138 MYTHOLOGY Sanskrit, and Zend philology " (Lect. on Lang., 2d ser., p. 406). A name may be intelligible in Sanskrit which has no sense in Greek. Thus Athene is a divine name with out meaning in Greek, but Mr Miiller advances reasons for supposing that it is identical with ahana, "the dawn," in Sanskrit. It is his opinion, apparently, that whatever story is told of Athene must have originally been told of the dawn, and that we must keep this before us in attempting to understand the legends of Athene. Thus (op. cit., p. 410), he says, "we have a right to explain all that is told of him " (Agni, " fire ") " as originally meant for fire." To take another example, Mr Miiller proves by aid of Sanskrit philology that Zeus originally meant the sky, and, as a result, "there was nothing that could be told of the sky that was not in some form or other ascribed to Zeus." If, then (to take an example of our own), we read that Zeus, to pacify the jealousy of Demeter, muti lated a ram, and pretended to have mutilated himself, are we to suppose that this story had originally a meaning in reference to the sky ? The system, at all events, is simply this : the original meaning of the names of gods must be ascertained by comparative philology. The names, as a rule, will be found to denote elemental phenomena. And the silly, savage, and senseless elements in the legends of the gods will be shown to have a natural significance, as descriptions of sky, storms, sunset, water, fire, dawn, twi light, the life of earth, and other celestial and terrestrial existences. Stated in the barest form, these results do not differ greatly from the conclusions of Theagenes of Rhegium, who held that " Hephaestus was fire, Hera was air, Poseidon was water, Artemis was the moon, KCU TO. AotTro, 6//.OIWS." But Mr Miiller s system is based on scien tific philology, not on conjecture, and is supported by a theory, which we shall try to explain, of the various pro cesses in the evolution of myths out of language. (3.) The following is an abstract of Mr Miiller s theory of one process by which myths were developed out of language. " The keenest eye of the antiquary and the philosopher" cannot see farther back, he says, than the period when expressions were coined for the most neces sary ideas, and when a grammar began which was destitute of national peculiarities, but contained the germ of all the Turanian as well as the Aryan and Semitic forms of speech. This age Mr Miiller calls the Rhematie period (Sel. Ess., i. 306). (As yet there were no myths, or none alluded to by Mr Miiller.) This was followed by an age in which at least two families of language, the Aryan and Semitic, left " the nomadic stage of grammar," and received once for all the peculiar impress of their formative system. There were as yet no such tongues as the Greek, Sanskrit, and Latin we know, only the Aryan speech from which these languages differentiated themselves. This was the Dialectic period. (As yet we understand that there were no myths, or none which Mr Miiller takes into account.) We now come to the Mythopoeic age. It was "half-way between the Dialectical period, presenting the human race gradually diverging into different families and languages, and the National period, exhibiting to us the earliest traces of nationalized language, and a nationalized litera ture in India, Persia, Greece, Italy, and Germany " (op. cit., i. p. 311). The Mythopoeic age, according to Mr Miiller (op. cit., i. 308), came after the "unavoidable divergence of dialects and languages," between that period and the age of "the establishment of laws and customs and the first beginnings of religion and poetry." Mr Miiller s next step is to examine the intellectual and social conditions of man in the Mythopoeic age. This he does by aid of philology. He analyses the words which, being common to Sanskrit, Greek, Latin, and German, must have existed in Aryan speech before the nations that talked these languages separated from the common stock. Although, as we have seen, the Mythopoeic age came before " the earliest concentrations of political societies, the establish ment of laws and customs," yet Mr Miiller demonstrates that man, in the Mythopoeic age, had political societies, and customs if not laws. Man, in the age called Mythopoeic, the age when myths were made, already possessed the modern, or at all events the patriarchal, form of the family. His life was "half nomadic, half pastoral." He had abundance of domesticated animals, he practised agriculture, and had invented the plough. His political institutions included kingship. He was a builder of cities and a constructor of roads. He could weave, and work the metals, includ ing iron. He possessed a system of decimal arithmetic, " which could only have been secured," says Mr Miiller, " by the wear and tear of language in literary and practical use." Thus, as he possessed a literature, his language must have been tolerably definite and, so to say, stereo typed in meaning. "This earliest period, then, previous to any national separation, is what I call the Mythopoeic period, for every one of these common Aryan words is, in a certain sense, a myth " (op. cit., i. p. 355). It will be observed that, in the Mythopoeic age, man was essentially civilized, and that his language had passed through " liter ary wear and tear." (4.) Having thus defined the social, political, and literary status of men in the Mythopceic age, Mr Miiller goes on to describe the style of their conversation, which in the long run was the source of their myths. In the language of that day (as we gather from an examination of Aryan words) a number of terms, which later became Abstract, " expressed something substantial, something open to sensuous perceptions." "In ancient languages every one of these words" (such as "day," "night," "earth," " spring," " dawn ") " had necessarily a termination expres sive of gender, and this naturally produced the corre sponding idea of sex, so that these names received not only an individual but a sexual character. . . . What must have been the result of this ? As long as people thought in language it was simply impossible to speak of morning or evening, of spring and winter, without giving to these conceptions something of an individual, active, sexual, and at last personal character. They were either nothings, as they are nothings to our withered thought, or they were something ; and then they could not be conceived as mere powers, but as beings powerful." Now let us take an example to show how, if the original sense of the names of those " powerful beings " were lost, while the names themselves survived in language as part of a traditional saying, a myth would arise where no myth was intended. Let us suppose that in the Mythopoeic age some one said "the shining one pursues the burning one," meaning the sun follows the dawn. Let it further be supposed that the word for " shining one " was an Aryan prototype of the Greek tjXio?, " the sun," and that the word for the "burning one" was, similarly, an Aryan prototype of the Sanskrit ahana or dahana, "the dawn." Then grant that the term for Helios came to be confused with Apollo (a god who has points in common with the sun), grant that the word for the " burning one " became, from some thing like ahana or dahana, Daphne, and admit that a certain tree was also called daphne, because its wood burned easily. When all these changes had happened and had been forgotten, Greeks would find in their language this expression, "Apollo pursues Daphne." They would see that " Apollo " was a masculine, " Daphne " a feminine, word. And they would thus be led to suppose that Apollo was a young amorous god who chased a fair reluctant nymph Daphne, and that Daphne, to avoid his pursuit, changed herself, or was changed, into a tree bearing the
