Edited by
EDITORIAL    ADVISORY    BOARD
Donald S. Lopez, Jr., Tomoko Masuzawa, Jack Miles, Jonathan Z. Smith, David Tracy, Edirh Wyschogrod
The University of Chicago Press Chicago & London
G k e g o R"r Schopen
Duval, Yvetre. 1988. Aiiprés des saints corps et Arne: L'inhwnmtion "ad sanctos" dans la chrétienté d'oiient ct d'occidmt du Illcttu Vile slide. Paris: Etudes Augustinlennes.
Duval, Yvette, and J.-Ch. Picard, eds. 1986. U Inhumation privilegiee du IVe au Vllle sleek en Occident. Paris: De Boccard.
Eire, Carlos M. N. 198(5. War against the Idols: The Reformation of Worship from Erasmus to Calvin. Cambridge: Cambridge Universit;' Press.
Trend, William H. C. 1996. Tise Archaeology of Early Christianity: A History. Minneapolis: Fortress Press.
Geary, Patrick J. 1994. Living with the Dead in the Middle Ages. Ithaca: Cornell University Press.
Geiger, Wilhelm, trans. 1953. Cülavamsa: Being the More Recent Part of the Mahavamsa. Colombo: Ceylon Government Information Department.
Gilman, W. H., ed. 1965. Selected Writings of Ralph Waldo Emerson. New York: Penguin Group.
Harries, Jill. 1992. "Death and the Dead in the Late Roman West." In Death in Towns: Urban Responses to the Dying and the Dead, edited by Steven Bassett. London: Leicester University Press.
Has rings, James. 1908-21. Encyclopaedia of'Religion and Ethics. Edinburgh: Clark.
Hillgarth, J. N. 19S6. Christianity and Paganism, 350-750: The Conversion of Western Europe. Philadelphia: University of Pennsylvania Press.
Lacy, John A., trans. 1955. The Care to Be Taken for the Dead. Vol. 27 of The Fathers of the Church, edited by Roy J. Deferrari. New York: Fathers of the Church.
MacGrcgor, Geddes, 1992. Images of Afterlife: Beliefs from Antiquity to Modern Times. New York: Paragon House.
McCulloh, John M. 1976. "The Cult of Relics in the Letters and 'Dialogues' of Pope Gregory the Great: A Lexicographical Study." Traditio 32.
Mulders, L, and R. Demeulenaere, eds. 1985. VictHcii Rotomagensis: De Laude Sanctorum. In Corpus Christianorum. Series Latina 64. Tumhout: Brepols.
Rhys Davids, T. W., and C. A. F. Rhys Davids. 1910. Dialogues of the Buddha, part 2. London: FL Frowde.
Ricciotti, Giuseppe. 1959. Julian the Apostate, translated by M. Joseph Costelloc. Milwaukee: Bruce.
Scans brick, J. J. 1984. The Reformation and the English People. Oxford: Basil B lackwell.
Schopen, Gregory. 1994. "Stüpa and Tírtha: Tibetan Mortuary Practices and an Unrecognized Form of Burial Ad Sanctos at Buddhist Sites in India." In The Buddhist Forum. Vol. 3,1991-1993, edited by Tadensz Skorupski. London: School of Oriental and African Studies.
---------. 1996. "The Suppression of Nuns and the Ritual Murder of Their Special Dead in Two Buddhist Monastic Texts." Journal of Indian Philosophy 24.
---------. 1997. Bones, Stones, and Buddhist Monks: Collected Papers on the Archaeology, Epigraphy, and
Texts of Monastic Buddhism in India. Honolulu: University of Hawaii Press,
Suzuki, D. T-, ed. 1955-61. Tlie Tibetan Tripitaka. Peking Edition. Kyoto: Otani University,
Van Dam, Raymond, trans. 1988. Gregory of Tours, Glory of the Martyrs. Liverpool: Liverpool University Press.
2Ó8
FJřTEE K
Religion, Religions, Religious
Jonathan Z. Smith
In the second earliest account of the "New World" published in English, A Trea-tyse of the Newe India (1553), Richard Eden wrote of the natives of the Canary Islands that, "At Columbus first comming thether, the inhabitantes went naked, without shame, religion or knowledge of God." In the same year, toward the beginning of the first part of his massive Crónica del Perú (1553), the conquistador historian Pedro Cíeza de León described the north Andean indigenous peoples as "observing no religion at all, as we understand it (no . . . religion alguna. d lo que entendemos), nor is there any house of worship to be found." Wliile both were factually incorrect, their formulations bear witness to the major expansion of the use and understanding of the term "religion" that began in the sixteenth century and anticipate some of the continuing issues raised by that expansion: (1) "Religion" is not a native category. It is not a first person term of self-characterization. It is a category imposed from the outside on some aspect of native culture. It is the other, in these instances colonialists, who are solely responsible for the content of the term. (2) Even in these early formulations, there is an implicit universality. "Religion" is thought to be a ubiquitous human phenomenon; therefore, both Eden and Cieza find its alleged absence noteworthy. (3) In constructing the second-order, generic category "religion," its characteristics are those that appear natural to the other. In these quotations this familiarity is signaled by the phrases "knowledge of God" and "religion ... as we understand it." (4) "ReHgion'Ms^anjritiTropological not a theological category. (Perhaps the only exception is the distinctively American nineteenth-centuiy coinages, "to get religion" or "to experience religion.") It describes human thought and action, most frequently in terms of belief and norms of behavior. Eden understands the content of "religion" largely in the former sense ("without. . . religion or knowledge of God"), whereas Cieza articulates it in the latter ("no religion ... nor ... any house of worship").
The term "religion" has had a long history, much of it, prior to the sixteenth century, irrelevant to contemporary usage. Its etymolog}' is uncertain, although one of tlie three current possibilities, that it stems from tire root *leig meaning "to bind" rather than from roots meaning "to reread" or "to be careful," has been the subject of considerable Christian homiletic expansion from Lactantius's Divine Institutes (early fourth century) and Augustine's On Trus Religion (early fifth century) to William Camden's Britannia (1586). In both Roman and early Christian Latin usage, the noun forms religio/religiones and, most especially, the adjectival religiosus and the adverbial religiose were cultic terms referring primarily to the careful performance of ritual obligations. This sense survives in the
269
Jonathan Z.
Smith
English adverbial construction "religiously" designating a conscientious repetitive action such as "She reads the morning newspaper religiously." The onlv distinctively Christian usage was the fifth-century extension of this cultic sense to the totality of an individual's life in monasticism: "religion," a life bound by monastic vows; "religious," a monk; "to enter religion," to join a monastery It is this technical vocabulary that is first extended to non-Cliristian examples'in the literature of exploration, particularly in descriptions of the complex civilizations of Mesoamerica. Thus Hernán Cortés, in his second Carva de Relation (1520, 64), writes of Tenochtitlan:
This great city contains many mosques [mezquitas, an eleventh century Spanish loan word from the Arabic, masjid], or houses for
idols-----The principal ones house persons of their religious orders
(persona* religiöses de su secu)-----All these monies (religiosos) dress
in black... from the time they enter the order (entmn en la religion).
Cortes's relatively thoughtless language of assimilation is raised to the level of a systemic category two generations later in the encyclopedic work of the Jesuit scholar Joseph de Acosta, The Natural and Moral History of the Indies (1590-English translation, 1604). While the vast majority of the occurrences of the term "religious" refer to either Catholic or native members of "religious orders," sometimes expanded to the dual category, "priests and monks of Mexico" (los sacerdotes y religiosos de Mexico), a number of passages strain toward a more generic conception. The work is divided into two parts, with the latter "moral history," chiefly devoted to religion, governance, and political history. "Religion" per se is never defined. Its meaning must be sought in words associated with it as well as its synonyms. For Acosta, "religion" is the belief system that results in ceremonial behavior. "Religion" is "that which is used (que man) m their rites." "Custom"  (costumbre), "superstition" (superstition), and "religion" (religion) form a belief series in conjunction with the action series of "deed" (hecho), "rite" (rito), "idolatry" (idolatria), "sacrifice" (sacrifice) "ceremony" (ceremónia), and "feasts" (fiestas y solemnidades).                        '
"Religion" in relation to ritual practice became an item in an inventory of cultural topics that could be presented either ethnographically in terms of a particular people, as in Eden or Cieza with reference to the "Indies," or in a cross-cultural encyclopedia under the heading of "ritual" or "religion." The encyclopedic version is illustrated by Joannes Boemus's popular Omnium gentium mores, leges et ritus (1520), in which ritus was translated as "customs" in the English translations by William Watreman, The Fardle ofFacions, Conteining the Aunciente Manners, Customes and Lawes of the People Inhabiting the Two Partes of the Earth (1555) and by Edward Aston, The Manners, Laws and Customs of all Nations (1611), and by Sebastian Muenster's Cosmographiae universalis. . Item omnium gentium mores, leges, religio (1550). This focus on ritual had an unintended consequence. The myths and befiefs of other folk could simply be
Religion, Religions, Religious
recorded as "antiquities," to use the term employed by Columbus. They raised no particular issues for thought. But ritual, especially when it seemed similar to Christian practice or when it illustrated categories of otherness such as "idolatry" or "cannibalism," gave rise to projects of comparative and critical inquiries. Similarity and difference, with respect to ritual, constituted a puzzle that required explanation by appeals to old patristic, apologetic charges of priesdy deceit or to equally apologetic, patristic theories of accommodation, demonic plagiarism, diffusion, or degeneration. In the case of belief and myth, "their" words were primary; with ritual, "our" account superseded theirs.
Some two centuries later, this essentially Catholic understanding of "religion" in close proximity to ritual has been decisively altered. Samuel Johnson, in his Dictionary of the English Language (1755), defines "religion" as "virtue, as founded upon reverence of God, and expectations of future rewards and punishments." The first edition of the Encyclopaedia Britannica (1771) titled its entry "Religion, or Theology," defining the topic in the opening paragraph: "To know God, and to render him a reasonable service, are the two principal objects of religion.... Man appears to be formed to adore, but not to comprehend, the Supreme Being." Terms such as "reverence," "service," "adore," and "worship" in these sorts of definitions have been all but evacuated of ritual connotations, and seem more to denote a state of mind, a transition begun by Reformation figures such as Zwingli and Calvin who understood "religion" primarily as "piety." The latter term takes on a less awesome cast in subsequent Protestant discourse, for example, "Piety, a Moral vertue which causes us to have affection and esteem for God and Holy Things" (Phillips 1696).
This shift to belief as the defining characteristic of religion (stressed in the German preference for the term Glaube over Religion, and in the increasing English usage of "faiths" as a synonym for "religions") raised a host of interrelated questions as to credibility and truth. These issues were exacerbated by the schismatic tendencies of the various Protestantisms, with their rival claims to authority, as well as by the growing awareness of the existence of a multitude of articulate, non-Christian traditions. The former is best illustrated by the first attempt to provide a distribution map for the various European Protestantisms: Ephraim Pagitt's Christianogmphie, or The Description of the Multitude and Sundry Sorts of Christians in the World Not Subject to the Pope (1635). The latter is the explicit subject of the anthropological work by Edward Brerewood, Enquiries Touching the Diversity of Languages and Religions through the Cbiefe Parts of the World (1614), which distinguished four "sorts" (i.e., "species") of the genus "religion"—"Christianity, Mohametanism, Judaism and Idolatry"—and provided statistical estimates for "the quantitie and proportion of the parts of the earth possessed by the several sorts" (118-19). It is the question of the plural religions (both Christian and non-Christian) that forced a new interest in the singular, generic religion. To cite what is perhaps the first widely read English book to employ the plural in its title, Purchas His Pilfjrimaj^e; or, Relations of the World
Jonathan Z. Smith
and the Religions Observed m All Ages and Places Discovered, "The true Religion can be but one, and that which God himselfe teacheth[,] ... all other religions being but strayings from him, whereby men wander in the darke, and in labyrinthine errour" (Purchas 1613, sig. D4r). What is implicit in Purchas becomes-explicit in later seventeenth- and eighteenth-century debates concerning "natural religion," a term that became common only in the latter half of the seventeenth century, beginning with works such as the one by the prolific Puritan controversialist Richard Baxter, The Reasons of the Christian Religion (1667), in two parts: "Of Natural Religion, or Godliness," and "Of Christianity, and Supernatural Religion." (Compare Baxter's earlier but congruent terminolog}', Of Saving Faith, That It Is Not Only Gradually but Specifically Distinct from All Common Faith [1658]).
Äs David Pailan (1994) has demonstrated, the notion of natural religion has been employed in the literature "to designate at least eleven significantly different notions, some of which have significant sub-divisions" ranging from "religious beliefs and practices that are based on rational understanding that all people allegedly can discover for themselves and can warrant by rational reflection" to "drat which is held to be common to the different actual faiths that have been and are present in the world." The former definition largely grew out of intra-Christian sectarian disputation and relied primarily on processes of introspection; the latter arose from study of the "religions," and involved processes of comparison. The essentially anthropological project of describing natural religion privileged similarity, often expressed by claims of universality or innate-ness; the explanation of difference was chiefly historical, whether it emphasized progressive or degenerative processes. This double enterprise had the effect of blurring the distinctions between questions of truth and questions of origins. For example, the title of Matthew Tindaľs fairly pedestrian but widely read treatise, published anonymously as Christianity As Old as the Creation; or, The Gospel, a Republication of the Religion of Nature (1730; six printings by 1732, and the British Museum General Catalogue lists more than forty replies in the 1730s), contains early English uses of the terms "religion of nature" and "Christianity." Tindal argues:
If God, then, from the Beginning gave Men a Reiigion[,]... he must have giv'n them likewise sufficient Means of knowing it. . . . If God never intended Mankind shou'd at any Time be without Religion, or have false Religions; and there be but One True Religion, which ALL have been ever bound to believe, and profess[,] ... All Men, at all Times, must have had sufficient Means to discover whatever God de-sign'd they shou'd know and practice. . . . [He] has giv'n them no other Means for this, but the use of Reason... .There was from the Beginning but One True Religion,-which all Men might know was their Duty to embrace... .By [this] Natural Religion, I understand the Be-
272
Religion, Religions, Religious
lief of the Existence of a God, and the Sense and Practice of those Duties, which result from the Knowledge, we, by our Reason, have of Him and his Perfections; and of ourselves, and our own Imperfections; and of the Relations we stand in to him, and to our Fellow-Creatures; so that the Religion of Nature takes in even' Thing that is founded on the Reason and the Nature of Things, (pp. 3-7,13)
While Tindal acknowledges some relativity—"I do not mean by This that All shou'd have equal Knowledge; but that All shou'd have what is sufficient for the Circumstances they are in" (p. 5)—his usual explanation for variation is the historical institution and wiles of "priestcraft":
Religion either does not concern the Majority, as being incapable of forming a Judgement about it; or must carry such internal Marks of its Truth, as Men of mean Capacity are able to discover; or else notwithstanding the infinite Variety of Religions, All who do not understand the Original Languages their traditional Religions are written in, which is all Mankind, a very few excepted, are alike bound in all Places to pin their Faith on their Priests, and believe in Men, who have an Interest to deceive them; and who have seldom fail'd to do so, when Occasion serves, (p. 232)
In Tindaľs self-description,
He builds nothing on a Thing so uncertain as Tradition, which differs in most Countries; and of which, in all Countries, the Bull; of Mankind are incapable of judging; but thinks he has laid down such plain and evident Rules, as may enable Men of the meanest Capacity, to distinguish between Religion, and Superstition, (p. iii)
When Tindal argued on logical grounds, the presumption of the unity of truth, that natural religion "differs not from Revmľd, but in the manner of its being communicated: The One being the Internal, as tire Other the External Revelation" (p. 3) he signaled the beginning of the process of transposing "religion" from a supernatural to a natural history, from a theological to an anthropological category. This process was complete only when die distinctions between questions of truth and questions of origin were firmly established. While not without predecessors, the emblem of this transposition is David Hume's essay The Natural History of Religion, written between 1749 and 1751 and first published in his collection Four Dissertations (1757).
The question Hume sets out to answer in the Natural History is that of religion's "origin in human nature." He begins by disposing of the innateness thesis. If "religion" is defined as "the belief of invisible, intelligent power," then, although widely distributed, it is not universal, nor is there commonality: "no two nations, and scarce any two men, have ever agreed precisely in the same
273
Jonathan Z.
Smith
sentiments." "Religion" fails the minimal requirements for ínnateness, that it be "absolutely universal in all nations and ages and has always a precise, determinate object, which it inflexibly pursues." Therefore, "religion" is not "an original instinct or primary impression of nature," and "the first religious principles must be secondary." In addition, because they are "secondary," religious principles "may easily be perverted by various accidents and causes" (p. 25). In this opening move, a major thesis is forecast. There may well be a primär)' and valid human experience that gives rise to the secondary religious interpretation, but the truth of the experience is no guarantee of the validity of the interpretation.
The rich details of Hume's exposition need not concern us here but only the argument with respect to this issue. "Polytheism or idolatry was ... the first and most antient religion of mankind." Its origin must be sought in "the ordinary affections of human life." Filled with anxiety, human beings seek the "unknown causes" that "become the constant object of our hope and fear." The primary human experience, "hope and fear," becomes a secondary religious interpretation when these "unknown causes" are personified through "imagination" (pp. 26, 31-33).
There is a universal tendency amongst mankind to conceive all beings like themselves, and to transfer to every object those qualities, with which they are familiarly acquainted, and of which they are intimately conscious... . No wonder, then, that mankind, being placed in such an absolute ignorance of causes, and being at the same time so anxious concerning their future fortunes, should immediately acknowledge a dependence on invisible powers, possest of sentiment and intelligence. The unknown causes, which continually employ their thought, appearing always in the same aspect, are all apprehended to be of the same kind or species [as themselves]. Nor is it long before we ascribe to them thought, and reason, and passion, and sometimes even the limbs and figures of men, in order to bring them nearer to a resemblance with ourselves, (pp. 33-34)
What Hume here raises is the issue of the adjectival form "religious." What sort of primary human experience or activity does it modify? What constitutes its distinctive secondary interpretation? How may religious interpretation be assessed in relation to other sorts of interpretation of the same experience or activity? The "religious" (the unknown that the scholar is seeking to classify and explain) becomes an aspect of some other human phenomenon (the known). As Walter Capps (1995, 9) has argued, in the eighteenth-century Enlightenment debates "the goal of the inquiry was to make religion intelligible by discovering precisely where it is situated within the wide range of interactive human powers and faculties." In which of the genera of common individual human capacities is the religious a species? Most frequently, the religious is identified with rationality, morality, or feeling.
274
Religion, Religions, Religious
A different set of taxonomie questions were raised by the "religions" and became urgent by the nineteenth century: Are the diverse "religions" species of a generic "religion"? Is "religion" the unique beginner, a summam genus, or is it best conceived as a subordinate cultural taxon? How might the several "religions" be classified?
The question of the "religions" arose in response to an explosion of data. Increased mastery of non-European languages led by the latter part of the eighteenth century to a series of translations and editions of religious texts. Missionaries, colonial officials, and travelers contributed ethnographic descriptions. Encyclopedias of religions, lexica, and handbooks (the latter, frequently bearing the title "History of Religions") were produced to organize these materials. One of the earliest handbooks, Historische-theologische Bericht vom Unterschied der Religionen die Heute zu Tage auf Erden sind, by the Lutheran scholar Johann Heinrich Ursin (1563), focused heavily on the various Christian denominations, establishing a pattern that holds to the present day: that the history of the major "religions" is best organized as sectarian history, thereby reproducing the apologetic patristic heresiologicai model. By the time of Brerewood's Enquiries Touching the Diversity of Languages and Religions (1614) this horizon had been extended to require inclusion of not only Christian data but also Jewish, Muslim, and "idolatry." This fourfold schema was continued by other writers from the seventeenth century (for example, Guebhart Meier, Históriu religionům, Christianae, Judaeae, Gentilis, Mahumedanae [1697]) until well into the nineteenth century (Hannah Adams, A Dictionary of All Religions and Religious Denominations, Jewish, Heathen, Mahometan, and Christian, Ancient and Modern [1817]; David Benedict, History of All Religions, As Divided into Paganism, Mahometism, Judaism, and Christianity [1824]; J. Newton Brown, Encyclopedia of Religious Knowledge: or, Dictionary... Containing Definitions of All Religious Terms; An Impartial Account of the Principal Christian Denominations that have Existed in the World from the Birth of Christ to the Present Day with their Doctrines, Religious Rites and Ceremonies, as well as those of the Jews, Mohammedans, and Heathen Nations, together with the Manners and Customs of the East [1835b]; Vincent Miiner, Religious Denominations of the World: Comprising a General View of the Origin, History and Condition of the Various Sects of Christians, the Jews, and Mahometans, As Well as the Pagan Forms of Religion Existing in the Different Countries of the Earth [1872]). The bulk of the subsequent expansion occurred in Brerewooďs fourth category, "Idolatry," with data added on Asian religions and on those of traditional peoples. Beginning with Alexander Ross, Pansebeia; or A View of All Religions in the World from the Creation to These Times (1614), there was a steady stream of reference works that undertook this task, including Bernard Picart and J. E. Bernard, Ceremonies et coutumes de tous peuples du monde (1723-43); Antoine Sanier, Historie general des ceremonies, moeurs, et coutumes religieuses de tous les peuples du monde (1741); Thomas Broughton, An Historical Dictionary of All Religions, from the Creation of the
275
jo^kr^-wr^st
World to the Pi emit Tune (1742), Christopher Meiners, Grundriss der Geschichte aller Religionen (178 5) and Allgemeine kritische Geschichte der Religionen (1806-7); John Beilemy, The History of All Religions (1812); and Benjamin Constant, De la religion considérée dans sa source, ses formes et ses développements (1824-31). This undertaking invented the familiar nomenclature, "Boudhism" (1821), "Hindooism" (1829, which replaced the earlier seventeenth-century usages "Gentoo [from "gentile"] religion" and "Banian religion"), "Taouism" (1839), and "Confucianism" (1862). The urgent agendum was to bring order to this variety of species. Only an adequate taxonomy would convert a "natural history" of religion into a "science."
The most common form of classifying religions, found both in native categories and in scholarly literature, is dualistic and can be reduced, regardless of what cUfferentium is employed, to "theirs" and "ours." By the time of the fourth-century Christian Latin apologists, a strong dual vocabulary was well in place and could be deployed interchangeably regardless of the individual histories of the terms: "our reIigion'/"their religion," with the latter often expressed through generic terms such as "heathenism," "paganism," or "idolatry"; "true reii-gion'/"false religion"; "spiritual (or "internal") religion'/"material (or "external") religion"; "monotheism" (although this term, itself, is a relatively late construction)/ "polytheism"; "religion'/"superstition"; "religion'/"magic." This language was transposed to intrareligious disputation with respect to heresies, and later revived in positive proposals of originär)' recover)' in Christian Renaissance hermetism as well as, most massively and insistently, in Protestant polemics against Roman Catholicism. As such, it was at hand for the evaluation of the newly encountered religions beginning in the sixteenth century. Lifting up the fourfold enumeration of religions—Christianity, Judaism, Islam, and "Idolatry"— Christianity, in some imagination of its ideal form, became the norm in which Judaism and Islam problematically share. Adopting a term from Muslim discourse, these three "Abrahamic religions" form one set over and against an undifferentiated other:
It is indeed probable, that all the idolatrous systems'of religion, which have ever existed in the world, have had a common origin, and have been modified by the different fancies and corruptions of different nations. The essence of idolatry is every where the same. It ís every where "abominable" in its principles and its rites, and every where the cause of indescribable and manifold wretchedness. (Brown 1835a, 229)
The initial problem for a classification of the religions is the disaggregation of this category.
One of the more persistent stratagems was the conversion of the epistemo-logical duality natural/supernatural into a characterization of the object of belief
276
Religion, Religions, Religious
{as in "nature worship") and the placement of these two terms in a chronological relationship.
The elements of nature were . . . the first divinities of man; he generally has commenced with adoring material beings. . . . Everything was personified, . . . Natural philosophers and poets [later distinguished] nature from herself—from her own peculiar energies— from her faculty of action. By degrees they made an incomprehensible being of this energy, which as before they personified: this abstract metaphysical being the)' called tire mover of nature, or God. (Mirabaud 1770,2:4)
This simple schema of two religions could be greatly extended by the addition of intermediate stages in the temporal series.
Nineteenth-cen tun' anthropological approaches focused on increasing the number of "natural" religious categories, especially for "primitive" peoples, those held to be "nature peoples" (Naturvölker). Often mistermed evolutionär)', these theories conceded no historical dimensions to those being classified but rather froze each ethnic unit at a particular "stage of development" of the totalit)' of human religious thought and activity. "Natural" religion was segmented into fetishism, totemism, shamanism, anthropomorphism, preanimism, animism, family gods, guardian spirits, ancestor worship, departmental gods, to name but a few. If the category "natural" were to be taken as including not only "primitives" but "antiquity," a set of peoples with whom the scholar more readily identified, then a meager note of historical dynamism would be introduced. For example, A. M. Fairbairn in his Studies in the Philosophy of Religion and History (1S76) divided "Spontaneous or Natural Religions" into two classes, "Primitive Naturalisms" (which included, among others, "primitives" and the "early" Greeks, Hindus, Teutons, and Slavs) and "Transformed Naturalisms" (e.g., "later" Greeks and Romans, Egyptians, and "ancient" Chinese). The "high religions," which could be designated "spiritual," required a different technique for their division, one that recognized history. One proposal, establishing an alternative duality that remains current to this da)', was set forth by the distinguished American Sanskritist, W. D. Whitney (1881, 451): "There is no more marked distinction among religions than the one we are called upon to make between a race religion—which, like a language, is die collective product of the wisdom of a community, the unconscious growth of generations— and a religion proceeding from an individual founder." He cites as examples of the latter, Zoroastrianism, "Mohammedanism," Buddhism, and Christianity, noting that the latter may be described as "growing out of one [Judaism] that was limited to a race." Whitney here makes clear the dilemma posed by the study of the "religions" from the perspective of the spiritual. The older fourfold enumeration of the three "Abrahamic religions" plus "Idolatry" required revision.
277
Jonathan Z.
Smith
Judaism was to be demoted in that from a Christian apologetic perspective, it was the very type of a "fleshly religion"; Buddhism was to be promoted because in the two-century history of the Western imagination of Buddhism, it had become the very type of "spiritual religion."
Fairbairn adjusted his model such that the ultimate duality was between "spontaneous or natural religions" and "instituted religions," with the latter having two classes, each characterized by the same powerfully positive Protestant term: "Reformed Natural" (including the archaic religion of Israel ["Mosa-ism"], Zoroastrianism, Confucianism, Taoism), and "Reformed Spiritual," limited only to the new triad (Buddhism, "Mohammedanism," and Christianity). All other "religions" fell into one of three classes of "natural," the replacement term for the older category, "idolatry."
The most enduring device was the invention of the taxon "world" or "universal religions," a division that appeared to recognize both history and geography. The term was introduced and placed in a classificatory scheme that synthesized previous taxonomie divisions in a work that stands as the first classic in the science of religion, Cornelius Petrus Tiele's work Outline of the History of Religion to the Spread of Universal Religions (1876), and was reworked in Tiele's article "Religions" in the ninth edition of the Encyclopaedia Britannica (1884). Tiele's "morphological" classification, which schematizes die "stage of development" each religion has "attained," has as its fundamental principle of division "natural religion" and "ethical religion," which he self-consciously correlates with Whitney's distinction between "race religion" and "founded religion." "Natural religion" has three families, one of which has two genera. The first family comprises "polydaemonistic magical religions under the control of animism." To this class "belong [ail] die religions of the so-called savages or uncivilized peoples." Recognizing, perhaps, the effects of colonialism, lie adds that their present forms are "only degraded remnants of what they once must have been."
The second family of "nature religions" is that of "purified or organized magical religions," which Tide terms "therianthropic polytheism," according to which the "gods are sometimes represented in human form, more frequendy in that of an animal." These are politically divided into two families, "unorganized" (tribal) and "organized" (imperial). The "unorganized" include the Japanese kami traditions, the Dravidians, the Finns, the "old Arabic religions, old Pelasgic religion, old Italiote religions, Etruscan religion before its admixture with Greek elements, [and] the old Slavonic religions." The "organized" include "the semi-civilized religions of America,... the ancient religion of the Chinese empire, ancient Babylonian (Chaldaean) religion, [and] die religion of Egypt."
The third family, "anthropomorphic polytheism," is characterized by the "worship of manlike but superhuman and semi-ethical beings" (the latter indicating that while the gods are often represented as being concerned with good and evil, they are also depicted as essentially amoral). Belonging to this class are
278
Religion, Religions, Religious
"the ancient Vaidic religion (India), the pre-Zarathustrian Iranic religion, the younger Babylonian and Assyrian religion, the religions of the other civilized Semites, the Celtic, Germanic, Hellenic and Graeco-Roman religions."
Distinct from these "nature religions" are those belonging to the second major division, "ethical religions," which are subdivided into "national nomistic (nomothetic) religious communities" characterized by being "founded on a law or holy scripture," that is, "Taoism and Confucianism .. . Brahmanism, with its various ancient and modern sects, Jainism and primitive Buddhism, Mazdaism (Zarathustrianism) with its sects, Mosaism [and] Judaism," and "universalistic religious communities," a class with only three members: Islam, Buddhism, Christianity. They are distinguished in not being devoted to the special interests of a nation or people but to humankind in general; they are proselytizing traditions.
After discussing at some length the relative merits of die labels "universalistic," "universal," and "world religions," Tiele employs blunt imperialistic language to defend his use of "world religions" to
distinguish the three religions which have found their way to different races and peoples and ail of which profess the intention to conquer the world, from such communities [that is, "national, nomistic religions"] as are generally limited to a single race or nation, and, where they have extended farther, have done so only in the train of, and in connection with, a superior civilization. Strictly speaking, there can be no more than one universal or world religion, and if one of the existing religions is so potentially, it has not yet reached its goal. This is a matter of belief which lies beyond the limits of scientific classification.... Modern history of religions is chiefly the history of Buddhism, Christianity and Islam, and of their wrestling with the ancient faiths and primitive modes of worship, which slowly fade away before their encroachments, and which, where they still survive in some parts of the world and do not reform themselves after the model of the superior religion, draw nearer and nearer to extinction.
Furthermore, he apologetically insists, the three "world religions" are not on an equal plane. Islam "is not original, not a ripe fruit, but rather a wild offshoot of Judaism and Christianity," "in its external features [it] is little better than an extended Judaism." Buddhism "neglects the divine" and while "atheistic in its origin, it very soon becomes infected by the most fantastic mythology and the most childish superstitions." Christianity "alone preaches a worship in spirit and in truth . . . the natural result of its purely spiritual character, Christianity ranks incommensurably high above both its rivals." Despite the latter assertion, Tiele insists that "we are giving here neither a confession of faith nor an apology. . . . we have here to treat Christianity simply as a subject of comparative study, from a scientific, not from a religious point of view." (Tiele 1884,20:358-71.)
279
I             t        I        ■:(        [        w        I             it        i        :f        i
Later scholars expanded the number of world religions to seven by collapsing Tide's two classes of "ethical religions" in an odd venture of pluralistic etiquette: if Christianity and Islam count as world religions, then it would be rude to ex-,elude Judaism (ironically, the original model for the opposite type, "national nomistic religions"). Likewise, if Buddhism is included, then Hinduism cannot be ignored. And again, if Buddhism, then Chinese religions and Japanese religions.
It is impossible to escape tlie suspicion that a world religion is simply a religion like ours, and that it is, above all, a tradition that has achieved sufficient power and numbers to enter our history to form it, interact with it, or thwart it. We recognize bodi the unity within and the diversity among the world religions because they correspond to important geopolitical entities with which we must deal. All "primitives," by way of contrast, may be lumped together, as may the "minor religions," because they do not confront our history in any direct fash-. ion. From the point of view of power, they are invisible.
Attempting to avoid such strictures and suspicions, other scholars have turned to alternative modes of classification. Following the implied correlation in Brere-wood's Enquiries Touching the Diversity of Languages and Religions, F. Max Müller (1873,143) argued "that the only scientific and truly genetic classification of religions is the same as the classification oflanguages," while Brerewood's interest in statistics has led to geographical taxonomies, either demographic (Haupt 1821 is an early example) or in terms of spatial distribution (for example, Deffontaines 1948). Others combine these elements with ethnographic classifications maintaining that any particular "religion derives its character from the people or race who develop it or adopt it" (Ward 1909, 64). All of these result in projects describing "the religion of" such and such a geographical region or folic, arguing that these eschew tine imposed universalisms or barely disguised apologetics of their predecessors in the name of a new ethic of locality that often favors native categories. Thus, Clifford Geertz introduces his early work The Religion of Java (1960) by emphasizing the copresence of nativistic, Islamic, and "Hinduist" elements, arguing that "these three main subtraditions ... are not constructed types, but terms and divisions the Javanese themselves apply. . . . Any simple unitary view is certain to be inadequate; and so I have tried to show . . . variation in ritual, contrast in belief, and conflict in values" (pp. 6-7). What remains uncertain is what he intends by the singular religion in his title.
As in the eighteenth century, so too in the late twentieth do the issues attending the religions force the definitional question of religion. Two definitions command widespread scholarly assent, one essentially theological, the other anthropological. Paul Tillich, reversing his previous formulation that religion is concern for the ultimate, argued that
religion, in the largest and most basic sense of die word, is ultimate concern . . . manifest in the moral sphere as the unconditional seri-
280

°N' Religions, Religious
ousness of moral demand[,] ... in the realm of knowledge as the passionate longing for ultimate reality[,J... in the aesthetic function of the human spirit as the infinite desire to express ultimate meaning." [Religion is not a] special function of man's spiritual life, but the dimension of depth in all its functions. (1959, 7-8)
As Tillich's earlier concern with topics such as idolatry and the demonic should suggest, this is not as generous and open ended a definition as might seem to be implied. There are insufficient, inadequate, and false convictions of "ultimacy." Tillich has in fact provided a definition of the religious, as a dimension (in his case, the ultimate, unconditioned aspect) of human existence. This is explicit in William A. Christian's reformulation: "Someone is religious if in his universe there is something to which (in principle) all other things are subordinated. Being religious means having an interest of this land" (1964, 61). If one removes Tillich's and Christian's theological criteria (as, for example, Robert D. Baird suggests in Category Formation and the History of Religions [1971]), then it becomes difficult if not impossible to distinguish religion from any other ideological category. Tin's would be the direction that Ninian Smart (1983) points to in suggesting that religion be understood as "worldview," with the latter understood as a system "of belief which, through symbols and actions, mobilize[s] the feelings and wills of human beings" (pp. 2-3).
The anthropological definition of religion that has gained widespread assent among scholars of religion, who both share and reject its functionalist frame,, is that formulated by Melford E. Spiro (1966, 96), "an institution consisting of culturally patterned interaction with culturally postulated superhuman beings." This definition requires acceptance of a broad theory of cultural creation, signaled by the phrases "culturally patterned" and "culturally postulated," and places human cultural-activities or instiaitions as the summum genus and religion as a subordinate taxon. This is made plain in Spiro's formulation that "religion can be differentiated from other culturally constituted institutions by virtue only of its reference to superhuman beings" (p. 98). Subsequent reformulations by scholars of religion have tended either to remove this subordination (for example, Penner 1989) or to substitute "supernatural" for "superhuman" (as in Stark and Bainbridge 1987).
It was once a tactic of students of religion to cite the appendix of James H. Leuba's Psychological Study of Religion (1912), which lists more than fifty definitions of religion, to demonstrate that "the effort clearly to define religion in short compass is a hopeless task" (King 1954). Not at all! The moral of Leuba is not that religion cannot be defined, but that it can be defined, with greater or lesser success, more than fifty \vays. Besides, Leuba goes on to classify and evaluate his list of definitions. "Religion" 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
281
Jonathan Z. Smith
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. .
Suggested Readings
Almond, Philip C. 1988. The British Discovery of Buddhism. Bianchi, Ugo, Fabio Mora, and Lorenzo Bianchi, eds. 1994. The Notion of "Religion" in Comparative Research: Selected Proceedings of the Sixteenth Congress of the International Association for the History of Religions, Rome,
3-8 September 1990. Capps, Waiter H. 1995. Religious Studies: Tioe Making of a Discipline. Despiand, Michael. 1979. La Religion en Occident. Feil, Ernst. 1986. Religio: Die Geschichtes eines neuzeitlichen Grundbegriffs vom
Frühchrisentum bis zur Reformation. Harrison, Peter, 1990. "Religion" andthe Religions inthe English Enlightenment. Manuel, Frank E. 1959. The Eighteenth Century Confronts the Gods. Pailan, David A. 1984. Attitudes to Other Religions: Comparative Religion in
Seventeenth- and Eighteenth-Century Britain. Penner, Hans H. 1989. Impasse and Resolution: A Critique of the Study of
Religion. Preus, J. Samuel. 1987. Explaining Religion: Criticism and Theory from Bodin
to Freud. Smith, Wilfred Cantwell. 1963. The Meaning and End of Religion. Spiro, Melford E. 1966. "Religion: Problems of Definition and Explanation."
In Anthropological Approaches to the Study of Religion, edited by Michael
Banton.
References
Acosta, Joseph de. 1ÓQ4. The Natural and Moral History of the Indies, translation of the original
Spanish edition, published in Seville in 1590. London: V. Sims for E. Blount and W. Aspiey. Adams, Hannah. IS 17. A Dictionary of 'All Religions and Religious Denominations, Jewish, Heathen,
Mahometan, and Christian, Ancient and Modern. 4th ed. Boston: Cummings and Hilliard. Aston, Edward, trans. 1611. The Manners, Laws, and Customs of All Nations, by Joannes Boemus.
London. Baird, Robert D. 1971. Category Formation and the History of Religions. The Hague: Mouton de
Gmyter. Banier, Antoine. 1741. Histoiregenerale des ceremonies, moeurs, et coutumes religieuses de touš lespea-
pies du monde. Paris. Baxter, Richard. 1658. Of Saving Faith, Tlmt It Is Not Only Gradually but Specifically Distinct from
All Common Faith. London.
-----—. 1667. The Reasons of the Christian Religion. London: Printed by R. White for Fran. Titon.
Bellemy, John. 1812. The History of All Religions, Comprehending the Different Doctrines, Customs,
and Order of Worship in the Churches. . . from the Beginning of Time to the Present Day.
London.
282
Religion, Religions, Religious
1
Benedict, David. 1824. History of All Religions, As Divided into Paganism, Mahometism, Judaism,
and Christianity. Providence, R.I. Bernard, J. F. 1723-43. Ceremonies et coutumes de tons peuples du monde. Amsterdam. Boemus, Joannes. 1520. Omnium gentium mores, leges, et rítus. Augsburg. Brerewood, Edward. 1614. Enquiries Touching the Diversity of Languages and Religions through the
Chiefe Parts of the World. London. Broughton, Thomas. 1742. An Historical Dictionary of All Religions, from the Creation of the World
to the Present Time. London. Brown, J. Newton. 1835a. "Buddhism." In Encyclopedia of Religious Knowledge. Brattleboro, Vt. —-----. 1835b. Encyclopedia ofReligious Knowledge; or, Dictionary... Containing Definitions ofAll
Religious Terms: AiiTmpartial Account of the Principal Christian Denominations Tlmt Have
Existed in the World from the Birth of Christ to the Present Day with Tlieir Doctrines, Religious
Rites, and Ceremonies, and Heathen Nations, Together with the Manners and Customs of the
East. Brattleboro, Vt. Camden, William. 1586. Britannia. London.
Capps, Walter H. 1995. Religious Studies: The Making of a Discipline. Minneapolis: Fortress Press. Christian, William A. 1964. Meaning and Truth in Religion. Princeton, N.J. Cieza de León, Pedro. 1553. Cronica del Perú. 4 vols. Seville. Reprint 1918, edited by D.Enrique
de Vedia, Historiadores primitives de Indias, 2 vols. Madrid: Imprenta de los Sucesores de
Hernando.
Constant, Benjamin, 1824-31. De la religion considérée dans sa source, ses formes, et ses développe-ments. Paris.
Cortés, Hernán. 1520. Cartas de Relacíon. Reprint 1971, edited by Manuel Akala. Sepan Cuantos,
no. 7. Mexico City: Editorial Porrúa. DetFontaines, Pierre. 1948. Geographie et religions. Paris. Eden, Richard. 1553.-d Treatyse of'the Newe India. London. Encyclopaedia Britannica. 1771. 1st ed. Edinburgh.
Fairbairn,A. M. 1876. Studies in the Philosophy of Religion and History. London. Geertz, Clifford. 1960. Tlie Religion of Java. Glencoe, III.: Free Press. Haupt, Karl G. 1821. Tabellarischer Abriss der vorzüglichsten Religionen und Religionsparteien der
Jetzigen Erdebewohner. Leipzig. Hume, David. 1757. The Natural History of Religion. In Four Dissertations. London: Printed for A.
Millar. Variorum edition, 1976, edited by A. Wayne Colver, Oxford, Clarendon Press. (Page
numbers cited in test are from the Variorum ed.) Johnson, Samuel, comp, and ed. 1755. A Dictionary of the English Language. London: W. Strahan,
for J. and P. Knapton. King, Winston L. 1954. Introduction to Religion. New York: Harper and Row. Leuba, James H. 1912. A Psychological Study of Religion. New York: Macmillan. Meier,  Guebhart.  1Ó97.  História religionům,  Christianae, Judaeae,  Gentilis, Mahumedanae.
Heimstadt.
Meiners, Christopher-1785. Grundriss der Geschichte aller Religionen. Lemgo.
———. 1806-7, Allgemeine kritische Geschichte der Rdigionm. Hannover: Helwing.
Milner, Vincent. 1872. Religious Denominations of the World: Comprising a General View of the Origin, History, and Condition of the Various Sects of Christians, the Jews, the Mahometans, as well as the Pagan Forms of Religion Existing in the Different Countries of the Earth. Philadelphia.
Mirabaud, M. [Paul Henry Thiery, Baron d'Holbach]. 1770. Systeme de la nature; on, Des lois du monde physique et du monde moral. [First English translation 1820, ii.p.]
Muenster, Sebasdan. 1550. Cosmographiae universalis...: Item omnium gentium mores, leges, religio. Basel.
Müller, F. Max. 1873. Introduction to the Science of Religion. London: Longmans, Green. Pagitt, Ephraim. 1635. Christianographu; or, The Description of the Multitude and Sundry Sorts of Christians in the World Not Subject to the Pope. London.
283
j.--------------^.^ AimiMi luccuiig ur nie American
Academy of Religion, Chicago.
Penner, Hans H. 1989. Impasse and Resolution: A Critique of the Study of Religion. New "York;
Peter Lang. Phillips, Edward. 1696. A New World of English Words; or, A General Dictionary Containing the
Interpretation of Such Hard Words As Are Derivedfrom Other Languages. London: E. Tyler
for Nathanael Brooke at the Sign of the Angel in Cornhill. Picart, Bernard, and ]. F. Bernard. 1723-43. Ceremonies et- eoutumes dc tons peupl&s du monde.
Amsterdam. Purchas, Samuel. 1613. Pur chas His Pilgrimage,- or, Relations of the World and the Religions Observed
in All Ages and Places Discovered, in FoureParts. London. Ross - Alexander. 1614. Pansebeia; or, A View of All Religions in the World from the Creation to These
Times, Together with the Discovery of All Known Heresies, in All Ages and Places. London. Smart, Ninian.  1983.   Worldviews:  Cross-Cultural Explorations of Human Beliefs. New York:
Scribner's.
Spiro, Meiford E. I960. "Religion: Problems of Definition and Explanation." In Anthropological Approaches to the Study of Religion, edited by Michael Banron. London: Tavistock.
Stark, Rodnej', and William S. Eainbridge. 1987. A Theory of Religion. New York: Peter Lang.
Tiele, Cornelius Petrus. 1877. Outline of the History of Religion to the Spread of Universal Religions, translation of the original Dutch edition, published in Amsterdam in 1876. London.
------—. 1884. "Religions," In Encyclopaedia Pritannica. 9th ed.
Tillich, Paul. 1959. Theology of Culture, edited by Robert C. Kimball. New York: Oxford University Press.
Tjndal, Matthew. 1730. Christianity As Old As the Creation; or, The Gospel, a Republication of the
Religion of Nature. London. Ursin, Johann Heinrich. 1563. Historische-theologische Bericht vom Unterschied der Religionen die
Heute zu Tage auf Erden sind. Nuremberg. Ward, Düren J. H. 1909. The Classification of Religions. Chicago: Open Court. Watreman, William, trans. 1555. ThePardleofFacions, Conteining the AuncienteManners, Customs,
and Lowes of the People Inhabiting the Two Partes of the Earth, by Joannes Boemus. London. Whitney, W. D. 1881, "On the So-Called Science of Religion." Princeton Revieiv 57, pt. 1;
429-52.
284
Jill Robbins
hile a comprehensive treament of the term, not to mention a theory of, "sacrifice," is not possible here, it will be useful to indicate the parameters and the points of reference that such a term and such a theory would comprise. After a general discussion of a historical and critical nature, I will turn brief!;' to an exemplar)' text from die Hebrew Bible, the binding of Isaac recounted in Genesis 22, in order to determine some of the questions it poses about sacrifice.
In the Hebrew Bible and in the cultic context of ancient Israelite religion, sacrifice, the offering up of slain animals for sacred purposes, holds a prominent place, although its full significance is not entirely understood. Leviticus, especially chapters 1 through 7 (general!;' attributed to the biblical author "P" or the Priesdy source), details the laws of sacrifice. It makes distinctions between such categories as propitiary offerings (as atonement for sins and as a purification ritual) and dedicator;' offerings (gifts for the deity). In the practice known as the sin offering, battn^at, the offering must be the property of the person making the sacrifice. The sacrificer lays his hand on the offering, thus identifying it with himself. The idea behind this practice was explained by the medieval commentator Nachmanides in his commentary on Lev. 1:9 as follows: the sinner's life is forfeit to God, but by a gracious provision, he is permitted to substitute an animal victim in his place. Lev. 17:11 explains that the substitution occurs precisely by the extraction of the animal's blood: "For the life of the flesh is in tire blood, and I have given it for you upon the altar to make atonement for your souls: for it is the blood that makes atonement, by reason of the life that is in it." While the idea of substitution and expiation is prominent in these cases, the dedicator;' offerings convey primarily the idea of gift giving, which will later be explored within anthropological theories of sacrifice. These offerings are distinguished in terms of the matter, the mode, and even the place of sacrifice. For example, there is die meal offering, the minku, sometimes translated as "gift." In Genesis 4, Cain brings such a gift to God, who in turn indicates his preference for Abel's animal sacrifice over Cain's cereal offering. The burnt offering, the olah, which means literally "that which goes up," namely, the smoke, describes the mode of deliver;' of the sacrifice. It refers to an offering that is entirely consumed in the fire, what the Septuagint translates as hoíok&ustus, and the King James, in its English equivalent, as "holocaust." (I will return later to the problematic use of this term to refer to the Nazi genocide of the Jews.) Ola-h is the word that is used in Genesis 8 for Noah's sacrifice, the odor of which is "pleasing" to God; it is also the word used in Genesis 22 when God commands Abraham to sacrifice his
285