PREANIMISM . In the years around 1900, the scholarly debate about the origins and evolution of religion was still in large measure dominated by the theories put forward by E. B. Tylor thirty years previously, notably in his Primitive Culture (London, 1871). The key concept was animism, which denoted both a primitive belief in spiritual beings and a belief in the "animation" of nonhuman beings—from the higher mammals down to trees, plants, and stones—by spirits or spirit forces. By 1900, however, Tylor's theory had been challenged by two of his Oxford disciples, both of whom were and remained his personal friends. In his Cock Lane and Common Sense (London, 1894) and definitively in his celebrated The Making of Religion (London, 1898), Andrew Lang had questioned the animistic hypothesis from one direction, suggesting that "perhaps there is no savage race so lowly endowed, that it does not possess, in addition to a world of 'spirits,' something that answers to the conception of God" (Cock Lane and Common Sense, p. 334). At a meeting of the British Association in 1899, the animistic theory was questioned from another direction, this time by the philosopher-anthropologist R. R. Marett. Whereas Lang was saying that adherents of the animistic theory had been prevented by their presuppositions from even noticing the evidence in favor of what he called "high gods" among peoples on a low level of material development, Marett claimed that the term animism was ambiguous and that the mental processes it assumed were too sophisticated to have been present at the lowest level of human evolution.
Marett's paper "Pre-Animistic Religion" was first published in the journal Folk-Lore (June 1900, pp. 162–182); it subsequently formed the first chapter of his book The Threshold of Religion (London, 1909; 2d exp. ed., London, 1914). Although brief, its argument was revolutionary. On the one hand, it suggested that in view of the double meaning of the word animism in Tylor's Primitive Culture, a distinction might be drawn between animism proper, as a belief in spiritual beings, and the belief in the "animation" of animals, plants, and natural objects, which he proposed to call "animatism." This of course had nothing to do with any theory of the origin of religion as such, but was merely a plea for greater terminological precision. On the other hand—and this appeared to be an outright challenge to the Tylorian hypothesis—Marett also ventured the opinion that animism was simply not "primitive" enough to represent the earliest form of religion. Beneath (though not necessarily chronologically prior to) the belief in spirits, he argued, there is a more amorphous sense of the world as being filled with the manifestations of supernatural power. This notion was unlikely to have been reasoned out in the first instance; rather it involved a "basic feeling of awe, which drives a man, ere he can think or theorize upon it, into personal relations with the supernatural" (Marett, 1914, p. 15). In search of a word to characterize this power, Marett settled finally upon the Melanesian word mana, as described by the missionary R. H. Codrington in his book The Melanesians (Oxford, 1891). Mentioned only in passing in his 1899 paper, alongside other "power-words," over the next few years mana came to eclipse the others as a terminus technicus to describe what lay at the root of preanimism.
Mana, however, was by no means an exclusively Melanesian concept. It was common to the whole of the Pacific, to Polynesia as well as Melanesia. It had been first noted by Captain James Cook in 1777 and long before Codrington's time had been fairly fully discussed in relation to the Maori of New Zealand. F. E. Maning in his book Old New Zealand (Auckland, 1863) had stressed, for instance, that mana had no single meaning but was associated with such diverse ideas as "virtue, prestige, authority, good fortune, influence, sanctity, luck" (Maning,  1927, pp. 239–240). However, the early preanimists remained generally unaware of the New Zealand material and were content to rely for the most part on Codrington's evidence as transmitted first by Marett and subsequently by the German and French sociologists.
Marett himself was most unwilling to "dogmatize" about religious origins and always expressed himself with great caution. Thus although in his 1899 paper he went so far as to suggest that what he there called "supernaturalism" might be "not only logically but also in some sense chronologically prior to animism" (Marett, 1914, p. 11), he did not say in what sense. Again—and this is important in view of the direction subsequently taken by the debate—he did not categorize mana as unambiguously impersonal. In a later paper, in fact, he stated explicitly that mana "leaves in solution the distinction between personal and impersonal" (1915, p. 119) and noted that although it may in some circumstances be used in a somewhat impersonal way, it is always necessary to take account of "the ambiguity that lies sleeping in mana " (p. 121). Other writers on the subject found this degree of ambiguity unmanageable and unwelcome.
In the wake of Marett's work, the first decade of the twentieth century saw the appearance of a great deal of writing on the subject of preanimism and on mana and its various equivalents. In Germany, Wilhelm Wundt of Leipzig wrote extensively in his Völkerpsychologie (1900) about "die präanimistische Hypothese," followed by K. T. Preuss in a series of articles in the journal Globus (1904–1905). Both, however, seem to have assumed Marett's theory to have been conceived in direct and complete opposition to Tylor—a charge that Marett, who admired Tylor greatly, strenuously denied. In France, the Année sociologique school (which included Durkheim, Hubert, and Mauss) produced a theory very similar to Marett's, perhaps independently, though Hubert and Mauss's article "Esquisse d'une théorie générale de la Magie" appeared in Année sociologique only in 1904, and Durkheim's magnum opus did not appear until 1912.
By this time, however, mana had been coupled with a bewildering variety of terms drawn from primal cultures in various parts of the world, all of which, it was claimed, conveyed the same basic sense of that supernatural power that had inspired an initial human response of awe. A proportion of these words had been culled from the vocabularies of various Amerindian peoples: From the Iroquois came orenda (as in Hewitt's "Orenda and a Definition of Religion," American Anthropologist, n.s. 4, 1902), from the Algonquin manitou, and from the Lakota wakan and wakanda. The Australian Aranda (Arunta) term arungquiltha/arúnkulta, the Malagasy andriamanitra, the Fijian kalou, and even the Old Norse hamingja and the Hindu brahman were added to the list, which by 1914 had assumed considerable proportions. Mana, however, continued to serve as the flagship of the preanimistic fleet.
It is important to remember that Marett had stated (not in his original article but at the Oxford Congress of the Science of Religion in 1908) that it was by now his express intention to endow mana with "classificatory authority to some extent at the expense of the older notion [i.e., animism]" (Marett, 1915, p. 102). Every new science had to create its own specialist terminology; this being so, Marett was proposing the use of mana whenever and wherever circumstances appeared to warrant it as a technical term expressive of preanimistic religions and virtually independent of the etymological meaning of the word in its original Pacific context. In the light of Marett's express intention, it is slightly embarrassing to note the solemnity with which some scholars have subsequently believed themselves to be demolishing Marett's argument by pointing out that the etymology of mana is not altogether what he supposed it to be.
Another critical point concerns the supposed impersonality of the power of mana. As has been seen, Marett was initially insistent that mana is an ambiguous concept, even as he knew perfectly well that his chief informant Codrington had stated that it was always associated with and derived from persons, spirits, or ghosts. On at least one later occasion, however, in his article "Mana" in Hastings's Encyclopaedia of Religion and Ethics (vol. 8, Edinburgh, 1915), he was prepared to state that mana was "in itself impersonal" while always associated with personal beings. (Often in such contexts he used the analogy of electricity, which remains latent until tapped and channeled.) The ambiguity between personal and impersonal remained in force nonetheless. But just as Marett read mana through the prism of Codrington, one feels that almost all later debaters have read Marett through the prism of the greater international celebrity Émile Durkheim.
To Durkheim, writing in The Elementary Forms of the Religious Life, first published in French in 1912 and in English in 1915, there were no ambiguities. Caution was replaced by assertion. According to Durkheim, Marett had shown "the existence of a religious phase which he called preanimistic, in which the rites are addressed to impersonal forces like the Melanesian mana and the wakan of the Omaha and Dakota" (1968 edition, p. 201). Durkheim categorically stated that mana was "an impersonal religious force" (pp. 192, 198), "an anonymous and diffused force" (p. 194); because it was not, according to Codrington, a supreme being, Durkheim concludes that it must possess "impersonality" (p. 194). One need look no further for the later impression that preanimism must of necessity involve belief in impersonal forces; it comes not from Marett but from Durkheim.
The preanimistic theory of the origin of religion (as it had developed between 1900 and 1914) first began to be called in question in the years following World War I. In 1914 Nathan Söderblom (who had been a professor in Leipzig from 1912 to 1914) published in the Archiv für Religionswissenschaft an article, "Über den Zusammenhang höherer Gottesideen mit primitiven Vorstellungen," in which the customary preanimistic points were discussed (see also Söderblom, Gudstrons uppkomst, Stockholm, 1914, pp. 30–108). One of his students, F. R. Lehmann, was inspired by this article to take up the question of mana and in 1915 presented his dissertation on the subject, in which he penetrated beyond Durkheim and Marett to Codrington, and beyond Codrington to the etymology and implications of the common Polynesian/Melanesian word mana itself.
Lehmann's researches had the effect of discrediting altogether the notion that the term mana had ever been used in the Pacific region to denote an impersonal force. Even when trees, stones, or other inanimate objects were declared to possess mana, this was because spirits had associated themselves with those objects, and not by virtue of their having an impersonal force of their own. Paul Radin had made substantially the same point in 1914, when he asked, "What warrant have we for thinking of the god as a deity plus power, and not merely as a powerful deity? Are we not committing the old error of confusing an adjective with a noun?" (Journal of American Folklore 27, 1914, p. 347). Following Lehmann, and in the increasingly antievolutionary atmosphere of the interwar years and beyond, more and more frequent criticisms were leveled against the preanimistic hypothesis, the interpretation of mana that had supported it, and against those who had written in these terms. A powerful broadside against the theory was produced by the Germanist Walter Baetke, in his book Das Heilige im Germanischen (Tübingen, 1942), and another by Geo Widengren, in a polemical article, "Evolutionism and the Problem of the Origin of Religion" (Ethnos 10, 1945, pp. 57–96). Widengren, incidentally, admired Baetke's work; and it was in the Baetke festschrift that Lehmann described the course of his research in the area of mana, in an article called "Versuche, die Bedeutung des Wortes 'Mana'… festzustellen" (pp. 215–240). Widengren summed up: "The best experts in the field of Melanesian religion have explicitly stated that mana is actually never an impersonal power"; it is "in reality a quality. It goes without saying that not mana in itself but persons and things possessing mana are the objects of worship" (p. 84). One last critic may be quoted. In his 1958 Patterns in Comparative Religion and in virtually identical terms in his 1968 Myths, Dreams and Mysteries, Mircea Eliade denies the existence of any such "impersonal and universal force" as mana was once thought to represent, not least because "impersonality" is "without meaning within the archaic spiritual horizon" (Eliade, 1968, p. 129). All these critics, however, have tended to attribute to Marett extreme opinions that were actually those of Durkheim.
It remains to be noted that Rudolf Otto, in his celebrated book Das Heilige, produced a theory of the origin of religion in an ineffable sensus numinis, in the course of which he praised Marett for coming "within a hair's breadth" of his own views. Otto, too, was criticized by Baetke and Widengren, who used arguments very similar to those they had used against Marett and the preanimists. Otto's numen could hardly be called "impersonal," however.
Preanimism and the debate about preanimistic religion belong less to the world of religion as such (and hardly, it would seem, to the area of primal religion at all) than to the intellectual history of the early twentieth century in the West. Possibly the popularity of the concept was not unrelated to the West's growing estrangement from fixed forms of religious belief and doctrine and its simultaneous maintenance of a sense that there might be "something" (rather than "someone") in charge of the world's destiny. It involved the evolutionists' conviction that religion had emerged out of something other than, and simpler than, religion. It also made assumptions about personality and (at least after Durkheim) impersonality that later critics found it all too easy to demolish. The critics, however, may have gone too far in the opposite direction. In their desire to disassociate themselves from the evolutionists, they have frequently misrepresented and misinterpreted them, without realizing that the evolutionists themselves were quite capable of raising objections—often the same objections—to their own work. Preanimism as such can be neither proved nor disproved as a rudimentary stage in the evolution of religion. There may, however, remain an area of religion within which supernatural (or at least uncontrollable) power is sensed, while remaining inchoate and unconnected with any firm notion of deity. This need not be a stage out of which more precise notions emerge. It is just as likely to be found at the end of a long process of decline, and thus to be as much posttheist as preanimist. There is no word that can be used as a technical term to describe this. Preanimism clearly will not do, because of the implicit sequence involved. Some use might however still be found for the term mana in this connection. In 1907 Marett wrote that "the last word about mana has not been said" (p. 219). By 1965, mana had almost been dismissed from the technical vocabulary of the study of religion. It may be high time for its reexamination.
References to the "preanimistic hypothesis" will be found scattered throughout the anthropological literature of the first half of the twentieth century. The seminal articles are gathered in R. R. Marett's The Threshold of Religion, 3d ed. (London, 1915), and Émile Durkheim's application of the theory is found in The Elementary Forms of the Religious Life, translated by Joseph Ward Swain (1915; reprint, New York, 1965), in which see especially pages 191–204. For the subsequent attempted demolition of the theory, reference must be made to the German works of F. R. Lehmann, beginning with Mana: Der begriff des "ausserordentlich wirkungsvollen" bei Südseevölkern (Leipzig, 1922) and ending with his essay "Versuche, die Bedeutung des Wortes 'Mana' … festzustellen," in Festschrift Walter Baetke, edited by Kurt Rudolph et al. (Weimar, 1966), pp. 215–240; and Walter Baetke's Das Heilige im Germanischen (Tübingen, 1942). See also Geo Widengren's "Evolutionism and the Problem of the Origin of Religion," Ethnos 10 (1945): 57–96, which follows substantially the same line.
Eric J. Sharpe (1987)