painful path of conversion. The heretic, having developed powers of rational choice, perceives his heresy, to wit, his want of adaptation to the moral environment, and turning round embraces the new faith that is the passport to survival.
Far otherwise is it with man at the stage of savagery—the stage of petty groups pursuing a self-centred life of inveterate custom, in an isolation almost as complete as if they were marooned on separate atolls of the ocean. Progress, or at all events change, does indeed take place, though very slowly, since the most primitive savage we know of has his portion of human intelligence, looks after and before, nay, in regard to the pressing needs of every day shows a quite remarkable shrewdness and resource. Speaking generally, however, we must pronounce him unprogressive, since, on the whole, unreflective in regard to his ends. It is the price that must be paid for social discreteness and incoherency. And the consequence of this atomism is not what a careless thinker might be led to assume, extreme diversity, but, on the contrary, extreme homogeneity of culture. It has been found unworkable, for instance, to classify the religions of really primitive peoples under a plurality of heads, as becomes necessary the moment that the presence of a distinctive basis of linked ideas testifies to the individuality of this or that type of higher creed. Primitive religions are like so many similar beads on a string; and the concern of the student of comparative religion is at this stage mainly with the nature of the string, to wit, the common conditions of soul and society that make, say, totemism, or taboo, very much the same thing all the savage world over, when we seek to penetrate to its essence.
This fundamental homogeneity of primitive culture, however, must not be made the excuse for a treatment at the hands of psychology and sociology that dispenses with the study of details and trusts to an a priori method. By all means let universal characterization be attempted—we are about to attempt one here, though well aware of the difficulty in the present state of our knowledge—but they must at least model themselves on the composite photograph rather than the impressionist sketch. An enormous mass of material, mostly quite in the raw, awaits reduction to order on the part of anthropological theorists, as yet a small and ill-supported body of enthusiasts. Under these circumstances it would be premature to expect agreement as to results. In regard to method, however, there is little difference of opinion. Thus, whereas the popular writer abounds in wide generalizations on the subject of primitive humanity, the expert has hitherto for the most part deliberately restricted himself to departmental investigations. Religion, for example, seems altogether too vast a theme for him to embark on, and he usually prefers to deal with some single element or aspect. Again, origins attract the littérateur; he revels in describing the transition from the pre-religious to the religious era. But the expert, confining his attention to the known savage, finds him already religious, nay, encumbered with religious survivals of all kinds; for him, then, it suffices to describe things as they now are, or as they were in the comparatively recent fore-time. Lastly, there are many who, being competent in some other branch of science, but having small acquaintance with the scientific study of human culture, are inclined to explain primitive ideas and institutions from without, namely by reference to various external conditions of the mental life of peoples, such as race, climate, food-supply and so on. The anthropological expert, on the other hand, insists on making the primitive point of view itself the be-all and end-all of his investigations. The inwardness of savage religion—the meaning it has for those who practise it—constitutes its essence and meaning likewise for him, who after all is a man and a brother, not one who stands really outside.
In what follows, then, we shall, indeed, venture to present a wholesale appreciation of the religious idea as it is for primitive man in general; but our account will respect the modern anthropological method that bids the student keep closely to the actualities of the religious experience of savages, as it can with reasonable accuracy be gathered from what they do and say. We have sought to render only the spirit of primitive religion, keeping clear both of technicalities and of departmental investigations. These are left to the separate articles bearing on the subject. There the reader will find the most solid results of recent anthropological research. Here is he merely offered a flimsy thread that, we hope, may guide him through the maze of facts, but alas! is only too likely to break off short in his hand.
Definition of Primitive Religion.—In dealing with a development of culture that has no immutable essence, but is intrinsically fluid and changing, definition must consist either in a definition of type, which indicates prevalence of relevant resemblance as between specimens more or less divergent, or in exterior definition, which delimits the field of inquiry by laying down within what extreme limits this divergence holds. Amongst the numberless definitions of religion that have been suggested, those that have been most frequently adopted for working purposes by anthropologists are Tylor’s and Frazer’s. Dr E. B. Tylor in Primitive Culture (1), i. 424, proposes as a “minimum definition” of religion “the belief in spiritual beings.” Objections to this definition on the score of incompleteness are, firstly, that, besides belief, practice must be reckoned with (since, as Dr W. Robertson Smith has made clear in his Lectures on the Religion of the Semites, 18 sqq., ritual is in fact primary for primitive religion, whilst dogma and myth are secondary); secondly, that the outlook of such belief and practice is not exclusively towards the spiritual, unless this term be widened until it mean next to nothing, but is likewise towards the quasi-material, as will be shown presently. The merit of this definition, on the other hand, lies in its bilateral form, which calls attention to the need of characterizing both the religious attitude and the religious object to which the former has reference. The same form appears in Dr J. G. Frazer’s definition in The Golden Bough (2nd ed.), i. 63. He understands by religion “a propitiation or conciliation of powers superior to man which are believed to direct and control the course of nature and of human life.” He goes on to explain that by “powers” he means “conscious or personal agents.” It is also to be noted that he is here definitely opposing religion to magic, which he holds to be based on the (implicit) assumption “that the course of nature is determined, not by the passions or caprice of personal beings, but by the operation of immutable laws acting mechanically.” His definition improves on Tylor’s in so far as it makes worship integral to the religious attitude. By regarding the object of religion as necessarily personal, however, he is led to exclude much that the primitive man undoubtedly treats with awe and respect as exerting a mystic effect on his life. Further, in maintaining that the powers recognized by religion are always superior to man, he leaves unclasped a host of practices that display a bargaining, or even a hectoring, spirit on the part of those addressing them (see Prayer). Threatening or beating a fetish cannot be brought under the head of magic, even if we adopt Frazer’s principle (op. cit. i. 64) that to constrain or coerce a personal being is to treat him as an inanimate agent; for such a principle is quite inapplicable to cases of mere terrorism, whilst it may be doubted if it even renders the sense of the savage magician’s typical notion of his modus operandi, viz. as the bringing to bear of a greater mana or psychic influence (see below) on what has less, and must therefore do as it is bidden. Such definitions, then, are to be accepted, if at all, as definitions of type, selective designations of leading but not strictly universal features. An encyclopaedic account, however, should rest rather on an exterior definition which can serve as it were to pigeon-hole the whole mass of significant facts. Such an exterior definition is suggested by Mr E. Crawley in The Tree of Life, 209, where he points out that “neither the Greek nor the Latin language has any comprehensive term for religion, except in the one ἱερά, and in the other sacra, words which are equivalent to ‘sacred.’ No other term covers the whole of religious phenomena, and a survey of the complex details of various worships results in showing that no other conception will comprise the whole body of religious facts.” It may be added that we have here no generalization imported from a
