This would indicate a mixed descent, and such, no doubt, it owned. It is absurd to suppose the contrary. The type of the proto-Aryac language is one which originates not early, but late in the history of human speech. The process of grammatical inflection is the highest stage of linguistic evolution. It is the result of a slow growth, in which the material elements of language are transformed into formal elements, and the “grammatical categories,” or parts of speech, gradually assume logical distinctness and independent expression. We can watch this growth in its imperfect form in the Nahuatl of Mexico and the Berber of Morocco; and when we see it completed, as in the Arabic or Latin, we may be sure it is a comparatively late fruit of the human intellect. The expressions common to all Aryac languages reveal a primitive social condition to correspond with this. It was above that of savagery. These common ancestors had domesticated dogs, cattle, and perhaps sheep; nomadic at times, they at some seasons tilled the soil; they were acquainted with copper, and brewed mead from honey; they had probably even invented a wagon, and milked their cows, and they certainly lived on or near the seashore, and used boats.

The conclusion is that the original inflected Aryac tongue arose from the coalescing of two or more uninflected agglutinative or semi-incorporative tongues, the mingling of the speeches being accompanied, as always, by a mingling of blood and of physical traits. This explains the fact that has puzzled so many ethnologists, that there is no fixed Aryac type.

Where should we look for this intermingling to have taken place? From the arguments already advanced you would naturally say, somewhere on the western coast of Europe.

This is supported by an unexpected piece of evidence of a strong character. The system of consonants is undoubtedly the most persistent part of a language, and there is no question but that the Celtic and Lithuanian, of all the Aryac tongues, have kept most closely to the primitive system of consonants once common to them all.[93] The Lithuanian is spoken by a limited community on the coast of the Baltic sea, while the Celtic, in proto-historic times, occupied the whole of Great Britain and northern Belgium, France and Spain. In the two latter areas it was from immemorial time in close connection with the Euskaric (Basque), and perhaps the Libyan (Berber) groups, and it is possible that in comparatively late (neolithic) times the Aryac with its inflections might have been developed from these partly agglutinative languages.

This suggestion is not so hazardous as it may seem. William von Humboldt, one of the ablest linguists of this century, suggested that the Basques and the Celts, the Ligurians and the Gauls, in spite of the contrasted structure of their languages, may have sprung from the same ethnic trunk, and derived their languages from a common source.[94]

Other scholars of eminence, such as Delitzsch, Ascoli, Raumer, Schultze and Abel, have pointed out numerous affinities between the Hamito-Semitic, Libyan, old Coptic and Assyrian tongues, and the oldest Aryac forms, and have argued for the existence of a fundamental “proto-Ayro-Semitic” speech which existed before the separation of the white race into its northern and southern branches.[95] There is evidence that this very ancient tongue was of the “isolating” character, with a tendency to agglutination by suffixes.

It is now recognized that inflection did not exist in the primitive Aryac dialects, but was gradually developed by means of such suffixes added to the stem, by different processes in the different dialects, many of which are in activity to-day.[96] These inflective processes bear closer resemblance to the Libyan, which has suffixes, and the old Egyptian, than to pure Semitic tongues, which leads to the suggestion, again, that the separation of the race was in the west rather than the east.

Proto-Aryac Migrations.—Leaving these speculations as to the origin of the Aryac stock, let us sketch its probable migrations, as indicated by linguistic research. It appears to have divided early into two main streams, the one occupying central and southern Europe, the other moving eastward on a northerly route, the two meeting as they neared the Bosphorus.

The central stream was of Celtic affinities. Its tribes having possessed themselves of the coast line from Cape Finisterre to the mouth of the Rhine and the islands of Great Britain, passed up the valleys of the Rhine and its affluents into southern Germany, the valleys of Switzerland and the Tyrol, quite to the Danube. Its easternmost tribes were probably the Dacians.

The Aryac Italic peoples, the Umbrians, the Oscans, the Latins, were the first offshoot of this southern migration; not that they were directly descended from the Celts, but that they sprang from the same division of the primitive Aryac stock. This is still so clear that I remember Matthew Arnold in his lectures on poetry quotes sentences from ancient Irish which are also intelligible Latin.