WE have now arrived at a point where we can more fully understand those curious ideas of sacrifice and sacrament which lie at the root of so much that is essential in the Jewish, the Christian, and most other religions.

Mr. Galton tells us that to the Damaras, when he travelled among them, all meat was common property. No one killed an ox except as a sacrifice and on a festal occasion; and when the ox was killed, the whole community feasted upon it indiscriminately. This is but a single instance of a feeling almost universal among primitive pastoral people. Cattle and other domestic animals, being regarded as sacred, are rarely killed; and when they are killed, they are eaten at a feast as a social and practically religious rite—in short, sacramentally. I need not give instances of so well-known a principle; I will content myself with quoting what Dr. Robertson Smith says of a particular race: “Among the early Semites generally, no slaughter was legitimate except for sacrifice.”

Barbaric herdsmen, indeed, can hardly conceive of men to whom flesh meat is a daily article of diet. Mr. Galton found the idea very strange to his Damaras. Primitive pastoral races keep their domestic animals mainly for the sake of the milk, or as beasts of burden, or for the wool and hair; they seldom kill one except for a feast, at which the gods are fellow-partakers. Indeed, it is probable, as the sequel will suggest, that domestic animals were originally kept as totems or ancestor-gods, and that the habit of eating 319the meat of sheep, goats, and oxen has arisen mainly out of the substitution of such a divine animal-victim for the divine human-victim of earlier usage. Our butchers’ shops have their origin in mitigated sacrificial cannibalism.

Sacrifice, regarded merely as offering to the gods, has thus, I believe, two distinct origins. Its earliest, simplest, and most natural form is that whose development we have already traced,—the placing of small articles of food and drink at the graves of ancestors or kings or revered fellow-tribesmen. That from a very early period men have believed the dead to eat and drink, whether as corpse, as mummy, as ghost of buried friend, or as ethereal spirit of cremated chieftain, we have already seen with sufficient frequency. About the origin of these simplest and most primitive sacrifices, I think, there can be little doubt. Savages offer at the graves of their dead precisely those ordinary articles of food which they consumed while living, without any distinction of kind; and they continue to offer them in the same na茂ve way when the ghost has progressed to the status of one of the great gods of the community.

But there is another mode of sacrifice, superposed upon this, and gradually tending to be more or less identified with it, which yet, if I am right, had a quite different origin in the artificial production of gods about which I have written at considerable length in the last three chapters. The human or animal victim, thus slaughtered in order to make a new god or protecting spirit, came in time to be assimilated in thought to the older type of mere honorific offerings to the dead gods; and so gave rise to those mystic ideas of the god who is sacrificed, himself to himself, of which the sacrament of the Mass is the final and most mysterious outcome. Thus, the foundation-gods, originally killed in order to make a protecting spirit for a house or a tribal god for a city or village, came at last to be regarded as victims sacrificed to the Earth Goddess or to the Earth Demons; and thus, too, the Meriahs and other agricultural 320victims, originally killed in order to make a corn-god or a corn-spirit, came at last to be regarded as sacrifices to the Earth, or to some abstract Dionysus or Attis or Adonis. And since in the last case at least the god and the victim were still called by the same name and recognised as one, there grew up at last in many lands, and in both hemispheres, but especially in the Eastern Mediterranean basin, the mystic theory of the sacrifice of a god, himself to himself, in atonement or expiation, which forms the basis of the Christian Plan of Salvation. It is this secondary and derivative form of sacrifice, I believe, which is mainly considered in Professor Robertson Smith’s elaborate and extremely valuable analysis.

I have said that the secondary form of sacrifice, which for brevity’s sake I shall henceforth designate as the mystic, is found in most parts of the world and in both hemispheres. This naturally raises the question whether it has a single common origin, and antedates the dispersal of mankind through the hemispheres; or whether it has been independently evolved several times over in many lands by many races. For myself, I have no cut-and-dried answer to this abstruse question, nor do I regard it, indeed, as a really important one. On the one hand, there are many reasons for supposing that certain relatively high traits of thought or art were common property among mankind before the dispersion from the primitive centre, if a primitive centre ever existed. On the other hand, psychologists know well that the human mind acts with extraordinary similarity in given circumstances all the world over, and that identical stages of evolution seem to have been passed through independently by many races, in Egypt and Mexico, in China and Peru; so that we can find nothing inherently improbable in the idea that even these complex conceptions of mystic sacrifice have distinct origins in remote countries. What is certain is the fact that among the Aztecs, as among the Phrygians, the priest who sacrificed, the victim he slew, and the image or great god to 321whom he slew him, were all identified; the killer, the killed, and the being in whose honor the killing took place were all one single indivisible deity. Even such details as that the priest clothed himself in the skin of the victim are common to many lands; they may very well be either a heritage from remote ancestral humanity, or the separate product of the human mind, working along like grooves under identical conditions. In one word they may perhaps be necessary and inevitable corollaries from antecedent conceptions.

I must further premise that no religion as we now know it is by any means primitive. The most savage creeds we find among us have still hundreds of thousands of years behind them. The oldest religions whose records have descended to us, like those of Egypt and of Assyria, are still remote by hundreds of thousands of years from the prime original. Cultivation itself is a very ancient and immemorial art. Few savages, even among those who are commonly described as in the hunting stage, are wholly ignorant of some simple form of seed-sowing and tillage. The few who are now ignorant of those arts show some apparent signs of being rather degenerate than primitive peoples. My own belief or suspicion is that ideas derived from the set of practices in connexion with agriculture detailed in the last two chapters have deeply coloured the life and thought of almost the whole human race, including even those rudest tribes which now know little or nothing of agriculture. But I do not lay stress upon this half-formed conviction, to justify which would lead me too far afield. I shall be content with endeavouring to suggest how far they have coloured the ideas of the greater number of existing nations.

Early pastoral races seldom kill a beast except on great occasions. When they kill it, they devour it in common, all the tribe being invited to the festival. But they also eat it in fellowship with their gods; every great feast is essentially a Theoxenion, a Lectisternium, a banquet in which the 322deities participate with mortals. It is this sense of a common feast of gods and men which gave, no doubt, the first step towards the complex idea of the sacramental meal—an idea still further developed at a later stage by the addition of the concept that the worshipper eats and drinks the actual divinity.

My own belief is that all sacrificial feasts of this godeating character most probably originated in actual cannibalism; and that later an animal victim was substituted for the human meat; but I do not insist on this point, nor attempt, strictly speaking, to prove it. It is hardly more than a deeply grounded suspicion. Nevertheless, I will begin for convenience’ sake with the cannibal class of sacrifice, and will come round in time to the familiar slaughter of sheep and oxen, which in many cases is known to have supplanted a human offering.

Acosta’s account of the Mexican custom is perhaps the best instance we now possess of the ritual of cannibal mystic sacrifice in its fullest barbarity. “They took a captive,” says that racy old author, “at random; and before sacrificing him to their idols, they gave him the name of the idol to whom he should be sacrificed, and dressed him in the same ornaments, identifying him with the god. During the time that the identification lasted, which was for a year in some feasts, six months or less in others, they reverenced and worshipped him in the same manner as the idol itself. Meanwhile, he was allowed to eat, drink, and make merry. When he went through the streets, the people came forth to worship him; and every one brought alms, with children and sick people that he might cure them and bless them. He did as he pleased in everything except that he had ten or twelve men about him, to prevent him from escaping. In order that he might be reverenced as he passed, he sometimes sounded upon a small flute, to tell the people to worship him. When the feast arrived, and he had grown fat, they killed him, opened him, and making a solemn sacrifice, eat him.” There, in the words 323of a competent authority, we have the simple cannibal feast in its fullest nakedness.

I need hardly point out how much this account recalls the Khond custom of the Meriah. The victim, though not really of royal blood, is made artificially into a divine king; he is treated with all the honours of royalty and godhead, is dressed like the deity with whom he is identified, and is finally killed and eaten. The last point alone differs in any large degree from the case of the Meriah. We have still to enquire, “Why did they eat him?”

The answer to this enquiry takes us into the very heart and core of the sacramental concept.

It is a common early belief that to eat of any particular animal gives you the qualities of that animal. The Miris of Northern India prize tiger’s flesh for men; it gives them strength and courage; but women must not eat it; ‘twould make them “too strong-minded.” The Namaquas abstain from eating hare; they would become faint-hearted if they swallowed it; but they eat the meat of the lion or drink the blood of the leopard, in order to gain their strength and courage. Among the Dyaks, young men and warriors must not eat deer; it would render them cowardly; but women and very old men are allowed to eat it. Men of the Buro and Aru Islands feed on the flesh of dogs in order to be bold and nimble. Mr. Frazer has collected an immense number of similar instances, which show both how widespread and how deep-seated are such beliefs. Even scrapings of the bones are sufficient to produce the desired result; in Corea, the bones of tigers fetch a higher price than those of leopards as inspirers of courage. The heart of a lion is also particularly good for this purpose; and the tongues of birds are recommended for eloquence.

Again, on the same analogy, the flesh and blood of brave men are eaten in order to inspire bravery. The Australian Kamilaroi eat the heart and liver of a valiant warrior in order to acquire his courage. The Philippine Islanders drink the blood of their bravest enemies. In the Shire Highlands 324of Africa, those who kill a distinguished fighter eat his heart to get his courage. Du Chaillu’s negro, attendants, we saw, scraped their ancestors’ skulls, and drank the powder in water. “Our ancestors were brave,” said they; “and by drinking their skulls, we shall be brave as they were.” Here again I can only refer the reader for numerous examples to Mr. Frazer’s inexhaustible storehouse.

The case of Du Chaillu’s warriors, however, takes us with one bound into the heart of the subject. Many savages for similar reasons actually eat their own dead fathers. * We learn from Strabo that the ancient Irish “deemed it honourable to devour the bodies of their parents.” So, Herodotus tells us, did the Issedones of Central Asia. The Massaget忙 used “from compassion” to club and eat their aged people. The custom was quite recently common among the Battas of Sumatra, who used “religiously and ceremonially to eat their old relations.” In Australia, it was usual to eat relatives who died by mischance. Of the Cucumas we read that “as soon as a relation died, these people assembled and eat him roasted or boiled, according as he was thin or fat.” The Tarianas and Tucanas, who drink the ashes of their relatives, “believe that thus the virtues of the deceased will be transmitted to the drinkers.” The Arawaks think it the highest mark of honour they could pay to the dead to drink their powdered bones mixed in water. Generally speaking, in a large number of cases, the parents or relatives were eaten in order “not to let the life go out of the family” or to preserve the bodies and souls in a kindred body; or to gain the courage and other qualities of the dead relation. In short, the dead were eaten sacramentally or, as one writer even phrases it, “eucharistically.” Mr. Hartland has collected many striking instances.

     * Since this chapter was written, the subject of honorific
     cannibalism has been far more fully treated by Mr. Sidney
     Hartland in the chapter on Funeral Rites, in the second
     volume of The Legend of Perseus.

How this strange custom originates we may guess from Mr. 325Wyatt Gill’s description of a New Guinea funeral. “The women lacerated their faces and beat their breasts most affectingly,” he says; “and then, in the madness of their grief, pressed the matter out of the wounded thigh, and smeared it over their faces and persons, and even licked it up.” Of the Koiari corpses he says: “A fire is kept burning day and night at the head and feet for months. The entire skin is removed by means of the thumb and forefinger, and the juices plastered all over the face and body of the operator,—parent, husband, or wife of the deceased. The fire gradually desiccates the flesh, so that little more than the skeleton is left.” This naturally leads on to eating the dead, which indeed is practised elsewhere in New Guinea.

But if men eat the bodies of their fathers, who are their family and household gods, they will also naturally eat the bodies of the artificial gods of cultivation, or of the temporary kings who die for the people. By eating the body of a god, you absorb his divinity; he and you become one; he is in you and inspires you. This is the root-idea of sacramental practice; you eat your god by way of complete union; you subsume him in yourself; you and he are one being.

Still, how can you eat your god if you also bury him as a corn-spirit to use him as seed? The Gonds supply us with the answer to that obvious difficulty. For, as we saw, they sprinkle the blood of the victim over the ploughed field or ripe crop, and then they sacramentally devour his body. Such a double use of the artificial god is frequently to be detected, indeed, through the vague words of our authorities. We see it in the Potraj ceremony, where the blood of the lamb is drunk by the officiating priest, while the remainder of the animal is buried beside the altar; we see it in the numerous cases where a portion of the victim is eaten sacramentally, and the rest burned and scattered over the fields, which it is supposed to fertilise. You eat your god in part, so as to imbibe his divinity; but you bury 326him in part, so as to secure at the same time his fertilising qualities for your corn or your vineyard.

I admit that all this is distinctly mystic; but mystery-mongering and strange reduplication of persons, with marvellous identifications and minute distinctions, have always formed much of the stock-in-trade of religion. If cults were all plain sailing throughout, what room for faith?—there would be less to engage the imagination of the votary.

And now let us return awhile to our Mexican instances.

At the annual feast of the great god Tezcatlipoca, which, like most similar festivals, fell about the same time as the Christian Easter, a young man was chosen to be the representative of the god for a twelvemonth. As in the case of almost all chosen victims, he had to be a person of unblemished body, and he was trained to behave like a god-king with becoming dignity. During his year of godship, he was lapped in luxury; and the actual reigning emperor took care that he should be splendidly attired, regarding him already as a present deity. He was attended by eight pages clad in the royal livery—which shows him to have been a king as well as a god; and wherever he went the people bowed down to him. Twenty days before the festival at which he was to be sacrificed, four noble maidens, bearing the names of four goddesses, were given him to be his brides. The final feast itself, like those of Dionysus, of Attis, and of Potraj, occupied five days—a coincidence between the two hemispheres which almost points to original identity of custom before the dispersion of the races. During these five days the real king remained in his palace—and this circumstance plainly shows that the victim belonged to the common class of substituted and temporary divine king-gods. The whole court, on the other hand, attended the victim. On the last day of the feast, the victim was ferried across the lake in a covered barge to a small temple in the form of a pyramid. On reaching the summit, he was seized and held down on a block of stone,—no doubt 327an altar of funereal origin,—while the priest cut open his breast with a stone knife, and plucked his heart out. This he offered to the god of the sun. The head was hung up among the skulls of previous victims, no doubt for oracular purposes, and as a permanent god; but the legs and arms were cooked and prepared for the table of the lords, who thus partook of the god sacramentally. His place was immediately filled by another young man, who for a year was treated with the same respect, and at the end of that time was similarly slaughtered.

I do not think I need point out the close resemblance of this ritual to that of the Khond Meriah, of the Potraj, and of the festivals of Dionysus, Osiris, Attis, and Adonis. But I would also call particular attention to the final destination of the skull, and its exact equivalence to the skull of the animal-god in India and elsewhere.

“The idea that the god thus slain in the person of his representative comes to life again immediately,” says Mr. Frazer, “was graphically represented in the Mexican ritual by skinning the slain man-god, and clothing in his skin a living man, who thus became the new representative of the godhead.” For example, at an annual festival in Mexico, a woman was sacrificed who represented Toci, the Mother of the Gods—a sort of yearly Mexican Cybele. She was dressed in the ornaments and bore the name of the goddess of whom she was believed to be an incarnation. After being feasted for several days, she was taken at midnight to the summit of a temple, and there beheaded. Her body was flayed, and one of the priests, clothing himself in the skin, became the representative of the goddess Toci. The skin of the woman’s thigh, however, was separately removed, and a young man who represented the god Cinteotl, the son of Toci, wrapt it round him like a mask. Ceremonies then followed, in which the two men, clad in the woman’s skin, enacted the parts of the god and goddess. In all this, there is much that seems to me reminiscent 328of Isis and Horus, of Cybele and Attis, of Semele and Dionysus, and of several other eastern rituals.

Still more significant is the yearly festival of the god Totec, who was represented in like manner by a priest, clad in the skin of a human victim, and who received offerings of first-fruits and first-flowers, together with bunches of maize which had been kept for seed. Here we have the closest possible analogy to the case of the Meriah. The offering of first-fruits, made sometimes to the king, sometimes to the ancestral spirits, is here made to the human god of cultivation, who represents both in his own person.

Many other cannibal sacrifices are recorded in Mexico: in more than one of them it was customary for the priest to tear out the warm throbbing heart of the victim, and present it to the idol. Whether these sacrifices in each particular case were of the ordinary or of the mystic type it is not always quite easy to decide; probably the worshippers themselves did not accurately discriminate in every instance. But however that may be, we know at least this much: when human sacrifices had been rare, the priests reminded the kings that the gods “were starving with hunger”; war was then made on purpose to take prisoners, “because the gods had asked for something to eat,” and thousands of victims were thus slaughtered annually. The blood of the victims was separately offered; and I may add in this connexion that as a rule both ghosts and gods are rather thirsty than hungry. I take the explanation of this peculiar taste to be that blood and other liquids poured upon the ground of graves or at altar-stones soon sink in, and so seem to have been drunk or sucked up by the ghost or god; whereas meat and solid offerings are seen to be untouched by the deity to whom they are presented. A minor trait in this blood-loving habit of the gods is seen in the fact that the Mexicans also gave the god to drink fresh blood drawn from their own ears, and that the priests likewise drew blood from their legs, and daubed it on the temples. Similar mitigations of self-immolation 329are seen elsewhere in the Attis-priest drawing blood from his arms for Attis, in the Hebrew Baal-priests “cutting themselves for Baal,” and in the familiar Hebrew rite of circumcision. Blood is constantly drawn by survivors or worshippers as an act of homage to the dead or to deities.

I might multiply instances of human sacrifices of the mystic order elsewhere, but I prefer to pass on to the various mitigations which they tend to undergo in various communities. In its fullest form, I take it, the mystic sacrifice ought to be the self-immolation of a divine priest-king, a god and descendant of gods, himself to himself, on the altar of his own divine foundation-ancestor. But in most cases which we can trace, the sacrifice has already assumed the form of an immolation of a willing victim, a temporary king, of the divine stock only by adoption, though sometimes a son or brother of the actual monarch. Further modifications are that the victim becomes a captive taken in war (which indeed is implied in the very etymology of the Latin word victima), or a condemned criminal, or an imbecile, who can be more readily induced to undertake the fatal office. Of all of these we have seen hints at least in previous cases. Still more mitigated are the forms in which the victim is allowed to escape actual death by a subterfuge, and those in which an image or effigy is allowed to do duty for the living person. Of these intermediates we get a good instance in the case of the Bhagats, mentioned by Col. Dalton, who “annually make an image of a man in wood, put clothes and ornaments on it, and present it before the altar of a Mahadeo” (or rude stone phallic idol). “The person who officiates as priest on the occasion says, ‘O, Mahadeo, we sacrifice this man to you according to ancient customs. Give us rain in due season, and a plentiful harvest.’ Then, with one stroke of the axe, the head of the image is struck off, and the body is removed and buried.” This strange rite shows us a surviving 330but much mitigated form of the Khond Meriah practice.

As a rule, however, such bloodless representations do not please the gods; nor do they succeed in really liberating a ghost or corn-god. They are after all but feeble phantom sacrifices. Blood the gods want, and blood is given them. The most common substitute for the human victim-god is therefore the animal victim-god, of which we have already seen copious examples in the ox and kid of Dionysus, the pig of Attis, and many others. It seems probable that a large number of sacrifices, if not the majority of those in which domestic animals are slain, belong in the last resort to the same category. Thus, indeed, we can most easily explain the theory of the so-called “thean-thropic” victim,—the animal which stands for a man and a god,—as well as the point of view of sacrifice so ably elaborated by Dr. Robertson Smith.

According to this theory, the domestic animals were early regarded as of the same kin or blood as the tribe; and the slaughter of an ox, a goat, or a sheep could only be permitted if it were done, like the slaughter of a king’s son, sacrificially and sacramentally. In my own opinion, this scarcely means more than that the sacred domestic animals were early accepted as substitutes for the human victim, and that they were eaten sacrificially and sacramentally as the human victim was also eaten. But I will waive this somewhat controversial point, and content myself with suggesting that the animal victim was habitually treated as in itself divine, and that its blood was treated in the same way as the blood of the original cannibal offering. At the same time, the sacrifice was usually offered at the altar of some older and, so to speak, more constant deity, while the blood of the victim was allowed to flow over the sacred stone. Certainly, both among the Arabs and the Hebrews, all slaughter of domestic animals appears to have been at one time sacrificial; and even when the slaughter ceased necessarily to involve a formal sacrifice, it was still thought 331necessary to slay the victim in the name of a god, and to pour out the blood in his honour on the ground. Even in the Gr忙co-Roman world, the mass of butcher’s meat was “meat offered to idols.” We shall see hereafter that among existing savages the slaughter of domestic animals is still regarded as a sacred rite.

I believe also that as a rule the blood-offering is the earliest and commonest form of slaughter to the gods; and that the victim in the earlier stages was generally consumed by the communicants, as we know the cannibal victim to have been consumed among the Mexicans, and as we saw the theanthropic goat or kid was orgiastically devoured by the worshippers of Dionysus. It is a detail whether the sacred victim happened to be eaten raw or cooked; the one usage prevailed in the earlier and more orgiastic rites, the other in the milder and more civilised ceremonies. But in either case, the animal-god, like the human god, was eaten sacramentally by all his worshippers, who thus took into themselves his divine qualities. The practice of burning the victim, on the other hand, prevailed mainly, I think, among cremationists, like the Tyrians and Hellenes, though it undoubtedly extended also to many burying peoples, like the Hebrews and Egyptians. In most cases even of cremated victims, it would appear, a portion at least of the animal was saved from the fire and sacramentally eaten by the worshippers.

Once more, the victim itself was usually a particular kind of sacred animal. This sacredness of the chosen beast has some more important bearings than we have yet considered. For among various pastoral races, various domesticated animals possess in themselves positive sanctity. We know, for example, that cows are very holy in the greater part of India, and buffaloes in the Deccan. Among the African peoples of the pastoral tribes, the common food is milk and game; cattle are seldom slaughtered merely to eat, and always on exceptional or sacred occasions—the very occasions which elsewhere demand a human 332victim—such as the proclamation of a war, a religious festival, a wedding, or the funeral of a great chieftain. In such cases, the feast is public, all blood-relations having a natural right to attend. The cattle-kraal itself is extremely sacred. The herd and its members are treated by their masters with affectionate and almost............