office had to be found. One such means was that adopted by the Mikado: it consisted in abdicating on the birth of a son and doing homage to the child, on whom thus fell all the restrictions, while the father, acting in the infant monarch's name, exercised all the power.1 It is in a similar way, we may conjecture, that the priesthoods administered by young men or children were transferred to them by their fathers; for the rules which would hamper the father in his daily life and work could be observed with less practical inconvenience in the case of the young or infant son. For, it need hardly be remarked, the priest, even when temporal power had passed to the kingship, still retained the divine character, and with it the incapacity for mixing in the affairs of daily life, which attached to the priest-king. Thus in Tartary, we find Father Grueber saying," Duo hoc in Regno Reges sunt, quorum prior Regni negotiis recte administrandis incumbit, et Dena dicitur; alter ab omni negotiorum extraneorum mole avulsus, intra secretos palatii sui secessus otio indulgens, Numinis instar adoratur . . hunc veluti Deum verum et vivum, quem et Patrem æternum et cœlestem vocant, . . . adorant." 2 In this connection we may note it as a further indication of the original indivisible unity of the office of priest and king, that even when the two functions have come to be exercised by different persons there is a perpetual tendency to revert to the old organic unity: it is not merely that each of the separate offices retains some part of the divine character that attached to the undivided office, but the functions themselves tend to reunite-reverting in their unity sometimes to the priest and sometimes to the king. If, for instance, the priesthood becomes (or remains) hereditary, and temporal rulers are appointed ad hoc and from time to time, the temporal functions naturally relapse into the priesthood in the intervals (longer or shorter) when no judge or war leader is forthcoming. Indeed, even in the latest times, the consecration of the king by a priest testifies to the original source of the king's office. On the other hand, if the kingship becomes hereditary but the priesthood not, then, in spite of the existence of priests, priestly functions tend to attach themselves to the kingly office; hence it is a 2 Thevenot, Divers Voyages, iv. 22.

1 Loc. cit.

very general feature of the kingship in ancient times that the king can offer sacrifice, like a priest. If this reunion of the two functions becomes so intimate as to amount to a reversion to the ancestral organism, so to speak, then the same process of fission which originally gave birth to the king will be repeated; and the temporal ruler, whose office originated in a delegation of power from the king-priest, will himself have to appoint a delegate to do those warlike duties which the sanctity of his office prevents him from discharging himself-by the side of a Baoiλeus we shall find a Tоλéμaρxos, by the side of the "king" a heretoga. The tendency to reversion, however, which manifests itself particularly when either of the derived offices is hereditary, may be averted without danger to the hereditary principle, if the hereditary priest (or king) delegates his temporal (or priestly) functions to his brother, or other relative and his descendants.

A further and remarkable fact which tends to connect kingship and priesthood together, and to prove their common origin, is the common fate to which divine kings and divine priests alike were liable: at the end of a certain period of time the king had to commit suicide or was put to death. In India, the king of Calicut had to cut his throat in public at the end of a twelve years' reign; so, too, the king of Quilacare in South India.1 The divine kings of Mero in Ethiopia could be ordered to die whenever the priests chose. In various parts of Africa, kings and priests having supernatural powers are put to death, sometimes when old age threatens, sometimes when they have developed the least bodily blemish, such as the loss of a tooth; and the executioner may be the destined successor of the king. Amongst the ancient Prussians, the ruler, whose title was God's Mouth, might commit suicide by burning himself in front of the sacred oak. Amongst other peoples death seems not to have been insisted on at all unless drought or pestilence or other calamities occurred. But even so, a difficulty was found in obtaining persons willing to take office. In Savage Island, " of old they had kings, but as they were the high priests as well, and were supposed to cause the

4

E.g. the Swedes, ibid. 47.

food to grow, the people got angry with them in times of scarcity, and killed them, so the end of it was that no one wished to be king." On the other hand, it is clear that peoples who wished to retain the institution of kingship would have to give up requiring suicide of the king. The requirement, however, would not have been made in the first instance if there had not been a very powerful motivewhatever the motive might have been-for making it; and the motive operated against the abolition of this condition of holding royal office, as also it must have operated in inducing the occupants of the office to comply with it. Eventually the condition was evaded. Amongst the Western Semites, in Babylon, the tenure of office seems to have been annualthe original term, as we shall hereafter argue and at the end of the year the king was put to death. In course of time the community seem to have consented to an evasion: when the time for execution came, the king abdicated, and a criminal was allowed to reign in his stead for five days, at the end of which time the criminal was executed and the king resumed his throne.2 Elsewhere the king abdicates annually, and a temporary king is appointed but is not killed, he is only subject to a mock execution.3 In two places (Cambodia and Jambi) the temporary kings come of a stock believed to be akin to the royal family. Sometimes the mock king is not appointed annually, but once for all for a few days at the beginning of the reign, which seems to indicate that in this case the custom of annually executing a substitute for the king had given way to the practice of executing one, once for all, at the accession of the king. Finally, it is suggested by Mr. Frazer that a criminal would probably not at first have been accepted by the community as an adequate substitute: hence possibly the original substitute was the king's firstborn son.5 The practice of sacrificing the firstborn to the gods is well known.

It seems probable that originally the office of divine priest-king was held for a year, because in that case the difficulty and cruelty of insisting on the fulfilment of the condition of tenure would naturally lead to an extension 2 Frazer, op. cit. 227.

1 Turner, Samoa, 304. 4 Ibid. 234.

5 Ibid.

3 Ibid. 228-31.

first to some definite period, as for instance to twelve years (or, since as some priesthoods were quinquennial, perhaps to five years), then for life, provided that natural death was not allowed to interfere with the suicide or execution which was in the bond. To prevent this last contingency, some peoples made the appearance of the first indication of old age, the first physical blemish, a sign for execution, and to the end a physical blemish in a priest was widely deprecated: "sacerdos non integri corporis quasi mali ominis res vitanda est." 1

It seems, then, that the functions habitually performed by the priest in the civilised states of ancient times, and the powers which he exercised less frequently, and the restrictions which were laid upon him, were all inherited by him from his predecessor the divine priest. It seems also that the similar restrictions and the similar sanctity of the ordinary king of historic times were inherited by him from his predecessor the divine king. And the existence of these divine priests and divine kings-in all quarters of the globe, as the instances accumulated by the learning of Mr. Frazer show-points to the fact that in the early history of the race, in patriarchal times, each wandering community of fellow-tribesmen had over it a person who was in some sense divine, both priest and king, and whose death, voluntary or imposed, at the end of a year, was regarded by the community and accepted by the victim as imperative in the highest interests of the community. We have therefore to inquire why this was believed; and it is only proper that we should begin by stating Mr. Frazer's answer to the question.

Mr. Frazer thinks that men began by believing themselves to be possessed of magical powers, and consequently that the distinction between men and gods was somewhat blurred―apparently that it was difficult or impossible for primitive man to tell whether a certain person, his own ruler in this case, was a very great magician or a god. Further, apparently the primitive community seem to have come to the conclusion that their chief was a god, and that, having got hold of a god, it was desirable to retain him for purposes of their own. But the god might grow old and feeble, which

1 Seneca, Controv. So in Mexico, Sahagun (pp. 62 and 97 of the French trans.).

would be a pity, and he might die and so slip through their hands altogether. Both misfortunes, however, could be averted by inducing his soul to migrate into another healthy young body. This was effected by killing the god: his soul then had perforce to leave its old body, and by some means, not quite clear, it was supposed to enter the body of the murderer, who thus became the new god. Eventually, however, according to Mr. Frazer, men learned to distinguish between magic and religion, and then they placed their faith in the former no longer, but in prayer and sacrifice—not now deeming themselves indistinguishable from gods.

The doctrine that magic is prior to, or even in origin coeval with, religion has already failed to win our assent,1 and we have also argued that the idea of man's coercing the gods for his own ends belongs to a different set of thoughts and feelings from those in which religion originates, and must be later in point of development, because gods must exist first before coercion can be applied to them.2 We do not, therefore, propose to repeat our arguments on the general question of the priority of religion or magic. Nor do we propose to traverse the statement that divine power can be transmitted by the person who possesses it to someone else. What we are here concerned to show is that, apart from these questions, there is evidence to show, first, that these kings and priests were not gods, and, next, that the divine powers they possessed were not native to them and inherent in them, in virtue of their magic, but communicated to them or derived by them from the gods.

This may take us a step further towards the answer to the main question of this chapter, namely, how and why did the community come to regard it as the privilege or duty of some one particular member to exercise the priestly function of dealing the first and fatal blow at the sacrificial victim? To answer that it was because that person was the chief of the tribe, will not advance us much now that we recognise the

1

Supra, p. 177-9.

2 If it be argued that the magical means of coercion may have existed before the gods did, we must refer the reader again to our attempt to show that all such magic is derived from, or rather a distortion or parody of, the worship of the gods.