Strange as it may sound to hear the language of Homer and Ennius spoken of as an offshoot of the Sandwich Islands, mere ridicule would be a very inappropriate and very inefficient answer to such a theory. It is not very long ago that all the Greek and Latin scholars of Europe shook their heads at the idea of tracing the roots of the classical languages back to Sanskrit; and even at the present moment there are still many persons who cannot realize the fact that, at a very remote, but a very real period in the history of the world, the ancestors of the Homeric poets and of the poets of the Veda must have lived together as members of one and the same race, as speakers of one and the same idiom.

There are other theories not less startling than this, which would make the Polynesian the primitive language of mankind. I received lately a Comparative Grammar of the South-African Languages, printed at the Cape, written by a most learned and ingenious scholar, Dr. Bleek. In it he proves that, with the exception of the Bushman tongue, which has not yet been sufficiently studied, the great mass of African languages may be reduced to two families. He shows that the Hottentot is a branch of the NorthAfrican class of languages, and that it was sepa

1 A Comparative Grammar of the South-African Languages, by W. H. J. Bleek, Ph. D. 1862.

2 When the Rev. R. Moffat was in England, a few years since, he met with a Syrian who had recently arrived from Egypt, and in reference to whom Mr. Moffat has the following note: "On my giving him a specimen and a description of the Hottentot language, he remarked that he had seen slaves in the market of Cairo, brought a great distance from the interior, who spoke a similar language, and were not near so dark-colored as slaves in general. This corroborates the statement of ancient authors, whose description of a people inhabiting the interior regions of Northern

rated from its relatives by the intrusion of the second great family, the Kafir, or, as Appleyard calls them, Alliteral languages, which occupy (as far as our knowledge goes) the whole remaining portion of the South-African continent, extending on the eastern side from the Keiskamma to the equator, and on the western side from 32° southern to about 8° northern latitude. But the same author claims likewise a very prominent place for the African idioms, in the general history of human speech. "It is perhaps not too much to say," he writes (Preface, page viii), "that similar results may at present be expected from a deeper study of such primitive forms of language as the Kafir and the Hottentot exhibit, as followed, at the beginning of the century, the discovery of Sanskrit, and the comparative researches of Oriental scholars. The origin of the grammatical forms, of gender and number, the etymology of pronouns, and many other questions of the highest interest to the philologist, find their true solution in Southern Africa."

Africa answers to that of the Hottentot and Bushman."- "It may be conceived as possible, therefore, that the people here alluded to form a portion of the Hottentot race, whose progenitors remained behind in the interior country, to the south or southwest of Egypt, whilst the general emigration continued its onward course. Should this prove not incorrect, it might be reasonably conjectured that Egypt is the country from which the Hottentot tribes originally came. This supposition, indeed, is strengthened by the resemblance which appears to subsist between the Copts and Hottentots in general appearance." (Appleyard, The Kafir Language. 1850.) — "Since the Hottentot race is known only as a receding one, and traces of its existence extend into the interior of South Africa, it may be looked upon as a fragment of the old and properly Ethiopic population, stretched along the mountain-spine of Africa, through the regions now occupied by the Galla; but cut through and now enveloped by tribes of a different stock." (J. C. Adamson, in Journal of the American Oriental Society, vol. iv. p. 449. 1354.)

But while we are thus told by some scholars that we must look to Polynesia and South Africa if we would find the clue to the mysteries of Aryan speech, we are warned by others that there is no such thing as an Aryan or Indo-European family of languages, that Sanskrit has no relationship with Greek, and that Comparative Philology, as hitherto treated by Bopp and others, is but a dream of continental professors.1 How are theories and counter-theories of this kind to be treated? However startling and paradoxical in appearance, they must be examined before we can either accept or reject them. "Science," as Bunsen 2 said, "excludes no suppositions, however strange they may appear, which are not in themselves absurd viz. demonstrably contradictory to its own principles." But by what tests and rules are they to be examined? They can only be examined by those tests and rules which the Science of Language has established in its more limited areas of research. "We must begin," as Leibnitz said, "with studying the modern languages which are within our reach, in order to compare them with one another, to discover their differences and affinities, and then to proceed to those which have preceded them in former ages, in order to show their filiation and their origin, and then to ascend step by step to the most ancient of tongues, the analysis of which must lead us to the only trustworthy con

1 See Mr. John Crawfurd's Essay On the Aryan or Indo-Germanic The ory, and an article by Professor T. Hewitt Key in the Transactions of the Philological Society, “The Sanskrit Language, as the Basis of Linguistic Science, and the Labours of the German School in that field, are they not overvalued?"

2 L. c. p. 256.

clusions." The principles of Comparative Philology must rest on the evidence of the best known and the best analyzed dialects, and it is to them that we must look, if we wish for a compass to guide us through the most violent storms and hurricanes of philological speculation.1

I thought it best, therefore, to devote the present course of lectures to the examination of a very limited area of speech, to English, French, German, Latin, and Greek, and, of course, to Sanskrit, — in order to discover or to establish more firmly some of the fundamental principles of the Science of Language. I believe there is no science from which we, the students of language, may learn more than from Geology. Now, in Geology, if we have once acquired a general knowledge of the successive strata that form the crust of the earth, and of the faunas and floras present or absent in each, nothing is so instructive as the minute exploration of a quarry close at hand, of a cave or a mine, in order to see things with our own eyes, to handle them, and to learn how every pebble that we pick up points a lesson of the widest range. I believe it is the same in the Science of Language. One word, however common, of our own dialect, if well examined and analyzed, will teach us more than the most ingenious speculations on the nature of speech and the origin of roots. We may accept it, I believe, as a general principle that what is real in modern formations is possible in more ancient formations; that what has been found to be true on a small scale

1 Lectures on the Science of Language, First Series, p. 136, note (4th edition).

may be true on a larger scale. Principles like these, which underlie the study of Geology, are equally applicable to the study of Philology, though in their application they require, no doubt, the same circumspectness which is the great charm of geological reasoning.

A few instances will make my meaning clearer. They will show how the solution of some of the most difficult problems of Comparative Grammar may be found at our very door, and how theories that would seem fanciful and incredible if applied to the analysis of ancient languages, stand before us as real and undeniable facts in the very words which we use in our every-day conversation. They will at the same time serve as a warning against too rapid generalizations, both on the part of those who have no eye for distinctive features and see nothing but similarity in all the languages of the world, and on the part of those who can perceive but one kind of likeness, and who would fain confine the whole ocean of living speech within the narrow bars of Aryan or Semitic grammar.

We have not very far to go in order to hear such phrases as "he is a-going, I am a-coming," &c., instead of the more usual "he is going, I am coming." Now the fact is, that the vulgar or dialectic expression," he is a-going," is far more correct than “he is going." Ing, in our modern grammars, is called the termination of the participle present, but it does not exist as such in Anglo-Saxon. In Anglo-Saxon the termination of that participle is ande or inde (Gothic,

1 Archdeacon Hare, Words corrupted by Fulse Analogy or Fulse Deriva tion, p. 65.