outward deed and act. Myths originate in a totally different psychological quarter: they are the work of the human reason, acting in accordance with the laws of primitive logic; or are the outcome of the imagination, playing with the freedom of the poetic fancy. In neither case are they primarily the product of religious feeling: it is not the function of feeling to draw inferences. It is for moral feeling, or religious, to reject what is alien to it, to penetrate what is compatible with it. Hence the selective function of the religious consciousness depends upon the sensitiveness of that consciousness. Where its sensitiveness was great, only those pieces of primitive science survived which were capable of being informed by the religious spirit. Far different was the case with those nations in whom the religious spirit was late in waking. The explanations which savages invent to account for things that puzzle them are of necessity, like their inventors, savage. If, then, a nation advances from savagery, through barbarism and semi-barbarism, to civilisation, and if the myths which were invented in the savage stage are not rejected by the religious consciousness, but continue to live, in virtue of their connection with the institutions which also are transmitted from the earlier to the later stages of the national life, the result will be that a civilised generation will find itself saddled with myths that attribute to the gods actions of a savage, irrational, and even disgusting description. Philosophers like Plato, then, may argue that tales of this kind, which cannot be true and must be demoralising, ought to be thrown overboard altogether; but the majority of people, to whom these tales have been taught as part of their traditional religion, cannot cast them away in this fashion. At the same time they cannot accept them wholly and literally. A via media, therefore, has to be sought, and this via media has always been found in allegory: the obvious meaning of the myths cannot be the true one, but they must have some meaning, therefore they must contain a hidden meaning, intentionally concealed by the authors of the myth. This was the explanation given of Sanskrit mythology in early times in India, and of Greek mythology by Anaxagoras and Empedocles in Greece and by the Stoics of Rome.

The assumption at the base of all forms of the allegorical theory is that in early times there existed a class of philosophers teaching profound philosophy, and conveying it in the form of fables. Now the existence of this caste of philosophers, if it is a historic fact, ought to be capable of being demonstrated in accordance with the ordinary canons of historical criticism; and it is Lobeck's contribution to the science of mythology that he proved, once and for all, the entire absence of any proof, or even presumption, in favour of the historical existence of these philosophers. Since Lobeck's time-his Aglaophamus was published in 1829the application of the theory of evolution to the science of man has enabled us to trace back civilised peoples through the Iron Age and the Bronze Age to the time when their ancestors had only flint implements, and were unacquainted even with the rudiments of agriculture. At the same time the study of savages still in the Stone Age has revealed the fact that not only are the implements made and used by them the same all over the world, but that the institutions and conceptions by which they govern their lives have an equally strong resemblance to one another. The presumption, therefore, that our Indo-European forefathers of the Stone Age had beliefs and practices similar to those of other peoples in the same stage of development, is very strong; and it is confirmed by the fact that amongst the most backward members of civilised communities, amongst those classes which have made relatively little advance in civilisation, folk-lore discovers abundant traces of superstitions which find exact analogues in savage customs. For the proof,

however, that the irrational elements in the mythology and folk-tales of civilised nations-the taboos and metamorphoses, the incest and bestiality-are survivals from savagery, we must refer the reader to the works of Mr. Andrew Lang, mentioned at the beginning of this chapter.

That the allegory theory of mythology survived to the present century, until it received its deathblow from Lobeck, was due partly to the belief that the inner, esoteric meaning of the myths was taught to the initiated at the Eleusinian and other mysteries by the priests, to whom it was handed down by their predecessors, the inventors of this mode of

teaching. This belief, which we shall have to examine shortly, derived considerable sustenance from two fallacies. One was based on the illicit importation of modern ideas into ancient institutions: it was naturally but erroneously inferred that because in modern religions great stress is laid upon what a man believes, the same importance was ascribed to this side of religion in ancient times, whereas "the antique religions had for the most part no creed; they consisted entirely of institutions and practices." Hence, then, the first fallacy, that of believing that the business of the ancient priest was to teach. There was no authoritative dogma for him to teach, and as a matter of fact he did not teach. The other fallacy consisted in the assumption that mythology was the work of the priests-which is but a form of the wider and coarser fallacy that religion is the invention of priestcraft.

1

It seems, therefore, to be desirable that, before resuming the direct thread of our argument, and showing how the mystic tendency, obscured under polytheism, was revived by the mysteries, we should indicate the place of the priesthood in early religion, and show that it was not the priest that made religion, but religion that made the priest.

1 Robertson Smith, Religion of the Semites, 16.

CHAPTER XX

PRIESTHOOD

IN all early religions, priests are marked off from other worshippers, partly by what they do, and partly by what they may not do; and there is so much agreement between the different religions on both points, that we obviously have to do with the effects of a cause or causes operating uniformly in all parts of the world. At the same time there are certain features of the priesthood which, though they recur in various religions, are not uniformly present in all: they are not essential parts of the antique conception of priesthood. It is clear, therefore, that any general theory on the subject must account for both the uniformity in certain characteristics and the want of uniformity in the other characteristics. The general cause which a theory postulates must be such that its operation would produce the complete uniformity of the one class and the only partial uniformity of the other class of features.

The most important point in which only partial uniformity prevails is tenure of office. Some priesthoods are annual, some tenable for five years, some for twelve, some for life; of some the tenure is terminable on certain contingencies; others are hereditary. Sometimes priests form an order apart, and in that case the order in some places consists of priests appointed for life, sometimes of hereditary priests. In one country there may be only one form of priesthood, e.g. an order of hereditary priests as in Israel, or an order of priests chosen for life, as amongst the negroes of the Gold Coast. In another, life-priests, annual and quinquennial priesthoods, and priesthoods terminable on certain contingencies, may all exist side by side, as, e.g., in ancient Greece. And the tenure of even hereditary priesthood may be made

terminable-as far as the individual is concerned- -on certain contingencies, or on attaining a certain age, e.g. manhood; for, whereas some priesthoods could not commence before manhood, others could only be held before that period.

Having illustrated the want of uniformity in this feature of the priesthood, and having noted that it will require explanation, we may proceed to examine the features in which uniformity prevails. First, we will take the fact that in all religions there are certain things which priests may not do there may be, there is, a want of agreement in details, as to the particular things, but the general principle is universal. When, however, we come to examine the details, we find that, though the particular things which are thus forbidden in antique religions vary, they all agree in certain points: they are prohibitions which have no spiritual value (e.g. the priestess of Athênê at Athens might not eat cheese1), no ethical import (e.g. the prohibition of attendance at funerals 2), and no practical utility (e.g. the prohibition of seeing an army under arms 3); in fine, they constitute the "irrational element" in the conditions of priesthood, and have exactly the same value for the historian as the irrational element of myth has: they indicate that the institution has been transmitted to civilised man from ancestors who were in a less advanced stage of culture than he, and to whom, consequently, these prohibitions appeared, when they made them, perfectly reasonable. It is clear, then, that any general theory of the priesthood must account for these prohibitions; and to be a satisfactory theory must account for them all. The nature of the class of facts requiring explanation may be inferred from the summary Mr. Frazer gives of the prohibitions or rules of life observed by the Flamen Dialis at Rome; "they were such as the following: the Flamen Dialis might not ride or even touch a horse, nor see an army under arms, nor wear a ring which was not broken, nor have a knot on any part of his garments; no fire except a sacred fire might be taken out of his house

1 Strabo, ix. 395.

2 Lev. x. 6, xxi. 1-5; Plato, Laws, 947 C. Festus, 2496, 22 for the Flamen Dialis, and Schömann, Antiquités Grecques, 11. ii. 507 for Greek priests.

4 G.B. i. 117.