Not so with language. The home of a language means merely the accidental place of its formation, and this has no relation to its inner form. Languages migrate in that they spread by carriage from tribe to tribe. Above all, they are capable of being, and are, exchanged--indeed, in studying the early history of races we need not, and should not, feel the slightest hesitation about postulating such speech-changes. It is, I repeat, the form-content and not the speaking of a language that is taken over, and it is taken over (as primitives are for ever taking over ornament-motives) in order to be used with perfect sureness as elements of their own form-language. In early times the fact that a people has shown itself the stronger, or the feeling that its language possesses superior efficacy, is enough to induce others to give up their own language and--with genuinely religious awe--to take its language to themselves. Follow out the speech-changes of the Normans, whom we find in Normandy, England, Sicily, and Constantinople with different languages in each place, and ever ready to exchange one for another. Piety towards the mother tongue--the very term testifies to deep ethical forces, and accounts for the bitterness of our ever-recurring language-battles--is a trait of the _Late_ Western soul, almost unknowable for the men of other Cultures and entirely so for the primitive. Unfortunately, our historians not only are sensible of this, but tacitly extend it as a postulate over their entire field, which leads to a multitude of fallacious conclusions as to the bearing of linguistic discoveries upon the fortunes of “peoples”--think of the reconstruction of the “Dorian migration,” argued from the distribution of later Greek dialects. It is impossible, therefore, to draw conclusions as to the fortunes of the race side of peoples from mere place-names, personal names, inscriptions, and dialects. Never do we know _a priori_, whether a folkname stands for a language-body, or a race-part, or both, or neither--besides which, folk-names themselves, and even land-names, have, as such, Destinies of their own. II