Besides the immense number of languages now spoken over the surface of the globe, we must remember that hundreds have now died away altogether, and left no trace behind them. Even in our own times, languages are dying out; the last person who could speak Cornish died almost within this generation,[254] and it is probable that Manx will not long survive, although it may be violently galvanized into a semblance of vitality. Many of the sporadic dialects, spoken by the North American Indians, have disappeared with the tribes that spoke them; and Humboldt even mentions that he had seen a parrot which was the only living thing that preserved the articulation of one forgotten tongue. Every extant language has grown out of the death of a preceding one.[255] “Like a tree, unobserved through the solitude of a thousand years, up grows the mighty stem, and the mighty branches of a magnificent speech. No man saw the seed planted; no eye noticed the infant sprouts; no register was kept of the gradual widening of its girth, or of the growing circumference of its shade, till the deciduous dialects of surrounding barbarians dying out, the unexpected bole stands forth in all its magnitude, carrying aloft in its foliage, the poetry, the history, and the philosophy of an heroic people.”[256]
Thus the Greeks and Romans[257] displaced by their dominant idioms numerous languages of Southern and Central Europe; the Arabs effaced the indigenous dialects of a large portion of Western Asia, and Northern and Eastern Africa; the Spanish and Portuguese have expelled a crowd of American languages. Again, the Visigoths and Alani lost in Spain both their name and their language; the Ostrogoths and Heruli suffered the same fate in Italy; and in short, we may fairly suppose that the dead languages of the world are nearly as numerous as those that are still living.
Passing over the dead languages, is it possible to deduce even all living languages from one primitive speech?
Even those who believe in a primitive language admit that the three families of language are irreducible, i.e. incapable of being derived from one another.
“These three systems of grammar (Arian, Semitic, and Turanian), are,” says Professor Max Müller, “perfectly distinct, and it is impossible to derive the grammatical forms of the one from those of the other, though we cannot deny that in their radical elements the three families of human speech may have had a common source.”
Attempts have, indeed, been made to connect Hebrew and Sanskrit, but the adduced points of osculation are so few and dubious, that such attempts must be pronounced to be egregious failures. Dr. Prichard endeavoured to prove a connection between Celtic and Hebrew, but “he succeeds no better than those who had made the same attempt before him. In nearly every case, the identity of the terms compared is questionable, and in many it is demonstrably imaginary.”[258]
It must then be allowed, that the Indo-European and Semitic families are in their grammatical system (which affords the truest, if not the only test of affinity) radically distinct, and can in no way be derived from each other. The motto of the old school, that “all languages are dialects of a single one,” must be abandoned for ever.
But even if it could be shown that there is an affinity between Hebrew and Sanskrit, a far more difficult task would remain for those who endeavour to prove from philology the original unity of the human race; for it would be still necessary for them to show further the Turanian unity, and the possibility of a primitive nucleus, not only for Semitic, and Arian, and Turanian languages, (assuming this to comprise even the Malay, Australian, Papuan, Kaffir, Esquimaux, &c.), but also for these languages and the ungrammatical, unagglutinative, monosyllabic Chinese. Yet, such is the task undertaken, with vast learning and marvellous ingenuity, by Professor Müller and Baron Bunsen. It will, however, be admitted, that the proved existence of great irreducible families is a strong à priori evidence against them. Let us examine some of their main arguments.
1. “Though in physical ethnology we cannot derive the Negro from the Malay, or the Malay from the Negro type, we may look upon each as a modification of a common and more general type. The same applies to the types of language. We cannot derive Sanskrit from Hebrew, or Hebrew from Sanskrit: but we can well understand how both may have proceeded from one common source.”[259]
Thus it is argued, that although these families of language cannot, in their present state, have been derived from each other, yet it is possible to suppose that they are widely diverging radii from the same original centre; that they may all have sprung from a primitive language, whose existence we may conjecture, just as we should have conjectured the existence of such a language as the Latin, to account for the numerous marks of affinity between the Romance dialects.