The observations, too, which Mr. Smith has made on the origin of verbs appear to me liable to strong objections. "Verbs," he says, "must necessarily have been. coëval with the very first attempts towards the formation of language;" and "probably," he adds, "existed first in an impersonal form."* But if all verbs were impersonal, how could a substantive noun be introduced into a sentence? or, indeed, what could have been the use of substantives? And yet, in the very first sentence of this dissertation, it is taken for granted that the invention of nouns substantive must have been the first step taken towards the formation of a language.†

It should seem, that the following may be laid down as a general rule with respect to the subject. Men were led to invent artificial signs from the defects of natural ones; and therefore it is probable that the first artificial signs would be employed to convey those ideas which it is most difficult to express by the language of nature. To judge by this rule, we must conclude, that substantives were prior to verbs; for an idea of individual objects would be conveyed with much greater difficulty than that of action or suffering in any particular mode. In confirmation of this, we may remark, that what we cali action in delivery is still chiefly connected with verbs; a proof that the notions they convey are more easily expressible by natural signs than the import of any of the other parts of speech.

language." The book, Monboddo informs us, is so extremely rare, that he could only hear of one copy of it in the Royal Library at Paris, for the use of which he expresses his obligation to the librarian, M. Capperonnier.—Origin and Progress of Language, Vol. I. pp. 471, 534.

Theory of Moral Sentiments, 6th edit. Vol. II. pp. 434, 435.

† A late very learned author has censured, with some severity, the whole of this Dissertation, and, in particular, has pronounced Mr. Smith's opinion concerning the origin of substantive nouns as antecedent to that of adjectives or names of qualities, to be altogether unsupported by facts in the history of language." The same author asserts, with some confidence, that "the first words were monosyllabic verbs,"" this," he says, "is discovered by analysis."-History of European Language, &c. by the late Alexander Murray, D. D. Vol. II. p. 489. The ingenious critic, in my opinion, would have been nearer the truth had he blamed Mr. Smith for not keeping his original and fundamenta! proposition more steadily in view in the sequel of his theory.

As for Dr. Murray's assertion, that " the first words were monosyllabic verbs," how is it to be reconciled with the fact, that, in most savage languages, the words are of so immoderate a length as to appear to our ears quite ludicrous?

Language, then, I apprehend, in its rudest state, would consist partly of natural and partly of artificial signs; substantives being denoted by the latter, and verbs by the former.

Mr. Smith says "a savage who saw a wolf or a bear approaching, would announce the event by using the word venit without a substantive." To me it appears much more probable, that he would exclaim Lupus or Ursus, without a verb.. Such an exclamation, accompanied with natural signs, would convey a complete idea of the event; and is, indeed, the very mode of expression which, on such an occasion, would probably be used, even in the present improved state of language; whereas the word venit, with whatever natural signs we may conceive it to be connected, could never convey any information concerning the particular animal whose approach was to be announced.

From these observations I am led to conclude, that as soon as verbs were introduced, they were used personally; excepting in those cases where a foundation is laid for the use of impersonal verbs in the nature of things—and, in such cases, those verbs which were once impersonal always continue so, under every progressive improvement of the art of speech. In most instances, it may be observed, there is a natural foundation for a separation of the agent and the action; because the same agent may act in an infinite variety of modes; or, in other words, the same substantive may be a nominative to an infinite variety of verbs. It is thus we say Petrus ambulat, Petrus sedet, Petrus dormit; these three verbs expressing three different states of the same person. In some cases, however, we see an event where the agent and action, and, consequently, the nominative and verb, are inseparably blended or combined together; and where, accordingly, we are naturally led to express ourselves by means of an impersonal verb. Of such cases the following examples may serve as a specimen, if they indeed do not comprehend all the varieties that

exist.

* Theory of Moral Sentiments, Vol. II. p. 437.

1. When the agent and action are always seen in a state of combination, or, in other words, when the agent only exists in that mode of action which the verb expresses. This is the case with rain, snow, wind, where the action is implied in the substantive nouns, and where, on the other hand, the substantive is implied or involved in the corresponding verbs. We do not, therefore, here, as in the former cases, make use of a mode of. speaking analogous to Petrus ambulat, Petrus sedet, but express the event in one word, pluit, flat, ningit.

2. When we mean to express an effect, without any reference to its cause; or to state a truth which is selfevident, or a fact which is universally admitted. Of this class are the following verbs,-tonat, turbatur, lucet, liquet, constat. In both of these cases, the origin of impersonal verbs may be easily deduced from the nature of the thing which the verb is employed to express.

With respect to a large proportion of impersonal verbs it may be remarked, that although they agree with those now mentioned in their form, they yet approach much nearer to personal verbs in the species of meaning they convey, and in the analogy of their construction. Such are the verbs pænitet, decet, oportet, which differ from other verbs only in this, that they have infinitives for nominatives; and hence the infinitive is called by some grammarians the noun of the verb. Now, with respect to all verbs of this description, it is evident, that their origin cannot be explained upon Mr. Smith's principle, (to wit, the difficulty of making a metaphysical separation between the subject and the action,) for a separation perfectly analogous takes place between the idea expressed by the infinitive, and that expressed by the impersonal verb.

In deciding upon the order in which the different parts of the verb were invented, a great deal must undoubtedly be left to conjecture; and of the various hypotheses that may be formed on the subject, there is perhaps none which, in point of probability, possesses such a decided advantage over the others, as to unite all suffrages in its favor. Mr. Smith thinks it natural to suppose, that verbs would first be made use of in the

third person singular. To this opinion he was led by his position, which I formerly animadverted upon, that all verbs were originally impersonal; and that they became personal by the division of the event into its metaphysical elements. In proof of this he observes, that, "in the ancient languages, whenever any verb is used impersonally, it is always in the third person singular. The termination of those verbs, which are still always impersonal, is constantly the same with that of the third person singular of personal verbs. The consideration of these circumstances, joined to the naturalness of the thing itself, may serve to convince us that verbs first became personal in what is now called the third person singular." *

For my own part, I am strongly inclined to agree with Leibnitz, the President de Brosses,† and Court de Gebelin, in thinking it probable, that the first of the tenses (or what grammarians call the root of the verb) was the imperative. The last of these writers, in particular, has supported this opinion by some considerations which appear to me equally ingenious and solid.† But on this very questionable point I must not enlarge.

It is somewhat remarkable, that, in this review of the origin of the parts of speech, no notice is taken of conjunctions.; the metaphysical nature of which furnishes as curious a subject of discussion, as that of any of the others. Some eminent grammarians (in order probably

Theory of Moral Sentiments, Vol. II. p. 441.

Traité de la Formation Mécanique des Langues, 1765.

"Avant qu'on pût penser à l'avenir ou qu'on cherchât à se rapeller le passé, il fallut pourvoir au moment présent: car comment se rapeller l'un ou rêver à l'autre, tandis qu'on eût été agité du plus pressant besoin, celui de pourvoir au moment ? Le premier soin des hommes fut donc de réunir leurs efforts pour se procurer ce qui leur étoit indispensable pour la vie; tel dut donc être le but de leurs premiers dis

cours.

"Les verbes commencèrent donc par l'impératif, par ce tems qui dit de la manière la plus courte et la plus promte, ce qu'on doit faire: car dans les choses pressées et où il faut exécuter sur le champ, on ne sauroit chercher de longs discours; et ce n'est pas dans le besoin qu'on s'amuse à haranguer.

"Aussi l'imperatif est-il comme les discours des muets; à peine est-il audessus du geste : il est comme lui isolé, décousu, l'affaire de l'instant, un simple son, comme l'autre est un simple mouvement; presque toujours composé d'une seule syllabe. Ama, aime; Lege, lis; Dic, dis; Fer, porte, sont plus courts qu'aucun autre tems de ces verbes."-Monde Primitif, &c. par M. Court de Gebelin, 1774, Vol. II. p. 240, et seq.

to elude the difficulty of explaining them) deny them to be parts of speech, and insist that they are only the mortar which cements the other parts of speech together; while others, in farther prosecution of the same idea, call them the nails and pegs of discourse. My own opinion is, that they were first explained in a satisfactory manner by Mr. Horne Tooke, in a letter addressed to Mr. Dunning in 1778;* the substance of which pamphlet he has since expanded into a large work, under the title of The Diversions of Purley.

The first conjunction to which Tooke turned his attention was the conjunction that, which he affirmed to be only a particular mode of using the article or pronoun of the same name, and consequently not to belong to a specifically different class of words. A few examples will sufficiently illustrate the scope of this theory.

Example.-"I wish you to believe THAT I would not wilfully hurt a fly."

Resolution.—" I would not wilfully hurt a fly, I wish you to believe THAT" (assertion.)

Example. "Thieves rise by night, THAT they may cut men's throats."

Resolution." Thieves may eut men's throats, (for) THAT (purpose) they rise by night."

"After the same manner may all sentences be resolved, where the supposed conjunction THAT (or its equivalent) is employed; and by such resolution it will always be discovered to have merely the same force and signification, and to be in fact nothing else but an article.

"And this is not the case in English alone, where THAT is the only conjunction of the same signification which we employ in this manner; but this same method of resolution takes place in those languages also, which have different conjunctions for this same purpose: for the original of my last example (where UT is employed, and not the Latin neuter article QUOD) will be resolved in the same manner.

UT jugulent homines, surgunt de nocte latrones.'

* A letter to John Dunning, Esq. by Mr. Horne. London, printed by J. Johnson, St. Paul's Church-Yard, 1778.