¹ See Gray, , Mill on Liberty: A Defence, London, 1983CrossRefGoogle Scholar; Gray, , Liberalism, Milton Keynes and Minneapolis, 1986Google Scholar; Gray, , Liberalisms: Essays in Political Philosophy, London, 1989Google Scholar; Berger, , Happiness, Justice and Freedom: The Moral and Political Philosophy of John Stuart Mill, Berkeley, 1984Google Scholar; and Rees, , John Stuart Mill's ‘On Liberty’, ed. Williams, G. L., Oxford, 1985.Google Scholar

² Liberalisms, p. 222Google Scholar, capitals in original.

³ Ibid., p. 220.

⁴ Ibid., p. 221, capitals in original.

⁵ Ibid., pp. 222 and 223, emphasis and capitals in original.

⁶ Ibid., p. 220.

⁷ Ibid., p. 221.

⁸ Ibid., p. 223.

⁹ Ibid., p. 224.

¹⁰ Ibid., pp. 220 and 224.

¹¹ Ibid., p. 224.

¹² Ibid., pp. 224–5.

¹³ Ibid., p. 225.

¹⁴ Ibid., p. 226.

¹⁵ Ibid.

¹⁶ Ibid., pp. 226–7.

¹⁷ Ibid., p. 229. Gray implies not merely that Mill's doctrine is incompatible with the standards of morality actually accepted and shared within modern societies; but also that the liberty principle lacks appeal as a critical principle for use in the criticism of actual social institutions including law and positive morality. In contrast H.L.A. Hart argues that ‘Mill's principles are still very much alive [as of 1963] in the criticism of law.’ Hart recognizes, however, that anti-Millian views such as those of Stephen, James Fitzjames (Liberty, Equality, Fraternity, 2nd ed., London, 1874)Google Scholar and Devlin, Lord Patrick (The Enforcement of Morality, Oxford, 1965)Google Scholar may actually be more popular (especially among lawyers) both in England and America. See Hart, , Law, Liberty and Morality, Oxford, 1963, pp. 15 and 17Google Scholar. For the distinction between critical and positive morality, see ibid., pp. 17–24. It should be noted that Hart himself departs in important ways from the ‘simple liberalism’ of Mill. For a useful discussion of this point in the context of the wellknown debate between Hart and Devlin, see Martin, Michael, The Legal Philosophy of H.L.A. Hart, Philadelphia, 1987, pp. 239–71.Google Scholar

¹⁸ Liberalisms, p. 92.Google Scholar

¹⁹ Ibid., pp. 26–36.

²⁰ Ibid., pp. 30–1.

²¹ Ibid., pp. 91–2. For a different argument that stresses the continuities between Whig constitutionalism and progressive liberalism of the sort associated with Mill, see Burrow, J. W., Whigs and Liberals: Continuity and Change in English Political Thought, Oxford, 1988.Google Scholar

²² Liberalisms, p. 234Google Scholar. Gray's various essays on Mill, Popper, Nozick, Rawls, Berlin, Hayek, Spencer and Oakeshott (among others) attempt to elucidate that general claim.

²³ Ibid., p. 232.

²⁴ Ibid., p. 233.

²⁵ Ibid., pp. 230 and 234.

²⁶ Ibid., p. 234.

²⁷ Ibid., emphasis added.

²⁸ Ibid., pp. 261–4.

²⁹ Ibid., p. 264. ‘Political Pyrrhonism’ seems to come dangerously close to claiming that there cannot be any ‘liberal’ critical morality for evaluating a society's institutions and practices, including its positive morality. The approach merely wishes to conserve ‘the historical inheritance of liberal practice’ without identifying any principles for distinguishing between liberal and non-liberal practices. Without such principles, it is difficult to see how political Pyrrhonism can be anything more than mere conservatism in the context of this or that particular cultural tradition. The approach bears strong affinities to Michael Oakeshott's thought as interpreted by Gray, at pp. 199–216Google Scholar. See also Oakeshott, , Rationalism in Politics, London, 1962Google Scholar. Non-doctrinal liberals (conservatives by another name?) are apparently included among ‘political Pyrrhonists’ because they do not defend any universal principle of liberty. Gray specifically directs our attention to Raz, Joseph, The Morality of Freedom, Oxford, 1986Google Scholar, and Flathman, Richard, The Philosophy and Politics of Freedom, Chicago, 1987Google Scholar. Perhaps Isaiah Berlin is also meant to be included because ‘a genuine pluralist about values cannot be a doctrinal liberal’ (p. 260). But, as I argue in Section III.C. of the text, Berlin does defend some sort of universal principle of liberty, despite his value pluralism.

³⁰ The two associated papers are ‘Individuality, Custom and Progress’ and ‘The Liberty Principle Applied’ respectively. The former argues that Mill's conception of individuality or autonomy is not by definition incompatible with social customs and traditions, and that his theory of progress fills in an important lacuna in alternative evolutionary explanations such as those of Herbert Spencer and Friedrich Hayek. The latter paper argues that Millian liberalism is not an overly simple doctrine as suggested by H.L.A. Hart and Joel Feinberg, among others. Some key practical implications of Mill's doctrine are illustrated with a view to defending its appeal relative to the more complex versions of liberalism attributable to Hart and Feinberg.

³¹ ‘On Liberty’, [1859], Essays on Politics and Society, ed. Robson, John M., 2 vols., Toronto, 1977Google Scholar, Collected Works of J. S. Mill, xviii. 223.Google Scholar

³² Ibid., 224.

³³ Ibid.

³⁴ Ibid.

³⁵ Ibid., 225.

³⁶ Ibid., 281.

³⁷ Ibid., 225–6.

³⁸ Ibid., 226.

³⁹ Ibid., 292.

⁴⁰ Ibid., 245.

⁴¹ In the introduction to his Four Essays on Liberty, Oxford, 1969Google Scholar, Berlin comments as follows: ‘Mill does seem to have convinced himself that there exists such a thing as attainable, communicable, objective truth in the field of value judgments; but that the conditions for its discovery do not exist save in a society which provides a sufficient degree of individual liberty, particularly of inquiry and discussion…. My thesis is not this at all, but that, since some values may conflict intrinsically, the very notion that a pattern must in principle be discoverable in which they are all rendered harmonious is founded on a false a priori view of what the world is like’ (pp. 1–li). Actually, Berlin here seems to be asserting what he takes to be a true a priori view of what the world is like. Mill, however, makes no such positive assertion. He merely claims that the possibility of a coherent notion of good (so-called ‘utility in the largest sense’) does not seem to be ruled out by any evidence warranted from our experience of the actual world to date.

⁴² ‘On Liberty’, Essays on Politics and Society, CW, xviii. 231.Google Scholar

⁴³ Ibid., 244.

⁴⁴ Ibid., 245. See also 250–2 relating to Socratic dialectics.

⁴⁵ Ibid., 220.

⁴⁶ Ibid., 223.

⁴⁷ Ibid., emphasis added.

⁴⁸ Ibid., 241.

⁴⁹ Ibid., 217.

⁵⁰ Ibid., 218.

⁵¹ Ibid.

⁵² Ibid., 219.

⁵³ Ibid.

⁵⁴ This view of liberty is emphasized in the Federalist Papers written during 1788–9 by James Madison, Alexander Hamilton and John Jay to encourage the people of the State of New York to ratify the U.S. Constitution. See The Federalist, ed. Cooke, Jacob E., Middletown, Conn., 1961Google Scholar. For further discussion of this notion of political liberty, see also Carey, George W., The Federalist: Design for a Constitutional Republic, Urbana and Chicago, 1989Google Scholar; and Riley, , ‘American Democracy and Majority Rule’, in Majorities and Minorities, Nomos, vol. xxxii., eds. Chapman, John W. and Wertheimer, Alan, New York, 1990, 267–307.Google Scholar

⁵⁵ ‘De Tocqueville on Democracy in America II’, [1840], Essays on Politics and Society, ed. Robson, John M., 2 vols., Toronto, 1977Google Scholar, Collected Works of J.S. Mill, xviii. 198Google Scholar. See also ‘On Liberty’, Essays on Politics and Society, CW, xviii. 272–5.Google Scholar

⁵⁶ ‘De Tocqueville’, Essays on Politics and Society, CW, xviii. 196.Google Scholar

⁵⁷ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 220.Google Scholar

⁵⁸ ‘De Tocqueville’, Essays on Politics and Society, CW, xviii. 198.Google Scholar

⁵⁹ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 220Google Scholar. For Hayek's view, see The Constitution of Liberty, Chicago, 1960Google Scholar; and Law, Legislation and Liberty, 3 vols., Chicago, 1973–1979.Google Scholar

⁶⁰ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 220–1Google Scholar. Lord Devlin relies on a variant of this practical principle to justify legal enforcement of a society's existing morality (independently of content). See The Enforcement of Morals, Chapter 1. Martin refers to Devlin's use of the principle as the ‘Disgust Test’ for legal moralism. See The Legal Philosophy of H.L.A. Hart, pp. 245–6.Google Scholar

⁶¹ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 261Google Scholar. For Mill's comments on the exceptional case of religious liberty, see ibid., p. 222. He may have in mind the distinctive American doctrine of religious liberty enunciated in such documents as the Virginia Statute for Religious Freedom (drafted by Jefferson, Thomas in 1777, enacted 1786)Google Scholar, Madison, 's ‘Memorial and Remonstrance Against Religious Assessments’ (written in 1785 to help enact Jefferson's Statute for Religious Freedom)Google Scholar, and the First Amendment to the U.S. Constitution. For discussion of this American doctrine, see The Virginia Statute for Religious Freedom: Its Evolution and Consequences in American History, eds. Peterson, Merrill D. and Vaughan, Robert C., Cambridge, 1988Google Scholar. As I argue elsewhere, Mill's liberty principle can be interpreted to be a generalization of the Jefferson-Madison argument for religious liberty. See Riley, , ‘The Liberty Principle Applied’.Google Scholar

⁶² ‘On Liberty’, Essays on Politics and Society, CW, xviii. 227Google Scholar. For an enlightening discussion of Mill's pessimism in this context, see Burrow, , Whigs and Liberals, pp. 77–124Google Scholar. As Burrow points out, there are loud civic republican and Country Whig echoes in Mill's pessimistic view that rising social homogeneity would likely lead to stagnation:

In both cases, because we are invited to think of a social agency on which the whole fate of the polity or society turns there is an accompanying anxiety. Public virtue and individuality are conditions of social health which may easily be lost. To sustain them calls for a constant exertion of moral energy; to allow this to weaken is a fatal enervation. ‘The consequence is the loss of the most cherished social good, which for the eighteenth-century constitutionalist is ‘public liberty’ and for Mill is ‘social progress’. The distinction between the two is important … But in each case the feared final state, the collapse or enervation of energy and will, is referred to in similar terms. In the eighteenth century the opposite of public liberty is often an Asiatic, or, more especially, ‘Turkish’ servitude. For Mill the antithesis of the state of progressive social energy, fuelled by freedom of thought and speech, and the vigorous cultivation of a spontaneous individuality, is ‘Chinese stationariness’…. It seems not an exaggeration to see, in Mill's notion of many-sidedness and exaltation of variety … the extension of older notions of balance, and of Parliament as the representation and harmonization of contending interests, to an increasingly mobile and self-consciously individualistic society (pp. 85–6 and 113).

⁶³ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 271.Google Scholar

⁶⁴ Ibid., 276, emphasis added.

⁶⁵ Ibid., 277.

⁶⁶ Ibid., 217.

⁶⁷ Ibid., 292.

⁶⁸ Ibid.

⁶⁹ Rees, , John Stuart Mill's ‘On Liberty’, pp. 168, 172.Google Scholar

⁷⁰ See Gray, , Mill on Liberty: A Defence.Google Scholar

⁷¹ Berger, , Happiness, Justice and Freedom, p. 250Google Scholar. Many other commentators subscribe to something like the Rees-Gray-Berger definition of harm. See, for example, Feinberg, Joel, The Moral Limits of the Criminal Law, 4 vols., Oxford and New York, 1984–1988Google Scholar. It should be noted that Feinberg, like Hart, believes legal coercion may be also justified on the basis of principles other than a harm principle thus interpreted, for example, a principle of offence to public standards of decency.

⁷² See Rees, , John Stuart Mill's ‘On Liberty’, pp. 148–55 and 168–74Google Scholar; and Gray, , Liberalisms, pp. 150–3 and 222–4.Google Scholar

⁷³ Hart, , Law, Liberty and Morality, p. 47.Google Scholar

⁷⁴ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 288.Google Scholar

⁷⁵ A more familiar example of this general approach is provided by Hayek, , The Constitution of Liberty, pp. 133–92Google Scholar. See also Gray's critique of Hayek on liberty, in Liberalisms, pp. 89–102Google Scholar; and Hayek on Liberty, London, 1984.Google Scholar

⁷⁶ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 276.Google Scholar

⁷⁷ ‘Utilitarianism’, [1861], Essays on Ethics, Religion, and Society, ed. Robson, John M., Toronto, 1969Google Scholar, Collected Works of J.S. Mill, x. 247.Google Scholar

⁷⁸ Ibid., 243–60.

⁷⁹ Ibid., 247.

⁸⁰ ‘Auguste Comte and Positivism’, [1865], Essays on Ethics, Religion, and Society, ed. Robson, John M., Toronto, 1969Google Scholar, Collected Works of J.S. Mill, x. 337–8Google Scholar. For further discussion, see Riley, , Liberal Utilitarianism, Cambridge, 1988, pp. 191–234.Google Scholar

⁸¹ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 292–3.Google Scholar

⁸² Ibid., 293.

⁸³ Ibid., 225.

⁸⁴ In my view, emotional distress is, at best, evidence of harm in the sense of perceptible injury. Distress is not harm itself, in other words, but rather a sign that harm may be present. I say that, ‘at best’, distress is evidence of harm because it may instead be evidence of mere dislike or even of mere like, however intense. In the case of purely selfregarding actions, others' distress is not evidence of any perceptible damage to them. It is evidence merely of their own attitudes and desires. Mill's liberty principle recommends that society should, in effect, ignore such self-induced evidence. Of course others' distress is not self-induced in the case of other-regarding actions that directly cause perceptible hurt to other people, or that carry a probability of doing so. Distress in that case is evidence of harm. Whether he likes or dislikes the other-regarding conduct in question, a victim is forced to live with its harmful consequences, including physical injury, financial loss, ruined expectations, and the like. These consequences do not depend solely on his own attitudes and desires, and give him a reason for feeling emotional distress. Thus, for example, serious lies and credible threats that literally create a probable perceptible hurt in someone else's life, are harmful other-regarding actions justifiably restrained by social authority.

⁸⁵ Hart seems to think that the proper definition of harm is ‘of subsidiary importance’. But he also admits that defenders of individual liberty are faced with an impossible task unless certain kinds of distress are in effect not counted as harm. In particular, ‘the utilitarian principle that coercion may be used to protect men from harm’ cannot be extended to protect men from ‘the distress which is inseparable from the bare knowledge that others are acting in ways you think wrong’ or from ‘the distress incident to the belief that others are doing what you do not want them to do. No social order which accords to individual liberty any value could also accord the right to be protected from distress thus occasioned.’ So far Hart apparently wants to restrict the meaning of harm in a way similar to that which I am attributing to Mill. But he also introduces an admittedly fine distinction between these excludable forms of distress and ‘shock or offence to feelings caused by some public display’. For Hart, mere dislike is in effect not harm if it is occasioned by another's activities in the privacy of his own home, but mere dislike is harm if it is occasioned by another's activities in public. Although Mill sometimes gives indications of having accepted a similar distinction, it is not easy to see why the source of a person's mere dislike should have this sort of importance. In my view, even serious offence at another's conduct in public is indistinguishable from mere dislike if the conduct is not directly associated with any ‘perceptible damage’ to others. The main differences between the Millian and Hartian doctrines are further discussed in Riley, , ‘The Liberty Principle Applied’Google Scholar. For present purposes, the important point is that Hart's definition of harm seems to be less restricted than Mill's definition.

⁸⁶ Berlin, , Four Essays on Liberty, p. xxxii.Google Scholar

⁸⁷ Ibid., pp. 164–5, emphasis added.

⁸⁸ Ibid., p. 155.

⁸⁹ Ibid., p. liii.

⁹⁰ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 278, emphasis added.Google Scholar

⁹¹ Ibid., 277.

⁹² Ibid.

⁹³ Ibid., 278. As Hart emphasizes, Mill certainly does not advocate indifference to selfregarding vices (Law, Liberty and Morality, pp. 76–7Google Scholar). Nevertheless, Mill holds that natural penalties are not the same thing as deliberate moral blame or punishment. Public reprobation does not legitimately extend to self-regarding conduct because such conduct is harmless to others.

⁹⁴ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 278.Google Scholar

⁹⁵ Ibid., 282.

⁹⁶ Ibid., 280.

⁹⁷ For Hume's distinction between natural and artificial virtues and vices, see A Treatise of Human Nature, [1739–1740], ed. Selby-Bigge, L. A., 2nd ed., Oxford, 1978, pp. 294–8, 474–5, 477–621Google Scholar; and also Enquiries Concerning Human Understanding and Concerning the Principles of Morals, [1777], ed. Selby-Bigge, L. A., 3rd. ed., Oxford, 1975, pp. 169–284, 303–11.Google Scholar

⁹⁸ ‘On Liberty’, Essays on Politics and Society, CW, xviii. 279.Google Scholar

⁹⁹ Ibid., 279–80.

¹⁰⁰ Ibid., 280.

¹⁰¹ Ibid.

¹⁰² Ibid., 279. Recall his related claim that thoughts and opinions are not properly termed immoral (ibid., 234).

¹⁰³ Ibid.

¹⁰⁴ Ibid., 280.

¹⁰⁵ Ibid.

¹⁰⁶ Ibid., 281

¹⁰⁷ Ibid.

¹⁰⁸ Ibid., 281–2.

¹⁰⁹ Ibid., 281.

¹¹⁰ Ibid.

¹¹¹ Ibid., 282.

¹¹² Ibid.

¹¹³ Ibid.

¹¹⁴ Ibid., 283.

¹¹⁵ Ibid.

¹¹⁶ Ibid.

¹¹⁷ Ibid., 224.

¹¹⁸ Ibid.

¹¹⁹ Ibid., 283.

¹²⁰ Ibid., 277.

¹²¹ See Riley, , Liberal Utilitarianism.Google Scholar

¹²² Mill seems to have this sort of argument in mind. See ‘On Liberty’, Essays on Politics and Society, CW, xviii. 294 f.Google Scholar

¹²³ Ibid., 294, 260, 226.

¹²⁴ Ibid., 261, 271 n.

¹²⁵ ‘Utilitarianism’, Essays on Ethics, Religion, and Society, CW, x. 247.Google Scholar

¹²⁶ Burke, to Depont, , [November 1789], in The Correspondence of Edmund Burke, ed. Cobban, Alfred and Smith, Robert A., 8 vols., Chicago, 1967, vi. 42Google Scholar; as quoted by Kammen, Michael, Spheres of Liberty: Changing Perceptions of Liberty in American Culture, Madison, 1986, p. 84Google Scholar. For a similar view, see Oakeshott, , Rationalism in Politics, pp. 40–1 and 120Google Scholar; and also Gray, , Limited Government: A Positive Agenda, Hobart Paper 113, London, 1989.Google Scholar

¹²⁷ The Octopus, [1901]Google Scholar, in Frank Norris: Novels and Essays, ed. Pizer, Donald, New York, 1986, p. 1017.Google Scholar