merely proves that, in some past age, the various members of the same linguistic group inhabited a common home, and maintained a close intercourse with one another. This, however, is all that we require; for as all races of mankind generate fertile hybrids with one another, residence in a common home is sufficient to produce a new mongrel race, even from stray portions of the human family physically unlike. But here again the consideration arises, that a common home may be inhabited by two physically distinct races united by a predominant language, and yet little or no admixture of blood may have taken place. We see these cases realized in the United States and in India, where admixture of blood only rarely occurs between white and coloured people, or between Aryans of high and natives of low caste. This consideration should be kept in view, though these instances are solitary. The aversion of the English and Germans to intermarriage with negroes is not shared to the same extent by Semites and Hamites, nor, among Europeans, by Spaniards, Portuguese, and French. This feeling of caste only restrains nations of very high culture from an admixture of blood with nations of a very low order. In the newer races of mankind nothing of the sort is to be apprehended. Moreover, as structure of language requires long periods for its development, during which the families with a common language maintain the closest intercourse of ideas, common descent or continued affinity may be inferred in the case of nations connected by a community of verbal structure and parts of speech. No one who has studied the subject any longer doubts that the so-called Indo-Europeans, the Semites, the Bantu nations of South Africa, all derive the rudiments of their languages from intercourse in a common home, where they used a common vocabulary. Yet no comparison of the bodily characters of Icelanders, of Hindoos of high caste, of the natives of Madagascar and Easter Island, would have suggested to us that they were all descendants of ancestors inhabiting a common home and intermarrying. But having observed every rule of critical caution, none but those who have formed exaggerated ideas of the persistency of physical characters will neglect language as a means of classification, or make light of the results of the philological researches of the present day. But where a comparison Connection of Characters with Language. 131 of languages gives results inconsistent with the racial characters, we necessarily suspect an admixture of blood. Hence, we have no hesitation in reckoning the inhabitants of Kashgar among the Turkish hybrid nations, for by their facial type they would otherwise be classed among the Indo-Germans. We must assume that the conquering Turkish-speaking race mingled to such a degree with the subjugated Tadshiks of Iranian stock, that their original bodily characters were entirely obliterated. Linguistic relations, founded on a community of defining auxiliary syllables, are recognized without dispute by all philologists. Those cases in which the similarity depends only on conformity of structure, are more suspicious and more liable to objection. But even in these relationship is admitted, at least with regard to the aborigines of America. Their common use of the "incorporative” system has induced all philologists to regard them as members of a single family of mankind, and to separate them from the Eskimo, who form their words by means of suffixes, especially as there are no distinct bodily characteristics which would suggest a real separation in the former. The association of the UralAltaic nations, in which the community of the various groups depends only on the type of linguistic structure, and its restriction to the suffix as its morphological element, is far more doubtful. Yet even in this instance we may assume a derivation from a common home, because the special character of their rules of euphony at least is peculiar to them; and we may conjecture that if the records of their language reached back some thousands instead of hundreds of years, as is the case, a closer kinship might probably be discovered, and, lastly, because their bodily structure favours this association. On the other hand, it seems inadmissible to elevate the Ural-Altaic groups into a Turanian family, and to assimilate with them the Dravida languages of the aboriginal Indians because they likewise observe laws of euphony in the formation of words. Since these laws differ from those of the Ural-Altaic languages, and also because their physical characters render it imperative, we shall treat these South Indian people as a separate branch of the human family. |