confounded the artificial classification and arbitrary divisions of language, as established by grammarians, with the natural and essential parts of language itself.

Language, so far as regards the philologist, may be arranged under four leading classes or divisions, one or the other of which will include everything he can possibly have to say on the subject.

1. Etymology, or that branch of language which treats of the derivation of words. Had there been but one language in the world, this branch of it would have had no existence; as to have recourse to etymology is to seek for the signification of a word in another, and if possible its original language, the use of which is altogether arbitrary in our own. The Greeks and Romans, who were perhaps the most finished and faultless writers the world has ever seen, were also the very worst etymologists, simply from the circumstance of their knowing no language but their own,-a disqualification which ought to have prevented them from entering on the task in limine. Their etymologies are sometimes utterly ridiculous, generally quite unfounded, and never satisfactory.

2. Prosody, which treats of the quantity of syllables, their combinations into feet, and the arrangement of the feet in lines, so as to form regular metres. What is usually understood by the term prosody, as a body or collection of laws, by means of which the length or shortness of every syllable in the language is fixed and determined, is almost peculiar to the Greek and Latin, and has nothing analogous in the languages of modern Europe, in which accented and unaccented syllables are very conspicuous, but in which long and short, in the sense in which those words were used by the Greeks and Romans, can hardly be said to have any existence. Prosody, therefore, in this sense, is not an essential but extraneous part of language, not a necessary but a luxury, and therefore does not naturally form an ingredient in philosophical grammar.

3. Inflexion, which treats of the various methods of modifying the meaning of that class of words denominated by

grammarians declinable, which are the Article, Noun, Pronoun, Adjective, Verb, and Participle. Inflexion may take place, either in the beginning, the middle, or the end of a word, or, in other words, may be accomplished by prefixes or augments as in the Greek verbs, by infixes as in the Egyptian nouns, or by affixes or terminations as in almost the whole class of declinable Latin words. Almost all these

changes are effected in the languages of modern Europe by means of Particles, and therefore cannot be regarded as essential parts of philosophical or general grammar.

4. Syntax, which is conversant with the laws which determine the formation of sentences, and their arrangement in periods, so as to constitute discourse or composition. Syntax decides on the position or collocation in the sentence of words of every class, and consequently of the indeclinable Prepositions, Adverbs, Conjunctions, and Interjections, as well as of the declinable Articles, Nouns, Pronouns, Adjectives, Verbs, and Participles. In the Hebrew, the Arabic, and most of the languages of modern Europe, the Syntax occupies but a few pages, in the Greek and Latin a much more considerable space; but it will not require a very copious notice in a treatise on general grammar.

VIII. If we take a comprehensive survey of the grammatical contrivances of some of the oldest written languages, and retrench all those which any particular language is destitute of, we shall have a clearer idea of what really are the essential ingredients, or component parts of language in general, which is the whole that philosophical grammar proposes to itself to give an account of. For instance, the Greek Verb has two Futures and two Aorists in the Indicative Mode; the Sanskrit two Futures and three Preterites, and the Egyptian Verb three Futures; but the Hebrew, the Arabic, and the whole class of Shemitic languages have only two tenses in the Indicative, the Preter and the Future. these, therefore, are all that are absolutely essential, it follows, as a matter of course, that all the rest in every other language are redundant, not indeed with reference to that par

As

ticular language, but so far as the actual wants of the species are concerned.

IX. Again, as to Moods, in the Greek we have the Indicative, the Imperative, the Optative, the Subjunctive, and the Infinitive. For the Optative and Subjunctive, the Sanskrit substitutes a Potential, a Precative, and a Conditional. The Latin rejected the Greek Optative, or rather did not distinguish it from the Subjunctive, while it retained the name of the Potential like the Sanskrit, without however giving it a distinct form from the Subjunctive. The Hebrew has but three Moods, the Indicative, the Imperative, and the Infinitive; and as all the purposes of language have been actually answered by these three, it follows that all the others in every language might have been dispensed with.

x. Once more as to Voices, in the Greek we find three, the Active, Passive, and Middle; the Sanskrit has but two, the Active and Passive, in which it agrees with the Latin; while the Egyptian cannot be said to have any form of the Active or Passive, as there is no mode of discovering the meaning, except by the context; and precisely the same word is used, sometimes in an Active and sometimes in a Passive sense. The distinction of Active and Passive Voices, therefore, though a beauty, a convenience, and an accuracy in language, cannot be said to be a necessary, as the Egyptians, a great, a polished, and a learned people, did without it. (Scholtz's Grammar, p. 74.)

XI. With respect to Numbers, the Greek, the Sanskrit, and the Slavonic have a Singular, a Dual, and a Plural. The Egyptian rejects the Dual, in which it is followed by the Latin; while the Arabic retains it, and the Hebrew has a Dual form for nouns, but none for verbs. As the Dual form is dispensed with by the greater number of languages, it cannot be regarded as an essential of language in general.

XII. As regards Cases, the Greek has five, the Nominative, the Genitive, the Dative, the Accusative, and the Vocative, to which the Latin joins the Ablative. The Sanskrit has

eight Cases, adding the Implementive and Locative to the Latin. The Slavonic follows the Sanskrit in its Implementive and Locative cases, but rejects the Ablative. With the exception of the German, most of the other languages of modern Europe cannot be said, strictly speaking, to have any cases of Substantives, as their different relations are denoted by Particles.

XIII. One of the most perplexing circumstances to the learner of a new language, is the distinction of genders, which in things without life, and in the very large class of words, significant of intellectual and moral ideas and relations, cannot be said to have any foundation whatever in the order of nature, or to be applicable to them at all without a violent metaphor, with which we should be more shocked if we were not accustomed to it by the ordinary use of language. Gender, however, is an adscititious and accidental and by no means an inherent and essential quality of speech, as two of the most beautiful and cultivated languages in the world do without it, the Persic and the English; the former so remarkable for its softness and harmony that it may be denominated the Italian of Asia, and might induce one to hazard a prediction that, if ever the march of intellect should again extend to that vast continent (where, alas! since the decline of the glories of Bagdad under the Arabian Caliphs, in the beginning of the ninth century, there has been no march except in the wrong direction), Saadi, Hafiz, and Firdousi will become, to the blue stockings of Asia, what Petrarch, Guarini, and Tasso have long been to those of Europe; and the latter so conspicuous for its copiousness, clearness, versatility, and vigour as to have proved itself fully equal to the development of some of the grandest efforts of the human mind the philosophy of Bacon, the poetry of Shakspeare, and the oratory of Burke, three names which it would be difficult to parallel in the annals of any age or country.

XIV. With respect to what are called the parts of speech, the terms of Noun Substantive and Noun Adjective, employed by the Greek and Latin grammarians, prove that they re

garded them as essentially the same word; and it is a wellascertained fact, that many of the Indian tribes of North America have no Adjectives in their language. In the Greek and Latin it is extremely difficult, in many instances, to distinguish the Article from the Pronoun, "as each seems either." The Participle is so clearly a mere modification of the Verb, as hardly to deserve the name of a distinct part of speech more than either of the tenses; and as to the Particles, or indeclinable parts of speech, it may be easily proved that they differ from the declinable solely in use, and not at all by nature. According to Plutarch, Plato was accustomed to assert, that all language consists of Nouns and Verbs; and I am disposed to go one step further, and state my opinion, that there must have been a period in human society when the language of mankind consisted solely of Nouns Substantive, and those the names of external objects.

xv. Dismissing every sort of hypothesis on the subject, the simplest and least artificial specimens of speech we can be said to be acquainted with, are the Sanskrit Dhatos, or Verbal Roots, and the great body of words composing the Chinese spoken language, which, we are informed by grammarians, are destitute of case, gender, and number, and may therefore, with propriety, be denominated Abstract Words.

XVI. Respecting the former, the literal meaning of the word Dhatu, as explained in Wilson's Sanskrit Dictionary, is a primary or elementary substance, earth, water, fire, air, and Akas, or atmosphere, and, as applied to language in a secondary or metaphorical sense, he says, "Dhatu, a grammatical root; in Sanskrit this radical performs no other office, and cannot be used as a word without undergoing some change;" and Wilkins, in the introduction to his Sanskrit Radicals, expresses himself on the subject as follows:-"The interpretations of the roots are given in Sanskrit by nouns put in the locative or seventh case, and in English by the second person of the Imperative, which must be considered as the root of verbs and verbal nouns in our language; but the student must not conclude from this, that these radical