only of social contact. We cannot argue from the exceptional phenomena of the stereotyped families of Aryan, Semitic, and Turanian speech. Savage and barbarous dialects are in a constant state of flux and change, while conquest, migration, and other causes, occasion the borrowing of new languages, and the loss of old ones. . . . The Basques, physically and linguistically, are the representatives of a race which preceded the Kelts, and were driven by them into the mountain fastnesses of the extreme West, just as the Finns were by other Aryan tribes in the North. Just as the existence of light-haired persons among the Basques shows only that mixture of blood which was to be expected, so, from the present state of the Basque language, we cannot draw any conclusion against the view that the primitive population with whom the Aryan Kelts came into contact spoke older but cognate dialects. The oldest Basque with which we are acquainted does not date back beyond three or four centuries; before that time there was no literature, and the changes undergone by languages other than literary are astonishingly rapid and extensive. The few native inscriptions of early date found in northern Spain, so far as they can be deciphered, show little resemblance to modern Basque, while Strabo[1] states that not only had the Iberians many different dialects, but several different alphabets as well. This points to want of intercourse, bringing with it a great diversity of language. Numerous as these languages were, however, they must have had a general resemblance to one another, since Strabo (Book iv.) says they were like the idioms of Aquitania, in contradistinction to those of Celtic Gaul. The modern
- ↑ iii. p. 139.
