Critical Responses to Phenomenological Theories of Religion
What Kind of Category Is “Religion”?
WILLIAM ARNAL
The conception of religious studies as a discipline, or even as a coherent area of study, requires not only that “religion” be capable of some kind of reasonable definition, but that it correspond to a specific set of entities that are, in some fashion, worth demarcating for some intellectual purpose. This need to define and defend an academic subject matter creates a pressure to fabricate religion as an autonomous and more or less “natural” or universal aspect of human existence.1 Nearly all modern theories of and approaches to religion therefore have tended to take for granted the given-ness, and the cultural universality, of religion, even when, as with, for example, Marx or Freud, religious phenomena are reduced to aspects of other (universal) human processes.2 Insofar as the dominant approach to religious studies today (at least as it affects the teaching and the organization of departments and professional associations) remains phenomenological in its orientation, this tendency is greatly exacerbated, and the purpose of study is imagined to be understanding religion as irreducibly religious. But there has been an increasing dis-ease with this set of assumptions for decades, beginning almost fifty years ago with the demurrals of Wilfred Cantwell Smith,3 and gathering increasing momentum in the almost thirty years since the publication of Jonathan Z. Smith’s Imagining Religion.4 Since that time, dozens of scholars have, in a variety of different and at times incompatible ways, questioned the cogency, validity, and utility of “religion” as a concept.5 Such is the present mood that Stanley Stowers asserts that “the rhetoric of despair about ‘religion as an object of study’ has become nearly hysterical.”6 This “despair” merits examination: Is there something odd, something misleading or problematic, about “religion” as a category? In fact, in my view, there is: the idea of “religion” as a bounded entity and a natural type not only represents an imposition on non-Western and nonmodern cultures (an imposition that could be tolerated if it bore intellectual fruit), but is an analytically incoherent concept, an idea that inhibits, rather than facilitates, our understanding of those phenomena we designate “religious.” “Religion” is not fundamentally a scholarly, academic, or analytic category (at least not usefully); it is a political and legal category demarcating cultural territory on which the modern secular state will not tread.
I
Wilfred Cantwell Smith’s dis-ease about the category of religion represents, in many ways, the limits of phenomenology, the approach’s pursuit to its ultimate conclusions. He takes seriously the phenomenological dictum that religion must be handled on its own (sui generis) terms.7 He likewise adopts the phenomenological viewpoint that observable religious phenomena are but historically variable expressions8 of the underlying force that motivates them, that is, a faith in, and relationship to, deity:9 “A lively faith involves a limpid sincerity of relationship to one’s fellow men, and to oneself, and to the Creator or ground or totality of the universe. For these things the formalities of one’s religious tradition are at best a channel, and at worst a substitute.”10 In addition to rather less important claims,11 this observation forms the backbone of Cantwell Smith’s critique of “religion” as a notion—it diminishes the experience and claims of the practitioner, and it confuses the manifestation or phenomenon (religion) with the invisible essence (faith, deity) of which the phenomenon is but the contingent shadow.12 Thus does phenomenology’s insistence on apprehending its subject matter “as it presents itself” render that subject matter invisible and impervious to investigation.13 This paradox also reveals to some degree the theological tendentiousness of the idea of religion, insofar as the very notion refers itself to and defines itself in terms of its own commitment and claims to transcendence, and, at the same time, manifests the ambivalence associated with the concept by the religious themselves, insofar as the notion is inadequate to the ineffable “reality” to which it purportedly refers.14
Jonathan Z. Smith, more recently and with greater effect, has turned Cantwell Smith’s arguments inside out, preserving the insight that classifying phenomena as religious involves a dimension of artificiality and even distortion, but defending the taxon nonetheless. J. Z. Smith agrees with Cantwell Smith that “religion” does not correspond to something “really there” in the world, apart from our formulation of it:
But man, more precisely western man, has had only the last few centuries in which to imagine religion. That is to say, while there is a staggering amount of data, phenomena, of human experiences and expressions that might be characterized in one culture or another, by one criterion or another, as religion—there is no data for religion. Religion is solely the creation of the scholar’s study. It is created for the scholar’s analytic purposes by his imaginative acts of comparison and generalization. Religion has no existence apart from the academy.15
J. Z. Smith’s comments represent an important and influential effort to denaturalize the idea of “religion.” But J. Z. Smith does not draw the conclusion from this that religion as a category should be avoided, cannot be defined, or must not be imposed on the data. Quite the contrary: if our goal is to understand those data we designate religious, such an active mental process must, of necessity, involve a transformation and manipulation of those data, classifying them according to our questions and agenda rather than merely describing them as they appear to insiders. Cantwell Smith’s observation that the religious individual would not recognize our characterizations is for Jonathan Smith not an argument against the concept’s utility, but an argument for it. In a variety of places, J. Z. Smith uses the metaphor of a map to defend and even actively promote the distorting effects of our efforts at classification and explanation. Just as a map’s utility consists precisely in its schematizing and condensing the territory it represents, in order to make that territory more cognitively apprehensible in particular ways, so also does scholarship necessarily do violence to the data it translates: “it is the very distance and difference of ‘religion’ as a second-order category that gives it cognitive power.”16 Thus the effect of denaturalization is not a repudiation of the concept, but rather increased self-consciousness about what intellectual processes might be served by demarcating some data as religious data. Cantwell Smith is found standing on his head, and is turned right side up: “religion” is not an insider category, but it is for that very reason a valuable intellectual tool for reconfiguring data in intellectually constructive ways.
II
Since the publication in 1982 of J. Z. Smith’s Imagining Religion, the idea that religion is not a universal, cross-cultural, or natural category has proliferated widely. While many (perhaps still the majority of) students of religion—not only phenomenologists, but also the more recent and quite radical cognitive approaches to religion17—continue to treat it as a given and essentially self-evident cross-cultural universal, more and more theoretically and methodologically self-conscious scholars have observed that religion as a category is a culture-specific, historically fixed, and perhaps ideologically loaded fabrication of more or less recent time, one that does not necessarily find parallels in non-Western or nonmodern cultures, and one that taxonomizes the world in ways that separate things those cultures associate, and associate things those cultures separate.18 More importantly, it may also be that such a division separates things whose association would provide analytical leverage, and associates things that convey no explanatory advantage at all.19 In the field of anthropology, from which religious studies has traditionally derived its more influential definitions of religion,20 more sophisticated understandings of religion as a culturally specific, recently fabricated, and perhaps misleading taxon are being explored, especially by Webb Keane and Maurice Bloch.21 Bloch asserts that “anthropologists have, after countless fruitless attempts, found it impossible to usefully and convincingly cross-culturally isolate or define a distinct phenomenon that can analytically be labeled ‘religion.’ ”22 Within religious studies as well as anthropology, Talal Asad’s earlier challenge to modern assumptions about religion has proven very influential.23 For Asad, the view of religion as distinctively personal-individual, interiorized, and focused on sincerity of belief is a direct reflection of the modern invention of the secular state:
with the triumphant rise of modern science, modern production, and the modern state, the churches would also be clear about the need to distinguish the religious from the secular, shifting, as they did so, the weight of religion more and more onto the moods and motivations of the individual believer. Discipline (intellectual and social) would, in this period, gradually abandon religious space, letting “belief,” “conscience,” and “sensibility” take its place.24
To some degree the lineage of “religion” as a taxon and the agenda behind its formulation (and the construction of the academic study thereof) are at issue in these more recent discussions. The problem is not merely that the category of religion fails to “carve nature at the joints,” but rather that it carves our data in accord with some misleading, mischievous, or at least nonacademic project.25 These more recent criticisms agree with Jonathan Z. Smith that religion is a fabricated category, but disagree that it is a scholarly fabrication. For those who take this position, “religion” imposes on the non-Western or premodern data normative, or political, or theological undertakings alien to them, reads those endeavors into the data, and turns them into little more than reflections of sociopolitical projects or conclusions native to the European West.
The problem may simply rest in an ethnocentric misreading of our issues as their issues, a problem not because it fails to take an insider viewpoint “seriously,” but because it distorts the data in unhelpful ways. Daniel Dubuisson frames the issue with admirable precision:
has this notion [that is, religion] nevertheless acquired, thanks to critical studies conducted by the history of religions for more than a century, an indisputable and rigorous definition, capable of aiding us in discovering and understanding … anthropological invariables … or, on the contrary, captive to its origins and history, has it instead remained a kind of native concept, typically European, gathering and summarizing under its aegis the struggles of a Western consciousness grappling with itself?26
His own answer is that “the human sciences (and among them the history of religions) have frequently been content, often unknowingly, sometimes naively, at other times arrogantly, consciously to revive a prejudice—and one of our dearest native categories.”27 Analogously, Daniel Boyarin, locating the origin of the Western notion of religion in the Christianity of the Patristic era,28 sees in it a potential Christian imposition on other cultures, a projection of one religion’s own self-image as a religion onto those other data sets it chooses to so designate—in the case of Boyarin’s analysis, Judaism.29 Similar claims have been made for Hinduism30 and Buddhism,31 among others. If these views have merit, the primary vector for what J. Z. Smith decries as the imposition of theological categories onto the analysis of religious data32 could be the notion of religion itself.
III
More may be going on, however, than mere oblivious ethnocentrism. Some scholars, particularly those focused on the study of religions of “the East,” have argued that “religion” is used as a stalking horse for colonial agenda, as is the identification of variegated national-ethnic traditions as reified and single “religions” (for example, “Hinduism”). Influenced the discourse criticism of Michel Foucault and by postcolonial theorists such as Edward Said, the very act of classification itself is seen as an act of power, an imposition on native cultural patterns and types of knowledge.33 The fact that it is first the colonial administrator and then secondarily the Western scholar who gets to classify types of native discourse as “religious” (or conversely as something else: “philosophical,” “legal,” “magical,” “terrorist,” and so on) or as manifestations of this and not that tradition shows this act of classification for what it is: a political activity, and one particularly related to the colonial and imperial situation of a foreign power rendering newly encountered societies digestible and manipulable in terms congenial to its own culture and agenda.
According to such views, the very point of religious studies as a field may be to process the data generated by the colonial project, implicating the scholar deeply in the mechanisms of state. Such a view of religion is nicely exemplified by Tomoko Masuzawa: “ ‘World religions’ as a category and as a conceptual framework initially developed in the European academy … quickly became an effective means of differentiating, variegating, consolidating, and totalizing a large portion of the social, cultural, and political practices observable among the inhabitants of regions elsewhere in the world.”34
Likewise, Daniel Dubuisson sees something insidious involved in the application of this native European category to outsiders: a mode of affirming our own sense of superiority. Insofar as “religion” is a Western and European concept, it will find its clearest and most developed manifestations in European avatars. As a result, to universalize the category is to create a universal human quality or characteristic that is, in fact, best manifested in the European person, hence marking that person’s superiority. Dubuisson describes this process as leading to overt rationalizations for imperialism:
The majority of other religions were henceforth viewed as rough drafts, archaic or primitive forms of our religion. The universal undeniably exists, but at different stages of development. By having these religions succeed one another along a single temporal axis, where the West clearly occupied the terminal position, the differences that were observed lost all capacity to subvert.… At a single stroke, imperialism and colonialism were equally justified and even, with the impetus of missionary activity, received an unanticipated moral guarantee.35
Or again, Timothy Fitzgerald sees the creation of “religion” as a way of sidelining and deprecating native traditions, while simultaneously naturalizing the functioning of the modern Western state and economy.36
According to such readings as these, the category religion and the choice to use that category to make sense of data derived from other cultures represent not the map-making of self-conscious scholarship, but at best the distorting and un-self-conscious imposition of native European categories onto cultures that do not share the same organizational principles of those of European modernity, making those cultures into little more than mirrors for our own prejudices and self-understanding. At worst, it represents a deliberate and self-interested exercise of discursive power, an imperial dissection of others’ cultures into manageable bits, and their relegation to the irrelevant realm of the supernatural. At issue, then, in much of the current discussion is not so much the artificiality of the category of religion as its implication in political agendas of questionable merit.
IV
With all of this in mind, my own view is that the most useful way of thinking about the category of religion is as a historically specific, emic product of the West, an anthropological folk category tied up with the development of the modern state, and the ways in which the modern state has defined itself as secular, and thereby needed to generate a shadow image of itself, a realm of collective voluntary commitment rooted in (irrational, variable, and uncompelled) personal belief, that the state would not partake of or constrain.37 Thus we may answer the question as to what kind of category religion is by asserting that religion is first and foremost a political category, albeit an ambivalent one, with its apotropaic function of referring to qualities from which the state seeks to dissociate itself. This fundamentally political role is reinforced by the fact that the most consequential and efficacious definitions of religion are those of the state, embodied in the tax code and in judicial decisions about what does and does not constitute religious observance. As Winnifred Sullivan points out, this means that the government itself defines the realm in which “religious freedom” is allowed to operate, thus making that “freedom” paradoxically subject to the edict of the state.38
Historically, this concept of religion is secondary to the development of states that dissociated themselves from ecclesiastical institutions to the point of distinguishing citizenship independently of church affiliation, a process begun in the Reformation and culminating in revolutionary secular states in America (1776) and France (1789). The idea of religion as a bounded, independent, distinguishable, and above all universal aspect of human sociality is the product of a specifically Western history.
It is true of course that there are periods for which and cultures in which concepts akin to our modern notion of religion do surface. Daniel Boyarin, for instance, points to expressions of identity in late antiquity that appear to revolve around types of allegiance to the divine.39 Or again, the category might be applied to any number of premodern discussions of observances with respect to the gods or cult (for example, Cicero’s de legibus and de natura deorum, Plutarch’s de superstitione). But it takes until the modern period for the notion of religion to be generalized. Only in the last five centuries or so, and with increasing force in the last two hundred years, has the liberal notion of the state as a negative entity developed, that is, the belief that the role of the state is to protect individual self-expression, rather than constitute it. Religion is created as the shadow image, the denied other, of that secular state, as the personal self-construction of an identity imagined as distinct from national identity; and this state of affairs is naturalized and universalized as the common condition of the human race. What we call “religion” is not theorized as an invariable and cross-cultural feature of human nature—never mind a feature with the kind of deracinated, apolitical, otherworldly force we moderns accord it—until the institutionalized secular state established as semipermanent its “religion’s” condition of possibility.
That condition of possibility can be elaborated more precisely. Maurice Bloch notes that those kinds of phenomena we tend to categorize as “religion” are in themselves little more than (an arbitrary subset of) representations of the social abstract, a “transcendental social consist[ing] of essentialized roles and groups.”40 As such they do not constitute a natural clade at all—they are intertwined with and to be understood alongside and as functionally identical to other varying forms of imaginative social self-reference such as nationalism,41 prejudice, kinship groupings, and so on. It is only on those occasions when some types of social self-reference—the mythology and symbology of who “we” are—for one reason or another come to be detached from the mechanisms and conception of the state, whatever those may be, that the strange deracinated cultural signifier we moderns call “religion” seems to rear its head. It is precisely for this reason that we do encounter phenomena that look like our notion of “religion” long before the invention of religion as an important modern taxon. At any time or in any place in which for one reason or another the coercive state power is detached from other types of social self-signification, we misrecognize the latter as a thing, “religion.” Again, however, it is only with the advent of the modern nation—in which the ecclesiastical frameworks for expressing the social imaginary are expelled from the purview of the state, or from state compulsion and discipline—as the normative state of affairs that this distinction comes to be permanent and institutional, and, hence, comes to be generalized as a universal and crucial category of human experience.
Religion has thus come to be, as a result of the Western invention of a form of state that deliberately dissociates itself from certain aspects of social identity (that is, those it identifies as “religious,” in part because of their past association with ecclesiastical institutions), a central cultural tool that we—modern, Western heirs to Reformation, Enlightenment, revolution, and the secular nation-state—use to describe and make sense of ourselves: our history; our commitments, subjectivities, and identities; and (above all else) our circumscribed political institutions as distinct from other types of social activity. Moreover, “the separation of religion from the transcendental social in general is, even in the places where it appears at first to exist, superficial and transient.”42 The unstable and arbitrary historical accident that creates a sharp division within the field of the social imaginary has thus generated a rather ambivalent and incoherent conceptual tool, a category that gets justified, demarcated, and defended in self-contradictory ways. Religion is identified with the quintessence of the human self, the most ineffable interiority, the ultimate concern so powerful as to be immune from criticism or coercion; and simultaneously it is peripheralized as irrational, as subject to individual and arbitrary choice, and as positively ruled out of state activity and effective social potency. It comes to refer to a realm of human desire that is simultaneously ideal and unreal; in fact, the category as a category implicitly asserts that what is ideal is unreal, is unrealizable socially. It is precisely for this reason that the religiously committed cannot concede that what they are doing is in fact “religion,” as Cantwell Smith noted. They are wary of the devil’s bargain that they are permitted complete freedom to pursue whatever “religious” practices and beliefs they wish,43 in exchange for which they must assent to the cost of those practices and beliefs being utterly irrelevant. The price of religious freedom is eternal insignificance.
V
This conclusion has ramifications for the role the imposition of “religion” as a discursive element might have had in European colonialism. In my view, postcolonialist readings of religion as itself a tool for colonization or as the direct product of reflection on newly “discovered” cultures’ (various) doctrines of and practices related to the gods may be somewhat overstated. The European West always had contact with peoples of differing views and cultural practices: Muslims and Jews, most obviously, but also pagan Europeans well into the Middle Ages (and beyond, in epic literature and the like), the classical pagans via history, literature, and philosophy, and even (albeit more problematically) Christian heretics and apostates. So it is not as though Europe lived in a dreamland of borderless Christian totality, like water in water, until the shocking discovery of Columbus that some people did not believe in Jesus. Certainly overseas exploration contributed data and additional impetus to develop categories with which to relate “their” practices to “our” practices, but it did not introduce the European awareness of religio-cultural difference. Nor again can the sometimes-romantic characterizations of precontact non-Europeans be sustained. It is not as though prior to the onslaught of Enlightenment disenchantment all the world lived in an undifferentiated cultural soup, a kind of unmediated, undivided integrity of spirit infusing all of daily life. All cultures and all societies carve up the world, make distinctions, classify both objects and social practices, albeit in different ways.44 So it avails us little—aside from producing a frisson of moral horror and self-loathing—to imagine the incursion of European conquistadors as a kind of Freudian trauma, a radical shock to a body politic earlier marked by primordial integration. Besides being excessively romantic, such an argument against religion as a category can (somewhat ironically) provide a high-minded rationalization for a return to the phenomenological insistence on taking “religion” as it presents itself, that is, as incomparable, transcendent, and ultimately unanalyzable. As Bruce Lincoln notes incisively, “When one permits those whom one studies to define the terms in which they will be understood, [and] suspends one’s interest in the temporal and contingent, … one has ceased to function as historian or scholar.”45
The European application of “religion” to societies first encountered in the modern period via voyages of exploration is probably not in the first instance a deliberate effort to dominate but to understand. This involves the simple, if flawed, process of Europeans somewhat naively applying their own (limited and culture-specific) categories onto “strange” data. The effect is ethnocentric, and can be unhelpfully distorting; among other things, such wholesale imposition of one’s own native categories can make one blind to the strangenesses of the data, to the genuinely illuminating differences between “us” and “them.” Moreover, it is certainly true that knowledge, manifested in taxonomizing the novelties one encounters, does express a will to power. But the main work being done with the category religion seems to concern European subjectivity, states, and self-identity. The imposition of these self-images on others,46 while distorting and even oppressive, is fundamentally a claim that those others are in some important way “like us,” are, in a word, human (and hence understandable). In fact, such likeness was seriously open to question among explorers of the early modern period, and some in fact denied “religion” to “the natives,”47 a position that, as Dubuisson notes, was openly dehumanizing.48 The use of religion as a cross-cultural category implies by contrast that the other is not sheer incomprehensible difference.
In fact, in light of what was said above about the origins of the idea of “religion,” it is easy to see why the imposition of colonial rule would bring with it a greater tendency to identify religion among those colonized: the state power has been seized by outsiders, but a persistent sense of “us” remains within the culture thus dominated. In this sense, then, colonialism—Western or not!—really does create “religion.” It does so not simply or primarily by imposing a Western taxon on non-Western cultures, but by creating a kind of ideological lacuna, a situation in which the social body is divided against itself and is imagined in a kind of bifurcated condition. The disempowered side of that bifurcation will then look to us moderns like what we call religion. The flaw or misconception at the heart of “religion” as a category is thus probably not to be found primarily in its oppressive importation into non-Western contexts by modern colonialism. Rather, the concept’s weakness derives from its origins as a political creation rather than an academic one, as well as the fact that the political agenda expressed in the concept is self-contradictory, confused, and ambivalent, making the concept itself enormously incoherent. As Maurice Bloch vividly expresses it: “To explain religion is therefore a fundamentally misguided enterprise. It is rather like trying to explain the function of headlights while ignoring what motorcars are like and for. What needs to be explained is the nature of human sociability, and then religion simply appears as an aspect of this that cannot stand alone.”49
VI
In sum, then, we might conclude that Jonathan Z. Smith is at least partly wrong: “religion” is not a scholarly category or construct.50 Most scholarly efforts to define religion have in fact simply been efforts to identify the natural type to which this Euro-Western folk category (perhaps imperfectly) must surely refer.51 At the same time, though, Jonathan Z. Smith is also partly right: we are trying to understand our data, and to do so requires us to some degree to frame that data in terms of the categories that make sense to us, even if in so doing we distort them somewhat or offend those whom we seek to understand. Self-critical reflection on the taxonomies of religious studies should not become an excuse for resorting to simple repetition and description of insider claims.
The value of “religion” will not depend on its isometry with the data it delineates and shapes,52 or on the validity of its claims to universal applicability. Rather, it will depend on the capacity the category might continue (or not) to have to surprise, to clarify, and to shape and transform our own organization of the human universe. If “religion” as an idea forces upon us new and helpful understandings of familiar data, if it leads to us juxtaposing phenomena we might not otherwise think to compare, and if in the process it leads us continually to rectify the taxonomy from which we started, it serves a valuable intellectual purpose, even if it does not refer to anything “real.” If, by contrast, the category does little more than provide a justification for comparing things we already see to be similar (or, worse, identical), if it simply reinforces our prejudices or turns foreign cultural data into evidence for the naturalness of our own practices and assumptions, then it is not simply intellectually unhelpful—it is positively pernicious. Which of these judgments is closer to the truth is a matter for debate. What is important—and what is promising about the present juncture in the field of religious studies—is that the question does get raised.
NOTES
Please note that a modified version of this article also appears as chapter 6 of William Arnal and Russell McCutcheon, The Sacred Is the Profane: The Political Nature of “Religion” (Oxford: Oxford University Press, 2012). Reprinted with permission from Oxford University Press.
1. So, for example, Russell T. McCutcheon, Manufacturing Religion: The Discourse on Sui Generis Religion and the Politics of Nostalgia (Oxford: Oxford University Press, 1997), 202; Tomoko Masuzawa, The Invention of World Religions; Or, How European Universalism Was Preserved in the Language of Pluralism (Chicago: University of Chicago Press, 2005), 316–17.
2. Durkheim constitutes something of an exception here; his definition of religion, whatever its problems, remains rigorously and thoroughly sociological, and hence leaves little “religious” in place to anchor it to the transcendent. See, for example, Émile Durkheim, The Elementary Forms of Religious Life, trans. Carol Cosman (Oxford: Oxford University Press, 2001), 36: “The division of the world into two comprehensive domains, one sacred, the other profane, is the hallmark of religious thought.… A rock, a tree, a spring, a stone, a piece of wood, a house, in other words anything at all, can be sacred.”
3. Wilfred Cantwell Smith, The Meaning and End of Religion: A New Approach to the Religious Traditions of Mankind (New York: Mentor, 1962).
4. Jonathan Z. Smith, Imagining Religion: From Babylon to Jonestown (Chicago: University of Chicago Press, 1988).
5. For example, Talal Asad, Genealogies of Religion: Discipline and Reasons of Power in Christianity and Islam (Baltimore: Johns Hopkins University Press, 1993); Daniel Dubuisson, The Western Construction of Religion: Myths, Knowledge, and Ideology, trans. William Sayers (Baltimore: Johns Hopkins University Press, 2003); Russell T. McCutcheon, The Discipline of Religion: Structure, Meaning, Rhetoric (London: Routledge, 2003); Timothy Fitzgerald, The Ideology of Religious Studies (Oxford: Oxford University Press, 2000); and others.
6. Stanley K. Stowers, “The Ontology of Religion,” in Introducing Religion, ed. Willi Braun and Russell T. McCutcheon (London: Equinox, 2008), 434. Interestingly, Stowers echoes here the observation of Cantwell Smith (Cantwell Smith, Meaning and End of Religion, 110) that the term “religion” carries a sense of malaise with it.
7. See, for example, Smith, Meaning and End of Religion, 154–55.
8. Ibid., 155.
9. Ibid., 114–17.
10. Ibid., 117.
11. That is, that religion as a concept is a culturally bounded notion specific to Islam and Christianity (Cantwell Smith, Meaning and End of Religion, 110–14), and that the term “religion” and, even more so, the names (and ideas) of individual “religions” (that is, Islam, Buddhism, Christianity) reify and make monolithic what are in fact extremely complex and historically changeable phenomena. Such an argument neglects the extent to which category formation is something we undertake precisely in order to simplify matters.
12. Cantwell Smith, Meaning and End of Religion, 124. This is stated clearly by Cantwell Smith, who claims that the student of religion runs the risk of “omitting not only the vitality but the most significant of all factors in that vitality, namely its relation with transcendence. The observer’s concept of a religion is by definition constituted of what can be observed. Yet the whole pith and substance of religious life lies in its relation to what cannot be observed.” Jonathan Z. Smith likewise characterizes Cantwell Smith’s argument as boiling down to this claim. See Jonathan Z. Smith, “Bible and Religion,” in Relating Religion: Essays in the Study of Religion (Chicago: University of Chicago Press, 2004), 214n28.
13. The similarities between the claims of phenomenology of religion about its subject matter and those of Calvinist theology about the Eucharist are striking and bear further investigation. For both, outer form is a vector for immaterial, spiritual content, which must not be confused or conflated with that content. See Webb Keane, Christian Moderns: Freedom and Fetish in the Mission Encounter (Berkeley: University of California Press, 2007), 61–65.
14. Strangely, Cantwell Smith’s demurrals are paralleled today in the popular phenomenon of individuals claiming to be “spiritual, not religious.” This suggests that Cantwell Smith, perhaps, is just as much a religious datum in need of explanation as are the self-described “spiritual.” On this terminology and its significance for our understanding of religion, see Janet M. Klippenstein, “Imagine No Religion: On Defining ‘New Age,’ ” Studies in Religion 34, nos. 3–4 (2005): 391–403.
15. Smith, Imagining Religion, xi.
16. Smith, “Bible and Religion,” 208.
17. This is a fatal weakness of the otherwise-promising cognitive approaches: their tendency to treat “religion” as a given, rather than the artificial and, especially, culturally specific (that is, Western and modern) category that it is. Note, for example, the title of Pascal Boyer, The Naturalness of Religious Ideas: A Cognitive Theory of Religion (Berkeley: University of California Press, 1994). This weakness of cognitive approaches is discussed brilliantly in Maurice Bloch, “Why Religion Is Nothing Special but Is Central,” Philosophical Transactions of the Royal Society 363 (February 2008): 2055–61.
18. See the helpful discussion in Malory Nye, Religion: The Basics (London: Routledge, 2003), 12–18; cf. Dubuisson, Western Construction, 112–15.
19. In other words, religion is a polyphyletic category—it associates and confuses objects that range across different categories. An excellent example of such categorization is found in the discussion of zebras offered under the rubric of “What, if anything, is a zebra?” in Stephen J. Gould, Hen’s Teeth and Horse’s Toes (New York: Norton, 1983), 355–65.
20. Especially that of Clifford Geertz, “Religion as a Cultural System,” in Anthropological Approaches to the Study of Religion, ed. Michael Banton (London: Tavistock, 1985), 1–46. I also note that the variety of attempts to cut the Gordian knot by resorting to more “common-sense” and abrupt definitions of religion as discourse pertaining to “nonobvious” or “counterintuitive” beings is little more than a restatement of E. B. Tylor’s—also anthropological—definition of religion as discourse pertaining to “spiritual beings.”
21. See, for example, Bloch, “Religion Is Nothing Special”; Keane, Christian Moderns.
22. Bloch, “Religion Is Nothing Special,” 2055.
23. Asad, Genealogies of Religion.
24. So ibid., 39; cf. also Talal Asad, Formations of the Secular: Christianity, Islam, Modernity (Stanford: Stanford University Press, 2003), 181–94; William E. Arnal, “Definition (of Religion),” in Guide to the Study of Religion, ed. Willi Braun and Russell T. McCutcheon (London: Cassell, 2000), 30–33; Keane, Christian Moderns, 213–14; McCutcheon, Manufacturing Religion, 127–57; and others.
25. Plato, Phaedrus, 265d–266a.
26. Dubuisson, Western Construction, 5–6.
27. Ibid., 115.
28. See Daniel Boyarin, “The Christian Invention of Judaism: The Theodosian Empire and the Rabbinic Refusal of Religion,” in Religion: Beyond a Concept, ed. H. de Vries (New York: Fordham University Press, 2008), 152–54, citing, among others, Gregory Nazianzen, Eusebius, and Epiphanius. This early identification of the “invention” of religion is somewhat odd, but not unique—a number of theorists see in the initial separation of ideological commitments from the state, found in nascent Christianity in part due to its illegality and in a definition of cultural entities in terms of commitments to and modes of worship of a given deity, a foreshadowing or even an initial fabrication of the modern notion of religion.
29. So, ibid.
30. See, for example, Richard King, Orientalism and Religion: Postcolonial Theory, India and “the Mystic East” (London: Routledge, 1999).
31. See the discussion in Masuzawa, Invention of World Religions, 121–46.
32. See, for example, his comments in Jonathan Z. Smith, Drudgery Divine: On the Comparison of Early Christianities and the Religions of Late Antiquity (Chicago: University of Chicago Press, 1990), 34.
33. See, for example, Nye, Religion, 12–15.
34. Masuzawa, Invention of World Religions, 20; cf. 29. For an example of how native political actions in a colonial context can be redrawn as “religious” and, in the process, attenuated, see McCutcheon, Manufacturing Religion, 167–77.
35. Dubuisson, Western Construction, 114–15.
36. For example, Fitzgerald, Ideology of Religious Studies,8.
37. For an elaboration of this view, see, among others, Asad, Genealogies of Religion; Arnal, “Definition”; William E. Arnal, “The Segregation of Social Desire: ‘Religion’ and Disney World,” Journal of the American Academy of Religion 69, no. 1 (March 2001): 1–19; Timothy Fitzgerald, “Encompassing Religion, Privatized Religions and the Invention of Modern Politics,” in Religion and the Secular: Historical and Colonial Formations, ed. Timothy Fitzgerald (London: Equinox, 2007), 211–40; Fitzgerald, Discourse on Civility and Barbarity: A Critical History of Religion and Related Categories (Oxford: Oxford University Press, 2007); McCutcheon, Discipline of Religion, 230–90.
38. Winnifred F. Sullivan, The Impossibility of Religious Freedom (Oxford: Oxford University Press, 2005).
39. So Boyarin, “Christian Invention of Judaism.”
40. Bloch, “Religion Is Nothing Special,” 2056.
41. Cf. Benedict Anderson’s characterization of nation as an “imagined community.” Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism, rev. ed. (London: Verso, 2006).
42. Bloch, “Religion Is Nothing Special,” 2060.
43. With the important caveat, noted earlier, that this “freedom” is in fact circumscribed by the state itself, insofar as a practice or belief must be legally defined as religion to be thus protected—the initial gesture of state noninvolvement is state definition.
44. Note, too, that among other things this means that cultural influences never move simply from the dominating to the dominated culture, but are reciprocal. This is a point made perhaps most forcefully by Marshall Sahlins, although others have noted it as well; see especially Marshall Sahlins, Islands of History (Chicago: University of Chicago Press, 1987).
45. So Bruce Lincoln, “Reflections on ‘Theses on Method,’ ” in Secular Theories on Religion: Current Perspectives, ed. Tim Jensen and Mikael Rothstein (Copenhagen: Museum Tusculanum Press, 2000), 120–21. See also his comments in the same piece specifically addressing the scholar’s descriptions of other cultures (ibid., 119): “Reverence is a religious, and not a scholarly virtue. When good manners and good conscience cannot be reconciled, the demands of the latter ought to prevail. Many who would not think of insulating their own or their parents’ religion against critical inquiry still afford such protection to other people’s faiths, via a stance of cultural relativism. One can appreciate their good intentions, while recognizing a certain displaced defensiveness, as well as the guilty conscience of western imperialism.”
46. Often in the deeply misleading form of “human nature,” on which see Marshall Sahlins, The Western Illusion of Human Nature: With Reflections on the Long History of Hierarchy, Equality, and the Sublimation of Anarchy in the West, and Comparative Notes on Other Conceptions of the Human Condition (Chicago: Prickly Paradigm, 2008).
47. For example, Jonathan Z. Smith, “Religion, Religions, Religious,” in Relating Religion: Essays in the Study of Religion (Chicago: University of Chicago Press, 2004), 179, refers to Pedro Cieza de León’s Crónica del Péru (1553) as denying religion to Andean indigenous peoples; Nancy Senior, “ ‘Sathans Inventions and Worships’: Two 17th-Century Clergymen on Native American Religions,” Studies in Religion 35, no. 2 (2006): 279–80, refers to Samuel de Champlain’s Des sauvages (1603) and Gabriel Sagard’s Le grand voyage du pays des Hurons (1632) for the idea that while North American natives may believe in God, they have no systemic worship.
48. Dubuisson, Western Construction, 114.
49. Bloch, “Religion Is Nothing Special,” 2060.
50. It may be used as such, as it clearly is in Smith’s own work, but this seems more the exception than the rule.
51. This, in my view, is all that is accomplished, for instance, by the excruciating definitional efforts of Geertz, “Religion as a Cultural System.”
52. As Smith stresses, it is precisely the difference between analytic concepts and the self-presentation of the data that gives those concepts explanatory force. So, for example, Smith, “Bible and Religion,” 208.