syllables are either nouns or verbs till they are inflected. The significations attributed to a large proportion of them will be found exceedingly vague and unsatisfactory, as if nothing more were intended than to convey some general idea of the meaning of each root. In the English explanations occasional deviations will be found from the Sanskrit. These for the most part have been made upon the authority of commentators."

XVII. As regards the Chinese spoken words, we have the following account of them in the articleChina,' in the Supplement to the Encyclopedia Britannica: "The colloquial language is not less singular than the symbolical characters; being, like the latter, exclusively their own, having borrowed nothing from, nor lent any thing to the rest of the world. The 330 monosyllables, each beginning generally with a consonant, and ending with a vowel, or liquid, or the double consonant ng, which, as we have observed, complete the catalogue of words in their language, are, by means of four modifications of sound or intonation to each syllable, extended to about 1300, beyond which not one of them is capable of the least degree of inflection, or change of termination; and the same unchangeable monosyllable acts the part of a Noun Substantive and Adjective, a Verb, and a Participle according to its collocation in a sentence, or the monosyllables with which it is connected. It is neither affected by number, case, nor gender; mood, tense, nor person; all of which in speaking are designated by certain affixes or prefixes, to mark the sense."

XVIII. Having traced language by a process of analysis to the rudest state in which we can discover it actually existing, or, at any rate, of which we have any clear and authentic account, a task comparatively easy, I must now proceed in another direction, and endeavour, by synthesis, to build up the vast fabric of language, tracing, if possible, the connecting links, the gradual changes,—the fine and almost insensible shades which unite the abstract words of the Chinese with

X

the Orations of Demosthenes and the Poems of Virgil, perhaps the most perfect models of human compositions.

Some such attempt as this appears to have engaged the mind of Dugald Stewart when he wrote the following passage:"I have mentioned but a few of the innumerable topics which crowd upon me as fit objects of inquiry for the rising generation; nor have I been guided in my selection of these by any other consideration than their peculiar adaptation to the actual circumstances of the philosophical world. Among these the most prominent is the Natural or Theoretical History of Language, including under this title written as well as oral language,—a subject which will probably continue to furnish new problems to human ingenuity, in the most improved state of human knowledge. It is not surprising that an art which lays the foundation of all the others, and which is so intimately connected with the exercise of reason itself, should leave behind it such faint and obscure traces of its origin and infancy." (First Dissertation, Enc. Brit. vol. v. p. 199.)

1. LOCKE has remarked in the third Book of his Essay, that the formation of language supposes two indispensable conditions:

1. A consciousness in man of his power to produce articulate sounds.

2. A perception of the possibility of those sounds becoming the signs of his ideas.

There is every reason to believe that a person born perfectly deaf, or secluded altogether from human converse, would possess neither the consciousness, nor the perception, as we find from experience, that dumbness is the invariable associate with deafness from infancy, and that gestures and motions are the imperfect substitutes for words. Though a healthy infant, therefore, begins to use its tongue very early, and clearly expects that the inarticulate sounds he utters will be understood by those about him, there are strong grounds for believing that the whole is to be referred to imitation. When his powers of perception and memory are sufficiently developed to enable his ear to discriminate, and his mind to retain the sounds which are addressed to him most frequently, words are gradually substituted for those inarticulate cries he had been in the habit of uttering, which appear to be the only language taught by Nature. She supplies the elements of speech; but without the assistance of her sister Art the native powers of the human mind seem to be altogether inadequate either to effect the combination those elements are intended to form, articulate sounds, or make any considerable advance towards the end that combination was destined to accomplish, language.

II. He has also observed, that communication by words has a double use.

1. Civil, or such an interchange of thoughts and ideas as may be rendered subservient to the ordinary business and pursuits of life, in the societies of men, one amongst another.

2. Philosophical, or such a use of words as may serve to convey the precise notions of things, and to express, in general propositions, certain and undoubted truths, which the mind may rest upon, and be satisfied with, in its search after true knowledge.

Whatever may be the possible attainments of the human mind, in some future and greatly improved state of society, I must confess that I can form no conception, either of language or of the use of language, essentially different from what we now experience them to be. That philology will partake of the benefit of the general advance of knowledge, I see no reason to doubt; but, on the contrary, feel confident that the inductive reasoning of Bacon is as applicable to that as to any other branch of human pursuit, and that our more accurate acquaintance with the various languages of mankind must be beneficially felt as regards language in general; but, in my most sanguine expectations of the possible improvement of language, I must persist in regarding any future change to be produced as one of degree, and not of kind. With the progress of philology the labours of the etymologist will become at once more comprehensive and more certain; but when every word in every language has been traced to its true and undoubted source, should such a period ever arrive, we shall still be very far from having got rid of all ambiguity in the use of language. Etymology, by ascertaining the origin, may undoubtedly do much towards determining the meaning, and fixing the use of words; but it must never be forgotten that etymology, in its most high and palmy state, can give us no certain information, except with regard to the past, little as to the present, and absolutely nothing as to the future; that it can only tell us what a word was, not what it actually is, which can be learnt in no other way than by a diligent observation of the various

modes and different connexions in which it is employed by the best speakers and writers. As the meaning of words can only be acquired by a reference to their actual use, and as the latter varies greatly in every country, not only from century to century, but from generation to generation, it appears to me that the efforts of the philologist, when directed towards the attainment of an immutable signification and use of words, are quite as chimerical and hopeless as the exertions of the political economist to devise an invariable standard of value, or the long exploded phantoms of the transmutation. of metals, the elixir vitæ, or the perpetual motion.

111. The philosophy of language does not pretend to account for the reason why particular ideas, or combinations of ideas, were expressed by particular sounds. If there had been any natural and necessary connexion between things and words, between ideas and the sounds which are the signs of them, there could have been but one language in the world. But there is a well-known volume which contains the Lord's Prayer in one hundred and fifty different languages, and in a general way one may be said to be as good as another, as the leading purposes of speech are answered by all. The connexion between words and things, therefore, is not natural, but entirely arbitrary, and any idea may be represented by any sound, when the two have been firmly connected by the mind in the way of association. But while, in this view, language appears to be the most flexible, in another it is the most intractable of all conceivable subjects. Augustus remarked very truly, that though he was the undisputed master of the Roman world, it was not in his power to introduce a new Latin word into general use, and Claudius was equally unable to give permanence to three new forms of letters, which we find in the inscriptions of his reign.

IV. We can derive little or no assistance from the Scriptures of the Old Testament, and gain nothing of moment by reverting to the origin of the human race as described by them, as, so far as the subject of language is concerned, all access to the tree of knowledge is closed. Although the