praise being neither associated with, nor distinguished from, any succeeding words, they ought to be united with those that precede, as both of them form a member sufficiently short to be pronounced with ease; but if distinctness had made it necessary to pause at praise, then notwithstanding the shortness of the phrase, it would have formed a distinct member, and have readily admitted a pause. Thus in the sentence, A violent and ungovernable passion for praise, rather than improvement in virtue, produces often the most ridiculous circumstances, &c. : here the word praise, being emphatically distinguished from improvement in virtue, demands a pause after it; and as this word, and its opposite, form a class more united together than both are with the word passion, a pause is necessary, to shew they belong to distinct classes; the pause between the opposing words shewing their distinction, and the pause before and after them shewing their union.

But it may be asked, how can we suppose words opposed to each other, and requiring a pause to shew that opposition, can be more united with each other than they are with the preceding words they modify? It may be answered, that the modifying word, when unaccompanied by adjuncts, and the word modified, form but one class, and do not admit of a pause, either when the modifying word precedes or succeeds the word modified. Thus in the phrases, It was from a prepense malice that he committed the action; and, It was from a malice prepense that he committed the action: In these phrases, I say, the substantive malice, and the adjective prepense, are equally inseparable by a pause; but in the following phrases:

It was from a preconceived and prepense malice that he committed the action: and, It was from a malice, preconceived and prepense, that he committed the action. In the former of these phrases, the mod

ifying words do not form a distinct class from the word modified; and in the latter they do, and, therefore, admit of a pause after the word malice, which can arise from nothing else but this: in one case, the modifying words, preceding the word modified, can signify nothing without being joined to it; and in the other, the modified word, preceding those that modify, does signify something independent on them; and this independent signification admits those words that equally depend on it, to form a distinct, though not an independent, class, by permitting a pause. Hence arises this general ruleThe word modified, and the words modifying, form but one class with relation to the rest of the words of the sentence; but if the modifying words precede the word modified, the modifying words are distinguished from each other by a pause, but not from the word modified; and if the modifying words succeed the word modified, they are not only distinguished from each other, but from the word which they modify; that is, they form distinct classes respecting each other, and one whole class respecting the rest of the words in the sentence.

Thus have we endeavoured to trace out the rea son for pausing differently in phrases differently constructed, though perfectly similar in meaning. In this enquiry, the ingenious researches of Lord Kaims upon this subject have been of great use. His idea of the connection between the adjective and the substantive in their natural order, and the separation they admit of when inverted, is the principal clue to the difficulties that have been proposed: his assertion, however, that the adjective and substantive in an inverted order admit of a pause, is true only when the adjective is single; for thousands of instances might be produced, where a pause is no more admissible between a substantive and an ad. jective in their inverted than in their natural order.

For example, in the following lines from the Rape of

the Lock:

Of these the chief the care of nations own,

And guard with arms divine the British throne.

Though the melody of the verse inclines us strongly to pause at arms, yet the adjective divine, immediately succeeding, forbids it. Nay, if the line Lord Kaims produces to prove we may pause between the adjective and the substantive in an invert. ed order

For thee the fates, severely kind, ordain......

If this line, I say, had been constructed in this man

ner,

For thee the fates severe, have this ordain'd,

it is evident no pause could be admitted between the substantive fates and the adjective severe, though they are here in their inverted order; it is not then merely the adjective being placed after the substan tive which makes it separable from it, but the adjective being joined by other words, which, when the substantive is understood, are more immediately connected with each other than with the substantive itself.

If these observations have any solidity, we may perceive how few are the grammatical connections which absolutely refuse a suspension of pronunciation, for the sake of breathing, where precision or energy require it: it is certainly to be presumed, that the breath of every person is nearly proportioned to the forcible pronunciation of so many words together as are necessary to preserve the sense unbroken; the contrary, however, would often be the case, if the integrity of the sense depended on the common rules for placing the comma. Let those,

however, who can pronounce a long sentence easily and forcibly, provided they preserve the pauses neces sary to the sense, take breath as seldom as they please. I have rather consulted the infirmities than the perfections of my fellow creatures; by endeavouring to point out those resources which are necessary to the weak, without imposing them as rules upon the strong;-Clausulas enim, says Cicero, atque interpuncta verborum animæ interclusio atque angustia spiritus adtulerunt. De Orat. Lib. iii.

But from studying the human voice, and not relying implicitly on the assertions of the ancients, we perceive the weakness of that common observation, that long sentences require a greater quantity of breath, and a much more forcible exertion in the lungs, than such sentences as are short. The folly of this opinion must evidently appear to those who have taken notice how often we may pause in a long sentence; and it will be shown hereafter, that the sense of a sentence depends much less on the pause. than on the inflexion of voice we adopt; and that, provided we pause in the proper place, and preserve the proper tone and inflexion of the voice, the sense runs no risk on account of the multiplicity or duration of the pauses.

To reduce what has been said into something like a system, we shall endeavour to bring together sentences in every variety of construction, and mark, as carefully as possible, such pauses, as are necessary to pronounce them with clearness, force, and variety.

A Practical System of Rhetorical Punctuation.

BEFORE We give such directions for pausing, or dividing a sentence, as will, in some measure, enable us to aviod the errors of common punctuation, and to point for ourselves, it will be necessary to inquire into the nature of a sentence, and to distinguish it into its different kinds: for this purpose, I shall make use of the words of a very ingenious author,* who has lately written on the Philosophy of Rhetorick: Complex sentences,' says this author, 'are of two 'kinds; first, they are either periods, or sentences of 'a looser composition, for which the language doth not furnish us with a particular name.

A period is a complex sentence, wherein the meaning remains suspended, till the whole is finish'ed: the connection, consequently, is so close be'tween the beginning and the end, as to give rise to 'the name period, which signifies circuit; the fol'lowing is such a sentence:

"Corruption could not spread with so much success, though reduced into system, and though some "ministers, with equal impudence and folly, avowed "it, by themselves and their advocates, to be the prin"cipal expedient by which they governed, if a long "and almost unobserved progression of causes and "effects did not prepare the conjuncture."

Bolingbroke's Spirit of Patriotism.

'The criterion of a period is this: If you stop any *where before the end, the preceding words will not 'form a sentence, and therefore cannot convey any 'determined sense.

This is plainly the case with the above example: "the first verb being could, and not can; the poten

Campbell's Philos. of Rhetorick, vok ii. p. 339.