NO earnest student of human culture can as yet have forgotten or wholly outlived the feeling of delight awakened by the first perusal of Max Muller's brilliant "Essay on Comparative Mythology,"—a work in which the scientific principles of myth-interpretation, though not newly announced, were at least brought home to the reader with such an amount of fresh and striking concrete illustration as they had not before received. Yet it must have occurred to more than one reader that, while the analyses of myths contained in this noble essay are in the main sound in principle and correct in detail, nevertheless the author's theory of the genesis of myth is expressed, and most likely conceived, in a way that is very suggestive of carelessness and fallacy. There are obvious reasons for doubting whether the existence of mythology can be due to any "disease," abnormity, or hypertrophy of metaphor in language; and the criticism at once arises, that with the myth-makers it was not so much the character of the expression which originated the thought, as it was the thought which gave character to the expression. It is not that the early Aryans were myth-makers because their language abounded in metaphor; it is that the Aryan mother-tongue abounded in metaphor because the men and women who spoke it were myth-makers. And they were myth-makers because they had nothing but the phenomena of human will and effort with which to compare objective phenomena. Therefore it was that they spoke of the sun as an unwearied voyager or a matchless archer, and classified inanimate no less than animate objects as masculine and feminine. Max Muller's way of stating his theory, both in this Essay and in his later Lectures, affords one among several instances of the curious manner in which he combines a marvellous penetration into the significance of details with a certain looseness of general conception. 155 The principles of philological interpretation are an indispensable aid to us in detecting the hidden meaning of many a legend in which the powers of nature are represented in the guise of living and thinking persons; but before we can get at the secret of the myth-making tendency itself, we must leave philology and enter upon a psychological study. We must inquire into the characteristics of that primitive style of thinking to which it seemed quite natural that the sun should be an unerring archer, and the thunder-cloud a black demon or gigantic robber finding his richly merited doom at the hands of the indignant Lord of Light.

Among recent treatises which have dealt with this interesting problem, we shall find it advantageous to give especial attention to Mr. Tylor's "Primitive Culture," 156 one of the few erudite works which are at once truly great and thoroughly entertaining. The learning displayed in it would do credit to a German specialist, both for extent and for minuteness, while the orderly arrangement of the arguments and the elegant lucidity of the style are such as we are accustomed to expect from French essay-writers. And what is still more admirable is the way in which the enthusiasm characteristic of a genial and original speculator is tempered by the patience and caution of a cool-headed critic. Patience and caution are nowhere more needed than in writers who deal with mythology and with primitive religious ideas; but these qualities are too seldom found in combination with the speculative boldness which is required when fresh theories are to be framed or new paths of investigation opened. The state of mind in which the explaining powers of a favourite theory are fondly contemplated is, to some extent, antagonistic to the state of mind in which facts are seen, with the eye of impartial criticism, in all their obstinate and uncompromising reality. To be able to preserve the balance between the two opposing tendencies is to give evidence of the most consummate scientific training. It is from the want of such a balance that the recent great work of Mr. Cox is at times so unsatisfactory. It may, I fear, seem ill-natured to say so, but the eagerness with which Mr. Cox waylays every available illustration of the physical theory of the origin of myths has now and then the curious effect of weakening the reader's conviction of the soundness of the theory. For my own part, though by no means inclined to waver in adherence to a doctrine once adopted on good grounds, I never felt so much like rebelling against the mythologic supremacy of the Sun and the Dawn as when reading Mr. Cox's volumes. That Mr. Tylor, while defending the same fundamental theory, awakens no such rebellious feelings, is due to his clear perception and realization of the fact that it is impossible to generalize in a single formula such many-sided correspondences as those which primitive poetry end philosophy have discerned between the life of man and the life of outward nature. Whoso goes roaming up and down the elf-land of popular fancies, with sole intent to resolve each episode of myth into some answering physical event, his only criterion being outward resemblance, cannot be trusted in his conclusions, since wherever he turns for evidence he is sure to find something that can be made to serve as such. As Mr. Tylor observes, no household legend or nursery rhyme is safe from his hermeneutics. "Should he, for instance, demand as his property the nursery 'Song of Sixpence,' his claim would be easily established,—obviously the four-and-twenty blackbirds are the four-and-twenty hours, and the pie that holds them is the underlying earth covered with the overarching sky,—how true a touch of nature it is that when the pie is opened, that is, when day breaks, the birds begin to sing; the King is the Sun, and his counting out his money is pouring out the sunshine, the golden shower of Danae; the Queen is the Moon, and her transparent honey the moonlight; the Maid is the 'rosy-fingered' Dawn, who rises before the Sun, her master, and hangs out the clouds, his clothes, across the sky; the particular blackbird, who so tragically ends the tale by snipping off her nose, is the hour of sunrise." In all this interpretation there is no a priori improbability, save, perhaps, in its unbroken symmetry and completeness. That some points, at least, of the story are thus derived from antique interpretations of physical events, is in harmony with all that we know concerning nursery rhymes. In short, "the time-honoured rhyme really wants but one thing to prove it a sun-myth, that one thing being a proof by some argument more valid than analogy." The character of the argument which is lacking may be illustrated by a reference to the rhyme about Jack and Jill, explained some time since in the paper on "The Origins of Folk Lore." If the argument be thought valid which shows these ill-fated children to be the spots on the moon, it is because the proof consists, not in the analogy, which is in this case not especially obvious, but in the fact that in the Edda, and among ignorant Swedish peasants of our own day, the story of Jack and Jill is actually given as an explanation of the moon-spots. To the neglect of this distinction between what is plausible and what is supported by direct evidence, is due much of the crude speculation which encumbers the study of myths.

It is when Mr. Tylor merges the study of mythology into the wider inquiry into the characteristic features of the mode of thinking in which myths originated, that we can best appreciate the practical value of that union of speculative boldness and critical sobriety which everywhere distinguishes him. It is pleasant to meet with a writer who can treat of primitive religious ideas without losing his head over allegory and symbolism, and who duly realizes the fact that a savage is not a rabbinical commentator, or a cabalist, or a Rosicrucian, but a plain man who draws conclusions like ourselves, though with feeble intelligence and scanty knowledge. The mystic allegory with which such modern writers as Lord Bacon have invested the myths of antiquity is no part of their original clothing, but is rather the late product of a style of reasoning from analogy quite similar to that which we shall perceive to have guided the myth-makers in their primitive constructions. The myths and customs and beliefs which, in an advanced stage of culture, seem meaningless save when characterized by some quaintly wrought device of symbolic explanation, did not seem meaningless in the lower culture which gave birth to them. Myths, like words, survive their primitive meanings. In the early stage the myth is part and parcel of the current mode of philosophizing; the explanation which it offers is, for the time, the natural one, the one which would most readily occur to any one thinking on the theme with which the myth is concerned. But by and by the mode of philosophizing has changed; explanations which formerly seemed quite obvious no longer occur to any one, but the myth has acquired an independent substantive existence, and continues to be handed down from parents to children as something true, though no one can tell why it is true: Lastly, the myth itself gradually fades from remembrance, often leaving behind it some utterly unintelligible custom or seemingly absurd superstitious notion. For example,—to recur to an illustration already cited in a previous paper,—it is still believed here and there by some venerable granny that it is wicked to kill robins; but he who should attribute the belief to the old granny's refined sympathy with all sentient existence, would be making one of the blunders which are always committed by those who reason a priori about historical matters without following the historical method. At an earlier date the superstition existed in the shape of a belief that the killing of a robin portends some calamity; in a still earlier form the calamity is specified as death; and again, still earlier, as death by lightning. Another step backward reveals that the dread sanctity of the robin is owing to the fact that he is the bird of Thor, the lightning god; and finally we reach that primitive stage of philosophizing in which the lightning is explained as a red bird dropping from its beak a worm which cleaveth the rocks. Again, the belief that some harm is sure to come to him who saves the life of a drowning man, is unintelligible until it is regarded as a case of survival in culture. In the older form of the superstition it is held that the rescuer will sooner or later be drowned himself; and thus we pass to the fetichistic interpretation of drowning as the seizing of the unfortunate person by the water-spirit or nixy, who is naturally angry at being deprived of his victim, and henceforth bears a special grudge against the bold mortal who has thus dared to frustrate him.

The interpretation of the lightning as a red bird, and of drowning as the work of a smiling but treacherous fiend, are parts of that primitive philosophy of nature in which all forces objectively existing are conceived as identical with the force subjectively known as volition. It is this philosophy, currently known as fetichism, but treated by Mr. Tylor under the somewhat more comprehensive name of "animism," which we must now consider in a few of its most conspicuous exemplifications. When we have properly characterized some of the processes which the untrained mind habitually goes through, we shall have incidentally arrived at a fair solution of the genesis of mythology.

Let us first note the ease with which the barbaric or uncultivated mind reaches all manner of apparently fanciful conclusions through reckless reasoning from analogy. It is through the operation of certain laws of ideal association that all human thinking, that of the highest as well as that of the lowest minds, is conducted: the discovery of the law of gravitation, as well as the invention of such a superstition as the Hand of Glory, is at bottom but a case of association of ideas. The difference between the scientific and the mythologic inference consists solely in the number of checks which in the former case combine to prevent any other than the true conclusion from being framed into a proposition to which the mind assents. Countless accumulated experiences have taught the modern that there are many associations of ideas which do not correspond to any actual connection of cause and effect in the world of phenomena; and he has learned accordingly to apply to his newly framed notions the rigid test of verification. Besides which the same accumulation of experiences has built up an organized structure of ideal associations into which only the less extravagant newly framed notions have any chance of fitting. The primitive man, or the modern savage who is to some extent his counterpart, must reason without the aid of these multifarious checks. That immense mass of associations which answer to what are called physical laws, and which in the mind of the civilized modern have become almost organic, have not been formed in the mind of the savage; nor has he learned the necessity of experimentally testing any of his newly framed notions, save perhaps a few of the commonest. Consequently there is nothing but superficial analogy to guide the course of his thought hither or thither, and the conclusions at which he arrives will be determined by associations of ideas occurring apparently at haphazard. Hence the quaint or grotesque fancies with which European and barbaric folk-lore is filled, in the framing of which the myth-maker was but reasoning according to the best methods at his command. To this simplest class, in which the association of ideas is determined by mere analogy, belong such cases as that of the Zulu, who chews a piece of wood in order to soften the heart of the man with whom he is about to trade for cows, or the Hessian lad who "thinks he may escape the conscription by carrying a baby-girl's cap in his pocket,—a symbolic way of repudiating manhood." 157 A similar style of thinking underlies the mediaeval necromancer's practice of making a waxen image of his enemy and shooting at it with arrows, in order to bring about the enemy's death; as also the case of the magic rod, mentioned in a previous paper, by means of which a sound thrashing can be administered to an absent foe through the medium of an old coat which is imagined to cover him. The principle involved here is one which is doubtless familiar to most children, and is closely akin to that which Irving so amusingly illustrates in his doughty general who struts through a field of cabbages or corn-stalks, smiting them to earth with his cane, and imagining himself a hero of chivalry conquering single-handed a host of caitiff ruffians. Of like origin are the fancies that the breaking of a mirror heralds a death in the family,—probably because of the destruction of the reflected human image; that the "hair of the dog that bit you" will prevent hydrophobia if laid upon the wound; or that the tears shed by human victims, sacrificed to mother earth, will bring down showers upon the land. Mr. Tylor cites Lord Chesterfield's remark, "that the king had been ill, and that people generally expected the illness to be fatal, because the oldest lion in the Tower, about the king's age, had just died. 'So wild and capricious is the human mind,'" observes the elegant letter-writer. But indeed, as Mr. Tylor justly remarks, "the thought was neither wild nor capricious; it was simply such an argument from analogy as the educated world has at length painfully learned to be worthless, but which, it is not too much to declare, would to this day carry considerable weight to the minds of four fifths of the human race." Upon such symbolism are based most of the practices of divination and the great pseudo-science of astrology. "It is an old story, that when two brothers were once taken ill together, Hippokrates, the physician, concluded from the coincidence that they were twins, but Poseidonios, the astrologer, considered rather that they were born under the same constellation; we may add that either argument would be thought reasonable by a savage." So when a Maori fortress is attacked, the besiegers and besieged look to see if Venus is near the moon. The moon represents the fortress; and if it appears below the companion planet, the besiegers will carry the day, otherwise they will be repulsed. Equally primitive and childlike was Rousseau's train of thought on the memorable day at Les Charmettes when, being distressed with doubts as to the safety of his soul, he sought to determine the point by throwing a stone at a tree. "Hit, sign of salvation; miss, sign of damnation!" The tree being a large one and very near at hand, the result of the experiment was reassuring, and the young philosopher walked away without further misgivings concerning this momentous question. 158

When the savage, whose highest intellectual efforts result only in speculations of this childlike character, is confronted with the phenomena of dreams, it is easy to see what he will make of them. His practical knowledge of psychology is too limited to admit of his distinguishing between the solidity of waking experience and what we may call the unsubstantialness of the dream. He may, indeed, have learned that the dream is not to be relied on for telling the truth; the Zulu, for example, has even reached the perverse triumph of critical logic achieved by our own Aryan ancestors in the saying that "dreams go by contraries." But the Zulu has not learned, nor had the primeval Aryan learned, to disregard the utterances of the dream as being purely subjective phenomena. To the mind as yet untouched by modern culture, the visions seen and the voices heard in sleep possess as much objective reality as the gestures and shouts of waking hours. When the savage relates his dream, he tells how he SAW certain dogs, dead warriors, or demons last night, the implication being that the things seen were objects external to himself. As Mr. Spencer observes, "his rude language fails to state the difference between seeing and dreaming that he saw, doing and dreaming that he did. From this inadequacy of his language it not only results that he cannot truly represent this difference to others, but also that he cannot truly represent it to himself. Hence in the absence of an alternative interpretation, his belief, and that of those to whom he tells his adventures, is that his OTHER SELF has been away and came back when he awoke. And this belief, which we find among various existing savage tribes, we equally find in the traditions of the early civilized races." 159

Let us consider, for a moment, this assumption of the OTHER SELF, for upon this is based the great mass of crude inference which constitutes the primitive man's philosophy of nature. The hypothesis of the OTHER SELF, which serves to account for the savage's wanderings during sleep in strange lands and among strange people, serves also to account for the presence in his dreams of parents, comrades, or enemies, known to be dead and buried. The other self of the dreamer meets and converses with the other selves of his dead brethren, joins with them in the hunt, or sits down with them to the wild cannibal banquet. Thus arises the belief in an ever-present world of souls or ghosts, a belief which the entire experience of uncivilized man goes to strengthen and expand. The existence of some tribe or tribes of savages wholly destitute of religious belief has often been hastily asserted and as often called in question. But there is no question that, while many savages are unable to frame a conception so general as that of godhood, on the other hand no tribe has ever been found so low in the scale of intelligence as not to have framed the conception of ghosts or spiritual personalities, capable of being angered, propitiated, or conjured with. Indeed it is not improbable a priori that the original inference involved in the notion of the other self may be sufficiently simple and obvious to fall within the capacity of animals even less intelligent than uncivilized man. An authentic case is on record of a Skye terrier who, being accustomed to obtain favours from his master by sitting on his haunches, will also sit before his pet india-rubber ball placed on the chimney-piece, evidently beseeching it to jump down and play with him. 160 Such a fact as this is quite in harmony with Auguste Comte's suggestion that such intelligent animals as dogs, apes, and elephants may be capable of forming a few fetichistic notions. The behaviour of the terrier here rests upon the assumption that the ball is open to the same sort of entreaty which prevails with the master; which implies, not that the wistful brute accredits the ball with a soul, but that in his mind the distinction between life and inanimate existence has never been thoroughly established. Just this confusion between things living and things not living is present throughout the whole philosophy of fetichism; and the confusion between things seen and things dreamed, which suggests the notion of another self, belongs to this same twilight stage of intelligence in which primeval man has not yet clearly demonstrated his immeasurable superiority to the brutes. 161

The conception of a soul or other self, capable of going away from the body and returning to it, receives decisive confirmation from the phenomena of fainting, trance, catalepsy, and ecstasy, 162 which occur less rarely among savages, owing to their irregular mode of life, than among civilized men. "Further verification," observes Mr. Spencer, "is afforded by every epileptic subject, into whose body, during the absence of the other self, some enemy has entered; for how else does it happen that the other self on returning denies all knowledge of what his body has been doing? And this supposition, that the body has been 'possessed' by some other being, is confirmed by the phenomena of somnambulism and insanity." Still further, as Mr. Spencer points out, when we recollect that savages are very generally unwilling to have their portraits taken, lest a portion of themselves should get carried off and be exposed to foul play, 163 we must readily admit that the weird reflection of the person and imitation of the gestures in rivers or still woodland pools will go far to intensify the belief in the other self. Less frequent but uniform confirmation is to be found in echoes, which in Europe within two centuries have been commonly interpreted as the voices of mocking fiends or wood-nymphs, and which the savage might well regard as the utterances of his other self.

With the savage's unwillingness to have his portrait taken, lest it fall into the hands of some enemy who may injure him by conjuring with it, may be compared the reluctance which he often shows toward telling his name, or mentioning the name of his friend, or king, or tutelar ghost-deity. In fetichistic thought, the name is an entity mysteriously associated with its owner, and it is not well to run the risk of its getting into hostile hands. Along with this caution goes the similarly originated fear that the person whose name is spoken may resent such meddling with his personality. For the latter reason the Dayak will not allude by name to the small pox, but will call it "the chief" or "jungle-leaves"; the Laplander speaks of the bear as the "old man with the fur coat"; in Annam the tiger is called "grandfather" or "Lord"; while in more civilized communities such sayings are current as "talk of the Devil, and he will appear," with which we may also compare such expressions as "Eumenides" or "gracious ones" for the Furies, and other like euphemisms. Indeed, the maxim nil mortuis nisi bonum had most likely at one time a fetichistic flavour.

In various islands of the Pacific, for both the reasons above specified, the name of the reigning chief is so rigorously "tabu," that common words and even syllables resembling that name in sound must be omitted from the language. In New Zealand, where a chiefs name was Maripi, or "knife," it became necessary to call knives nekra; and in Tahiti, fetu, "star," had to be changed into fetia, and tui, "to strike," became tiai, etc., because the king's name was Tu. Curious freaks are played with the languages of these islands by this ever-recurring necessity. Among the Kafirs the women have come to speak a different dialect from the men, because words resembling the names of their lords or male relatives are in like manner "tabu." The student of human culture will trace among such primeval notions the origin of the Jew's unwillingness to pronounce the name of Jehovah; and hence we may perhaps have before us the ultimate source of the horror with which the Hebraizing Puritan regards such forms of light swearing—"Mon Dieu," etc.—as are still tolerated on the continent of Europe, but have disappeared from good society in Puritanic England and America. The reader interested in this group of ideas and customs may consult Tylor, Early History of Mankind, pp. 142, 363; Max Muller, Science of Language, 6th edition, Vol. II. p. 37; Mackay, Religious Development of the Greeks and Hebrews, Vol. I. p. 146.

Chamisso's well-known tale of Peter Schlemihl belongs to a widely diffused family of legends, which show that a man's shadow has been generally regarded not only as an entity, but as a sort of spiritual attendant of the body, which under certain circumstances it may permanently forsake. It is in strict accordance with this idea that not only in the classic languages, but in various barbaric tongues, the word for "shadow" expresses also the soul or other self. Tasmanians, Algonquins, Central-Americans, Abipones, Basutos, and Zulus are cited by Mr. Tylor as thus implicitly asserting the identity of the shadow with the ghost or phantasm seen in dreams; the Basutos going so far as to think "that if a man walks on the river-bank, a crocodile may seize his shadow in the water and draw him in." Among the Algonquins a sick person is supposed to have his shadow or other self temporarily detached from his body, and the convalescent is at times "reproached for exposing himself before his shadow was safely settled down in him." If the sick man has been plunged into stupor, it is because his other self has travelled away as far as the brink of the river of death, but not being allowed to cross has come back and re-entered him. And acting upon a similar notion the ailing Fiji will sometimes lie down and raise a hue and cry for his soul to be brought back. Thus, continues Mr. Tylor, "in various countries the bringing back of lost souls becomes a regular part of the sorcerer's or priest's profession." 164 On Aryan soil we find the notion of a temporary departure of the soul surviving to a late date in the theory that the witch may attend the infernal Sabbath while her earthly tabernacle is quietly sleeping at home. The primeval conception reappears, clothed in bitterest sarcasm, in Dante's reference to his living contemporaries whose souls he met with in the vaults of hell, while their bodies were still walking about on the earth, inhabited by devils.

The theory which identifies the soul with the shadow, and supposes the shadow to depart with the sickness and death of the body, would seem liable to be attended with some difficulties in the way of verification, even to the dim intelligence of the savage. But the propriety of identifying soul and breath is borne out by all primeval experience. The breath, which really quits the body at its decease, has furnished the chief name for the soul, not only to the Hebrew, the Sanskrit, and the classic tongues; not only to German and English, where geist, and ghost, according to Max Muller, have the meaning of "breath," and are akin to such words as gas, gust, and geyser; but also to numerous barbaric languages. Among the natives of Nicaragua and California, in Java and in West Australia, the soul is described as the air or breeze which passes in and out through the nostrils and mouth; and the Greenlanders, according to Cranz, reckon two separate souls, the breath and the shadow. "Among the Seminoles of Florida, when a woman died in childbirth, the infant was held over her face to receive her parting spirit, and thus acquire strength and knowledge for its future use..... Their state of mind is kept up to this day among Tyrolese peasants, who can still fancy a good man's soul to issue from his mouth at death like a little white cloud." 165 It is kept up, too, in Lancashire, where a well-known witch died a few years since; "but before she could 'shuffle off this mortal coil' she must needs TRANSFER HER FAMILIAR SPIRIT to some trusty successor. An intimate acquaintance from a neighbouring township was consequently sent for in all haste, and on her arrival was immediately closeted with her dying friend. What passed between them has never fully transpired, but it is confidently affirmed that at the close of the interview this associate RECEIVED THE WITCH'S LAST BREATH INTO HER MOUTH AND WITH IT HER FAMILIAR SPIRIT. The dreaded woman thus ceased to exist, but her powers for good or evil were transferred to her companion; and on passing along the road from Burnley to Blackburn we can point out a farmhouse at no great distance with whose thrifty matron no neighbouring farmer will yet dare to quarrel." 166

Of the theory of embodiment there will be occasion to speak further on. At present let us not pass over the fact that the other self is not only conceived as shadow or breath, which can at times quit the body during life, but is also supposed to become temporarily embodied in the visible form of some bird or beast. In discussing elsewhere the myth of Bishop Hatto, we saw that the soul is sometimes represented in the form of a rat or mouse; and in treating of werewolves we noticed the belief that the spirits of dead ancestors, borne along in the night-wind, have taken on the semblance of howling dogs or wolves. "Consistent with these quaint ideas are ceremonies in vogue in China of bringing home in a cock (live or artificial) the spirit of a man deceased in a distant place, and of enticing into a sick man's coat the departing spirit which has already left his body and so conveying it back." 167 In Castren's great work on Finnish mythology, we find the story of the giant who could not be killed because he kept his soul hidden in a twelve-headed snake which he carried in a bag as he rode on horseback; only when the secret was discovered and the snake carefully killed, did the giant yield up his life. In this Finnish legend we have one of the thousand phases of the story of the "Giant who had no Heart in his Body," but whose heart was concealed, for safe keeping, in a duck's egg, or in a pigeon, carefully disposed in some belfry at the world's end a million miles away, or encased in a wellnigh infinite series of Chinese boxes. 168 Since, in spite of all these precautions, the poor giant's heart invariably came to grief, we need not wonder at the Karen superstition that the soul is in danger when it quits the body on its excursions, as exemplified in countless Indo-European stories of the accidental killing of the weird mouse or pigeon which embodies the wandering spirit. Conversely it is held that the detachment of the other self is fraught with danger to the self which remains. In the philosophy of "wraiths" and "fetches," the appearance of a double, like that which troubled Mistress Affery in her waking dreams of Mr. Flintwinch, has been from time out of mind a signal of alarm. "In New Zealand it is ominous to see the figure of an absent person, for if it be shadowy and the face not visible, his death may erelong be expected, but if the face be seen he is dead already. A party of Maoris (one of whom told the story) were seated round a fire in the open air, when there appeared, seen only by two of them, the figure of a relative, left ill at home; they exclaimed, the figure vanished, and on the return of the party it appeared that the sick man had died about the time of the vision." 169 The belief in wraiths has survived into modern times, and now and then appears in the records of that remnant of primeval philosophy known as "spiritualism," as, for example, in the case of the lady who "thought she saw her own father look in at the church-window at the moment he was dying in his own house."

The belief in the "death-fetch," like the doctrine which identifies soul with shadow, is instructive as showing that in barbaric thought the other self is supposed to resemble the material self with which it has customarily been associated. In various savage superstitions the minute resemblance of soul to body is forcibly stated. The Australian, for instance, not content with slaying his enemy, cuts off the right thumb of the corpse, so that the departed soul may be incapacitated from throwing a spear. Even the half-civilized Chinese prefer crucifixion to decapitation, that their souls may not wander headless about the spirit-world. 171 Thus we see how far removed from the Christian doctrine of souls is the primeval theory of the soul or other self that figures in dreamland. So grossly materialistic is the primitive conception that the savage who cherishes it will bore holes in the coffin of his dead friend, so that the soul may again have a chance, if it likes, to revisit the body. To this day, among the peasants in some parts of Northern Europe, when Odin, the spectral hunter, rides by attended by his furious host, the windows in every sick-room are opened, in order that the soul, if it chooses to depart, may not be hindered from joining in the headlong chase. And so, adds Mr. Tylor, after the Indians of North America had spent a riotous night in singeing an unfortunate captive to death with firebrands, they would howl like the fiends they were, and beat the air with brushwood, to drive away the distressed and revengeful ghost. "With a kindlier feeling, the Congo negroes abstained for a whole year after a death from sweeping the house, lest the dust should injure the delicate substance of the ghost"; and even now, "it remains a German peasant saying that it is wrong to slam a door, lest one should pinch a soul in it." 172 Dante's experience with the ghosts in hell and purgatory, who were astonished at his weighing down the boat in which they were carried, is belied by the sweet German notion "that the dead mother's coming back in the night to suckle the baby she has left on earth may be known by the hollow pressed down in the bed where she lay." Almost universally ghosts, however impervious to thrust of sword or shot of pistol, can eat and drink like Squire Westerns. And lastly, we have the grotesque conception of souls sufficiently material to be killed over again, as in the case of the negro widows who, wishing to marry a second time, will go and duck themselves in the pond, in order to drown the souls of their departed husbands, which are supposed to cling about their necks; while, according to the Fiji theory, the ghost of every dead warrior must go through a terrible fight with Samu and his brethren, in which, if he succeeds, he will enter Paradise, but if he fails he will be killed over again and finally eaten by the dreaded Samu and his unearthly company.

From the conception of souls embodied in beast-forms, as above illustrated, it is not a wide step to the conception of beast-souls which, like human souls, survive the death of the tangible body. The wide-spread superstitions concerning werewolves and swan-maidens, and the hardly less general belief in metempsychosis, show that primitive culture has not arrived at the distinction attained by modern philosophy between the immortal man and the soulless brute. Still more direct evidence is furnished by sundry savage customs. The Kafir who has killed an elephant will cry that he did n't mean to do it, and, lest the elephant's soul should still seek vengeance, he will cut off and bury the trunk, so that the mighty beast may go crippled to the spirit-land. In like manner, the Samoyeds, after shooting a bear, will gather about the body offering excuses and laying the blame on the Russians; and the American redskin will even put the pipe of peace into the dead animal's mouth, and beseech him to forgive the deed. In Assam it is believed that the ghosts of slain animals will become in the next world the property of the hunter who kills them; and the Kamtchadales expressly declare that all animals, even flies and bugs, will live after death,—a belief, which, in our own day, has been indorsed on philosophical grounds by an eminent living naturalist. 173 The Greenlanders, too, give evidence of the same belief by supposing that when after an exhausting fever the patient comes up in unprecedented health and vigour, it is because he has lost his former soul and had it replaced by that of a young child or a reindeer. In a recent work in which the crudest fancies of primeval savagery are thinly disguised in a jargon learned from the superficial reading of modern books of science, M. Figuier maintains that human souls are for the most part the surviving souls of deceased animals; in general, the souls of precocious musical children like Mozart come from nightingales, while the souls of great architects have passed into them from beavers, etc., etc. 174

The practice of begging pardon of the animal one has just slain is in some parts of the world extended to the case of plants. When the Talein offers a prayer to the tree which he is about to cut down, it is obviously because he regards the tree as endowed with a soul or ghost which in the next life may need to be propitiated. And the doctrine of transmigration distinctly includes plants along with animals among the future existences into which the human soul may pass.

As plants, like animals, manifest phenomena of life, though to a much less conspicuous degree, it is not incomprehensible that the savage should attribute souls to them. But the primitive process of anthropomorphisation does not end here. Not only the horse and dog, the bamboo, and the oak-tree, but even lifeless objects, such as the hatchet, or bow and arrows, or food and drink of the dead man, possess other selves which pass into the world of ghosts. Fijis and other contemporary savages, when questioned, expressly declare that this is their belief. "If an axe or a chisel is worn out or broken up, away flies its soul for the service of the gods." The Algonquins told Charlevoix that since hatchets and kettles have shadows, no less than men and women, it follows, of course, that these shadows (or souls) must pass along with human shadows (or souls) into the spirit-land. In this we see how simple and consistent is the logic which guides the savage, and how inevitable is the genesis of the great mass of beliefs, to our minds so arbitrary and grotesque, which prevail throughout the barbaric world. However absurd the belief that pots and kettles have souls may seem to us, it is nevertheless the only belief which can be held consistently by the savage to whom pots and kettles, no less than human friends or enemies, may appear in his dreams; who sees them followed by shadows as they are moved about; who hears their voices, dull or ringing, when they are struck; and who watches their doubles fantastically dancing in the water as they are carried across the stream. 175 To minds, even in civilized countries, which are unused to the severe training of science, no stronger evidence can be alleged than what is called "the evidence of the senses"; for it is only long familiarity with science which teaches us that the evidence of the senses is trustworthy only in so far as it is correctly interpreted by reason. For the truth of his belief in the ghosts of men and beasts, trees and axes, the savage has undeniably the evidence of his senses which have so often seen, heard, and handled these other selves.

The funeral ceremonies of uncultured races freshly illustrate this crude philosophy, and receive fresh illustration from it. On the primitive belief in the ghostly survival of persons and objects rests the almost universal custom of sacrificing the wives, servants, horses, and dogs of the departed chief of the tribe, as well as of presenting at his shrine sacred offerings of food, ornaments, weapons, and money. Among the Kayans the slaves who are killed at their master's tomb are enjoined to take great care of their master's ghost, to wash and shampoo it, and to nurse it when sick. Other savages think that "all whom they kill in this world shall attend them as slaves after death," and for this reason the thrifty Dayaks of Borneo until lately would not allow their young men to marry until they had acquired some post mortem property by procuring at least one human head. It is hardly necessary to do more than allude to the Fiji custom of strangling all the wives of the deceased at his funeral, or to the equally well-known Hindu rite of suttee. Though, as Wilson has shown, the latter rite is not supported by any genuine Vedic authority, but only by a shameless Brahmanic corruption of the sacred text, Mr. Tylor is nevertheless quite right in arguing that unless the horrible custom had received the sanction of a public opinion bequeathed from pre-Vedic times, the Brahmans would have had no motive for fraudulently reviving it; and this opinion is virtually established by the fact of the prevalence of widow sacrifice among Gauls, Scandinavians, Slaves, and other European Aryans. 176 Though under English rule the rite has been forcibly suppressed, yet the archaic sentiments which so long maintained it are not yet extinct. Within the present year there has appeared in the newspapers a not improbable story of a beautiful and accomplished Hindu lady who, having become the wife of a wealthy Englishman, and after living several years in England amid the influences of modern society, nevertheless went off and privately burned herself to death soon after her husband's decease.

The reader who thinks it far-fetched to interpret funeral offerings of food, weapons, ornaments, or money, on the theory of object-souls, will probably suggest that such offerings may be mere memorials of affection or esteem for the dead man. Such, indeed, they have come to be in many countries after surviving the phase of culture in which they originated; but there is ample evidence to show that at the outset they were presented in the belief that their ghosts would be eaten or otherwise employed by the ghost of the dead man. The stout club which is buried with the dead Fiji sends its soul along with him that he may be able to defend himself against the hostile ghosts which will lie in ambush for him on the road to Mbulu, seeking to kill and eat him. Sometimes the club is afterwards removed from the grave as of no further use, since its ghost is all that the dead man needs. In like manner, "as the Greeks gave the dead man the obolus for Charon's toll, and the old Prussians furnished him with spending money, to buy refreshment on his weary journey, so to this day German peasants bury a corpse with money in his mouth or hand," and this is also said to be one of the regular ceremonies of an Irish wake. Of similar purport were the funeral feasts and oblations of food in Greece and Italy, the "rice-cakes made with ghee" destined for the Hindu sojourning in Yama's kingdom, and the meat and gruel offered by the Chinaman to the manes of his ancestors. "Many travellers have described the imagination with which the Chinese make such offerings. It is that the spirits of the dead consume the impalpable essence of the food, leaving behind its coarse material substance, wherefore the dutiful sacrificers, having set out sumptuous feasts for ancestral souls, allow them a proper time to satisfy their appetite, and then fall to themselves." 177 So in the Homeric sacrifice to the gods, after the deity has smelled the sweet savour and consumed the curling steam that rises ghost-like from the roasting viands, "the assembled warriors devour the remains." 178

Thus far the course of fetichistic thought which we have traced out, with Mr. Tylor's aid, is such as is not always obvious to the modern inquirer without considerable concrete illustration. The remainder of the process, resulting in that systematic and complete anthropomorphisation of nature which has given rise to mythology, may be more succinctly described. Gathering together the conclusions already obtained, we find that daily or frequent experience of the phenomena of shadows and dreams has combined with less frequent experience of the phenomena of trance, ecstasy, and insanity, to generate in the mind of uncultured man the notion of a twofold existence appertaining alike to all animate or inanimate objects: as all alike possess material bodies, so all alike possess ghosts or souls. Now when the theory of object-souls is expanded into a general doctrine of spirits, the philosophic scheme of animism is completed. Once habituated to the conception of souls of knives and tobacco-pipes passing to the land of ghosts, the savage cannot avoid carrying the interpretation still further, so that wind and water, fire and storm, are accredited with indwelling spirits akin by nature to the soul which inhabits the human frame. That the mighty spirit or demon by whose impelling will the trees are rooted up and the storm-clouds driven across the sky should resemble a freed human soul, is a natural inference, since uncultured man has not attained to the conception of physical force acting in accordance with uniform methods, and hence all events are to his mind the manifestations of capricious volition. If the fire burns down his hut, it is because the fire is a person with a soul, and is angry with him, and needs to be coaxed into a kindlier mood by means of prayer or sacrifice. Thus the savage has a priori no alternative but to regard fire-soul as something akin to human-soul; and in point of fact we find that savage philosophy makes no distinction between the human ghost and the elemental demon or deity. This is sufficiently proved by the universal prevalence of the worship of ancestors. The essential principle of manes-worship is that the tribal chief or patriarch, who has governed the community during life, continues also to govern it after death, assisting it in its warfare with hostile tribes, rewarding brave warriors, and punishing traitors and cowards. Thus from the conception of the living king we pass to the notion of what Mr. Spencer calls "the god-king," and thence to the rudimentary notion of deity. Among such higher savages as the Zulus, the doctrine of divine ancestors has been developed to the extent of recognizing a first ancestor, the Great Father, Unkulunkulu, who made the world. But in the stratum of savage thought in which barbaric or Aryan folk-lore is for the most part based, we find no such exalted speculation. The ancestors of the rude Veddas and of the Guinea negroes, the Hindu pitris (patres, "fathers"), and the Roman manes have become elemental deities which send rain or sunshine, health or sickness, plenty or famine, and to which their living offspring appeal for guidance amid the vicissitudes of life. 179 The theory of embodiment, already alluded to, shows how thoroughly the demons which cause disease are identified with human and object souls. In Australasia it is a dead man's ghost which creeps up into the liver of the impious wretch who has ventured to pronounce his name; while conversely in the well-known European theory of demoniacal possession, it is a fairy from elf-land, or an imp from hell, which has entered the body of the sufferer. In the close kinship, moreover, between disease-possession and oracle-possession, where the body of the Pythia, or the medicine-man, is placed under the direct control of some great deity, 180 we may see how by insensible transitions the conception of the human ghost passes into the conception of the spiritual numen, or divinity.

To pursue this line of inquiry through the countless nymphs and dryads and nixies of the higher nature-worship up to the Olympian divinities of classic polytheism, would be to enter upon the history of religious belief, and in so doing to lose sight of our present purpose, which has merely been to show by what mental process the myth-maker can speak of natural objects in language which implies that they are animated persons. Brief as our account of this process has been, I believe that enough has been said, not only to reveal the inadequacy of purely philological solutions (like those contained in Max Muller's famous Essay) to explain the growth of myths, but also to exhibit the vast importance for this purpose of the kind of psychological inquiry into the mental habits of savages which Mr. Tylor has so ably conducted. Indeed, however lacking we may still be in points of detail, I think we have already reached a very satisfactory explanation of the genesis of mythology. Since the essential characteristic of a myth is that it is an attempt to explain some natural phenomenon by endowing with human feelings and capacities the senseless factors in the phenomenon, and since it has here been shown how uncultured man, by the best use he can make of his rude common sense, must inevitably come, and has invariably come, to regard all objects as endowed with souls, and all nature as peopled with supra-human entities shaped after the general pattern of the human soul, I am inclined to suspect that we have got very near to the root of the whole matter. We can certainly find no difficulty in seeing why a water-spout should be described in the "Arabian Nights" as a living demon: "The sea became troubled before them, and there arose from it a black pillar, ascending towards the sky, and approaching the meadow,.... and behold it was a Jinni, of gigantic stature." We can see why the Moslem camel-driver should find it most natural to regard the whirling simoom as a malignant Jinni; we may understand how it is that the Persian sees in bodily shape the scarlet fever as "a blushing maid with locks of flame and cheeks all rosy red"; and we need not consider it strange that the primeval Aryan should have regarded the sun as a voyager, a climber, or an archer, and the clouds as cows driven by the wind-god Hermes to their milking. The identification of William Tell with the sun becomes thoroughly intelligible; nor can we be longer surprised at the conception of the howling night-wind as a ravenous wolf. When pots and kettles are thought to have souls that live hereafter, there is no difficulty in understanding how the blue sky can have been regarded as the sire of gods and men. And thus, as the elves and bogarts of popular lore are in many cases descended from ancient divinities of Olympos and Valhalla, so these in turn must acknowledge their ancestors in the shadowy denizens of the primeval ghost-world.

August, 1872.