by natural means: one bed cannot receive the two lovers, Æneas and Dido, without the immediate interposition of superior powers. The ridiculous in such fictions must appear even through the thickest veil of gravity and solemnity.

Angels and devils serve equally with Heathen deities as materials for figurative language; perhaps better among Christians, because we believe in them, and not in Heathen deities. But every one is sensible, as well as Boileau, that the invisible powers in our creed make a much worse figure as actors in a modern poem, than the invisible powers in the Heathen creed did in ancient poems; the cause of which is not far to seek. The Heathen deities, in the opinion of their votaries, were beings elevated one step only above mankind, subject to the same passions, and directed by the same motives; therefore not altogether improper to mix with men in an important action. In our creed, superior beings are placed at such a mighty distance from us, and are of a nature so different, that with no propriety can we appear with them upon the same stage: man, a creature much inferior, loses all dignity in the comparison.

There can be no doubt that an historical poem admits the embellishment of allegory, as well as of metaphor, simile, or other figure. Moral truth, in particular, is finely illustrated in the allegorical manner: it amuses the fancy to find abstract terms, by a sort of magic, metamorphosed into active beings: and it is highly pleasing to discover a general proposition in a pictured event. But allegorical beings should be confined within their own sphere, and never be admitted to mix in the principal action, nor to co-operate in retarding or advancing the catastrophe. This would have a still worse effect than invisible powers; and I am ready to assign the reason. The impression of real existence, essential to an epic poem, is inconsistent with that figurative existence which is essential to an allegory; and, therefore, no means can more effectually prevent the impression of reality than to introduce allegorical beings co-operating with those whom we conceive to be really existing. The love-episode in the Henriade,† insufferable by the discordant mixture of allegory with real life, is copied from that of Rinaldo and Armida, in the Gierusalemme Liberata, which hath no merit to entitle it to be copied. An allegorical object, such as Fame in the Eneid, and the Temple of Love in the Henriade, may find place in a description; but to introduce Discord as a real personage, imploring the assistance of Love, as another real personage, to enervate the courage of the hero, is making these figurative beings act beyond their sphere, and creating a strange jumble of truth and fiction. The allegory of Sin and Death in the Paradise Lost is, I presume, not generally relished, though it is not entirely of the same nature with what I have been condemning: in a work comprehending the achievements of superior beings, there is more room for fancy than where it is confined to human actions.

What is the true notion of an episode? or how is it to be distinguished from the principal action? Every incident that promotes or retards the catastrophe, must be part of the principal action. This clears the nature of an episode; which may be defined, "An inci* See chap. 20. sect. 6.

+ Canto 9.

dent connected with the principal action, but contributing neither to advance nor to retard it." The descent of Eneas into hell doth not advance nor retard the catastrophe, and therefore is an episode. The story of Nisus and Euryalus, producing an alteration in the affairs of the contending parties, is a part of the principal action. The family scene in the sixth book of the Iliad is of the same nature; for by Hector's retiring from the field of battle to visit his wife, the Grecians had opportunity to breathe, and even to turn upon the Trojans. The unavoidable effect of an episode, according to this definition, must be, to break the unity of action; and therefore it ought never to be indulged, unless to unbend the mind after the fatigue of a long narration. An episode, when such is its purpose, requires the following conditions: it ought to be well connected with the principal action; it ought to be lively and interesting; it ought to be short and a time ought to be chosen when the principal action relents.*

In the following beautiful episode, which closes the second book of Fingal, all these conditions are united:

Comal was the son of Albion; the chief of a hundred hills. His deer drank of a thousand streams; and a thousand rocks replied to the voice of his dogs. His face was the mildness of youth; but his hand the death of heroes. One was his love, and fair was she! the daughter of mighty Conloch. She appeared like a sunbeam among women, and her hair was like the wing of the raven. Her soul was fixed on Comal, and she was his companion in the chase. Often met their eyes of love, and happy were their words in secret. But Gormal loved the maid, the chief of gloomy Ardven. He watched her lone steps on the heath, the foe of unhappy Comal.

One day tired of the chase, when the mist had concealed their friends, Comal and the daughter of Conloch met in the cave of Ronan. It was the wonted haunt of Comal. Its sides were hung with his arms; a hundred shields of thongs were there, a hundred helms of sounding steel. Rest here, said he, my love Galvina, thou light of the cave of Ronan: a deer appears on Mora's brow; I go, but soon will return. I fear, said she, dark Gormal my foe: I will rest here; but soon return, my love.

He went to the deer of Mora. The daughter of Conloch, to try his love, clothed her white side with her armour, and strode from the cave of Ronan. Thinking her his foe, his heart beat high, and his colour changed. He drew the bow the arrow flew: Galvina fell in blood. He ran to the cave with hasty steps, and called the daughter of Conloch. Where art thou, my love? but no answer.—He marked, at length, her heaving heart beating against the mortal arrow. O Conloch's daughter, is it thou? he sunk upon her breast.

The hunters found the hapless pair. Many and silent were his steps round the dark dwelling of his love. The fleet of the ocean came; he fought, and the strangers fell: he searched for death over the field; but who could kill the mighty Comal? Throwing away his shield, an arrow found his manly breast. He sleeps with his Galvina: their green tombs are seen by the mariner, when he bounds on the waves of the north.

Next, upon the peculiarities of a dramatic poem. And the first I shall mention is a double plot; one of which must resemble an episode in an epic poem; for it would distract the spectator, instead of entertaining him, if he were forced to attend, at the same time, to two capital plots equally interesting. And even supposing it an under-plot like an episode, it seldom hath a good effect in tragedy,

Homer's description of the shield of Achilles is properly introduced at a time when the action relents, and the reader can bear an interruption. But the author of Telemachus describes the shield of that young hero in the heat of battle; a very improper time for an interruption.

of which simplicity is a chief property; for an interesting subject that engages our affections occupies our whole attention, and leaves no room for any separate concern.* Variety is more tolerable in comedy, which pretends only to amuse, without totally occupying the mind. But even there, to make a double plot agreeable, is no slight effort of art: the under-plot ought not to vary greatly in its tone from the principal; for discordant emotions are unpleasant when jumbled together; which, by the way, is an insuperable objection to tragi-comedy. Upon that account, the Provoked Husband deserves censure; all the scenes that bring the family of the Wrongheads into action, being ludicrous and farcical, are in a very different tone from the principal scenes, displaying severe and bitter expostulations between Lord Townley and his lady. The same objection touches not the double plot of the Careless Husband; the different subjects being sweetly connected, and having only so much variety as to resemble shades of colours harmoniously mixed. But this is not all. The under-plot ought to be connected with that which is principal, so much at least as to employ the same persons: the under-plot ought to occupy the intervals or pauses of the principal action; and both ought to be concluded together. This is the case of the Merry Wives of Windsor.

Violent action ought never to be represented on the stage. While the dialogue goes on, a thousand particulars concur to delude us into an impression of reality; genuine sentiments, passionate language, and persuasive gesture: the spectator, once engaged, is willing to be deceived, loses sight of himself, and without scruple enjoys the spectacle as a reality. From this absent state, he is roused by violent action he awakes as from a pleasing dream, and gathering his senses about him, finds all to be a fiction. Horace delivers the same rule, and founds it upon the same reason:

Nec pueros coram populo Medea trucidet;
Aut humana palam coquat exta nefarius Atreus:
Aut in avem Progne vertatur, Cadmus in anguem,
Quodcumque ostendis mihi sic, incredulus odi.

Racine, in his preface to the tragedy of Berenice, is sensible that simplicity is a great beauty in tragedy, but mistakes the cause. Nothing (says he) but verisimilitude pleases in tragedy: but where is the verisimilitude, that, within the compass of a day, events should be crowded which commonly are extended through months?" This is mistaking the accuracy of imitation for the probability or improbability of future events. I explain myself. The verisimilitude required in tragedy is, that the actions correspond to the manners, and the manners to nature. When this resemblance is preserved, the imitation is just, because it is a true copy of nature. But I deny that the verisimilitude of future events, meaning the probability of future events, is any rule in tragedy. A number of extraordinary events are, it is true, seldom crowded within the compass of a day; but what seldom happens may happen; and when such events fall out, they appear no less natural than the most ordinary accidents. To make verisimilitude in the sense of probability a governing rate in tragedy, would annihilate that sort of writing altogether; for it would exclude all extraordinary events, in which the life of tragedy consists. It is very improbable or unlikely, pitching upon any man at random, that he will sacrifice his life and fortune for his mistress or for his country: yet when that event happens, supposing it conformable to the character, we recognize the verisimilitude as to nature, whatever want of verisimilitude or of probability there was a priori that such would be the event.

The French critics join with Horace in excluding blood from the stage; but overlooking the most substantial objection, they urge only, that it is barbarous, and shocking to a polite audience. The Greeks had no notion of such delicacy, or rather effeminacy: witness the murder of Clytemnestra by her son Orestes, passing behind the scene as represented by Sophocles: her voice is heard calling out for mercy, bitter expostulations on his part, loud shrieks upon her being stabbed, and then a deep silence. I appeal to every person of feeling, whether this scene be not more horrible than if the deed had been committed in sight of the spectators upon a sudden gust of passion. If Corneille, in representing the affair between Horatius and his sister, upon which murder ensues behind the scene, had no other view but to remove from the spectators a shocking action, he was guilty of a capital mistake; for murder in cold blood, which in some measure was the case as represented, is more shocking to a polite audience, even where the conclusive stab is not seen, than the same act performed in their presence by violent and unpremeditated passion, as suddenly repented of as committed. I heartily agree with Addison,* that no part of this incident ought to have been represented, but reserved for a narrative, with every alleviating circumstance in favour of the hero.

A few words upon the dialogue; which ought to be so conducted as to be a true representation of nature. I talk not here of the sentiments, nor of the language; for these come under different heads: I talk of what properly belongs to dialogue-writing; where every single speech, short or long, ought to arise from what is said by the former speaker, and furnish matter for what comes after, till the end of the scene. In this view, all the speeches, from first to last, represent so many links of one continued chain. No author, ancient or modern, possesses the art of dialogue equal to Shakspeare. Dryden, in that particular, may justly be placed as his opposite: he frequently introduces three or four persons speaking upon the same subject, each throwing out his own notions separately, without regarding what is said by the rest take for an example the first scene of Aurenzebe. Sometimes he makes a number club in relating an event, not to a stranger, supposed ignorant of it, but to one another, for the sake merely of speaking: of which notable sort of dialogue we have a specimen in the first scene of the first part of the Conquest of Granada. In the second part of the same tragedy, scene second, the King, Abenamar, and Zulema, make their separate observations, like so many soliloquies, upon the fluctuating temper of the mob. A dialogue so uncouth puts one in mind of two shepherds in a pastoral, excited by a prize to pronounce verses alternately, each in praise of his own mistress.

This manner of dialogue-writing, beside an unnatural air, has another bad effect: it stays the course of the action, because it is not productive of any consequence. In Congreve's comedies the action is often suspended to make way for a play of wit. But of this more particularly in the chapter immediately following.

No fault is more common among writers, than to prolong a speech

Spectator, No. 44.

after the impatience of the person to whom it is addressed ought to prompt him or her to break in. Consider only how the impatient actor is to behave in the mean time. To express his impatience in violent action without interruption would be unnatural; and yet to dissemble his impatience, by appearing cool where he ought to be highly inflamed, would be no less so.

Rhyme being unnatural and disgusting in a dialogue, is happily banished from our theatre: the only wonder is that it ever found admittance, especially among a people accustomed to the more manly freedom of Shakspeare's dialogue. By banishing rhyme, we have gained so much, as never once to dream of any further improvement. And yet, however suitable blank verse may be to elevate characters and warm passions, it must appear improper and affected in the mouths of the lower sort. Why then should it be a rule, That every scene in tragedy must be in blank verse? Shakspeare, with great judgment, has followed a different rule; which is, to intermix prose with verse, and openly to employ the latter where it is required by the importance or dignity of the subject. Familiar thoughts and ordinary facts ought to be expressed in plain language: to hear, for example, a footman deliver a simple message in blank verse, must appear ridiculous to every one who is not biassed by custom. In short, that variety of characters and of situations, which is the life of a play, requires not only a suitable variety in the sentiments, but also in the diction.

CHAP. XXIII.

THE THREE UNITIES.

In the first chapter is explained the pleasure we have in a chain of connected facts. In histories of the world, of a country, of a people, this pleasure is faint; because the connexions are slight or obscure. We find more entertainment in biography; because the incidents are connected by their relation to a person who makes a figure, and commands our attention. But the greatest entertainment is in the history of a single event, supposing it interesting; and the reason is, that the facts and circumstances are connected by the strongest of all relations, that of cause and effect; a number of facts that give birth to each other form a delightful train; and we have great mental enjoyment in our progress from the beginning to the end.

But this subject merits a more particular discussion. When we consider the chain of causes and effects in the material world, independent of purpose, design, or thought, we find a number of incidents in succession, without beginning, middle, or end every thing that happens is both a cause and effect: being the effect of what goes before, and the cause of what follows: one incident may affect us more, another less; but all of them are links in the universal chain : the mind, in viewing these incidents, cannot rest or settle ultimately upon any one; but is carried along in the train without any close.