Skip to main content
Relating Religion

"Jonathan Z. Smith is one of the two or three most widely read and discussed living theorists of religion. He has been publishing in that field for almost forty years. . . . An excellent sense can be had from this volume—and most especially from its opening autobiographical chapter—of the state of play in the academic study of religion."–Paul J. Griffiths, First Things





Buy this book: Relating Religion
The First Chapter from
Relating Religion
Essays in the Study of Religion
by Jonathan Z. Smith



You and I ought not to die, before We have explained ourselves to each other.
John Quincy Adams to Thomas Jefferson, July 15, 1813.


WHEN MIRCEA ELIADE occasionally recounted the tale of our early meetings together, he always remarked that he had found it “very amusing” that I so often used the phrase, “when the chips are down,” when commenting on one or another’s work. He professed to have not previously encountered it. For my part, I had seen a version of the tag first as the English translation (1948) of the title of J.-P. Sartre’s play, Les Jeux sont faits (1947). Both the French and the English phrases are gaming terms that signal finality. With less urgency, I tended to employ it with the connotation, “when all is said and done,” as, for example, “despite so-and-so’s claims, when the chips are down, he’s a functionalist.”In this essay, I want to turn the phrase on myself and account for my most persistent interests as a student of religion.

After several abortive attempts, I have settled on writing this account in the form of that awkwardly entitled genre, a bio-bibliographical essay.I am enough of a residual New Critic to largely discount the relevance of the bios part of that compound term, but chronology has its conveniences along with its conventional limitations.

I: Student

Growing up in Manhattan in the ’40s and ’50s provided both a foil for my fascination with natural history, especially botany, and a nurturing setting for my initial interests in history and philosophy. Both sets of concerns have strongly influenced the sorts of problems and approaches I have taken when I turned to the study of religion.

My interest in natural history was at once both moral and intellectual. The former goes back to an early acceptance of the categorical imperative, “do no harm.” Whether expressed in public gestures such as vegetarianism, conscientious objection, and passive resistance activities, or in my vocational plans to become an agrostologist, a grass breeder, with the hope of an atoning reclamation of those deserts that were the products of human failures to take care—it seemed clear to me, as a preteen, that those western religious traditions with which I had some superficial acquaintance provided no intellectual resources for such an ethic of “do no harm,” insofar as they appeared to claim that the earth was ours to “subdue” and exploit. Writings from the late 40s and early 50s are filled with my excited reports of readings in Asian traditions, ranging from Buddhism and Jainism to Gandhi, as well as works on native religious traditions.

At the same time, agrostology led, among other things, to a deep interest in taxonomy. My two bedside books during high school were the seventh edition of Asa Gray’s New Manual of Botany and A. S. Hitchcock’s Manual of the Grasses of the United States, soon replaced by the second edition, revised by Agnes Chase. This interest remains today; taxonomic journals are the only biological field I still regularly read in.

Putting these naturalist interests together with a concern for teaching, I started, and maintained for fifteen years, a small trailside museum in Westport, Connecticut. My first appearance in a scholarly publication is a 1954 report of a survey of the average number of milkweed plants (Asclepias sp.) per acre, undertaken in support of the Royal Ontario Museum’s monarch butterfly migration project.

In 1956, I abandoned my long held plans to study agrostology at Cornell Agricultural School. I instead went to Haverford College, where a remarkable philosopher and teacher, Martin Foss, inspired in me such a desire to emulate that I became a philosophy major. Haverford, at that time, still had a palpable connection with its origin as a Quaker school. The pacifism of the Society of Friends, as well as the austerity of their Meetings, was extraordinarily attractive and moved me to begin systematic readings in western religious traditions, working in Haverford’s Rufus Jones collection of Christian mystical literature and studying with Douglas Steere.

These sorts of concerns coexisted with another set of philosophical and historical interests. By 1950, I had already discovered and taken full advantage of that distinctive West Side intellectual and political environment, bounded by Union Square and Brooklyn Heights, with Greenwich Village as its center. Within that universe of discourse, whether one’s interlocutor be an old-time member of the International Jewish Labor Bund or a young Stalinist or Trotskyite, Marx was taken for granted, although what “school” of Marxism remained a matter for fierce debate. While accepting the model of both Marx and Lenin, that Hegel had always to be seriously and critically read, a central division of intellectual labor was the split between those who sought to reconcile Marx and Freud and those who worked on relations between Marx and Kant. In either case, there were strong fissures between the several Freudianisms or neo-Kantianisms. Although the project of reconciliation with Kant interested me more, I happily took part in both projects. I was especially intrigued by the so-called Austrian Marxists (especially Max Adler) who had made an industry of works with titles such as Kant und Marx (K. Vorländer) or Kant und der Marxismus (M. Adler). Their writings were studied in informal translations prepared by those who read German and distributed on purple hectograph or sepia-toned thermofax sheets.

By 1950–52, I had come to the view that language per se was a way of reconceptualizing the issues of Marxism and culture. For most of the anti-Stalinist groups, if the topic of language entered the discussion at all, it did so with reference to George Orwell’s writings on the political manipulation of language—most famously in the twelve-page appendix to 1984 but more forcefully in some of his essays, especially, “Politics and the English Language.”(I do not recall anyone knowing of the journal Politics and Letters, coedited by Raymond Williams, until the 1958 publication of Culture and Society, 1780–1950, when he became a central figure). I found far more interesting one group that spent several months reading Joseph Stalin’s Marxism and Linguistics, which had just become available in English translation. Stalin’s insistence that “a Marxist cannot regard language as a superstructure on the base,” and that language is not constituted by its vocabulary but rather by its grammar, which, like geometry, treats “relations” of entities “in general, without any concreteness,”was a liberation from the mechanics of base/superstructure that dominated most of our discussions of cultural productions. Besides, this seemed to offer the possibility of a connection with Kant’s formalism, which appeared hospitable to thinking about general linguistics.In this connection I regularly quoted Hamann’s remark on Kant, “the question is not ‘What is Reason?’ but ‘What is Language?’. . . What we want is a Grammar of Reason”—a challenge paraphrased by Max Müller as a call for a “Critique of Language” to follow after Kant’s first Critique.

In the discussion of Marx and Kant, Marx and Freud, the under standing of Kant seemed enhanced by an appeal to language as grammar. On the other hand, Freud seemed more focused on a highly specific semantics. This had taught me to be shy of a quest for ‘deeper’ meaning (an abstention that largely remains intact), but the chapter “The Dream Work” in The Interpretation of Dreams was a brilliant syntactical proposal that informed my understanding of the project of relating Freud to Marx.

All of these inchoate musings became clearer when I accidentally came across a copy of a journal with E. Cassirer’s 1945 article, “Structuralism in Modern Linguistics,” in a 10¢ barrel in a Fourth Avenue used bookstore.This introduced me to the names of Saussure, Trubetzkoy and Jakobson. In 1951–52, very little of their work was available in English, but I began to read in linguistic theory, especially L. Bloomfield, Z. Harris, and L. Hjelmslev.

Cassirer’s article did more. He made a crucial analogy between morphology in biology and structuralism in linguistics—both of which would become lifelong preoccupations. (It was as a result of Cassirer that I read, for the first time, Goethe’s Metamorphosis of Plants).

I don’t think I had ever been as impressed with a mind at work as I was with Cassirer’s. As I came to read the bulk of his writings, he persuaded me of five foundational presuppositions.

First, symbols are not expressive, they are a mode of thought.

Second, anthropological thought begins with thinking about language itself, for it is in the linguistic project that we see most clearly the creation of a distinctively human world, our “second environment.” In crude terms, language “creates” the world; it does not merely “reflect” it. The various cultural forms that preoccupied Cassirer are differentiated modes of this linguistic project. To cite the peroration of Cassirer’s Essay on Man:

Language, art, religion, science are various phases of this process. In all of them man discovers and proves a new power—the power to build up a world of his own, an ‘ideal world.’ Philosophy cannot give up its search for a fundamental unity in this ideal world. But it does not confound this unity with simplicity. It does not overlook the tensions and deep conflicts between the various powers of man . . . They tend in different directions and obey different principles.

The point at which I have learned to differ from this paragraph is with respect to its opening sentence, which repeats one of Cassirer’s fundamental presuppositions: “human culture taken as a whole may be described as the process of man’s progressive self-liberation.” I would now reject the implications of the term “progressive” and would rather speak of a human cultural attempt at “liberation” from culturally created, culturally imposed, constraints through efforts at thought.

Third, Cassirer demonstrated that myth could be an object of inquiry that presumed its rationality, rather than denigrating it as irrational (as was common in so many triumphalist histories of philosophy and science) or celebrating it for its irrationality (as was common in so many Romantic works). I would now differ from Cassirer’s grounds for postulating this rationality and seek to extend the logic to other modes of thought and action, especially to ritual, but for me the goal of Cassirer’s project remains intact.

Fourth, Cassirer exhibited ways of thinking philosophically about ethnographies. They did not represent exotic or limit cases, extreme exaggerations of features we recognize in our own culture, but were to be taken as quite ordinary data for interrogation and constructive inquiry.

Finally, Cassirer illustrated by his practice that a prime way of doing philosophy was to think philosophically within the history of ideas, broadly construed, rather than against it (the latter, a mood rampant in the ’50s, largely under British influence). What I learned from Cassirer was an ethic of careful reading, both contextually and critically, but without suspicion.

Beyond these fundamental elements, Cassirer sealed (as Durkheim would later confirm) my allegiance to neo-Kantianism and its relationship to aspects of the Enlightenment project, a conviction that persists to this day.

Cassirer was as well the source of what appears, with hindsight, to have been an enormous detour. When, in the preface to the first volume of the English translation of The Philosophy of Symbolic Forms, he described his project of studying cultural forms as a “morphology.” I was certain I knew what he meant in light of both his 1945 article and my subsequent readings in the Germanic tradition of biological morphology. However, in the prefaces to the second and third volumes of the Philosophy, Cassirer went on to change his terminology from a Goethean one to a Hegelian, now describing his project as a “phenomenology.”

Beginning in 1955 and continuing through my four years at Haverford College, I read and reread Hegel, Husserl, Heidegger, and Sartre, each, always, in relation to Kant and the neo-Kantian tradition, in order to understand Cassirer’s turn. By then, I had shifted from being a biology to a philosophy major and had the advantage of a series of reading courses with Martin Foss on these figures. (I shall never forget Foss sitting, week after week, in an old armchair in his living room, translating, line by line, Heidegger’s Sein und Zeit, while I sat on a footstool furiously scribbling it down. This remains for me the archetypal image of a teacher at work.)

By the summer of 1958, both the readings in western religious and mystical traditions as well as those in phenomenology came together, and I began to outline a project on “the phenomenology of time and the structure of western mystical experience.” I spent the summer writing a dense manuscript intended as a “methodological prolegomenon.” While the attempt is, now, interesting largely for the bibliography, writing it convinced me that this was not a fruitful area for concentration.

Paul Desjardins had joined the philosophy faculty at Haverford and introduced, among other things, a measure of focus on ancient Greek materials (along with careful readings of Kant’s first and third Critiques). His passion was Plato; I found Aristotles’s logical and taxonomic work far more intriguing. (Frank Parker, the fourth member of the department, was primarily interested in Aristotelian and Thomistic logic). Almost by way of a defense, I became fascinated with Francis M. Cornford’s theory of the relationships between myth and philosophy, which, in turn, led to reading that collection of writers inappropriately collected together under the title the Myth-Ritual School. The Cornford thesis appeared to advance, in a historical way, Cassirer’s philosophical interests in taking myth seriously. This led to a decision to focus on the possibility of treating Hesiod as a philosopher by setting his works within the larger framework of what I termed a “phenomenology of myth.” At first, the older project was joined with this new one under the title “The Mythic Representation of Time: A Phenomenological Investigation” (1959), but was later generalized as “A Prolegomenon to a General Phenomenology of Myth,” which was submitted as my senior thesis (1960).

Because I was convinced that myth could not be studied simply on the basis of antiquarian texts but required ethnographic materials as well, I attended during Columbia University’s summer session (1958) a course in cultural anthropology with Morton Fried.The course gave me some beginning measure of confidence in reading anthropology. The most striking feature of the much-revised 1960 “Prolegomenon” is the decline of citations of philosophical and theological writings and the increase in references to specific ethnographies and anthropological theories (including some early writings by Lévi-Strauss), along with an enlarged repertoire of Continental historians of religion. In the “Prolegomenon,” the term “morphology” largely replaces “phenomenology.” “Morphology” had been reencountered in Eliade but was equally resonant with past readings in Cassirer and in works cited by him.

I do not regret the “detour” into phenomenology. I read some wonderful books along the way. (I have read or reread Heidegger each summer for more than forty years). But by the end of 1959 I knew that the Romantic ontology and epistemology which characterized so much of the thought of both the phenomenologists and the historians of religions was profoundly nonanthropological and antihistorical and was, at its base, in curious ways, disturbingly nonrational. Neither Marx nor Kant could be satisfied.

In several papers written in 1959 –60, I sharply criticized what was taken to be ‘phenomenology of religion,’coupled with a long paper in which I argued that Sartre’s Being and Nothingness was a useful model for a properly anthropological phenomenology, even though I ended by confessing that I could not see how to apply it to problems of mythic thought. My work on Hesiod still retained too much phenomenological vocabulary; it was my excited reading of Lévi-Strauss’s “Four Winnebago Myths” that purged that!I rewrote the Hesiod piece several times during 1960–61, striving to make it into a purely structuralist essay as I understood structuralism at that time.

Having finished the B.A. at Haverford, I was not much interested in the analytic tradition that then dominated graduate programs in philosophy. Besides, my interests had clearly shifted to myth. After some uncertainty, I ended up going to Yale Divinity School with the initial intention of working in the New Testament where, thanks to Bultmann and his demythologizing project, I thought myth would be a topic of conversation. (I had seen Brevard S. Childs’s 1955 dissertation, “A Study of Myth in Genesis I-XI,” later reprinted as Myth and Reality in the Old Testament).

As, at that time, the B.D. was the entry to M.A.-Ph.D. studies in religion, a two-year portion of the B.D.’s academic program was worked out for 1960 –62. The required Divinity sequences were the first formal courses I took in religion.

In retrospect I have come to see that “the Protestant seminary curriculum, for all its inadequacies, constituted a general education in theological studies; religious studies, whether at the collegiate or graduate level, has been unable to construct a general education in religion.” Rather, religious studies “made a decision to give up a (limited) coherence for a (limitless) incoherence.”In 1962, I moved to residency in the PhD. Program in the Department of Religion, newly severed from the Divinity School.

Yale was for me both a complex and an exhilarating experience. In many ways, interacting on a daily basis with tribal Protestants was analogous to an anthropologist’s fieldwork.I was, to a considerable degree, a participant-observer, making all the telling mistakes no native would, yet finding their indigenous quotidian practice and speech remarkable. (There was a set of friends who sometimes served as “native informants,” willing to help me out on such awkward occasions as, for example, when I thought that the economic understanding of the Trinity might have something to do with Marx.) Furthermore, my fellow students’ automatic assumption that, being Jewish, I would be as committed to that religious tradition as they were to theirs, prompted a considerable effort at reading in and thinking about Judaism.

Both this undertaking and my studies in the New Testament and early Christianities—I had not read the New Testament, in its entirety, before coming to Yale—led to the question as to whether some of the concerns I had brought to the study of myth and ethnographic sources could be applied as well to Jewish and Christian materials. That is to say, in terminology I began to be comfortable with, was a history of religions approach to these western traditions possible?Could such an approach overcome the exceptionalism with which they were commonly treated? At the same time, could such an approach be responsible to the thick history of scholarship on these traditions, a history whose extent I was beginning to learn?I think it was generally understood that these issues would frame my future work. Largely unnoted by others was the fact that these questions had become more complicated due to a deeply felt tension between the strong focus of Yale’s department on languages and history, which seemed to privilege particularities, and my growing acquaintance with complex statistical modes of comparison associated with Yale’s Human Relations Area Files.In retrospect, this discomfort was the beginning of what would become an interest in religion as a generic category, not limited to any particular tradition or canon.

This was clarified when I shifted from my initial plans to work at a degree in New Testament studies (I had drafted a dissertation that took as its starting point A. Guilding’s work on the relation of the rabbinic lectionary cycle to the Gospel of John). I enrolled, instead, in a field listed in the catalog as a Ph.D. area but having neither faculty nor students at that time; History of Religions. By 1963–64, the problematics of comparison emerged as the place at which I could bring together responsibility to specific languages and histories with those philosophical issues concerning culture, religion, and language that I found most interesting. Having a first-class research library’s stacks to rummage in,I began a program of systematic reading in the history of comparison in the study of religion and in anthropology, abstracting both premodern and modern texts.

At the same time, I began to think about a dissertation that would use Frazer’s The Golden Bough as a “laboratory for comparison.”The problem was the right one; but Frazer, as I already knew while drafting the dissertation, was the wrong choice. In my conclusion to the dissertation, written some six years later, I argued that the problem is neither with Frazer’s data nor with his ever-changing and always weak theories, it is that “Frazer had no method, either explicit or implicit, for his innumerable comparisons.” He offers no answer to the question, how shall we compare?

II: Teacher

My first academic appointment was in the Department of Religion at Dartmouth College, for the year 1965–66, replacing faculty on leave. The teaching assignments were predetermined, courses that had already been scheduled.

That year at Dartmouth sealed my already existing friendship with Jacob Neusner. It was also where I met Hans Penner, who had come that year from the University of Vermont to begin teaching at Dartmouth. Hans was trained in History of Religions at Chicago; our association was the first opportunity I had to talk about the field with an individual rather than with a book. And talk we did . . . for hours each day. Whatever growing sense of security I felt as a neophyte in the history of religions I owed to these conversations.

In 1966, I joined the faculty of the Department of Religious Studies, University of California, Santa Barbara. This was a considered move. I had known its chair, Robert Michaelsen, through membership in the Society for Religion in Higher Education as well as through his writings. I shared his sense that teaching religion in state universities provided a new environment in which to think through our subject matter, and had hoped to have a career in public education.The Department was small (Michaelsen, Walter Capps, Richard Comstock) and vigorous. I was expected to teach the large lecture class on world religions, to give an introductory course to the study of religion, and to offer introductions to the Old Testament, the New Testament, and Judaism.

While writing the bulk of the dissertation at Santa Barbara and reflecting on my growing sense of Frazer’s inability to answer my questions on comparison, I had come to see that the problem needed to be framed differently. Implicit in the comparative project was a prior question, how can a morphological/structural approach be historically and anthropologically responsible? My initial experiments with this issue were the formats of the world religions course and the introduction to the study of religion. I had experienced, as well as taught in, survey courses, and I was convinced that their principles of coverage and of chronological ordering were uninteresting. One damn thing after another simply fails at problematizing the subject matter with respect to any intellectual capacity other than mnemonics. The two devices finally settled on, the exemplum and the test case, not only served as organizing principles for the courses but came to be stratagems I would deploy in most of my subsequent work.

The world religions course was structured around the notion that each of the major religious traditions should be taken as exemplifying particular structures of religious experience, behavior, and expression as identified by students of religion. (For example, Hinduism was paired with temple construction and sacrifice; Islam, with holy book and pilgrimage; Judaism, with holy land and the duality, pure/impure; Christianity, with myth and history and life-cycle rituals). The idea was to first understand the structure by acquiring what I termed a “vocabulary,” a set of cross-cultural examples, quickly read, that gave some sense of a structure’s range and attendant interpretative issues, then to study relevant materials within the particular religious tradition.

The introduction to the study of religion was structured around the notion of test cases, significant texts that raised sets of questions as to both their meaning within the cultures that produced them, and their meaning within the history of scholarship on them.This became a characteristic of much of my later work, a double archaeology of situating a text or artifact both in ‘their’ history and in ‘ours.’ The intellectual challenges of organizing the syllabi for both of these courses, the efforts at choice with respect to materials, gave rise to my careerlong preoccupation with the pedagogy of the introductory course.

When I moved to the University of Chicago in 1968–69, the second course served, through the mid-1970s, at times as an offering in the College (“Basic Problems in the Study of Religion”), at times as the introductory course in the Divinity School to the History of Religions field (“Texts and Contexts in the History of Religions”). Each version had the same topics and primary readings, but had different supplementary bibliographies and final exercises. In 1974, I used some of these class materials as the armature for the Arthur O. Clarke Lectures at Pomona College. These were later rewritten as several of the essays subsequently reprinted in Imagining Religion (1982),hence the reference in its introduction:

The essays collected together in this volume are as well the efforts of a teacher.... Each had its origin in a specific classroom situation as I attempted to describe, through concrete example, the particular angle of vision of the historian of religion. As such, each essay has a double pedagogic intent: to cast light upon a specific religious phenomenon and to do so in such a way that the characteristic preoccupations and strategies of the historian of religion be better revealed.

The more than thirty years I have spent teaching at the University of Chicago has defined my work. With but two exceptions (1966, 1968), everything published has appeared during my tenure there. The University’s commitment to letting a mind go where it will resulted in a diversity of appointments, until the ultimate freedom, granted in 1982, of being without departmental affiliation as the Robert O. Anderson Distinguished Service Professor of the Humanities.

I have no intention, here, of reviewing these years in any chronological fashion. I shall only focus on a few items, postponing until the third part of this essay discussion of those persistent elements in my work which were formulated at Chicago. I have titled this third section “Persistent Preoccupations.” As one who claims to be a generalist, I determined early on, at Chicago, that this entailed doing work in relation to the agenda of others. For this reason, the vast majority of papers, lectures, seminar contributions, and responses (largely unpublished) have been on topics that have been assigned. The question has always been what generalizing or comparative perspective could be brought to issues and to data stipulated as representing their interests. Thus my work represents less of a coherent system than it does a series of continuing foci, hence “Preoccupations.”

It is not without significance that I was recruited by the New Collegiate Division of the College at the University of Chicago. Given my initial choice to work in a public university, I doubt I would have accepted appointment solely in the Divinity School. Charles Long was the voice behind the initiative, which was undertaken by James Redfield. The appointment was to a small, experimental program, the History and Philosophy of Religion, which had as its faculty Charles Long, Marshall Hodgson (who died, tragically, just before my arrival), and Henry Rago (the editor of Poetry). There were no set courses; teaching was in seminars or tutorials.

This said, it would be dissembling not to acknowledge that the Divinity School (to which I was offered a joint appointment in the History of Religions field) was a strong motivation in my decision to relocate. I had first met Mircea Eliade on February 14, 1968, in Santa Barbara, where he had arrived to begin a term as a Visiting Professor, the evening of my return from the interview at Chicago. We struck up an immediate relationship with many hours of conversation, both about the history of religions and about literature. I had known Chuck Long through the Society for Religion in Higher Education since 1965; I had first met Joseph Kitagawa at a conference the year before. While I was not particularly eager to teach at the graduate level, I viewed an opportunity to work with these individuals as equivalent to a postdoctoral education in the history of religions, a field in which I was entirely an autodidact. Certainly the central role of Chicago in the profession, in those years, guaranteed a constant stream of visitors, an extraordinary group of graduate students, and entry into a variety of associations that would not have occurred had I been located elsewhere. By 1971–72, the program in the New Collegiate Division had been suspended, and I shifted my primary appointment, for a time, to the Divinity School, becoming chair of the History of Religions field.

I was restless. My interest was chiefly in college teaching, and I was increasingly being defined as a graduate instructor. In 1972, I received an offer from another university that specifically exempted me from having to participate in their graduate program. I was prepared to leave Chicago and had so informed the president and the dean of the Divinity School, when, fortuitously, a letter arrived at their offices from James Gustafson, who was about to assume an appointment at Chicago as University Professor, inquiring as to his College teaching responsibilities. (There were none.) In a hastily arranged meeting, President Levi and Dean Kitagawa asked me if I would stay at the University and form a College religion program.

I remained and worked on what would become the College’s concentration in Religion and the Humanities—a program whose design both reflected my conception of religious studiesand my growing discontent with the traditional organization of the college major.

The new concentration was approved in 1973, and I began offering seminars in it; I also designing two yearlong sequences, independent of the concentration, that would serve in the College’s general education program.In 1975, I was appointed to one of the first new chairs in the College, as the William Benton Associate Professor (later, Professor) of Religion and the Human Sciences. Apparently I had displayed sufficient political savvy in gaining approval for Religion and the Humanities to be asked to join the College administration, first as Master of the Humanities Collegiate Division, Associate Dean of the graduate Division of the Humanities, and Associate Dean of the College, 1973–77, then as Dean of the Faculty of the College, 1977–82. (I resigned from the Divinity School in 1977.) The decade spent in administration, in effect, provided me with a second career, as I began to speak at educational conferences and write on liberal learning nearly as much as I did on religion.

III: Persistent Preoccupations

Moving to Chicago in the late 60s was to enter into Mircea Eliade’s orbit at the height of his influence. Eliade had been, for me, a model of what it might be to be a historian of religion. He seemed to have read everything and to be able to place the most variegated data within coherent structures. While a graduate student, I had set out to read nearly every work cited by Eliade in his extraordinary bibliographies in Patterns in Comparative Religion, hiring tutors to teach me the requisite languages. These readings constituted my education in the field. I took Eliade to be my master, his power all the more palpable in that we had never met. Once, in 1965, while driving to a conference at Notre Dame, I had stopped off in Chicago and sat in a telephone booth for several hours trying, unsuccessfully, to gain enough courage to call Eliade and ask to meet with him. I doubt that I have ever been as apprehensive as on that first night’s drive, two and a half years later, to meet Eliade at his villa outside of Santa Barbara. But Eliade proved to be a gentle giant, learned and funny, filled with curiosity, a master of exotica as well as a lively purveyor of gossip. He became the most generous of senior colleagues, both supporting my work and including our family within his wide circles of sociability. I count my association with him as one of the great gifts in my life and miss our conversations dearly.

Given this respect and affection, the issue became one of how to separate oneself from aspects of Eliade’s thought without distancing oneself from the man. This dilemma came to speech in a paper delivered in 1971, with Eliade present, at a symposium on his work.

In making these [few critical remarks], I feel acutely the stance of the pygmy standing on the giant’s shoulders but without the attendant claim of having seen further (if one dares to so stretch this tortured phrase, fondly contracted as OTSOG, following Robert K. Merton’s brilliant investigation of its history). The giant, in this case, has taught all of us how and what to see; and far more important, how to understand what we have learned to see.... It is for us, his students, only to bring forth the questions, blurrings, and shadows which result from our more peripheral vision.

Much of my work in the early Chicago years was devoted to finding both constructive alternatives to the points of disagreement with Eliade and the appropriate rhetoric with which to express them.

1. Reversal and Rebellion; Locative and Utopian

In this latter regard, I place particular importance on a paper written for the 1968 annual meeting of the Society for Religion in Higher Education and delivered that August, quite literally while in transit from Santa Barbara to Chicago, “Birth Upside Down or Right Side Up?” (1970), treating the apocryphal traditions concerning Peter’s upside down crucifixion.

From one perspective, the paper is a continuation of a series of explorations of gnostic themes and texts begun in 1962–63.Here, the only novelty was a shift from a genetic interest in gnosticism to one more appropriately morphological: “My understanding of gnosticism is that it is a structural possibility within a number of religious traditions in the hellenistic-Mediterranean world, that it is not a new religion, or a Christian heresy, but rather a structure analogous to mysticism or asceticism.”

The terms I chose to characterize this structure were “reversal” and “rebellion,” especially with respect to archaic traditions, and the examples cited make clear that these elements were in no way limited to the “hellenistic-Mediterranean world.”

From another perspective, the paper represents a series of breaks with previous work. (1) The first departure concerned the scope of the comparative materials, representing fidelity to the Eliadean tradition as I understood it. In “The Garments of Shame” (1966), I had ranged widely over traditions of nudity, and more narrowly, over exorcistic rituals on animal skins, but the examples were confined to either Mediterranean or Christian texts.That is to say, there was a presumption of a genetic relationship guaranteed by spatio-temporal contiguities—a presumption I would later label as that of homology.In “Birth,” I was not so confined. The paper began with an Eliadean move to the “immediate apprehension” of the meaning of a symbol or archetype,asking what the historian of religions could say about Peter’s upside down crucifixion, “What does it mean to be upside down?” and answering, with an appeal to the phenomenological psychiatrist E. W. Straus, “Its most basic sense is to be nonhuman.” It then goes on to give a variety of exempla from western and non-western traditions, from mythic, cultic, and literary texts.

(2) Unlike “Garments,” which set the problem posed by the Gospel of Thomas within the context of a problem in a history of traditions ap-proach to early Christianities, “Birth” placed the problem posed by the

Acts of Peter within a broad, largely anthropological, theoretical context, as represented preeminently by Claude Lévi-Strauss, Mary Douglas, and Victor Turner, but also by E. Durkheim and M. Mauss, R. Hertz, American ethnoscientists, and G. Dumézil.

(3) The paper goes on to state a disagreement with Eliade, one that came to dominate a number of essays written through 1974. After summarizing Eliade on the cosmogonic myth and arguing for its applicability to aspects of ancient Mediterranean cultures (using Cornelius Loew’s summary of the conclusions of the Panbabylonian school),I went on to argue the alternative:

One finds in many archaic cultures a profound faith in the cosmos as ordered in the beginning and a joyous celebration of the primordial act of ordering as well as a deep sense of responsibility for the maintenance of that order through repetition of the myth, through ritual, through norms of conduct, or through taxonomy. But it is equally apparent that in some cultures the structure of order, the gods that won or ordained it, creation itself, are discovered to be evil and oppressive. In such circumstances, one will rebel against the paradigms and seek to reverse their power . . .

Rather than renewing the creation, reestablishing the patterns of destiny, the patterns are seen to be fundamentally perverse.... Reality is discovered to lie not within the cosmos as ordered through creation but above the world, beyond it, and the aim of existence is seen to be to escape the constricted confines of one’s place.

I concluded that what was required was a “phenomenology of rebellion,” adding that “It is my conviction that such a study is long overdue and, in our present situation, might be of more than strictly academic relevance.”

All that was needed was to name this duality. After reviewing a number of options, I settled on ‘locative’ for the first, more Eliadean, pattern in “The Influence of Symbols on Social Change” (1970), and ‘utopian’ for the second, rebellious pattern in “The Wobbling Pivot” (1972), yielding the dichotomy, locative/utopian that I continue to use to the present day.

The culmination of this first effort at distancing was the 1971 conference on Eliade’s work that resulted in “The Wobbling Pivot” (1972).The essay opens with an attempt to summarize, as fairly and accurately as possible, Eliade’s categories of sacred space and sacred time—a necessary prelude to criticism—before turning to the critique, diplomatically expressed in the form of a series of questions. Is chaos best understood as the equivalent of the profane? Has the category of the “Center” been too narrowly discussed in terms of geographical symbolism? Can one pay such attention to the “Center” without giving equal attention to the periphery? Are all mythic first times paradigmatic and to be ritually repeated? What about reversals and rebellions? What about those myths which express a fundamental tension in the cosmos? “Clearly these mythologies, many of which are extremely archaic, point to a different spiritual horizon than that described by Eliade as the fundamental ‘archaic ontology.’” Are the materials Eliade describes best organized under the categories ‘archaic’ and ‘modern?’ Does the dichotomy between mythic-cyclical time and linear-historical time do justice to the rich patterns of temporal significance in various cultures? My attempt, here, was not so much to insist on a negative answer to each question as to propose a coherent alternative.

“Wobbling Pivot” is of interest in one other respect. After introducing for the first time the distinction, locative and utopian, it goes on to caution against imposing on this dichotomy “an explicit evolutionary scheme of development,” as, for example, identifying the locative with primitive, archaic societies and utopian with modern. “Both have been and remain coeval existential possibilities.... While in this culture, at this time or in that place, one or the other view may appear to be the more dominant,” this does not affect the “postulation of the basic availability of both at any time in any place.”This needs to be stressed as the Mediterranean examples I have frequently employed might lead to the perception that the locative, Classical traditions are always prior to the utopian Late Antique materials. (Unfortunately, my own language has, at times, given comfort to such an understanding.)

2. Situation, Incongruity, and Thought

In “Wobbling Pivot,” I observed, almost by way of an aside, that “I have a sense that much will be learned from relating the cosmic views [i.e., locative and utopian] to the social worlds in which they are found.” I clearly had in mind the sort of situating of texts characteristic of my introductory classes. Beginning in 1972, I began to experiment with ways of carrying these efforts over into my writing.In May 1974, when I was asked by the College of the University of Chicago to give an inaugural lecture to acknowledge receiving the William Benton Chair in Religion and the Human Sciences, I used the conventions of such an address both to summarize work already done and to forecast new directions. The lecture, “Map Is Not Territory” (1978), both bid farewell to my double project of appreciation and criticism of Eliade, and signalled a different tack. Rather than rebellion and reversal, the point of departure for understanding myth and ritual in a non-replicatory sense would be “incongruity” as experienced within quite particular “situations,” with the premise that it is “the perception of incongruity that gives rise to thought.”

The lecture begins with a rehearsal of the locative cosmology, now termed a “map,” but with a twist.The argument is no longer one of accepting the pattern but juxtaposing it to another; rather, it has become one of suspicion with respect to the pattern in terms of its social situation, both within archaic cultures and within contemporary scholarship:

I would term this cosmology a locative map of the world and the organizer of such a world, an imperial figure. It is a map of the world which guarantees meaning and value through structures of congruity and conformity.

Students of religion have been most successful in describing and interpreting this locative, imperial map of the world— especially within archaic, urban cultures.... Yet, the very success of these topographies should be a signal for caution. For they are largely based on documents from urban, agricultural, hierarchical cultures. The most persuasive witnesses to a locative, imperial world-view are the production of well organized, self-conscious scribal elites who had a deep vested interest in restricting mobility and valuing place. The texts are, by and large, the production of temples and royal courts and provide their raison d’être . . . In most cases one cannot escape the suspicion that, in the locative map of the world, we are encountering a self-serving ideology which ought not to be generalized into the universal pattern of religious experience and expression.

The singular “universal pattern” was intended to refer to Eliade’s “archaic ontology.” But the critique was expanded: “I find the same conservative, ideological element strongly to the fore in a variety of approaches to religion which lay prime emphasis upon congruency and conformity, whether it be expressed through phenomenological descriptions of repetition [or] functionalist descriptions of feedback mechanisms... .”

With this said, the lecture moved on to the chief implication of the criticism, not so much in terms of its ideological situation but, rather, in terms of its high valuation of congruency leading to a denial of the significance of efforts at thought and of intellectual criticism, especially as projected upon so-called primitives. Using the perception of incongruency as the point of contention, the lecture went on to argue against the consequences of such a view:

On the conceptual level it robs them of their humanity, of those perceptions of discrepancy and discord which give rise to the symbolic project that we identify as the very essence of being human. It reduces the primitive to the level of fantasy where experience plays no role in challenging belief . . . where discrepancy does not give rise to thought but rather is thought away.

One implication of this emphasis on the relationship between discrepancy and thought was that, rather than according to Romantic theories of pristine myth,myth was to be understood as a thoughtful “strategy for dealing with a situation.”This was illustrated, in the course of the lecture, by an Oceanic punning account of origins to account for a newly introduced domesticated animal; African divination; the juxtaposition between expectation and reality in Australian and African initiatory scenarios, and a lengthy discussion of the Ceramese myth of Hainuwele.The Late Antique materials previously used to establish the utopian themes of rebellion and reversal were briefly mentioned but not discussed.

The lecture had two sorts of conclusions. One reflected back on the distinction of locative and utopian which had previously preoccupied me, and suggested that these were but two “maps” among many. The lecture had sketched out one more of these maps: traditions which “are more closely akin to the joke in that they neither deny nor flee from disjunction, but allow the incongruous elements to stand”; traditions that seek “to play between the incongruities, and to provide an occasion for thought.” The other conclusion was offered with respect to the Hainuwele myth, but intended to be generalized:

The Ceramese myth of Hainuwele . . . does not solve the dilemma, overcome the incongruity or resolve the tension. Rather it provides the native with an occasion for thought. It is a testing of the adequacy and applicability of native categories to new situations and data. As such, it is preeminently a rational and rationalizing enterprise, an instance of an experimental method. The experiment was a failure. The white man was not brought into conformity with native categories, he still fails to recognize a moral claim of reciprocity. But this is not how we judge the suc cess of a science. We judge harshly those who have abandoned the novel and the incongruous to a realm outside of the confines of understanding and we value those who (even though failing) stubbornly make the attempt at achieving intelligibility, who have chosen the long, hard road of understanding.

When I identify myself as having an intellectualist understanding of religion, it is the sense suggested by this quotation.

Several of the essays in Imagining Religion were devoted to the elaboration of these themes: in situations of intercultural conflict (“A Pearl of Great Price and a Cargo of Yams” [1976]; “The Unknown God” [1982]); in situations of intracultural discrepancies (“The Bare Facts of Ritual” [1980]);as well as in scholarly discourse, when it classifies some phenomenon as unique and, therefore, unintelligible (“The Devil in Mr. Jones” [1982]).

The most sustained investigation of situational incongruity was To Take Place: Toward Theory in Ritual (1987), based on the Merril L. Hassenfeld Memorial Lectures delivered at Brown University in 1985.Here the issue was the displacement of two sets of systemic rituals, one Judaean, the other Christian, that had been carefully crafted for particular loci in Jerusalem, the Temple and the Church of the Holy Sepulchre. Each group, through the contingencies of history, lost access to the place. For each, the rituals had to be replaced by new intellectual constructions: the spatial systematics of pure/impure, as represented by texts such as Mishnah, and the systematics of temporality, as represented by the liturgical calendar.

3. Taxonomy and Comparison

My early interest in botany and fascination with taxonomy, especially of the Gramineae, led directly to a concern with comparison. I spent hours puzzling through the implications of phrases such as “a natural classification of plants is one in which the different kinds of species are arranged in groups according to their resemblances, as shown by their structure”; “ideas of the relations of groups to each other are largely inferences based upon morphological resemblances”; “those individuals which are so much alike as to appear to be of one kind . . . are regarded as belonging to the same species”; “the question always arises whether there are several closely related but distinct species or a few distinct species each of which shows great variation,” leading the author to distinguish between “well-marked varieties,” “less well-marked varieties,” and “additional forms.”

In the summer of 1952, one of the more ambitious exhibits I mounted in my small trailside museum was entitled Same, Like, Different, attempting to illustrate the issues attendant on taxonomy with examples of common wild plants. Put simply, taxonomy seemed a comparative enterprise which sought similarity across obvious individual variations and which asserted significant difference even in the face of apparent resemblances. Some thirty years later, this early perception was echoed in statements such as

It is axiomatic that comparison is never a matter of identity. Comparison requires the acceptance of difference as the grounds of its being interesting, and a methodical manipulation of that difference to achieve some stated cognitive end. The questions of comparison are questions of judgment with respect to difference: What differences are to be maintained in the interests of comparative inquiry? What differences can be defensibly relaxed and relativized in light of the intellectual tasks at hand?

By the time, as a graduate student at Yale, I began to take courses in anthropology, I was well prepared for their preoccupation with kinship. Kinship systems, after all, are one of humankind’s more elaborate taxonomic constructions, and the comparative study of kinship systems by anthropologists have yielded some of their most impressive and successful efforts at classification. As with any young student of Marx and Engels, I was led, by them, to Lewis H. Morgan’s pioneering studies, first, Ancient Society, and then, Systems of Consanguinity and Affinity of the Human Family. I had found my way, later, to Frazer, Durkheim, and Robertson Smith on totemism, and, by college, to Durkheim and Mauss on classification, as well as some of Lévi-Strauss’s early writings on kinship. As I worked on a structural study of Hesiod’s genealogies, I began systematic readings in the literature, published since the mid-1940’s, on kinship in Native American, African, Australian and other Oceanic societies. At Yale, it was the controversies over the various formalizations of these systems that most interested me, an interest reawakened when the longawaited second edition of Lévi-Strauss’s Elementary Structures of Kinship appeared. Throughout this period, I continued, as well, to read in the theoretical literature on biological taxonomy and in journals devoted to botanical classification.

Given this history, I have always viewed comparison and classification as inseparable. In my published work, the one topic always entails the other. There are a set of writings more focused on taxonomic issues, often

relying on biological systematics: “Animals and Plants in Myth and Legend” (1974), “Fences and Neighbors” (1980), “What a Difference a Difference Makes” (1985), and “Classification” (2000). One subset concerns the problematic taxon, “world religions”: “Map Is Not Territory” (1978), “A Matter of Class” (1996), “Religion, Religions, Religious” (1998), “Classification” (2000); and another concerns native taxonomic categories: “Earth and Gods” (1969), “Birth Upside Down or Rightside Up?” (1970), “Towards Interpreting Demonic Powers” (1978), To Take Place, esp. chapter 3 (1987), “Differential Equations” (1991), “Wisdom’s Place” (1995), “Close Encounters of Diverse Kinds” (2001).There are, as well, a group of programmatic essays more focused on comparison, which includes “Adde Parvum” (1971), “In Comparison a Magic Dwells” (1982), Drudgery Divine (1990), “Epilogue: The ‘End’ of Comparison” (2000), and “Acknowledgments: Morphology and History,” parts 1–2 (2000).

When first working on the problem of comparison, my chief concern was with the possibility of a rigorous method that would exclude what seemed all too typical of the practice of many scholars: x (which they happened to have noticed) reminds them of y (which they happen to have already known), therefore x must in some way be like y. The initial enthusiasm for the statistical procedures associated with George P. Murdock and the Human Relations Area Files was based on the hope that it offered such rigor. In fact, it was procedurally flawed, most tellingly with respect to its “sample.” This said, the project’s summary conclusion remains provocative, more than twenty thousand statistically significant correlations across a sample of sixty societies. By contrast, most students of religion, if they are interested in comparison at all, treat, at best, a few dozen.

Then, too, so many comparisons in the study of religion were between single traits despite the fact that they were embedded in complex phenomena. Already in my historical studies (“The Garments of Shame” [1966]; “The Prayer of Joseph” [1968]) I had insisted on comparing a cluster of elements. But, even as I wrote these pieces Lévi-Strauss had convinced me of their inadequacy, that it is not the elements but the relations between the elements that are significant.I worked through a good part of the ’70s convinced that both the problems with comparison and, to a lesser degree, the remedies were correct: rigor through something like statistical procedures was desirable; multiple characteristics were crucial.

Renewed interest in these matters was prompted by an invitation to give one of the plenary addresses in honor of the centennial of the Society of Biblical Literature at the society’s 1978 annual meeting. The hope was expressed that I might say something about the definitional problems associated with a field of increasing importance and activity but of unsatisfactory nomenclature—denoted by a variety of titles including intertestamental Judaism, postbiblical Judaism, early Judaism, late Judaism, hellenistic Judaism—such as would be suitable for the projected centennial volume, Early Judaism and Its Modern Interpreters, to be edited by Robert A. Kraft and G. W. E. Nickelsburg. I wrote several drafts of a lecture entitled “To Draw the Line” in response to this assignment, but it soon became clear that the problem was not confined to the anticipated one of the biases and interests reflected in the diversity of names. Of more gravity was a set of theoretical issues related to classification, comparison, and by implication, definition. About the same time (1975–76), in preparation for the possibility of reoffering the Problems in the History of Comparison course, I had been reading through a set of newer writings in phenetic (cladistic) and numerical classification which challenged the evolutionary concerns of more traditional biological systematics, insisting on taking all characteristics into account without regard for phylogenetic history. I was intrigued by echoes of Kant’s early distinction between complex contemporaneous similarities (Naturbeschreibung) and similarities that resulted from historical and genealogical relations (Naturgeschichte), which I had taught in the “Problems” course, with special reference to the invention of the category of ‘race.’

In the centennial lecture, “Fences and Neighbors” (1980), I summarized only the new numerical taxonomic proposals as representing a “selfconsciously polythetic mode of classification which surrendered the idea of perfect, unique, single differentia—a taxonomy which retained the notion of necessary but abandoned the notion of sufficient criteria for admission to a class.”Comparison would be based on a multiplicity of traits, not all of which might be possessed by any individual member of the class. I then went on to describe two “experiments” toward a future polythetic description of the taxon, Judaism. The first mapped the “range” of one taxic indicator, circumcision; the second, the variety of indicators of Jewish identity gleaned from a corpus of funerary inscriptions. The conclusion was programmatic:

What has animated these reflections and explorations is the conviction that students of religion need to abandon the notion of “essence,” of a unique differentium for early Judaism.... The cartography appears far messier. We need to map the variety of Judaisms, each of which appears as a shifting cluster of characteristics which vary over time.

As the anthropologist has begun to abandon a functionalist view of culture as a well-articulated, highly integrated mechanism . . . so we in religious studies must set about an analogous dismantling of the old theological and imperialistic impulses toward totalization, unification, and integration. The labor at achieving the goal of a polythetic classification of Judaisms, rather than a monothetic definition of early Judaism, is but a preliminary step toward this end.

I returned to the problem of singularity in comparisons ten years later, in the 1988 Louis H. Jordan Lectures, gaining an assist, this time, from philosophical resemblance theory rather than from biology. Here the suggestion was more formal, that rather than considering a comparative statement as being dyadic, ‘x resembles y,’ it should be thought of as, at least, triadic, as a “multiterm” expression such as ‘x resembles y more than z with respect to ... ,’ or ‘x resembles y more than w resembles z with respect to... .’

In 1979, I brought together reflections growing out of these twin concerns with rigor and singularity in response to an invitation from Jacob Neusner to deliver an opening paper to the newly reconstituted history of Judaism section of the American Academy of Religion. The paper, “In Comparison a Magic Dwells” (1982),was less important in itself than in the response it provoked by the distinguished anthropologist and specialist in Papua New Guinea, Fitz John Porter Poole. His paper, presented to a subsequent meeting of the same section of the Academy, remains, for me, the most suggestive treatment of comparison of the past two decades. The sentence in it that interested me most reads, “Comparison does not deal with phenomena in toto or in the round, but only with an aspectual characteristic of them.”This was a far better way of framing the issue than my usual language of suspicion concerning singularity and essence. It connected well with earlier criticisms of the holistic assumptions of conservative ideologies in the study of religion,and the term “aspectual” had, for me, a series of implications. I was able to take up some of these in a paper for the 1988 annual meeting of the American Society for the Study of Religion on the assigned, somewhat oddly phrased, theme, “What Makes Some Categories Better for Comparison Than Others?” This paper served as a partial rough draft for the second chapter of Drudgery Divine (1990), “On Comparison.”

“On Comparison” sketches out the lineaments of what remains my understanding of the comparative enterprise. Comparisons are not given; they are the result of thought. Comparison in the service of disciplined inquiry

brings differences together within the space of the scholar’s mind for the scholar’s own intellectual reasons . . . [C]omparison does not necessarily tell us how things ‘are’ . . . like models and metaphors, comparison tell us how things might be conceived, how they might be ‘redescribed’ . . . [Comparison] lifts out and strongly marks certain features within difference as being of possible intellectual significance, expressed in the rhetoric of their being ‘like’ in some stipulated fashion. Comparison provides the means by which we ‘re-vision’ phenomena as our data in order to solve our theoretical problems.... Comparison, as seen from such a view, is an active, at times even a playful, enterprise of deconstruction and reconstitution which, kaleidoscopelike, gives the scholar a shifting set of characteristics with which to negotiate the relations between his or her theoretical interests and data stipulated as exemplary.

In a passage such as this, the notion of the “aspectual” has been extended. It refers not only to the data but to the scholar’s intellectual interests as well.

In the course of this discussion in “On Comparison,” there is a subordinate clause, “recalling . . . that a model is useful precisely when it is different from that to which it is being applied.”At the time, I thought chiefly of the map metaphor and of the notion of “defamiliarization,” introduced by Soviet ‘formalist’ literary critics, which I thought to be one of the important consequences of the juxtaposition inherent in comparison.Later, in several of the essays reprinted in this volume, the notion of an intellectual “object” being required to be different from the phenomenon in the interest of cognitive gain came to be a central element in thinking about the study of religion.(See section 5 below).

In Drudgery Divine, these latter implications were not taken up. Rather, I revisted and expanded Richard Owen’s influential eighteenthcentury distinction between homology and analogy to make a parallel point.For Owen, homology, resemblances explained by common descent, were “real.” That is to say, they were the sorts of genealogical comparisons favored by historians in order to demonstrate filiation, contact, diffusion. Analogies, by contrast, are “ideal.” That is to say, they are mental constructions, they rest on postulated relations stipulated with respect to particular points of interest. In Drudgery, I cited, with approval, J. S. Mill’s dictum, “if we have the slightest reason to suppose any real connection between . . . A and B; the argument is no longer one of analogy.”As already observed, while earlier essays such as “The Prayer of Joseph” (1968) were homological in intent, I now insisted that analogy was the proper paradigm for comparison.

Both To Take Place (1987) and Drudgery Divine (1990) represented ambitious attempts at comparison. To Take Place had as its major interest a comparison between the systemic constructions of rabbinic Judaism and post-Constantianian Christianities, between the sort of thoughtful endeavor represented by Mishnah and by the elaboration of the Christian liturgical calendar. In its initial discussion of the Tjilpa pole, it addressed comparison in relation to Eliade’s taxonomic category of the “Center.” Drudgery Divine is a thick critical description of the history of comparisons between earliest Christianities and the religions of Late Antiquity (especially the so-called “mysteries”), on a variety of topics. Drudgery goes on to sketch out a new comparison through redeploying the distinction of locative and utopian.

The issue of the manipulation of difference attendant on both comparison and classification led to a new focus on difference as the term under which I could bring together older vocabulary of ‘reversal,’ ‘rebellion,’ ‘incongruity,’ and ‘gap’ and older interests in the histories of comparison in anthropology and the study of religion with newer historical and theoretical concerns that have become characteristic of my work over the last decade. These are represented by a number of the essays reprinted in this collection: the relations of culture and difference; the ‘discovery’ of the Americas as the site where difference became an urgent intellectual problem, and its relations to the invention of the category of race; ‘redescription’ as the replacement term for older usages such as ‘map,’ ‘model,’ ‘paradigm,’ and the ways in which a set of intellectual operations, ranging from definition to explanation, may be understood as modes of redescription.

4. Culture, Difference, and Thought

In 1976, I was invited to give a presentation for a theological series sponsored by the United Methodist Foundation at the University of Chicago on the assigned topic, “The Persistence of the Sacred.” As formulated, it seemed to me to be but a variant of the tired theme, humans have always been religious. After an initial interrogation of the title—a characteristic move in many lectures—I inverted the title to read “Sacred Persistence”and chose the category of canon as its exemplification.In the lecture, by way of preparation for the discussion of the latter, an illustration was introduced calling attention to three interrelated observations: the plenum of possible natural foodstuffs available to anyculture; the all but arbitrary limitation of what any culture will, in fact, eat; and the fact that this limitation is, then, overcome by the variety of the foodstuffs’ preparation. “If food is a phenomenon characterized by limitation, cuisine is a phenomenon characterized by variegation.” While this was employed in “Sacred Persistence” primarily as an analogy to canon, it contains, as well, a more “general model of cultural activity” and, through the use of the term ‘ingenuity,’ a suggestion as to the relation of this cultural dynamic to thought.

An almost limitless horizon of possibilities that are at hand (in nature) is arbitrarily reduced (by culture) to a set of basic elements.... This initial arbitrariness is, at times, overcome by secondary explanations which attempt to account for the reduction . . . Then a most intense ingenuity is exercised to overcome the reduction . . . to introduce interest and variety. This ingenuity is usually accompanied by a complex set of rules.

That is to say, the dynamic of reduction and elaboration connects with earlier formulations on situation, incongruity, and thought (see section 2 above) but adds the notion that what was there called a “situation” could be internally generated as well as externally imposed and is a regular rather than an unanticipated occurrence. The response, a thoughtful process, was in earlier essays called “application,” “experimentation,” “rationalizing,” “rationalization, accommodation and adjustment,” or “casuistry.” Here, it is called “ingenuity.”There is parity, as well, between the food/cuisine analogy and the earlier description of myth as a “limited collection of elements with a fixed range of cultural meanings which are applied, thought with, worked with, experimented with in particular situations.”

The most explicit deployment of this model is the Arizona State University lecture in religion, “Differential Equations: On Constructing the ‘Other’” ([1992], reprinted in this volume), where I wrote that, beyond the “ubiquity of the construction of difference in human culture,” it would seem that:

culture itself is constituted by the double process of both making differences and relativizing those very same distinctions. One of our fundamental social projects appears to be our collective capacity to think of, and to think away, the differences we create.

Earlier, in “Adde Parvum” (1978), I had generated a rudimentary classification of the sorts of constructed human differences, building on Robert Redfield’s discussion of the “primitive world view” as consisting of a pair of binary oppositions, often correlated, man/not-man and we/they. I proposed “four specifications of the we/they duality . . . (1) They are like-us, (2) They are not-like-us, (3) they are too-much-like-us (Robert Frost’s “Good fences make good neighbors”), or (4) we are notlike-them (expressions and polemics concerning ‘uniqueness’).”It is the third formulation that continued to fascinate me, for it is often the one that is most provocative of thought. In “What a Difference a Difference Makes” (1985) I termed this the problem of the “proximate other,” arguing that ‘otherness’

is a matter of relative rather than absolute difference. Difference is not a matter of comparison between entities judged to be equivalent, rather difference most frequently entails a hierarchy of prestige and ranking. Such distinctions are found to be drawn most sharply between “near neighbors” . . . This is the case because “otherness” is a relativistic category inasmuch as it is, necessarily, a term of interaction. A “theory of otherness,” is, from this perspective, essentially political.

[W]e know of thousands of societies and world views which are “different,” but in most cases, their “remoteness” guarantees our indifference. By and large, Christians and Jews qua Christians and Jews have not thought about the “otherness” of the Kwakiutl or, for that matter, of the Taoist. The bulk of Christian theological thinking about “otherness” (starting with Paul) has been directed toward “other Christians” and, more occasionally, towards those groups thought of as “near-Christians,” preeminently Jews and Muslims. Today, as in the past, the history of religious conflicts, of religious perceptions of “otherness” is largely intraspecific: Buddhists to Buddhists, Christians to Christians, Muslims to Muslims, Jews to Jews. The only major exceptions occur in those theoretically unrevealing but historically common moments when “proximity” becomes more a matter of territoriality than of thought.

I have persistently returned to this theme of the “proximate other” in later works,but it is in an unpublished manuscript (dated 1999) that an explicit connection was made with the 1976 food/cuisine analogy. I appealed to the discriminations of connoisseurship. There is no interest, for example, in distinguishing between red and white wine. They are sheerly different; nothing more needs be said. All the efforts at thought are directed toward distinguishing between two examples of the same varietal from different vineyards or between different years (vintages) of same vineyards’s production. These differences among items that, to a lay palate, appear to be the same are the differences that count. As I argued in “Differential Equations” (1992), “the deepest intellectual issues are not based upon perceptions of alterity, but, rather, of similarity, at times, even, of identity.”

This is not to deny that there are occasions when a difference that is unrelated to the proximate other appears as a severe cognitive shock, when the question of placing the different becomes a pressing one because it calls into question prior anthropological and cosmological assumptions. I have long suggested that, in western discourse, such an occasion was the ‘discovery’ of the Americas and the unanticipated encounter with the Native Americans. It was this occasion that gave rise to the issues as to ‘human nature’ that conferred urgency on the human sciences. It led, as well, to the intellectual problematics associated with the invention of the category of ‘race.’ Although I have published several preliminary studies, reprinted in this volume, the full historical arguments for this case, as well as the implications for anthropological theory, remain a work still in progress (see below, note 95).

5. Redescription, Translation and Generalization

In 1995–96, two events occurred that pushed me to articulate a group of previously inchoate notions that had already begun to influence my way of speaking and writing in a number of unpublished lectures and seminar papers as well as in the college classroom. The first was the opening session of the Ancient Myths and Modern Theories of Christian Origins Consultation at the 1995 annual meeting of the Society of Biblical Literature.The second was an invitation to give a keynote address to a 1996 conference of largely younger scholars, “Reconstructing a History of Religions: Problems and Possibilities,” at Western Maryland College, in which my role appeared to be that of a ghost of Christmas past. The first event was as significant for Burton Mack’s paper, which undertook in part to summarize aspects of my work with a clarity and conciseness I had not achieved, as it was for my own intervention. The second event, at which I was asked to speak on the question, “Why Imagine Religion?” I took as if it were a final testament.Both events led to papers with a new interlocking set of elements and formulations that have continued to reappear in subsequent work, framing my discussions of a wide variety of topics.

I first introduced the term ‘redescription’ in “Sacred Persistence” (1979), drawing on Max Black’s and Mary Hesse’s comparisons of metaphors and scientific models on the grounds that

both invite us to construe one thing in terms of another (most usually that which is problematic in terms of that which is relatively better understood) so that we may see things in a new, and frequently unexpected, light. A model, in short, is a “redescription.”

As I used the term, it expressed a central goal, the redescription of classical categories in the study of religion to the end that these be ‘rectified.’Thus “Sacred Persistence” was subtitled “Toward a Redescription of Canon.”As that article made plain, the route to redescription and rectification is comparison, the latter predicated upon careful description. Burton Mack first formulated these “four operations” in my work in his contribution to the Consultation.In a later restatement in 2000, I explained the relations between these operations as follows:

The ‘end’ of comparison cannot be the act of comparison itself. I would distinguish four moments in the comparative enterprise: description, comparison, redescription and rectification. Description is a double process which comprises the historical or anthropological dimensions of the work: First, the requirement that we locate a given example within the rich texture of its social, historical, and cultural environments that invest it with its local significance. The second task of description is that of reception history, a careful account of how our second-order scholarly tradition has intersected with the exemplum. That is to say, we need to describe how the datum has become accepted as significant for the purpose of argument. Only when such a double contextualization is completed does one move on to the description of [at least] a second example undertaken in the same double fashion. With at least two exempla in view, we are prepared to undertake their comparison both in terms of aspects and relations held to be significant, and with respect to some category, question, theory, or model of interest to us. The aim of such a comparison is the redescription of the exempla (each in light of the other) and a rectification of the academic categories in relation to which they have been imagined.

Because thought would be impossible without comparison, we could not stop comparing even if we wished to. The task, then, for those committed to the comparativist enterprise becomes one of clarifying our assumptions, rectifying our procedures, and justifying our goals.

In both my “Response” at the 1995 Consultation and in “Why Imagine Religion?” a new topic is advanced, partly with respect to the comparativist agendum, but extends beyond this role. I had long been fascinated with translation theories, accumulating a bulging file, offering a college seminar on the subject in the late 70s, and introducing translation as a pedagogical device for teaching Durkheim’s project in Elementary Forms to college students in the early 80s. Now it was expanded to become a key notion in relation to that most challenging goal of the study of religion: explanation. The same sorts of arguments and, indeed, similar phraseology recur in both papers.

Both explanations and interpretations are occasioned by surprise.... Surprise, in both the natural and the human sciences, is reduced by bringing the unknown into relations to the known— relations of similarity and difference, relations of metonymy and metaphor. The process by which this is accomplished, again in both the natural and the human sciences, is translation. That is to say, the proposal that the theoretical second-order language appropriate to one domain (the known/the familiar) may translate the theoretical second-order language appropriate to another domain (the unknown/the unfamiliar). Perhaps the strongest example of this procedure in the study of religion is Durkheim’s translation in Elementary Forms of the language appropriate to “religion” (for him, in this work, the unknown) into the language appropriate to “society” (for him, in this work, the known). The point at which one might differ from Durkheim’s project is with respect to his acceptance of a goal of explanatory simplicity. Better, here, Lévi-Strauss’s formulation: “scientific explanation consists not in a movement from the complex to the simple but in the substitution of a more intelligible complexity for another which is less.”

The argument goes on to offer one consequence of such an understanding:

Too much work by too many scholars of religion takes the form of a description or paraphrase—particularly weak modes of translation, insufficiently different from their subject matter for purposes of thought. To summarize: a theory, a model, a conceptual category cannot be simply the data writ large.

I need only add that the preference for description and paraphrase is one more instance of the conservative interest in totality discussed above in connection with the notion of ‘aspectual’ (see part 3 above). Translation, by contrast, is necessarily incomplete. “Whether of a conceptual or a natural language, whether intercultural or intracultural, translation can never be fully adequate, it can never be total. There is always discrepancy. If there is not, then one is not translating but rather speaking the other language.”One cannot escape the suspicion that it is precisely this latter possibility (speaking the other language) that defines the goal of many students of religion. It is one means for avoiding the consequences of the necessary inadequacy of translation, submission to the “double requirement of comparison and criticism.”

This necessary incompleteness of translation means, as well, that it is corrigible. This interest in corrigibility leads to the last new term introduced into my work in the mid-1990s, that of ‘generalization.’The ‘general’ differs from the ‘universal’ in that it admits to exceptions; it differs from the ‘particular’ in that it is highly selective. Both characteristics guarantee that generalizations are always corrigible. As with comparison, generalization brings disparate phenomena together in the space of the scholar’s intellect. As such, it often results in surprise, which calls forth efforts at explanation and rectification. One goal of the study of religion is the proposal of comparative generalizations based on a careful description of data that, nevertheless, remain firmly situated; generalizations that are advanced in the service of some stated intellectual task.

With the introduction of these newer notions of redescription, translation, and generalization, coupled with a more complex view of difference and its relation to thought, the sorts of problems I first associated with issues of taxonomy and then developed in my early essays on reversal, rebellion, and situational incongruity come together. The efforts at thinking about the status of theoretical constructs in the study of religion converge with interests in the data from religion and join with the long-standing critique of those scholars who would emphasize totality and congruency when describing religion. What these persistent preoccupations have in common is the insistence on the cognitive power of distortion, along with the concomitant choice of the map over the territory.

I have attempted in this essay to provide a sense of both the career and the preoccupations of an individual who has been as much a student of the study of religion as a student of religion. I have sought to convey, through all the movements and experimentations, something of what has been constant, the argument that religion is not best understood as a disclosure that gives rise to a particular mode of experience. To the contrary, religion is the relentlessly human activity of thinking through a ‘situation,’ an understanding that requires assenting to Lévi-Strauss’s dictum, “man has always been thinking equally well.”

Such a view carries, as well, as another consequence, the refusal to accept as an epitaph that phrase from one of Rudyard Kipling’s Plain Tales, “Theory killed him dead.” I am not so compulsive an academician as to claim that theory revivifies one or revives one’s chosen field. Theory is not life, but I know with perfect surety that it is liveliness.

Jonathan Z. Smith
© 2004, 424 pages
Paper $37.00 ISBN: 9780226763873 Cloth $99.00 ISBN: 9780226763866

Be the first to know

Get the latest updates on new releases, special offers, and media highlights when you subscribe to our email lists!

Sign up here for updates about the Press