Western Secularity - Rethinking Secularism - Craig J. Calhoun

Rethinking Secularism - Craig J. Calhoun (2011)

Chapter 1. Western Secularity

Charles Taylor


We live in a world in which ideas, institutions, artistic styles, and formulas for production and living circulate among societies and civilizations that are very different in their historical roots and traditional forms. Parliamentary democracy spread outward from England, among other countries, to India; likewise, the practice of nonviolent civil disobedience spread from its origins in the struggle for Indian independence to many other places, including the United States with Martin Luther King Jr. and the civil rights movement, Manila in 1983, and the Velvet and Orange Revolutions of our time.

But these ideas and forms of practice don’t just change place as solid blocks; they are modified, reinterpreted, given new meanings, in each transfer. This can lead to tremendous confusion when we try to follow these shifts and understand them. One such confusion comes from taking a word itself too seriously; the name may be the same, but the reality will often be different.

This is evident in the case of the word “secular.” We think of “secularization” as a selfsame process that can occur anywhere (and, according to some people, is occurring everywhere). And we think of secularist regimes as an option for any country, whether or not they are actually adopted. And certainly, these words crop up everywhere. But do they really mean the same thing in each iteration? Are there not, rather, subtle differences, which can bedevil cross-cultural discussions of these matters?

I think that there are and that they do create problems for our understanding. Either we stumble through tangles of cross-purposes, or else a rather minimal awareness of significant differences can lead us to draw far-reaching conclusions that are very far from the realities we seek to describe. Such is the case, for instance, when people argue that since the “secular” is an old category of Christian culture and since Islam doesn’t seem to have a corresponding category, therefore Islamic societies cannot adopt secular regimes. Obviously, they would not be just like those in Christendom, but maybe the idea, rather than being locally restricted, can travel across borders in an inventive and imaginative way.

Let’s look at some of the features of the “secular” as a category that developed within Latin Christendom. First, it was one term of a dyad. The secular had to do with the “century”—that is, with profane time—and it was contrasted with what related to the eternal, or to sacred time.1 Certain times, places, persons, institutions, and actions were seen as closely related to the sacred or higher time, and others were seen as pertaining to profane time alone. That’s why the same distinction could often be made by use of the dyad “spiritual/temporal” (e.g., the state as the “temporal arm” of the church). Ordinary parish priests are thus “secular” because they operate out there in the “century,” as against those in monastic institutions—“regular” priests—who live by the rules of their order.

So there is an obvious meaning of “secularization” that dates from the aftermath of the Reformation. It refers specifically, in this sense, to when certain functions, properties, and institutions were transferred from church control to that of laymen.

These moves were originally made within a system held in place by the overarching dyad; things were moved from one niche to another within a standing system of niches. This configuration of the “secular,” where it still holds, can make secularization a relatively undramatic affair, a rearrangement of the furniture in a space whose basic features remain unchanged.

But from the seventeenth century on, a new possibility gradually arose: a conception of social life in which the “secular” was all there was. Since “secular” originally referred to profane or ordinary time, in contradistinction to higher times, what was necessary was to come to understand profane time without any reference to higher times. The word could go on being used, but its meaning was profoundly changed, because its counterpoint had been fundamentally altered. The contrast was no longer with another temporal dimension, in which “spiritual” institutions had their niche; rather, the secular was, in its new sense, opposed to any claim made in the name of something transcendent of this world and its interests. Needless to say, those who imagined a “secular” world in this sense saw such claims as ultimately unfounded and only to be tolerated to the extent that they did not challenge the interests of worldly powers and human well-being.

Because many people went on believing in the transcendent, however, it was necessary for churches to continue to have a place in the social order. They could be essential in their own way to the functioning of society, but this functioning was to be understood exclusively in terms of “this-worldly” goals and values (peace, prosperity, growth, flourishing, etc.).

This shift entailed two important changes: first, it brought a new conception of good social and political order, which was unconnected to either the traditional ethics of the good life or the specifically Christian notion of perfection (sainthood). This was the new post-Grotian idea of a society formed of and by individuals in order to meet their needs for security and the means to life. The criterion of a good society in this outlook, mutual benefit, was not only emphatically “this-worldly” but also unconcerned with “virtue” in the traditional sense.

The hiving off of a specifically “earthly” criterion figured within a broader distinction, that which divided “this world,” or the immanent, from the transcendent. This very clear-cut distinction is itself a product of the development of Latin Christendom and has become part of our way of seeing things in the West. We tend to apply it universally, even though no distinction this hard and fast has existed in any other human culture in history. What does seem, indeed, to exist universally is some distinction between higher beings (spirits) and realms and the everyday world we see immediately around us. But these are not usually sorted out into two distinct domains, such that the lower one can be taken as a system understandable purely in its own terms. Rather, the levels usually interpenetrate, and the lower cannot be understood without reference to the higher. To take an example from the realm of philosophy, for Plato, the existence and development of the things around us can only be understood in terms of their corresponding Ideas, and these exist in a realm outside time. The clear separation of an immanent from a transcendent order is one of the inventions (for better or worse) of Latin Christendom.

The new understanding of the secular that I have been describing builds on this separation. It affirms, in effect, that the “lower,” immanent or secular, order is all that there is and that the higher, or transcendent, is a human invention. Obviously, the prior invention of a clear-cut distinction between these levels prepared the ground for the “declaration of independence” of the immanent.

At first, the independence claimed on the part of the immanent was limited and partial. In the “Deist” version of this claim, widespread in the eighteenth century, God was seen as the artificer of the immanent order. Since he is the creator, the natural order stands as a proof of his existence; and since the proper human order of mutual benefit is one that he designs and recommends, we follow his will in building it. Furthermore, it is still affirmed that he backs up his law with the rewards and punishments of the next life.

Thus, some religion, or a certain piety, is a necessary condition of good order. Locke will thus exclude from toleration not only Catholics but also atheists. This is the positive relation of God to good order, but religion can also have negative effects. Religious authority can enter into competition with secular rulers; it can demand things of the faithful that go beyond, or even against, the demands of good order; it can make irrational claims. So it remains to purge society of “superstition,” “fanaticism,” and “enthusiasm.”

The attempts of eighteenth-century “enlightened” rulers, such as Frederick the Great and Joseph II, to “rationalize” religious institutions—in effect, treating the church as a department of the state—belong to this earlier phase of secularization in the West. So, too, in a quite different fashion, does the founding of the American republic, with its separation of church and state. But the first unambiguous assertion of the self-sufficiency of the secular came with the radical phases of the French Revolution.

The polemical assertion of secularity returns in the Third Republic, whose laïcité is founded on the ideas of the self-sufficiency of the secular and the exclusion of religion. Marcel Gauchet shows how Renouvier laid the grounds for the outlook of the Third Republic radicals in their battle against the church. The state has to be “moral et enseignant.” It has “charge d’âmes aussi bien que toute Église ou communauté, mais à titre plus universel.” Morality is the key criterion. In order not to be subordinate to the church, the state must have “une morale indépendante de toute religion” and must enjoy a “suprématie morale” in relation to all religions. The basis of this morality is liberty, and in order to hold its own before religion, the morality legitimizing the state has to be based on more than just utility or feeling; it needs a “théologie rationnelle,” like that of Kant.2

Needless to say, this spirit goes marching on in contemporary France, as one can see in the ongoing debate over banning the Muslim head scarf. The insistence is still that the public spaces in which citizens meet must be purified of any religious reference.

And so the history of this term “secular” in the West is complex and ambiguous. It starts off as one term in a dyad that distinguishes two dimensions of existence, identifying them by the particular type of time that is essential to each. But from the foundation of this clear distinction between the immanent and the transcendent, there develops another dyad, in which “secular” refers to what pertains to a self-sufficient, immanent sphere and is contrasted with what relates to the transcendent realm (often identified as “religious”). This binary can then undergo a further mutation, via a denial of the transcendent level, into a dyad in which one term refers to the real (“secular”), and the other refers to what is merely invented (“religious”); or where “secular” refers to the institutions we really require to live in “this world,” and “religious” or “ecclesial” refers to optional accessories, which often disturb the course of this-worldly life.

Through this double mutation, the dyad itself is profoundly transformed; in the first case, both sides are real and indispensable dimensions of life and society. The dyad is thus “internal,” in the sense that each term is impossible without the other, like right and left or up and down. After the mutations, the dyad becomes “external”; secular and religious are opposed as true and false or necessary and superfluous. The goal of policy becomes, in many cases, to abolish one while conserving the other.

In some ways, the post-Deist modes of secularism transpose features of the Deist template described above. In the Jacobin outlook, the designer is now nature, and so the “piety” required is a humanist ideology based on the natural. What is unacceptable, in turn, is any form of “public” religion. Faith must be relegated to the private sphere. Following this view, there must be a coherent morale indépendante, a self-sufficient social morality without transcendent reference. This demand, in turn, encourages the idea that there is such a thing as “reason alone” (die blosse Vernunft), that is, reason unaided by any “extra” premises derived from Revelation or any other allegedly transcendent source. Variants of these claims resurface often in contemporary discussions of secularism in the West.3

The Deist template has helped to define “good,” or “acceptable,” religion for much of the Western discussion of the last few centuries. A good, or proper, religion is a set of beliefs in God or some other transcendent power, which entails an acceptable or, in some versions, a “rational” morality. It is devoid of any elements that do not contribute to this morality and thus of “superstition.” It is also necessarily opposed to “fanaticism” and “enthusiasm,” because these involve by definition a challenge by religious authority to what “reason alone” shows to be the proper order of society.4

Religion can thus be an aid to social order by inculcating the right principles, but it must avoid becoming a threat to this order by launching a challenge against it. Thus, Locke is ready to tolerate various religious views, but he excepts from this benign treatment atheists (whose nonbelief in an afterlife undermines their readiness to keep their promises and respect good order) and Catholics (who could not but challenge the established order).

In both of these ways, positive and negative, the essential impact of good religion takes place in foro interno: on one hand, it generates the right moral motivation; on the other, by remaining within the mind and soul of the subject, it refrains from challenging the external order. So public ritual can be an essential element of this “rational” religion only if it can help by celebrating public order or by stimulating inner moral motivation.

Eventually, this constellation of terms, including “secular” and “religious,” with all of its baggage of ambiguity and deep assumptions concerning the clear division between immanent and transcendent on one hand and public and private on the other, begins to travel. It is no surprise, then, that it causes immense confusion. Westerners themselves are frequently confused about their own history. One way of understanding the development of Western secularism is to see the separation of church and state and the removal of religion into a “private” sphere where it cannot interfere with public life as a result of the earlier distinction between the secular (or temporal) and the sacred (or eternal). The former would thus be, in retrospect, the ultimately satisfactory solution, whereby religion is finally relegated to the margins of political life.

But these stages are not clearly distinguished.5 Thus, American secularists often totally confuse the separation of church and state from that of religion and state (Rawls at one point wanted to ban all reference to the grounds of people’s “comprehensive views”—these included religious views—from public discourse. Moreover, the whole constellation generates disastrously ethnocentric judgments. If the canonical background for a satisfactory secularist regime is the three-stage history described above—distinction of church and state, separation of church and state, and, finally, sidelining of religion from the state and from public life—then obviously, Islamic societies can never make it.

Similarly, one often hears the judgment that Chinese imperial society was already “secular,” totally ignoring the tremendous role played by the immanent/transcendent split in the Western concept, which had no analogue in traditional China. Ashis Nandy, in discussing the problems that arise out of the multiple uses of the term “secular,” shows the confusions that are often involved in analogous statements about the Indian case (e.g., that the emperor Asoka was “secular” or that the Mughal emperor Akbar established a “secular” form of rule).

But this kind of (mis-)statement can also reflect a certain wisdom. In fact, Nandy distinguishes two quite different notions that consciously or unconsciously inform the Indian discussion. There is, first, the “scientific-rational” sense of the term, in which secularism is closely identified with modernity, and, second, a variety of “accommodative” meanings, which are rooted in indigenous traditions. The first attempts to free public life from religion; the second seek rather to open spaces “for a continuous dialogue among religious traditions and between the religious and the secular.”6

The invocation of Akbar’s rule as “secular” can thus function as a creative and productive way of redefining the term. Such redefinitions, starting from the problems that contemporary societies have to solve, often conceive of secularity as an attempt to find fair and harmonious modes of coexistence between religious communities and leave the connotations of the word “secular” as they have evolved through Western history quietly to the side. This takes account of the fact that formulas for mutually beneficent living together have evolved in many different religious traditions and are not the monopoly of those whose outlook has been formed by the modern, Western dyad, in which the secular lays claim to exclusive reality.7


What to do? We might think of starting again with another term, one less identified with a particular civilizational trajectory. But that is probably utopian. The word “secular” is much too entrenched in all sorts of discussion, historical and normative, to be displaced.

Obviously, we need a great deal of close study of other, non-Western contexts in order to help here, and I find myself very ill equipped to offer further useful contributions. But there might be a point in trying to give a more fine-grained account of the Western trajectory, so fine-grained that one would both lose any remaining temptation to see it as the universal road on which humanity as a whole is embarked and gain some interesting points of contrast with other civilizational histories. This is the project on which I have been embarked in recent years,8 and I turn in the rest of this chapter to offer a sketch of this Western path and, in particular, of the important role that the construct we often refer to as Deism has played in it.

One of the main vectors during the last six or seven centuries in this civilization has been a steadily increasing emphasis on a religion of personal commitment and devotion, as opposed to forms centered on collective ritual. We can see this in the growth of a more Christocentric religion in the High Middle Ages. It is further evident both in devotional movements and associations, such as the Brethren of the Common Life in the fifteenth century, and in the nature of the demands made by church hierarchies and leaders on their members. An early example of the latter is the decision of the Lateran Council in 1215 to require all of the faithful to confess to a priest and be shriven, so as to receive communion at least once a year.

From that point on, the pressure to adopt a more personal, committed, and inward form of religion continues—through the preaching of the mendicant friars and others, through the devotional movements mentioned above—eventually reaching a new stage with the Reformation. The point of declaring that salvation comes through faith was radically to devalue ritual and external practice in favor of inner acknowledgment of Christ as savior. It was not just that external ritual was of no effect, but relying on it was tantamount to a presumption that we could control God. The Reformation also tended to delegitimize the distinction between fully committed believers and other, less devoted ones. As against a view of the church in which people operated at many different “speeds,” with religious “virtuosi,” to use Max Weber’s term on one end and ordinary intermittent practitioners on the other, all Christians were expected to be fully committed.

But this movement toward the personal, committed, and inward didn’t exist only in the Protestant Churches. There is a parallel development in the Counter-Reformation, with the spread of different devotional movements and the attempts to regulate the lives of the laity according to increasingly stringent models of practice. The clergy were reformed, their training was upgraded, and they were expected, in turn, to reach out and demand a higher level of personal practice from their flocks. A striking figure illustrates this whole movement. In the history of Catholic France, the moment at which the level of practice, as measured by baptisms and Easter communions, reached its highest has been estimated to fall around 1880.9 This is well after the anticlericalism of the Revolution and its attempts at dechristianization and after a definite movement toward unbelief had set in among the educated classes. In spite of this incipient loss, the apogee of practice came this late because it stood at the end of a long process in which ordinary believers had been preached at, organized, and sometimes bullied into patterns of practice that reflected heightened personal commitment.

They had been pressed, we might be tempted to say, into “taking their religion seriously.” To take my religion seriously is to take it personally—that is, more devotionally, inwardly, and committedly. Just taking part in external rituals—those that don’t require the kind of personal engagement that, for example, auricular confession, with its self-examination and promises of amendment, entails—is devalued on this understanding. That is no longer what religion is really about.

Now, a striking feature of the Western march toward secularity is that it has been interwoven from the start with this drive toward a personal religion, as has frequently been remarked.10 The connections are multiple. It is not just that the falling off of religious belief and practice has forced a greater degree of reflection and commitment on those who remain faithful (this has perhaps become more evident in more recent times). It is much more that the drive to personal religion has itself been part of the impetus toward different facets of secularization. It was this drive, for instance, that powerfully contributed to the disenchantment of the world of spirits and higher forces in which our ancestors lived. The Reformation and the Counter-Reformation repressed magical practices and then those facets of traditional Christian sacramental ritual that they began to deem magical (for Calvinists, this even included the mass). Later, at the time of the early American republic, a separation of church and state was brought about, mainly to make space for, and avoid the contamination of, personal religion, which itself had been given further impetus through the Great Awakening.

We might identify two closely connected vectors here: toward personal commitment and toward the repression of what came to be understood as the “magical” elements in religion: practices that suppose and draw on various intracosmic spirits, good or bad, and higher powers inhering in things (relics, for instance). I want to use the word “disenchantment” for this movement of repression; this is a narrower sense than the one the word often bears, for it is frequently synonymous with the sidelining of religion as such, but my usage has some warrant in the original Weberian term Entzauberung.

Everyone can agree that one of the big differences between us and our ancestors of 500 years ago is that they lived in an “enchanted” world and we do not. We might think of this as our having “lost” a number of beliefs and the practices they made possible. Essentially, we become modern by breaking out of “superstition” and becoming more scientific and technological in our stance toward our world. But I want to accentuate something different. The “enchanted” world was one in which spirits and forces defined by their meanings (the kind of forces possessed by love potions or relics) played a big role. But more, the enchanted world was one in which these forces could shape our lives, both psychical and physical. One of the big differences between our forerunners and us is that we live with a much firmer sense of the boundary between self and other. We are “buffered” selves. We have changed. We sometimes find it hard to be frightened the way they were, and, indeed, we tend to invoke the uncanny things they feared with a pleasurable frisson, as if sitting through films about witches and sorcerers. They would have found this incomprehensible.

Here you see the difference between a subtraction story and one that considers not only loss but also remaking.11 On the subtraction story, there can be no epistemic loss involved in the transition from enchantment to disenchantment; we have just shucked off some false beliefs, baseless fears, and imagined objects. Looked at my way, the process of disenchantment involves a change in sensibility; one is open to different things, yet one has lost one important way in which people used to experience the world.

It is this sense of loss that underlies many attempts in our day to “reenchant” the world. This goal is frequently invoked, but it ought to be clear that what would be regained here is not what we have “lost.”12 People are talking of quite other ways of recovering an analogue of the original sensibility, whether in the sense of the forces moving through nature in the poems of Hölderlin or Wordsworth or through contact with spirits of the dead.

Disenchantment in my use really translates Weber’s term Entzauberung, the kernel concept of which is Zauber, or “magic.” In a sense, moderns constructed their own concept of magic from and through the process of disenchantment. Carried out first under the auspices of Reformed Christianity, the condemned practices all involved using spiritual forces against, or at least independently of, our relation to God. The worst examples were things such as saying a black mass for the dead to kill off your enemy or using the communion host as a love charm. But in the more exigent modes of reform, the distinction between white and black magic tended to disappear, and all recourse to forces independent of God was seen as culpable. The category “magic” was constituted through this rejection, and this distinction was then handed on to post-Enlightenment anthropology, as with Frazer’s distinction between “magic” and “religion.”13

The process of disenchantment, which involved a change in us, can be seen as the loss of a certain sensibility, which is really an impoverishment (as opposed to the simple shedding of irrational feelings). And there have been frequent attempts to “reenchant” the world, or at least admonitions and invitations to do so. In a sense, the Romantic movement can be seen as engaged in such a project. Think of Novalis’s “magic realism”; think of the depiction of the Newtonian universe as a dead one, shorn of the life it used to have (Schiller’s “The Gods of Greece”).

But it is clear that the poetry of Wordsworth, or of Novalis or Rilke, can’t come close to the original experience of porous selfhood. The experience it evokes is more fragile, often evanescent, and subject to doubt. It also draws on an ontology that is highly undetermined and must remain so.14

Indeed, “enchantment” is something that we, the products of the first vector—toward the personal, the committed, and the inward—have special trouble understanding. In Latin Christendom, movement along this vector increasingly tended to privilege belief, as opposed to unthinking practice. “Secular” people have inherited this emphasis and often propound an “ethics of belief.”15 So we tend to think of our differences from our remote forebears in terms of different beliefs, whereas there is something much more puzzling involved here. It is clear that for our forebears, and for many people in the world today who live in a similar religious world, the presence of spirits and of different forms of possession is no more a matter of (optional, voluntarily embraced) belief than the presence of this computer and its keyboard at the tips of my fingers is for me. There is a great deal that I don’t understand about the inner workings of this computer (almost everything, in fact) and about which I could be induced by experts to accept various theories; but the encounter with a computer is not a matter of “belief”—it’s a basic feature of my experience.

So it must have been for an elderly woman named Celestine, interviewed by Birgit Meyer, who, as a young woman, “walked home from Aventile with her mother, accompanied by a stranger dressed in a white northern gown.”16When asked afterward, her mother denied having seen the man. He turned out to be the Akan spirit Sowlui, and Celestine was pressed into his service. In Celestine’s world, the identification of the man with this spirit might be called a “belief,” in that it came after the experience in an attempt to explain what it was all about, but the man accompanying her was just something that happened to her, a fact of her world.

We have great trouble getting our minds around this, and we rapidly reach for intrapsychic explanations, in terms of delusions, projections, and the like. But one thing that seems clear is that the whole situation of the self in experience is subtly, but significantly, different in these worlds and in ours. We make a sharp distinction between inner and outer, between what is in the “mind” and what is out there in the world. Whatever has to do with thought, purpose, and human meanings has to be in the mind, rather than in the world. Some chemical can cause hormonal changes and thus alter the psyche. There can be an aphrodisiac but not a love potion, that is, a chemical that determines the human, or moral, meaning of the experience it enables. A vial of liquid can cure a specific disease, but there can’t be such vials as those brought back from pilgrimage at Canterbury, which contained a minuscule drop of the blood of Thomas à Becket and which could cure anything and even make us better people; that is, the liquid was the locus not of certain specific chemical properties but of a generalized beneficence.

Modern Westerners have a clear and firm boundary between mind and world, even mind and body. Moral and other meanings are “in the mind”; they cannot reside outside. But it was not so formerly. Let us take a well-known example of influence inhering in an inanimate substance, as this was understood in earlier times. Consider melancholy: black bile was not the cause of melancholy; rather, it embodied, it was, melancholy. The emotional life was porous here; it didn’t simply exist in an inner, mental space. Our vulnerability to the evil and the inwardly destructive extended to more than just malevolent spirits; it went beyond them to things that have no wills but are nevertheless redolent with evil meanings.

See the contrast. A modern is feeling depressed, melancholy. He is told it’s just his body chemistry; he’s hungry, or there is a hormone malfunction, or whatever. Straightaway, he feels relieved. He can take a distance from this feeling, which is declared ipso facto unjustified. Things don’t really have such a meaning; it just feels that way, which is the result of a causal process utterly unrelated to the meanings of things. This disengagement depends on our modern mind/body distinction and the relegation of the physical to being “just” a contingent cause of the psychical.

But a premodern might not be helped by learning that his mood comes from black bile, because this doesn’t permit any distancing. Black bile is melancholy; now he just knows that he’s in the grip of the real thing.

Here is the contrast between the modern, bounded self—I want to say “buffered” self—and the “porous” self of the earlier, enchanted world. What difference does this make? It makes, in short, for a very different existential condition. The last example, about melancholy and its causes, illustrates this well. For the modern, buffered self, the possibility exists of taking a distance, disengaging, from everything outside the mind. My ultimate purposes are those that arise within me; the crucial meanings of things are those defined in my responses to them. These purposes and meanings may be vulnerable to manipulation in various ways, including the use of chemicals, but this can, in principle, be met with a countermanipulation: I avoid distressing or tempting experiences, I don’t shoot up the wrong substances, and so on.

This is not to say that the buffered self-understanding requires that one take this stance, but it does allow it as a possibility, whereas the porous one does not. By definition, for the porous self, the source of its most powerful and important emotions are outside the “mind,” or, better put, the very notion that there is a clear boundary, allowing us to define an inner base area, grounded in which we can disengage from everything else, has no sense.

As a bounded self, I see the boundary as a buffer, such that the things beyond don’t need to “get to me,” to use the contemporary expression. That’s the sense of my use of the term “buffered” here. This self can see itself as invulnerable, as master of the meanings that things have for it.

These two descriptions get at the two important facets of this contrast. First, the porous self is vulnerable (to spirits, demons, cosmic forces, etc.), and along with this go certain fears, which can grip it in the right circumstances. The buffered self has been taken out of the world of this kind of fear—removed, for instance, from the sorts of terrors vividly portrayed in some of Bosch’s paintings.

It is true that something analogous can take its place. These images can also be seen as coded manifestations of inner depths, of repressed thoughts and feelings. But the point is that in this quite transformed understanding of self and world, we define these as inner, and, naturally, we deal with them very differently. Indeed, an important part of this treatment of the self is designed to make disengagement possible.

Perhaps the clearest sign of this transformation in our world is that many people today look back to the world of the porous self with nostalgia, as though the creation of a thick emotional boundary between us and the cosmos were now lived as a loss. The aim is to try to recover some measure of this lost feeling. So people go to movies about the uncanny in order to experience a frisson. Our peasant ancestors would have thought us insane. You can’t get a frisson from what is really terrifying you.

The second facet is that the buffered self can form the aspiration of disengaging from whatever is beyond the boundary and of giving its own autonomous order to its life. The absence of fear can be not just enjoyed but also seen as an opportunity for self-control or self-direction.

And so the boundary between agents and forces is fuzzy in the enchanted world, and the boundary between mind and world is porous, as we see in the way in which charged objects could influence us. I have been speaking about the moral influence of substances, such as black bile, but a similar point can be made about the relation to spirits. The porosity of the boundary emerges here in the various kinds of “possession,” all the way from a full taking over of the person, as with a medium, to various kinds of domination by, or partial fusion with, a spirit or God.17 Here, again, the boundary between self and other is fuzzy, or porous, and this has to be seen as a fact of experience, not a matter of “theory” or “belief.”


I want now to place this double vector (commitment-disenchantment) in an even deeper and broader historical context, that of the rise and forward march of what Jaspers called “axial” religions and spiritualities. The whole sweep, as it continues up to and into Western modernity, can be seen as a great disembedding of the “merely human,” even of the human individual.18 The full scale of this millennial change becomes clearer if we focus first on some features of the religious life of earlier, smaller-scale societies, insofar as we can trace these. There must have been a phase in which all humans lived in such small-scale societies, even though much of the character of life in this epoch can only be guessed at.

But if we focus on what I will call early religion (which partly covers what Robert Bellah, for instance, calls “archaic religion”), we note how profoundly these forms of life “embed” the agent and that they do so in three crucial ways.19

The first way is socially. In Paleolithic and even certain Neolithic tribal societies, religious life was inseparably linked with social life. The primary agency for important religious action—invoking, praying to, sacrificing to, or propitiating gods or spirits; coming close to these powers; receiving healing and protection from them; divining under their guidance, and so on—was the social group as a whole or some more specialized agency recognized as acting for the group. In early religion, we primarily relate to God as a society.

We see both aspects of this in, for example, ritual sacrifices among the Dinka, as they were described a half-century ago by Godfrey Lienhardt. On one hand, the major agents of the sacrifice, the “masters of the fishing spear,” were in a sense “functionaries,” acting for the whole society; while on the other, the whole community became involved, repeating the invocations of the masters until everyone’s attention was focused and concentrated on the single ritual action. It was at the climax that those attending the ceremony were “most palpably members of a single undifferentiated body.” This participation often took the form of possession by the divinity being invoked.20

Nor is this just the way things happen to be in a certain community. This collective action is essential for the efficacy of the ritual. You can’t mount a powerful invocation of the divinities this way on your own in the Dinka world. In Lienhardt’s words, the “importance of corporate action by a community of which the individual is really and traditionally a member is the reason for the fear which individual Dinka feel when they suffer misfortune away from home and kin.”21

This kind of collective ritual action, in which the principal agents act on behalf of a community, which also becomes involved in its own way in the action, seems to figure virtually everywhere in early religion and continues in some ways up to our day. It certainly goes on occupying an important place as long as people live in an “enchanted” world—the world of spirits and forces, which is prior to what I am calling “disenchantment.” The medieval ceremony of “beating the bounds” of the agricultural village, for instance, involved the whole parish and could only be effective as a collective act of this whole.

This embedding in social ritual usually carries with it another feature. Just because the most important religious action was that of the collective, and because it often required that certain functionaries—priests, shamans, medicine men, diviners, chiefs, and so on—fill crucial roles in the action, the social order in which these roles were defined tended to be sacrosanct. This is, of course, the aspect of religious life that was most centrally identified and pilloried by the radical Enlightenment. The crime laid bare here was the entrenchment of forms of inequality, domination, and exploitation through their identification with the untouchable, sacred structure of things; hence the longing to see the day “when the last king had been strangled in the entrails of the last priest.” But this identification is, in fact, very old and goes back to a time when many of the later, more egregious and vicious forms of inequality had not yet been developed, before there were kings and hierarchies of priests.22

Behind the issue of inequality and justice lies something deeper, which touches what we would call today the “identity” of the human beings in those earlier societies. Just because their most important actions were those of whole groups (tribe, clan, subtribe, lineage), articulated in a certain way (the actions were led by chiefs, shamans, masters of the fishing spear), they couldn’t conceive of themselves as potentially disconnected from this social matrix. It would probably never even occur to them to try.

What I’m calling social embeddedness is thus partly a matter of identity. From the standpoint of the individual’s sense of self, it means the inability to imagine oneself outside of a certain matrix. But it also can be understood as a social reality, and in this sense, it refers to the way we collectively imagine our social existence—for instance, that our most important actions are those of the whole society, which must be structured in a certain way so as to carry them out. And we can see that it is growing up in a world where this kind of social imaginary reigns that sets the limits to our sense of self.

Embedding thus pertains to society, but this also brings with it an embedding in the cosmos. For in early religion, the spirits and forces with which we are dealing are involved in the world in numerous ways. We can see examples of this if we refer back to the enchanted world of our medieval ancestors. For all that the God they worshipped transcended the world, they nevertheless also trafficked with intracosmic spirits and dealt with causal powers that were embedded in things: relics, sacred places, and the like. In early religion, even the high gods are often identified with certain features of the world; and where the phenomenon that has come to be called totemism exists, we can even say that some feature of the world—an animal or plant species, for instance—is central to the identity of a group.23 It may even be that a particular geographical terrain is essential to religious life. Certain places are sacred. Or the layout of the land speaks to us of the original disposition of things in sacred time. We relate to the ancestors and to this higher time through this landscape.24

Besides this relation to society and the cosmos, there is a third form of embedding in existing reality, which we see in early religion. This is what makes for the most striking contrast with what we tend to think of as the “higher” religions. What the people ask for when they invoke or placate divinities and powers is prosperity, health, long life, and fertility; what they ask to be preserved from is disease, dearth, sterility, and premature death. There is a certain understanding of human flourishing here that we can immediately understand, and that, however much we might want to add to it, seems to us quite “natural.” What there isn’t, and what seems central to the later (“higher”) religions, is the idea that we have to question radically this ordinary understanding, that we are called in some way to go beyond it.

This is not to say that human flourishing is the end sought by all things. The divine may also have other purposes, some of which have harmful impacts on us. There is a sense in which, for early religions, the divine is never simply well disposed toward us; the gods (or some of them) may also be in certain ways indifferent. or there may also be hostility, jealousy, or anger, which we have to deflect. Although benevolence, in principle, may have the upper hand, this process may have to be helped along by propitiation or even by the action of “trickster” figures. But through all of this, what remains true is that the divinity’s benign purposes are defined in terms of ordinary human flourishing. Again, there may be capacities that some people can attain, which go way beyond the ordinary human ones, which, say, prophets or shamans have. But these, in the end, subserve human well-being as ordinarily understood.

By contrast, with Christianity or Buddhism, for instance, there is a notion of the good that goes beyond human flourishing, which we may gain even while failing utterly in terms of the latter and even through such a failing (such as dying young on a cross) or which involves leaving the field of flourishing altogether (ending the cycle of rebirth). The paradox of Christianity, in relation to early religion, is that on one hand, it seems to assert the unconditional benevolence of God toward humans—there is none of the ambivalence of early divinity in this respect—and yet, on the other, it redefines our ends so as to take us beyond flourishing.

In this respect, early religion has something in common with modern exclusive humanism, and this has been felt and expressed in the sympathy of many modern post-Enlightenment people for “paganism.” “Pagan self-assertion,” John Stuart Mill thought, was as valid as, if not more valid than, “Christian self-denial.”25 (This is related to, but not quite the same as, the sympathy felt for “polytheism.”) What makes modern humanism unprecedented, of course, is the idea that this flourishing involves no relation to anything higher.26

Now, as earlier mentions suggest, I have been speaking of “early religion” in contrast with what many people have called “postaxial” religions.27 The reference is to what Karl Jaspers called the “axial age,”28 the extraordinary period in the last millennium BCE when various “higher” forms of religion appeared, seemingly independently in different civilizations, marked by such founding figures as Confucius, Gautama, Socrates, and the Hebrew prophets.

The surprising feature of the axial religions, compared with what went before—what, in other words, would have made them hard to predict before-hand—is that they initiate a break in all three dimensions of embeddedness: social order, cosmos, and human good. Yet this is not so in all cases and all at once; perhaps in some ways, Buddhism is the most far-reaching, because it radically undercuts the second dimension: the order of the world itself is called into question, because the wheel of rebirth means suffering. In Christianity, there is something analogous: the world is disordered and must be made anew. But some postaxial outlooks keep the sense of relation to an ordered cosmos, as we see in very different ways with Confucius and Plato; however, they mark a distinction between this and the actual, highly imperfect social order, so that the close link to the cosmos through collective religious life is made problematic.

Perhaps most fundamental of all is the revisionary stance toward the human good in axial religions. More or less radically, they all call into question the received, seemingly unquestionable understandings of human flourishing and hence, inevitably, the structures of society and the features of the cosmos through which this flourishing was supposedly achieved.

We might try to put the contrast in this way: unlike postaxial religion, early religion involved an acceptance of the order of things, in the three dimensions I have been discussing. In a remarkable series of articles on Australian aboriginal religion, W. E. H. Stanner speaks of “the mood of assent” that is central to this spirituality. Aboriginals had not set up the “kind of quarrel with life” that springs from the various postaxial religious initiatives.29 The contrast is in some ways easy to miss, because aboriginal mythology, in relating the way in which the order of things came to be in the Dream Time—the original time out of time, which is also “everywhen”—contains a number of stories of catastrophe, brought on by trickery, deceit, and violence, from which human life recouped and reemerged but in an impaired and divided fashion, so that there remains the intrinsic connection between life and suffering, and unity is inseparable from division. Now, this may seem reminiscent of other stories of a fall, including that related in Genesis; but in contrast with what Christianity has made of this last, for the Aboriginals, the imperative to “follow up” the dreaming, to recover through ritual and insight their contact with the order of the original time, relates to this riven and impaired dispensation, in which good and evil are interwoven. There is no question of reparation of the original rift or of a compensation for the original loss. Moreover, ritual and the wisdom that goes with it can even bring them to accept the inexorable and to “celebrate joyously what could not be changed.”30 The original catastrophe doesn’t separate or alienate us from the sacred or higher, as in the Genesis story; rather, it contributes to shaping the sacred order we are trying to “follow up.”31

Now, axial religion didn’t do away with early religious life. In many ways, it modified features of the latter to define majority religious life for centuries. Modifications arose, of course, not just from the axial formulations but also from the growth of larger-scale, more differentiated, and often urban-centered societies, with more hierarchical forms of organization and embryonic state structures. Indeed, it has been argued that these, too, played a part in the process of disembedding, because the very existence of state power entails some attempt to control and shape religious life and the social structures it requires and thus undercuts the sense of intangibility surrounding this life and these structures.32 I think that there is a lot to this thesis, and, indeed, I invoke something like it below, but for the moment, I want to focus on the significance of the axial period.

This doesn’t totally change the religious life of whole societies all at once, but it does open new possibilities of disembedded religion: seeking a relation to the divine or the higher, which severely revises the going notions of flourishing, or even goes beyond them, and can be carried through by individuals on their own and/or through new forms of sociality, unlinked to the established sacred order. So monks, bhikhus, sanyassi, or devotees of some avatar or God strike out on their own, and from this springs unprecedented modes of sociality: initiation groups, sects of devotees, the sangha, monastic orders, and so on.

In all of these cases, there is some kind of hiatus, difference, or even break in relation to the religious life of the whole society. This itself may be to some extent differentiated, with different strata or castes or classes, and a new religious outlook may lodge itself in one of them. But very often, a new devotion may cut across all of these, particularly where there is a break in the third dimension, with a “higher” idea of the human good emerging.

There is inevitably a tension here, but there often is also an attempt to secure the unity of the whole, to recover some sense of complementarity among the different religious forms, so that those who are fully dedicated to the “higher” forms can be seen at once as a standing reproach to those who remain in the earlier forms, supplicating the powers for human flourishing, and nonetheless as in a relationship of mutual help with them. The laity feed the monks, and by this they earn “merit,” which can be understood as taking them a little farther along the “higher” road, but it also serves to protect them against the dangers of life and increases their health, prosperity, and fertility.

So strong is the pull toward complementarity that even in those cases in which a “higher” religion took over the whole society—as we see with Buddhism, Christianity, and Islam—and there is nothing supposedly left to contrast with, the difference between dedicated minorities of religious “virtuosi” and the mass religion of the social sacred, still largely oriented toward flourishing, survived or reconstituted itself, with the same combination of tension on one hand and hierarchical complementarity on the other.

One can argue that all of the “higher” civilizations experienced similar tensions between axial spiritualities and earlier religious forms but that these took on a particular nature, and frequently a greater intensity, in ancient Judaism and the religions that sprang from it, including Christianity and Islam. The ban on idolatry (or shirk, in Islam) can generate a drive to reform, even to abolish, earlier cults as modes of forbidden, or “false,” religion, generating (in Christendom) the category of illicit “magic” that I alluded to above. We don’t seem to find an analogous animus against popular cults in India or China until relatively recently, and then as part of a response to real or threatened invasion by Western imperial powers.33 This important difference may be seen as a challenge to the whole category of “axial revolutions” as a class of similar transformations occurring in quite different societies, but we can’t go into this question here.

In any case, from our modern perspective, with 20/20 hindsight, it might appear as though the axial spiritualities were prevented from producing their full disembedding effect because they were, so to speak, hemmed in by the force of the majority religious life, which remained firmly in the old mold. They did bring about a certain form of religious individualism, but this was what Louis Dumont called the charter for “l’individu hors du monde”—that is, it was the way of life of elite minorities, and it was in some ways marginal to, or in some tension with, the “world,” where this means not just the cosmos, which is ordered in relation to the higher or the sacred, but also society, which is ordered in relation to both the cosmos and the sacred.34 This “world” was still a matrix of embeddedness, and it still provided the inescapable framework for social life, including that of the individuals who tried to turn their backs on it, insofar as they remained, in some sense, within its reach.

What had yet to happen was for this matrix to be transformed, to be made over according to some of the principles of axial spirituality, so that the “world” itself would come to be seen as constituted by individuals. This would be the charter for l’individu dans le monde, in Dumont’s terms, the agent who in his ordinary “worldly” life sees himself as primordially an individual, that is, the human agent of Western modernity.

This project of transformation is the one that I believe has been carried out in Latin Christendom. The vectors of commitment and disenchantment came about through a series of attempts at reform. The goal was to make over the lives of Christians, and also their social order, in a thoroughgoing way, so as to make them conform to the demands of the Gospel. I am talking not of a particular revolutionary moment but of a long, ascending series of attempts to establish a Christian order, of which the Reformation is a key phase. These attempts show a progressive impatience with older modes of postaxial religion in which certain collective, ritualistic forms of earlier religions coexisted uneasily with the demands of individual devotion and ethical reform that came from the “higher” revelations. In Latin Christendom, the attempt was to recover and impose on everyone a more individually committed and Christocentric religion of devotion and action and to repress or even abolish older, supposedly “magical” or “superstitious” forms of collective ritual practice. Social life was to be purged of its connection to an enchanted cosmos and all vestiges removed of the old complementarities between spiritual and temporal, between a life devoted to God and life in the “world,” between order and the chaos on which it draws.

This project was thoroughly disembedding just by virtue of its mode of operation, which took the form of a disciplined remaking of behavior and social forms through objectification and an instrumental stance toward human action. But its ends were also intrinsically inclined to disembed. This is clear with the drive to disenchantment, which destroys the second dimension of embeddedness. We can also see it in the specifically Christian context. In one way, Christianity here operates like any axial spirituality—indeed, it operates in conjunction with another, namely, Stoicism—but there also were specifically Christian modes. The New Testament is full of calls to leave or at least to relativize the solidarities of family, clan, and society and to become part of the Kingdom. We see this reflected in the way of operating common to certain Protestant churches, in which one was not simply a member in virtue of birth but which one had to join by answering a personal call. This understanding in turn helped to give force to a conception of society as founded on covenant and hence as ultimately constituted by the decision of free individuals.

This is a relatively obvious filiation, but my thesis is that the effect of the Christian, or Christian-Stoic, attempt to remake society in bringing about the modern “individual in the world” was much more pervasive and multitracked. It helped to nudge first the moral, then the social imaginary in the direction of modern individualism. I believe that this is what we see emerging in the new conception of moral order of seventeenth-century Natural Law theory. This was heavily indebted to Stoicism, and its originators were arguably the Netherlands neo-Stoics Justus Lipsius and Hugo Grotius. But this was a Christianized Stoicism, and a modern one, in the sense that it gave a crucial place to a willed remaking of human society.

We could say that both the buffered identity and the project of reform contributed to the great disembedding. Embeddedness, as I said above, is both a matter of identity—the contextual limits to the imagination of the self—and of the social imaginary, or the ways in which we are able to think or imagine the whole of society. But the new buffered identity, with its insistence on personal devotion and discipline, increased the distance, the disidentification, and even the hostility to the older forms of collective ritual and belonging, while the drive to reform came to envisage their abolition. In both their sense of self and their project for society, the disciplined elites moved toward a conception of the social world as constituted by individuals.

So, to the two linked vectors of personal commitment and disenchantment we can add two more, also closely related: those of the movements of reform and disembedding, or the rise of modern individualism. And these are connected to a fifth, which I think is one of the basic features, if not the basic feature, of modern secularity.

What do we mean when we speak of Western modernity as “secular”? There are all sorts of ways of describing it: the separation of religion from public life, the decline of religious belief and practice. But while one cannot avoid touching on these, my main interest here lies in another facet of our age: belief in God, or in the transcendent in any form, is contested; it is an option among many; it is therefore fragile; for some people in some milieus, it is very difficult, even “weird.” Five hundred years ago in Western civilization, this wasn’t so. Unbelief was off the map, close to inconceivable, for most people. But that description also applies to the whole of human history outside the modern West.

What had to happen for this kind of secular climate to come about? First, there had to develop a culture that marks a clear division between the “natural” and the “supernatural,” and second, it had to come to seem possible to live entirely within the natural. The first condition was something striven for, but the second came about at first quite inadvertently.

Very briefly, I believe that it came about as the by-product of the series of actions in the vector that I have called reform. Its attempt was to make individuals and their society over so as to conform to the demands of the Gospel. Allied with a neo-Stoical outlook, this became the charter for a series of attempts to establish new forms of social order, drawing on new disciplines (Foucault enters the story here), which helped to reduce violence and disorder and to create populations of relatively pacific and productive artisans and peasants, who were increasingly induced, or forced, into the new forms of devotional practice and moral behavior, whether this was in Protestant England, Holland, Counter-Reformation France, or, later, the American colonies and the Germany of the Polizeistaat.

My hypothesis is that this new creation of a civilized, “polite” order succeeded beyond what its originators could have hoped for and that this, in turn, led to a new reading of what a Christian order might be, one that was seen increasingly in “immanent” terms (e.g., the polite, civilized order is the Christian order). This version of Christianity was shorn of much of its “transcendent” content and was thus open to a new departure, in which the understanding of good order (what I call the “modern moral order”) could be embraced outside of its original theological, Providential framework and in certain cases even against it (as with Voltaire, Gibbon, and, in another way, Hume).

Disbelief in God arises in close symbiosis with this belief in a moral order of rights-bearing individuals who are destined (by God or nature) to act for one another’s mutual benefit, an order that thus rejects the earlier honor ethic, which exalted the warrior, as it also tends to occlude any transcendent horizon. (We see one good formulation of this notion of order in Locke’s Second Treatise. This understanding of order has profoundly shaped the forms of social imaginary that dominate in the modern West: the market economy, the public sphere, the sovereign “people.”35

In other words, the crucial change here could be described as the possibility of living within a purely immanent order; that is, the possibility of really conceiving of, or imagining, ourselves within such an order, one that could be accounted for on its own terms, which thus leaves belief in the transcendent as a kind of “optional extra”—something it had never been before in any human society. This presupposed the clear separation of natural and supernatural as a necessary condition, but it needed more than that. There had to develop a social order, sustained by a social imaginary that had a purely immanent character, which we see arising, for instance, in the modern forms of public sphere, market economy, and citizen state.


So the vectors of personal religion and disenchantment work to marginalize collective ritual. I am speaking, of course, of ritual in the strong sense, in which prayer, sacrifice, exorcism, or anything else transforms us, our world, or our relation to God and some higher realm. Ritual effects something in the higher realm or in our relation to it. Of course, we go on having rituals—we salute the flag, we sing the national anthem, we solemnly rededicate ourselves to the cause—but the efficacy here is inner: we are, in the best case, “transformed” psychologically; we come out feeling more dedicated.

All ritual consists of bodily action, which has some “symbolic” meaning; that is, it invokes or makes palpable something that has an important life meaning: our cause or God’s mercy, and so on. But the upshot of these two vectors is to reconceive of the efficacy of our action as its inward effect on our thoughts, emotions, and dispositions. The “symbol” now invokes in the sense that it awakens the thought of the meaning in us. We are no longer dealing with a real presence. We can now speak of an act as “only symbolic.”

The movement fits well with a slide toward dualism, as the bodily is merely external, and what is important happens inwardly.

The two vectors generate a new understanding of “religion,” which has affinities to what we often call Deism. One facet of this lies in the development of the modern moral order. The second facet makes impersonal orders paramount.

True religion in this view consists in a doctrine that is rationally defensible and that generates a morality that is endorsed by reason. It envisions a creator God, who gave us a universe in which we can read natural law, and then, as an optional extra, threw in an extra incentive to obey the law, in that he distributes rewards and punishments in an afterlife, based on how well we have fulfilled its demands. Everything else is superfluous and based on falsehood, like ritual, which is supposed to effect something real, and, indeed, forms of collective life that stifle the individual conscience, where the moral law has to reside.

This creates a template of true religion, which I described as Deist at the beginning of this chapter. People in other cultures can then take it up, or can find it taken up on their behalf, to show that they, too, have true religion or, indeed, even better examples of it than (orthodox) Christianity. Hence the reform proposals of Ram Mohan Roy in early-nineteenth-century Bengal and also Peng Guanyu, presenting at the 1893 Parliament of Religions in Chicago the orthodox view of Qing Dynasty scholars that Confucianism is not a religion (zongjiao) but, rather, a law and teaching (jiao) of proper human relations. The same status was claimed for Buddhism and Daoism and had been for some centuries. What has to be shorn off “religion” in order to have the pure stuff is wu, or shamanism, which includes magic, witchcraft, sorcery, and the like.36 Perhaps we have a case here of parallel development, which influences and is influenced by the Western move to Deism.


1. See the discussion of profane and sacred times in Charles Taylor, A Secular Age (Cambridge, Mass.: Harvard University Press, 2007), 54-61.

2. Marcel Gauchet, La religion dans la démocratie (Paris: Gallimard, 1998), 47-50.

3. I have discussed this at greater length in Charles Taylor, “Die blosse Vernunft,” in Dilemmas and Connections (Cambridge, Mass.: Harvard University Press), pp. 326-346.

4. For a fuller discussion of the modern idea of moral order, see Charles Taylor, Modern Social Imaginaries (Durham, N.C.: Duke University Press, 2004).

5. Thus, American secularists often totally confuse the separation of church and state from that of religion and state. For instance, Rawls at one point wanted to ban all reference to the grounds of people’s “comprehensive views” (these, of course, included religious views) from public discourse.

6. Ashis Nandy, Time Warps (Piscataway, N.J.: Rutgers University Press, 2002), chap. 3, esp. pp. 68-69, 80.

7. Ibid., 85. Amartya Sen also makes use of a similar point about Akbar’s rule to establish the roots of certain modes of secularism in Indian history. See Amartya Sen, The Argumentative Indian: Writings on Indian History, Culture, and Identity (New York: Farrar, Straus and Giroux, 2005). For an excellent example of such a creative redefinition, see Rajeev Bhargava, “What Is Secularism For?” in R. Bhargava, ed., Secularism and its Critics (Delhi: Oxford University Press, 1998), 486-522 (see especially 493-494 and 520 for “principled distance”); and Rajeev Bhargava, “The Distinctiveness of Indian Secularism,” in T. N. Srinavasan, ed., The Future of Secularism (Delhi: Oxford University Press, 2007), 20-58, especially 39-41.

8. See Taylor, A Secular Age.

9. See Robert Tombs, France: 1814-1914 (London: Longman, 1996), 135.

10. See John McManners, “Enlightenment: Secular and Christian (1600-1800),” in John McManners, ed., The Oxford History of Christianity (Oxford: Oxford University Press, 1993), 277-278.

11. “Subtraction story” refers to the thesis of religion’s ineluctable excision in the modern world, as people liberate themselves from prescientific, or prerational, systems of belief. See Taylor, A Secular Age, 26-29 et passim.

12. See, for instance, Akeel Bilgrami, “When Was Disenchantment?” in Craig Calhoun, Jonathan VanAntwerpen, and Michael Warner, eds., Varieties of Secularism in a Secular Age (Cambridge, Mass.: Harvard University Press, 2010).

13. Peter van der Veer shows how a not-dissimilar category—wu, which can be translated as either “shamanism” or “magic” and which emerged out of a parallel process of supposedly rational reform—was developed in modern China as a category for what was rejected as inferior, as not really religion. See Peter van der Veer, The Spirit of Asia: Comparing Indian and Chinese Spirituality (Princeton, N.J.: Princeton University Press, forthcoming).

14. See Taylor, A Secular Age, chap. 10.

15. See W. K. Clifford, The Ethics of Belief and Other Essays (Amherst, Mass.: Prometheus, 1999).

16. Birgit Meyer, Translating the Devil (Trenton, N.J.: Africa World Press, 1999), 181.

17. See the discussion of possession in ibid., 205-206.

18. See Taylor, A Secular Age, 146-158.

19. See Robert Bellah, Beyond Belief (New York: Harper & Row, 1970), chap. 2.

20. Godfrey Lienhardt, Divinity and Experience (Oxford: Oxford University Press, 1961), 233-235.

21. Ibid., 292.

22. As a matter of fact, it has been argued that the earliest forms of this religion were highly egalitarian in relation to later developments, just because the pervasive sense of a sacred order left little room for personal decision on the part of those charged with special functions. They couldn’t yet parlay these into personal power. See, for instance, Pierre Clastres, Les sociétés sans état (Paris: Minuit, 1974).

23. See, for instance, Lienhardt, Divinity and Experience, chap. 3; Roger Caillois, L’homme et le sacré (Paris: Gallimard, 1963), chap. 3.

24. This is a much-commented-on feature of aboriginal religion in Australia; see Lucien Lévy-Bruhl, L’expérience mystique et les symboles chez les primitifs (Paris: Alcan, 1937), 180 and ff.; Caillois, L’homme et le sacré, 143-145; W. E. H. Stanner, “On Aboriginal Religion,” a series of six articles in Oceania 30-33 (1959-63). The same connection to the land has been noted with the Okanagan in British Columbia; see Jerry Mander and Edward Goldsmith, The Case against the Global Economy: And for a Turn toward the Local (San Francisco, Calif.: Sierra Club Books, 1996), chap. 39.

25. John Stuart Mill, On Liberty, in John Stuart Mill, Three Essays (Oxford: Oxford University Press, 1975), 77.

26. In Taylor, A Secular Age, I define exclusive humanism as “accepting no final goals beyond human flourishing, nor any allegiance to anything else beyond this flourishing” (18).

27. See, for instance, S. N. Eisenstadt, ed., The Origins and Diversity of Axial Age Civilizations (Albany: State University of New York Press, 1986); see also Bellah, Beyond Belief, chap. 2.

28. Karl Jaspers, Vom Ursprung und Zielder Geschichte (Zurich: Artemis, 1949).

29. W. E. H. Stanner, “On Aboriginal Religion” Oceania 30, no. 4 (June 1960): 276. See also W. E. H. Stanner, “The Dreaming,” in W. Lessa and E. Z. Vogt, eds., Reader in Comparative Religion (Evanston, Ill.: Row, Peterson, 1958), 158-167.

30. W. E. H. Stanner, “On Aboriginal Religion,” Oceania 33, no. 4 (June 1963): 269.

31. I have been greatly helped here by the much richer account of religious development in Bellah, Beyond Belief. My contrast is much simpler than the series of stages that Bellah identifies; the “primitive” and the “archaic” are fused in my category of “early” religion. My point is to bring into sharp relief the disembedding thrust of the axial formulations.

32. See Marcel Gauchet, Le désenchantement du monde (Paris: Gallimard, 1985), chap. 2.

33. See Peter van der Veer, “Smash Temples, Burn Books,” 270-281, chap. 12 below.

34. Louis Dumont, Essaissur l’individualisme (Paris: Seuil, 1983), chap. 1.

35. I have developed this at greater length in Taylor, Modern Social Imaginaries.

36. Van der Veer, “Smash Temples, Burn Books: Comparing Secularist Projects in India and China.”