Ch. 1. is expreffed merely by the words standing together in the sentence.

As to the firft, words, in the learned languages, indicate themfelves their connections, by genders, numbers, and cafes; and this in two different ways, either by concord or by regimen. Concord is when the words agree together in certain accidents, fuch as thofe above mentioned. Regimen is when they do not fo agree, but the one, which is called the governing word, produces fome alteration upon the other.

This kind of fyntax, whereby the words themfelves fhew how they are to be connected, is by far the moft artificial, particularly with respect to the cafes; the relations expreffed by which are, as I have fhewn, very difficult to be explained, and comprehended in general definitions. I fhall only add here, that as the great use of cafes, as well as of genders and numbers, is for the purpose of fyntax, what I have already faid of thefe accidents of words, will make it unneceffary for me to fay any thing further here upon the fubject.

But fuppofe a language, of which the

art

art is fo imperfect that it has no cafes at Ch. 1. all, no genders of fubftantives, neither genders nor numbers of adjectives, and very little expreffion of numbers, even in their verbs, which is the cafe of the Englifh, and, for the greater part, of all the modern languages of Europe; in what manner are the words to be connected in fuch a language? It is evident it can only be in one or other, or both the two ways laft mentioned, namely, either by feparate words, or by juxtapofition of the words to be connected together; which last way, as we have elsewhere observed, is almost the only fyntax of the barbarous languages; and as it is a great part of the syntax of the modern languages of Europe, fo far at least it must be admitted, that these languages approach to barbarity.

But, with respect even to the learned languages, let us fuppofe that the relations betwixt things are fuch, that they cannot be expreffed conveniently by cafes, or flections of the word of any kind; what is then to be done? In that cafe the relations are to be expreffed by feparate words, called prepofitions, of which, as I have already explained the nature and use, and

diftinguished

Ch. 1. distinguished betwixt the relations expreffed by cafes, and thofe expreffed by prepofitions, I will fay no more of them at prefent.

Again, fuppofe the word is indeclinable, fo that its connection with the other word cannot be marked by any change of it, which is the cafe of adverbs, conjunctions, and the prepofitions themselves, then must even the learned languages of neceffity fubmit to the barbarous syntax above mentioned, and connect the words together by juxtapofition. In this way adverbs in Greek and Latin are connected with the verbs or adjectives to which they belong, and in like manner prepofitions and conjunctions.

And thus it appears from induction, that there can be no fyntax but in one or other of the three ways above mentioned. And it also appears, how far art in this matter can go, and where it must neceffarily ftop.

The general rules of fyntax, as well as of every other part of the grammatical art, and indeed of every art, must be founded in the nature of things.

For, as the bu

finefs of fyntax is to connect words, which

ftand

41

ftand for ideas, and as ideas represent Ch. 1. things, it is evident that words must be joined together according to the nature of the things they ftand for. For this reafon it is that in fyntax, as well as in nature, the fubftance is joined with the accident, the action with the agent, or with the qualities and circumstances of the action, and both the action and agent with the fubject of the action. But all this is fo fully, accurately, and elegantly explained by Mr Harris*, that I will not say a word more upon the subject.

The differences betwixt the antient and modern syntax I have already noted; but it is very well worth while further to obferve the difference thereby produced, betwixt antient and modern compofition. The difference which it makes with respect to found, and the pleasure of the ear, I fhall obferve when I come to speak of the compofition of the founds of language. But, in the next chapter, I will obferve what difference it makes in conveying the fense.

Hermes, book 2. cap. 3.

CHAP.

344

Ch. 2.

CHA P. II.

The difference betwixt the arrangement of words in antient and modern languages con= fidered. The most natural order of are rangement, whether the antient or the mo dern. The advantages of the antient arrangement in conveying the fenfe of the Speaker.

Know it is the opinion of many, that the antient compofition, or arrange ment of words in fentences, however pleafing it might be to the ear, was hurtful to the fenfe, as it diftracted the attention, and hindered the mind from apprehending the fense fo clearly and fully as it would other wife do; that the modern arrangement was more natural, conveyed the meaning bet ter, and was therefore on that account preferable. And I remember I heard one of thofe gentlemen go fo far as to fay, that it was impoffible that Demofthenes's orations could have been understood, if the words had been spoken in the unnatural order