We have to distinguish the destiny of a people from its name. The latter is often the only thing about which information remains to us; but can we fairly conclude from a name anything about the history, the descent, the language, or even merely the identity of those who bore it? Here again the historical researcher is to blame, in that, whatever his theory may have been, he has in practice treated the relation between name and bearer as simply as he would treat, say, the personal names of to-day. Have we any conception of the number of unexplored possibilities in this field? To begin with, the very act of name-giving is of enormous importance in early associations. For with a name the human group consciously sets itself up with a sort of sacral dignity. But, here, cult- and war-names may exist side by side; others the land or the heritage may provide; the tribal name may be exchanged for that of an eponymous hero, as with the Osmanli;[220] lastly, an unlimited number of alien names can be applied along the frontiers of a group without more than a part of the community ever hearing them at all. If only such names as these be handed down, it becomes practically inevitable that conclusions about the bearers of them will be wrong. The indubitably sacral names of Franks, Alemanni, and Saxons have superseded a host of names of the period of the Varus battle--but if we did not happen to know this, we should long ago have been convinced that an expulsion or annihilation of old tribes by new intruders had taken place here. The names “Romans” and “Quirites,” “Spartans” and “Lacedæmonians,” “Carthaginian” and “Punic” have endured side by side--here again there was a risk of supposing two peoples instead of one. In what relation the names “Pelasgi,” “Achæans,” “Danai,” stand to one another we shall never learn, and had we nothing more than these names, the scholar would long ago have assigned to each a separate people, complete with language and racial affinities. Has it not been attempted to draw from the regional designation “Doric” conclusions as to the course of the Dorian migration? How often may a people have adopted a land-name and taken it along with them? This is the case with the modern Prussians, but also with the modern Parsees, Jews, and Turks, while the opposite is the case in Burgundy and Normandy. The name “Hellenes” arose about 650, and, therefore, cannot be connected with any movement of population. Lorraine (Lothringen) received the name of a perfectly unimportant prince, and that, in connexion with the decision of a heritage and not a folk-migration. Paris called the Germans Allemands in 1814, Prussians in 1870, Boches in 1914--in other circumstances three distinct peoples might have been supposed to be covered by these names. The West-European is called in the East a Frank, the Jew a Spaniole--the fact is readily explained by historical circumstances, but what would a philologist have produced from the _words alone_?