festivals on Midsummer Eve or Midsummer Day which survive so generally all over Europe,1 and the African custom of worshipping the moon by shooting flaming arrows towards her.2 Mr. Frazer, however, who apparently inclines to regard religion as developed out of magic, consistently enough says, "The best general explanation of these European fire-festivals seems to be the one given by Mannhardt, namely, that they are sun-charms or magical ceremonies intended to ensure a proper supply of sunshine for men, animals, and plants; and, following Mannhardt, he also explains the custom of burning the representative of the vegetation spirit as a piece of sympathetic magic, having the same object as the Midsummer bonfires. But sympathetic magic implies that an effect is produced in virtue of the similarity between that effect and its cause, and without the intervention of any supernatural being-there is nothing religious about it. Now, neither is there anything religious in the Midsummer rites as at present practised by European peasants; but then these rites are survivals, and in religion a survival consists in the continued performance of acts, originally having a religious significance, after all religious significance has departed from them. Thus no one doubts that streams and wells were once considered supernatural powers, or the abodes of supernatural spirits, having, amongst other powers, that of curing disease. Nor can it be doubted that originally the worshipper placed himself in contact with, and under the protection of, the spirit by bathing in the water. That the "sacred" wells, which are common enough now, were originally worshipped as gods is tolerably clear. But the practice of resorting to them is now a survival-it is, in the proper sense of the word, a superstition; that is to say, those who believe that water from a certain well will cure diseases of the eye, believe so, not because they suppose any spirit to dwell in the water, but simply because it is the tradition that that water does, as a matter of fact, cure eye-disease. But it would be erroneous to infer that, because now no spirit is supposed to effect the cure, therefore the belief never had a religious element in it; and in the same way it is not safe to

1 For instances, see Frazer, G. B. ii. 58 ff., and Mannhardt, W. F. K. 309. Réville, Peuples non-civilisés, i. 58. 3 G. B. ii. 267-8.

infer that because there is now no element of religion in the Midsummer rites, therefore there never was. Rather, I would suggest, the inference is that the fire-festivals, occurring as they do at the summer solstice, are like other festivals occurring on that day, survivals of early sun-worship; while the burning of the vegetation spirit's representative is the early cultivator's method of commending his crops to the sunspirit, as immersion is his method of placing them under the care of the sky-spirit or rain-god. On the other hand, if

we regard these fire-festivals and water-rites as pieces of sympathetic magic, they are clear instances in which man imagines himself able to constrain the gods-in this case the god of vegetation-to subserve his own ends. Now, this vain imagination is not merely non-religious, but antireligious; and it is difficult to see how religion could have been developed out of it. It is inconsistent with the abject fear which the savage feels of the supernatural, and which is sometimes supposed to be the origin of religion; and it is inconsistent with that sense of man's dependence on a superior being which is a real element in religion.

CHAPTER XVIII

SYNCRETISM AND POLYTHEISM

THE material progress made by man, as he advanced from the material basis of subsistence on roots, fruits, and the chase, first to pastoral and then to agricultural life, required that he should make an ever-increasing use for his own ends of natural forces. These forces were to him living beings with superhuman powers, of whom he stood in dread, but whose co-operation he required. Without some confidence that it was possible, if he set about it in the right way, to secure their favour and assistance, his efforts would have been paralysed. That confidence was given him by religion; he was brought into friendly relations with powers from which, in his previously narrow circle of interests, he had had little to hope or to gain; and thus the number of his gods had been increased.

Pastoral life and even a rudimentary form of agriculture are compatible with a wandering mode of existence, in which the sole ties that can keep society together are the bonds of blood-kinship and a common cult. But the development of agriculture is only possible when the tribe is permanently settled in a fixed abode; and then it becomes possible for neighbours, not of kindred blood, to unite in one community. In a word, political progress becomes possible; and political progress at this stage consists in the fusion or synoikismos of several tribes into a single State. This process also had its effect upon religion: a clan is a religious community as well as a body of kinsmen, and the fusion of two clans implied the fusion of their respective cults. In many cases the resemblances of the two cults may well have been so great as rather to promote than hinder the alliance; thus when

we find, as occasionally happens, that in some villages two May-poles (survivals of tree-worship) are used at a harvest festival instead of one, the inference 1 rightly seems to be that two communities, both worshipping trees, if not the same species of tree, have in neighbourliness united their worship. Or, again, when we find that the branch which the treeworshipper annually carries round the community, in order that the spirit present in it may confer blessings on all to whom it is presented, is hung with various kinds of fruits and associated with cereals, we may infer that tree-worshipper and plant-worshipper have found no difficulty in uniting in a joint festival and common act of worship. So, too, in the Lithuanian Samborios, the Athenian Pyanepsion, and the Mexican offering to Chicomecoatl, the common feature is that cereals and leguminous plants of all kind are combined in one offering; and the implication is that the festival was one common to all the cultivators and worshippers of the various plants represented in the offering.

Again, two communities might happen to agree, though for different reasons, in offering the same kind of animal X in their annual sacrifice. Thus the moon-worshipper seems very generally to have believed that the moon-spirit manifested herself on earth in the shape of a cow, and that a cow was therefore the proper victim to offer, on the principle that the deity is to be offered to himself. A fusion, therefore, between a family of moon-worshippers and a family whose original totem and traditional deity was the cow, would meet with no difficulty on the ground of religion, if prompted by neighbourliness or political reasons. So, too, the clan that bred horses would be prepared to recognise fellow-worshippers in a clan that was in the habit of offering horses to the sun; one that owned bulls, to unite with one whose river-deity was bull-shaped.

Or neighbourhood and neighbourliness might lead to the use of a common altar and sacred place by two or more clans, each offering a different victim, because having different totems, and each sacrificing at different times; until the fusion became complete, and nothing more would be required but a myth to explain how it was that the one god worshipped 1Mannhardt, W.F.K. 260,

at the altar appeared in different animal shapes, or had different animals offered to him.

Fusion of this kind-syncretism-would be materially facilitated at first by the fact that the gods had originally no proper names. As long as the clan had only one god, no name was required, the gods of hostile clans were sufficiently distinguished by the fact that they were the gods of other clans: "the deity" was the deity of our clan, just as "the river" is that near which I dwell, and whose geography-book name I may not know. But the possession of a proper name gives more individuality to a god; and fusion between two gods, each possessing this higher degree of personality, is more difficult than between two nameless spirits. On the other hand, fusion is not impeded, if of the two gods one possesses a name and the other does not, only the advantage is with the one having a personal name. He readily absorbs Greek god of wine

the nameless one: thus the cult of the was a combination of the worship of a vegetation spirit and of the spirit of the vine, but the former was nameless, and therefore it was the latter, Dionysus, that gave its name to the god. So, too, when we find that in different places half a dozen different animals-wolf, roe, goat, ram, mouse, grasshopper, lizard, swan, hawk, eagle-to say nothing of plants (e.g. the laurel), were associated with or offered to Apollo, we are justified in inferring that as many different nameless totem gods, plant or animal, have been absorbed by the spirit which was fortunate enough to possess the personal name Apollo. Whether that spirit was or was not a sungod is a question to which no decisive answer is forthcoming. But it is clear that fusion between the cult of the sun-god and the worship of other gods would be considerably facilitated by the fact that burnt-offerings played a part in the ritual both of the sun-god and of other gods. The agriculturist, whose crops required sunshine, acknowledged his dependence on the sun and worshipped him. In many cases the sun-god might continue to be consciously distinguished from the plant totem or vegetation spirit, but in many, perhaps most, cases the agriculturist would worship both gods in a common festival, and combine their ritual: he had to make offerings to both, and to both it was possible to convey his offerings