But this proposition is hedged in by difficulties. The very unity of the great Arian and Semitic families tells powerfully against it. If the members of these families retain, after the separation of many hundred years, the most striking similarity, in the roots of the words which refer to the relations of life, and to the primitive acts of weaving and the working of metals, how is it possible to believe that the points of resemblance between Sanskrit and Hebrew, or between Chinese and Greek, are so extremely few, and so dubiously vague, that they hardly afford the shadow of a presumption in favour of the hypothesis which they are adduced to support? Even if we grant the postulated length of time—thousands and thousands of years—which take us back to a period when historical “chronology borders on the geologic eras,” which will alone render such a diversity of sister languages possible, we confess that it still appears to us so improbable, that it rather wears the appearance of an arbitrary hypothesis, than an inductive conclusion.
2. The main affinities supposed to exist between language of the different families, will be found at large in the “Outlines of the Philosophy of Universal History.” Great stress is there laid (i.) on the supposed discovery of certain non-Sanskritic elements in Celtic, which form the link by which the Indo-European family approaches the Turanian formations; and (ii.) the establishment of a connection between the Arian and Semitic families, by a reduction of the Hebrew triliteral roots to biliteral ones.
(i.) While wishing to allow the fullest weight to everything which has been adduced by Dr. Meyer in proof of this discovery, and not professing to be fully able to weigh the value of the evidence, we cannot think that his researches have at all settled the question. Beyond certain accidental and vague resemblances, a few lexicographical similarities[260] easily explicable by onomatopœia, and a few words[261] adopted in consequence of foreign influences, and that general affinity which we should expect from the ascertained fact of the psychological unity of the human race, nothing that we have hitherto met with seems at all adequate to counterbalance the enormous difficulty of supposing that families, closely united together, yet radically distinct from each other, could, even during thousands of years, have diverged so widely from a common source. Again, we must ask, if it was possible for one primitive language to pass through stages of development so irreconcilably different as those represented by Hebrew and Sanskrit, what cause can be adduced sufficient to account for the fact that after the lapse of three millenniums, a Lithuanian peasant could almost understand the commonest of Sanskrit verbs?[262]
The Chinese must always remain a stumbling-block in the way of all theories respecting a primitive language. Radical as is the dissimilarity between Arian and Semitic languages, and wide as is the abyss between their grammatical systems, yet they almost appear like sisters when compared with the Chinese, which has nothing like the organic principle of grammar at all. Indeed, so wide is the difference between Chinese and Sanskrit, that the richness of human intelligence in the formation of language receives no more striking illustration than the fact that, as we have already observed, these languages have absolutely nothing in common except the end at which they aim. This end is in both cases the expression of thought, and it is attained as well in Chinese as in the grammatical languages, although the means are wholly different.
(ii.) Very great stress has been laid on the general lexicographical affinity between Hebrew and Sanskrit, produced by the reduction of the Hebrew triliteral roots to biliteral ones. This was suggested by Klaproth, and supported with great learning and industry by Fürst and Delitzsch. We have already alluded to it, and can only repeat here, that it is not accepted as certain, or even as probable, by some high authorities. We cannot now recount the numerous and weighty objections brought against this attempt by the historian of the Semitic languages[263]—objections derived mainly from the extreme laxity of the process which even involves the extraordinary hypothesis, that these triliteral roots were formed by prefixes and suffixes, and that the prefixes have nothing determinate about them, but that every letter in the alphabet might be used for the purpose,—an hypothesis contrary to the most essential principles of language. It will be sufficient to repeat his questions. How can we conceive the passage from the monosyllabic to the triliteral stage? What cause can be assigned for it? At what epoch did it take place? Was it due to the multiplication of ideas or the invention of writing? Was this stage of grammatical innovation the result of chance, or of a common agreement? To these inquiries, no answer ever has been or can be given. The supposition of an original biliteralism must be considered (as we said before) simply as a convenient hypothesis, and must not be taken for an historical fact.
Languages, of course, develop; but it is, as we have seen, by the germinal development of a rudimentary idea, and not by this process of gross exterior concretion for which no single parallel can be suggested. The only monosyllabic dialects which we know, viz., those of Eastern Asia, have continued monosyllabic for unknown ages. Chinese cannot attain to a grammar, and the Semitic languages could never arrive either at regularly written vowels, or at a satisfactory system of moods and tenses. Grammar is to a language its unalterable individuality. The growth and change of language has nothing analogous to grammatical revolution; it is due to a silent, a spontaneous, an unconscious genius, not to deliberate reflexion, or conscious alteration. All idioms which have been artificially altered (e.g. Rabbinic Hebrew), betray the fact by their harshness and awkwardness,—their want of harmony and flexibility; they bear no resemblance to those languages which are the genuine instrument of a nation’s thoughts.
3. Undoubtedly the strongest argument in favour of a Primitive Language arises from the phenomenon of several languages which appear to occupy an anomalous position on the frontier of the great kingdoms of speech, and to present a lexicographical affinity with one family, and a grammatical affinity with another family. Such languages are the Egyptian, the Berber, the Touareg, and generally the languages of Northern and Eastern Africa, which resemble the Semitic tongues, in some parts of their vocabulary, but differ widely from them in all the rest. Similarly, the Tibetian and Burmese stand on the confines of the monosyllabic languages.
Perhaps the only way to account for these strange appearances is to suppose that language had a period of primitive fusibility,[264] during which they were susceptible of great modification from contact with other languages also in an ante-historical and embryonary state. It is impossible, otherwise, to explain the identity, for instance, of the pronouns and numerals in Coptic and the Semitic languages, or to account for the fact that among different races t is the sign of the second person singular, and n, of the first person plural. The analogies which guided the first men in such cases entirely escape our power of perception. Philology in its present state has not sufficient materials to decide how can it be that a few essential elements in a vocabulary should be nearly the same in two languages, while yet they differ totally in so important a particular as the flexions of the noun and verb. We know, however, as an historical fact, that wide as is the difference between the Semitic and Egyptian systems of civilisation, and different as are the physical traits of the two nations, yet that for many ages the Semitic influence was very strongly felt in Egypt.[265] Egypt, indeed, was only a narrow valley, surrounded by Semitic Nomads, who lived side by side with the sedentary population; sometimes victorious, sometimes subject,—always detested. The Egyptian language belongs then to a Chamitic family, to which also belong the Berber, and other indigenous languages of Northern Africa; a family which is spread in Africa from the Red Sea to Senegal, and from the Mediterranean to the Niger.
Of these languages, the Berber presents numerous grammatical affinities with the Hebrew, but is completely distinct in its vocabulary. This, too, may be accounted for by the fact, that it has also been submitted to long ages of Semitic influence, in consequence of its relations with Carthaginian and Arabic. The possibility of a state of language so incomplete as to admit of these radical influences from contact with superior idioms, is an important subject for philological inquiry.
We are forced then to conclude that whatever may be the other arguments, physiological and historical, for a material unity of the human race, a belief, which understood in a high psychological sense, will meet with universal acceptation, philology alone, so far as it has yet proceeded, adds no contribution to the probability of such a view. Of the primitive men we know little or nothing, nor can we advance beyond the region of conjecture; but language does reveal to us something about the origin of nations, and the apparition of the main races of humanity would appear to have been in the following succession.