6

the person, and as actual or contingent, and the time of the action. The personal terminations in the verb are generally recognized as abbreviated fragments of the personal pronouns."

6

This

is clearly the case in the Semitic languages; and Mr. Conybeare suggests, that it will probably be found to be the case in every class of languages, if the different dialects be collated with the requisite attention.

If we look at the Greek, &c., we shall see good reason to derive the first singular, μ, from μ; the second, o, from σ; the third, Ti, from Tos, Ta, To,—the ancient form of the pronoun, afterwards employed as the article, and thus incorporated into the digammatized form of αυτός. The first plural, es, seems to be from αμες, the ancient form of us the second and third are more obscure. We are at least sure that the Welsh second person plural is from the corresponding pronoun chwi. The root of this plural person in Sanscrit, and all the other Indo-European languages, appears to be Yu; hence, Yus, and the Latin and Slavonic Vos and Vas.' p. 88.

The following is a comparative view of the present tense of the verb substantive, in the principal dialects of the Indo-European family.

If we compare the other root employed in the verb substantive, the Latin fui, (from the Greek Quva,) with the corresponding Sanscrit bhavami and the Teutonic beon, we shall find the result equally satisfactory.' p. 89.

The tenses of the verb in Hebrew and Arabic, are only two, which are commonly represented as a preterite and a future. Mr. Conybeare, however, considers them as more properly

aorists.

For although, when simply used, the first has generally a past, and the second a future sense, yet, if both be connected by the conjunction, whichever is placed last, assumes the force of that which preceded. This is, by the common grammars, attributed to some mysterious power of the conjunction, by which it is supposed to convert the general sense of the tense, and is therefore called the conversive Vau; whereas, in truth, it does but subject (as a connective

particle) a tense in itself indeterminate to the general force of the context, and thus determine its exact acceptation. The present tense is often expressed by the first aorist, and often by the participle, the verb substantive being understood.' pp. 85, 6.

6

The best explanation of this singular rule in the Hebrew Syntax, Mr. C. suggests, is, to regard the future tense, or second aorist,' as implying simply a succession of time in the action.

In narrative passages, this tense merely denotes that the action was subsequent to the one with which the conjunction connects it, and thus it acquires from the context a preterite sense. Under other circumstances, notwithstanding the conjunction prefixed, it has necessarily a future sense: e. g. "Teach me the way of thy statutes, and I will keep it," &c. That is, thou teachest me;-afterwards I keep it.

Mr. Conybeare expresses his conviction, that the general principles which govern the grammar of any particular language cannot be rightly apprehended, except by instituting a careful and extensive comparison between the grammars of different languages. In this conviction of the valuable results that may be anticipated from the study of comparative philology, we entirely accord; and we cannot cease to regret the loss which philological science has sustained in the early death of the late Mr. Greenfield, who had paid particular attention to the philosophy of grammar, and was distinguished as much by his faculty of analysis, as by his surprising facility of acquirement. His projected grammar in twenty languages would, there can be no doubt, have thrown important light on the structure and common principles of the compared languages, and have tended greatly to diminish the preliminary difficulties which embarrass the student in entering upon the study of a new grammar.

Whatever theory we adopt as to the origin of language, the laws which govern it would seem to be as fixed and regular as those which we call the laws of nature in any other class of physical phenomena; and the further we pursue our analysis, the more what appears, on a superficial survey, arbitrary, variable, and complicate, is discovered to be the result of simple, uniform, and intelligible principles.

The general conclusion to which Mr. Conybeare is anxious to conduct the student, by the exemplification of the affinities between the Semitic and other languages, is; that numerous dialects may be readily reduced to a comparatively few mother tongues, all the descendants of which still attest their common origin by the exact identity of their grammatical mechanism in the sys'tems of declension and conjugation, and by the close agreement ' of whole classes of words forming the great elementary materials ' of speech.'

!

There seems,' he remarks, no good reason to ascribe diversities of language to the original ramifications of the Noachian family; whether we ascribe that diversity to the dispersion of Babel, or, with many orthodox commentators, suppose the miracle then recorded to have consisted in a temporary confusion of mind, producing as its effect a corresponding confusion of expression, rather than in any miraculous change of the permanent dialects, and refer their subsequent diversities to the operation of gradual causes arising from long separation, distant emigrations, and new associations, constantly modifying the simplicity of earlier language. Whichever of these views we may adopt, there seems no authority whatever for attributing distinct tongues to the immediate descendants of Noah's first descendants, rather than to subsequent causes, which may have blended together in a course of common emigration, the members of different nations.'

p. 94.

Of the languages usually termed Semitic, some were certainly spoken by the descendants of Ham, the progenitor of many of the Arabic tribes. The Canaanitish nations and the Phenicians assuredly spoke dialects of the same mother speech. Mr. Conybeare thinks that we may recognise the following subdivisions of the Semitic family of languages :-1. Aramaic ; subdivided into two dialects, the Eastern Aramaic, improperly termed Chaldean, and the Western Aramaic, or Syrian. 2. The Punic, or dialect of the Canaanites and Phenicians. 3. The Hebrew. 4. The Arabic. 5. The Ethiopic. To these, the Coptic ought probably to be added, if not the Libyan or Berber. Of this great family of languages, the region lying between the Persian Gulf and the Mediterranean, and extending from Ararat northward to the mountains of Tigre on the southwest, appears to be the proper country. The Indo-Germanic or Indo-European languages form another family, which includes the dialects of the greater portion of Europe, besides those of Persia and India. The Finnic, the Biscayan, the dialects of Northern Asia, North America, and Mexico, the Chinese, the Polynesian, and the African dialects, form a third class of independent languages, the relations of which to each other, or to the other two great families, is extremely obscure, owing mainly to two causes; first, that they are all unwritten languages, (for the Chinese character is not phonetic, and does not express the language,) and secondly, because, not being written, they are for the most part purely conventional, not systematic in their formation. It is only,' as Mr. Conybeare remarks, when the structure of languages becomes more arti'ficially complicated, and a single root is preserved in multitudes ' of varying forms and compounds, that their affinities are so 'marked as to attract observation. Such languages are termed polysynthetic; and their relation to each other is confessedly to be traced far more clearly than that of the ruder monosyllabic tongues. Even between these, however, and the languages of

6

the two great families, coincidences may be traced, far too numerous and of a nature too marked to be the result of either accident or mere intercourse.* In a postscript' to the Appendix under consideration, the learned Author has attempted to determine, by an application of the mathematical doctrine of chances, the probable number of accidental coincidences which may be expected to occur, in a given number of languages, in any single term. We must confess that the problem seems to us more curious than useful. The conclusion to which it leads us is, that little dependence can be placed on general comparisons of all the known languages, considered in a mass, as an argument to support the opinion of the original unity of our species,' (we employ Mr. Conybeare's language, with a protest against the apparent concession to the infidel, implied in such a use of the term opinion,) since, thus considered, so large a number of 'coincidences may probably be accidental; but the true point to which we ought to direct our attention, is the comparison in 'detail of each definite pair.' Verbal resemblances, however, though indefinitely multiplied, are less satisfactory marks of original unity of language, than the prevailing uniformity of certain general principles of universal grammar.

6

All languages employ similar classes of general terms, such as pronouns. All languages connect these pronouns with terms indicating action, so as to produce verbs varying through persons and numbers. And this identity of the general principles of the mechanism of language is often far greater than can be accounted for by ascribing it only to an identity in the general metaphysical operations of the human mind. In the forty distinct mother tongues of America, for instance, though few marks of verbal coincidence can be traced, yet the elaborate mechanism which pervades the whole, and the methods by which they express very complicated relations, and various modifications of the original ideas, evince the most remarkable identity of design.'

p. 105. Of all the absurdities which Scepticism has framed into an objection against the truth of the Mosaic Scriptures, that which would infer from the existing diversity of language a diversity of species in mankind, is perhaps the most flimsy and futile. The original unity of the human race is established by an authority which no physiological investigations can affect, but which they have tended hitherto only to verify. We do not, however, think that unity of species could be proved by establishing a universal consanguinity of origin in all existing languages; since it is conceivable that one and the same language might have been spoken by races of mankind primarily distinct. Again, to repel the

* Some curious coincidences have been pointed out between the Algonquin and the Irish, and between the American dialects and the Hebrew.

:

infidel objection against the statement, that, prior to the dispersion at Babel, all the people of the earth had but one language and one speech, we are not required to establish a consanguinity between all existing languages; since, to render the objection valid, the sceptic is bound to shew that those languages were spoken at the remote period referred to, which are now adduced as monuments of an original diversity. We are well persuaded, indeed, that man could never have invented language, any more than he could have invented breathing or thinking; because the essence of language is the communication of ideas to another by means of significant sounds, implying a faculty for understanding speech, which can be explained only as an original law of our constitution. Yet, the vocabulary of different languages must be considered as falling within the province of human invention in this sense, new language is continually being invented; and it is highly reasonable to suppose that the existing diversities in the languages of the world, are of comparatively modern origin. In fact, we uniformly find diversities of dialect to be multiplied in proportion to the retrogression of a people towards barbarism. What can more clearly prove the modern origin of the exist ing diversity of language, and that it is the result of a deteriorated civilization, than the fact, that while Asia, the primeval seat of the human family, has been ascertained to contain not more than twenty-three parent tongues, which are spoken by seven-eighths of the population of the world, Africa is supposed to contain a hundred, or even a hundred and fifty languages, (Mr. Conybeare would reduce them, by conjecture, to twenty-five parent tongues,) spoken by a few scattered semi-barbarous nations ? The distinct parent tongues of the New Continent, respecting the population of which we have no certain records that ascend higher than the sixth century of the Christian era, have been reckoned to amount to upwards of a hundred*, although the indigenous races do not number more than between six or seven millions. In the neighbourhood of Caucasus and the Caspian Sea, a few barbarians, ignorant of the art of writing, are said to be divided into more nations, speaking peculiar languages radically different from each other, than the whole of civilized Europe. In languages destitute of inflections, and of the alphabetic mechanism, the composition of words is so arbitrary, that the widest scope is given for the most capricious varieties. Now it is certain that the aboriginal nations of Africa and America are not less closely related to each other, and exhibit not less affinity in all their manners and traditions, as well as in their physiological characteristics, than those of the great Semitic and Indo-European

* Humboldt says, some hundreds. In Mexico, upwards of twenty languages are spoken by less than three millions of Indians.