In the abridgment I have neither added new matter nor altered the views expressed in the last edition; for the evidence which has come to my knowledge in the meantime has on the whole served either to confirm my former conclusions or to furnish fresh illustrations of old principles. Thus, for example, on the crucial question of the practice of putting kings to death either at the end of a fixed period or whenever their health and strength began to fail, the body of evidence which points to the wide prevalence of such a custom has been considerably augmented in the interval. A striking instance of a limited monarchy of this sort is furnished by the powerful mediaeval kingdom of the Khazars in Southern Russia, where the kings were liable to be put to death either on the expiry of a set term or whenever some public calamity, such as drought, dearth, or defeat in war, seemed to indicate a failure of their natural powers. The evidence for the systematic killing of the Khazar kings, drawn from the accounts of old Arab travellers, has been collected by me elsewhere. 1 Africa, again, has supplied several fresh examples of a similar practice of regicide. Among them the most notable perhaps is the custom formerly observed in Bunyoro of choosing every year from a particular clan a mock king, who was supposed to incarnate the late king, cohabited with his widows at his temple-tomb, and after reigning for a week was strangled. 2 The custom presents a close parallel to the ancient Babylonian festival of the Sacaea, at which a mock king was dressed in the royal robes, allowed to enjoy the real king's concubines, and after reigning for five days was stripped, scourged, and put to death. That festival in its turn has lately received fresh light from certain Assyrian inscriptions, 3 which seem to confirm the interpretation which I formerly gave of the festival as a New Year celebration and the parent of the Jewish festival of Purim. 4 Other recently discovered parallels to the priestly kings of Aricia are African priests and kings who used to be put to death at the end of seven or of two years, after being liable in the interval to be attacked and killed by a strong man, who thereupon succeeded to the priesthood or the kingdom. 5 With these and other instances of like customs before us it is no longer possible to regard the rule of succession to the priesthood of Diana at Aricia as exceptional; it clearly exemplifies a widespread institution, of which the most numerous and the most similar cases have thus far been found in Africa.
Note 1. J. G. Frazer, The Killing of the Khazar Kings, Folk-lore, xxviii. (1917), pp. 382-407.
Note 2. Rev. J. Roscoe, The Soul of Central Africa (London, 1922), p. 200. Compare J. G. Frazer, The Mackie Ethnological Expedition to Central Africa, Man, xx. (1920), p. 181.
Note 3. H. Zimmern, Zum babylonischen Neujahrsfest (Leipzig, 1918). Compare A. H. Sayce, in Journal of the Royal Asiatic Society, July 1921, pp.
440-442.
Note 4. The Golden Bough, Part VI. The Scapegoat, pp. 354 sqq., 412 sqq.
Note 5. P. Amaury Talbot in Journal of the African Society, July 1916, pp. 309 sq.; id., in Folk-lore, xxvi. (1916), pp. 79 sq.; H. R. Palmer, in Journal of the African Society, July 1912, pp. 403, 407 sq.
How far the facts point to an early influence of Africa on Italy, or even to the existence of an African population in Southern Europe, I do not presume to say. The pre-historic historic relations between the two continents are still obscure and still under investigation.
Whether the explanation which I have offered of the institution is correct or not must be left to the future to determine. I shall always be ready to abandon it if a better can be suggested. Meantime in committing the book in its new form to the judgment of the public I desire to guard against a misapprehension of its scope which appears to be still rife, though I have sought to correct it before now. If in the present work I have dwelt at some length on the worship of trees, it is not, I trust, because I exaggerate its importance in the history of religion, still less because I would deduce from it a whole system of mythology; it is simply because I could not ignore the subject in attempting to explain the significance of a priest who bore the title of King of the Wood, and one of whose titles to office was the plucking of a boughthe Golden Boughfrom a tree in the sacred grove. But I am so far from regarding the reverence for trees as of supreme importance for the evolution of religion that I consider it to have been altogether subordinate to other factors, and in particular to the fear of the human dead, which, on the whole, I believe to have been probably the most powerful force in the ****** of primitive religion. I hope that after this explicit disclaimer I shall no longer be taxed with embracing a system of mythology which I look upon not merely as false but as preposterous and absurd. But I am too familiar with the hydra of error to expect that by lopping off one of the monster's heads I can prevent another, or even the same, from sprouting again. I can only trust to the candour and intelligence of my readers to rectify this serious misconception of my views by a comparison with my own express declaration.
J. G. FRAZER.
1 BRICK COURT, TEMPLE, LONDON, June 1922.