Now this is precisely what the writer, with a sure scientific instinct, carefully guards against, and explicitly informs us he did not intend. He tells us that he gives the '_generations_ of the sons of Noah' and their descendants, and at the ends of the three lists relating to these sons, he is careful to say that he has given them 'in their lands, each according to his language, after their families, in their nations,' or the formula is slightly varied into 'after their families, after their tongues, in their lands, in their nations.' Lastly, in the conclusion of the whole table he reiterates, 'These are the _families_ of the sons of Noah, according to their generations, after their nations.' All these statements, let it be observed, are acknowledged to be parts of one (Elohistic) document. It is clear, therefore, that the writer intends us to understand that the determining elements of his classification are neither physical characters nor accidents of geographical distribution, but descent and original language--two primary and scientific grounds of classification, and which common sense requires us to adhere to in interpreting the document, whose value will depend on the certainty with which the writer could ascertain facts as to these criteria: criteria which are, of course, less open to the observation of later inquirers, who may find difficulty in ascertaining either descent or _original_ language, and in default of these may be obliged to resort to other grounds of classification.
[Illustration: MAP SHOWING LINES OF POSTDILUVIAN MIGRATIONS FROM SHINAR, AS IN GENESIS X.]
Among modern archæologists it has been a fruitful source of controversy whether we should classify men according to their skulls or to their tongues; in other words, whether physical characters or linguistic should be dominant in our classifications. Neither ground is absolutely certain. We may find long and short skulls in the same grave-mound, and there are intermediate forms which defy certain arrangement. In like manner history assures us that people of one race have often adopted the language of another. True science warns us that we may err unless we give a fair valuation to every available character. The ethnologist of Genesis considers both physical and linguistic characters, but bases his arrangement mainly on the sure ground of descent along with _original_ language.