University Press, 1972.

Spitz, David, ed. On Liberty. Norton Critical Edition. New York: W.W. Norton, and Co., 1975.

Stephen, James Fitzjames. Liberty, Equality, Fraternity. Edited by R.J. White. Cambridge: Cambridge University Press, 1967.

Stephen, Leslie. The English Utilitarians. London: Duckworth, 1900.

Strawson, P.F. "Social Morality and Individual Ideal." Philosophy: The Journal of the Royal Institute of Philosophy 36 (Jan. 1968): 1-17.

Urmson, J.O. "The Interpretation of the Moral Philosophy of J.S. Mill." Philosophical Quarterly 3 (1953): 33-39.

Watkins, J.W.N. "Three Views Concerning Human Freedom." In Nature and Conduct. London: Royal Institute of Philosophy Lectures Vol. 8, 1974.

Wollheim, Richard. "Mill on the Limits of State Interference." Social Research 40 (1973): 1-30.