INDO-EUROPEAN RELIGIONS: HISTORY OF STUDY

Strictly speaking, the history of comparative Indo-European studies begins in the late eighteenth century as a direct result of the momentous discovery that the ancient languages now classified as "Indo-European" (e.g., Latin, classical Greek, Sanskrit, Old English, Old Persian, Old Icelandic, Old Church Slavonic, Old Irish, Hittite, etc.) all stemmed ultimately from a common source, that is, Proto-Indo-European. As shall be seen, it soon became apparent that the speakers of these languages, which can be considered along with their progeny as members of a grand "family" of languages, shared more than simply a common linguistic heritage, and that among the most important features of this extralinguistic, Indo-European heritage was a common body of religious beliefs and practices.

To be sure, the taproots of the discipline can be traced back to classical antiquity, to the theories of Euhemerus (fl. 300 bce) and other Greek and Roman scholars who attempted to come to grips with the origin and meaning of myth. It is also possible to trace the immediate source of the ideas that flowered in the nineteenth century to the ideas of such eighteenth-century precursors as Bernard de Fontenelle (1657–1757), Giovanni Battista Vico (1668–1744), and Charles de Brosses (1709–1777), who first suggested that a search for natural metaphors might be preferable to the traditional euhemeristic and allegorical approaches that had heretofore been the rule. These ideas may also be traced as well to that curious (albeit all-pervasive) philosophical, literary, and artistic movement called Romanticism, adumbrated in the works of J. G. Herder (1744–1803), which profoundly influenced most of the scholars who first began to conceive of a distinctly Indo-European religious tradition in the early nineteenth century. But these ideas belong properly to the general history of comparative mythology and religion; for the purposes of this article, the survey begins with the discovery of the Indo-European language family.

Discovery of the Indo-European Language Family

Until the last quarter of the eighteenth century, most theories about the nature and origin of language were grounded in philosophical speculation, much of it centering on the idea of degeneration. Thus, the primordial language was often held to be Hebrew, since it must have been spoken in the Garden of Eden. Following the ancient notion of degeneration from an assumed "Golden Age," many writers on the subject maintained that Greek was a degenerate form of Hebrew, Latin a degenerate form of Greek, and that the modern languages of Europe were all degenerate offspring of Latin.

However, thanks to the voyages of discovery and the rapid expansion in European awareness of the range and diversity of human languages, and impelled by the romantic emphasis on national origins, which effectively precluded the notion that all languages were necessarily descended from Hebrew, scholars had begun to suspect that the degeneration hypothesis, whether secular or religious, was inadequate to explain the historical relationships among languages. Nowhere was this more obvious than in India, which, by the latter part of the eighteenth century, had become in effect the private preserve of the British East India Company. As European awareness of this vastly complex region deepened, it became clear that Sanskrit, the ancient language of the Hindu sacred texts, occupied a position in religious and literary affairs similar to that occupied by Latin in Europe during the Middle Ages. Indeed, several scholars, beginning with Filippo Sassetti in 1600, had remarked on the curious similarities between Latin and Sanskrit, but these similarities defied explanation in terms of the "degeneration hypothesis," as Sanskrit was patently as ancient as either Latin or classical Greek. Furthermore, the modern languages of North India—Hindi, Bengali, and the rest—seemed to bear the same immediate relationship to Sanskrit as French, Spanish, Italian and other members of what later came to be called the Romance languages did to Latin.

The problem was finally solved in 1786 by William Jones (1746–1794), who is generally considered the founder of scientific linguistics. An amateur philologist (he was reputed to have been fluent in some twenty-two languages), Jones had recently been appointed chief justice of the East India Company's establishment at Calcutta, and in his off-hours he immediately set about learning Sanskrit. In September 1786, at a meeting of the Royal Asiatic Society of Bengal, he gave an after-dinner speech in which, for the first time, the idea of the language family was first clearly articulated. As Jones saw it, the relationship among Sanskrit and the ancient languages of Greece and Rome, as well as those spoken by the ancient Germans and Iranians, was that of a set of orphaned siblings: all were descended from a common parent language that had long since disappeared. That parent language, however, might be reconstructed by rigorously comparing the grammars and lexicons of these attested languages. The whole ensemble could be described as a family tree, one to which Hebrew, Arabic, and other Semitic languages did not belong, for they were members of another, wholly distinct language family.

Thus was born both comparative philology and the idea of the Indo-European language family. Although Jones himself never followed up his monumental discovery, others soon did, and by the beginning of the third decade of the next century the science of comparative philology, together with the discipline now referred to as comparative Indo-European religious studies, was well under way.

Almost from the outset, the practitioners of this new science, almost all of them steeped in romantic idealism, found themselves confronted by more than simply a set of linguistic similarities. The primary source materials—the Ṛgveda, the Mahābhārata, the Iliad, the Iranian Avesta, the Icelandic Eddas, and so forth—were religious and/or mythological texts, and it soon became apparent that the gods, heroes, rituals, and events described in these texts could be compared using the same basic methodology that Jones and others had developed, that is, the comparative method, which is predicated on the assumption that anterior stages and/or prototypes can be systematically reconstructed from attested evidence, linguistic or otherwise. Thus, comparative mythology, and especially comparative Indo-European mythology, rapidly took its place as a sister discipline of comparative philology.

Early Nineteenth Century

As might be expected, many early nineteenth-century scholars, even those who were not directly concerned with Indo-European linguistic studies, had something to say about various aspects of the newly discovered parallels among the several Indo-European pantheons. This was especially true in Germany, where romantic concern with the origins of the Volk (German and otherwise) had become almost a national passion. Thus, Karl O. Müller (1797–1840) and G. F. Creuzer (1771–1858) drew heavily, albeit selectively, upon the linguistic evidence in their attempts to reconstruct the prototypes of Greek and other Indo-European gods and heroes. Even the philosopher G. W. F. Hegel (1770–1831), in whose works Romantic idealism reached the apex of its development, seems to have been strongly influenced by the new comparativism, and, as Richard Chase puts it, "longed for a 'polytheism in art' and imagination, a plastic and mythological philosophy" (Quest for Myth, 1949, p. 39).

Most of the pioneer philologists, among them Franz Bopp (1791–1867), Friedrich Schlegel (1772–1829), and Rasmus Rask (1787–1832), also made important contributions to comparative Indo-European mythological and religious studies. In many respects, the most distinguished member of this group was Jacob Grimm (1785–1863), who, with his brother Wilhelm (1786–1859), was responsible for amassing the great collection of tales that bears their name. However, Jacob Grimm was more than a mere collector of folk tales; he was also a preeminent philologist, and in 1823 he articulated the principle that later came to be known as "Grimm's law," which firmly established the phonological connections among Latin, Greek, and the ancient Germanic languages. His most important single contribution to Indo-European religious studies was a two-volume work entitled Deutsche Mythologie (1835). In it he developed the thesis that the Märchen he and his brother had collected were the detritus of pre-Christian Germanic mythology. This argument is bolstered by a host of etymologies, as well as comparisons to other Indo-European traditions. A good example of the latter is Grimm's suggestion that the ancient Scandinavian account of a war between the gods (Óðinn, Vili, and Vé) and an earlier generation of giants (Ymir et al.) is cognate to the Greek Titanomachy, or the war between the Olympians and the Titans (that is, between Zeus and his siblings and the supernatural beings of the previous generation, led by Kronos).

Elsewhere in Europe and in America interest in mythology, if not exclusively Indo-European mythology, also ran high. In Britain, for example, most of the Romantic poets—Wordsworth, Coleridge, Blake, Shelley, Keats, Byron, and others—drew extensively upon mythological themes; and Thomas Bulfinch's The Age of Fable (1855) popularized the study of mythology like no other work before it. Thus, by the middle of the nineteenth century, the science of comparative philology had reached maturity, interest in mythology and the history of religions had become widespread, and the stage was set for the appearance of the first grand paradigm in the history of Indo-European religious studies.

The First Grand Paradigm: F. Max MÜller and the Naturists

In his seminal book, The Structure of Scientific Revolutions (1970), the eminent philosopher of science Thomas S. Kuhn makes a persuasive case for the proposition that all scientific knowledge expands in what amounts to an ascending and ever-widening spiral. In its earliest stages a new discipline necessarily finds itself groping for a central focus, for an overarching model in terms of which theories can be generated. Eventually, however, thanks to the efforts of a few scholars, a breakthrough is made, and there emerges a grand paradigm, which not only organizes the knowledge heretofore gained, but by its very nature generates a host of new discoveries and/or interpretations. The emergence of such a paradigm is revolutionary in its impact, and constitutes a quantum leap forward in the history of a discipline.

Kuhn, of course, focuses his attention upon the growth of the physical sciences, which so far have known at least two grand paradigms—Newtonian mechanics and quantum mechanics/relativity—and which may well be on the verge of a third. But the model applies generally. For example, in the history of linguistics, William Jones's discovery led to that discipline's first grand paradigm, which indeed precipitated the study of Indo-European religions. However, it was not until the 1850s, almost sixty years after Jones's death, that Indo-European religious studies finally achieved its own grand paradigm.

The person most responsible for this "revolution" was F. Max Müller (1823–1900), a German-born Sanskrit scholar, philologist, and student of Indian religions who had studied with Bopp and the eminent French Sanskritist Eugène Burnouf. Shortly after completing his formal studies, Müller accepted a position at Oxford University as a lecturer in Sanskrit and Indian religions; as it turned out, he spent the rest of his career there, eventually becoming one of the Victorian era's most distinguished men of letters.

In 1856, seven years after arriving at Oxford, Müller published a long essay entitled simply "Comparative Mythology" (published in Oxford Essays, 1856), and the revolution was launched. Although he went on to publish a veritable library of books, as well as innumerable collections of essays, articles, introductions, and so forth, most of his basic ideas were laid out in "Comparative Mythology."

Solar mythology and the "disease of language"

First and foremost among Müller's ideas was the notion that the gods and heroes of the "Aryan" (i.e., Indo-European) peoples were basically metaphors for the sun, in all its aspects. To be sure, this was not a brand-new idea. In 1795, Charles-François Dupuis (1742–1809) had suggested that Jesus Christ was a solar metaphor and that the twelve apostles could be interpreted as the signs of the zodiac. But it was Müller who escalated the notion into a full-blown paradigm, one that had special relevance to the ancient Indo-European-speaking domain. Moreover, as a philologist, Müller insisted that the key to understanding these solar metaphors lay in the etymologies of divine names.

Müller asserted that language, including Proto-Indo-European, which he identified in effect with the earliest form of Sanskrit, was in its pristine state eminently rational. Objects such as the sun, the moon, stars, and other natural phenomena were labeled without reference to any divine beings or concepts, as the earliest dialects were incapable of expressing abstractions. But as time went on, Müller concluded, a curious malady set in, a "disease of language," the prime symptom of which was metaphor. What had begun as simple, descriptive terms gradually evolved into increasingly complex and abstract metaphors, and these in turn came to take on a life of their own. In short, by the time the earliest religious texts (e.g., the Ṛgveda and Hesiod's Theogony ) were composed, the disease of language had become terminal; myth and religion had replaced reason and rationality. By judicious use of the comparative method, however, one could cut through the layers of metaphoric accretion and arrive at the root meanings underlying divine and heroic names. Thus, for example, the equation between Zeus and the Indian figure Dyauh, which clearly stemmed from a Proto-Indo-European conception of the sky god, could be traced back to a series of abstract conceptions relating to light, brightness, dawn, and so on, which, in turn, ultimately derived from metaphors for various solar attributes. Although he admitted that other natural phenomena play a part in generating mythical metaphors, Müller constantly emphasized the sun as the prime source of Indo-European religious inspiration: "I am bound to say that my own researches have led me again and again to the dawn and the sun as the chief burden of the myths of the Aryan race" (Lectures on the Science of Language, 1864, p. 520).

Müller's solar mythology rapidly began to gain adherents, both in Great Britain and abroad. Perhaps the most important of these was the English classicist George W. Cox (1827–1902), author of The Mythology of the Aryan Nations (1887). Despite his obsession with "pan-Aryanism" and with solar and other natural metaphors, Cox added a new and important dimension to comparative Indo-European mythology through his emphasis upon structural as well as etymological equations. As shall be seen, this prefigured more recent theories about the nature of the Indo-European religious tradition. Another major disciple was the Semitist Robert Brown (b. 1844), who extended the paradigm far beyond the Indo-European domain and used it to explain the ancient Near Eastern divinities as well as those of the Ṛgveda. Two American scholars, John Fiske (1842–1901) and Daniel G. Brinton (1837–1899), also made significant contributions to the literature of solar mythology. In Myths and Mythmakers (1888) Fiske attempted to reconcile the meteorological and solar varieties of naturism, and Brinton, in The Myths of the New World (3d ed., 1896), sought to demonstrate the parallels between North American Indian and Indo-European mythological figures.

It should be pointed out that Müller's was by no means the only naturistic school of comparative Indo-European mythology to flourish in the late nineteenth century. Indeed, the "first paradigm," as it has been termed here, actually included several rather distinct subparadigms, all of which shared essentially the same methodology and basic assumptions. For example, in 1859 Adalbert Kuhn (1812–1881) published his famous Die Herabkunft des Feuers und des Göttertranks, in which thunderstorms and their attendant bolts of lightning, rather than the sun, were conceived to be the prime source of Indo-European (and other) mythological and religious metaphors. Kuhn's most famous onomastic equation, later shown to be totally incorrect, was the assumed etymological connection between Prometheus and the Indian figure Pramantha. Both were seen as archetypal "fire bringers," and Kuhn and his followers were as assiduous in discovering other Indo-European fire gods as Müller and others were in discovering their solar divinities. Another prominent naturist was the Italian philologist Angelo de Gubernatis (1840–1913), who emphasized animal metaphors; thus, where Müller and Kuhn saw the sun and the lightning bolt, Gubernatis saw wild beasts, especially beasts of prey. Still others sought to find lunar and/or stellar metaphors in the Indo-European and other ancient mythological traditions.

Collapse of the first paradigm

While Müller, Kuhn, Cox, and the rest were developing their naturistic models, another scholarly approach to myth and religion per se was quietly taking a shape that would ultimately prove to be the undoing of these models. This approach was fostered by the pioneer anthropologists, such as E. B. Tylor (1832–1917), John Lubbock (1834–1913), and John McLennan (1827–1881), who, as might be expected, came to focus their attention not on the Indo-European tradition, but rather on the vast corpus of data that had come to light relative to the beliefs and practices of contemporary "primitive" peoples. In his Primitive Culture (1871), for example, Tylor laid the foundations for the theory of animism, that is, the notion that all religious beliefs are rooted in the concept of the human soul. The anthropologists were for the most part not trained philologists—although they did, of course, make use of the comparative method in its broadest sense—and therefore were not as attuned to etymologies and the metaphoric significance of names. The result was a profoundly different conception of the origin and evolution of human religious beliefs.

By the late 1880s the naturists and the anthropologists found themselves on a collision course. The anthropological attack was led by a brilliant and iconoclastic Scotsman, Andrew Lang (1844–1912). A sometime disciple of Tylor, Lang set about to destroy naturism in general and the theories of Max Müller in particular. In a series of books, essays, and popular articles he hammered at Müller's assumptions and etymologies, and by the end of the century had effectively demonstrated the weaknesses in the naturistic paradigm so effectively that it did not long survive the death of its chief proponent in 1900.

It would be impossible here to trace all of the thrusts and counterthrusts that marked this famous scholarly debate, but Lang's principal objections can be summed up as follows: (1) Müller's theory—and, by extension, the theories of Kuhn, Cox, Fiske, and the rest—was implicitly based on the fallacious linkage of "degradation" to Original Sin, which, although the chosen people in this instance were the so-called Aryans (i.e., Indo-Europeans) rather than the Jews, was modeled on traditional Judeo-Christian historiography and did not take into account the comparative data from contemporary non-Western cultures; (2) too much emphasis was placed upon language and linguistic processes, especially metaphor and etymology, and too little on the differential effects of the social, cultural, and physical setting wherein myths and religious concepts originated; and (3) there was too much concern with origins and not enough with the historical development of myths and mythmaking, nor was enough attention paid to the universal, evolutionary stages evident in the Indo-European tradition. Needless to say, Müller attempted to answer these charges as best he could, and indeed his criticisms of unilineal evolutionism are remarkably similar to those of later critics. But in the end Lang was triumphant, and solar mythology, together with the other varieties of naturism that had flourished since the middle of the century, went into a permanent eclipse.

Empirical Reaction and Emergence of New Models: 1900–1920

Thus passed the first grand paradigm in comparative Indo-European religious studies. As the new century dawned, the majority of scholars working in the field—classicists, Indologists, Germanists, Celticists, and so forth—rapidly abandoned the naturistic/etymological approach in favor of more intense efforts to explain the various Indo-European religious traditions on their own terms. As in other disciplines at this time, including anthropology, a new spirit of empiricism came to the fore, marked by a growing distrust of comparativism. Most of these specialists, as they may be termed (e.g., the Celticist Joseph Vendryes), relied heavily on the methods of textual criticism, phrasing their analyses in terms of new translations, new specific etymologies, and the like. Indeed, save for the purposes of linguistic reconstruction, the idea of a common Indo-European religious and/or mythological heritage was rarely mentioned in the first two decades of the twentieth century.

At the same time, unrelated, for the most part, to Indo-European studies, several new theoretical models for the study of religion emerged, two of which were to have an important impact on the future development of this discipline. In his massive survey of primitive religion, The Golden Bough (3d ed., 12 vols., 1911–1915), James G. Frazer (1854–1941) came to the conclusion that religion everywhere was rooted in magic, and that all belief systems, including those of the ancient Indo-European-speaking communities, were predicated on a sacrificial ritual wherein a god was killed and replaced so as to renew the world. Among Frazer's prime examples was the death of Baldr, the Apollo-like son of the chief Norse god, Óðinn, who, thanks to the machinations of Loki, was unintentionally killed at the peak of his youthful vigor by his sibling, the blind god Ho̜ðr. Thus, through a form of "sympathetic magic" the gods, and the forces they incarnated, were periodically manipulated so as to keep them perpetually vigorous and fertile. Although largely rejected by subsequent generations of anthropologists, Frazer's influence lingered on in the so-called ritualist school of mythology associated with Jane E. Harrison, Francis M. Cornford, Jessie L. Weston, Gilbert Murray, F. R. S. Raglan, and H. J. Rose, all of whom drew heavily on Greco-Roman beliefs and practices in the formulation of their theories (indeed, most were classicists by academic training).

A second theoretical development occurred in France under the aegis of Émile Durkheim (1858–1917), one of the founding fathers of contemporary social science. In 1903, in collaboration with his principal student and disciple, Marcel Mauss (1872–1950), Durkheim published a short monograph entitled "De quelques formes primitives de classification: Contribution à l'étude des représentations collectives" (Année sociologique 6, 1903, pp. 1–72), which argued that social classification systems are necessarily "collectively represented" in a society's belief systems. This was followed in 1912 by his magnum opus, Les formes élémentaires de la vie religieuse (translated as The Elementary Forms of the Religious Life, 1917), in which he persuasively demonstrated that society itself is the stuff of the divine and that humans necessarily fashion their gods as collective representations of fundamental "social facts." To be sure, Durkheim's prime examples were drawn from the belief systems of the Australian Aborigines, but the implications for the study of religion per se were clear: a new primary source of religious metaphors had been identified, and the immediate implications for the study of Indo-European belief systems were also clearly present from the outset, as Durkheim's ideas themselves were in some measure influenced by the demonstration (1907) of Antoine Meillet (1866–1936) that the Iranian god Mithra (equivalent to the Vedic god Mitra) was the personification of the idea of "contract." Indeed, as shall shortly be demonstrated, Meillet, perhaps the most eminent Indo-European philologist of his time, had more than a little to do with the development of the second grand paradigm in Indo-European studies.

Neo-Comparativists and the Search for a New Paradigm: 1920–1938

Although comparative Indo-European religious studies suffered a marked decline in the generation following Müller's death, the basic questions he and his colleagues had addressed regarding the fundamental similarities among the several ancient Indo-European pantheons remained, and in the early 1920s the pendulum began to swing once again in the direction of what can best be labeled neo-comparativism. For example, Albert Carnoy began to speak in no uncertain terms about a "religion indo-européenne," and shortly thereafter, although they differed widely in inspirations and orientation, a number of German scholars, among them Walter F. Otto, Hermann Güntert, Friedrich Cornelius, and F. R. Schröder, came to the same general conclusion: that it is impossible to understand any single ancient Indo-European religious system without reference to a common set of deities, rituals, and myths, and that it is indeed possible to conceive of such a common Indo-European tradition without reference to the discarded theories of Müller and Kuhn. Another driving force in this new effort was provided by Meillet, who, although he himself never attempted with Indo-European mythological materials the kind of broad synthesis that characterizes his Introduction à l'étude des langues indo-européennes (1922), encouraged his students to undertake such studies. One of these students was Georges Dumézil (1898–1986), a young philologist and historian of religions who took his doctorate under Meillet in 1924.

Like the other neo-comparativists, Dumézil sought to find a viable theoretical basis upon which to build a new paradigm for comparative Indo-European mythology. In his early studies, for example, Le festin d'immortalité (1924), Le crime des Lemniennes (1924), and Le problème des Centaures (1929), which focused on what he came to call the "ambrosia cycle," that is, the common Indo-European traditions surrounding the preparation and consumption of a deified beverage (soma, mead, ambrosia, and so forth), he drew heavily on Frazer's theory of death and rebirth and of the ritual sacrifice of the king. But as he himself later observed, the Frazerian model ultimately proved to be insufficient for his purposes; it simply could not explain the multitude of common motifs that pervaded the several Indo-European traditions.

After a decade of grappling with the problem, Dumézil took an extended leave from his academic duties in the early 1930s and undertook the study of ancient Chinese religion under the guidance of Marcel Granet (1884–1940), an eminent Sinologist who had also been one of Durkheim's most devoted disciples. Yet although the project began as an attempt to gain a perspective on the Indo-European tradition by coming to grips with a wholly different ancient belief system, it ended by providing Dumézil with the framework he had been searching for and that he came to call la méthode sociologique. Thus, in 1938, not long after he had completed his studies with Granet, Dumézil achieved the breakthrough he had been seeking, and the second grand paradigm in Indo-European studies was born.

The Second Grand Paradigm: Dumézil and the New Comparative Mythology

Although the breakthrough itself came in 1938, the first hint of what Dumézil now refers to as the tripartite ideology actually surfaced shortly before he began his Chinese studies. In 1930 he published an article comparing the three divisons of ancient Scythian society—the "Royal Scyths," the "Warrior Scyths," and the "Agricultural Scyths," each of which was believed to have descended from one of the sons of the primeval figure Targitaus (Herodotos, 4.5–4.6)—with the three varṇa s, or classes (later to become full-fledged castes) of Vedic India: the Brāhmas (priests), the Kṣhatriyas (warriors), and the Vaiśhiyas (herders and cultivators, that is, the food producers). He also recognized that the sovereignty of the Royal Scyths was based on the myth that their ancestor, Targitaus's youngest son, had managed to recover three fiery golden objects, a cup, an ax, and a yoked plow, each symbolic of one of the social divisions, that had fallen from the sky, although the full import of this symbolism did not become apparent until the new paradigm had fully crystallized. Two years later, in 1932, the linguist Émile Benveniste arrived independently at a similar conclusion relative to the parallels not only between the Scythian and Indian situations, but also among these two and the social classes of ancient Iran. However, all of the societies concerned belonged to the Indo-Iranian substock, and at the time there seemed to be no reason to conclude that this tripartite hierarchy of priests (or priest-kings), warriors, and cultivators was necessarily pan-Indo-European.

Nevertheless, in the years that followed, Dumézil began to pick up hints of an analogous structure in the Roman tradition (see, for example, Flamen-Brahman, 1935), especially in the makeup of the most ancient of the Roman priestly colleges, the flamines maiores. Could the distinctions between the flamen Dialis, or chief priest of Jupiter, the flamen Martialis, who presided over the cult of the war god Mars, and the flamen Quirinalis, who served the popular divinity Quirinus, an incarnation of the mass of Roman society, reflect the same structure he and Benveniste had discovered in the Indo-Iranian tradition, especially in light of the probable etymological connections between the two terms flamen and brahman ? It was not until he had focused his attention upon the ancient Germanic pantheons in the course of giving a series of lectures at the University of Uppsala in Sweden in the fall of 1938 that he finally came to the realization that this threefold hierarchy was in fact pan-Indo-European, and that it was reflected in both the structure of the pantheons and the structure of society itself, especially in the system of social stratification. And here, of course, his recent exposure to Durkheimian theory in the course of his studies with Granet served him well. The Old Norse gods Óðinn, Thórr, and Freyr reflected the same basic type of social organization, even though the priestly, or Brahmanic, level had long since disappeared as a viable social entity by the time the myths were transcribed by Snorri Sturluson and Saxo Grammaticus. Óðinn (Odin), like Jupiter and the Vedic god Varuṇa, was a collective representation of ultimate sovereignty; Thórr was the incarnation of the warrior stratum and thus was cognate to Mars and Indra; while Freyr (together with his father Njorðr), like Quirinus and the Vedic Asvins ("divine horsemen"), represented the producing classes, that is, the herders and cultivators upon whom the other two classes depended for nourishment.

Dumézil's discovery was in large measure confirmed by his Swedish colleague Stig Wikander's conclusive demonstration that among the most prominent features of ancient Indo-European social organization was the comitatus ("war band"), which typically formed itself around the person of a chief. According to Wikander, the comitatus was mythologically reflected by such otherwise diverse phenomena as Indra and his Marut (i.e., the Rudriyas) and the war bands that followed Irish heroes like Cú Chulainn and Finn (see Der arische Männerbund, 1938). Thus, thanks to Wikander, who became one of Dumézil's earliest and most productive supporters (see below), a major piece of the puzzle had fallen into place.

A preliminary statement of the new model appeared in Les dieux des Germains (1939), which was based on the lectures Dumézil had given in Sweden, and for the next decade the discoveries came thick and fast. Dumézil rapidly came to the conclusion that the sovereign level, shortly to be labeled the "first function," was in fact represented by two complementary divinities: Varuṇa, Jupiter, and Óðinn were primarily concerned with the maintenance of cosmic order (e.g., the Vedic concept of ṛta ), while Mitra, Týr, and the otherwise obscure Roman divinity Dius Fidius were concerned with social and juridical sovereignty. This idea of the "joint sovereignty" formed the major focus of Mitra-Varuna: Essai sur deux représentations indo-européennes de la souveraineté (1940). The first comprehensive statement of the new paradigm appeared a year later in a book entitled Jupiter, Mars, Quirinus (1941). Although Dumézil here focuses on Rome and its mythological origins, this book spelled out in detail for the first time the concept that came to be known as the "three functions" of social organization, that is, the "first function" (cosmic and juridical sovereignty in all its manifestations), the "second function" (the exercise of military prowess), and the "third function" (the provision of nourishment, health, physical well-being, wealth, the welfare of the masses, etc.).

In short, by the end of the 1940s, in a remarkable series of books, monographs, and shorter works, Dumézil had fully articulated the basic elements of the second grand paradigm in comparative Indo-European religious studies. The Iranian and Celtic traditions had been brought into the picture, and a great many secondary themes had been discovered; for example, the recognition that the juridical sovereign (e.g., Mitra and Týr) typically had two ancillary manifestations, each of whom was concerned with an aspect of this function. In the Vedic texts, these were the figures Aryaman and Bhaga, who represented, respectively, the Aryan community itself, along with its most basic social relationship, marriage; and the equitable distribution of goods and rewards. This idea was first enunciated in Le troisième souverain (1949).

In his first articulations of the new paradigm Dumézil had relied heavily on the previously mentioned Durkheimian proposition that "social facts" give rise to "supernatural facts," or "collective representations." However, as he himself observed, around 1950 his orientation began to shift, and he took what amounted to a long step beyond strict Durkheimianism and "la méthode sociologique." Adopting what in retrospect may be called a more structuralist perspective, he began to conceive of the three functions as expressions of a deep-seated, tripartite ideology that was manifest in both social and supernatural contexts, but which ultimately lay outside either sphere. Thus, the functions were gradually redefined as "un moyen d'analyser," a method of analysis, and this revised orientation is, in some respects, not dissimilar to the structuralist vision espoused by Claude Lévi-Strauss (b. 1908). There is, however, a major difference between the two French scholars: Lévi-Strauss (in such works as Le cru et le cuit, 1964) is concerned primarily with the "deep structure" of the human mind per se, while Dumézil remains committed to the proposition that the tripartite ideology is uniquely Indo-European, and that other major language families, such as the Sino-Tibetan, the Hamito-Semitic, and the Uto-Aztecan, are probably characterized by their own unique ideologies. Perhaps the best way to describe this approach is to label it "structural relativism."

In the course of the next three decades more important discoveries were made, not only by Dumézil himself, but also by the scholars who have come to adopt the paradigm. One of the earliest of these was Stig Wikander, who in 1947 demonstrated the extent to which the heroes of the Mahābhārata (Yudhiṣṭhira, Arjuna, Bhīma, Nakula, and Sahadeva) were at bottom transpositions of the major Vedic divinities (Mitra, Varuṇa, Indra, and the Nāsatya) and showed that the tripartite ideology could be detected at the epic as well as the mythological level. Other early followers of Dumézil were Lucien Gerschel, Jan de Vries, Edgar Polomé, Robert Schilling, Jacques Duchesne-Guillemin, François Vian, and Marie-Louise Sjoestedt.

In the late 1950s and early 1960s a new generation of scholars was attracted to the Dumézilian model, including Jaan Puhvel, Donald J. Ward, Françoise Le Roux, and myself, and the paradigm was extended even more broadly. Among the major subthemes discovered by Dumézil and his colleagues over the years, in addition to Wikander's 1947 breakthrough, were (1) the "three sins of the warrior," that is, the recognition that Indo-European warrior figures (e.g., Indra, Herakles, and the Norse figure Starkaðr) typically commit three canonical "sins," one against each of the functions, and (2) the "war between the functions," manifested principally in the Roman and Germanic traditions, wherein representatives of the first two functions defeat representatives of the third and incorporate them into the system, rendering it complete (e.g., the Sabine war and the conflict between the Æsir and Vanir).

In the early 1970s Dumézil pushed the paradigm in yet another important direction (see Mythe et épopée, vol. 2, 1971, especially "L'enjeu du jeu des dieux: Un héros") through his discovery that Indo-European warrior figures such as the Vedic character Śiśupāla are in the final analysis but counters in a game played by the gods, and that the gods themselves can be sorted into "dark" and "light" categories—that is, those who represent the chaotic forces of nature and those who seek to control these forces. In the Indian tradition this dichotomy is reflected in the difference between the "dark" divinity Rudra and the "light" divinity Viṣṇu; in ancient Scandinavia it appears in that between Óðinn and Thórr. The full implications of this discovery are being probed by several of Dumézil's disciples, among them Udo Strutynski and the author of this article.

In the course of what may be termed his phase de bilan, Dumézil's remarkable scholarly output continued unabated. Among his subsequent books were a reexamination of the Indo-European concept of sovereignty (Les dieux souverains des Indo-Européens, 1977) and a disquisition on Indo-European attitudes toward marriage (Mariages indo-européens, 1979). He also published several collections of earlier writings, all of which bear on one or another aspect of the tripartite ideology. Dumézil's career was capped in 1979 when he was elected to the Académie Française.

This is not to imply that the "new comparative mythology" has become universally accepted by Indo-Europeanists. Indeed, almost from the outset it has been the subject of intense and persistent criticism from a variety of scholars, many of whom have suggested that Dumézil imposed the tripartite model on the data, and that it has no existence save in the minds of the researchers concerned. Among the most persistent of these critics was Paul Thieme, an Indologist, who asserted on numerous occasions that Dumézil's interpretation of the Indic pantheon, especially the role played by the god Aryaman, was wholly incorrect. Thieme interpreted the Sanskrit root ari- to mean "stranger" rather than "the people" (or "the shining ones"), the common meaning of most ethnic self-identification terms, modern as well as ancient—for example, Hopi, Diné (Navajo), and so forth. Other prominent critics have included H. J. Rose (who took Dumézil to task for ignoring the "manaistic" basis of Roman religion), Jan Gonda, Angelo Brelich, the Germanist E. A. Philippson, and John Brough, a Sanskrit scholar who claimed to have discovered the tripartite ideology in the Bible and therefore asserted that it was not uniquely Indo-European.

Dumézil vigorously responded to these and other criticisms, and to date no single critic has emerged as a potential "Andrew Lang" as far as this paradigm is concerned. Indeed, it is fair to say that the majority of contemporary scholars in the field of comparative Indo-European mythology and religion continue to make effective use of the general theoretical and methodological framework developed by Dumézil and his colleagues in the course of the last five decades. A good example is Joël Grisward, whose brilliant analysis of the medieval French legends of Aymeri de Narbonne and the extent to which they have Indo-Iranian counterparts (see his Archéologie de l'épopée médiévale, 1981) is, as Dumézil himself noted, perhaps the most important contribution to the new comparative mythology since Wikander's discovery of epical transposition in 1947. Another excellent example can be seen in Udo Strutynski's convincing demonstration that the English weekday names, at least from Tuesday through Friday, and their cognates in other modern Germanic languages, represent a persistence of a tripartite ideological formula—that is, "Týr's day," "Óðinn's day," "Thórr's day," and "Frigg's day" (see his "Germanic Divinities in Weekday Names," Journal of Indo-European Studies 3, 1975, pp. 363–384).

It would be impossible in the space of this brief article even to mention, let alone discuss in any detail, all of the significant research that has been pursued since the late 1960s by specialists in comparative Indo-European religion and mythology who have oriented their work around the Dumézilian paradigm. For example, Atsuhiko Yoshida, a Japanese Hellenist who studied with Dumézil for the better part of a decade, has demonstrated the strong probability that the development of Japanese mythology was profoundly influenced, either directly or indirectly, by Indo-European themes in the late prehistoric period (that is, the fourth and fifth centuries ce), and that the most likely source of this influence was one or another tribe of North Iranian-speaking steppe nomads (Scythians, Alans, etc.) that managed to reach East Asia during this period (Yoshida, 1977). The late Ōbayashi Taryō, an anthropologist at the University of Tokyo, and the author of this article subsequently joined Yoshida in this effort. Bruce Lincoln has published a book comparing Indo-Iranian and contemporary East African religious attitudes toward cattle (Priests, Warriors, and Cattle, 1981). David Cohen has expanded the understanding of the "three sins" typically committed by the Indo-European warrior (see above) in a penetrating analysis of the Irish hero Suibhne ("Suibhne Geilt," Celtica 12, 1977, pp. 113–124).

In France, Daniel Dubuisson, who took his doctorate under Dumézil in 1983, has attempted to develop a quasi-mathematical approach to Indo-European myth, based in large part on his Indological research. More recently, in 1994, he published a major overview of the current status of comparative Indo-European mythology. Bernard Sergent has illuminated the dual kingship at Sparta by judicious application of the Dumézilian paradigm ("La représentation spariate de la royauté," Revue de l'histoire des religions 189, 1976, pp. 3–52). And in 1984 Dean A. Miller investigated the trifunctional implications of the "three kings" in Sophocles' Oedipus at Colonus from what can best be termed a neo-Dumézilian standpoint. Other scholars who have extended the paradigm in a variety of new and potentially important directions include Steven O'Brien, Miriam Robbins, Alf Hiltebeitel, David B. Evans, and Jean-Claude Rivière.

Some Recent Developments

Like all grand paradigms that have been pushed to their effective limits, the Dumézilian paradigm is fraying a bit at the edges, and several of the most important recent advances in Indo-European religious studies have involved matters that transcend the tripartite ideology. One of these is the matter of "dark" and "light" divinities mentioned earlier. Indeed, Dumézil himself suggested that this dichotomy cuts across the three functions, and perhaps reflects a more fundamental binary structure that underlies social and supernatural tripartition. If this proves to be the case, it may well be that the ideological model Dumézil first detected some fifty-odd years ago is but a special case of a broader and more deep-seated mental template, as it were, that is shared by homo religiosus as a whole. Such a template, if it exists, would closely parallel the presumably universal "deep structure" of the human psyche posited by Lévi-Strauss.

Another extremely significant development involves the nature of the common Indo-European cosmology, something Dumézil never really came to grips with and which, heretofore, had defied all attempts at elucidation via the tripartite ideology. In 1975, Puhvel and Lincoln, working independently, reached compatible conclusions; they agreed that the elusive cosmology was in fact embedded in a theme, present in the Roman, Indo-Iranian, and Norse traditions, wherein a primeval being kills his twin and makes the world from the latter's remains. This theme closely approximates the nearly universal concept of what Adolf E. Jensen calls the "dema deity," that is, a sacrificial victim whose body parts provide the materia prima of either the world itself or some important part thereof (as in the Ceramese myth of Hainuwele; see Jensen, Myth and Cult among Primitive Peoples, 1963).

For Puhvel, the point of departure was the pseudo-historical account of Romulus and Remus, in which the latter is killed shortly after the founding of Rome. Underlying the names Romulus and Remus, Puhvel suggests, are *Wironos ("man") and *Yemo(no)s ("twin"), to which may be compared Yama (Skt.), Yima (Av.), and Ymir (ON), as well as Mannus and Tuisto, mentioned in Tacitus's Germania. Although Romulus/*Wironos did not explicitly "make the world" from Remus/*Yemos's remains, Remus's death seems clearly to have been somehow essential to the building of the city, like a sacrificial offering, and the fact that Remus's "crime" consists of jumping over the newly dug foundation for the city wall implies that the victim's essence was in one way or another mixed with the mortar that eventually filled the ditch. Lincoln's point of departure was the Indic manifestation of the theme and its implications as they relate to the dema- deity concept, that is, the account in Ṛgveda 10.90 wherein Manu (i.e., "man") sacrifices Yama (or Puruṣa, as he is called in the Vedic text) and creates the world from his corpse. (Unlike Remus/*Yemos, Puruṣa was a willing victim, and Manu is credited with originating the institution of religious sacrifice; however, the basic context of the two accounts is remarkably similar.) As luck would have it, Lincoln sent a draft of his manuscript to Puhvel for comment and criticism, and the result was a pair of seminal articles that in 1975 appeared back-to-back in History of Religions.

The paradigmatic implications of this discovery are still under investigation, and various questions have been raised by scholars. Does the ideology itself spring from this primordial sacrifice? Is it possible that the account of Romulus and Remus, who began life as the foster children of a shepherd, became warriors, and finally went off to found a city, is a euhemeristic survival of an ontological myth wherein the three functions emerge successively after a primeval fratricide? And is there a connection between the dramatis personae of this primeval drama and the dark/light dichotomy (see above)? Or does the theme in question lie totally outside the parameters of the paradigm? As yet no clear answers have been provided to these questions.

Yet another extremely significant discovery relating to Indo-European religion was N. J. Allen's 1987 compelling case for the existence of a "fourth" ideological function (or "F4," as he labels it) that lies outside the tripartite paradigm per se and can thus be described as "other." A good exmaple of a "fourth function" phenomenon in the Indic tradfition the Shurdra caste, that is, the non-Aryan outisders, who are "other" to the three twice-born Aryan varna (that is, the Brāhmaṇs, Kṣhatriyas and Vaiśhiyas, who reflect "F1," "F2" and "F3," respectively). This concept adds a new and extremely important dimesion to the Dumézilian paradigm, the implications of which are only just beginning to be appreciated.

The common denominators among the the several Indo-European epic traditions have also been the subject of some important recent research. For example, Dean A. Miller's book The Epic Hero (2000) has materially advanced the understanding of this most important aspect of the Indo-European worldview, while Julian Baldick's (1994) convincing demonstration that the Iliad is fundamentally cognate not to the Mahābhārata, as has long been held, but rather to the Rāmāyana, has led some scholars, including the author of this article, to the conclusion that the two epics in question are, at bottom, reflexes of a common Indo-European concern with abducted brides and their rescurers. Other reflexes of this concern may include the medieval European tale of Tristan and Isolde and the Middle High German Kudrun epic, both of which involve figures broadly similar to Helen, Sītā, Rāma, Menelaus, Agamamnon, and the rest. Moreover, Baldick has gone on to suggest that the Mahābhārata is cognate to the Odyssey, in that they both involve accounts of exiled kings who eventually return to reclaim their thrones after a climactic battle (e.g., the Battle of Kurukshetra in the Mahābhārata, in which the Pāndava defeat their enemies and restore Yudhisthira to his rightful throne, and Homer's account of the slaying of Penelope's suitors by Odysseus and Telemachus).

Finally, the importance of binarism in the Indo-European tradition, that is, the all-pervasive difference between "light" and "dark," which, as has been noted, Dumézil came to recognize late in his career, is something that may link it to far older Nostratic and even, perhaps, Eurasiatic traditions dating as far back as the Upper Paleolithic (that is, prior to 10,000 bce), has begun to be recognized (see Littleton 2002). Indo-European religion is thus beginning to be grounded in the broader context of the Eurasian tradition, which took shape in Central Asia millennia before anything identificable as Proto-Indo-European appeared on the scene.

In sum, as the field of Indo-European religious studies enters its third century it remains a vigorous and intellectually viable discipline. In the course of the last two hundred-plus years it has managed to develop and then transcend one grand paradigm (naturism) and is currently dominated by a second (the new comparative mythology). How long this second paradigm will continue to reign is uncertain; as has been indicated, there are already signs that it may have begun to outlive its usefulness. But whatever may be the ultimate fate of the Dumézilian model—and one suspects that it will eventually become a "special case" of a much broader paradigm, the outlines of which cannot yet be clearly perceived, although Allen's aforementioned seminal discovery of a fouth function does provide a glimpse of what may lie ahead—the discipline itself will almost certainly persevere, and will continue to contribute important insights not only into a fundamental aspect of the heritage shared by all Indo-European speakers, but also into the nature of religion per se.

See Also

Comparative Mythology.

Bibliography

Allen, N. J. "The Ideology of the Indo-Europeans: Dumézil's Theory and the Idea of a Fourth Function." International Journal of Moral and Social Studies 2 (1987): 23–39. A significant discovery that adds an important new dimension to the study of Indo-European religion.

Baldick, Julian. Homer and the Indo-Europeans: Comparing Mythologies. London, 1994. The author contends that the Iliad is cognate to the Rāmāyana in that both reflect an "abducted bride" mythologem, while the Mahābhārata is cognate to the Odyssey. An important reassessment.

Dorson, Richard. "The Eclipse of Solar Mythology." In Myth: A Symposium, edited by Thomas A. Sebeok, pp. 25–63. Bloomington, Ind., 1965. The definitive study of the Müller-Lang controversy.

Dubuisson, Daniel. Mythologie du xxeme siecle (Dumézil, Lévi-Strauss, Eliade). Lille, 1993. An important overview of contemporary comparative Indo-European mythology and religion by one of Dumézil's chief students.

Dumézil, Georges. L'idéologie tripartie des Indo-Européens. Brussels, 1958. Remains the most succinct overview of Dumézil's thesis.

Dumézil, Georges. Camillus. Berkeley, Calif., 1980. A translation by Annette Aronowicz and Josette Bryson of the "Camillus" sections from Mythe et épopée, vol. 3 (Paris, 1973) and related passages from Fêtes romaines d'été et d'automne (Paris, 1975). Includes a definitive introduction by Udo Strutynski.

Dumézil, Georges. The Stakes of the Warrior. Berkeley, 1983. A translation by David Weeks of the "L'enjeu du jeu des dieux: Un héros" section of Mythe et épopée, vol. 2 (Paris, 1971). Includes a masterful introduction by Jaan Puhvel, which puts the "dark/light" dichotomy into its proper perspective.

Feldman, Burton, and Robert D. Richardson, eds. The Rise of Modern Mythology, 1680–1860. Bloomington, Ind., 1972. A comprehensive anthology of the major eighteenth- and early nineteenth-century contributions to comparative mythology, from Vico and Fontenelle to F. Max Müller.

Grisward, Joël H. Archéologie de l'épopée médiévale. Paris, 1981. A brilliant application of the tripartite model to a medieval French epic, the saga of Aymeri de Narbonne.

Larson, Gerald James, C. Scott Littleton, and Jaan Puhvel, eds. Myth in Indo-European Antiquity. Berkeley, 1974. A symposium on various aspects of the Dumézilian model and related subjects. Includes papers by Littleton, Puhvel, Strutynski, David Evans, Mary R. Gerstein, Steven E. Greenebaum, Edgar Polomé, Marija Gimbutas, Jeannine Talley, Matthias Vereno, and an essay by Dumézil entitled "'Le Borgne' and 'Le Manchot': The State of the Problem." This essay concerns yet another Indo-European subtheme: the loss of an eye and a hand, respectively, by the cosmic and juridical representatives of the first function, for example, Óðinn, who gives up an eye in exchange for wisdom, and Týr, who loses a hand while swearing what amounts to a false oath, as well as the Roman figures Horatius Cocles, who is one-eyed, and Mucius Scaevola, who also loses a hand while swearing falsely.

Lincoln, Bruce. "The Indo-European Myth of Creation." History of Religions 15 (1975): 121–145. A seminal contribution to the understanding of the Indo-European cosmogonic myth; see the article by Puhvel listed below.

Littleton, C. Scott. The New Comparative Mythology: An Anthropological Assessment of the Theories of Georges Dumézil. 3d ed. Berkeley, 1980. A comprehensive review of the origins and current state of the "second paradigm," that is, the Dumézilian model. Includes a discussion of the major criticisms that have been directed against the model, as well as an essay comparing Dumézil and Lévi-Strauss, and an extensive bibliography of works by Dumézil and other contributors to the new comparative mythology.

Littleton, C. Scott. "Gods, Myths and Structures: Dumézil." Encyclopedia of Continental Philosophy, edited by Simon Glendinning, pp. 558–568. Edinburgh, 1999. A recent overview.

Littleton, C. Scott. "The Binary 'Spine' of Dumézil's Tripartite Indo-European Ideology: A Pan-Nostratic/Eurasiatic Feature?" Cosmos 14 (2001): 69–84. Discusses the extent to which Indo-European binarism has deep Eurasian roots.

Meillet, Antoine, "Le dieu indo-iranien Mitra." Journal asiatique 9 (1907): 143–159. A seminal article on the Vedic god Mithra and the extent to which he is a "collective representation" of the idea of "contract"; had an impact on Durkheim and later on the development of Dumézil's theory.

Miller, Dean A. The Epic Hero. Baltimore, 2000. An important contribution to the understanding of the Indo-European warrior figure as manifested in epics.

Puhvel, Jaan. "Remus et frater." History of Religions 15 (1975): 146–157. Reprinted in Puhvel's Analecta Indoeuropaea (Innsbruck, 1981), pp. 300–311. Together with the article by Lincoln listed above, this paper probes the Indo-European cosmogonic myth and concludes that it is based on a primeval sacrifice of "Twin" by "Man." An extremely significant contribution to the new comparative mythology.

Puhvel, Jaan, ed. Myth and Law among the Indo-Europeans. Berkeley, 1970. A symposium on the new comparative mythology. Includes papers by Puhvel, Littleton, Strutynski, Donald Ward, Jacques Duchesne-Guillemin, Edgar Polomé, Calvert Watkins, James L. Sauvé, Robert L. Fisher, and Stephen P. Schwartz.

Rivière, Jean-Claude, ed. Georges Dumézil à la découverte des Indo-Européens. Paris, 1979. A symposium marking Dumézil's election to the Académie Française in 1979. Includes essays by several of his colleagues and former students, including Rivière, Robert Schilling, François-Xavier Dillmann, Jean Varenne, Joël Grisward, Georges Charachidzé, and Alain de Benoist, editor of Nouvelle école. Original versions of some of these essays were published, together with other materials, in a 1973 issue of Nouvelle école devoted to Dumézil.

Vries, Jan de. Perspectives in the History of Religions. Translated by Kees W. Bolle. Berkeley, 1977. A succinct survey of the history of religious and mythological thought from classical antiquity to modern times. De Vries was an early disciple of Dumézil, and he includes an interesting section on his theories (pp. 182–186).

Ward, Donald. The Divine Twins: An Indo-European Myth in Germanic Tradition. Berkeley, 1968. An important contribution to the study of the "third function" and the role played in dioscurism in the Indo-European ideology.

Wikander, Stig. "Pandava-sagan och Mahabharata's mytiska förutsättningar." Religion och Bibel (Lund) 6 (1947): 27–39. Demonstrates that the heroes of the Mahābhārata reflect the tripartite ideology and were derived from the Vedic divinities. Wikander's essay was a major step forward in the development of the new comparative mythology.

Yoshida, Atsuhiko. "Japanese Mythology and the Indo-European Trifunctional System." Diogenes 98 (1977): 93–116. Summarizes the evidence suggesting that Japanese mythology, as expressed in the Kojiki (712 ce) and the Nihonshoki (720 ce), was influenced by the tripartite Indo-European ideology at some point in the late prehistoric period.

C. Scott Littleton (1987 and 2005)