LECTURE IV.

In the previous Lectures, after briefly describing the steps which led to the great discovery of Comparative Philology, I have endeavoured to set before you a few of its most remarkable results, and to call your attention to the progress of the religious ideas of the Semite, and the arms and civilisation of the Aryan, across the world. In doing this I wished to sum up the chief historical conclusions to which the Science of Language has hitherto attained; but it is probable that discoveries no less startling, and inferences even more important, may await her in the vast field to which I now invite your attention, a field so vast that it might well occupy a series of many Lectures, and to which it is utterly impossible to do justice in the hour which alone remains to me of my present course.

You will see it stated in many modern treatises that the languages of the world may be divided into three great Families-the Semitic, the Aryan, and the Turanian. Now, unless the word Family be used in two entirely different senses, I must at the very outset protest against any such classification as

L

illusory and unscientific. The Aryan is a family; the Semitic is a family; the so-called Turanian, unless it be confined within very narrow limits, is in no sense of the word a family, but a vast seething mass of human languages hitherto most imperfectly known, and most superficially compared together. These languages are spoken by tribes and nations which have no ethnographical affinities, and many of them differ from each other as completely and fundamentally as it is possible for languages to do.* To speak of them as forming a family is to force a number of gratuitous hypotheses into the shadowy semblance of a scientific generalization. And the very name Turanian is altogether unfortunate, for at the best it has a mere geographical significance, and can only be correctly applied to the natives of Turkestan. It is simply unwarrantable, as Professor Pott demonstrated fourteen years ago,† to open it like a great convenient bag' and fling promiscuously into it languages so radically diverse as Basque, Malay,

* I have already touched on this subject in a paper read before the Ethnological Society (Mag. 1865), and printed in their Transactions. No doubt the dissemination and (as I hold) misapplication of the name Turanian is due in a great measure to Bunsen, whose glowing human sympathies were delighted by what seemed to him a splendid and well-established generalisation. Professor Max Müller, although he holds what may be called the Turanian theory, has always written of it, and especially in all his later works, with conspicuous caution and moderation; he has even checked the unscientific zeal of philologians who were too hasty in adducing arguments in its favour. See Stratification of Language, pp. 42, 43. In the Deutsch. morgenl. Zeitsch. ix. 417.

Polynesian, American, African, Australasian, with Chinese underlying them all as an 'inorganic' Turanian structure. In fact, so much has this unfortunate word been abused, so completely have English and American writers made it a sort of hypothetical sandrope to tie together languages absolutely alien to each other, that the original inventor of the word, the venerable Omalius d'Halloy, in the last edition of his Élémens d'Ethnographie, deliberately abandoned it for the name Alatyan, a name applied to themselves by the Tatars of Siberia. He says that the name Tatar should be banished from Ethnology, because under it are confused together people of two great races, the white and the yellow; consequently in 1840 he adopted instead the name Turanian ; but in 1859, regretting the wholly undue extension which had been given to the term, and finding from M. Levchine's travels among the Khirgiz-Kazaks, that the Tatars of Siberia call themselves Alatys, he deliberately adopted this name and excluded the other.*

Great as is the discovery that languages mutually unintelligible may yet be rigorously proved to be connected, by the close similarity of their grammatical structure and the rigid identity of their roots, it is a discovery which will lose half its value if it be hastily and impetuously applied to languages wholly distinct from each other. It is far too early in the day to talk of the Turanian Unity,' unless we limit the Omalius d'Halloy, Élém. d'Ethnographie, p. 52.

6