Page:The Tarikh-i-Rashidi - Mirza Muhammad Haidar, Dughlát - tr. Edward D. Ross (1895).djvu/111

 80 quently the distinctive features of their race would take longer to undermine. As already observed, the life of the steppes and the comparative isolation of the aul, would tend rather to preserve the purity of the race. It may not be possible to form an estimate of the length of time that would be needed to bring about a change of type by gradual intermarriage, but we know, at any rate, of one instance where this same Mongol people, from living in more or less isolated positions, and mixing with neighbouring races only to a very slight extent, have preserved all the physical characteristics of their original type, as well as the language, down to our own day—or some six and a half centuries from the date of their transplantation, during the era of the Mongol conquests. I refer to the Hazáras of Afghanistan, most of whom are still as unmistakably Mongol in feature and build as the inhabitants of Mongolia itself. According to the most trustworthy accounts of them, they descend from the remnants of the army of Nikudar Oghlán, a son of Hulaku, who invaded the region in which they dwell now, about the latter half of the thirteenth century; while Professor von der Gabelentz has shown that, in spite of a slight mixture of Persian words, their language is still strictly Mongolian, or more particularly, West Mongolian—i.e., Kalmák.

On the general question of the rise and decay of languages, enough is known of the process which a nation has to go through before it can completely change its tongue, to justify the belief that a very long period is needed for the transfer to become finally accomplished. The first step is that the people should become bi-lingual—that the mass of them (not a few of the chiefs) should come to use both the old and the new language with equal facility—and this alone is a process requiring many generations. The next step is that the old language should fall into disuse and be forgotten. The second