On this part of speech the improvements have not been so considerable (nor was there equal need) as on the conjunctions and the relatives. Yet even here the progress of taste hath not been entirely without effect. The until and unto are now almost always, and the upon very often, contracted into till and to, and on. The to and the for are in some cases, without occasioning any inconvenience, and with a sensible advantage in point of energy, discarded altogether. Thus we say, "Forgive us our debts," and not "forgive to us our debts"-"I have gotten you a license," and not "I have gotten a license for you." The same manner hath also obtained in some other modern tongues. What I am next to mention is peculiar to us the preposition of is frequently supplied by the possessive case of the noun. Lastly, which is a real acquisition in respect of vivacity, when two or more nouns are conjoined in the same construction, it is not necessary in English, as in French, that the preposition of the first be repeated before each of the subsequent nouns. This ought to be done only in those cases wherein either perspicuity or harmony requires it.

Now that I am on the subject of the prepositions, it will not be improper to consider a peculiarity which is often to be found with us in their arrangement. In every other language the preposition is almost constantly prefixed to the noun which it governs; in English it is sometimes placed not only after the noun, but at a considerable distance from it, as in the following example: "The infirmary was indeed never so full as on this day, which I was at some loss to account for, till, upon my going abroad, I observed that it was an easterly wind."* Here no fewer than seven words intervene between the relative which and the preposition for belonging to it. Besides, the preposition doth not here precede its regimen, but follows it. One would imagine, to consider the matter abstractly, that this could not fail in a language like ours, which admits so few inflections, to create obscurity. Yet this, in fact, is seldom or never the consequence. Indeed, the singularity of the idiom hath made some critics condemn it absolutely. That there is nothing analogous in any known tongue, ancient or modern, hath appeared to them a sufficient reason. I own it never appeared so to me.

If we examine the matter independently of custom, we person who loves her," or "The person who loves it." Nay, more, though there is a difference in writing between qui l'aime and qu'il aime, there is no difference in sound, and therefore the same phrase spoken may also mean "The person whom he loves." In Italian there are several periphrastic prepositions in the same taste with the French, as a l'intorno di, di la di, in mezzo di, dentro di, fuori di, di sopra di, di sotto di. There are only two prepDsitions in French which we are obliged to express by circumlocution These are. chez, at the house of, and selon, according to.

Spectator, No. 440, C.

shall find that the preposition is just as closely connected with the word, whether verb or noun, governing, as with the word, whether noun or pronoun, governed. It is always expressive of the relation which the one bears to the other, or of the action of the one upon the other. And as the cause in the order of Nature precedes the effect, the most proper situation for the preposition is immediately after the word governing, and before the word governed. This will accordingly, in all languages, be found the most common situation. But there are cases in all languages wherein it is even necessary that the word governing should come after the word governed. In such cases it is impossible that the preposition should be situated as above described. Only half of the description is then attainable, and the speaker is reduced to this alternative, either to make the preposition follow the word governing, in which case it must be detached from the word governed, or to make it precede the word governed, in which case it must be detached from the word governing. The choice, in itself arbitrary, custom hath determined in every tongue.

But will it be admitted as a maxim that the custom of one language, or even of ever so many, may be urged as a rule in another language, wherein no such custom hath ever obtained? An argument founded on so false a principle must certainly be inconclusive. With us, indeed, either arrangement is good; but I suspect that to make the preposition follow the word governing is more suitable than the other to the original idiom of the tongue, as in fact it prevails more in conversation. The most common case wherein there is scope for election is with the relatives whom and which, since these, as in the example quoted, must necessarily precede the governing verb or noun. But this is not the only case. Vivacity requires sometimes, as hath been shown above, that even the governed part, if it be that which chiefly fixes the attention of the speaker, should stand foremost in the sentence. Let the following serve as an example: "The man whom you were so anxious to discover, I have at length got information of." We have here, indeed, a considerable hyperbaton, as grammarians term it, there being no less than thirteen words interposed between the noun and the preposition. Yet whether the expression can be altered for the better, will perhaps be questioned. Shall we say, "Of the man whom you were so anxious to discover, I have at length got information?" Who sees not that by this small alteration, not only is the vivacity destroyed, but the expression is rendered stiff and formal, and therefore ill adapted to the style of conversation? Shall we, then, restore what is called the grammatical, because the most common order, and say, 'I have at length gotten information of the man whom you

were so anxious to discover?" The arrangement here is unexceptionable, but the expression is unanimated. There is in the first manner something that displays an ardour in the speaker to be the messenger of good news. Of this character there are no traces in the last; and in the second there is a cold and studied formality which would make it appear intolerable. So much is in the power merely of arrangement. Ought we, then, always to prefer this way of placing the preposition after the governing word? By no means. There are cases wherein this is preferable. There are cases wherein the other way is preferable. In general, the former suits better the familiar and easy style which copies the dialect of conversation; the latter more benefits the elaborate and solemn diction, which requires somewhat of dignity and pomp. But to what purpose, I pray, those criticisms which serve only to narrow our range, where there would be no danger of a trespass though we were indulged with more liberty? Is it that the genius of our language doth not sufficiently cramp us without these additional restraints? But it is the unhappiness of the generality of critics, that when two modes of expressing the same thing come under their consideration, of which one appears to them preferable, the other is condemned in gross, as what ought to be reprobated in every instance. A few contractions have been adopted by some writers which appear harsh and affected; and all contractions, without exception, must be rejected, though ever so easy and natural, and though evidently conducing to enliven the expression.* One order of the words in a particular example

*About the beginning of the present century, the tendency to contract our words, especially in the compound tenses of the verbs, was undoubtedly excessive. The worst of it was, that most of the contractions were effected by expunging the vowels, even where there was no hiatus, and by clashing together consonants of most obdurate sound, as Swift calls them. This produced the auimadversion of some of our ablest pens, Addison, Swift, Pope, and others, whose concurring sentiments have operated so strongly on the public, that contractions of every kind have ever since been in disgrace, even those of easy pronunciation, and which had been in use long before. Yet our accumulated auxiliaries seemed to require something of this kind. And though I am sensible that wasn't, didn't, shouldn't, and couldn't are intolerably bad, there are others of more pleasant sound, to which our critics, without any injury to the language, might have given a pass. On the contrary, even those elisions whereby the sound is improved, as when the succession of an initial to a final vowel is prevented (which in all languages men have a natural propensity to avoid by contracting), as I'm for I am; or when a feeble vowel is suppressed without harshness, as in the last syllable of the preterits of our regular verbs (which without a contraction we can never bear in verse), or when some of our rougher consonants are cut off after other consonants, as 'em for them (these, I say), have all shared the same fate. Some indulgence, I think, may still be given to the more familiar style of dialogues, letters, essays, and even of popular addresses, which, like comedy, are formed on the dialect of conversation. In this dialect, wherein all languages originate, the eagerness of conveying one's sentiments, the rapidity and ease of utterance, necessarily produce

seems worthy of the preference; and it must be established as a rule, that no other order in any case is to be admitted.

But we are not peculiar in this disposition, though we may be peculiar in some of our ways of exerting it. The French critics, and even the Academy, have proceeded, if not always in the same manner, on much the same principle in the improvements they have made on their language. They have, indeed, cleared it of many, not of all their low idioms, cant phrases, and useless anomalies; they have rendered the style, in the main, more perspicuous, more grammatical, and more precise than it was before. But they have not known where to stop. Their criticisms often degenerate into refinements, and everything is carried to excess. If one mode of construction, or form of expression, hath been lucky enough to please those arbitrators of the public taste, and to obtain their sanction, no different mode or form must expect so much as a toleration. What is the consequence? They have purified their language; at the same time, they have impoverished it, and have, in a considerable measure, reduced all kinds of composition to a tasteless uniformity. Accordingly, in perhaps no language, ancient or modern, will you find so little variety of expression in the various kinds of writing as in French. In prose and verse, in philosophy and romance, in tragedy and comely, in epic and pastoral, the difference may be very great in the sentiments, but it is nothing, or next to nothing, in the style.

Is this insipid sameness to be envied them as an excel such abbreviations. It appears, indeed, so natural, that I think it requires that people be more than commonly phlegmatic, not to say stupid, to be able to avoid them. Upon the whole, therefore, this tendency, in my opinion, ought to have been checked and regulated, but not entirely crushed. That contracting serves to improve the expression in vivacity is manifest; it was necessary only to take care that it might not hurt it in harmony or in perspicuity. It is certainly this which constitutes one of the greatest beauties in French dialogue, as by means of it, what in other languages is expressed by a pronoun and a preposition, they sometimes convey, not by a single syllable, but by a single letter. At the same time, it must be owned, they have never admitted contractions that could justly be denominated harsh; that they have not, on the other hand, been equally careful to avoid such as are equivocal, hath been observed already. We are apt to imagine that there is something in the elision of letters and contraction of syllables that is particularly unsuitable to the grave and solemn style. This notion of ours is, I suspect, more the consequence of the disuse than the cause, since such abbreviations do not offend the severest critic when they occur in books written in an ancient or a foreign language. Even the sacred penmen have not disdained to adopt them into the simple, but very serious style of Holy Writ. Witness the καγω for και εγω, απ' εμου for απο εμού, κακείνος for Kaι EKELVOS, and many others. No doubt desuetude alone is sufficient to create an unsuitableness in any language. I will admit farther, that there is some convenience in discriminating the different characters of writing by some such differences in the style. For both these reasons, I should not now wish to see them revived in performances of a serious or solemn na

ture.

lence? Or shall we Britons, who are lovers of freedom almost to idolatry, voluntarily hamper ourselves in the trammels of the French Academy? Not that I think we should disdain to receive instruction from any quarter, from neighbours, or even from enemies. But as we renounce implicit faith in more important matters, let us renounce it here too. Before we adopt any new measure or limitation, by the practice of whatever nation it comes recommended to us, let us give it an impartial examination, that we may not, like servile imitators, copy the bad with the good. The rules of our language should breathe the same spirit with the laws of our country. They ought to prove bars against licentiousness, without being checks to liberty.

SECTION III.

MODERN LANGUAGES COMPARED WITH GREEK AND LATIN, PARTICULARLY IN REGARD TO THE COMPOSITION OF SENTENCES.

BEFORE I Conclude this chapter, I must beg leave to offer a few general remarks on the comparison of modern languages with Greek and Latin. This I am the rather disposed to do, that it will serve farther to illustrate the principles above laid down. I make no doubt but the former have some advantage in respect of perspicuity. I think not only that the disposition of the words, according to certain stated rules, may be made more effectually to secure the sentence against ambiguous construction than can be done merely by inflection, but that an habitual method of arranging words which are in a certain way related to one another, must, from the natural influence of habit on the principle of association, even where there is no risk of misconstruction, more quickly suggest the meaning than can be done in the freer and more varied methods made use of in those ancient languages. This holds especially with regard to Latin, wherein the number of equivocal inflections is considerably greater than in Greek; and wherein there are no articles, which are of unspeakable advantage, as for several other purposes, so in particular for ascertaining the construction. But while the latter, though in this respect inferior, are, when skilfully managed, by no means ill adapted for perspicuous expression, they are, in respect of vivacity, elegance, animation, and variety of harmony, incomparably superior. I shall at present consider their advantage principally in point of vivacity, which in a great measure, when the subject is of such a nature as to excite passion, secures animation also.

In the first place, the brevity that is attainable in these languages gives them an immense superiority. Some testimonies in confirmation of this remark may be obtained by comparing the Latin examples of antithesis quoted in the notes