Unfortunately, we entirely lack the means of carrying back the history of human speech to its first beginnings.  In the latter half of the last century, whilst the ferment of Darwinism was freshly seething, all sorts of speculations were rife concerning the origin of language.  One school sought the source of the earliest words in imitative sounds of the type of bow-wow; another in interjectional expressions of the type of tut-tut.  Or, again, as was natural in Europe, where, with the exception of Basque in a corner of the west, and of certain Asiatic languages, Turkish, Hungarian and Finnish, on the eastern border, all spoken tongues present certain obvious affinities, the comparative philologist undertook to construct sundry great families of speech; and it was hoped that sooner or later, by working back to some linguistic parting of the ways, the central problem would be solved of the dispersal of the world’s races.

These painted bubbles have burst.  The further examination of the forms of speech current amongst peoples of rude culture has not revealed a conspicuous wealth either of imitative or of interjectional sounds.  On the other hand, the comparative study of the European, or, as they must be termed in virtue of the branch stretching through Persia into India, the Indo-European stock of languages, carries us back three or four thousand years at most—­a mere nothing in terms of anthropological time.  Moreover, a more extended search through the world, which in many of its less cultured parts furnishes no literary remains that may serve to illustrate linguistic evolution, shows endless diversity of tongues in place of the hoped-for system of a few families; so that half a hundred apparently independent types must be distinguished in North America alone.  For the rest, it has become increasingly clear that race and language need not go together at all.  What philologist, for instance, could ever discover, if he had no history to help him, but must rely wholly on the examination of modern French, that the bulk of the population of France is connected by way of blood with ancient Gauls who spoke Celtic, until the Roman conquest caused them to adopt a vulgar form of Latin in its place.  The Celtic tongue, in its turn, had, doubtless not so very long before, ousted some earlier type of language, perhaps one allied to the still surviving Basque; though it is not in the least necessary, therefore, to suppose that the Celtic-speaking invaders wiped out the previous inhabitants of the land to a corresponding extent.  Races, in short, mix readily; languages, except in very special circumstances, hardly at all.