that the contrasted features in this moral painting serves to ascertain the direction and boundaries of one another with greater precision than could otherwise be accomplished. It is too nice a matter, without the aid of this artifice, for even the most copious and expressive language. For a specimen in this way take these lines of Pope :

Should such a man, too fond to rule alone,
Bear, like the Turk, no brother near the throne,
View him with scornful, yet with jealous eyes,
And hate for arts that caus'd himself to rise;
Damn with faint praise, || assent with civil leer,
And without sneering, teach the rest to sneer;
Willing to wound, || and yet - afraid to strike,
Just hint a fault, || and hesitate dislike;
Alike reserved to blame, or to commend,
A tim'rous foe, || by flatterers besieged,
And so obliging, that he ne'er obliged."

With what a masterly hand are the colours in this picture blended! and how admirably do the different traits, thus opposed, serve, as it were, to touch up and shade one another! I would not be understood by this to signify my opinion of its likeness to the original. I should be sorry to think that it deserves this praise. The poet had received, or fancied he had received, great provocation. A perfect impartiality in one under the influence of resentment is more than can be expected from human nature. I only speak of the character here exhibited, as one, who, speaking of a portrait, without knowing the person for whom it was drawn, says it is well painted, and that there is both life and expression in the countenance.

If there be any style of composition which excludes antithesis altogether, (for I am not positive that there is,) it is the pathetic. But the true reason which hath induced some critics immoderately to decry this figure is, that some authors are disposed immoderately to employ it. One extreme naturally drives those who perceive the error to the opposite extreme. It rarely leaves them, even though persons of good sense and critical discernment, precisely where they were before. Such is the repulsive power of jarring tastes. Nay, there is a kind of mode, which in these, as well as in other matters, often influences our censures without our knowing it. It is this which sometimes leads us to condemn as critics what as authors we ourselves practise. Witness the following reproach from the author just now quoted.

I see a chief who leads my chosen sons,
All arm'd with points, antitheses, and puns.†

On the other hand, it is certain, that the more agreeable the apposite and temperate use of this figure is, the more offensive is the abuse, or, which is nearly the same, the immoderate use of it. When used moderately, the appearance of art, which it might otherwise have, is veiled, partly by the energy of the expression, which doth not permit † Dunciad.

"Prologue to the Satires.

the hearer at first to attend critically to the composition, and partly by the simplicity, or at least the more artless structure, both of the preceding sentences and of the following. But if a discourse run in a continued string of antithesis, it is impossible the hearer should not become sensible of this particularity. The art is in that case quite naked. Then indeed the frequency of the figure renders it insipid, the sameness tiresome, and the artifice insufferable.

The only original qualities of style which are excluded from no part of a performance, nay, which ought, on the contrary, to pervade the whole, are purity and perspicuity. The others are suited merely to particular subjects and occasions. And if this be true of the qualities themselves, it must certainly be true of the tropes and figures which are subservient to these qualities. In the art of cookery, those spiceries which give the highest relish must be used the most sparingly. Who then could endure a dish, wherein these were the only ingredients? There is no trope or figure that is not capable of a good effect, I do not except those which are reckoned of the lowest value, alliteration, paronomasis, or even pun. But then the effect depends entirely on the circumstances. If these are not properly adjusted, it is always different from what it was intended to be, and often the reverse.

The antithesis in particular gives a kind of lustre and emphasis to the expression. It is the conviction of this that hath rendered some writers intemperate in the use of it. But the excess itself is an evidence of its value. There is no risk of intemperance in using a liquor which hath neither spirit nor flavour. On the contrary, the richer the beverage is, the danger is the greater, and therefore it ought to be used with the greater caution. Quintilian hath remarked concerning the writings of Seneca, which are stuffed with antithesis, that "they abound in pleasant faults."* The example had not been dangerous, if the faults had not been pleasant. But the danger here was the greater, as the sentiments conveyed under these figures were excellent. The thought recommended the expression. An admiration of the former insinuated a regard to the latter, with which it was so closely connected, and both very naturally engaged imitation. Hence Seneca is justly considered as one of the earliest corrupters of the Roman eloquence. And here we may remark by the way, that the language of any country is in no hazard of being corrupted by bad writers. The hazard is only when a writer of considerable talents hath not a perfect chastity of taste in composition; but, as was the case of Seneca, affects to excess what in itself is agreeable. Such a style, compared with the more manly elocution of Cicero, we call effeminate, as betraying a sort of feminine fondness for glitter and ornament. There is some danger that both French and English will be corrupted in the same manner. There have been some writers of eminence in both, who might be charged, perhaps as justly as Seneca, with abounding in pleasant faults.

But enough of the antithesis; I return to the consideration of periods in general. And on this head I shall only further remark, that when they consist of complex members, we must follow the same

• Instit. Lib. x. Cap. 1. Abundant dulcibus vitiis.

rule in arranging the clauses of each member, in order to give all possible energy to the sentence, that we do in arranging the members of the period. By doing this, we shall never be in danger of thinking that the member is complete till it actually be so, just as by the structure of the period we are prevented from thinking the sentence finished before the end. A disappointment in the former case is of less moment, but it is still of some. In each it occasions a degree of languor which weakens the expression.

I shall give an example of a period where, in one of the members, this rule is not observed. "Having already shown how the fancy is affected by the works of Nature, and afterwards considered in general both the works of Nature and of Art || how they mutually assist and complete each other, in forming such scenes and prospects || as are most apt to delight the mind of the beholder; I shall in this paper throw together some reflections on that particular art, || which has a more immediate tendency than any other, || to produce those pleasures of the imagination, || which have hitherto been the subject of this discourse. ""* This sentence is a period, agreeably to the definition formerly given. Wherever we stop, the sentence is imperfect till we reach the end. But the members are not all composed according to the rule laid down. It consisteth of three members. The first ends at Nature, is a single clause, and therefore not affected by the rule; the second is complex, consisting of several clauses, and ends at beholder; the third is also complex, and concludes the sentence. The last number cannot be faulty, else the sentences would be no period. The fault must then be in the structure of the second, which is evidently loose. That member, though not the sentence, might conclude, and a reader naturally supposes that it doth conclude, first at the word art, afterwards at the word other, both which are before its real conclusion. Such a composition, therefore, even in periods, occasions, though in a less degree, the same kind of disappointment to the reader, and consequently the same appearance of feebleness in the style, which result from long, loose, and complex sentences. A very little alteration in the faulty member will unite the clauses more intimately, and entirely remove the exception; as thus, "and afterwards considered in general, how, in forming such scenes and prospects as are most apt to delight the mind of the beholder, the works both of Nature and of Art mutually assist and complete each other."

It may be thought, and justly too, that this care will sometimes make the expression appear elaborate. I shall only recommend it as one of the surest means of preventing this effect, to render the members as simple as possible, and particularly to avoid synonymas and redundancies, of which there are a few in the member now criticised. Such are, scenes and prospects, assist and complete, mutually and each other. With the aid of this reformation also, the whole period will appear much better compacted as follows: "Having already shown how the fancy is affected by the works of Nature e; and afterwards considered in general, || how in forming such scenes as are

*Spectator, No. 415. O.

most apt to delight the mind of the beholder, the works both of Nature and of Art assist each other; I shall in this paper throw together some reflections on that particular art, || which has a more immediate tendency than any other, || to produce those primary pleasures of the imagination, which have hitherto been the subject of this discourse."

PART III. Observations on loose sentences.

IN complex sentences of looser composition, there is, as was observed, a much greater risk of falling into a languid manner. This may arise from different causes. First, even where the sentence is neither long nor complex, the members will sometimes appear disjointed. The consequence always is, that a hearer will at first be in doubt, whether it be one sentence or more. Take the following for an example: "However, many who do not read themselves, | are seduced by others that do; and thus become unbelievers upon trust, and at second hand; and this is too frequent a case."* The harmony of the members taken severally contribute to the bad effect of the whole. The cadence is so perfect at the end both of the first member and of the second, that the reader is not only disappointed, but surprised, to find the sentence still unfinished. The additional clauses appear out of their proper place, like something that had been forgotten.

Another cause of languor here is the excessive length of a sentence, and too many members. Indeed, wherever the sentiments of an author are not expressed in periods, the end of a member or clause, or even an intermediate word, as hath been observed already, may be the end of the sentence. Yet the commonness of such sentences, when they do not exceed an ordinary length, prevents in a great measure a too early expectation of the end. On the contrary, when they transgress all customary limits, the reader begins to grow impatient, and to look for a full stop or breathing-place at the end of every clause and member. An instance of this excess you have in the succeeding quotation: "Though in yesterday's paper, we considered how every thing that is great, new, or beautiful, is apt to affect the imagination with pleasure, we must own that it is impossible for us to assign the necessary cause of this pleasure, because we know neither the nature of an idea, nor the substance of a human soul, which might help us to discover the conformity or disagreeableness of the one to the other; and therefore, for want of such a light, all that we can do, in speculations of this kind, is, to reflect on those operations of the soul that are most agreeable, and to range, under their proper heads, what is pleasing or displeasing to the mind, without being able to trace out the several necessary and efficient causes from whence the pleasure or displeasure arises."t The reader will observe, that in this passage I have distinguished by italics all those words in the body of the sentence, no fewer than seven, at any of

Swift's Sermon on the Trinity.

† Spectator, No. 413. O.

which, if there were a full stop, the construction of the preceding part would be complete. The fault here is solely in the length of the whole, and in the number of the parts. The members themselves are well connected.

In the next example we have both the faults above mentioned in one sentence: "Last year a paper was brought here from England, called a Dialogue between the Archbishop of Canterbury and Mr. Higgins, which we ordered to be burnt by the common hangman, as it well deserved, though we have no more to do with his Grace of Canterbury, than you have with the Archbishop of Dublin, whom you tamely suffer to be abused openly, and by name, by that paltry rascal of an observator; and lately upon an affair wherein he had no concern; I mean the business of the missionary of Drogheda, wherein our excellent primate was engaged, and did nothing but according to law and discretion."* Hardly will you find in any of the worst English writers a more exceptionable sentence in point of composition than the preceding, which is taken from one of the best. The stops which might be in it will be found, on an attentive perusal, to be no fewer than fourteen; the clauses are exceedingly unequal, abrupt, and ill-compacted. Intricacy in the structure of a complex sentence might also be here exemplified as a cause of languor. But as this errror never fails to create obscurity, it hath been considered already under a former head.

PART IV. - Review of what has been deduced above in regard to arrangement.

I HAVE now briefly examined how far arrangement may contribute to vivacity, both in simple sentences and in complex, and from what principles in our nature it is that the effect ariseth.

In this discussion I have had occasion to consider, in regard to simple sentences, the difference between what may properly be called the rhetorical and natural order, and that which I have denominated the artificial and grammatical, or the customary way of combining the words in any particular language. I have observed, as to the former, and taken some pains to illustrate the observation, that it is universal, that it results from the frame of spirit in which the sentiment, whatever it be, is spoken or written, that it is by consequence a sort of natural expression of that frame, and tends to communicate it to the hearer or the reader. I have observed, also, that this order, which alone deserves the name of Natural, is in every language more or less cramped by the artificial or conventional laws of arrangement in the language; that, in this respect, the present languages of Europe, as they allow less latitude, are considerably inferior to Greek and Latin, but that English is not a little superior in this particular to some of the most eminent of the modern tongues. I have shown also, that the artificial arrangement is different in different languages, and seems chiefly accommodated to such simple explanation, narra

*Swift's Letter concerning the Sacramental test,