The outline of the genealogy of languages which I gave you in some of my former lectures will be equally useful for the genealogy of mythology. It will in fact be the chief object of this and the next following lectures to show that what we call myth is a natural and inevitable phase in the development of language; that in its initial stages that phase showed itself before the different languages belonging to the same family had become finally separated, and that therefore, besides much that is peculiar to each, we find in all a common fund of mythology which we may look upon as the earliest stratum likely to contain the germs of religious thoughts.

If we use myth and mythology synonymously, we have the authority of Greek writers for doing so, for mythology (μυθολογία) with them does not mean, as it often does with us, a study of myths, but it is used in the sense of a telling of mythic legends, and afterwards of these legends and tales themselves.

Few words, however, have of late changed their meaning so completely as myth and mythology. Not very long ago Greek mythology meant Greek religion, Roman mythology meant Roman religion, and each was supposed to consist of a body of traditions and doctrines which a Greek or Roman had to believe, just as Christians believe in the New, or the Jews in the Old Testament. As mythology was taught at school chiefly from manuals, a very general impression prevailed that the legends collected in them existed in this collective form in Greece and Italy, that they formed in fact a complete system, and were known as such by every Greek and Roman, man, woman, and child; the fact being that hardly a single Greek or Roman could have passed an examination in our manuals of mythology, nay that the very names of many of the gods and heroes therein mentioned would have been utterly unknown to the majority of the inhabitants of Greece and Italy.

Before we discuss the meaning which mythology has assumed, chiefly owing to the discovery that myth is a phase of language, inevitable in the early development of speech and thought, it may be well to ask in what sense μυ̑θος was used by the Greeks themselves.

The etymology of μυ̑θος is unknown, or at all events doubtful. It is well to be reminded from time to time how many words there are still in Greek and Latin, to say nothing of Sanskrit, of which we cannot render any etymological account. Of course, we can guess that μυ̑θος is derived from μύω, to shut, to close. This is used of shutting the eyes, as in μύωψ, μύωωος, literally closing the eyes, then shortsighted; and it is likewise used of shutting the lips. From this a secondary base might be derived, μυάω, which means to compress the lips, to express contempt. In Sanskrit we have a root mû, to bind, from which mû-ka, dumb, lit. tongue-bound, and likewise Latin mû-tus, dumb, and Greek μύ-τις, which Hesychius mentions in the sense of ἄϕωνος, as well as μύτης and μύττός. Possibly μυέω, to initiate, to teach secrets, may like-wise come from that root, while μύστης and μυστήριον might owe their s to analogy. Still it would be strange if μυ̑θος, word, had meant originally a muttering with closed lips, even though we can appeal to Latin muttum, a muttering, muttire, or mutîre, to mumble. The Gothic rûna, secret counsel, has like-wise been mentioned as a parallel case, because it is derived from a root RU, to whisper¹.

All we can say is that a derivation of μυ̑θος from the root mû, to bind, to close, is phonetically possible, and this is more than can be said for another etymology which connects μυ̑θος with μύζω, to murmur, for in μύζω the final of the root is guttural, not dental, as is shown by μυγμός, muttering.

Though the etymology of μυ̑θος is somewhat doubtful, its meaning in Greek is clear enough. It means word as opposed to deeds, and hardly differs originally from ἔπος and λόγος. Afterwards², however, a distinction is made between μυ̑θος in the sense of a story, a fable, and λόγος, an historical account, and this distinction has been preserved in modern times.

If the original meaning of the Greek λόγος, as both word and thought, has revealed to us a forgotten truth which must become the foundation of all true philosophy, namely the identity of thought and language, the original meaning of μυ̑θος, word, will teach us an equally useful lesson for the study of mythology, and indirectly, of religion.

Let us take myth in its original sense, and we shall see that here too the Greeks saw rightly. A myth was at first a word. The formation of such a word as Eos, dawn, seems at first sight not very different from the formation of any other word. But if you remember that all roots expressed originally an action, you will see that we require for every word an agent. Now so long as we deal with verbs, we always have our agent; namely, I, thou, or he—I strike, thou strikest, he strikes. But when we have to deal with a word like Eos—who is the agent there?

We know that Eos is the Sanskrit Ushas, and we know that ushas is derived from a root VAS, which means to shine. So Eos meant originally ‘shining-it,’ or ‘shining-he,’ or ‘shining-she.’ But who was it, or he, or she? Here you have at once the inevitable birth of what we call a myth. What our senses perceive and what we are able to name is only an effect, it is the illumination of the sky, the brightness of the morning or, as we now should say, the reflection of the rays of the sun on the clouds of the sky. But such were not the thoughts of the early framers of language. After they bad framed a word which meant shining there, or light, namely Eos, they would go on to say, that Eos has returned, Eos has fled, Eos will return, Eos wakens the sleepers, Eos lengthens our life, Eos makes us grow old, Eos rises from the sea, Eos is the daughter of the sky, Eos is followed by the sun, Eos is loved by the sun, Eos is killed by the sun, and so on ad infinitum.

Now what is all this? You may say, it is language, it is mythos—yes, and it is what I called the inevitable myth, and a myth that will grow on for ever. For, if Eos is followed by the sun, or, as we should say, if she has the sun for her follower or lover, she would naturally be conceived as a woman, and as a bright and beautiful woman. If she appeared veiled in clouds, she would be conceived as a veiled bride; if she was seen in her naked beauty, she would be celebrated for her brilliant charms. Now let us look at all the epithets which Greek poets have bestowed on Eos, and every one of them will become intelligible. If she is called the daughter of Hyperion, who can doubt that Hyperion, like Summanus, was the high heaven? If her mother is called Euryphaessa, the wide-shining, do we want an explanation to tell us that that is only another name for the dawn or for the East or for the morning? If she is called the sister of Helios and Selene, is that mythology, or is it plain truth? As the gloaming seemed but a repetition of the dawn, nothing was more natural than to suppose, as the Greeks did, that Eos had followed Helios through the whole of his course, and that she followed him at last to his watery grave. If Helios or the sun was conceived as driving from East to West, nothing seemed more natural than to assign to Eos also two horses, and to call them Lampros and Phaeton. When chariots were drawn by four horses in Greece, Eos also received four instead of two steeds³

Her epithets require hardly any commentary. Αἰγλήήϵσσα is the brilliant; χαροπή is the joyful-eyed, the Sanskrit haryaksha; χρυσόθρονος is the dawn sitting on a golden throne; ἠριγγένϵια is the early-born; γϵυκόπωλος is Eos drawn by white horses; λϵυκόπτϵρος, Eos with white wings; ϕαϵσίμβροτος, Eos who brings light to mortals. The rest is added by poets who speak of her as ῥοδοδɕκτυλος, rosy-fingered; κροκόπϵπλος, clothed in yellow garments; ϵὐπλόκαμος, with beautiful ringlets; χιονοβλέϕαρος, with eyelids white as snow. Latin poets add new epithets, such as lutea, rosy; pallida, pale; purpurea, purple-coloured; roscida, dewy; vigil, wakeful.

You can see from these epithets, which gathered round the name of Eos in Greek, and Aurora in Latin, how inevitably what we call mythology springs up from the soil of language. As soon as a name, such as Eos, was thrown out, it grew and gathered new materials round itself, and without any special intention or effort became what we call a myth. Even such simple sentences as ‘Eos is born,’ ‘Eos brings light,’ ‘Eos dies or disappears,’ are changed at once into myth, fable, and legend, and it seems impossible to draw a line between what is simple language and what is myth.

It was long supposed that much of what we call mythology was due to the peculiar poetical genius of the Greeks. Our first acquaintance with mythology came from Greece, and we were accustomed from our school-days to look upon the Greeks as a nation endowed with such wonderful gifts that we thought we might safely credit them with the invention of all the beauty and wisdom of their mythology. That there were dark sides to that bright picture also, could not be denied; but it was thought possible by classical scholars, unacquainted with the mythology of other nations, that all that was hideous and foolish in classical mythology might be explained as a survival of barbarous ages, when the barbarous ancestors of Greeks and Romans were not above committing themselves those crimes and follies which they fondly ascribed to their gods.

It is here that Comparative Mythology has stepped in, and helped us to solve many difficulties which could not be removed by any other theory.

What is Comparative Mythology?

Comparative is a name which has been assumed of late by nearly all historical and natural sciences, though, if we once understand the true method and purpose of any single science, it would seem to be almost superfluous to qualify it by that predicate. There is no science of single things, and all progress in human knowledge is achieved through comparison, leading on to the discovery of what different objects share in common, till we reach the widest generalisations and the highest ideas that are within the ken of human knowledge.

Thus with regard to languages, the very first steps in our knowledge of words are made by comparison. What does grammar consist in but a collection of words which, though they differ from each other, share certain formal elements in common? These formal elements are called grammatical elements, or suffixes, affixes, prefixes. etc., and we are said to know the grammar of a language when we have learnt under what conditions different words undergo the same formal modifications.

Thus comparison leads in the first instance to a grammatical knowledge of a single language.

When, however, we proceed from a study of one to a study of many languages, a new process of comparison begins. We observe that words in different languages undergo the same or nearly the same modifications, and by placing the paradigms of their declension and conjugation side by side, we try to find out on what points they agree and on what points they differ, and we hope thus to discover in the end the reasons why they should agree on certain points, and why they should differ on others.

Comparative Philology deals partly with facts, that is, the differences and coincidences that can be observed in the material and formal elements of language; partly with laws,—using that word in the humble sense of ‘something which is true of many objects,’ not, as some scholars imagine, in that of νόμοι ὑψίποδϵς οὐρανίαν δι᾽ αἰθέρα τϵκνωθέντϵς, ώ̑ν ῎Ολυμποςπατὴρ μόνος, οὐδέ νιν θνατȤ ϕύσις ἀνέρρων ἔτικτϵν. These laws are to account for such peculiarities as give to each language its own distinctive character.

This science of Comparative Philology, however, very soon assumed three different aspects, and was cultivated in three schools, which may be called (1) the Etymological, or genealogical, (2) the Analogical, and (3) the Psychological.

In comparing such languages as Latin, Greek, and Sanskrit, it was soon found that they were really varieties only of one and the same historical prototype, that they pointed to a common origin, and that all their differences must be accounted for either by Phonetic Corruption, or by Dialectic Growth. The comparative study of these languages became therefore genealogical, or, in grammatical phraseology, etymological.

Starting from a certain number of radical and formal elements (the latter being frequently radical elements of an earlier period), the principal object of the genealogical or etymological school has always been to discover the system according to which these elements were combined into words, and to determine the laws which regulate the phonetic changes of words, either in the same or in different languages. When these laws are sometimes treated as natural laws, this means in reality no more than that they admit of no exception, except such as can again be accounted for by new laws.

The next school, the analogical, or as it might also be called, the dialectic, tries to discover what in the same or in different languages is not identical, but yet analogous. While the genealogical school looks upon all cognate languages as dialects developed from one ideal κοινή, the dialectic school looks upon each language as the result of a previous independent growth, and is thus able to account for freedom and variety in single languages as well as in whole families of speech, as against the iron laws of phonetic change, established by the etymological school.

It would be impossible, for instance, or at allevents undesirable⁴, to treat say the Ionic dialect as a corruption of the Aeolic, or the Aeolic as a corruption of the Ionic. The same applies to High German and Low German, to Sanskrit and Prâkrit, to Cymric and Goidhelic. These are all independent streams of language, which it is as hopeless to trace back to one common source, as it is to discover the one small source of the Nile, or even of the Thames. They spring indeed from the same geological stratum, and they follow parallel courses under similar conditions, but they are not yet one stream of water or of speech, kept in by the same shores and moving on in the same bed. Even after their confluence, the peculiar colours of what I call Dialectic Growth remain, and help us to account, whether by true or by false analogy, for that want of uniformity or regularity which the etymological school postulates with unyielding severity.

Thus dvau in Sanskrit, δύω in Greek, duo in Latin are phonetic varieties of one and the same type. They are identical in origin, and their differences can be accounted for by phonetic laws. But Sk. dvitîya, Polydeukes, Apollon and Greek δϵύτϵρος are not identical in origin. They are dialectic forms, sprung from the same etymological stratum, not the products of one and the same creative act.

Nevertheless, it is in cognate languages only that we could account for such words as Sk. prathama, the first, Greek πρω̑τος, Latin primus, and Gothic fruma. These are all analogous formations, only they must not be treated as varieties of one common prototype. Their differences are not due to the influence of phonetic modification, which can be reduced to a law, but to the freedom of dialectic growth, which must be accepted as a fact.

I go even further. We can hardly doubt, for instance, that the words for twenty were formed by a composition of words meaning two and ten. In Chinese shῐ is ten, eúl is two, therefore eúl-shῐ, twenty. Our own twenty comes from Anglo-Saxon tuên-tig, which corresponds to the Gothic tvai tigjus, and to the modern German zwanzig.

In Sanskrit we should expect a form like dvi-dasa, and in Latin duo-decem. But instead of this we find in Sanskrit vimsati, in Latin vîgintî, in Greek ϵἴκοσι, and the older form Ϝϵικατι. According to strict phonetic laws, these forms are all irregular. Dvi does never lose the initial d in Sanskrit, nor does dvi in Latin become vi, or in Greek ϵἰ. In Sanskrit dvi ought to have remained; in Greek dvi ought to have become δι, in Latin di or bi. Yet the fact remains that in one of the ancient Aryan dialects dvi was replaced by vi, for thus and thus only can we explain vi in vimsati, ϵἰ in Greek in ϵἴκοσι, vî in Latin in vîgintî. The stem for ten or decad, namely dasa or dasat, was shortened to sat⁵, which is likewise without any phonetic excuse or analogy.

Here then we see what I call dialectic influence, as different from the independent working of phonetic laws. Vimsati is not a phonetic corruption of *dvi+dasati, nor vîgintî of *dvî+decinti, nor Ϝϵικατι of *δϜι+ δϵκατι, but they are dialectic forms in which some old compound of twice-ten was fixed and retained, and was afterwards modified according to the peculiar phonetic instincts of Sanskrit, Greek, and Latin.

I cannot enter more fully into this subject at present, but I may remark that it is the disregard of this distinction between phonetic modification and dialectic growth which at the present moment seems to me to have led to a series of misunderstandings between the most prominent representatives of Comparative Philology⁶.

The comparison of various languages, after it had led to the discovery of the great families of human speech and settled the principles according to which cognate languages should be analysed and explained, opened in the end a still wider prospect and disclosed before our eyes, not only what was common to Greek and Latin, to Hebrew and Arabic, to Finnish and Hungarian, but what was common and essential to all languages, what constitutes in fact the nature of language in general, and indirectly the nature of thought.

This kind of study, comparative in the widest sense, though it aimed at the discovery of the highest philosophical truth, does not depend for that discovery on abstract reasoning, but, differing thereby from all former attempts to construct a science of general grammar and of logic, it takes its materials entirely from the facts supplied by that infinite number of languages in which the power of language and thought has become realised. It matters little whether we call this branch of Comparative Philology psychological or ethno-psychological, as long as we see clearly that it aims at explaining that intellectual development which has its outward form in language, and that it derives its materials entirely from a careful study of the different types of human speech, so far as they are still accessible to the student of the present day. To me, that branch of the Science of Language seems almost to transcend the powers of the present generation, and to belong to the future of our race. But I look to it as the final consummation of all that has ever claimed the name of philosophy, as the solution of all psychological, logical, and metaphysical problems, and in the end as the only true key to our knowledge of the Self.

What applies to Comparative Philology, applies mutatis mutandis to Comparative Mythology. I do not mean to say that the science of Comparative Mythology is as yet so firmly established as the Science of Comparative Philology. There is an honest difference of opinion with regard to many minor points, but the fundamental principles of Compartitive Mythology, such as I tried to lay them down in my first essay on Comparative Mythology in the year 1856, are now generally admitted. I say generally, I do not say universally. There are still some philosophers who deny that the languages of the Greeks and Hindus, and the mythology of the Greeks and Hindus have anything in common; but I do not know of any scholar of any authority who denies that the Greek Zeus finds its true explanation in the Vedic Dyaus, and that our first duty as students of Comparative Mythology must be to discover the etymology of as many mythological proper names as possible. To say that critics disagree among themselves, and that they need not be listened to till they agree, is one of those lazy commonplaces which no true scholar would dare to employ. I know full well that several mythological etymologies have been contested, and I have always been most grateful for any criticisms proceeding from scholars who really care for the progress of our science.

Among them few have a better right to be listened to than M. A. Barth. He has often criticised what Kuhn and others have written on the origin of mythology, but after making all necessary reservation, he sums up as follows:—

‘No one contests any longer that myths are from the first the natural and popular expression of very simple facts; that particularly the most ancient have reference to the most common phenomena of nature; that they depend very closely on language, being often no more than an antiquated form of it; and that what applies to words in general applies likewise to the immense variety of myths, namely that they can be reduced to a small number of elements, as words are to a small number of roots. In spite of the state of flux they are in, and their apparent confusion, they possess a certain cohesion and are held together by a kind of hidden logic. They do not migrate so easily nor so wildly as had been supposed, from one nation to another, from one race to another, but, like language, they are transmitted by inheritance only, and there are characteristic signs by which borrowed foreign myths can be discovered quite as well as borrowed foreign words…By applying these principles students of comparative mythology have established the fact that the common ancestors of the Celts, the Italians, the Hellenes, the Germans, the Slaves, the Iranians and the Hindus, at a time when they were still settled side by side in some for ever forgotten region of the old continent, adored the same deities; and they have succeeded in restoring at least some persons of that prehistoric pantheon. Of these two series of results, which together constitute Aryan Mythology, the one, that which establishes the unity of beliefs, is certain, quite as certain as the corresponding result established by the Science of Language, namely the unity of an Indo-European mother tongue. The other series, however, the partial restoration of those beliefs, is far less certain.’

If it is considered that this judgment was delivered by a very independent judge so far back as 1880 in the Bulletin Critique de la Mythologie Aryenne in the Revue de l'Histoire des Religions, it can hardly be called exaggeration when in 1886 I expressed my conviction that the fundamental principles of Comparative Mythology were now beyond the reach of cavil or criticism.

But in order to avoid misunderstandings and barren discussions, we ought to divide Comparative Mythology also into three branches, which may be defined as, (1) the Etymological or genealogical, (2) the Analogical, (3) the Psychological or ethno psychological.

The Etymological branch of Comparative Mythology places the names and stories of certain gods and heroes side by side, and tries to prove that these names were derived from prototypes common to certain families of speech. As its object is not only to compare, but to identify these names, and the persons to whom they belong, it is clear that this branch of Comparative Mythology can deal with the traditions of such languages only as have been proved to be connected genealogically. It is natural, therefore, that this special domain of research should have been almost exclusively cultivated by critical scholars, and that the evidence to which they appeal should be entirely etymological, and under the sway of the strictest phonetic laws.

The second branch, the Analogical, might claim for itself the principal right to the name of Comparative Mythology, for it is chiefly occupied with comparing myths and legends, without attempting to prove that they are actually identical in origin. Like the etymological school, it confines itself to the myths of cognate languages, but after having shown how many different names and personifications may attach themselves to the principal objects of mythological thought, such as the sun, the moon, the sky, the earth, fire, and water, storms and lightning, and in bow many different ways the same story may be told of these polynomous objects, it proceeds to a comparison of myths which, though not identical, must have sprung from the same common stratum, and thus takes possession of a far larger area of mythological thought as the common property of a race than could be claimed by purely etymological tests. This analogical process has its dangers, like all purely morphological comparisons, but it forms nevertheless an almost indispensable supplement in the genealogical treatment of mythology.

While both the Genealogical and the Analogical schools confine themselves to a comparison of mythologies which are handed down to us in languages held together by the ties of a common origin, the Psychological or Ethno-psychological school soars higher, and comprehends the mythologies of all mankind. There is nothing in all the mythologies of the world that cannot be compared. What Heine said to an Ethno-psychological lover,

may be said to an Ethno-psychological Mythologist:

It is a most fascinating, though, no doubt, at the same time, a somewhat dangerous study, unless it is carried on by men of scholarlike instinct and historical tact. Its charm consists in the discovery of the most surprising coincidences in the mythologies, the customs, and traditions of distant races, distant in space as well as in time, unconnected by any relationship, whether genealogical or linguistic, civilised and uncivilised, ancient and modern. And it becomes still more attractive when it leads us on to the discovery of general motives which alone can account for such similarities. It becomes, in fact, an historical psychology of the human race (Völkerpsychologie), and promises in time results of the highest value, not only to the historian, but to the philosopher also.

Comparative Mythology rests, as we saw, and can only rest, on Comparative Philology, and such has been the constant advance of that science, particularly with regard to the laws which regulate the interchange of consonants and vowels, that many etymological identifications which seemed quite legitimate fifty years ago, cannot be considered so any longer. My own conviction has always been that phonetic laws cannot be administered in too Draconian a spirit, and that there ought to be no difference made in applying them either to vowels or to consonants. It is far better to leave an etymology, however tempting, as not proven for a time, than to tamper with a single phonetic law.

But, with regard to mythological names, I confess that I myself have been guilty sometimes of pleading for circonstances atténuantes, and I must do so once more. I pointed out many years ago, first, that all mythology was in its origin local or dialectic, and that therefore we must be prepared in mythological names for dialectic variations, which we should not tolerate in other nouns and verbs. In one of my latest papers (Internationale Zeitschrift für allgemeine Sprachwissenschaft, vol. i. p. 214), where I compare Zephyros with the Vedic Gâhusha, I had to remark: ‘Scholars might differ as to Sanskrit g being represented by Greek ζ; but that on Greek soil γ and ζ vary dialectically, can be seen from γϵύσασθαι and ζϵύσασθαι, ἐπιζαρέω by the side of βαρύς, Sanskrit guru; πϵϕυζότϵς and πϵϕυγότϵς,’ etc.

I quite appreciate the motives which have led some scholars to dispute this principle and to object to the granting of any greater latitude in the comparison of mythological names than of ordinary words. But we cannot shut our eyes to facts. Now it is a fact that many of the Greek mythological names appear under dialectic forms which it would be difficult to reduce to general phonetic laws.

By the side of ᾽Απόλλων we find ΅Απλου̑ν (Thessalian), and in Etruscan Aplu; by the side of Ποσϵιδω̑ν, Ποσϵίδης, Ποτίδας, Ποτϵιδÿων; by the side of ῞Αιδης, gen. ῞Αιδου, ᾽Αίδης, gen. ᾽Αίδαο (also an old gen. ῎Αȹδος); by the side of Δημήτηρ, Δα̑, Δηώ. These are parallel or dialectic forms which cannot be derived one from the other. We cannot derive, for instance, Poseidon from Potidas, nor Potidas from Poseidon. Supposing then that the form Potidas bad not been preserved to us, but that Poseidon would yield an intelligible etymology if only we could trace it back to a form like Poteidon, we should have taken a phonetic liberty which might be without any excuse in Greek, and which nevertheless might have yielded an explanation of the myth of Poseidon in accordaace with facts. If we take even so well-known a name as Ζϵύς, I doubt whether its various forms, such as Ζϵύς, gen. Διός, Δɕν, Δηνός, Δίς and Δϵύς, could have been traced satisfactorily to a common base without the help of Sanskrit.

Secondly, I pointed out, likewise many years ago, that it was almost an essential condition, before a name could assume a truly mythological character, that, by some accident or other, its etymological meaning should have been somewhat obscured. Words like Hemera, day, Nyx, night, Helios, sun, Selene, moon, may send out a few mythological off-shoots, but it is chiefly round dark and decaying names such as Kastor and Polydeukes, Apollon and Athene that the mythological ivy grows most luxuriantly⁷.

Let us now return for a moment to Eos. Her name in Sanskrit is Ushas, and we saw that it means the bright one, from a root VAS, to shine. The Dawn, of course, might have been called by many names, and we shall soon see that she was called by many names and that some of them have survived though under strange disguises. But that Ushas should have remained her principal name in Sanskrit, and that the identical name, Eos, should appear in Greek, shows that she was known as a definite personality before the Aryan Separation.

A comparison of Eos, as we saw her in Greece, with what the Veda tells us of Ushas, will show us that much in her character, which we are inclined to look upon as simply Greek poetry, is far more than that—is in fact the outcome of Aryan thought before it was divided into various national streams. Though I am rather anticipating, yet I think that a study of Ushas, as she actually appears in the Veda, may prove useful as giving you a clear idea of what Comparative Mythology can really do for us.

We read, Rv. I. 48, 7: ‘She has harnessed (her the steeds) from a far distance, from the rising of the sun; on a hundred chariots Ushas spreads herself out, going towards men.’

Rv. I. 124, 5: ‘In the eastern half of the watery sky she has put forth her banner; she spreads far and wide, and fills the two laps of her parents.’

You see how mythology is beginning. She has already chariot and horses, she carries a banner of light, and she has two parents, heaven and earth, whose lap she fills with her light.

In the same verse she is also called gávâm gánitrî, ‘the mother of the cows,’ these cows being either the clouds which pour water on the fields, or the bright mornings which, like cows, are supposed to step out one by one from the stable of the night.

Thus we read again, Rv. III. 61, 4: ‘Ushas, the mistress of the stable, comes, the wealthy, as if loosening her girdle.’ Here she has clearly assumed the character of a woman, she has a girdle which she can put off like the clouds that surround her, and she is a rich woman, the mistress of her own stable, the stable constituting the chief wealth of the Vedic Aryas.

Thus she goes on growing before our very eyes. Sometimes she is still simply light, or like streams of light, or like rivers of clouds. Then it is said that she has no feet, and that she became the first of those who have feet (Rv. I. 152, 3), nay, that she is like a beautiful woman (Rv. I. 48, 5, yόshâ-iva sûnárî). Soon she is compared to a bride, adorned by her mother (I. 123, 10. 11); as smiling on approaching her beloved (I. 123, 10; VII. 76, 3). Sometimes she is clothed in bright garments (I. 192, 1–4); at other times, proud of her faultless limbs, she reveals her beautiful body (I. 124, 6; 1. 123, 10; V. 80, 5).

One feature peculiar to the Veda is the plurality of the dawn. Whether the many dawns were meant for the manifold rays of light playing across the sky in the morning, or for the dawns that come and go every day, certain it is that by the side of the one Dawn, poets constantly speak of the many Dawns also.

So far, however, all that is said of Ushas in the Veda might be called mere mythological poetry. But there are some expressions which contain religious germs. It is often said that the dawn is the sister of the dark night, that the two, day and night, succeed each other regularly, that the dawn is in fact always the same, always returning, the old, and yet the ever-young. Then follow such sayings as that she who makes every one else to grow old, remains young her-self. Soon she is called the young, the never-aging, the never-dying—and at last, the immortal.

This will show you how language, by the mere formation of a certain class of words, leads us on to myth, and from myth to religion. Ushas, the bright, devî, has now become Ushas, the immortal, and after that stop has been taken, what is more natural than that she should become an attractive centre for other religious sentiments and thoughts? Even with us a bright morning raises our spirits, and rouses a sense of happiness and gratitude in our heart, though the object of our gratitude may remain nameless. Think what it must have been in early times, when life and everything was felt to depend on the kindly light of the morning! A bright sunrise was a new life, a sunless, cold, stormy morning meant suffering, often starvation or even death. Need we wonder then that some words should have been stammered forth at the rise of a bright dawn, words of joy and gratitude, addressed not to a nameless being, but to the kind and brilliant Ushas, or Eos, or Aurora?

Moral ideas also would soon cluster round such names as the Dawn. If a crime, a dark crime, as we call it, had been committed during the night, who was to discover or to revenge it? Again the Dawn, under one of her many names, the Sanskrit Saranyu, the Greek Erinys.

You may say, But why did not people ask who that Eos really was, before they praised her and thanked her? This is a difficult question to answer. But do we always ask what a name really means, what is behind a name, what is the true substance of a name? We speak of Angels, without asking what they are made of. The ancients did the same, and when they had called Ushas, the daughter of Dyaus or Zeus, their mind seemed satisfied, at least for a time. Names have a wonderfully satisfying power, and few only venture to lift the veil which language has thrown over nature. And when they do, what do they find? They find the infinite hidden under a name, and they find that all they can know of the infinite is what is signified by these names. Ushas, the morning light, is as good a name for the Infinite as Dyaus, the sky, who became Zeus and Jupiter, only that its history took a different direction. And remember that we ourselves also, though we may no longer use the name of Morning-light for the Infinite, the Beyond, the Divine, still find no better expression than Light, when we speak of the manifestations of God whether in nature or in our mind.

So far the way of Comparative Mythology is smooth and easy. But etymology, if only kept under proper control, can lead us over more rugged roads, and give us light in darker passages.

I said before that though Ushas was the oldest name of the Dawn, having been fixed before the Aryan Separation, there were many other names given to the same phenomenon, as looked upon from different points of view. Some of these names might be used by one poet only, others might become traditional in one family or clan, and those dialectic names would lend themselves most easily to mythological phraseology, on account of the very uncertainty of their original meaning. The dawn as Ushas has become mythological, but, as we saw in the Veda as well as in Homer, its natural character was never quite forgotten.

Now there is in the Veda another name for the Dawn, which is Ahanâ. It occurs but once, in a hymn addressed to Ushas, and there can be no doubt that it is one of the many epithets of the dawn.

When we ask why Ahanâ should mean the dawn, the answer is easy. Ahan and ahar mean the day, and ahanî in the dual means day and night. In Sanskrit mythology this name of Ahanâ has remained sterile, but in Greek, as we shall see, it has become the germ of a magnificent growth. When we ask under what form Ahanâ, could appear in Greek we should say at first ᾽Αχανα or ᾽Αχνα. Neither of these forms exists. But we must remember that Sanskrit h represents three original sounds, namely gh, dh, and bh. It represents gh, for instance, in dah, to burn, by the side of which we have Sanskrit ni-dâgha, heat. It represents dh, as, for instance, in NAH, to bind together, nectere, by the side of which we have *NADH, in the present naddhá It represents bh, as in the same root NAH, by the side of which we have NABH in nâbhi, in GRAH and GRABH, both meaning to take, to grab⁸.

In Greek itself we find the aspirates changing dialectically. We have not only ὄρνις, ὄρνιθος, but also ὄρνις, ὄρνιχος We have ἴθμα and ἵχμα and similar forms.

We have therefore a perfect right to expect Athanâ or Aphanâ instead of Achanâ⁹. Now ᾽Αθᾰ̑να exists in Greek as an old name of Athene. We have also ᾽Αθα̑ναι, and ᾽Αθᾱναία. In Athene we have the same suffix as in selene, and the change between the two suffixes ᾰna and âna has been shown to be very common¹⁰. Phonetically therefore the identification of Ahanâ in the Veda and Athene in Greek is beyond the reach of criticism and cavil. If after that we identify Ahanâ with Athene mythologically also, we must see clearly what we mean. First of all, we cannot mean that there ever was a real being, a woman or a goddess, who was known in India and in Greece and hail received there the same name, Ahanâ and Athanâ.

Secondly, we cannot mean that whatever was told of Athene in Greek was told of Ahanâ also in Sanskrit.

Thirdly, and least of all, can we mean that the worship of Ahanâ was carried from India to Greece, or the worship of Athene from Greece to India.

All we can mean is that Ahanâ, as a name of the dawn, was known before Greek and Sanskrit separated, and that while in India this mythological germ withered away, it developed into a splendid growth in Greece.

We see the same with common words. Bhag, for instance, in Sanskrit, means to divide, and one of the Vedic gods, Bhaga, meant originally the divider and benefactor. In Zend also Baga appears in the same character, and in the Slavonic languages the Old Slav. bogῠ has become the general name for god. In Greek the same root ϕαγ has completely lost its meaning of dividing, and has entered into a new channel. It means to eat, whether in the sense of dividing the meat with our teeth (ϕαγόντϵς, teeth, Hesych.) or in the sense of sharing a meal with others (as in δαίς, δαίνυμι, δαιτρός, etc.).

All this must be fully admitted, but nevertheless, as little as we could explain why ϕαγ in Greek means to eat, without a reference to the Sanskrit bhag, to divide, could we understand why the great Greek goddess should be called Athene, unless we knew the Sanskrit Ahanâ, and its meaning of dawn.

It is often urged by Greek scholars that the Greeks themselves had no idea that Athene meant originally the dawn, or the verb ϕαγϵι̑ν, to divide. That, no doubt, is true, and it is quite as true that few only of the Greeks knew that Zeus meant originally the sky, and Zephyros the wind blowing from the setting of the sun, or Boreas the wind blowing from the northern mountains. We do not know that Lord meant originally bread-giver, or Duke a man of leading and light; but it is only after knowing it that we can understand the historical growth of the later meanings of Lord and Duke.

Nor is it impossible to discover certain traces in the mythological stories told of Athene which point to her original character as dawn-goddess. Her birth from the head of Zeus is like the rising of the dawn in the Veda from the head of Dyaus (mûrdhâ± Diváh); and it may be in the same sense that she was called Koryphasia, as coming ἐκ κορυϕη̑ς¹¹, and that her counterpart in Italy was called Cap(i)ta. Her purity points to the purity of the dawn, her wisdom to the brightness of the light of the morning, her valour to the irresistible light of her rays. Everything else in her character may be called Greek, and cannot be explained by any reference to Vedic ideas. But what is most interesting to the student of mythology, the germinal idea of the goddess, can be found nowhere else but in the name of Ahanâ, which would have been forgotten in India also, if it had not been for the single verse of the Rig-veda which I quoted to you.

So far, I believe, we are on safe ground. But I think we may venture a step beyond. We saw that the name for morning or day in Sanskrit was ahan or ahar, meaning originally brightness. Now the Teutonic words for day are derived from a root dah, to burn, to be hot. The Gothic dog-s, A.S. dæg, English day, presuppose a root DHAGH, and this exists in Sanskrit as dah, to burn.

Whether the two roots, AH, from which ahan, day, and DAH, from which Goth. dag-s¹², day, are parallel roots, is a question that can only be decided by a full discussion of general principles. To say that an initial d in dah is lost, is saying nothing, for initial d's are never lost without a reason. The same applies to the opposite theory that an initial d was added to the root AH. All we can say is that there are other cases where we find parallel roots, one with, the other without, an initial d. Whether this is mere accident, we cannot tell at present; all we can say is that there are analogies for that process. For instance, we find in Sanskrit asru, tear, probably derived from a root as;, to be sharp and cutting; and we find in Greek δɕκρυ, tear, being evidently derived from a root DAS, to bite¹³. Are we to believe that these two words have nothing in common, and that they do not owe their origin to a common metaphorical concept, namely of sharp and biting, and therefore to a common creative act? Both roots, AS and DAS, exist and have proved prolific in different Aryan languages. From AS, to be sharp (in every sense of the word), we have in Sanskrit asra and asri, point, edge, in Latin acus, âcer, in Greek ἄκρος and ἄκρις. As acidus, from meaning sharp, comes to mean bitter and sour, asru in Sanskrit and Zend, aszara in Lituanian, came to mean a bitter tear.

From DAS, to bite (bitter comes from to bite, Sk. bhid, Lat. findo), we have in Greek δÿκρυ, in Lat. lacruma, in Gothic tagr, in English tear, and who can doubt that all these words mean originally the biting tear? Of course, we can doubt anything, and as it always looks much more learned to doubt than to accept, the temptation to shake one's head is very great. But for that very reason this cheap scepticism deserves a sharp rebuke, such as Professor Pott, for instance, has lately administered to a learned colleague. ‘Naturally,’ he writes, ‘the determined tone of the professor's veto, when be says “the comparison of asru with dasru is as little justified as that of ahan with day,” signifies nothing¹⁴.’

But even those scholars who maintain that the root AH is in no way connected with the root DAH, cannot deny that Dahanâ would be a perfectly legitimate derivation of the root DAH, which root has given us the names for day in the Teutonic languages. That root DAH presupposes a root DHAGH, and belongs to a whole class of roots in which, according to Grassmann's observation, an initial and final aspirate are necessarily represented in Gothic by initial and final media.

As the final h in the root DAH may represent an original gh or bh, we get two possible varieties, DAGH and DABH. DAGH exists in Sanskrit nidâgha, heat; DABH would in Greek appear as δαϕ. From this δαϕ the Greek by a most regular process could have derived δαϕ-νη¹⁵, and the meaning of that name would have been the same as that of Ahanâ in Sanskrit, namely the burning one, the bright, the brilliant. By the side of Δÿϕνη we have the Thessalian form Δαύχνη, with the guttural final of DAH, and Hesychios mentions δαυχμόυ as a name of the wood of the laurel-tree, because it burns easily (ϵὕκαυστου).

If then we know that Phoebos meant the sun, and few scholars will deny that, and that Daphne may have meant the dawn, we shall probably not look very far for an explanation of the Greek saying, that the Dawn fled before the Sun, and vanished when he wished to embrace her.

But why, it may be asked, was Daphne supposed to have been changed into a laurel-tree? Ethno-psychological mythologists will tell us that in Samoa, Sarawak, and other savage countries, men and women are supposed to be capable of turning into plants, and that, as the Greeks were savages once, they no doubt believed the same, and we need therefore inquire no further. Now, with all possible respect for Ethno-psychologists, or as they are sometimes called, Folk-lorists, I cannot think that this would be much more than explaining ignotum per ignotius. The question that everybody would ask is, Why then did the Samoans and Sarawakians and other savages believe that men and women turned into trees? Neither Totemism, surely, nor Fetishism, nor Tabuism, would help them to that belief. Then why should not the classical scholar be allowed to look for a key nearer home, and when he finds that the laurel, being a wood that burns easily, was therefore called δɕϕνη¹⁶, or fire-wood, why should be not be allowed to say that the legend of Daphne, the dawn, being changed into daphne, the laurel-tree, may have been due to the influence of language on thought, to some self-forgetfulness of language—in fact, to the same influence which induced people to adopt an ox passing a ford as the arms of Oxford?

Whether cases of identity of name, like that of Daphne and daphne, are likewise at the bottom of the Samoan and Sarawakian belief that men and women can be turned into plants, is a far more difficult question to answer, and before we generalise on such matters, it is far better to inquire into a number of single cases, such as those, for instance, of Hyacinthus, Narcissus, and others in Greece and elsewhere. We shall find, I believe, here as elsewhere, that the same effect is not always due to the same cause, but unless we find some kind of cause, Comparative Mythology might indeed be called a collection of rubbish, and not a museum of antiquities. To say that ‘a legend of a woman being changed into a tree is explained when we have shown that it is natural to a race which believes in woman being changed into trees,’ is surely not saying very much.

Professor Gruppe has a curious way of dealing with these mythological etymologies. He asks whether they can be true, and then leaves the matter alone. ‘Are we allowed,’ he writes (p. 90), ‘to declare Daphne, the laurel, the beloved of Δαϕνηϕόρος, to be the dawn, because this name, by no means clear as yet, corresponds perhaps to Sk. dahanâ, which is said to be identical with ahanâ, an adjective of the dawn? This is a combination which ignores the atoning and purifying power ascribed to the laurel not by the Greeks only.’ What can be gained, by such remarks? Daphne, the dawn, was called Daphne on account of her blazing light, and not because she was originally a laurel-tree. The laurel-tree was called δɕϕνη, because, if used as fire-wood, it blazes up quickly. These were two quite distinct acts of naming, and their synonymy produced, as often, a later legend. We might as well reject the identification of Dyaus and Zeus, because it ignores the moral character of Ζϵὺς ξέυιος!

But although nothing really important could be brought forward against my equation Ahanâ= Athene, the fact that another scholar had propounded another etymology seemed to offer a great opportunity to those who imagine that by simply declaring themselves incompetent to decide between two opinions, they can prove both to be wrong. Now Benfey's etymology¹⁷ of Athene is certainly extremely learned, ingenious, and carefully worked out; yet whoever will take the trouble to examine its phonetic foundation, will be bound in common honesty to confess that it is untenable. We are dealing here with facts that admit of almost mathematical precision, though, as in mathematics, a certain knowledge of addition and subtraction is certainly indispensable for taking part in the discussion. I speak of the phonetic difficulties only, for if they should prove un surmountable we need not inquire any further.

Benfey (p. 21) places his equation before us, as follows:—

Leaving aside the etymology of Trîtonia, which may be right, quite independently of that of Athene, we have to consider whether ᾽Αθȥνα or ᾿Αθήνη or ᾿Αθηναίη can be the same word as the Zend áthuyánô. And here, though willing to make every allowance for local and dialectic irregularities, I must say decidedly, No.

Ɣthwyânô is a peculiarly Zend modification which presupposes a Sanskrit ptyânâ. This is therefore the only word we can deal with, when looking for an etymology of the Greek word Athana. It is true that even this âptyânâ does not exist in Sanskrit, but we find there âp-tya, i.e. aquaticus, an epithet of Trita, of Indra, and, later on, of a whole class of legendary beings. From this âp-tya, however, no road leads to Athana, and even Benfey himself is obliged to confess at almost every step, that the phonetic changes which he postulates are without any analogy whatsoever.

He first maintains that Athene is connected with Atthis. But though Atthis, or Attica, is under the patronage of Athene, the two names are quite distinct. This becomes still clearer, when he traces Atthis back to Attike, for how can tt ever stand for th? I admit that there is no proof of Attike being derived from ἀκτή, shore, which would have rendered a transition to Atthis and Athene quite hopeless. But even after rejecting the derivation from ἀκτή, how shall we get from âptya to Attike? Benfey says the pt in âptya may become tt, as in πέπτϵιν=πέπτϵιν But in πέπτϵιν (pak) the original final consonant of the root is a guttural, not a labial. Benfey himself feels this, and he therefore appeals to the base at for ap, which appears in Sanskrit ad-bhis instead of abbhis; and postulating a further case apt, he changes ap-tya into apt-tya. He then argues that in pty, p and t are assimilated to tt, that the last t is aspirated through the influence of y, so that τθ is=pty. But here again his conscience smites him, for he admits that ty in Sanskrit never becomes θ in Greek. If so, it follows that pty cannot become τθ. Again, when he postulates the loss of τ, in order to arrive from Atthis at Athene, he honestly confesses that no analogy can be found for this, and yet he adds: ‘the connection of Atthis and Athene is so completely beyond the reach of doubt that it is not injured in the least by this defect.’

Lastly, when he perceives that the, first vowel in Athene is short, while it is long in âptya, he tries to explain this by the accent, which is again impossible; or he simply postulates a form ᾰptya, by the side of âptya, which, however, has no existence.

I doubt whether after this, a single Sanskrit scholar would put his name to the equation âptyânâ= Athana, and there is no necessity therefore to examine the further speculations, which are based on it. If Athana, according to Benfey, is the lightning, and not the dawn, or if she is, according to others, both the lightning and the dawn, this would have to be established by other evidence; it cannot be established by her name. The equation Ahanâ=Athene, on the contrary, is phonetically irreproachable, and mythologically perfectly intelligible¹⁸. I do not wish to deny the principle to which Benfey appeals so frequently in his essay on Athana, namely that dialectic irregularities must by necessity abound in mythological names. There are limits, no doubt, to our respect for phonetic laws, but this applies chiefly to cases where the full bearing of a law has not yet been settled, not where we know the law and knowingly break it. If, for instance, we are told that there is no phonetic law sanctioning the change of nis or nakta into νύξ, νυκτός of ἐπτÿ and ὀκτώ into ἔβδομος and ὄγδοος, all I can say is, that though an adequate cause of the change of a into i and v, of π into β, and of κ into γ, is not yet known, it will be known in time. I am old and bold enough to declare that, in spite of all that has been written on the subject, I still believe in the relationship of θϵός and deus, because, though I cannot fully account for it phonetically, it seems to me far more unaccountable that the Aryan word for God should have been lost in Greek, and been replaced afterwards by another, nearly identical in form and meaning, but totally distinct in origin¹⁹. And even if we yielded on the point of θϵός, and admitted that it could not be connected with Sanskrit deva, bright, and Latin deus, god, how could we separate the brilliant and heavenly goddess Theia from the root div or dyu, to shine, she who is the wife of Hyperion, the mother of Helios (Thiae clara progenies, Cat. 66, 44), of Selene, and Eos, and the daughter of Uranos and Ge? What can be the meaning of θϵι̑ος, θέοϵιδής Lakon. σϵι̑ος when applied to men like Odysseus, if not θϵοϵί-κϵλος, god-like, or θϵοϵιδής, of godly kind, or θϵογϵνής, born of god? If then the same Odysseus is called Διογϵνής, sprung from Zeus, or δι̑ος, divine, excellent, if we find in Homer θϵι̑νϵ γένος and δι̑ον side by side, are we to suppose that Διο and θϵο have no connection whatever with each other²⁰? By all means let us put a mark against all these names, for they still require justification; but let us not suppose that to be dogmatic negatively is less objectionable than to be dogmatic positively.

If it could be proved that Greek and Sanskrit had no mythological names in common, there would, of course, be an end of Comparative Mythology in the narrow sense of the word. We might still be able to compare, but we could no longer think of identifying gods and heroes, having no common name, and therefore no common origin. We can, if we like, compare Jupiter, Jehovah, and Unkulunkulu, but we cannot identify them. We should find many things which these three supreme deities share in common, only not their names, that is, not their original conception. We should have in fact morphological comparisons, which are very interesting in their way, but not what we want for historical purposes, namely genealogical identifications.