________________
LECTURE III.
105
exist in the same form in French, Italian, and Spanish, to show what words, and therefore what things, inust have been known to the people who did not as yet speak French, Italian, and Spanish, but who spoke that language which preceded these Romance dialects. We happen to know that language: it was Latin; but if we did not know a word of Latin or a single chapter of Roman history, we should still be able, by using the evidence of the words which are common to all the Romance languages, to draw some kind of picture of what the principal thoughts and occupations of those people must have been who lived in Italy a thousand years at least before the time of Charlemagne. We could easily prove that those people must have had kings and laws, temples and palaces, ships and carriages, high roads and bridges, and nearly all the ingredients of a highly civilised life. We could prove this, as I said, by simply taking the names of all these things as they occur in French, Spanish, and Italian, and by showing that as Spanish did not borrow them from French, or Italian from Spanish, they must have existed in that previous stratum of language from which these three modern Romance dialects took their origin.
Exactly the same kind of argument enabled us to put together a kind of mosaic picture of the earliest civilisation of the Aryan people before the time of their separation. As we find in Greek, Latin, and Sanskrit, also in Slavonic, Celtic, and Teutonic, the same word for house, we are fully justified in concluding that before any of these languages had assumed a separate existence, a thousand years at least before Agamemnon and before Manu, the ancestors of the
