John Rawls’s A Theory of Justice (1970) gave a new start to political philosophy, which had been in the doldrums for many decades, having been overshadowed by developments in the philosophy of logic, the philosophy of language, and the philosophy of mind. The study of the past masters of political philosophy was dutifully prolonged in universities, but without much hope of a new vision that would be likely to stir the interest of a wider public. The dust jacket says that A Theory of Justice has been translated into every major European language as well as into Chinese, Japanese, and Korean: a wider public indeed.
After Rawls followed not only a rush of books and articles in learned journals commenting on his work but also at least two well-argued parallel theories, each conveying a distinctive vision of liberalism: by Ronald Dworkin in Taking Rights Seriously and by Michael Walzer in Spheres of Justice. All three writers have been trying to formulate, as clearly as they can, the moral foundations of liberalism, as this slippery term is understood in political arguments within the US.
It is important, I think, to acknowledge that Rawls’s theories and arguments have to be understood, in the first place, as having their origin in the special setting of American history and of the American Constitution. Obviously liberalism is a political creed, with a set of political attitudes, which has flourished in Europe since Napoleon and which there has been supported by the arguments of Kant, Constant, Humboldt, Mill, and Sidgwick, among others. But American liberalism calls upon specific historical memories of the War of Independence, the debates on the Constitution, the Bill of Rights, and the Emancipation, all relatively recent events that have no equivalents in Europe. For this reason the search in the US for the justification of coercive government, and for the grounds of legitimacy, has employed a distinctive moral vocabulary which places a greater stress upon essential or primary human rights than has been usual in European thought about politics. Reading Rawls one is impressed by his distinctively American account of republican virtue, and of the political conception of persons as free and equal participants in a democracy.
Rawls’s great achievement in international thought was to restore the notion of justice to its proper place at the center of arguments about politics, the place that it had occupied at the very beginning of theorizing in Plato’s Republic. Justice is a necessary virtue of individuals both in their day-to-day conduct and in their personal relations, and it is the principal virtue of institutions and the social order. Plato described justice in the city state, social justice, as both reflecting and promoting the wisdom and moral soundness of those individual citizens who are themselves just, and Rawls’s account is in this tradition.
The second great contribution of A Theory of Justice was to detach the definition of social justice from the diverse moral aspirations and goals of individuals within the society. Moral ideals, in the sense of conceptions of the good life, are one thing, and justice is another: and precisely this separation constitutes the problem for a liberal theory of justice.
How can persons with contrary moral ideals coexist in harmony within a single society, possessing a unitary conception of justice? All the most authoritative political philosophers within the Western tradition, including Plato, Aristotle, Kant, Bentham, and Mill, had assumed that the virtue of justice, including social justice, had to be defined within a comprehensive theory of the human virtues, which each philosopher had in his turn supplied. A society ordered in accordance with the principles of justice must be so ordered as to sustain the best life for its citizens, and what this form of life is must be independently determined by philosophical argument. The utilitarian philosophies of Bentham and Mill have been the justificatory foundation of movements of social reform for many, perhaps most, secular liberals in Britain for a century or more, and utilitarianism was a comprehensive morality prescribing clear principles of action which are to be equally valid in public and in private life, for individuals and also for institutions.
But a well-argued doubt about the claims of all comprehensive moral theories has spread around the world in the last fifty years. It has come to seem both impossible and undesirable that anyone, whether a philosopher or not, should be able to identify one form of life as the best for all mankind at all times, unless he happens to have some direct access to the creator’s intentions. The distinction of humanity, and its interest in its own eyes, lies in the variety and unending competition of ideals and languages, and in the absurdity of a moral Esperanto. There have been many sources for this pluralist faith, including the evidences of social anthropology, the history of art, and the history of sexuality and family structures. The ideals of the monk and the soldier, of the revolutionary and the poet, of the aesthete and the politician, seem incurably at odds with each other, even as ideal types, and even more so when individuals of these types are inserted into a particular historical setting.
The clash of moral and religious loyalties has come to seem, in the light of recent history, much more than a temporary accident of human development, to be dispelled by the spreading natural sciences and by healthy enlightenment. Rather the deep-seated spiritual antagonisms have come to seem the essence of humanity, and it is an accident of history if, in some regions and for some period of time, a relative harmony of shared values prevails within a modern society. The natural condition of men and women is to cling to their distinguishing languages and to their divisive customs and rituals, and thereby to ensure that they will always misunderstand and distrust their neighbors anxiously clustered in their rival churches and assemblies.
A liberal philosopher who still wants to wrest some harmony and a degree of tolerance from this otherwise interesting chaos must not make the simple move of the utilitarians, and call on everyone to come together in assessing the consequences of their practices and institutions, as being favorable or unfavorable to the general welfare. This summons will not work, if only because every faction will interpret welfare differently, and in any case they normally also want justice as fairness, and the right to be inconveniently different from their neighbors, and even the right to be unhappy for their own peculiar reasons.
To meet these difficulties Rawls thought of an imaginary assembly of abstract persons who are supposed to choose the ideally just institutions for a society, without presupposing any specific conception of the good: the choosing persons are abstract choosers, because they are required to discount any knowledge that they have of their own position in society, of their own abilities and dispositions, and of their particular advantages and disadvantages. Behind the “veil of ignorance” they are to decide what sort of society they would consider to be just if they had to live in it.
Such an undifferentiated person speaks with the voice of the man within the breast, as Adam Smith first imagined him, which is the voice of shared humanity, and he knows enough of the general tendencies of people to enable him to agree with his peers on a set of principles, in an order of priority, that will define the just institutions for a society. For example, the abstract persons will, as reasonable thinkers, all agree that the liberty of the individual should have an overriding importance. They will agree also that the worst-off members of society, whoever they may be, must be protected against any worsening of their situation; this principle must be built into the basic structure of a liberal society from the beginning.
Critics of A Theory of Justice remarked that the abstract persons who are choosing the basic institutions seem already to be carrying with them the full liberal ideal of treating all persons as free and equal in a democratic state. Without this assumed liberal ideal there could be no constraint that would compel every rational man to prefer liberty to all other human goods, or that would compel him to mitigate, or never to increase, natural human inequalities.
In the eight lectures collected in his new book, Rawls has, step by step, modified and restated his theory in response to the critics who had accused him of presupposing, or of bluntly asserting, what he had undertaken to demonstrate: that any reasonable body of men, if they disregarded their own particular interests and characters, would arrive at principles of social justice which would be recognizably liberal.
It was not altogether clear how much had to be included in “reasonable”: Did this refer to enlightened calculations of self-interest or did it also include some minimum notion of moral decency? If the former, there is always the possibility that an apparently reasonable person might decide to gamble on the chance of his having natural advantages and not to bother too much about the liberties and the deprivations of others. Such a reasonable, but illiberal, person might cling to the principle that persons should be rewarded in proportion to their gifts and attainments without further correction. Or, if the constraint on the choice of institutional structures is a moral constraint, then it seems likely that some conception of the good has crept in, and Rawls would have allowed some exception to his respect for moral pluralism—for the great variety of reasonable opinions about what is good in human life.
For example, a person may have reasons to believe that a small group of human beings has access to the truth about good and evil, and that all the others will be ruined if they are not induced to respect the authority of the few; such a person will not believe that absolute priority should be given to the liberty of the individual, and yet he may be equipped with a battery of philosophical arguments, and with acute observations of human nature and of history, and he may be a calm and soft-spoken person; in all these senses he is eminently “reasonable.” After reflection he chooses a hierarchical society which engenders respect for principles of justice allotting appropriate privileges and constitutional powers to the group of superior and discerning persons. This would be illiberal and perhaps also un-American, but certainly not, in the ordinary sense of the word, crazy or unreasonable.
These were some of the simple and straightforward objections to the original theory of justice, and over the years Rawls has published more or less effective replies to them. But he finally came to recognize, as he explains here, that he had misdescribed his own enterprise, which is something that very easily happens in philosophy. He saw that he had been formulating the constitutional principles underlying the sense of justice that is employed in political situations and for political purposes within a democratic society by all those who acknowledge that citizens should be equally free to pursue their own good in their own way. The principles are the principles of political liberalism, and not of liberalism in a wider sense, which might include, for example, permissiveness in sexual morality or laissez faire in economic arrangements. Reasonable men and women will arrive at a consensus about a fair structure of institutions if they share a political conception of persons as free and equal for all political purposes and “a certain natural political virtue.” These reasonable men and women need to be sensitive to “the political value of a public life” and to the duty of civility as one duty among others.
These dispositions amount to “civic humanism,” an interesting and suggestive phrase. In a well-ordered liberal society the values of public life, including justice in the democratic city-state, flourish alongside a multiplicity of rationally defensible conceptions of the good life which citizens are free to form and to pursue on their own. A complete human being has two faces, one a communicative consensus-seeking, politically active, reasonable face, in the sense of “reasonable” that is opposed to fanatical, the other a private and autonomous, perhaps detached and secretive, uncompromising face of a person pursuing his own distinctive good, perhaps guided in this by a comprehensive morality. Such a person knows that political justice requires that he should not try to build his private moral ideas into the basic institutions of his society: this is the first commandment of political liberalism. Institutions are not individuals (earlier philosophers called them concrete universals), and the criteria of right and wrong that are applicable to them are not those that are applicable to persons. Justice is merely one among the many other virtues to which persons for different reasons aspire, while it is the great and prime virtue of social institutions, and hence of public life in general.
This is a most attractive picture, I think, of the authentic liberal democrat, and it is in many ways an advance on Benjamin Constant and John Stuart Mill as a vivid restatement of the first liberal aspirations. It also has most pleasant echoes of the founding fathers’ intentions for the future of the United States. A true liberal democrat goes to the assemblies and meeting-places, joins in party politics, and votes and accepts compromises, precisely because he wants to be left alone to cultivate his own garden, and because he respects the same desire in everyone else. Like Cincinnatus once he has performed his republican duties, he returns to the plough, and follows his yeoman’s conception of the good, and he expects his fellow citizens from the city similarly to pursue their goals, however unattractive these interests and goals may be to him. The political participation is not a cost and price paid, but an equal satisfaction of his moral nature, alongside his other aspirations.
Equal? There is the problem. How is the reasonable person to assess the claims upon him of his political self, the claims of democratic civility, when they come into conflict with the duties arising from his own distinctive conception of the good? Such conflicts are the stuff of common experience as they are represented in history and also in the lives of many, probably of most, of Rawls’s older readers. It will quite often happen that in “a well-ordered society” which is “a fair system of cooperation between reasonable and rational citizens regarded as free and equal,” a group of citizens, probably a minority, finds its strongest moral convictions overridden by policies that have been fairly chosen within the basic institutions recognized by the dissenters themselves as fair. The dissenters are not irrational fanatics, and they agree that fair and just institutions ought to be respected and obeyed, except when they engender policies that are for them morally unacceptable.
To take an example: in a well-ordered democratic society, there are likely now to be liberal dissenters who reject some of the demands that the nation-state makes upon its citizens when it calls for the forceful defense of national greatness and national interests. Nationalist sentiment, apt to be dominant in a democracy, may be found morally repugnant by some of those who accept Rawls’s definition of political justice in the basic institutions of their society. They are perhaps passionate internationalists or passionate pacifists, because their comprehensive morality is a version of Tolstoyan Christianity and it leads them to repudiate violence and aggression wherever possible.
Rawls can immediately reply that, in defining political justice from the liberal point of view, he is not called upon to propose some general method of resolving casuistical problems of public policy and of private conflicts of duty. He is writing about the basic structure of institutions in a just society, not about the choice of policies. From beginning to end these lectures remain on this level of high abstraction, and deliberately so. In spite of the recurrent toll of the word “liberal,” actual liberals struggling against actual conservatives, adversarial politics, scarcely enter into the argument. But even on this level of abstract general principle the problem of public versus personal morality remains; it does so because the separation between the political domain of overlapping consensus, and the private domain of protected moral diversity, superficially so attractive, is precarious and unstable.
At one point Rawls asserts, to me alarmingly, that “under reasonably favorable conditions that make democracy possible, political values normally outweigh whatever nonpolitical values conflict with them.” This meta-statement surely conveys a very substantial and definite moral point of view, putting the duties owed to basic institutions that are just ones ahead of all other human commitments. This ruling brings political liberalism, supposedly not itself a comprehensive morality, into direct conflict with many comprehensive moralities that are likely to flourish in a modern democracy. It suggests that a voluntary exile or emigrant, a hermit or wanderer, a detached aesthete or artist, is a delinquent, short of full humanity, insofar as he is careless of the justice of the institutions governing his own society. This may be a defensible opinion, among the multiplicity of possible moral opinions, but abstract philosophical argument is surely not by itself sufficient to support it.
Again the barrier between public political and private values breaks down when one very specific and concrete moral issue comes into Rawls’s argument in these lectures: abortion. Is the familiar dispute about the state’s provision or prohibition of abortion properly a matter for “public reason,” therefore to be referred to the basic principles of justice governing the institutions of a democratic society? Alternatively, does it fall within the domain of private morality, involving, as it surely does, the deepest moral convictions of individuals about their private lives? Certainly one may expect that the various comprehensive moralities, some religious, some secular, will each yield their typical and differing rulings on this issue, as it affects individuals and as it affects democratic legislation and the law. The comprehensive morality of believing Roman Catholics famously demands the state’s prohibition of abortion as being a form of murder.
At this point Rawls is firm in his liberal convictions: he writes—
Suppose…we consider the question [of abortion] in terms of these three political values: the due respect for human life, the ordered reproduction of political society over time, including the family in some form, and finally the equality of women as equal citizens…. I believe any reasonable balance of these three values will give a woman a duly qualified right to decide whether or not to end her pregnancy during the first trimester…. Any comprehensive doctrine that leads to a balance of political values excluding that duly qualified right in the first trimester is to that extent unreasonable…. Thus, assuming that this question is either a constitutional essential or a matter of basic justice, we would go against the ideal of public reason if we voted from a comprehensive doctrine that denied this right.
This passage makes it clear that Rawls’s new “political liberalism” still preserves the full creed of traditional liberalism and that this is the effect of a very extended and substantial use of the words “unreasonable” and “public reason.” The political liberal leaves space for the plurality of moral views to be found in any society, but only if they can be called reasonable, and this means reasonable as judged by the traditional standards of liberalism itself. Therefore one is inclined to ask—Is not political liberalism itself, in virtue of its exclusions under the heading of “unreasonable,” just one among other comprehensive moral doctrines? To which Rawls replies, “No,” because “the requirements of public reason belong to an ideal of democratic citizenship.” Public reason by definition aims at an overlapping consensus, and its ultimate justification is that it can plausibly be supposed to achieve this consensus. This is not true of comprehensive moral theories, such as utilitarianism, which in a polemical spirit dismisses all intuitive moralities whether democratic or not. Rawls’s whole argument here turns on the notion of “public reason” which, in pursuit of a consensus among free and equal citizens, abstracts from all other and more specific moral concerns: for example, from the association suggested between the killing of a fetus and the crime of murder.
The final authority rests with consensus—seeking “reason,” which is attributed to all citizens equally as free and autonomous agents. We must still feel some awe or fear (Achtung) before these magnificent words “reason” and “unreasonable” brandished in front of us in Kantian style. But I, as a liberal, cannot start here, because I deem myself to detect an ancient and disastrous illusion, descending unashamed to Kant and Rawls from Plato’s Republic.
Rawls rightly insists that it is helpful to look for the most general, and therefore abstract, underpinnings of our moral and political attitudes and beliefs, even though the support that we find will at best be persuasive but will never be any kind of proof; and persuasion is what he tries in this book and in A Theory of Justice, a theory, one among other possible theories. My objection is not against the abstractness of his argument, but against the notions of “reason” and “reasonable” employed in it. Why should an overlapping consensus among “reasonable” persons about basic liberal values be either required or expected?
The answer is to be found in the history of the myth of reason itself. Plato, discussing justice in The Republic, threw off the brilliant and entertaining idea that the soul is divided into three parts, just as the city-state is to be divided into three social classes, and in a just person’s soul the upper part, reason, ensures harmony and stability, and in a just city the upper class, philosophers trained in mathematics, will impose order in a well-ordered society. This fiction is the decorative part of a polemic against democracy and in favor of oligarchy, a polemic claiming that government and the administration of justice are serious matters, and that it is obvious that human beings are not equally endowed, but rather noticeably differ in their capacity to construct and to follow arguments, to study and to learn. Only if the intelligent and well-educated govern will society be stable and harmonious.
Unfortunately this fairy tale about the parts of the soul and their hierarchy passed into common speech and into the vocabulary of the Latin language and hence into much of Christian philosophy. The corollary in ordinary and conventional speech has been that the desires and emotions of persons are supposed to issue from the quarrelsome and insubordinate underclass in the soul, and that they should be left in their proper place and kept away from the serious business of self-control.
Of course there have been honorable dissenters from this establishment myth, among them Montaigne, Machiavelli, and Hobbes, heirs of Heraclitus rather than of Plato. They each looked at the facts on the surface of their own experience and of ancient and modern history, and noticed that everywhere, both in the soul and in the city, the mark of vitality is conflict, so much so that it seems a law of life that any individual’s desires and feelings should be at all times in a state of conflict and properly unstable, and that in public life social classes should at all times be in conflict and society should be properly unstable. The narrative of a person’s life shows the constantly shifting predominance of one set of contrary dispositions and ambivalent feelings over another, as she matures and decays, and the historical development of a state or society depends on the competition between different social groups in an unceasing struggle for power. This is the engine of history and we do not expect it to come to a dead stop, although in some moods of despair we may indulge in a fantasy of a final stage, of a Utopia, or we may dream of stability and harmony, as Plato did after the failures of Athenian democracy. Nowhere is there evidence, whether in the individual soul or in society, of a sovereign reason which can secure a consensus, the end of conflict, a uniform order, a harmony of interests, the heavenly city of the philosophers.
If the moral and religious sentiments of human beings are in their essence exclusive and divisive, how is the war of all against all to be avoided and how can that degree of consensus necessary for public order ever come into existence? The most plausible and historically defensible answer is by political compromise, by rule-governed negotiation, by arbitration, sometimes adjudication, in institutions that have grown up to serve this purpose, usually by slow stages over a long period of time. This is the sphere of public reason, of political values and virtues, and of the duties of civility. In a section entitled “Permissible Conceptions of the Good and Political Virtues,” Rawls mentions the possibility that public conflicts can be dealt with by the use of “neutral procedure,” which can be justified by an appeal to “neutral values, that is, to values such as impartiality, consistency in the application of general principles to all reasonably related cases…and equal opportunity for the contending parties to present their claims.”
The principal procedural value is justice and fairness in an adversarial process, whether in parliament or in the courts; and justice and injustice in the procedures used in settling a conflict can be distinguished from the evaluation of the settlement finally reached as itself just or unjust. A liberal living in an imperfectly liberal democracy, such as the US, may be convinced that it is unjust that the law should deny a pregnant woman the freedom to terminate her pregnancy in the first trimester; this is his conviction because, within his comprehensive morality, the freedom of the individual has an overriding priority. His Roman Catholic adversary may be certain that it is unjust to allow an innocent living soul, as determined by the Church, to be destroyed for any reason whatever: this is his belief because he is convinced that the Church’s account of God’s will is correct.
On the substantial nonprocedural, issue of justice, there is outright conflict. In a just and perfectly liberal society the basic institutions will ensure that both sides have the opportunity to argue their case on equal terms, and this will mean to argue in Congress or parliament and in the Supreme Court, if there is one. In a just society neither side can impose its moral convictions on the society by forceful domination and without following the recognized procedures of equal debate and of consistent and rule-governed adjudication which are already established in that society.
If this distinction between procedural and substantial justice is understood, between fairness in the process and fairness in the result, the claims for political liberalism and public reason and for the duty of civility become plausible. Within different moralities, liberal and conservative, the fairness of the actual outcome of a conflict will be evaluated differently, even though both sides recognize the fairness of the adversarial process. Outcomes are of their nature open to dispute, but processes need not be. One begins to see why Rawls can think that political values normally outweigh other values, such as the value of friendship, although the qualification “normally” is important here, because sometimes the claims of justice, whether procedural or substantial, are outweighed by other claims. In a universe pictured as one of perpetual conflict, private and public, one can never say of any virtue or moral duty that it is never outweighed. Rawls is implicitly recognizing the distinction between fairness in the procedure and fairness in the result when he makes democracy an essential element in the structure of political liberalism. Democracy is a procedure, and, more specifically, an adversarial procedure, at least if one regards one-party democracy as a sham. The purpose of democratic politics is to elevate any serious social conflict into debate and to provide some institutional clothing to cover moral and natural hostilities.
Loosened from any single conception of the good, including the divisive values which are cherished by liberals, political liberalism, so modified, has, I believe, a great future in many parts of the world, in spite of conservative propaganda stirring up fear on behalf of the more solid and rigid faiths. There is one obstacle. Since liberal philosophy appeared on the scene soon after 1800, it has accepted from some of its natural enemies, the churches and the traditionalists, an excessive valuation of the part that reasonable beliefs attached to carefully drafted propositions actually play, and ought to play, in human affairs generally, and, more particularly, in politics and in sustaining a decent social order.
In these lectures Rawls goes some way, but not all the way, to free himself from this misplaced rationalism by insisting that he does not claim truth, only (once again) “reasonableness,” for his principles of political liberalism. But fairness and justice within a liberal society first require that there should exist respected institutions for adversarial argument, and equal access to them, accepted manners in negotiation, and entrenched rules and habits of advocacy, a full ritualization of public conflicts. Articulate beliefs are secondary for two well-known reasons: first because they are supported by uncertain arguments and probable evidence and not by proofs, and therefore will not generally be agreed upon. Second, where ideals are at issue, the particular passions and memories of the particular individuals involved will largely determine their beliefs.
Political Liberalism is certainly a very important book. Parts of it have been delivered as lectures on various occasions, and there are repetitions. It is not an easy book to read. One tends to be lulled into acquiescence because the noise and muddle of actual politics are altogether absent, and history is scarcely called upon at all. Rawls has recently, in a public lecture at Oxford, extended his defense of political liberalism to international politics and tested its principles in Machiavelli’s world. He discusses the possible recognition of a “law of peoples” which would ideally be recognized by the community of nations.
Even though this law of peoples is put forward only as an ideal to mark the goal of political reason in international affairs, the argument still seems unconvincing because it starts from principles rather than from the institutions that might ideally be developed as props to a liberal international order. Such ideal institutions could possibly be imagined and projected on the basis of the very few, very weak, and very imperfect institutions for the worldwide protection of human rights which are already in existence—ranging from Amnesty International to various agencies of the UN. The customary constraints of traditional diplomacy, taken together with the rules and conventions of the United Nations and its subsidiary organs, make a possible base from which an ideal of international procedural justice might be projected. Certainly it is a frail and scanty base, but such institutions are more likely to point the way toward a future international order, however sketchily, than any set of general principles.
Finally it may be asked how a choice is made between the two pictures of human nature that I have mentioned and between the resulting two pictures of the structure of politics: first, the traditional and Rawlsian picture of a dominant practical reason finding an overlapping consensus in the embedding of liberal values of freedom and equality in basic institutions. Second, the Heraclitean picture of the unavoidable and life-giving clash and friction of competing moral ideals and contrary visions of life founded on memories and passions, a clash that is precariously controlled and regulated by imperfect institutions of adversarial argument and arbitration. The distinctively liberal requirement, in this second picture, is that all sides should have equal access and be heard (the legal tag “audi alteram partem“), not only the powerful and dominant groups.
The choice between the two pictures of personality and of society can only be tested by experience, and by experience in two senses: by observation of one’s own feelings and by the study of history. The perpetual clash and friction of divisive attachments and of memories and of emotions in conflict seems to me to make up the internal life of a person, and the perpetual clash and friction of ethnic loyalties and religious loyalties and cultural loyalties and class loyalties make up the life and development of societies, cities, and states.
Rawls’s new political liberalism, as he presents it, seems to me too gentle and too temperate in tone. Liberalism was born around 1800 in hostility to the established order of Church and state, expecting perhaps occasional victories and more frequent defeats, but not expecting just a tranquil acceptance. Liberals are trampled on at the same time as comprehensive moralists and patriots march to and fro in pursuit of their fantasies—while liberals fight for the habit of allowing each side of each case to be heard. That is their passion, a passion for civility within conflict.
August 12, 1993