the origin of a belief in god beyond the ken of history and of speculation — sketch of conjectural theories — two elements in all beliefs, whether of backward or civilised races — the mythical and the religious — these may be coeval, or either may be older than the other — difficulty of study — the current anthropological theory — stated objections to the theory — gods and spirits — suggestion that savage religion is borrowed from europeans — reply to mr. tylor’s arguments on this head — the morality of savages.

“the question of the origin of a belief in deity does not come within the scope of a strictly historical inquiry. no man can watch the idea of god in the making or in the beginning. we are acquainted with no race whose beginning does not lie far back in the unpenetrated past. even on the hypothesis that the natives of australia, for example, were discovered in a state of culture more backward than that of other known races, yet the institutions and ideas of the australians must have required for their development an incalculable series of centuries. the notions of man about the deity, man’s religious sentiments and his mythical narratives, must be taken as we find them. there have been, and are, many theories as to the origin of the conception of a supernatural being or beings, concerned with the fortunes of mankind, and once active in the making of the earth and its inhabitants. there is the hypothesis of an original divine tradition, darkened by the smoke of foolish mortal fancies. there is the hypothesis of an innate and intuitive sensus numinis. there is the opinion that the notion of deity was introduced to man by the very nature of his knowledge and perceptions, which compel him in all things to recognise a finite and an infinite. there is the hypothesis that gods were originally ghosts, the magnified shapes of ancestral spectres. there is the doctrine that man, seeking in his early speculations for the causes of things, and conscious of his own powers as an active cause, projected his own shadow on the mists of the unknown, and peopled the void with figures of magnified non-natural men, his own parents and protectors, and the makers of many of the things in the world.

“since the actual truth cannot be determined by observation and experiment, the question as to the first germs of the divine conception must here be left unanswered. but it is possible to disengage and examine apart the two chief elements in the earliest as in the latest ideas of godhead. among the lowest and most backward, as among the most advanced races, there coexist the mythical and the religious elements in belief. the rational factor (or what approves itself to us as the rational factor) is visible in religion; the irrational is prominent in myth. the australian, the bushman, the solomon islander, in hours of danger and necessity ‘yearns after the gods,’ and has present in his heart the idea of a father and friend. this is the religious element. the same man, when he comes to indulge his fancy for fiction, will degrade this spiritual friend and father to the level of the beasts, and will make him the hero of comic or repulsive adventures. this is the mythical or irrational element. religion, in its moral aspect, always traces back to the belief in a power that is benign and works for righteousness. myth, even in homer or the rig-veda, perpetually falls back on the old stock of absurd and immoral divine adventures.1

“it would be rash, in the present state of knowledge, to pronounce that the germ of the serious homeric sense of the justice and power of the divinity is earlier or later than the germ of the homeric stories of gods disguised as animals, or imprisoned by mortals, or kicked out of olympus. the rational and irrational aspects of mythology and religion may be of coeval antiquity for all that is certainly known, or either of them, in the dark backward of mortal experience, may have preceded the other. there is probably no religion nor mythology which does not offer both aspects to the student. but it is the part of advancing civilisation to adorn and purify the rational element, and to subordinate and supersede the irrational element, as far as religious conservatism, ritual and priestly dogma will permit.”

such were the general remarks with which this chapter opened in the original edition of the present work. but reading, reflection and certain additions to the author’s knowledge of facts, have made it seem advisable to state, more fully and forcibly than before, that, in his opinion, not only the puzzling element of myth, but the purer element of a religious belief sanctioning morality is derived by civilised people from a remote past of savagery. it is also necessary to draw attention to a singular religious phenomena, a break, or “fault,” as geologists call it, in the religious strata. while the most backward savages, in certain cases, present the conception of a being who sanctions ethics, and while that conception recurs at a given stage of civilisation, it appears to fade, or even to disappear in some conditions of barbarism. among some barbaric peoples, such as the zulus, and the red indians of french canada when first observed, as among some polynesians and some tribes of western and central africa little trace of a supreme being is found, except a name, and that name is even occasionally a matter of ridicule. the highest religious conception has been reached, and is generally known, yet the being conceived of as creative is utterly neglected, while ghosts, or minor gods, are served and adored. to this religious phenomenon (if correctly observed) we must attempt to assign a cause. for this purpose it is necessary to state again what may be called the current or popular anthropological theory of the evolution of gods.

that theory takes varying shapes. in the philosophy of mr. herbert spencer we find a pure euhemerism. gods are but ghosts of dead men, raised to a higher and finally to the highest power. in the somewhat analogous but not identical system of mr. tylor, man first attains to the idea of spirit by reflection on various physical, psychological and psychical experiences, such as sleep, dreams, trances, shadows, hallucinations, breath and death, and he gradually extends the conception of soul or ghost till all nature is peopled with spirits. of these spirits one is finally promoted to supremacy, where the conception of a supreme being occurs. in the lowest faiths there is said, on this theory, to be no connection, or very little connection, between religion and morality. to supply a religious sanction of morals is the work of advancing thought.2

this current hypothesis is, confessedly, “animistic,” in mr. tylor’s phrase, or, in mr. spencer’s terminology, it is “the ghost theory”. the human soul, says mr. tylor, has been the model on which all man’s ideas of spiritual beings, from “the tiniest elf” to “the heavenly creator and ruler of the world, the great spirit,” have been framed.3 thus it has been necessary for mr. tylor and for mr. spencer to discover first an origin of man’s idea of his own soul, and that supposed origin in psychological, physical and psychical experiences is no doubt adequate. by reflection on these facts, probably, the idea of spirit was reached, though the psychical experiences enumerated by mr. tylor may contain points as yet unexplained by materialism. from these sources are derived all really “animistic” gods, all that from the first partake of the nature of hungry ghosts, placated by sacrifices of food, though in certain cases that hunger may have been transferred, we surmise, by worshippers to gods not originally animistic.

in answer to this theory of an animistic or ghostly origin of all gods, it must first be observed that all gods are not necessarily, it would seem, of animistic origin. among certain of the lowest savages, although they believe in ghosts, the animistic conception, the spiritual idea, is not attached to the relatively supreme being of their faith. he is merely a powerful being, unborn, and not subject to death. the purely metaphysical question “was he a ghost?” does not seem always to have been asked. consequently there is no logical reason why man’s idea of a maker should not be prior to man’s idea that there are such things as souls, ghosts and spirits. therefore the animistic theory is not necessary as material for the “god-idea”. we cannot, of course, prove that the “god-idea” was historically prior to the “ghost-idea,” for we know no savages who have a god and yet are ignorant of ghosts. but we can show that the idea of god may exist, in germ, without explicitly involving the idea of spirit. thus gods may be prior in evolution to ghosts, and therefore the animistic theory of the origin of gods in ghosts need not necessarily be accepted.

in the first place, the original evolution of a god out of a ghost need not be conceded, because in perhaps all known savage theological philosophy the god, the maker and master, is regarded as a being who existed before death entered the world. everywhere, practically speaking, death is looked on as a comparatively late intruder. he came not only after god was active, but after men and beasts had populated the world. scores of myths accounting for this invasion of death have been collected all over the world.4 thus the relatively supreme being, or beings, of religion are looked on as prior to death, therefore, not as ghosts. they are sometimes expressly distinguished as “original gods” from other gods who are secondary, being souls of chiefs. thus all tongan gods are atua, but all atua are not “original gods”.5 the word atua, according to mr. white, is “a-tu-a”. “a” was the name given to the author of the universe, and signifies: “am the unlimited in power,” “the conception,” “the leader,” “the beyond all”. “tua” means “beyond that which is most distant,” “behind all matter,” and “behind every action”. clearly these conceptions are not more mythical (indeed a does not seem to occur in the myths), nor are they more involved in ghosts, than the unknown absolute of mr. herbert spencer. yet the word atua denotes gods who are recognised as ghosts of chiefs, no less than it denotes the supreme existence.6 these ideas are the metaphysical theology of a race considerably above the lowest level. they lend no assistance to a theory that a was, or was evolved out of, a human ghost, and he is not found in maori mythology as far as our knowledge goes. but, among the lowest known savages, the australians, we read that “the creator was a gigantic black, once on earth, now among the stars”. this is in gippsland; the deities of the fuegians and the blackfoot indians are also beings, anthropomorphic, unborn and undying, like mangarrah, the creative being of the larrakeah tribe in australia. “a very good man called mangarrah lives in the sky. . . . he made everything” (blacks excepted). he never dies.7 the melanesian vui “never were men,” were “something different,” and “were not ghosts”. it is as a being, not as a spirit, that the kurnai deity munganngaur (our father) is described.8 in short, though europeans often speak of these divine beings of low savages as “spirits,” it does not appear that the natives themselves advance here the metaphysical idea of spirit. these gods are just beings, anthropomorphic, or (in myth and fable), very often bestial, “theriomorphic”.9 it is manifest that a divine being envisaged thus need not have been evolved out of the theory of spirits or ghosts, and may even have been prior to the rise of the belief in ghosts.

again, these powerful, or omnipotent divine beings are looked on as guardians of morality, punishers of sin, rewarders of righteousness, both in this world and in a future life, in places where ghosts, though believed in, are not worshipped, nor in receipt of sacrifice, and where, great grandfathers being forgotten, ancestral ghosts can scarcely swell into gods. this occurs among andamanese, fuegians and australians, therefore, among non-ghost-worshipping races, ghosts cannot have developed into deities who are not even necessarily spirits. these gods, again, do not receive sacrifice, and thus lack the note of descent from hungry food-craving ghosts. in australia, indeed, while ghosts are not known to receive any offerings, “the recent custom of providing food for it”— the dead body of a friend —“is derided by the intelligent old aborigines as ‘white fellow’s gammon’”.10

the australians possess no chiefs like “vich ian vohr or chingachgook” whose ghosts might be said to swell into supreme moral deities. “headmen” they have, leaders of various degrees of authority, but no vich ian vohr, no semi-sacred representative of the tribe.11 nor are the ghosts of the headmen known to receive any particular posthumous attention or worship. thus it really seems impossible to show proof that australian gods grew out of australian ghosts, a subject to which we shall return.

some supporters of the current theory therefore fall back on the hypothesis that the australians are sadly degenerate.12 chiefs, it is argued, or kings, they once had, and the gods are surviving ghosts of these wholly forgotten potentates. to this we reply that we know not the very faintest trace of australian degeneration. sir john lubbock and mr. tylor have correctly argued that the soil of australia has not yet yielded so much as a fragment of native pottery, nor any trace of native metal work, not a vestige of stone buildings occurs, nor of any work beyond the present native level of culture, unless we reckon weirs for fish-catching. “the australian boomerang,” writes mr. tylor, “has been claimed as derived from some hypothetical high culture, whereas the transition-stages through which it is connected with the club are to be observed in its own country, while no civilised race possesses the weapon.”13

therefore the australian, with his boomerang, represents no degeneration but advance on his ancestors, who had not yet developed the boomerang out of the club. if the excessively complex nature of australian rules of prohibited degrees be appealed to as proof of degeneration from the stage in which they were evolved, we reply that civilisation everywhere tends not to complicate but to simplify such rules, as it also notoriously simplifies the forms of language.

the australian people, when discovered, were only emerging from palaeolithic culture, while the neighbouring tasmanians were frankly palaeolithic.14 far from degenerating, the australians show advance when they supersede their beast or other totem by an eponymous human hero.15 the eponymous hero, however, changed with each generation, so that no one name was fixed as that of tribal father, later perhaps to become a tribal god. we find several tribes in which the children now follow the father’s class, and thus paternal kin takes the place of the usual early savage method of reckoning kinship by the mother’s side, elsewhere prevalent in australia. in one of these tribes, dwelling between the glenelg and mount napier, headmanship is hereditary, but nothing is said of any worship of the ghosts of chiefs. all this social improvement denotes advance on the usual australian standard.16 of degeneration (except when produced recently by european vices and diseases) i know no trace in australia. their highest religious conceptions, therefore, are not to be disposed of as survivals of a religion of the ghosts of such chiefs as the australians are not shown ever to have recognised. the “god idea” in australia, or among the andamanese, must have some other source than the ghost-theory. this is all the more obvious because not only are ghosts not worshipped by the australians, but also the divine beings who are alleged to form links between the ghost and the moral god are absent. there are no departmental gods, as of war, peace, the chase, love, and so forth. sun, sky and earth are equally unworshipped. there is nothing in religion between a being, on one hand (with a son or sons), and vague mischievous spirits, boilyas or mrarts, and ghosts (who are not worshipped), on the other hand. the friends of the idea that the god is an ancient evolution from the ghost of such a chief as is not proved to have existed, must apparently believe that the intermediate stages in religious evolution, departmental gods, nature gods and gods of polytheism in general once existed in australia, and have all been swept away in a deluge of degeneration. that deluge left in religion a moral, potently active father and judge. now that conception is considerably above the obsolescent belief in an otiose god which is usually found among barbaric races of the type from which the australians are said to have degenerated. there is no proof of degeneracy, and, if degeneration has occurred, why has it left just the kind of deity who, in the higher barbaric culture, is not commonly found? clearly this attempt to explain the highest aspect of australian religion by an undemonstrated degeneration is an effort of despair.

while the current theory thus appears to break down over the deities of certain australian tribes and of other low savages to be more particularly described later, it is not more successful in dealing with what we have called the “fault” or break in the religious strata of higher races. the nature of that “fault” may thus be described: while the deities of several low savage peoples are religiously regarded as guardians and judges of conduct both in this life and in the next, among higher barbarians they are often little, or not at all, interested in conduct. again, while among australians, and andamanese, and fuegians, there is hardly a verifiable trace, if any trace there be, of sacrifice to any divine being, among barbarians the gods beneath the very highest are in receipt even of human sacrifice. even among barbarians the highest deity is very rarely worshipped with sacrifice. through various degrees he is found to lose all claim on worship, and even to become a mere name, and finally a jest and a mockery. meanwhile ancestral ghosts, and gods framed on the same lines as ghosts, receive sacrifice of food and of human victims. once more, the high gods of low savages are not localised, not confined to any temple or region. but the gods of higher barbarians (the gods beneath the highest), are localised in this way, as occasionally even the highest god also is.

all this shows that, among advancing barbarians, the gods, if they started from the estate of gods among savages on the lowest level, become demoralised, limited, conditioned, relegated to an otiose condition, and finally deposed, till progressive civilisation, as in greece, reinstates or invents purer and more philosophic conceptions, without being able to abolish popular and priestly myth and ritual.

here, then, is a flaw or break in the strata of religion. what was the cause of this flaw? we answer, the evolution, through ghosts, of “animistic” gods who retained the hunger and selfishness of these ancestral spirits whom the lowest savages are not known to worship.

the moral divine beings of these lowest races, beings (when religiously regarded) unconditioned, in need of no gift that man can give, are not to be won by offerings of food and blood. of such offerings ghosts, and gods modelled on ghosts, are notoriously in need. strengthened and propitiated by blood and sacrifice (not offered to the gods of low savages), the animistic deities will become partisans of their adorers, and will either pay no regard to the morals of their worshippers, or will be easily bribed to forgive sins. here then is, ethically speaking, a flaw in the strata of religion, a flaw found in the creeds of ghost-worshipping barbarians, but not of non-ghost-worshipping savages. a crowd of venal, easy-going, serviceable deities has now been evolved out of ghosts, and animism is on its way to supplant or overlay a rude early form of theism. granting the facts, we fail to see how they are explained by the current theory which makes the highest god the latest in evolution from a ghost. that theory wrecks itself again on the circumstance that, whereas the tribal or national highest divine being, as latest in evolution, ought to be the most potent, he is, in fact, among barbaric races, usually the most disregarded. a new idea, of course, is not necessarily a powerful or fashionable idea. it may be regarded as a “fad,” or a heresy, or a low form of dissent. but, when universally known to and accepted by a tribe or people, then it must be deemed likely to possess great influence. but that is not the case; and among barbaric tribes the most advanced conception of deity is the least regarded, the most obsolete.

an excellent instance of the difference between the theory here advocated, and that generally held by anthropologists, may be found in mr. abercromby’s valuable work, pre-and proto-historic finns, i. 150-154. the gods, and other early ideas, says mr. abercromby, “could in no sense be considered as supernatural”. we shall give examples of gods among the races “nearest the beginning,” whose attributes of power and knowledge can not, by us at least, be considered other than “supernatural”. “the gods” (in this hypothesis) “were so human that they could be forced to act in accordance with the wishes of their worshippers, and could likewise be punished.” these ideas, to an australian black, or an andamanese, would seem dangerously blasphemous. these older gods “resided chiefly in trees, wells, rivers and animals”. but many gods of our lowest known savages live “beyond the sky”. mr. abercromby supposes the sky god to be of later evolution, and to be worshipped after man had exhausted “the helpers that seemed nearest at hand . . . in the trees and waters at his very door”. now the australian black has not a door, nor has he gods of any service to him in the “trees and waters,” though sprites may lurk in such places for mischief. but in mr. abercromby’s view, some men turned at last to the sky-god, “who in time would gain a large circle of worshippers”. he would come to be thought omnipotent, omniscient, the creator. this notion, says mr. abercromby, “must, if this view is correct, be of late origin”. but the view is not correct. the far-seeing powerful maker beyond the sky is found among the very backward races who have not developed helpers nearer man, dwelling round what would be his door, if door he was civilised enough to possess. such near neighbouring gods, of human needs, capable of being bullied, or propitiated by sacrifice, are found in races higher than the lowest, who, for their easily procurable aid, have allowed the maker to sink into an otiose god, or a mere name. mr. abercromby unconsciously proves our case by quoting the example of a samoyede. this man knew a sky-god, num; that conception was familiar to him. he also knew a familiar spirit. on mr. abercromby’s theory he should have resorted for help to the sky-god, not to the sprite. but he did the reverse: he said, “i cannot approach num, he is too far away; if i could reach him i should not beseech thee (the familiar spirit), but should go myself; but i cannot”. for this precise reason, people who have developed the belief in accessible affable spirits go to them, with a spell to constrain, or a gift to bribe, and neglect, in some cases almost forget, their maker. but he is worshipped by low savages, who do not propitiate ghosts and who have no gods in wells and trees, close at hand. it seems an obvious inference that the greater god is the earlier evolved.

these are among the difficulties of the current anthropological theory. there is, however, a solution by which the weakness of the divine conception, its neglected, disused aspect among barbaric races, might be explained by anthropologists, without regarding it as an obsolescent form of a very early idea. this solution is therefore in common use. it is applied to the deity revealed in the ancient mysteries of the australians, and it is employed in american and african instances.

the custom is to say that the highest divine being of american or african native peoples has been borrowed from europeans, and is, especially, a savage refraction from the god of missionaries. if this can be proved, the shadowy, practically powerless “master of life” of certain barbaric peoples, will have degenerated from the christian conception, because of that conception he will be only a faint unsuccessful refraction. he has been introduced by europeans, it is argued, but is not in harmony with his new environment, and so is “half-remembered and half forgot”.

the hypothesis of borrowing admits of only one answer, but that answer should be conclusive. if we can discover, say in north america, a single instance in which the supreme being occurs, while yet he cannot possibly be accounted for by any traceable or verifiable foreign influence, then the burden of proof, in other cases, falls on the opponent. when he urges that other north american supreme beings were borrowed, we can reply that our crucial example shows that this need not be the fact. to prove that it is the fact, in his instances, is then his business. it is obvious that for information on this subject we must go to the reports of the earliest travellers who knew the red indians well. we must try to get at gods behind any known missionary efforts. mr. tylor offers us the testimony of heriot, about 1586, that the natives of virginia believed in many gods, also in one chief god, “who first made other principal gods, and then the sun, moon and stars as petty gods”.17 whence could the natives of virginia have borrowed this notion of a creator before 1586? if it is replied, in the usual way, that they developed him upwards out of sun, moon and star gods, other principal gods, and finally reached the idea of the creator, we answer that the idea of the maker is found where these alleged intermediate stages are not found, as in australia. in virginia then, as in victoria, a creator may have been evolved in some other way than that of gradual ascent from ghosts, and may have been, as in australia and elsewhere, prior to verifiable ghost-worship. again, in virginia at our first settlement, the native priests strenuously resisted the introduction of christianity. they were content with their deity, ahone, “the great god who governs all the world, and makes the sun to shine, creating the moon and stars his companions. . . . the good and peaceable god . . . needs not to be sacrificed unto, for he intendeth all good unto them.” this good creator, without sacrifice, among a settled agricultural barbaric race sacrificing to other gods and ghosts, manifestly cannot be borrowed from the newly arrived religion of christianity, which his priests, according to the observer, vigorously resisted. ahone had a subordinate deity, magisterial in functions, “looking into all men’s actions” and punishing the same, when evil. to this god sacrifices were made, and if his name, okeus, is derived from oki = “spirit,” he was, of course, an animistic ghost-evolved deity. anthropological writers, by an oversight, have dwelt on oki, but have not mentioned ahone.18 manifestly it is not possible to insist that these virginian high deities were borrowed, without saying whence and when they were borrowed by a barbaric race which was, at the same time, rejecting christian teaching.

mr. tylor writes, with his habitual perspicacity: “it is the widespread belief in the great spirit, whatever his precise nature and origin, that has long and deservedly drawn the attention of european thinkers to the native religions of the north american tribes”. now while, in recent times, christian ideas may undeniably have crystallised round “the great spirit,” it has come to be thought “that the whole doctrine of the great spirit was borrowed by the savages from missionaries and colonists. but this view will not bear examination,” says mr. tylor.19

mr. tylor proceeds to prove this by examples from greenland, and the algonkins. he instances the massachusett god, kiehtan, who created the other gods, and receives the just into heaven. this was recorded in 1622, but the belief, says winslow, our authority, goes back into the unknown past. “they never saw kiehtan, but they hold it a great charge and duty that one age teach another.” how could a deity thus rooted in a traditional past be borrowed from recent english settlers?

in these cases the hypothesis of borrowing breaks down, and still more does it break down over the algonkin deity atahocan.

father le jeune, s.j., went first among the algonkins, a missionary pioneer, in 1633, and suffered unspeakable things in his courageous endeavour to win souls in a most recalcitrant flock. he writes (1633): “as this savage has given me occasion to speak of their god, i will remark that it is a great error to think that the savages have no knowledge of any deity. i was surprised to hear this in france. i do not know their secrets, but, from the little which i am about to tell, it will be seen that they have such knowledge.

“they say that one exists whom they call atahocan, who made the whole. speaking of god in a wigwam one day, they asked me ‘what is god?’ i told them that it was he who made all things, heaven and earth. they then began to cry out to each other, ‘atahocan! atahocan! it is atahocan!’”

there could be no better evidence that atahocan was not (as is often said) “borrowed from the jesuits”. the jesuits had only just arrived.

later (1634) le jeune interrogated an old man and a partly europeanised sorcerer. they replied that nothing was certain; that atahocan was only spoken of as “of a thing so remote,” that assurance was impossible. “in fact, their word nitatohokan means, ‘i fable, i tell an old story’.”

thus atahocan, though at once recognised as identical with the creator of the missionary, was so far from being the latest thing in religious evolution that he had passed into a proverb for the ancient and the fabulous. this, of course, is inconsistent with recent borrowing. he was neglected for khichikouai, spirits which inspire seers, and are of some practical use, receiving rewards in offerings of grease, says le jeune.20

the obsolescent atahocan seems to have had no moral activity. but, in america, this indolence of god is not universal. mr. parkman indeed writes: “in the primitive indian’s conception of a god, the idea of moral good has no part”.21 but this is definitely contradicted by heriot, strachey, winslow, already cited, and by pere le jeune. the good attributes of kiehtan and ahone were not borrowed from christianity, were matter of indian belief before the english arrived. mr. parkman writes: “the moment the indians began to contemplate the object of his faith, and sought to clothe it with attributes, it became finite, and commonly ridiculous”. it did so, as usual, in mythology, but not in religion. there is nothing ridiculous in what is known of ahone and kiehtan. if they had a mythology, and if we knew the myths, doubtless they would be ridiculous enough. the savage mind, turned from belief and awe into the spinning of yarns, instantly yields to humorous fancy. as we know, mediaeval popular christianity, in imagery, marchen or tales, and art, copiously illustrates the same mental phenomenon. saints, god, our lord, and the virgin, all play ludicrous and immoral parts in christian folk-tales. this is mythology, and here is, beyond all cavil, a late corruption of religion. here, where we know the history of a creed, religion is early, and these myths are late. other examples of american divine ideas might be given, such as the extraordinary hymns in which the zunis address the eternal, ahonawilona. but as the zuni religion has only been studied in recent years, the hymns would be dismissed as “borrowed,” though there is nothing catholic or christian about them. we have preferred to select examples where borrowing from christianity is out of the question. the current anthropological theory is thus confronted with american examples of ideas of the divine which cannot have been borrowed, while, if the gods are said to have been evolved out of ghosts, we reply that, in some cases, they receive no sacrifice, sacrifice being usually a note of ghostly descent. again, similar gods, as we show, exist where ghosts of chiefs are not worshipped, and as far as evidence goes never were worshipped, because there is no evidence of the existence at any time of such chiefs. the american highest gods may then be equally free from the taint of ghostly descent.

there is another more or less moral north american deity whose evolution is rather questionable. pere brebeuf (1636), speaking of the hurons, says that “they have recourse to heaven in almost all their necessities, . . . and i may say that it is, in fact, god whom they blindly adore, for they imagine that there is an oki, that is, a demon, in heaven, who regulates the seasons, bridles the winds and the waves of the sea, and helps them in every need. they dread his wrath, and appeal to him as witness to the inviolability of their faith, when they make a promise or treaty of peace with enemies. ‘heaven hear us to-day’ is their form of adjuration.”22

a spiritual being, whose home is heaven, who rides on the winds, whose wrath is dreaded, who sanctions the oath, is only called “a demon” by the prejudice of the worthy father who, at the same time, admits that the savages have a conception of god — and that god, so conceived, is this demon!

the debatable question is, was the “demon,” or the actual expanse of sky, first in evolution? that cannot precisely be settled, but in the analogous chinese case of china we find heaven (tien) and “shang-ti, the personal ruling deity,” corresponding to the huron “demon”. shang-ti, the personal deity, occurs most in the oldest, pre-confucian sacred documents, and, so far, appears to be the earlier conception. the “demon” in huron faith may also be earlier than the religious regard paid to his home, the sky.23 the unborrowed antiquity of a belief in a divine being, creative and sometimes moral, in north america, is thus demonstrated. so far i had written when i accidentally fell in with mr. tylor’s essay on “the limits of savage religion”.24 in that essay, rather to my surprise, mr. tylor argues for the borrowing of “the great spirit,” “the great manitou,” from the jesuits. now, as to the phrase, “great spirit,” the jesuits doubtless caused its promulgation, and, where their teaching penetrated, shreds of their doctrine may have adhered to the indian conception of that divine being. but mr. tylor in his essay does not allude to the early evidence, his own, for oki, atahocan, kiehtan, and torngursak, all undeniably prior to jesuit influence, and found where jesuits, later, did not go. as mr. tylor offers no reason for disregarding evidence in 1892 which he had republished in a new edition of primitive culture in 1891, it is impossible to argue against him in this place. he went on, in the essay cited (1892) to contend that the australian god of the kamilaroi of victoria, baiame, is, in name and attributes, of missionary introduction. happily this hypothesis can be refuted, as we show in the following chapter on australian gods.

it would be easy enough to meet the hypothesis of borrowing in the case of the many african tribes who possess something approaching to a rude monotheistic conception. among these are the dinkas of the upper nile, with their neighbours, whose creed russegger compares to that of modern deists in europe. the dinka god, dendid, is omnipotent, but so benevolent that he is not addressed in prayer, nor propitiated by sacrifice. compare the supreme being of the caribs, beneficent, otiose, unadored.25 a similar deity, veiled in the instruction of the as yet unpenetrated mysteries, exists among the yao of central africa.26 of the negro race, waitz says, “even if we do not call them monotheists, we may still think of them as standing on the boundary of monotheism despite their innumerable rude superstitions”.27 the tshi speaking people of the gold coast have their unworshipped nyankupon, a now otiose unadored being, with a magisterial deputy, worshipped with many sacrifices. the case is almost an exact parallel to that of ahone and oki in america. these were not borrowed, and the author has argued at length against major ellis’s theory of the borrowing from christians of nyankupon.28

to conclude this chapter, the study of savage and barbaric religions seems to yield the following facts:—

1. low savages. no regular chiefs. great beings, not in receipt of sacrifice, sanctioning morality. ghosts are not worshipped, though believed in. polytheism, departmental gods and gods of heaven, earth, sky and so forth, have not been developed or are not found.

2. barbaric races. aristocratic or monarchic. ghosts are worshipped and receive sacrifice. polytheistic gods are in renown and receive sacrifice. there is usually a supreme maker who is, in some cases, moral, in others otiose. in only one or two known cases (as in that of the polynesian taaroa) is he in receipt of sacrifice.

3. barbaric races. (zulus, monarchic with unkulunkulu; some algonquins (feebly aristocratic) with atahocan). religion is mainly ancestor worship or vague spirit worship; ghosts are propitiated with food. there are traces of an original divine being whose name is becoming obsolescent and a matter of jest.

4. early civilisations. monarchic or aristocratic. (greece, egypt, india, peru, mexico.) polytheism. one god tends to be supreme. religiously regarded, gods are moral; in myth are the reverse. gods are in receipt of sacrifice. heavenly society is modelled on that of men, monarchic or aristocratic. philosophic thought tends towards belief in one pure god, who may be named zeus, in greece.

5. the religion of israel. probably a revival and purification of the old conception of a moral, beneficent creator, whose creed had been involved in sacrifice and anthropomorphic myth.

in all the stages thus roughly sketched, myths of the lowest sort prevail, except in the records of the last stage, where the documents have been edited by earnest monotheists.

if this theory be approximately correct, man’s earliest religious ideas may very well have consisted, in a sense, of dependence on a supreme moral being who, when attempts were made by savages to describe the modus of his working, became involved in the fancies of mythology. how this belief in such a being arose we have no evidence to prove. we make no hint at a sensus numinis, or direct revelation.

while offering no hypothesis of the origin of belief in a moral creator we may present a suggestion. mr. darwin says about early man: “the same high mental faculties which first led man to believe in unseen spiritual agencies, then in fetichism, polytheism and ultimately monotheism, would infallibly lead him, so long as his reasoning powers remained poorly developed, to various strange superstitions and customs”.29 now, accepting mr. darwin’s theory that early man had “high mental faculties,” the conception of a maker of things does not seem beyond his grasp. man himself made plenty of things, and could probably conceive of a being who made the world and the objects in it. “certainly there must be some being who made all these things. he must be very good too,” said an eskimo to a missionary.30 the goodness is inferred by the eskimo from his own contentment with “the things which are made”.31

another example of barbaric man “seeking after god” may be adduced.

what the greenlander said is corroborated by what a kaffir said. kaffir religion is mainly animistic, ancestral spirits receive food and sacrifice — there is but an evanescent tradition of a “lord in heaven”. thus a very respectable kaffir said to m. arbrousset, “your tidings (christianity) are what i want; and i was seeking before i knew you. . . . i asked myself sorrowful questions. ‘who has touched the stars with his hands? . . . who makes the waters flow? . . . who can have given earth the wisdom and power to produce corn?’ then i buried my face in my hands.”

“this,” says sir john lubbock, “was, however, an exceptional case. as a general rule savages do not set themselves to think out such questions.”32

as a common fact, if savages never ask the question, at all events, somehow, they have the answer ready made. “mangarrah, or baiame, puluga, or dendid, or ahone, or ahonawilona, or atahocan, or taaroa, or tui laga, was the maker.” therefore savages who know that leave the question alone, or add mythical accretions. but their ancestors must have asked the question, like the “very respectable kaffir” before they answered it.

having reached the idea of a creator, it was not difficult to add that he was “good,” or beneficent, and was deathless.

a notion of a good powerful maker, not subject to death because necessarily prior to death (who only invaded the world late), seems easier of attainment than the notion of spirit which, ex hypothesi, demands much delicate psychological study and hard thought. the idea of a good maker, once reached, becomes, perhaps, the germ of future theism, but, as mr. darwin says, the human mind was “infallibly led to various strange superstitions”. as st. paul says, in perfect agreement with mr. darwin on this point, “they became vain in their imaginations, and their foolish heart was darkened”.

among other imaginations (right or wrong) was the belief in spirits, with all that followed in the way of instituting sacrifices, even of human beings, and of dropping morality, about which the ghost of a deceased medicine-man was not likely to be much interested. the supposed nearness to man, and the venal and partial character of worshipped gods and ghost-gods, would inevitably win for them more service and attention than would be paid to a maker remote, unbought and impartial. hence the conception of such a being would tend to obsolescence, as we see that it does, and would be most obscured where ghosts were most propitiated, as among the zulus. later philosophy would attach the spiritual conception to the revived or newly discovered idea of the supreme god.

in all this speculation there is nothing mystical; no supernatural or supernormal interference is postulated. supernormal experiences may have helped to originate or support the belief in spirits, that, however, is another question. but this hypothesis of the origin of belief in a good unceasing maker of things is, of course, confessedly a conjecture, for which historical evidence cannot be given, in the nature of the case. all our attempts to discover origins far behind history must be conjectural. their value must be estimated by the extent to which this or that hypothesis colligates the facts. now our hypothesis does colligate the facts. it shows how belief in a moral supreme being might arise before ghosts were worshipped, and it accounts for the flaw in the religious strata, for the mythical accretions, for the otiose creator in the background of many barbaric religions, and for the almost universal absence of sacrifice to the god relatively supreme. he was, from his earliest conception, in no need of gifts from men.

on this matter of otiose supreme gods, professor menzies writes, “it is very common to find in savage beliefs a vague far-off god, who is at the back of all the others, takes little part in the management of things, and receives little worship. but it is impossible to judge what that being was at an earlier time; he may have been a nature god, or a spirit who has by degrees grown faint, and come to occupy this position.”

now the position which he occupies is usually, if not universally, that of the creator. he could not arrive at this rank by “becoming faint,” nor could “a nature-god” be the maker of nature. the only way by which we can discover “what that being was at an earlier time” is to see what he is at an earlier time, that is to say, what the conception of him is, among men in an earlier state of culture. among them, as we show, he is very much more near, potent and moral, than among races more advanced in social evolution and material culture. we can form no opinion as to the nature of such “vague, far-off gods, at the back of all the others,” till we collect and compare examples, and endeavour to ascertain what points they have in common, and in what points they differ from each other. it then becomes plain that they are least far away, and most potent, where there is least ghostly and polytheistic competition, that is, among the most backward races. the more animism the less theism, is the general rule. manifestly the current hypothesis — that all religion is animistic in origin — does not account for these facts, and is obliged to fly to an undemonstrated theory of degradation, or to an undemonstrated theory of borrowing. that our theory is inconsistent with the general doctrine of evolution we cannot admit, if we are allowed to agree with mr. darwin’s statement about the high mental faculties which first led man to sympathetic, and then to wild beliefs. we do not pretend to be more darwinian than mr. darwin, who compares “these miserable and indirect results of our higher faculties” to “the occasional mistakes of the instincts of the lower animals”.

the opinion here maintained, namely, that a germ of pure belief may be detected amidst the confusion of low savage faith, and that in a still earlier stage it may have been less overlaid with fable, is in direct contradiction to current theories. it is also in contradiction with the opinions entertained by myself before i made an independent examination of the evidence. like others, i was inclined to regard reports of a moral creator, who observes conduct, and judges it even in the next life, as rumours due either to christian influence, or to mistake. i well know, however, and could, and did, discount the sources of error. i was on my guard against the twin fallacies of describing all savage religion as “devil worship,” and of expecting to find a primitive “divine tradition”. i was also on my guard against the modern bias derived from the “ghost-theory,” and mr. spencer’s works, and i kept an eye on opportunities of “borrowing”.33 i had, in fact, classified all known idola in the first edition of this work, such as the fallacy of leading questions and the chance of deliberate deception. i sought the earliest evidence, prior to any missionary teaching, and the evidence of what the first missionaries found, in the way of belief, on their arrival. i preferred the testimony of the best educated observers, and of those most familiar with native languages. i sought for evidence in native hymns (maori, zuni, dinka, red indian) and in native ceremonial and mystery, as these sources were least likely to be contaminated.

on the other side, i found a vast body of testimony that savages had no religion at all. but that testimony, en masse, was refuted by roskoff, and also, in places, by tylor. when three witnesses were brought to swear that they saw the irishman commit a crime, he offered to bring a dozen witnesses who did not see him. negative evidence of squatters, sailors and colonists, who did not see any religion among this or that race, is not worth much against evidence of trained observers and linguists who did find what the others missed, and who found more the more they knew the tribe in question. again, like others, i thought savages incapable of such relatively pure ideas as i now believe some of them to possess. but i could not resist the evidence, and i abandoned my a priori notions. the evidence forcibly attests gradations in the central belief. it is found in various shades, from relative potency down to a vanishing trace, and it is found in significant proportion to the prevalence of animistic ideas, being weakest where they are most developed, strongest where they are least developed. there must be a reason for these phenomena, and that reason, as it seems to me, is the overlaying and supersession of a rudely theistic by an animistic creed. that one cause would explain, and does colligate, all the facts.

there remains a point on which misconception proves to be possible. it will be shown, contrary to the current hypothesis, that the religion of the lowest races, in its highest form, sanctions morality. that morality, again, in certain instances, demands unselfishness. of course we are not claiming for that doctrine any supernatural origin. religion, if it sanctions ethics at all, will sanction those which the conscience accepts, and those ethics, in one way or other, must have been evolved. that the “cosmical” law is “the weakest must go to the wall” is generally conceded. man, however, is found trying to reverse the law, by equal and friendly dealing (at least within what is vaguely called “the tribe”). his religion, as in australia, will be shown to insist on this unselfishness. how did he evolve his ethics?

“be it little or be it much they get,” says dampier about the australians in 1688, “every one has his part, as well the young and tender as the old and feeble, who are not able to get abroad as the strong and lusty.” this conduct reverses the cosmical process, and notoriously civilised society, christian society, does not act on these principles. neither do the savages, who knock the old and feeble on the head, or deliberately leave them to starve, act on these principles, sanctioned by australian religion, but (according to mr. dawson) not carried out in australian practice. “when old people become infirm . . . it is lawful and customary to kill them.”34

as to the point of unselfishness, evolutionists are apt to account for it by common interest. a tribe in which the strongest monopolise what is best will not survive so well as an unselfish tribe in the struggle for existence. but precisely the opposite is true, aristocracy marks the more successful barbaric races, and an aristocratic slave-holding tribe could have swept australia as the zulus swept south africa. that aristocracy and acquisition of separate property are steps in advance on communistic savagery all history declares. therefore a tribe which in australia developed private property, and reduced its neighbours to slavery, would have been better fitted to survive than such a tribe as dampier describes.

this is so evident that probably, or possibly, the dampier state of society was not developed in obedience to a recognised tribal interest, but in obedience to an affectionate instinct. “ils s’entr’ aiment les une les autres,” says brebeuf of the hurons.35 “i never heard the women complain of being left out of feasts, or that the men ate the best portions . . . every one does his business sweetly, peaceably, without dispute. you never see disputes, quarrels, hatred, or reproach among them.” brebeuf then tells how a young indian stranger, in a time of want, stole the best part of a moose. “they did not rage or curse, they only bantered him, and yet to take our meat was almost to take our lives.” brebeuf wanted to lecture the lad; his indian host bade him hold his peace, and the stranger was given hospitality, with his wife and children. “they are very generous, and make it a point not to attach themselves to the goods of this world.” “their greatest reproach is ‘that man wants everything, he is greedy’. they support, with never a murmur, widows, orphans and old men, yet they kill hopeless or troublesome invalids, and their whole conduct to europeans was the reverse of their domestic behaviour.”

another example of savage unselfish ethics may be found in mr. mann’s account of the andaman islanders, a nomad race, very low in culture. “it is a noteworthy trait, and one which deserves high commendation, that every care and consideration are paid by all classes to the very young, the weak, the aged, and the helpless, and these being made special objects of interest and attention, invariably fare better in regard to the comforts and necessaries of daily life than any of the otherwise more fortunate members of the community.”36

mr. huxley, in his celebrated romanes lecture on “evolution and morality,” laid stress on man’s contravention of the cosmic law, “the weakest must go to the wall”. he did not explain the evolution of man’s opposition to this law. the ordinary evolutionist hypothesis, that the tribe would prosper most whose members were least self-seeking, is contradicted by all history. the overbearing, “grabbing,” aristocratic, individualistic, unscrupulous races beat the others out of the field. mr. huxley, indeed, alleged that the “influence of the cosmic process in the evolution of society is the greater the more rudimentary its civilisation. social progress means a checking of the cosmic process at every step and the substitution for it of another, which may be called the ethical process. . . . as civilisation has advanced, so has the extent of this interference increased. . . . ”37 but where, in europe, is the interference so marked as among the andamanese? we have still to face the problem of the generosity of low savages.

it is conceivable that the higher ethics of low savages rather reflect their emotional instincts than arise from tribal legislation which is supposed to enable a “tribe” to prosper in the struggle for existence. as brebeuf and dampier, among others, prove, savages often set a good example to christians, and their ethics are, in certain cases, as among the andamanese and fuegians, and, probably among the yao, sanctioned by their religion. but, as mr. tylor says, “the better savage social life seems but in unstable equilibrium, liable to be easily upset by a touch of distress, temptation, or violence”.38 still, religion does its best, in certain cases, to lend equilibrium; though all the world over, religion often fails in practice.