It is said that we live in a secular society. But what is meant by “secular”? All relevant usages of the word, deriving as it does from saecularis (of an age, or of a generation), point to a concern with the affairs of our time, as opposed to some other; to a concern with this world of ours, as opposed to some other. But a great deal hangs on the nature of the contrast. There are, I believe, three possibilities, two of which have considerable currency today, one of which does not. I will call them the supersessionist, the liberal, and the eschatological. Or, if you prefer, the immodest, the falsely modest, and the genuinely modest.




The supersessionist or immodest approach implies a contrast between eras, an essentially negative contrast generally associated with nineteenth-century antireligious humanism. Here the “secular” is defined over against the “religious” in much the same way as late Renaissance and early Enlightenment thought was defined over against the thinking of the so-called Dark Ages. Darkness gives way to dawn. Priestcraft gives way to science. Religion gives way to reason (“the very voice of God,” as the Rev. Benjamin Whichcote put it already in the seventeenth century). Secularization, intones the sociologist, is the process by which religious ideas and institutions lose their public significance. Secular principles are those that operate without any appeal to religious beliefs or justification, that indeed are uncontaminated by any such beliefs or justifications.




Although the supersessionist definition is still widely employed in Western society, and is often implicit or explicit in the language of the media, the academy, and even the courts, it has many difficulties. Not the least of these is the disconcerting persistence of religious ideas and practices even in “enlightened” cultures. Religion in the West is far from moribund, even where it is in turmoil or distress. But I have not described the supersessionist view as “immodest” merely because it is inclined to exaggerate its own success. Its immodesty runs deeper than that.




Consider the contribution of Jean-Jacques Rousseau. Like Thomas Hobbes, Rousseau rejected the medieval doctrine of overlapping ecclesiastical and civil authorities. Yet Rousseau rejected Hobbes’ substitute, which assimilated church to state, because it could not ensure the state’s supremacy to the Church. Rousseau thus proposed a solution that would eliminate the competition between the two once and for all. His first move was to redefine the spiritual realm, from which the Church purported to derive its authority, in such a way as to render it politically impotent. The spiritual and the political, the eternal and the temporal, were henceforth to be held apart by the repelling force of a powerful dualism. The spirit or the conscience might belong to God, but the body would belong to the state.




To say the least, this was a curious take on Jesus’ saying, “Render to Caesar what is Caesar’s, and to God what is God’s.” It gets worse, however, for Rousseau believed that the state”if it is to claim and to hold the allegiance of the people”must itself provide some form of bonding backed by theological and legal sanctions. He therefore called for a civil religion that he hoped would do just that. Revolutionary France experimented with such a religion during the days of its triumph over the Church, an experiment that proved in its immodesty to be the debacle any sane person might have predicted.




Supersessionism, in subsequent incarnations, has tended to shy away from religious aspirations, civil or otherwise. In the West it is usually deemed enough simply to resist any attempt by organized religion at a political resurrection. Occasionally, however, especially in the East, a still more militant supersessionism has substituted for religion an oppressive ideology of state authoritarianism such as that which created the infamous Gulag Archipelago. But such immodesty, though secretly or openly admired by some Western supersessionists, has always proved to be self-destructive. The state that exalts itself “above every so-called god” is bound to fail, and to fail dramatically.




The second, “liberal” approach owes more to John Locke than to Hobbes or Rousseau; more recently, it is deeply indebted to John Rawls. It still works with a negative contrast between eras: the past world, with its putative “wars of religion,” and the present world, which is ever so much more peaceful”or would be, if only religious people at home and abroad would catch up with the Enlightenment. Moreover, it also assumes a sharp dichotomy between the spiritual and the material, as between the private and the public. Yet it is distinct in a number of ways.




For the liberal, a secular society may be proud of its different cultures and religions. Liberalism is not in principle against religion, nor has it lost all interest in religion. Rather it is interested in religion, albeit in a disinterested way. Above all, it is committed chiefly to the cultivation of tolerance”the one remaining virtue that strives, in valiant combat with Nietzsche’s Will to Power, to govern all values”so that everyone may be free to pursue whatever he or she happens to believe makes for peace and happiness, whether in this world or the next.




Like a neutral referee, such a society’s aim is simply to observe and to regulate the competition between the religious and the nonreligious, refusing to take sides with this or that worldview or way of life, but making sure that there is some kind of space for adherents of each. In a liberal society the state and its functionaries have no mandate to deliver us from the seductive power of religion, nor yet to create a religion that serves the state, but only to maintain a just peace, an essential equality; which means, in practice, to enforce tolerance. Thus, for example, in the face of differences over the education of its youth, its officials will be instructed to take no decision that rests upon religious reasoning, or that cannot be justified without reference to some creed or other. Put positively, their task is to establish policies that enfranchise the disenfranchised, without endorsing specific community values or counter-values (on human sexuality, say, or on worship of God or gods).




The modesty of this view is at first glance obvious. It is so modest, in fact, that it is doubtful whether it can possibly sustain itself. Where is the mutual binding, the religare, in the pursuit of a “justice” or an “equality” or a “liberty” or a “peace” that has no definite end or object, no historic content, and no particular measure except tolerance? No society, nor any state, has ever yet attempted to rest itself on so meager a basis for union, or for unified action.




At the end of the day such a society is no society at all; it is merely a disparate collection of quite distinct societies competing for limited resources. Insofar as this disparate collection is held together by liberal ideals, it is not held together as a society but in spite of its societies. And it is not at all clear why this particular collection of societies should attempt to coexist within a single governmental or legal structure. The logic of liberal secularism is thus at odds with itself.




But why describe this view as falsely modest? Because it conceals a whole series of metaphysical commitments that compete with those of ordinary religions and religious philosophies. It has been noted, for example, that the doctrine of “liberal neutrality” detaches the pursuit of justice from pursuit of the good. That is, it detaches means, which are public (the state’s distribution of resources, including “the social bases of respect and dignity”), from ends, which are private. In this way public actions can be justified independently of any particular”and, as such, sectarian”idea of the good, which is said to be necessary if the problem of pluralism is to be solved. Yet to sever the bond between justice and the good entails the reconstruction of both concepts; perhaps even the equating of the liberal concept of justice with the good.




“Liberal neutrality” is thus the product of a comprehensive (though not necessarily a coherent) worldview and so is not neutral at all. As Oliver O’Donovan remarks in The Desire of the Nations: “The false self-consciousness of the would-be secular society lies in its determination to conceal the religious judgments that it has made.” Indeed, liberal secularism conceals a deep antipathy toward its competitors. Its “pluralism” is a cover for hegemony. It has taken to heart the negative article of Rousseau’s own political creed”that intolerance is the one thing which must not be tolerated”and learned to deploy this article against those with firm convictions about the good that differ from its own.




The third approach to secular society”the eschatological”derives from the Jewish and the Christian conviction that we must contrast ourselves not so much to the era that preceded us as to that which is yet to come. This is not to be confused with utopianism, nor with a doctrine of manifest destiny, whether national or global, nor with a theocratic theory of the state. All such theories give to the state both too much and too little. Too much, because they mistake some formal feature of social life in the saeculum”for example, the rule of law, or the advance of democracy, or the triumph of techno-capitalism”for the main instrument by which the ideal society is to be achieved. Too little, because they do not require of society or of the state the pursuit of a de facto, but only of a de jure, truth, justice, righteousness, or freedom. The eschatological alternative makes neither of these mistakes.




On this view society is “secular” in the sense and to the degree not that it divides body from soul or means from ends, but to the extent that it is conscious of, and consistent with, its entirely provisional and temporary character. For it belongs to an age that is entirely provisional: an age with the highest possible expectations”the coming of the kingdom of God”and the gravest of reservations”an intuition of “the mystery of lawlessness” at work even within the law itself. Eschatological secularism does not respond by demanding a state that renders no metaphysical or religious judgments (this it regards as an impossibility) but rather by demanding a state that renders these judgments more openly, precisely so that they can be subjected to public scrutiny and debate. For on this view the religious judgments of the state are not and cannot be absolute, and their reach cannot be anything like comprehensive. Likewise, then, their enactment and enforcement must be tempered by a profound sense of provisionality, and tested accordingly. The Church may, if it will, say extra ecclesiam nulla salus, but the state may never make a comparable claim.




On the other hand, such a state enjoys a new freedom. Unlike its supposedly neutral counterpart, it does not have to play the nanny. It does not have to burden itself with the impossible task of appearing to balance every claim to public resources or respect. It is free to respond to the truth, and is not obligated to treat all would-be truths equally.




The objection will be made, however, that this “eschatological” alternative contains its own bid for hegemony. Do we really want a society that makes, or a state that operates on the basis of, religious commitments? Whose commitments? And which eschatology exactly will be in play? Do eschatologies not vary widely within religions, and between them? Is our society not generally skeptical of religious eschatologies? How can such things possibly be decided anyway? Is this just nostalgia for a past”a Christian past”that is no longer either accessible or desirable? A nostalgia that is in fact threatening to Jews, to Muslims, to agnostics? Should it not be dismissed out of hand for that reason, if for no other?




I would reply to these objections in two ways. First, by pressing the claim that there is no alternative to a society, or to a state, that makes religious judgments. It is true that the supersessionist approach tends to be hostile to much traditional religion. It is equally true that the liberal approach (while repudiating this hostility) intends to reject in principle any state-sanctioned religion. But that does not alter the fact that religious judgments will be made in both cases. Certainly negative judgments cannot be avoided. The state will operate as if it were not the case that Moses speaks for God, or that Jesus is Lord, or that the Koran is the highest authority in all human affairs, and so forth. But I have argued that positive judgments of a religious nature will also be made, since no society can evade the kind of questions that produce such judgments. Some warrant must be offered for every attempt (including those of political theorists) to regulate human affairs, and this warrant will sooner or later imply an anthropology or a worldview that rivals its “religious” predecessors and competitors in comprehensiveness. Moreover, certain rituals will in time be developed to signal and to reinforce this warrant and this worldview, since communal action requires such rituals.




So there is no way out. Neither the need to make religious decisions, nor the extreme difficulty in doing so, can be avoided. And these decisions, in one fashion or another, will have to be enforced, making perfectly clear that what we are dealing with is not apples and oranges (secular judgments and religious judgments) but apples and apples (secular judgments that are at the same time religious). The only way in which the illusion of apples and oranges, or the myth of liberal neutrality, or the so-called wall of separation can be maintained, even for a time, is by redefining religious judgments as purely “spiritual” in nature, that is, as private opinions or beliefs that have no bearing on public behavior. This, of course, is to define religion out of existence. But to define religion out of existence is to revert, after all, to the crudest kind of supersessionism.




Second, I would reply by reversing the charge that the eschatological model smacks of nostalgia for a lost Christian hegemony. The fact is that the eschatological model is not for hegemony at all, but for restraint. It will have to be allowed that the restraint it calls for”the restraint it urges upon every society and every worldly government”is one that is based upon a sense of provisionality that is informed by the particular religious traditions I have identified. It will also have to be allowed that these traditions have sometimes been at odds with each other, and with themselves, and that they have witnessed (or even been party to) quite disastrous attempts to interpret their basic eschatological convictions in ways that have effectively undermined political restraint. It will have to be confessed that, when this has happened, it has sometimes been accompanied by great evils, from which Jews especially (but not only Jews) have suffered.




This only helps to make the point. For it is in the messianic (that is, in the Jewish and Christian) traditions that the problem, and the possibility, of political restraint has come to the fore, generating these very debates about secularism. And it is in the intersection of these traditions with their Islamic counterpart”an intersection which lies, quite literally, in Jerusalem and the West Bank”that the problem will have ultimately to be solved. What I have suggested, however, is that neither supersessionist nor liberal secularism can solve it, whether in the Middle East or here at home, because both are hegemonic in nature. Ironically, it is they that represent the extension, not the correction, of what was false in Christendom.




This is not the place to say what eschatological secularism might contribute to the Middle East. But is this model either accessible or desirable even here at home? Our level of comfort with it might grow if we gave it another name: “deliberation secularism,” say, or “accountability secularism,” both of which would appeal to the best intentions of the supersessionist or of the liberal, and might point the way to common ground. Or, if it is not too hubristic, “modest secularism.” Whatever we called it, it would still be the same thing if it required of us the will to determine openly, by due process of civil debate, what are the goods (not merely the “values”) to which we ought to hold, without denying that the identification of these goods may in itself involve a collective act of religious judgment. And if it cultivated the appropriate caution in choosing means to pursue or protect these goods, so as to give, as far as possible, the benefit of the doubt to dissenting voices and communities.




For such a “modest” secularism, it is not necessary that our society be largely united in religion. It is not even necessary to have a certain sort of constitution or government. It is necessary only to have a government that fosters free debate, and that is prepared to allow that, since it is secular, it must act on the basis that it is accountable, not only to its citizens and to its laws, but to God. This means precisely that it does not imagine itself to be a theocracy. But it also means that, while it can be a government friendly to agnostics and atheists, it cannot be an atheistic government. “One nation under God,” or “founded upon principles that recognize the supremacy of God and the rule of law”: such constitutional references cannot become dead letters without modest secularism passing over into immodest secularism. As to exactly who or what is indicated by “God,” that will be debated between societies, and also within them, until God determines that the time for debating is done.





Douglas Farrow is Associate Professor of Christian Thought at McGill University in Montreal. An earlier version of this article was presented to the international conference on Pluralism, Religion, and Public Policy held at McGill in October 2002 .