The Very Idea of Religion


Paul J. Griffiths


Copyright (c) 2000 First Things 103 (May 2000): 30-35.

The study of religion as an academic discipline is now close to a century and a half old. In the United States, at least, it appears to be flourishing. There are scores of university departments devoted to it, several hundred doctoral degrees earned in it each year, a professional organization (the American Academy of Religion) that draws almost ten thousand people to its yearly meetings, and numerous journals (most, it is true, with a small circulation and a smaller readership) for publication of the writings of those who identify themselves with the discipline.

And yet it is a scholarly enterprise without an identity, one that lacks any widely shared understanding of its central topic, or of the methods appropriate to the study of this topic. This may seem puzzling. After all, the idea that there is such a thing as religion (and that it is important) is enshrined in the foundational texts of the American republic, and in its legislative, bureaucratic, and judicial procedures. The First Amendment guarantees freedom of it (or at least that Congress shall neither establish it nor prevent its free exercise); the IRS must often decide whether an institution is sufficiently religious to be granted tax privileges; Congress and the states have often made laws in which religion figures prominently (recall the 1993 Religious Freedom Restoration Act and its more recent clones passed by several state legislatures); and, most obviously, the courts must often rule as to whether some belief or practice is or is not religious.

In their public forums and official voices, then, as well as in their private discussions, America and Americans have a lot to say about religion. Listening to these discussions, though, rapidly suggests the conclusion that hardly anyone has any idea what they are talking about—or, perhaps more accurately, that there are so many different ideas in play about what religion is that conversations in which the term figures significantly make the difficulties in communication at the Tower of Babel seem minor and easily dealt with. These difficulties are apparent, too, in the academic study of religion, and they go far toward an explanation of why the discipline has no coherent or widely shared understanding of its central topic.

Superficially, perhaps, matters are clear enough at the formal level. "Religion" picks out a genus of which there are many species, and we all know what at least some of the species are: Judaism, Buddhism, Christianity, Islam, and so on. It is, in this formal respect, a term just like "currency," a genus whose species include dollars, deutschmarks, pounds, and yen. But while there is fairly widespread agreement about what makes it reasonable to say that dollars and deutschmarks are species of the same genus (and about what genus they’re a species of), there is much less about what makes it reasonable to say that Islam and Buddhism are species of the same genus (and about what genus they’re a species of). There are few impassioned arguments about whether something is or is not a currency, but many about whether something is or is not a religion and even more about whether some action, event, or belief does or does not merit the label "religious."

This state of affairs wouldn’t have worried Parson Thwackum, an aggressively insular Anglican priest in Henry Fielding’s novel Tom Jones, first published in 1749. Thwackum takes the high stipulative road in talking about religion: "When I mention religion," he says, "I mean the Christian religion; and not only the Christian religion, but the Protestant religion; and not only the Protestant religion, but the Church of England." Thwackum uses the term "religion" as a label for the beliefs and practices of his preferred form of Christianity, his church; this provides him with the essential characteristics of the genus, and thereby also the means to judge whether anything else can be said to be a species of it. On this view, the Church of England is the ideal type of religion; other forms of life can be called religions just to the extent that they approximate or approach it. This is a theological (or, as much contemporary scholarship in religion would have it, an ideological) understanding and use of the term "religion." It is also, arguably, just the kind of understanding of religion evident in the thought of such otherwise very different figures as the founding fathers of this republic and the present Pope—though they would have agreed neither with Thwackum nor among themselves about what the ideal type of religion is.

Such understandings of religion are, however, anathema to practitioners of the academic discipline of the study of religion. This discipline, also variously known as "religious studies," "history of religions," Religionswissenschaft (and so on), developed in self–conscious opposition to theological understandings of religion like Thwackum’s. Its founding fathers wanted (and some of its practitioners still want) an understanding of religion not derived from the self–understanding of some church, without theological commitments, sufficiently subtle and interesting to permit interesting intellectual work to be done, and capable of being agreed upon by a community of scholars sufficiently large to make possible the development of a discipline with religion as its central topic. But no such understanding has yet been found: Parson Thwackum’s embrace has proved difficult to escape. A little history will show why.

The Latin word religio, from which we get "religion," was not an especially important term for pre–Christian thinkers who used that language; and there is no single Greek word for which it is the obvious translation. This lack of importance is reflected, too, in early Christian literature. Jerome’s fourth–century Latin version of the New Testament, for instance—the version that, with some revisions, was the standard text for Western Christians for a thousand years—uses religio and its derivatives only six times, and to render a number of different Greek words. The word is simply not an important part of the biblical lexicon, and this is evident, too, in the decisions made by early translators of the Bible into English. The King James Version (1611), for instance, uses "religion" or "religious" only five times in its rendering of the New Testament from Greek, and for three different Greek terms (not always the same ones that Jerome chose to render with religio).

The term is, of course, sometimes used by Christians in late antiquity. Augustine, for instance, gave detailed attention to it at the end of the fourth century in his work De vera religione (On True Religion), and occasionally elsewhere in his work. For him, religio meant worship, those patterns of action by which, in public, we self–consciously turn ourselves toward God in homage and praise. There could, he thought, be a right and proper ("true") way of worshiping God, just as there could be improper and damnable ("false") ways of doing so; and so there could be a true and many false "religions." Since Augustine was a Catholic Christian, he also thought that Christian worship was, on the whole, identical with true religion; and that, although true religion (proper public and communal worship) was not found only within the bounds of the Christian church, it was found preeminently and most perfectly there.

This equation of religion with public and communal worship was not unique to Augustine. It was almost standard in the pre–Christian Mediterranean world, and it became the ordinary understanding of religio among those Christians of late antiquity who thought and wrote in Latin. This understanding of the word is evident, too, in the etymology of religio most commonly given by Latin–using intellectuals (Christian and otherwise) in late antiquity. This etymology derives religio from re + ligare, "to bind back," or "to re–bind," meaning to reestablish by worship a lost or broken intimacy between God and worshipers. (There are other etymologies of what Walker Percy calls this "peculiar word" defended by a minority both ancient and modern, the most interesting of which derives religio from re + legere, "to re–read"; but this etymology has entered less deeply into the soul of the West.)

It is important to notice, too, that Western Christians from the fourth century onward, after Constantine’s Edict of 313 had made Christianity the mode of worship and theological thought favored by the state, had little occasion to think or write about those things that we now usually call "religions." Islam did not come into existence until the seventh century, and until the Renaissance was most often thought of by Christians as a Christian heresy rather than a non–Christian religion: this is how John of Damascus, the first Christian to give it serious attention, categorized it. The religions of India, China, Japan, Africa, and America were effectively unknown until the sixteenth century; and Judaism, in spite of the many lively Jewish communities in Europe, was a topic of interest to Christians largely as a precursor to Christianity, a praeparatio evangelica. And so Christianity was rarely, if ever, thought of by Christians as one religion among many: the idea that there is a genus called "religion" of which there are many species did not gain much currency until the seventeenth century. It is, by and large, a modern invention.

Insofar as there was a standard use of religio in Europe between the effective end of Roman hegemony in the fifth century and the cataclysm of the Reformation in the sixteenth, it was to denote the activities and members of the monastic orders. These were typically called religious orders, and their members were simply the religious. This usage has survived, in somewhat attenuated form, in the Catholic Church, where it is common to hear people speak of "the religious life" and mean by it life within a religious order.

Modern (post–Reformation) understandings of religion differ from these premodern uses most dramatically in that they see religion exactly as a genus of which there are many species. One influence upon the acceptance of this idea was the pressing necessity in the seventeenth century of creating political forms of life in Europe that could peacefully accommodate a wide variety of Christian groups with incompatible understandings of what it is to be a Christian, and often with a deep hatred of one another. After the Thirty Years’ War in Europe (1618–1648) and the Civil War in England (1642–1648), in which such differences showed themselves clearly in large–scale and long–lived violence, it was clear that the political forms which had served Europe fairly well for the preceding millennium would no longer do, and that any new ones would have to find a way of dealing with the violent splintering of Christendom brought about by the Reformation.

The political solutions that emerged were of two kinds. The first affirmed the idea that a sovereign state could and should accommodate only one Christian group, and that your religion (now it began commonly to be called that) should therefore be determined by geography, by where you happened to live. Calvin’s Geneva provides one instance of this solution, as does the English settlement of 1688; both use the idea that there are many religions, and that the state should establish and give special privileges to one among them. The second kind of political resolution used the idea that the state should be neutral with respect to religion (which usually meant neutral with respect to the various brands of Protestant Christianity; Jews and Catholics were usually beyond the pale, and Buddhists and Muslims did not enter into consideration). The passage of the First Amendment to the American Constitution in 1791 provides a paradigm here. This second kind of political resolution, like the first, required (usually in very explicit terms) the view that religion is a genus with many species—although the understanding of "religion" in play was always entirely Thwackum–like. "Religion" always referred to one or another form of Protestant Christianity as its ideal type.

Almost equally important to this new understanding of "religion" was the vast increase in European knowledge of the history, languages, and practices of non–European civilizations. Beginning in the fifteenth century (and increasing almost exponentially in the sixteenth and seventeenth) reports of the habits and practices of the Indians, the Chinese, the Japanese, and the inhabitants of Meso–America began to be available to the literati of Europe. The earliest among these reports were written by Catholic missionaries for the use of the Church in its efforts to propagate itself; but these were soon followed by work sponsored by the European states with interests in empire–building, first the Portuguese, Spanish, and Dutch, and later the English and French. By the seventeenth century, grammars and lexicons of hitherto exotic and unknown languages (Sanskrit, Chinese) began to become available, and throughout the eighteenth and nineteenth centuries works in these languages were translated in ever–increasing numbers into those of Europe.

Much of the information gathered in these ways seemed to European intellectuals, Christian and otherwise, to reveal forms of life and patterns of belief both deeply like and importantly unlike Christian forms and patterns. The Indians wrote hymns and prayers to many gods, and seemed to worship images and statues of them; the Chinese had temples, canonical works, and a highly developed ritual system; and so on. It began to seem natural to European historians, philosophers, and theologians to think of these forms of life as the religions of India and China, and also to think of Christianity as the religion of Europe. It was then not difficult to move to the more abstract theoretical view that there is a genus called religion of which there are many species; and this was effectively the standard position by the eighteenth century.

Over these intellectual moves, too, Parson Thwackum’s spirit hovers: "religion" means "things like Christianity." The likeness often became quite strained, but the controlling power of the paradigm case can be seen clearly in the endless nineteenth–and early–twentieth–century debates about whether such things as Buddhism (no God?) or State Shinto (the Japanese Emperor as God and no theology?) are really religions—for they aren’t really very much like Christianity. Even in the analyses of religion offered by such resolutely anti–Christian figures as David Hume (in The Natural History of Religion, first published in 1757), or by such quasi–Christian thinkers as Immanuel Kant (in Religion Within the Bounds of Reason Alone, 1793) and G. W. F. Hegel (in Lectures on the Philosophy of Religion, first delivered in 1821), "religion" is analyzed by a process of abstraction from features of Protestant Christianity.

This history, sketchy though it is, should help to explain why scholars of religion who don’t find Parson Thwackum appealing as a founding father have found it so difficult to construct an idea of religion that will meet the needs mentioned above. The historian of religion Jonathan Z. Smith puts the matter very clearly in an essay published in 1998: "‘Religion,’" he writes, "is not a native term; it is a term created by scholars for their intellectual purposes and therefore is theirs to define. It is a second–order generic concept that plays the same role in establishing a disciplinary horizon that a concept such as ‘language’ plays in linguistics or ‘culture’ plays in anthropology. There can be no disciplined study of religion without such a horizon." Smith might have added, though he does not, that such a horizon is, as a matter of fact, entirely lacking.

Others do add this conclusion. There are now effectively only two camps among the anti–Thwackums about the possible scholarly uses of the category of religion. The first argues that the attempt to find a nontheological understanding of religion around which a scholarly discipline can be constructed has failed, and that as a result the category ought to be dropped. On this view, work hitherto done under the religion umbrella ought to disperse quietly (but quickly) in search of new shelter under other academic umbrellas. History, literary studies, anthropology (a discipline with its own deep identity problems), and cultural studies are the usual suggestions. This is the avant–garde position; it amounts to a concession that Thwackum was right, and a concomitant agreement that if thinking from within a theological perspective is to be abandoned (as this camp would wish), use of the category "religion" must go with it. The other camp, fighting a desperate defensive action and struggling not to appear quaintly positivist, argues that a scientific (nontheological) understanding of religion can still be productively used, and that in order to do so the avant garde must be fended off on the eastern front, as it were, while the advance of the theologians is blocked to the west. Two recent books illustrate the divide very nicely.

The first is Timothy Fitzgerald’s The Ideology of Religious Studies (Oxford University Press). Fitzgerald is Associate Professor in the Department of International Culture of Aichi Gakuin University in Japan. He has worked also in Maharashtra, India, and appears to have lived in both England and the United States (clues dropped at various points in the book suggest that he may be English by birth and education). His book has two central purposes. The first is to show, historically, that those who have tried to develop a scientific understanding of religion capable of providing the necessary horizon for the discipline have failed to do so. The second is to show that when the category of religion is used in contemporary (nontheological) scholarship, it has no analytical uses and confuses rather than illuminates what is studied.

The book largely succeeds in both these purposes. The analysis of the work of the nineteenth–and twentieth–century founding fathers of Religionswissenschaft (Max Mueller, P. D. Chantepie de la Saussaye, Gerardus Van der Leeuw, Joachim Wach, Mircea Eliade), though brief, is convincing in its claim that the understandings of religion developed by all of them were informed by theological premises and goals. Fitzgerald does not mean by this that each of them was thinking explicitly about religion as Parson Thwackum did; neither does he mean that they were all Christians (or Jews). He means, rather (and rightly), that they assumed, sometimes implicitly and sometimes explicitly, "that a transcendental ontology provided the universal reference point . . . for all those forms of life that could be categorized as religious." Such an assumption, as Fitzgerald rightly says, is crypto–Christian, and thus at least implicitly theological. He doesn’t say, though I wish he had, that it is vastly less interesting and less theoretically productive than openly and fully Christian assumptions.

Fitzgerald is equally convincing in his analysis of the work of more recent scholars in religious studies, such as Ninian Smart. His discussion of Smart shows that, while Smart has moved further from explicitly theological assumptions than had (for example) Van der Leeuw, they can still be found in the subtext of his work; moreover, they provide the only reasonable explanation for his unwillingness to abandon the concept of religion. The difficulties are in fact more pressing for Smart and his like. They want, as Fitzgerald shows, to preserve the idea that religion is a sui generis dimension of human experience, recognizable cross–culturally and not to be analytically reduced to other categories such as culture, economics, and power. And yet at the same time they want to deny the ontological commitments that would make sense of such an idea, most especially the commitment to "belief in a transcendent being which provides individuals with a unique and irreducible category of experience, an intelligent purpose for humankind."

Fitzgerald’s view, in summary form, is that when theological assumptions about religion are jettisoned, the category ceases to be either believable (why should we think there is any such thing?) or analytically useful (there is nothing we can do with it in the way of explanation or prediction that we can’t do better with other categories). Continuing to employ the category in the absence of theology is, for him, a form of mystification that prevents useful analytical attention to cultural, economic, and political particularities. He provides interesting analyses of the Ambedkar–inspired Buddhist movement in India, and of contemporary (so–called) religious movements in Japan, in which the mystifying effects of applying the category "religion" in the absence of theological commitments are amply demonstrated.

It is important to see that what Fitzgerald argues for is the truth of a conditional. The conditional is: if Thwackum is left behind, the category of religion must go too. He doesn’t argue for the truth of the antecedent, which is that Thwackum should be left behind. In spite of his occasional protestations that he is not against theology, though, he would clearly rather Thwackum were left behind. It’s just that his book is not concerned to argue that point. With this reservation in mind, then, I suggest that theologians, whether Christian, Jewish, or Islamic, ought to give their support to Fitzgerald’s argument. It affirms, after all, only what theologians ought already to think, which is that an attempt to make sense and use of an idea of religion that systematically rejects theological assumptions will fail. It is heartening to see that at least some of those formed by the attempt to find a non–Thwackum–like understanding of religion are beginning, at last, to see that this is true.

Some, however, are not yet convinced. There are still some in the camp of those who think that a coherent and useful scientific (nontheological) concept of religion can be saved. Among the most vociferous of them is Donald Wiebe, whose book The Politics of Religious Studies: The Continuing Conflict with Theology in the Academy (St. Martin’s) gives this line of reasoning as good a defense as it is likely to get. The book contains revisions of essays by Wiebe (who is Professor of the Philosophy of Religion and Dean of the Faculty of Divinity at Trinity College, Toronto) published at various times between 1986 and 1997. In the Preface to the book he says: "Taken together, therefore, these essays constitute a much–needed book that advocates recovery of the scientific agenda in the academic study of religion, on which basis it achieved cognitive legitimation." An author who describes his own book as "much–needed" offends at least against good taste, and the book is saturated throughout with the author’s sense of his own importance. The book does however have the virtue of clarity, a virtue given to it in part by the endless repetition of its central claim.

This central claim is that the academic study of religion both can and ought to be understood as a science, which is to say that its only goals are understanding and explanation. Theology, by contrast, understood generically as "any kind of confessional or religious orientation," cannot be scientific in this sense precisely because it attempts more than understanding and explanation: it includes involvement in the object of study, much as a political activist cannot be scientific in his approach to politics because he intends more than understanding and explanation. The (public) university, furthermore, is the proper place for the scientific study of religion, because the university is interested only in science and has no place for theology.

Wiebe is concerned throughout with what he sees as theology’s continuing tendency to corrupt the pure science of religion. The chief defense, he suggests, ought to be the adoption of naturalism (not defined, but certainly taken to rule out appeal to or use of God as an explanation of anything), which is a way of viewing the world "implicit in a scientific methodology." Science, naturalism, and the public university therefore belong inextricably together in Wiebe’s thought, and this explains his passionate objection to the intrusion of theology upon the hallowed ground of the public university. The ramparts must be high and well defended if the public university is not to become fouled by the impurities of theology. This is stronger language than any Wiebe uses, but not by much: his work is informed by passions at least as deep as those that produce odium theologicum, and it is one of the interesting paradoxes of the book that a resolutely naturalistic position is assumed in terms that suggest a commitment of faith to it. Another is that the position is argued by a Dean of Divinity. What this means for the current state of theological thinking at Trinity College can only be imagined.

The book’s argument, then, is profoundly anti–Thwackum. Oddly, though, there is almost no serious engagement with perspectives such as Fitzgerald’s. Wiebe scarcely acknowledges that there are many, probably now a majority, among those who reject theological understandings of religion who also take the next step and reject the category altogether. These are not his opponents; he is concerned only with the theologians, but this fact drastically limits the usefulness of his book since it doesn’t permit engagement with the more interesting nontheological theorists of religion currently working. It also gives his work a deeply old–fashioned, even quaint air, and it means that he makes no attempt to show what can be done, intellectually speaking, with the scientific concept of religion he advocates. The challenge of work like Fitzgerald’s is, as a result, not met. But this is not the only nor even the most important difficulty.

A deeper problem is that Wiebe understands the public university to be capable of accommodating only those who adopt naturalism as the frame for their work. But why think this? Wiebe’s reasons appear to be historical. He seems to think that the public university in the United States, because it has roots in the German research universities of the nineteenth century, just is the kind of place in which naturalism is the frame for all intellectual work. If this is his view, it carries no implication as to the truth or falsehood of naturalism. The bowling alley, because of its history and design, is the place where you go to bowl; you don’t play tennis there, but this has nothing to say about the desirability or otherwise of playing tennis. If this is Wiebe’s view (and the book reads as though it is), then it is deeply inadequate. There are lots of non–and anti–naturalists working in public universities, and there have been for a long time. The question of whether naturalism (as Wiebe appears to understand it) is true has, for example, exercised philosophers working in such universities for several generations at least, and continues to do so. Were commitment to naturalism an entrance requirement, a kind of quasi–religious test, this would not be the case. I hope that Wiebe does not mean to imply that all those who take naturalism to be false should be purged from our public universities. If he does, then he has a deeply impoverished understanding of what a university, public or otherwise, is; he does not, for example, see that it may also be a place in which those with deeply different philosophical commitments may work with (and against) one another in mutually productive and interesting ways.

The other possibility on this matter is that Wiebe wishes to cleanse the universities of the anti–naturalists because he thinks naturalism is true. But he offers no reasons for thinking it true; and, more damagingly still, no reasons for thinking that judging someone’s views to be false is a sufficient reason for excluding them from the public university. And so he provides no remotely convincing arguments for his view that the study of religion in public universities should proceed as if naturalism were true. Neither does he seem to see that his own uncritical presentation of naturalism as though it were obviously true, or obviously to be adhered to by students of religion in a public university, makes it seem formally and functionally indistinguishable from the confessional commitments he excoriates—though it is of course substantively much less interesting than they.

Parson Thwackum, then, has not been escaped. Those, like Wiebe, who have tried the hardest to escape him have not managed to constitute a discipline because they have not managed to construct what Jonathan Smith so clearly sees the need for: a horizon within which intellectual work can be done. Others, clearer–sighted, see that the idea of religion is so deeply intertwined with a theological and specifically Christian response to modernity that attempts to disentangle it by abstraction will produce a cipher; they therefore recommend its abandonment by those who cannot or will not adopt a theological frame for their work. These are the only two possible options for those who want to renounce theology, and that this is so shows that the scientific study of religion is without a future. This is no news, of course. What is new is that more and more of those who find their academic home in departments of religion and their principal guild affiliation in the American Academy of Religion are happy to say it.

A final question remains. Is there an intellectual future for the idea of religion among those who are happy to return Parson Thwackum’s embrace? Is there, that is to say, an intellectually productive Christian (or Jewish, or Islamic, or Buddhist) use of the term? Can it become native to these traditions of reasoning and practice? I suspect that, in the Christian case, it can; but arguing that position is for another occasion.


Paul J. Griffiths is Professor of the Philosophy of Religions in the Divinity School at the University of Chicago. His most recent book is Religious Reading (Oxford University Press).