Skip to navigation | Skip to content

Facing religion, from anthropology

Anthropology was founded by freeing itself from the confines of religious authority. At least, such is anthropology’s original self-understanding, part of its project and ideology. Hence if today anthropology reports on the boundaries between religion and secularism, it has also been complicit in formulating and reproducing them.

This essay reflects on the making of distinctions between the religious and the secular.[1]  I am interested in the boundary work carried not only by religion and by the law, but by academic disciplines, such as my own. However, I will argue that the kinds of distinctions and truths produced by religion, law, and social science, respectively, are incommensurable with one another. By “secular” I mean roughly the realization or substantiation of claims that neither the state, science, philosophy, nor possibly everyday life are legitimated by what is considered religion, hence that one can—and sometimes must—think and act outside the sphere of religion and, in fact, with respect to religion.

The argument proceeds in three successively shorter phases. First I will discuss some of the paradoxes in anthropology’s response, relation, and reference to what it takes to be religion. One could say that this is one specific version, angle on, or component of the mutual constitution of secularism and religion and I hope to show that it is a particularly salient one. Next I will offer some anthropological insights into the boundary problem. And last I will suggest how these might influence considerations of the secular university

I claim, first, that anthropology is definitively a secular discipline and could be nothing else. But second, I complicate this picture by suggesting that anthropology inevitably blurs the boundaries between the religious and the secular and challenges the authority of the distinction. I take anthropology to be exemplary in this regard but also use it as something of a synecdoche for the other human disciplines, hence for the modern university more broadly. In a final portion of the paper I suggest that an exclusive focus on the interface of religion with the secular is misleading, itself part of the mystification occasioned by secularism, insofar as it neglects the other faces of religion, in particular the Abrahamic religions’ campaigns against what I will roughly call animism or polytheism, that is, religion in less objectified and less rationalized manifestations. It is attention to these other faces that perhaps most clearly distinguishes anthropology from neighbouring disciplines that study religion. It could doubtless be said of the secular too that it is multi-faceted, but that is not something I can address.

The distinction between religion and secularism is of course more complicated than an image of interfacing polyhedrons. It is complicated in the first instance by the fact that the divisions in the world of practice outside the university that are the ostensible subject of investigation by the academic disciplines are not entirely distinct from the divisions made by and within the disciplines, nor even unconnected from the very constitution of the modern academic disciplines and of the university of which the disciplines are a part. Insofar as the university itself has become secular in its constitution and governance and insofar as the various academic disciplines are secular in their orientation, modes of procedure, and especially, in the means by which they establish truth and hence in the kinds of truth they establish, so they are always already implicated in the question of the boundaries of religion and secularism. Insofar as religion (or its more holistic predecessor) historically produced the space from which secularism emerged and from which to demarcate it, and insofar as that space grew so that secularism now ostensibly encapsulates religion, rather than the reverse, and offers a space outside religion from which to turn back and study what has been left behind, as it were, so one may well ask whether there is any space available within Western thought beyond the secularism/religion distinction itself, that is, any space that is not always already implicated by the distinction, any thought that could be distinctly meta-secular (as the secular is meta-religious). In other words, it is immensely difficult to find a neutral or objective place, a place that is not already self-identified as secular or religious, especially within the academy, from which to think about the distinction.


The hermeneutic problems of reflexivity and objectivity that characterize all the human sciences are thus particularly acute when it comes to demarcating both religion and the space from which to discern and carry out such demarcations. And perhaps nowhere among the disciplines have these problems been more acute than within anthropology.[2]

This is not to say that anthropologists hold a unified position on these issues. Unlike a Lévi-Straussian opposition, the tension between religion and science or secularism is thoroughly shaped by the historical conditions of its time and it has changed for anthropology over the course of its own history, each phase leaving significant traces in successive layers of theoretical debate.[3] I paint the historical picture with extremely broad brush strokes as a series of three phases of anthropological thinking in conjunction with wider intellectual and sociopolitical processes.

In its inception as an academic discipline in the late Nineteenth Century, anthropology was comparative and, to draw on MacIntyre’s term (1990), ‘encyclopaedist.’ Anthropology saw itself as an objective, empirical, neutral science of reason that challenged the obfuscations and misapprehensions of religion and tried to locate religion’s place in the long trajectory of human history that evolutionary theory and archaeology had opened out. The separation from theology was not easy, as manifest in the career of Robertson-Smith (Beidelman 1974, Masuzawa 2010), but Tylor had a kind of assurance in stating that religion was rational but grounded in error and that it had its roots in animism.  The general project, too easily denigrated today for its affinity with, if not actual modality as, colonial ideology, was also part of the radical Enlightenment programme of locating humankind  (“man”) as a creature of nature rather than of God.

Evolutionism was eventually surpassed by functionalism, cultural particularism, and structuralism. There was a general deconstruction of the overly objective evolutionary typologies and categories of the earlier period (especially magic vs. religion, but also taboo and totemism[4]) and the relocating of humans now less as a product of either God or nature than of themselves (“culture”). As Gordon Childe (1965) memorably put it Man makes himself. For much of the 20th century the progressive task of anthropology was to show the order, logic, ethical consistency, meaningfulness, and beauty in what seemed to the majority of Europeans and North Americans to be exotic or uninteresting, primitive, backward, disorderly, disappearing, and generally unworthy societies and systems of thought.[5] The role of anthropology was no longer to critique religion but to appreciate it from a distance, i.e. indirectly, by means of “other” societies. In displaying the varieties of religious life, anthropology was also at least implicitly challenging the superiority of Christianity over other forms of thought and practice. However, anthropology’s subjects were for the most part distant, quiescent, and relatively powerless and there was an ethical imperative to represent them in face of the onslaught of change, whether one saw it as “modernization” or the “development of underdevelopment.” Attitudes to social change and, especially, to the acts of missionaries and the facts of Christian conversion, were ambivalent. Lévi-Strauss’s title Tristes Tropiques (1970) conveys the sensibility of the period.

During both these phases, modernity was identified with the growth of secularism and anthropology understood itself as a secular and largely scientific discipline, sometimes concealing from itself its strong romanticist tendencies. But by the end of the 20th Century, with the resurgence of religion in the United States and within national, transnational, and global politics (a resurgence perhaps first acknowledged in the surprise occasioned by the Iranian revolution), but also with the rise of scepticism within the academy about the nature of science and secularism themselves (as phrased by diverse strands of poststructuralist, postmodernist, and postcolonial thought, including, in anthropology, the seminal work of Asad 2003) and the concomitant affirmation of history or Foucauldian genealogy as the master paradigms, anthropology finds itself squirming, no longer content or able either simply to champion religion against science and modernization narratives or, with the exception of a vocal minority, to develop in full confidence an ostensibly value-free objectivist (and secular) science of religion.

The tension between anthropology’s scientific rationalism and its humanistic relativism was particularly acute (at the time of writing the essay from which I have been drawing) when President Bush advocated the teaching of  “intelligent design” alongside natural selection in American schools, refused subsidies for AIDS prevention programs that promoted the use of condoms, and referred to an axis of evil. Do anthropologists simply interpret the coherence of conservative Christianity and analyze the power of its rhetoric or do we try to fight, as secularists, for the naturalist and evolutionary premises on which anthropology and the life sciences are built? The inverse question in Europe has been how anthropologists represent the worlds and rights of Muslim religious communities in face of a secularist ethos as well as sheer prejudice and fear, questions that produce conferences such as the one for which this essay was written, which ought then to be part of its subject matter. If there is a compass to the anthropological direction perhaps it lies in unmasking or decentring hegemonic assumptions, undue power, unfairness, and dogmatic or absolutist thinking, from whatever quarter. These are, of course, not the special province of religion, science, or the state per se but are characteristic of certain manifestations of each.

If, during the first two phases of anthropological thought delineated above, science itself was unproblematic, and if, in the first phase anthropology saw itself unproblematically as science, these facts are not true of the present age. A number of things have changed beyond the political fortunes of religion. First, anthropology has increasingly questioned its own status as a science, and second, science itself—and, of course, secularism—have become objects of anthropological inquiry alongside and roughly equivalent to that of religion.[6] Both religion and science can be described as systems of human thought and disciplined practice, each with strengths and weaknesses, neither perfect according to their own standards, the practitioners of each struggling with issues of moral judgment, creativity versus iteration, and the prejudices entailed by their own means of production and reproduction and modes of seeing the world. Moreover, both can be seen as strong political forces, whether countering, supporting, or encapsulated by the state and market. The politics of religion, or religion as politics (hence as ‘identity,’ ‘resistance,’ ‘opposition,’ etc.), have become central modes of anthropological investigation and are one of the factors that have pushed anthropology itself toward a much more self-consciously political identity. The theoretical challenge lies in how to balance these insights with the recognition that religion may also be one of the few locations from which it is possible to stand outside politics or to regulate or ground the political in a different kind of order. Religion must be treated in a way that neither excessively politicizes nor excessively depoliticizes it.

In general, then, and with the exception of the new champions of a cognitive and evolutionary science of religion, the current phase of anthropological thought has meant a move away from claiming a particular expertise or understanding of the nature or essence of religion—and from participation in the theological or scientific debates that such claims entail—toward analysing the politics ‘of religion,’ as though from outside (and perhaps, outside ‘science’) but thereby necessarily ‘inside’ some other, at times seemingly inchoate, disciplinary mode of practice that, according to at least some accounts, would still be specifically secular. Insofar as secularism can, in one formulation, be defined as that position or standpoint from which religion can be viewed from outside (Lambek n.d.) or alongside other, equivalent competing forms (Taylor 2007), so anthropology, the outsider and pluralist perspective par excellence, must be secular vis-à-vis religion, even though vis-à-vis itself it could then be only meta-secular.

One current articulation of the recurrent epistemological fault line of anthropology is that between the sceptical genealogical observer and the complicit, but possibly critical, hermeneutic participant (i.e., the person who accepts Gadamer’s [1985] arguments that we are all located within traditions and that all traditions entail their prejudices) who accepts that anthropology shares horizons with both religion and science or secularism.[7] Between these positions anthropologists must construct both their research programs and their politics. When they are understood as incommensurable and hence co-present rather than binary and mutually exclusive, the co-existence of scepticism and conviction or objectivism and relativism gives rise to a state that, for at least this informant, must be described as a kind of irony. This is irony understood in the sense of the dramaturgical or novelistic polyphony of Kenneth Burke (1945) or Mikhail Bakhtin (1981) and the Socratic/Platonic acknowledgment, so well elaborated by Alexander Nehamas (1998), that there are limits to what one knows, an irony that I find manifest also in, if not constitutive of, the spirit possession that I study and that I am at some pains to know whether and on what criteria to classify as religion.


One could say that the relationship between religion and secularism is intrinsic to the discipline, that religion/secularism stands, like nature/culture in Lévi-Strauss’s model of myth, as the irresolvable opposition around which anthropological thought builds itself.[8]  This is so for both epistemological and historical reasons. Epistemologically, anthropology is constituted by means of an intrinsic tension between objectivism and relativism (Bernstein 1983). The religion/secularism opposition stands in anthropology for large questions, notably, the debate between relativism and rationalism, and the contrast and transition between holistic but diverse, particular ‘primitive’ or ‘traditional’ worlds and the disenchanted, fragmented, but ultimately singular and universal ‘modern’ one. In this picture, which is the inverse, but not the contradiction, of Asad (1993), ‘religion’ once stood for the holistic worlds of pre- or non-modern societies rather than the relatively disembedded and encapsulated phenomena characteristic of modern states. Religion in this understanding is neither a particular institution or sensibility, nor a set of comparable beliefs in something specifically spiritual or super-natural, but a kind of ontology, a model of the world in its essence and a model for existing in it and reasoning about it in a meaningful and ethical fashion; religions are the fictive (not fictitious) worlds in which people live and the techniques that render those worlds real and realizable. The proximity of this depiction to Geertz’s famous definitional essay (1973a) is no coincidence.[9]  One possible inference to make from this is to say that secularism itself is the comparable ontology of modernity, our own form of enchantment. This move of course destabilizes the opposition between secularism and religion, an opposition that may then be said to be part of secularism’s own ontology. Alternatively, one can discriminate between kinds of ontologies, or world-making, referring to only some of them as religious.

The science/religion or objectivism/relativism opposition finds its theoretical expression in the ways in which anthropologists think through the relationship between cultural particularism and universalism—as for instance in the question of whether we can find something one can call ‘religion’ universally, i.e., within all human societies, past and present, with or without taking modernity as an exception, or in relation to some universal capacities, needs, or other features of the human condition, or, after Durkheim, as an intrinsic feature or dialectical movement of human society.[10]

Anthropology’s central opposition finds its practical expression in the common injunction for ethnographic research, namely participant observation. The issue is the relationship of ethnographic observation to participation, or to borrow (after Geertz) terms from psychoanalysis (which, in this respect is similar to anthropology), experience-near to experience distant analysis, as well as the effects of transference and counter-transference. With respect to religion this has been expressed in the debate over whether, in order to understand the religious ideas and practices of a particular group (and hence to identify them as religious), it is necessary in some sense to share them, that is, maximally, to experience, believe, or accept them, or minimally, to understand oneself as ‘religious,’ whether in sensibility, practice, or belief–or whether, to the contrary, it is better to situate oneself firmly outside religion (Firth 1996). And if one were to do the latter, what else would one call one’s location but secular? More generally, insofar as anthropology is characterized by the ascetic calling or stance of understanding natives without oneself going native, how is the anthropologist to conceptualize her relationship to her subject of study (and articulate her research to the rest of her life)?

In sum, much as anthropology would like to see itself as an objective observer of the range of human institutions and transformations, it is itself situated within the broad discursive field constituted by secularism and religion and it has always also been an interested party in the debate between them, pulled between explanation and interpretation, demystification and appreciation, transcendent reason and immanent experience. As a result, despite many insightful contributions and developments, clarifying the relationship between the secular and religion remains an ongoing therapeutic task or feature of anthropology rather than a fully realized or realizable scientific goal. It is internal as well as external to the practice of anthropology.


One way to summarize the argument so far would be to say that the very conceptualization of the subjects, questions, problems, and practices characteristic of anthropology as a discipline presuppose some kind of bracketing, or standing outside of something. This is both analogous to the way secularism and religion can be understood, respectively, as standing outside one another or outside ordinary life, and continuous with the division between religion and the secular as standpoints. In other words, anthropology is implicated in this distinction and the distinction is implicated in anthropology. One could go further and say that anthropology is intrinsically secular or that it is the emergence of the secular that enables the possibility for anthropology.[11] Is it not the case that a nonsecular anthropology would be impractible or even inconceivable?

I take this view but at the same time I take an alternative view, holding, ironically, to two points of view. Let me try and explain. The various oppositions I have enunciated—universal/ particular, objectivism/ relativism, anthropologist/ native, observation/ participation, and even secular/ religion imply that they are each binary oppositions, such that each term is mutually exclusive of the other, and hence that they are possible only of various, ultimately unsatisfactory, forms of mediation, of the order that Lévi-Strauss (1963a) has described as mythological. It is clear that in theory-making the pairs can form analogies of the sort a : b :: c : d that Lévi-Strauss describes. But there is a bit of a worry here because Lévi-Strauss describes the play of concrete signifiers—drawn from the material sensory world of animals, colours, and the like—that are displayed precisely in lieu of abstractions. Abstractions themselves do not have intrinsically the distinctive features that are enabled by the sensory and material world, and so there may be a category mistake entailed in setting abstractions themselves directly in binary opposition to one another in the way we do with properties like raw and cooked.

The alternative is to conceive the items in our pairs of abstract concepts as incommensurable to one another, in which case the ostensible binary opposition with its implication of mutual exclusion is itself the product of a category mistake of the order that Gilbert Ryle describes with respect to mind and body. My inclination is to follow this alternative while also being mindful of the ease with which human beings make binary oppositions and category mistakes and the consideration that, with respect to some matters, it may be impossible not to do so. That is to say, there are some domains in which we recognize simultaneously our category mistakes and the impossibility of avoiding them. Category mistakes that are incapable of correction are thereby incapable of authoritative and conclusive resolution. They remain either as mysteries, in discourses we call religious, or as paradoxes in discourses of philosophy. Where they simply go unrecognized, they lead to continuing conversation, conversation that is perpetuated insofar as we speak simultaneously to each other and past each other. Each ostensible resolution is bound to be provisional and eventually contested. The situation of inherent contestability of certain concepts once noted by Gallie (as noted by Geertz 1973b) is here the effect or product of incommensurability.[12]

If we take incommensurability to be the case we accept that neither objectivism and relativism, observation and participation, nor the religious and the secular are in simple binary opposition and hence do not divide the world in two, nor offer clear, exclusive boundaries between their referents, nor clear-cut choices of affiliation for would-be adherents. Instead, there are merely contingent, provisional boundaries, imperfectly conceived and ineffectively policed. I am suggesting that insofar as religion and the secular are not discrete objects they need not be mutually exclusive. In other words, insofar as their ideology of mutual exclusion is the product of a category mistake, persons, objects, and acts may be located or described simultaneously with respect to both religion and the secular (possibly as both inside and outside religion or as more or less religious and secular). As did the workshop on Religious-Secular Distinctions, we can explore the emergence, constitution and effects of provisional boundaries, noticing in the first instance that they emerge in multiple discourses, practices, institutions, and forums—from within religion, the state or law, and the academy. Although these discourses acknowledge and influence each other, their activities and effects differ in location, force and meaning.[13]  There is no single correct or uniquely authoritative boundary to be discerned, established, or justified.

Whence come such category mistakes? One of the specific sources of confusion is that we tend to deploy these terms as nouns and by means of a semiotic ideology that supposes that words, in the first instance, refer to autonomous things in the world. This semiotic ideology itself has a history, being elucidated within anthropology by Webb Keane (2007) who shows its relationship to changes in religion itself and hence to the emergence of secularism. Indeed, we could call this semiotic ideology the ideology of modernity and hence of secularism. I have argued elsewhere (Lambek forthcoming) that it was followed in anthropology itself by alternative semiotic ideologies that emphasize respectively the poetic and perlocutionary and the indexical and illocutionary functions of language. Without elaborating these distinctions here, let me throw out three, related suggestions.

First, while the locutionary and referential semiotic ideology certainly has its place, the academic predilection for reified abstractions overextends it. This semiotic ideology has underpinned the idea of religion and the secular as separate spaces, indeed as discrete res extensa, things extending in space, and thus has had an affinity with the political ideology that has sought to circumscribe and govern religion. Second, what happens if we began to take seriously other semiotic ideologies and specifically ones that recognize the significance of the illocutionary function of utterances? I suggest that the boundaries between religion and its others look different, and must be marked differently, according to the semiotic ideologies deemed respectively at play. But I would also want to distinguish the semiotic ideologies from actual semiotic practices. Thus, I have argued that the identification and realization of spirits in Mayotte and Madagascar is partially the product of the mystification of performativeness (2006b). Here a semiotic ideology of reference in fact conceals a semiotic practice of performativeness. Some of the conflict between science and religion comes when their semiotic ideologies are construed as identical to one another. When they are not so construed and, as inspection of their practice largely demonstrates, when they operate with different semiotic ideologies, this difference is sufficient to make them incommensurable with one another rather than in direct contradiction or competition. However, practitioners themselves may refuse to recognize the semiotic specificity of their own practices and may be wont to borrow what appears to be the most authoritative semiotic ideology in the prevailing climate.

Third, what would happen if we changed our own semiotic practices and began to treat words like religion and the secular—and here note the awkwardness of adding the definite article—not as nouns, with the implications of bounded entities, but as verbs, adjectives or adverbs? We have not done so in part due to the deep grammar of our Indo-European languages and in part because we have tended to view religion as comprised of or produced by a series of things, more nouns. In some versions these things are called beliefs and in other versions experiences. It is worthy of note that both belief and experience tend to start with the individual. They are thus congenial to modernity, and indeed they are the approaches that most modernists tend to take. Charles Taylor is an obvious case in point, moving in his introduction to A Secular Age (2007) from an experience of “fullness” to faith in the transcendent. Belief and experience, as Taylor would be the first to admit, are also close to Christianity. Here it is important to note the recent self-critique of anthropology, articulated especially by Cannell (2006), that its own views have been implicitly shaped by Christianity. Christianity has served as the model for what religion is, as the paradigm case, hence establishing what constitutes the core of religion and the framework through which, and standard against which, other practices have been viewed. Hence too, Christianity has been the model for how this core might relate to other features of the social world, including, where it exists, the state or political realm, hence to the very definition of secularism.

Perhaps the most pernicious appropriation from Christianity—actually from the Abrahamic tradition more generally–is the idea that Religion is manifest as specific religions and that these religions are bounded units demanding exclusive loyalty from their adherents. Closely attached to this has been the idea that religion concerns primarily belief in God, personified and preferably singular (Lambek 2008). In some respects the concepts of ‘belief’ and of ‘God’ appear to mutually imply each other, at least they do so in the Christian tradition (Ruel 1982); so too the Christian concepts of belief and doubt (Pouillon 1982). And of course, they find their flowering in Protestantism. One of the challenges for anthropology has been how to view and compare phenomena in ways that do not make these concepts or principles foundational, thereby neither excluding the practices of other people from the high moral ground of religion nor unduly distorting them so they fit in.

Less specifically Abrahamic and, I would say, more successful anthropological approaches to religion begin with acts of making and doing rather than in the articulation of personal experience or beliefs in persons. One approach examines symbolic classification and poiesis; another begins with ritual acts. Both are rooted in Durkheimian sociological theory, which begins with the collective rather than the individual. From this perspective, belief and experience are secondary phenomena when it comes to discerning, distinguishing (or, for that matter, explaining) religion, despite the fact that they may be salient both within specific religious formations and for particular subjects. When we start with ritual or symbolic classification rather than specific beliefs or experiences the boundary problem looks quite different.

Since it is especially Protestant Christianity that emphasizes belief over classificatory schemes, ritual, or structures of authority and then it is secularism that, in response, emphasizes reason and scepticism, perhaps the boundary looks less sharp where the prevailing religious context is not a Protestant one and where our own models are not ones of belief or inspired directly by Christianity.

Whatever approach we take, the adequacy of our explanatory or interpretive frameworks exists in relation to their calibration with the phenomena. Put another way, we can only understand religion insofar as we take it seriously, perhaps acknowledging religion itself as the vehicle for acknowledging the seriousness of life (Tugendhat 2006), and once we do so, have we not already begun to compromise our secular stance that made the conceptualization of our project possible at the outset?

It is thus evident that insofar as anthropology must take seriously and respectfully the concepts and practices of those whom it studies, attempting to understand their full meaning and meaningfulness, while not embracing them as an exclusive version of the truth and thereby shifting from an anthropological to a native standpoint, or from a pluralist to a monological one, and insofar as it must take an equally sceptical, anthropological eye to the concepts and practices of its own milieu, so it is inevitably both intrinsically secular and verging on the non-secular. Put another way, it is inevitably both sympathetic to and sceptical of varieties of religious practice while being both in practice secular and in principle sceptical of secularism. And insofar as it is sceptical of scepticism, perhaps it is attracted, after all, to the order provided by ritual, the seriousness afforded by the transcendental, and the play enabled by symbolic classification. What I think this means is that anthropologists are forced to exercise judgment and to attempt to find a middle path between the temptations and risks that lie to either side. In this respect anthropology is an ethical vocation, by which I do not mean a vocation that does good or advocates justice (let us leave the former to advocates of religion and the latter to proponents of secularism) but one in which phronesis is the central virtue. And perhaps it is a vocation that is neither fully religious from the beginning nor exactly secular at the end.


Building on the ideas of distinctive semiotic ideologies and kinds of speech acts, and linking these back to Foucauldian insights, we might say that one of the ways to differentiate kinds of discursive regimes is the manner by which they produce truth and hence the kind of truth they do produce. It should be evident that different kinds of truth—for example, referential, poetic, and performatively established–may be incommensurable with one another rather than in contradiction. In other words, a given truth from one kind of discourse need not imply the falsity of something produced in a different kind of discourse. For example, logical and experiential truths cannot be readily compared with one another. One of the significant features of the work of Roy Rappaport lies in his analysis of how the enactment of ritual entails acceptance rather than belief and produces what he calls “the truth of things” (1999: 293ff) or “the truths of sanctity” (304ff).[14] Sacred truth is different from the truths of logic, empiricism, or experience, of a different order as it were, but it may have bearing on them. The question is, what are the truths and truth-making procedures of secularism? Some of them come from science; here the kind of ‘truth’ generated by means of religion will differ from that generated through the scientific study of religion. But an equally significant truth-making regime of secularism is the law, and the law operates performatively, much as do the ritual acts and practices I call religious.

Hence it seems it me that the central question about discerning religion today concerns not so much its distinction from science but its distinction from law. Historically, of course, this is partly about the establishment of divine law in place of the merely human and eventually the dislodging and circumscription of divine law as it is replaced by the human. Perhaps I have simply arrived by my own path at Asad’s original point. But my question is less how law circumscribes or competes with religion than with how we might analyze the differences between the ways in which religion and law respectively construct their truths. And it is of considerable interest that human law is still sanctified, in the last instance, by God’s word. That witnesses continue to swear on a Bible is not simply a vestige of Christianity in an ostensibly neutral and pluralist state but equally an indication that sanctification still has a function that cannot be replaced, underpinning the referential truth of testimony with a different kind of truth that establishes the act of witness as one of truth-telling in the first place.

Insofar as the university and academic research draw upon a different semiotic ideology and different semiotic practices, and insofar as they seek and produce a different kind of truth, they are not religious. They are, we may say, secular. But because their truths and forms of truth-making are incommensurable with those of religion, so they do not necessarily replace, supersede, or contradict religion. Stack and Fitzgerald (2010) have asked “what the idea of the secular university does, including legitimating our scholarship, and whether or not our debates about the religious-secular distinction call the secular university into question.” My answer, in short, is that irrespective of the idea of the secular university, the practice of the secular university offers a kind of truth and a kind of knowledge incommensurate with that of religion. And yet I believe that religion, in a broad sense, continues to play a role in the university. I will not discuss the complex ways in which values shape research and teaching or how symbolic classification operates in the academy, but close with a simple illustration of the legitimation of scholarship. Rituals of graduation are constituted as performative acts and draw on various sources of sacred authority—archaic language (e.g. Latin), venerable patriarchs, peculiar vestments, and the like.

The real question facing the university is not whether to continue to solemnize and sanctify such occasions, but by means of which particular tradition to do so. Calls for abstract forms of civil religion not withstanding, in practice the rituals of state of most countries draw upon the religious traditions of their founders. And it remains that the rituals of each university appear to be drawn from a specific religious or ritual tradition; indeed, the less they appear to do so, the thinner and less effective they are.[15] The problem with religion, as distinct from scholarship, is that it is not obvious how to universalize it without homogenizing it, and thus enable it to belong, at least potentially, equally to everyone. Universality is certainly one of the missions of the secular university.

There is one way in which the university finds its position analogous rather than opposed to religion and that has to do with the manner in which both are subject to regulation by the state (and market). In this sense both are subject to secularism. In its imposition of various auditing techniques the state has gone too far in regulating the university and has, unfortunately, infected the authority structure of the university itself. In principle, the university and religion might be allies in a struggle to retain a degree of autonomy from the state.


In conclusion I want to mention the fact that the anthropological study of the relation of religion to secularism retains a kind of ethnocentric bias since, on the one hand, it is interested almost exclusively in the Abrahamic traditions[16] and, on the other, it tends to see these religions as intruded upon rather than intruding. We need to acknowledge that the encounter or dialectic of religion with secularism is only one direction to which religion is turned. If the secular constitutes itself as being able to view, encapsulate, or govern religion, so religion, conceived in an Abrahamic sense, constitutes itself as able to disparage, denigrate, exploit, patronize, or convert those who are not its members. I refer here less to a possible conflict with the secular world than with respect to encroaching on non-secular practices that are nevertheless neither fully Christian nor Muslim, on forms of belief and practice that are not disembedded as formal religion, or that present themselves as incorporative rather than exclusive, one could say, polytheist or animist rather than monotheist. From the perspective of secularism one could speak of these as respectively the front and back faces of religion.

Much hangs on how we depict the practices of smaller-scale and non-Western societies. Anthropologists are divided over whether they “have” religion, and indeed whether it is the sort of thing one has and what its boundaries are.[17] This indicates a Rylian question of the sort addressed above. But additionally, it could be dangerous for the health of a community for an outside expert to claim that they have no religion or that what they have is no religion. This has left such people ripe for the disdain of their neighbours and in need of conversion and possibly conquest. The fact that most such people do accede to one of the so-called world religions is a symptom of the problem. Perhaps they do so in part in order to participate or be seen to participate in religion at all, that is to participate in what is precisely a bounded object, recognized by the state and seen by the world to be a religion. Participation in a transnational religion gets recognized as a sign of civilization, modernity, or citizenship in the global community—which, curiously, has become defined increasingly by means of religion, rather than as Freud, Marx, and a number of present day Darwinists suppose, by getting beyond it.



Asad, Talal. 1993. Genealogies of Religion. Baltimore: Johns Hopkins UP.

_____2003. Formations of the Secular: Christianity, Islam, Modernity. Stanford: Stanford University Press.

Bakhtin, M. 1981. The Dialogic Imagination. Austin: U Texas Press.

Beidelman, Thomas. 1974. William Robertson-Smith and the Sociological Study of Religion. Chicago.

Bernstein, Richard J. 1983. Beyond Objectivism and Relativism: Science, Hermeneutics, and Praxis. Philadelphia: University of Pennsylvania Press.

Bubandt, Nils and Martijn van Beek, 2012. Varieties of Secularism in Asia: Anthropological Explorations of Religion, Politics and the Spiritual. New York: Routledge.

Burke, Kenneth, 1945. Four Master Tropes. In A Grammar of Motives, by Kenneth Burke. New York: Prentice Hall, pp. 503-17.

Cannell, Fenella, 2006. Introduction to The Anthropology of Christianity. London: Duke University Press.

_____2010. The Anthropology of Secularism. Annual Review of Anthropology 39:85–100.

Childe, V. Gordon, 1965. Man Makes Himself. 4th edition. London: Watts.

Das Veena, 2006. Secularism and the argument from nature. In Charles Hirschkind and David Scott, eds. Power of the Secular Modern: Talal Asad and his Interlocutors. Stanford, CA: Stanford Univ. Press. Pp. 93–112

Firth, Raymond, 1996. Religion: A Humanist Interpretation. London: Routledge.

Gadamer, Hans-Georg. 1985 [1960]. Truth and Method. New York: Seabury.

Geertz, Clifford, 1973a [1966]. Religion as a Cultural System. In The Interpretation of Cultures. New York: Basic Books. Pp. 87-125.

_____1973b. Thick Description: Toward an Interpretive Theory of Culture. In The Interpretation of Cultures. New York: Basic Books. Pp. 1-30.

Keane, Webb, 2007. Christian Moderns: Freedom & Fetish in the Mission Encounter. Berkeley: University of California Press.

Keller, Eva, 2005. The Road to Clarity: Seventh-day Adventism in Madagascar. New York: Palgrave-Macmillan

Lambek, Michael, 1993. Knowledge and Practice in Mayotte: Local Discourses of Islam, Sorcery, and Spirit Possession. Toronto: University of Toronto Press.

_____ 1998. Body and Mind in Mind, Body and Mind in Body: Some Anthropological Interventions in a Long Conversation . In Michael Lambek and Andrew Strathern, eds. Bodies and Persons: Comparative Perspectives from Africa and Melanesia. Cambridge: CUP. Pp. 103-123.

_____2006a. Anthropology and Religion. Oxford Handbook of Religion and Science. Philip Clayton, ed. Oxford University Press, pp. 271-289.

_____2006b. What’s In a Name? Name Bestowal and the Identity of Spirits in Mayotte and Northwest Madagascar. In Gabriele vom Bruck and Barbara Bodenhorn, eds. The Anthropology of Names and Naming. Cambridge: Cambridge University Press. Pp. 116-38.

_____2008. Provincializing God? Provocations from an Anthropology of Religion. In Hent de Vries, ed. Religion: Beyond a Concept. New York: Fordham University Press.  Pp. 120-138.

_____forthcoming. Word as Act: Varieties of Semiotic Ideology in the Interpretation of Religion. In Words, Ernst van den Hemel and Asja Szafraniec, Eds. New York: Fordham University Press (presented at the NWO conference on Words, Groningen, Netherlands, June 15-16, 2009).

_____n.d. On Standing ‘Outside’ Religion: A View from the Pre-Alps. Panel on Secularism. American Anthropological Association. Chicago, November 2003.

Lévi-Strauss, Claude, 1963a. Structural Anthropology I. New York.

_____1963b. Totemism. Boston: Beacon.

_____1970 [1955] Tristes Tropiques. New York.

Macintyre, Alasdair. 1990. Three Rival Versions of Moral Enquiry. London: Duckworth.

Masuzawa, Tomoko, 2005. The Invention of World Religions: Or, How European Universalism Was Preserved in the Language of Pluralism. Chicago: University of Chicago Press.

_____2010. Promise of the Secular: William Robertson Smith and the historical constitution of Biblical Studies. Paper presented to the conference on Religious-Secular Distinctions. London: The British Academy, January 14-16 2010.

Meyer, Birgit, 2009. Response to ter Haar and Ellis. Africa 79(3): 413-15.

Nehamas, Alexander, 1998. The Art of Living. Berkeley: University of California Press.

Pouillon, Jean, 1982. Remarks on the Verb “To Believe.” In Michel Izard and Pierre Smith, eds. Between Belief and Transgression: Structural Essays on Religion, History, and Myth. Chicago: University of Chicago Press. Pp. 1-8.

Ranger, Terence, 2007. Scotland Yard in the Bush: Medicine Murders, Child Witches, and the Construction of the Occult: A Literature Review. Africa 77(2): 272-83.

Rappaport, Roy. 1999. Ritual and Religion in the Making of Humanity. Cambridge: CUP.

Ruel, Malcolm, 1982. Christians as Believers. In John Davis, ed. Religious Organization and Religious Experience. London: Academic Press. PP. 9-31.

Stack, Trevor and Tim Fitzgerald, 2010. Religious-Secular Distinctions. Conference programme, The British Academy. London, January 14-16. last accessed March 25 2012.

Stambach, Amy, 2009. Spiritual Warfare 101: Preparing the Student for Christian Battle. Journal of Religion in Africa 39: 137-157.

Taylor, Charles, 2007. A Secular Age.  Cambridge: Harvard University Press.

ter Haar, Gerrie and Stephen Ellis, 2009. The Occult Does Not Exist: A Response to Terence Ranger. Africa 79(3): 397-412.

Tugendhat, Ernst, 2006. Whom to thank? German philosopher Ernst Tugendhat on religion as a need and the difficulty of satisfying it. The article originally appeared in German in the Neue Zürcher Zeitung on December 9, 2006.

  1. [1]This paper was written for and first presented at the interdisciplinary conference on Religious-Secular Distinctions held at the British Academy, January 14-16, 2010 (Stack and Fitzgerald 2010). My work is supported by the Social Sciences and Humanities Research Council of Canada and the Canada Research Chair in the Anthropology of Ethical Life in the Department of Social Sciences, University of Toronto at Scarborough. I am grateful to Trevor Stack and Simon Lambek for astute comments on a first draft and Charles Stafford on the second.
  2. [2]They may be equally—but differently—acute in Religious Studies.
  3. [3]I am no expert in the Reformation, the emergence of the liberal state, or the origins of the academic study of religion. For wider and deeper perspectives, turn to the pioneering works of Asad (1993, 2005) and Masuzawa (2005).
  4. [4]Lévi-Strauss’s deconstruction of ‘totemism’ as ‘a’ religion (1963b) prefigures Asad’s deconstruction of the larger category of religion itself.
  5. [5]Here, as Trevor Stack rightly notes (pers. comm. January 11, 2010), ‘religion’ was often taken by anthropologists as “the limit of the ‘culture’ they take as their object. . . Religion is what defines the outside for anthropologists – that which they cannot even aspire to share. It is the most intractable domain of culture.”
  6. [6]As far as I am aware, secularism owes its salient presence in anthropology largely to the prescient interventions of Asad.
  7. [7]Indeed, the relationship between the sceptical or critical and the secular is perhaps not as well thought through as it might be. If, as is evidently the case, secularism does not have exclusive rights to critical thought or to scepticism and as readily sinks from self-reflection into complacency as does religion, surely not all scepticism concerning the secular need be defined by default as religious, just as not all scepticism of received religion need be defined as secular. Remember that for Kierkegaard doubt and its leap of faith may be of the very essence for religion—but this too has its genealogy, and the dialectic of faith and doubt could be considered a specifically Christian phenomenon (Ruel 1982).
  8. [8]Several of the following paragraphs draw liberally from (but also change and add to) Lambek (2006a). Among the emendations is the occasional transposition of the word ‘secularism’ for that of ‘science.’
  9. [9]This position was supported by both symbolic anthropology and structuralism. However the Weberian turn in anthropology to focus on ‘world’ religions, a turn in which Geertz himself played a central role, and also the increasing consciousness of traditions like Islam and Hinduism that they are religions shifted the intellectual terrain considerably.
  10. [10]If one can, the question is how to describe and account for the privileged position from which the universal becomes visible and objectified. If one cannot, the question is how, and with what terms, one could describe, characterize, compare, and understand historically constituted practices, traditions, differences and similarities.
  11. [11]Conversely, perhaps it is the emergence of anthropology that makes necessary the secular.
  12. [12]I have developed these ideas more fully elsewhere. See Lambek 1993 for a fuller ethnographic exposition and 1998 for a discussion of the mind/body problem.
  13. [13]Anthropology shows this empirically. For two very different but equally interesting examples concerning Christianity in Africa see Keller 2005 and Stambach 2009.
  14. [14]For Rappaport sacred truth is an intrinsic product of the enactment and acceptance of ritual order. I would add that it was the invention of Abrahamic religion to say that loyalty to the ostensible originator of this truth must be exclusive and then of Christianity to suggest that the truth itself must be exclusive.
  15. [15]To take my institution, the University of Toronto was formed by an affiliation of a number of Christian denominational institutions and some of the constitutive colleges retain a religious identity while others do not. It is no accident that the most prestigious college, Trinity, is also the Anglican one. It has the nicest architecture and gardens and, I suspect, the closest connections to the governing and corporate elite of the country and the richest or most generous alumni. The dining hall retains high table and gowns in self-conscious imitation of Oxbridge; religious and colonial hegemony are marked. In the past there has been a quiet struggle at Toronto and elsewhere to make the university a hospitable place for women and for Jews and this struggle now concerns those of other backgrounds. Practicing Muslims need space and time for prayer and facilities for ablutions but the larger issues have to do with respect. The question is: if the university is a secular place, how can it also be cosmopolitan and make all its members feel as comfortable and as entitled as those of ‘old boy’ heritage? Each new cohort encounters this challenge, more or less happily.
  16. [16]But see Bubandt and van Beek 2012, Cannell 2010, Das 2006, inter alia.
  17. [17]An index of the current state of confusion is the recent debate in the journal Africa over whether ‘witchcraft’ and the ‘occult’ should be part of an anthropological and historical vocabulary and contrasted with ‘religion,’ or subsumed within a universal and bounded category of religion, defined as “a belief in the existence of an invisible world, distinct but not separate from the visible one, that is home to spiritual being with effective powers over the material world” (ter Haar and Ellis 2009: 400; cf. Meyer 2009 and Ranger 2007).

Please join our mailing list to receive notification of new issues