1851.

Superficial Classification of Languages.

11

that it would be impossible to advance one step towards a scientific solution of the problem in question, without first trying to arrange the whole mass in certain groups or families.

If in the cases mentioned before, we had taken German or English, instead of Italian and Spanish, to explain the formation of the Future in French, we should have found no explanation at all, or probably a wrong one. Yet both these languages are immediately bordering on the French, and both show by their vocables, that they are largely indebted to the Latin, from which the French also is derived. With regard to modern languages, indeed, a knowledge of the political history of the nations by whom they were spoken, is generally sufficient to indicate their genealogical connexion; but we have no such help for classifying the languages of old nations.

This classification had been attempted long before the rise of Comparative Philology, but it had never before been founded on the nature of language itself. Languages had been classified either according to their geographical distribution, (such as the languages of Europe, Asia, Africa, America, and Australia,) or according to the physical races of man by whom they were spoken (such as the Caucasian, Mongolian, Negro and Malay). It was usual also, to speak of sacred and profane, classical and oriental, living and dead languages, but all these divisions were based only on external accidents. By undertaking, for the first time, a classification of languages according to the peculiar character of their etymological and grammatical structure, Comparative Philology has found that tongues, spoken in the most distant regions of the world, and by nations apparently unconnected by any historical intercourse, may yet belong to the same family, while, in other cases, languages, spoken in one and the same district, can be shown to be of a totally different origin. Great results have been obtained in this manner, and other sciences, like Ethnology and History, have largely availed themselves of these new discoveries. Yet, however interesting and important the facts may be, which have been elicited from a comparison of languages, this subject is as yet very far from being exhausted. There are still languages in the world which have never been studied at all. Many others are known only by scanty and often untrustworthy lists of words. And although the characteristic features and broad outlines of several groups have been established by philological research, yet the number of languages which have been subjected to a careful analysis of their grammatical and etymological structure, is comparatively small. It is true that for general purposes, lists of words, when drawn up carefully, are sometimes sufficient, if not

to prove, at least to indicate, the connexion of languages. This process, however, has so frequently been found unsuccessful, that Comparative Philology has altogether discouraged it. It is true also, that as a first attempt a division of languages, according to their general character, may be instructive. Yet by knowing that certain tongues are monosyllabic, agglutinative, or inflectional, we know little more than a scholar of natural history, who has observed, that some animals have two, and others four legs, while some have no legs at all. It is much the same, as if we were to classify men, birds, and whales as bipeds, or eels and serpents as fishes. This is not meant, to deny that terms such as monosyllabic, agglutinative, and inflectional, synthetic and analytic, are very useful and appropriate for a classification of languages. But such terms have a meaning only after languages have been subjected to the most careful analysis, and, so to say, to a microscopic anatomy of their grammar. They mean nothing at all in the mouths of people, who do not know even the alphabets of the very languages which they venture to classify." * The less we know of lan

* A striking confirmation of this is offered in a work lately published by Dr. Latham, On the Varieties of Man.' This gentleman, to whom we owe already a history of the English language, embodying the results of Grimm's celebrated Teutonic Grammar, has also thought it necessary in his present work to avail himself of the results of Comparative Philology, and to bring them to bear on the natural history of man. But, instead of following Dr. Prichard's excellent work,-'Researches into the Physical History of Man,' which is by no means antiquated, Dr. Latham has adopted a division of languages which seems to be entirely his own. He divides all the languages of the world into four classes, which he calls aptotic, agglutinate, amalgamate, and anaptotic. He admits, however, of only three methods of grammar-the Classical, English, and Chinese. All the languages, dead or living, are referred to one of these classes with astonishing rapidity. There remains but one family of languages, which Dr. Latham considers hypothetical, the Arian Indo-Germans.' Sanskrit is to him a very doubtful language, still more its modern descendants, Hindi, Bengali, Mahratti, &c. According to him, 'the nation that is at one and the same time Asiatic and Indo-Germanic remains to be discovered.' This prejudice against Sanskrit is not peculiar to Dr. Latham. It is, or at all events it was, shared by many who found it troublesome to learn this new language. Sanskrit was called a factitious idiom, concocted by the Brahmins after the expedition of Alexander into India; a theory which Schlegel considers 'as happy as that which would account for 'the Egyptian pyramids as natural crystallisations.' There is another point, however, where Dr. Latham seems to have a fair claim on originality. We must quote his own words, because we might be suspected of misrepresenting his opinions. The criticism, or, rather,

6

1851. Ethnological Results of Comparative Philology. 13

guages, the easier, of course it is, to classify them, and to apply to them scientific names in an unscientific manner. But Comparative Philology is not a study for amateurs, and has nothing in common with the premature attempts of those precocious systematisers, who see no difficulty in bringing all the languages of the world under certain categories, although they would find it difficult to translate but one sentence from the idioms, which they have so hastily cast into the crucible.

It is, however, not so much an interest in language itself, which has given rise to these systems, as the pressing importance of other questions, which are more or less dependent on the results of Comparative Philology. Of these, the most remarkable are the problems relating to the early diffusion of nations in times not reached by history, and to the common origin of mankind. It is clear, that as soon as all languages, spoken by man, can be traced back to one common source, it will be in vain to maintain any longer that the physical varieties of man necessitate the admission of an independent origin for each race. It was natural, therefore, that the advocates both of the monogenetic and polygenetic theory should have tried

'scepticism,' he says, which has been extended by others to the Indo-Gangetic languages of Hindostan, is extended by the present 'writer to the Persian.' He afterwards maintains, that the language ' of the arrow-headed inscriptions is Sanskrit.' Colonel Rawlinson, Burnouf, and Lassen, might have saved themselves their trouble if they had been informed of this before. But Dr. Latham has allowed himself to be misled into a still greater mistake. Colonel Rawlinson, Burnouf, and Lassen have shown, that the Persian branch of the Indo-European stock has preserved, particularly in its oldest literary document, the Zend avesta, ancient forms, which occur in the Veda, but have been modified in the more modern Sanskrit. Dr. Latham, not knowing that the language of the cuneiform inscriptions differs from that of the Veda nearly as much as that of Cicero from Homer, has misunderstood this grammatical observation, and imagines that the language of Darius approaches so much to the Vedic dialect, as to prove that the Veda cannot be older than Darius. The premises are wrong, but still more the conclusion. For if we applied this principle to other facts of Comparative Philology, we might say, because the Lithuanian, as spoken at the present day, approaches so much to the Sanskrit as to possess in its declensions Sanskrit terminations, which have been modified in the other IndoEuropean idioms; therefore Sanskrit may not be much older than the Lithuanian, which any traveller may still hear spoken in parts of Prussia. But there is a Nemesis in every thing; and in the only instance where Dr. Latham attempts to give an authentic specimen of cuneiform writing every letter stands TOPSY-TURVY.

to support their views by evidence, derived from a comparative study of languages; and it was natural also, that, in their impatience to generalise, they should have anticipated results, where all was still conjectural, or taken for granted whatever in different works on Comparative Philology seemed best to suit their own theories. These great questions, however, must wait for their final solution, until the principal languages shall have all passed through the ordeal of Comparative Grammar. The results of Comparative Philology have always been progressive. Different scholars beginning with a diligent study of the organism of one or two languages, have gone on successively to compare them together, and to point out their essential differences or their radical resemblance. Families of languages have thus been established, and the members belonging to one or the other have continually increased in number. The families which have been traced out in this manner, comprise already most of the principal nations of the world; and an admirable statement of the results, arrived at by the combined labours of English and Continental scholars, may be seen in a paper of the late Dr. Prichard' On the various Methods of Research which contribute to the Advancement of Ethnology, and on the. Relations of that Science to other Branches of Knowledge.' This article, which is incorporated in the Report of the British Association for 1848, is the last word which its lamented author has left on the classification of languages and the varieties of man; and it is remarkable, not only for the vastness and accuracy of its learning, but also for that noble spirit of truthfulness and fairness which pervades all the works of Dr. Prichard. After having enumerated the different languages, belonging to what he calls the Indo-European (Arian), the Ugro-Tatarian* (Turanian), the Chinese, and Syro-Arabian (Semitic) families, he candidly admits that in several cases the inter-connexion rests on unsatisfactory grounds. This applies in particular to the languages of Africa and America, and to several branches of the Ugro-Tatarian family, under which the Chinese and the Indo-Chinese are ranged by some authors. In fact, as before stated, wherever comparative grammar is least advanced, we find the most vague and changeful ethnological con

6

* It is high time that the false spelling of Tartar, instead of Tatar, which is found not only in Dr. Latham's Varieties of Man,' but even in Dr. Prichard's works, should be given up. The Tatars have nothing to do with Tartarus and the Titans, but they are called Tata (and Tatan) from a Turanian root, which means to stretch, to draw the bow, to pitch tents, ri being the plural termination in the Tungusian languages.

1851.

6

Common Origin of Languages.

15

clusions, while languages, whose connexion rests on the firm basis of a grammatical comparison, place the ethnological relation of the nations, by whom they were spoken, beyond all contradiction. Although great progress has been made in a comparative analysis of the Arian or Indo-European, the Polynesian, the Semitic, and some branches of the Ugro-Tatarian family, yet the time for approaching the great problem of the common origin of languages is not yet come. No conclusions on this subject can be drawn from casual points of coincidence or difference, between single members of each family, but only from an inter-comparison of all families, and it will require the labour of centuries before this 'comparison of comparative grammars' can be carried out. In the mean time, it is worthy of remark, that the scholars, who are best competent to give an opinion as to the final results of Comparative Philology, believe, that all researches are tending more and more to the establishment of the common origin of language. Dr. Prichard concludes his paper, alluded to before, with the following words: I may venture to remark that with the increase of knowledge in every direction, we find continually less and less reason for believing that the diversified races of 'men are separated from each other by insurmountable barriers.' He remarks, that the same is the ultimate conviction of the great author of Kosmos;' and in the course of his paper he points out himself, that some of the barriers, by which families of languages seemed to be insurmountably separated, are already beginning to give way. In connexion with this subject, he alludes to a work by Professor Keyser of Christiania, in which its author endeavours to prove the wide extension of the Iberian people through Western Europe in remote times, and connects them with the Lapponic aborigines of Scandinavia. Nay, Dr. Prichard maintains further, that there are phenomena both in language and history which tend to favour the conjecture, that the Celtic nations (whose connexion with the Indo-European family he was himself the first to point out) were in part of Finnish or Lappish descent, and sprang from a mixture of this race with a tribe of Indo-European origin. And lastly, he refers to Egypt, where, as he says, it was reserved to a distin'guished scholar of the present day (the Chevalier Bunsen), to erect the edifice of the most ancient history of the world, a monument of the intelligence of modern Europe more exalted than the royal pomp of the pyramids.' Now Chevalier Bunsen's great discovery, stated in his own words, is, that the Egyptian, and perhaps the African man in general, is a scion of the Asiatic stock, which gradually degenerated into the 'African type. The Egyptian language attests a unity of blood