he wished the fruit to be given him? If language was of human invention, we think this is the course which would be followed in the particular instance supposed, and in all others of an analogous description; and hence it is evident, that, in the order of nature, nouns or names of sensible objects necessarily take precedence of all other kinds of words, and constitute the basis of language.

But Dr Murray thinks very differently. Forgetting that every action presupposes an agent and an object, he assumes that action may be conceived independently of either the one or the other; and consequently, that general or abstract terms would take precedence of particular terms, or names, in the formation of language. The first step, according to him, would be the invention of a species of nominalism, from which, by some inexplicable process, particular appellatives would be afterwards derived. But the very examples he has himself giventhe tree grows, the fire burns, the stone hurts, the plant poi'sons,'-are sufficient to demonstrate the fallacy of the principle on which his system is grounded: for the verbs, grows, burns, hurts, poisons, would evidently convey no determinate meaning without the appropriate substantives, tree, fire, stone, plant, and could never have been formed at all, until these physical objects had been distinguished by names, and their most obvious qualities ascertained. Growth, burning, hurting, poisoning, could never possibly be predicated, or even imagined, until it was previously known that things or objects existed, which grew, burned, hurt, and poisoned.

Our conviction therefore is, that if language was invented by man, the theory of its formation proposed by Dr Smith,* is not only clear and satisfactory in itself, but, we may add, the only one which proceeds on strictly philosophical principles, and accords with the order and sequence of human ideas. On the question, however, as to the divine or human origin of language, we mean to offer no opinion at present, because, besides being incapable of determination by mere reason, we conceive it wholly immaterial to the enquiry in which we are at present engaged: our object is, not to indulge in vague speculation on a subject which does not admit of even an approximation to cer

See Considerations concerning the First Formation of Languages, and the Different Genius of Original and Compounded Language, subjoined to the Theory of Moral Sentiments; a tract displaying much depth and ingenuity in the purely theoretical part, relative to the formation of languages, but exceedingly defective in that portion of it in which the illustrious author endeavours to determine the characteristics of original, as contradistinguished from those of derivative languages.

I

t

f

ainty, but to examine and analyze the various theories which ave at different times been proposed to account for the original ormation and the actual affinities and relations of languages. In assing, however, we may observe, that Dr Smith's theory is vidently applicable to the language of one people only; for, as Colonel Kennedy judiciously remarks, 'it attempts not to explain the causes which have occasioned that variety of names which are given by different nations to one and the same object;' and consequently, it leaves untouched and unaccounted for one of the nost remarkable circumstances in the history of mankind. On his subject no profane writer has been able to give us any inormation; and hence, no alternative remains but to adopt implicitly the account contained in the Mosaic history, which ascribes to a miraculous interposition of the Supreme Being, a phenomenon of which human ingenuity has utterly failed to supply any adequate explanation.

And, most certainly, if mankind ever spoke only one language, as we have no reason to doubt they did, such a miracle seems alone sufficient to account for its utter extinction, and the creation of a vast multitude of tongues, in none of which can the smallest trace of its existence be discovered for there is no authentic instance in profane history of a tongue which once existed being irretrievably lost. Language, which seems so evanescent and perishable, is nevertheless the most durable of human monuments. The nation who spoke it may be destroyed or cease to be; its history may be effaced from all human registers, and even its very name may be forgotten; but still some fragments of its language may outlast all this ruin, and evident traces of its existence be detected in other and newer dialects, which have been enriched with its spoils, and have perhaps acquired variety, compass, and force, from adopting its roots and inflections. Human speech, considered in relation to the communities of mankind, may be compared to the monumental ivy, which clings to the edifice to which it has been attached, with greater tenacity in its ruin than in its splendour, and when the last fragment has crumbled into dust, and its natural support is thus destroyed, creeps along until it encounter some humble dike or wall, into which it immediately strikes its roots, and thus prolongs its existence after the very ruins of the original edifice to which it adhered have utterly perished.

II. So much, then, for the origin and formation of languages. We now proceed to consider some of their more remarkable characteristics and affinities; beginning with Sanscrit, which, from a variety of causes, and, particularly, from the striking resemblance it bears, both in its words and grammatical struc

ture, to the Greek, is eminently deserving a prominent share of our attention.

India, formerly the home and birth-place' of all sorts of prodigies, contains nothing at the present day half so marvellous, or calculated to strike an enlightened enquirer with so much surprise and admiration, as the sacred idiom, to which the guardianship of all its treasures, of religion, science, and literature has been, in a great measure, confided. The Sanscrit, whether it be considered with reference to the remarkable peculiarities of its structure, the unrivalled perfection of its grammar, or its more general and distinguishing qualities as an organ or vehicle for the expression of thought, and the communication of knowledge, is now admitted, upon all hands, to be one of the most finished languages extant, or that ever existed among mankind, in any age or period of the world. This opinion was long ago expressed, in the strongest manner, by Sir William Jones. The 'Sanscrit language,' he observes, whatever be its antiquity, is of a wonderful structure; more perfect than the Greek, more 'copious than the Latin, and more exquisitely refined than either, yet bearing to both of them a stronger affinity, both in the roots of verbs, and in the forms of grammar, than could 'possibly have been produced by accident; so strong, indeed, that no philosopher could examine them all three, without be lieving them to have sprung from some common source, which no longer exists.' *

In fact, it is the wonderful structure' here spoken of, which pre-eminently distinguishes Sanscrit from other languages, and impresses upon it a character of originality, which cannot be mistaken; it is this circumstance chiefly which convinces us that it is not, as some have maintained, a derivative form of the Greek, or of any other tongue extant. Secondary languages are, in almost every case, less complex and involved in their grammatical structure, than those from which they have sprung; and, in most instances, the progress has been from languages rich in all the varieties of inflection to dialects of simple formation.+ Compare the Romaic with the Greek, the Italian with the Latin, the Teutonic with the Sanscrit, and the truth of this observation will be immediately apparent.

* Works of Sir William Jones, vol i. p. 26.

Dr Smith, with his usual discrimination, observes, in relation to this subject, that, in general, it may be laid down for a maxim, that the more 'simple any language is in its composition, the more complex it must be in its declensions and conjugations; and, on the contrary, the more simple it is in its declensions and conjugations, the more complex it must be ' in its composition,'-Formation of Languages, ubi supra.

C

pr

.0

G

31

is

P

C

0

0

E

Latin, indeed, which is generally supposed to be the oldest existing dialect of the Greek, may, at first view, be thought an exception to the rule here laid down; inasmuch as it retains inflections, and resembles, in the analogy of its general structure, the language of which it is believed to be a derivative. But we may be permitted to observe, that it is just as probable that Greek and Latin have sprung from a common source, as that the one of these languages is the parent of the other; that the most ancient specimens of Latin extant exhibit scarcely any inflections; that the subsequent introduction and use of these arose out of peculiar circumstances, which in no degree affect the truth of the general principle we have stated; and, at all events, that the Latin, even in its most improved form, is a language greatly inferior to the Greek, in variety, versatility, rhythm, compass, force, and dignity. The supposed exception, in the case of Latin, therefore, is more apparent than real, and, at the very utmost, is only an exception in degree, not in kind: For, to make out an analogy at all applicable to the Sanscrit, viewed as a derivative of the Greek, or of any other tongue, an instance must be produced of a secondary form of speech, which has not only retained all, or the greater part, of the inflections of the original language, but augmented greatly their number, complexity, and variety, and, at the same time, subjected the whole to the most perfect system of grammar and prosody which it is possible to conceive. This, however, cannot be done; there is no other example of a derivative language attaining to a higher degree of perfection than that from which it sprung, or adopting, extending, improving, and systematizing those very peculiarities which, in every other case, have been, in a great measure, lost, in the course of transference or descent; and hence we are disposed to consider the anomaly in regard to Sanscrit, as just one more proof of the originality of that remarkable language.

Nor are there wanting other considerations which seem to lead irresistibly to the same conclusion. Sanscrit contains no exotic terms, and throughout is perfectly homogeneous. Its roots, which are evidently the work of grammarians, afford a strong proof of the diligence with which it has been subjected to grammatical rules. They are merely monosyllables, consisting of the radical letters composing the words derived from them; they have in themselves no distinct meaning; and, consequently, they must have been formed long after the origin of the language."

* Sanscrit has one radical quality which it shares in common with the Greek, and other languages of the West, and which distinguishes it from most other Oriental tongues. For example, in the words, cañas, oiñas, oi

Had they been coeval with it, they would, beyond all question, have been significative; and the fact of their not being so appears to us a demonstrative proof that these roots were the work of grammarians, subsequently to the formation of the language. But although the roots do not form a constituent part of the language, still its words show that they must have been all formed by the people who once spoke it, (for that it was, at one period, spoken, seems, on many grounds, very probable,) according to some principle well known to them, although we may not now be able to discover it. What can be more evident, for instance, than that bhara, that which supports; bharata, a servant; bharanium, wages; bharaniu, a master; bharatha, a king; bharta, a busband; bharia, a wife; bhara, a burden; bhrita, hired; bhriti, wages; bhritya, a servant; are all cognates of the verb bibharti or bharati, he supports, maintains, or bears? But still no one, except a grammarian, would have thought of deriving all these words from the monosyllable bhri, which is given as the root. Again, the suffixes (termed kridantas) and affixes, employed in the formation of derivatives, appear to be nothing more than a classification by grammarians of letters and syllables of actually existing words, which could not conveniently be included among the roots; another proof that the present perfection of the grammatical system of Sanscrit has been the result of long consideration of the subject, and of the most anxious and painful elaboration; without which, indeed, it could never have been rendered of any practical use. For what hopes,' Mr Foster asks, would the unremitted toil of a protracted life, even of one endowed with the intuitive genius, the all-embracing faculties, of a Sir William Jones, hold out of attaining such an incredible language, was not every step directed by ety'mological rules, at once general, simple, and comprehen

Ais, oudas, in Greek ; malus, miles, moles, in Latin; tan, ten, tin, tone, tun, tune, in English; pati, pita, puta, pota, in Sanscrit ; it is manifest that the signification is determined by the vowels alone, the consonants being the same in each set of instances, λs, in the Greek, mls, in the Latin, în in the English, and pt in the Sanscrit. But the very opposite of this obtains in Hebrew, Arabic, and other Oriental tongues, in which the root consists exclusively of consonants, and never undergoes the slightest change of meaning in what way soever the vowel may be interposed. Thus, in the Arabic words, harama, hirman, muharram, hiramat, muhtarim, the meaning depends entirely, not on the vowels, but on the radical consonants hrm; these, and a variety of other derivatives, being merely modifications of the sense of the radical word. See Researches, &c. p. 91.

*Researches, &c. p. 194.