language known to us. It had no grammatical form. So fluctuating were its phonetics, and so much depended on gesture, tone, and stress, that its words could not have been reduced to writing, nor arranged in alphabetic order.' To give an idea of what he supposes the phonetic chaos of that palaeolithic language to have been, he mentions that in the Araucanian of Chili the following letters are permutable. B may become W, WF, FU, U Ú, Ú 1, IE, EG, G GH, ÚI,

GH Hu1.

But that is not all. 'These palaeolithic words often signified logical contradictories, and which of the antithetic meanings was intended could be guessed only from the accent or a sign.' This will delight Dr. Abel. It possessed no prepositions nor conjunctions, no numerals, no pronouns of any kind, no forms to express singular and plural, male nor female, past nor present. The different vowel sounds and the different consonantal groups conveyed specific significance, and were of more import than the syllables which they formed.'

This last rather mysterious theory of vowels and consonants being more significant than the syllables which consist of them is illustrated by some remarks made by Bishop Faraud 2, on the Tinné or Athapascan language, spoken widely in British America, and closely allied to the Apache and Navaho dialects, spoken in the United States. Being, as we are told, a thorough master of Tinné, the Bishop states that its significant radicals are the five primitive vowels, A, E, I, O, U. Of these A expresses matter, E existence,

1 Dr. Darapsky, La Lengua Araucana, Santiago de Chile, 1888, p. 15. 2 Dix-huit ans chez les Sauvages, p. 85.

Р

I force or energy, O existence doubtful, and U existence absent, non-existence, negation, or succession. These vowels are put in action by single or double consonants, which have more or less value in proportion as the vowel is more or less strong. Father Petitot1 tells us that there are sixty-three consonants, divided into nine classes, each of which conveys a series of related or associated ideas in the native mind. Labials express the idea of time and space, as age, length, distance, and also whiteness-the lastmentioned, perhaps, through association with the white hair of age, or the endless snow-fields of their winter. The dentals express all that relates to force, &c. &c.'

Here I stop, and though I am afraid it will sound most audacious, I cannot help expressing my conviction that all this is simply wrong, and that language could never have been built up with such materials, as little as it was built up with interjections. I know this audacity will seem quite intolerable. My only excuse is that I could produce books published during the prehistoric times of Comparative Philology, in which English and other Aryan languages have been reduced as triumphantly to significant vowels and significant consonants.

The Historical School therefore leads us up to a certain point, up to where all is safe, but beyond which all is darkness, at least without the light of hypothetical illustration. It never pretends to prove that the roots which it leaves as ultimate facts were the primordial elements of human speech. It admits the possibility of aeons after aeons between the first man,

1 Petitot, Dictionnaire de la Langue Dénée Dindjie.

fresh from the hands of nature, and the roots of the Aryan or Semitic family of speech. All it does is to venture on a guess. We found that nearly all the concepts expressed by these roots are significant of acts. Now as the great difficulty, which is hardly ever realised by anthropologists, consists in our having to account for the origin of concepts, and sounds expressive of concepts, and not merely of percepts, and sounds expressive of percepts, the suggestion first made by Noiré is that these roots were originally sounds uttered by men while performing certain acts in common. How little, the real character of this theory has been understood is best shown by the fact that it has been actually mistaken for what is called the Yo-heoic theory. No doubt it is a suggestion, and no more, for who would dare to speak with positive certainty on matters so distant from us in time, and still more distant from us in thought? All we can say is that such a suggestion would fulfil three essential conditions; it would explain the simultaneous origin of concepts and roots; it would account for their intelligibility among fellow-workers, and it would explain what has to be explained, viz. conceptual, not perceptual language; language such as it is, not language such as it might have been. If any one has anything better to suggest, let him do so; if not, his utere mecum.

Advantages of both Theories.

We

I certainly do not wish to throw unmerited contempt on the Theoretical School. Far from it. want the theorist quite as much as the historian. The one must check the other, nay, even help the other,

just as every government wants an opposition to keep it in order, or, I ought perhaps to say, to give it from time to time new life and vigour. I only wished to show, by an example or two, what is the real difference between these two schools, and what I meant when I said that, whether by temperament, or by education, or by conviction, I myself have always belonged to the Historical School.

Science of Religion.

If now we return to the Science of Religion, we shall find here again the same difference of treatment between the historian and the theorist.

The theorist begins by assuring us that all men were originally savages, or, to use a milder term, children. Therefore, if we wish to study the origin of religion, we must study children and savages.

Now at the present moment some savages in Africa, Australia, and elsewhere are supposed to be fetishworshippers and nothing else. Therefore we are assured that five thousand or ten thousand years ago religion must have begun with a worship of fetishes -that is, of stones, and shells, and sticks, and other inanimate objects.

Again, children are very apt not only to beat their dolls, but even to punish a chair or a table, if they have hurt themselves against it. This shows that they ascribe life and personality-nay, something like human nature-to inanimate objects. Hence we are told that savages would naturally do the same, or have actually done the same from the earliest time to the present day. A savage is, in fact, the most obliging creature, for he does everything that any

But, even then, the

anthropologist wishes him to do. question of all questions, why he does what he is supposed to do, is never asked. We are told that he worships a stone as his god, but how he came to possess the idea of God, and to predicate it of a stone, is called a metaphysical question of no interest to the student of anthropology-that is, of man. Neyertheless it is the primary question that is of interest, and the most vital interest to us.

If then we press for an answer to this all-important question, we are informed that animism, personification, and anthropomorphism are the three well-known agencies which fully account for the fact that the ancient inhabitants of India, Greece, and Italy believed that there was life in the rivers, the mountains, and the sky; that the sun, and the moon, and the dawn were cognisant of the deeds of men, and, finally, that Jupiter and Juno, Mars and Venus, were endowed with the form and the beauty, the feelings and passions of men We might as well be told that all animals are hungry, because they have an appetite.

We read in many of the most popular works of the day how, from the stage of fetishism, there was a natural and necessary progress to polytheism, monotheism, and atheism, and after these stages have been erected one above the other, all that remains is to fill each stage with illustrations taken from every race that ever had a religion, whether these races were ancient or modern, savage or civilised, genealogically related to each other, or perfect strangers.

Again, I must guard most decidedly against being supposed to wish to throw contempt or ridicule on this