On Poetry, its influence and species. LET us speak concerning poetry itself, and its [different] species1; what power each possesses, and how fables must be composed, in order that poetry may be such as is fitting: further still, [let us show] of how many and what kind of parts it consists; and in like manner [let us treat] concerning such other things as pertain to this method, beginning, conformably to nature, first from such things as are first.

2. The imitative arts. The epopee, therefore, and tragic poetry, and moreover comedy, and dithyrambic poetry, and the greatest part of the art pertaining to the flute and the lyre , are all entirely imitations 3. 3. Their difference. They differ, however, in three things; for [ they differ] either by imitating through means different in kind, or by imitating different objects, or in a different, and not after the same manner. 4. For as certain persons assimilating, imitate many things by colours and figures, some indeed through art, but others through custom, [ and others through voice]4; thus also in the afore- " Poesis in species sive formas (εἴδη), formae rursus in partes (μόρια, μέρη) dividuntur." Ritter. Cf. Plato Theag. p. 10. C. Cithern-playing was one of the favourite accomplishments of the Athenian youth, ibid. p. 8. E. Alcib. Pr. p. 26. D. Laem. and Olympiodor. in eund. t. ii. p. 65. ed. Creuz. See Ritter, and Twining's Dissertations and first note. Hermann reads φύσεως for φωνη̃ς with Madius, but Ritter condemns the words as spurious. mentioned arts, all of them indeed produce imitation in rhythm, words, and harmony; and in these, either distinctly, or mingled together, as, for instance, the arts of the flute and the lyre alone employ harmony and rhythm5; and this will also be the case with any other arts which possess a power of this kind, such as the art of playing on reed-pipes. 5. Dancing imitates by rhythm only. But the arts pertaining to dancing imitate by rhythm, without harmony ; for dancers, through figured rhythms, imitate manners, and passions, and actions. 6. The epopee by words. But the epopee alone imitates by mere words or metres, and by these either mingling them with each other, or employing one certain kind of metres, which method has been adopted up to the present time. 7. Etymology of the name Poet. For otherwise we should have no common name, by which we could denominate the Mimes of Sophron and Xenarchus, and the dialogues of Socrates; or those who imitate by trimetres, or elegies, or certain other things of this kind; except that men joining with metre the verb to make8, call some of these makers of elegies, but others epic makers, not as poets according to imi- Cf. Rhet. iii. 8. There is much difficulty about this definition of ἐποποιία, as λόγοις ψιλοῖς is supposed by some to mean prose, (see Robortell. p. 14,) by others, verse without music. The former opinion is advocated by Hermann, the latter by Buhle. Ritter seems to have clearly shown that ἢ τοῖς μέτροις is added, not as an explanation of λόγοις ψιλοῖς, but disjunctively, and that ἐποποιία is used in a new sense for certain parts of ing as well as verse. The sense is, therefore, " by prose or by metre, but unaccompanied by song." Ritter observes that the following passage is an excuse for the new signification of ἐποποιία. It may be necessary to observe, that the Greek word (ποιητής—poiétes) whence poeta, and poet, is, literally, maker; and maker, it is well known, was once the current term for poet in our language; and to write verses, was, to make. Sir Philip Sidney, speaking of the Greek word, says, " wherein, I know not whether by luck or wisdom, we Englishmen have met with the Greeks, in calling him maker." Defense of Poesy. So Spenser: The god of shepherds, Tityrus, is dead, Who taught me, homely, as I can, to MAKE. Shep. Cal. June. Twining. tation, but denominating them in common according to measure. For they are accustomed thus to denominate them, if they write any thing medical or musical in verse. 8. Which name is common to writers on other subjects. There is, however, nothing common to Homer and Empedocles except the measure; on which account, it is right indeed to call the former a poet; but the latter, a physiologist rather than a poet. 9. But the adherence to metre does not make the poet. In like manner, though some one, mingling all the measures, should produce imitation, as Chaeremon has done in his Centaur, a mixed rhapsody of all the metres, yet he must not be called a poet. 10. Means respectively employed by the different kinds of poetry. Let it then be thus laid down concerning these particulars. But there are some kinds of poetry which employ all the before-mentioned means, I mean, rhythm, melody, and measure, such as dithyrambic poetry and the Nomes10, and also tragedy and comedy. But these differ, because some of them use all these at once, but others partially. I speak, therefore, of these differences of the arts in respect to the means by which they produce imitation.

Objects of imitation either good or bad, as in painting. BUT since imitators imitate those who do something, and it is necessary that these should either be worthy or depraved persons; (for manners nearly always depend on these alone, since all men differ in their manners by vice and virtue;) it is necessary either [to imitate] those who are better than we are, " For imitation makes the poet, not the metre." Ritter. In dithyrambic or Bacchic hymns, and in the Nomes, which were also a species of hymns to Apollo and other deities, all the means of imitation were employed together, and throughout: in tragedy and comedy, separately ; some of them in one part of the drama, and some in another. In the choral part, however, at least, if no where else, all, melody, rhythm, and words, must probably have been used at once, as in the hymns. Twining. or those who are worse, or such as are like ourselves 1, in the same manner as painters do. For Polygnotus, indeed, painted men more beautiful than they are, but Pauson less so, and Dionysius painted them as they are. 2. And imitations different. But it is evident that each of the before-mentioned imitations will have these differences; and imitation is different, by imitating different things after this manner. 3.For there may be differences of this kind in dancing, in playing on the flute, on the lyre, and also in orations and mere measure. Thus Homer imitates better men8 [than exist], but Cleophon men as they are; and Hegemon the Thasian, who first made parodies, and Nicochares, who wrote the Deliad, imitate worse characters. 4. In like manner in dithyrambics and the Nomi, [as Timotheus and Philoxenus have imitated the Persians and the Cyclops,] one may imitate. By this very same difference, also, tragedy differs from comedy. For the one seeks to imitate worse, but the other better men than are.

Imitations different in the manner. THERE is also a third difference of these, consisting in the manner in which one may imitate each of Or, “those who are commonly found." Polygnotus and Dionysius lived about 01. 80; Pauson about 01. 90. On the following poets see Ritter. Superior, that is, in courage, strength, wisdom, prudence, etc.-in any laudable, useful, or admirable quality, whether such as we denominate moral, or not. If superiority of moral character only were meant, the assertion would be false.-It is necessary to remember here, the wide sense in which the ancients used the terms virtue, vice—good, bad, etc. See note 19.—The difference between moral and poetical perfection of character, is well explained by Dr. Beattie, Essay on Poetry, etc., Part I. ch. 4. The heroes of Homer, as he well observes, are "finer animals" than we are, (p. 69,) not better men. Twining. Ritter would throw out the words enclosed in brackets. See his note. them. For by the same instruments the same things may be imitated, the poet sometimes himself narrating, and sometimes assuming another person [as Homer does]1 ; or speaking as the same person without any change; or as all imitate [who do so] by deed and action. But imitation consists in these 2. And the differences of imitation are three. three differences, as we said in the beginning; viz. in the means, the objects, or the manner. Hence, Sophocles will in one respect be the same imitator as Homer, for both of them imitate elevated characters ; and in another the same as Aristophanes, for both of them imitate persons engaged in acting; 3. Etymology of the drama and claims to its invention. [2 whence also it is said that certain persons call their works dramas, because they imitate those who are engaged in doing something. On this account the Dorians lay claim to the invention of tragedy and comedy; of comedy indeed the Megarians, as well those who are natives of Greece, as being invented by them at the time when their government was a democracy, as those of Sicily. For thence was the poet Epicharmus, who was much prior to Chonides and Magnes. But some of those Dorians who inhabit Peloponnesus lay claim to tragedy, making names an evidence. For they allege that they call their villages komai, but the Athenians demo;; as if comedians were not so denominated from komazein, [i. e. to revel,] but from their wandering through villages, being ignominiously expelled from the cities. The verb poiein also, or to make, is by the Dorians denominated dran, but by the Athenians prattein.] 4. And thus much concerning the differences of imitation, as to their number and quality. But this assertion is not correct, and Ritter shows that the words are spurious. The learned note of Ritter seems to condemn the whole of this passage as spurious.

Two causes Two causes, however, and these physical, appear to have produced poetry in general. 2. of poetry, and those natural: the first, the natural taste for imitation,For to imitate is congenial to men from childhood. And in this they differ from other animals, that they are most imitative, and acquire the first disciplines through imitation; and that all men delight in imitations. But an evidence of this is that which happens in the works [of artists]. 3. seen in respect to the works of art; For we are delighted on surveying very accurate images, the realities of which are painful to the view; such as the forms of the most contemptible animals, and dead bodies. 4. because to learn is ever pleasant. The cause, however, of this is, that learning is not only most delightful to philosophers, but in like manner to other persons, though they partake of it but in a small degree. 5. Hence a delight in images.For on this account, men are delighted on surveying images, because it happens that by surveying they learn and infer what each particular is; as, that this is an image of that man; since, unless one happen to have seen [the reality], it is not the imitation that pleases 1, but [it is through] either the workmanship, or the colour, or some other cause of the like kind. 6. And from imitation, harmony, and rhythm, men originally produced poetry, But imitation, harmony, and rhythm being natural to us, (for it is evident that measures or metres are parts of rhythms, ) the earliest among mankind, making a gradual progress in these things Ritter well reads οὐχὶ μίμημα ᾖ μίμημα. " RHYTHM differs from METRE, inasmuch as RHYTHM is proportion, applied to any motion whatever ; METRE is proportion, applied to the motion of WORDS SPOKEN. Thus, in the drumming of a march, or the dancing of a hornpipe, there is rhythm, though no metre.; in Dryden's celebrated Ode there is METRE as well as RHYTHM, because the poet with the rhythm has associated certain words. And hence it follows, that, though ALL METRE is RHYTHM, yet ALL RHYTHM is NOT METRE." Harris's Philol. Inquiries, p. 67,—where it is also observed, very truly, that “no English word expresses rhythmus better than the word time." P. 69, note. Twining. from the beginning, produced poetry from extemporaneous efforts. 7. which was varied according to suitable manners. But poetry was divided according to appropriate manners. For men of a more venerable character imitated beautiful actions, and the actions of such men; but the more ignoble imitated the actions of depraved characters, first composing vituperative verses, in the same manner as the others composed hymns and encomiums. 8. No satirical poetry before Homer's Margites. Of the authors, therefore, before Homer, we cannot mention any poem of this kind; though it is probable that there were many such writers. But if we begin from Homer, there are such for instance as his Margites3, and some others, in which, as being suited, the measure is Iambic. Hence, also, the Iambic verse is now called, because in this metre they used to Iambize (i. e. defame) each other. 9. The old poets thence were either heroic or Iambic. Of ancient poets, likewise, some composed heroic poems, and others Iambic. But as Homer was the greatest of poets on serious subjects, (and this not only because he alone imitated well, but also because he made dramatic imitations,) thus too he first demonstrated the figures of comedy, not dramatically exhibiting invective, but ridicule. For the Margites bears the same analogy to comedy, as the Iliad and Odyssey to tragedy. 10. Progress of the two styles, upon the introduction of tragedy and comedy. But when tragedy and comedy had appeared, those poets who were naturally impelled to each kind of poetry, some, instead of writing Iambics, became comic poets, but others, instead of [writing] epic poems, became the authors of tragedies, because these forms [of poetry] are greater and more esteemed than those. 11. To consider, therefore, whether tragedy is now perfect in its species or not, regarded as well with reference to itself as to the theatres, is the business of another treatise. 12. Both at first extemporaneous. Both tragedy and comedy, therefore, at first originated from extemporaneous efforts. The character of the hero is handed down to us thus: Τὸν ὃ οὔτ᾽ ἂν σκαπτῆρα θεοὶ θέσαν οὗτ᾽ ἀροτῆρα οὔτ᾽ ἄλλως τε σοφόν, πάσης δ᾽ ἡμάρτανε τέχνης. And again, πολλ᾽ ἡπίστατο ἐργα, κακως ὁ ἠπίστατο πάντα. A character not unlike Sir Abel Handy in Morton's "School of Reform." And tragedy, indeed, originated from those who led the dithyramb, but comedy from those who sung the Phallic verses, which even now in many cities remain in use; and it gradually increased as obvious improvements became known. And tragedy, having experienced many changes, rested when it had ar- rived at its proper nature. 13. Aeschylus improves tragedy, and then Sophocles. Aeschylus, also, first increased the number of players from one to two4, abridged the functions of the chorus, and made one of the players act the chief part5. But Sophocles introduced three players into the scene, and added scenic painting. 14. It becomes elevated in style, Further still, the magnitude [of tragedy increased] from small fables and ridiculous diction, in consequence of having been changed from satyric composition, it was late before it acquired dignity. The metre also of tragedy, from tetrameter, became Iambic (for at first they used tetrameter in tragedy, because poetry was then satyrical, and more adapted to the dance, but dialogue being adopted, nature herself discovered a suitable metre; for the Iambic measure is most of all adapted to conversation. Suitableness of the Iambic metre. And as an evidence of this, we most frequently speak in Iambics in familiar discourse with each other; but we seldom speak in hexameters, and then only when we depart from that harmony which is adapted to conversation). Again, tragedy is said to have been further adorned, with a multitude of episodes, and 15. other particulars. Let, therefore, thus much said suffice concerning these things; for it would perhaps be a great toil to discuss every particular. See my introduction to Aeschylus, p. vii. ed. Bohn. See ibid. note. Satyric, from the share which those fantastic beings called Satyrs, the companions and play-fellows of Bacchus, had in the earliest Tragedy, of which they formed the chorus. Joking and dancing were essential attributes of these rustic semi-deities. Hence the " ludicrous language” and the " dancing genius " of the old Tragedy, to which the TROCHAIC or running metre here spoken of was peculiarly adapted; being no other than this : " Jolly mortals, fill your glasses, noble deeds are done by wine." The reader will not confound satyric with satiric; nor the Greek satyric drama with the satire of Roman origin. See Harris's Phil. Arrang. p. 460, note. Dacier's Preface to Horace's Satires. The two words are of different derivations. Twining.

CHAP. V.]

Comedy an imitation of the ridiculous. BUT comedy is, as we have said, an imitation indeed of bad characters, yet it does not imitate them according to every vice, [but the ridiculous only;] since the ridiculous is a portion of turpitude. For the ridiculous is a certain error, and turpitude unattended with pain, and not destructive. Thus, for instance, a ridiculous face is something deformed and distorted without pain. 2. Its changes not known. The transitions, therefore, of tragedy, and the causes through which they are produced, are not unknown; but [those of] comedy have escaped our knowledge, because it was not at first an object of attention. For it was late before the magistrate gave a chorus to comedians 1; but prior to that period, the choruses were voluntary. Comedy, however, at length having obtained a certain form, those who are said to have been poets therein are commemorated. But it is unknown who it was 3. Nor the inventors of masks, etc. that introduced masks or prologues, or a multitude of players, and such like particulars. But Epicharmus and Phormis [were the first] to compose fables ; which, therefore, originated from Sicily. But among the Athenians, Crates, rejecting the Iambic This was almost equivalent to the modern "licensing" of plays, but was probably conducted with more taste and less absurdity. The poet was said αἰτεῖν χορὸν, the choragus δtoovat, and, if the piece was approved by the archon, the poet χορὸν ἔλαβε. See Ritter. But this is evidently corrupt. Ritter reads λόγους with mann, understanding those passages which the single actor either recited, or spoke in conversation with the chorus, opposing λόγοι ὑποκρ ιτ ω̃v to the ἄσματα κωμψδω̃ν. Iambic, i. e. satirical, and personally so, like the old Iambi, form, first began generally to compose speeches and fables. 4. Similarity of tragedy to epic poetry. The epopee, therefore, is an attendant on tragedy, [with the exception of the long metre 4,] since through this it is an imitation of worthy characters and actions. But it differs from tragedy in that it has a simple metre, and is a narration. It also [differs from it] in length. For tragedy is especially limited by one period of the sun, or admits but a small variation from this period; but the epopee is not defined within a certain time, and in this it differs; though at first they observed the same conduct with tragedy5, no less than epic poetry. 5. Their common and distinctive parts. With respect to the parts, however, [of the epopee and tragedy,] some are the same in both, but others are peculiar to tragedy. Hence he who knows what is a good or bad tragedy, knows the same in respect to epic poetry. For those things which the epopee possesses are to be found in tragedy; but every thing which tragedy contains is not in the epopee.

CONCERNING, therefore, imitative poetry in hexameters, and comedy, we shall speak hereafter. Let us now, however, speak concerning tragedy, assuming the definition of its essence as arising from what has been already said. 2. Definition of tragedy. Tragedy, therefore, is an imitation of a worthy or illustrious and perfect action, invectives, or lampoons, of which Aristotle speaks above, and from which the Iambic metre, which is not here alluded to, took its name. Twining. The words μέχρι μόνου μέτρου μεγάλου or μετὰ λόγου are thrown out by Ritter, and can have no meaning. On the question of the unities, see Twining, note 43, and my own note on Aesch. Eum. 235. possessing magnitude, in pleasing language, using separately the several species of imitation in its parts, by men acting, and not through narration, through pity and fear effecting a purification from such like passions1. 3. Its language, and different manners of imitation : But by pleasing language, I mean language possessing rhythm, harmony, and melody. And it uses separately the several species [of imitation], because some parts of the tragedy are alone perfected through metres, and others again through melody. 4. of scenery, melopoeia, and diction. But since they produce imitation by acting, in the first place the ornament of the spectacle will be a certain part of the tragedy, and in the next place the melopoeia and the diction. For by these they produce imitation. But I call diction, indeed, the composition of the metres; and melopoeia that, the whole power of which is apparent. 5. Two causes of action, sentiments, and moral habit. Since, however, [tragedy] is an imitation of action, and action is effected by certain agents, who must needs be persons of a certain description both as to their manners and their sentiments 4, (for from these we say that On the different interpretations of this difficult passage, see Twining, note 45. Ritter has followed the views of Lessing, in a note too full of argument to admit of condensation. Taylor's note is, as usual, a blundering Neo-Platonic attempt to reconcile the discrepancies of Aristotle and Plato. Decoration—literally, the decoration of the spectacle, or sight. In other places it is called the spectacle, or sight only —ὀψις. It comprehends scenery, dresses—the whole visible apparatus of the theatre. I do not know any single English word that answers fully to the Greek word. Twining. Melopoeia-literally, the making, or the composition, of the Music; as we use Epopoeia, or according to the French termination, which we have naturalized, Epopee, to signify epic poetry, or epic-making, in general.—I might have rendered it, at once, the MUSIC ; but that it would have appeared ridiculous to observe, of a word so familiar to us, even that " its meaning is obvious.” Twining. Dianoia—διανοία, in a general way, may be defined to be διεξοδικὴ του̃ λόγου ἐνεργεία, i. e. the discursive energy of son. But accurately speaking, it is that power of soul which reasons scientifically, deriving the principles of its reasoning from intellect. This latter definition, however, pertains to it, so far as it is not influenced in its reasouings by imagination and false opinions. TAYLOR, who objects to translating it" sentiments." I prefer following Twining, understanding sentiments not ex- actions derive their quality,) hence there are naturally two causes of actions, sentiments and moral habit, and through these actions all men obtain or fail of the object of their wishes. 6. Definition of fable, manners, and sentiment. But a fable, indeed, is an imitation of action; for I mean by a fable here, the composition of incidents. By manners, I mean those things according to which we say that agents are persons of a certain character; and by sentiment, that through which those who speak demonstrate any thing, or explain their meaning. 7. Six parts in every tragedy. It is necessary, therefore, that the parts of every tragedy should be six, from which the tragedy derives its quality. But these are, fable and manners, diction and sentiment, spectacle and melopoeia. Of these parts, however, two pertain to the means by which they imitate; one, to the manner; and three, to the objects5. And besides these, there are no other. 8. Some tragic poeta use all these parts. [Not a few [tragic poets], therefore, as I may say, use all these parts 6.] For every tragedy has scenic apparatus, manners, and a fable, and melody, and, in a similar manner, sentiment. But the greatest of these is the combination of the incidents. 9. Tragedy an imitation of actions. For tragedy is an imitation not of men, but of actions, [of life, and of felicity. For infelicity consists in action, and the end is a certain action, and not a quality7.] 10. What constitutes character. Manner not the end of tragedy; Men, however, are persons of a certain character, according to their manners ; but according to their actions, they are happy, or the contrary. The end of tragedy, therefore, does not consist in imitating manners, but it embraces manners on account of actions; so that the action and the fable are the end of tragedy. 11. but action. But the end is the greatest of all things. Moreover, without action, tragedy cannot exist; but it may exist without pressed, (γνώμαι, cf. Rhet. ii. 21,) but in the mind, and forming the mainspring of action. i. e. λέξις and μελοποιία are the means or instruments, ὄψις the manner, μυ̃θος, ἤθη, and διανοία, the objects. An evident interpolation. See Ritter. Thus stands the text, as freed from the additions made by Aldus, or other Italian critics. But the whole passage is probably an interpolation, as Ritter seems to have clearly shown. See his judicious note. manners. For most modern tragedies are without manners; and in short, many poets are such as among painters Zeuxis is when compared with Polygnotus. For Polygnotus, indeed, painted the manners of the good ; but the pictures of Zeuxis are without manners. 12. Proof. Further still, if any one place in a continued series moral speeches, sayings, and sentiments well framed, he will not produce that which is the work of tragedy; but that will be much more a tragedy, which uses these things as subordinate, and which contains a fable and combination of incidents. 13. The revolution and discovery cause most pleasure. Add to this, that the greatest parts by which fable allures the soul, are the revolutions and discoveries. Again, it is likewise an evidence of this, that those who attempt to write tragedies, acquire the power of expressing a thing in tragic diction and manners accurately, before they can compose a fable, as was the case with nearly all the first poets. 14. Proof from the difficulty of forming a plot. The fable, therefore, is the principal part, and as it were the soul of tragedy; but the manners are next in rank8. 15. Comparison of painting. [Just as in painting, if any one were to spread the most beautiful pigments on promiscuously, he would not please the view so much as by outlining an image with white colour only. Tragedy also is an imitation of action, and on this account, especially, [an imitation of agents. But the sentiments rank third. 16. Definition of sentiment. And by them [I mean] the power of explaining what is inherent in the subject, and adapted to it, which is the peculiar province of politics and rhetoric. For the ancient poets represent those whom they introduce as speaking politically; but those of the present day, rhetorically. 17. Of manners and of sentiment. But the manners are, whatever shows what the deliberate choice is. Hence those speeches are without manners, in which there is altogether The rest of this chapter is condemned by Ritter as an interpolation. The reader, here, must not think of our modern politics. -The political, or civil art, or science, was, in Aristotle's view, of wide extent and high importance. It comprehended ethics and eloquence, or the art of public speaking; every thing, in short, that concerned the well-being of a state.—See note 57. Twining. nothing that the speaker may choose or avoid 10. But sentiment is that through which they show that a certain thing is, or is not, or by which they universally enunciate something. 18. Of diction. And the fourth part of tragedy is diction. But I say, as was before observed, that diction is interpretation by the means of words, and which also has the same power in verse and prose. 19. Melopoeia and spectacle, the latter least pertaining to art. But of the remaining five, the melopoeia is the greatest of the embellishments. But the scenic decoration is alluring indeed; yet it is most inartificial, and is in the smallest degree akin to poetry. For the power of tragedy remains, even when unaccompanied with scenic apparatus and players. And further still, the art of the mechanic possesses more power in constructing the scenic apparatus than that of the poet.]

Definition of a plot. THESE things being defined, let us in the next place show what the combination of the incidents ought to be, since this is the first and greatest part of tragedy. 2. But it is granted to us, that tragedy is the imitation of a perfect and whole action, and of one which possesses a certain magnitude; for there may be a whole which has no magnitude. 3. Of a dramatic whole, and its parts. But a whole is that which has a beginning, middle, and end. And the beginning is that which necessarily is not itself posterior to another thing; but another thing is naturally expected to follow it. On the contrary, the end is that which is itself naturally adapted to be posterior to another thing, either from necessity, or for the most part; but after this there is nothing else. But the middle is that which is itself after another thing, and after which there is something else. Hence, it is necessary that those who compose fables properly, should neither These words are differently placed in some editions. begin them casually, nor end them casually, but should employ the above-mentioned forms [of beginning, middle, and end]. 4. Magnitude and order the constituents of beauty. Further still, since that which is beautiful, whether it be an animal, or any thing else which is composed from certain parts, ought not only to have these parts arranged, but a magnitude also which is not casual For the beautiful consists in magnitude and order. Hence, neither can any very small animal be beautiful; for the survey of it is confused, since it is effected in a time nearly insensible. Nor yet a very large animal; for it is not surveyed at once, but its subsistence as one and a whole escapes the view of the spectators; such as if, for instance, it should be an animal of ten thousand stadia in length. 5. Magnitude in reference to plots; Hence, as in bodies and in animals it is necessary there should be magnitude; but such as can easily be seen; thus also in fables, there should be length, but this such as can easily be remembered 1. The definition, however, of the length 6. 1stly. as to dramatic exhibitions. with reference to contests and the senses, does not fall under the consideration of art. For if it were requisite to perform a hundred tragedies, [as is said to have been the case more than once 3,] the performance ought to be regulated by a clepsydra. 7. 2ndly, as to intrinsic value. But the definition [of the length of the fable] according to the nature of the thing, is this, that the fable is always more beautiful the greater it is, if at the same time it is perspicuous. Simply defining the thing, however, we may say, [that a fable has an appropriate magnitude,] when the time of its duration is such as to render it probable that there can be a transition from prosperous to adverse, or from adverse to prosperous fortune, according to the necessary or probable order The unity here spoken of, it must be remembered, is not absolute and simple, but relative and compound, unity; a unity consisting of different parts, the relation of which to each other, and to the whole, is easily perceived at one view. On this depends the perception of beauty inform.—In objects too extended, you may be said to have parts, but no whole: in very minute objects the whole, but no parts. Twining. i. e. to its representation at the dramatic contests. These words are condemned by Ritter. of things as they take place. This is a sufficient definition of magnitude. of

On dramatic unity. THE fable, however, is one, not as some suppose, if one person is the subject of it; for many things which are infinite in kind happen [to one man], from a certain number of which no one event arises. Thus, also, there are many actions of one man, from which no one action is produced. 2. Previous misconceptions on the subject. Hence all those poets appear to have erred who have written the Heracleid, and Theseid, and such like poems. For they suppose that because Hercules was one person, it was fit that the fable should be one. 3. Conduct of Homer. Homer, however, as he excelled in other things, appears likewise to have seen this clearly, whether from art, or from nature. For in composing the Odyssey, he has not related every thing which happened to Ulysses; such as the being wounded in Parnassus 1, and pretending to be insane at .-the muster of the Greeks; one of which taking place, it was not necessary or probable that the other should happen; but he composed the Odyssey, as also his Iliad, upon one action. 4. Application of unity to the plot. It is requisite, therefore, that as in other imitative arts one imitation is the imitation of one thing, thus, This incident is, however, related, and at considerable length, in the xixth book of the Odyssey, (v. 563 of Pope's translation,) but digressively, and incidentally; it made no essential part of his general plan. Twining. A ridiculous story.—“ To avoid going to the Trojan war, Ulysses pretended to be mad; and, to prove his insanity, went to plough with an ox and a horse; but Palamedes, in order to detect him, laid his infant son, Telemachus, in the way of the plough; upon which Ulysses immediately stopped, and thereby proved himself to be in his right senses.”—(Hyginus, etc.) Twining. I follow Ritter's text. i. e. one imitative work. Thus one picture represents, or also, [in tragedy,] the fable, since it is an imitation of action, should be the imitation of one action, and of the whole of this, and that the parts of the transactions should be so arranged, that any one of them being transposed, or taken away, the whole would become different and changed. For that which when present or not present produces no sensible [difference], is not a part of the fable 5.

The poet does not relate real events, BUT it is evident from what has been said, that it is not the province of a poet to relate things which have happened, but such as might have happened, and such things as are possible according to probability, or which would necessarily have happened1. 2. for then he would clash with the historian. For an historian and a poet do not differ from each other, because the one writes in verse and the other in prose; for the history of Herodotus might be written in verse, and yet it would be no less a history should represent, but one thing;-a single object, or a single action. etc. So, every poem (the Orlando Furioso as much as the Iliad) is one imitation—one imitative work, and should imitate one action, in Aristotle's sense of unity, like the poems of Homer; not a number of actions unconnected with each other, or connected merely by their common relation to one person, as in the Theseids, etc., or to one time, as in the poem of Ariosto; or, by their resemblance merely, as in the Metamorphoses of Ovid. Twining. " The painter will not inquire what things'may be admitted without much censure. He will not think it enough to show that they may be there, he will show that they must be there; that their absence would render his picture maimed and defective.—They should make a part of that whole which would be imperfect without them.”—Sir J. Reynolds, Disc. on Painting, p. 106. Twining. Cf. Sheridan's Critic, vii. 1. " What the plague! a play is not to show occurrences that happen every day, but things just so strange, that though they never did, they might happen." with metre, than without metre. But they differ in this, that the one speaks of things which have happened, and the other of such as might have happened. 3. Hence poetry is universal and philosophic. 4. Definition of universal and particular. Hence, poetry is more philosophic, and more deserving of attention, than history. For poetry speaks more of universals, but history of particulars. But universal consists indeed in relating or performing certain things which happen to a man of a certain description, either probably or necessarily, [to which the aim of poetry is directed in giving names 2;] but particular consists in narrating what, [for example,] 5. As is evident in comedy. Alci blades did, or what he suffered. in comedy, therefore, this is now become evident. For [comic poets] having composed a fable through things of a probable nature, they thus give whatever names they please to their characters, and do not, like Iambic poets, write poems about particular persons. 6. Real names retained in tragedy, But in tragedy they cling to real names. The cause, however, of this is that the possible is credible. Things, therefore, which have not yet been done, we do not yet believe to be possible; but it is evident that things which have been done are possible; for they would not have been done, if they were impossible. 7. but not real names only. Not, indeed, but that in some tragedies there are one or two of known names, and the rest are feigned; but in others there is no known name; as, for instance, in the Flower of Agatho. For in this tragedy, the things and the names are alike feigned, and yet it delights no less. 8. Nor must we entirely confine ourselves to traditional fables. Hence, one must not seek to adhere entirely to traditional fables, which are the subjects of tragedy. For it is ridiculous to make this the object of search, because even known subjects are known but to a few, though at the same time they Ritter well observes that the perspicuity of this otherwise clear passage is destroyed by this absurd interpolation. Thus nearly all the names in the tragedies of Terence and Plautus, thus Dromo and Sosia are applied to slaves, Pamphilus to a lover, Glycerium or Philumena to a lady, Pyrgopoli- Dices or Thraso to soldiers. Also the names in Petronius and Apuleius, as Pannychis, Meroe, Fotis, etc. So Ben Jonson has personified the virtues and vices in " Cynthia's Revels," and elsewhere. delight all men. 9. Hence the plot, rather than the metre, is the poet's object. From these things, therefore, it is evident that a poet ought rather to be the author of fables than of metres, inasmuch as he is a poet from imitation, and he imitates actions. Hence, though it should happen that he relates things which have happened, he is no less a poet. For nothing hinders but that some actions which have happened, are such as might both probably and possibly have happened, and by [the narration of] such he is a poet.

10. Objection to, and definition of, episodic fables. But of simple plots and actions, the episodic are the worst. But I call the plot episodic, in which it is neither probable nor necessary that the episodes follow each other. Such plots, however, are com- It may appear to the reader to be a strange observation, that "some true events MAY be probable.” But he will recollect what sort of events, and what sort of probability, Aristotle here speaks of: i. e. of extraordinary events, such as Poetry requires, and of that more strict and perfect probability, that closer connexion and visible dependence of circumstances, which are always required from the poet, though in such events, not often to be found in fact, and real life, and therefore not expected from the historian. Aristotle alludes to these two lines of Agatho :

Even this, it may be said, is probable, That many things improbable should happen, In human life.- See Rhet. ii. 24. 10, and Bayle's Dict. Art. AGATHON, note [F], who mentions a similar maxim of St. Bernard's: " Ordinatissimum est, minus interdum ordinate fieri." " 11 est tout à fait de l'ordre, que de tems en tems il se fasse quelque chose contre l'ordre." This general, and, if I may call it so, possible sort of probability, may be termed, the probability of romance; and these lines of Agatho furnish a good apologetical motto for the novel writer. It might be prefixed, perhaps, without impropriety, even to the best productions of the kind-to a CLARISSA, or a CECILIA. Nothing is so commonly complained of in such works, as their improbability; and often, no doubt, the complaint is well-founded : often, however, the criticism means nothing more, than that the events are uncommon, and proves nothing more, than the want of fancy, and an extended view of human life, in the reader. If the events were not uncommon, where would the book find readers ? Twining. posed by bad poets indeed, through their own want of ability5; but by good poets, on account of the players 6. For, introducing [dramatic] contests, and extending the plot beyond its capabilities, they are frequently compelled to distort the connexion of the parts. 11. Definition of the terrible. But, since tragedy is not only an imitation of a perfect action, but also of actions which are terrible and piteous, and actions principally become such, [and in a greater degree, when they happen contrary to opinion, 7] on account of each other. 12. Of chance circumstances. * * * * For thus they will possess more of the marvellous, than if they happened from chance and fortune; since, also, of things which are from fortune, those appear to be most admirable, which seem to happen as it were by design. Thus the statue of Mityus at Argos, killed him who was the cause of the death of Mityus by falling as he was surveying it. For such events as these seem not to take place casually. Hence it is necessary that fables of this kind should be more beautiful.

Of fables simple and complex. OF fables, however, some are simple, and others complex; for so also are the actions of which fables are the imitations. 2. But I call the action simple, Such is the sense of αὐτοὺς=“propter ingenii tarditatem." Ritter. i. e. that the play may not show off only one actor. Thus a modern dramatist writes a play for Charles Kean, and puts in a few episodes for Buckstone or Keeley. It is clear from this that plays were often written to suit particular performers. These words are condemned as interpolated. But the apodosis to the whole sentence is wanting, and I have therefore marked the lacuna with Ritter. A similar story of the death of a cruel beloved being killed by the falling of a statue of Cupid, is told by Theocritus, Id. xxiii. sub fin. from which taking place, as it has been defined, with continuity and unity, there is a transition without either revolution or discovery; but complex, from which there is a transition, together with discovery, or revolution, or both. 3. It is necessary, however, that these should be effected from the composition itself of the fable, so that from what has formerly happened it may come to pass that the same things take place either necessarily or probably. For it makes a great difference whether these things are effected on account of these, or after these.

Definition of revolution ; Now, revolution is a mutation, as has been stated, of actions into a contrary condition; and this, as we say, according to the probable, or the necessary. Thus in the " Œdipus, 1” the messenger who comes with an intention of delighting Oedipus and liberating him from his fear respecting his mother, when he makes himself known, produces a contrary effect. Thus, too, in the “Lynceus," he indeed is introduced as one who is to die, and Danaus follows with an intention of killing him ; but it happens from the course of incidents, that Lynceus is saved, and Danaus is slain. 2. of discovery. And discovery is, as the name signifies, a change from ignorance to knowledge, or into the friendship or hatred of those who are destined to prosperous or adverse fortune. The discovery, however, is most beautiful, when at the same time there are, as in the “Œdipus, revolutions. 3. Other recognitions. There are, therefore, other discoveries also. For sometimes it happens, as has been before observed, that there are discoveries of things inanimate3, and casual; or if some one has Cf. Soph. Œd. Tyr. 1014, sqq. Dind. Such is the discovery of Joseph by his brethren, Gen. xlv. -the most beautiful and affecting example that can be given. Twining. I do not understand Aristotle to be here speaking of such performed, or has not performed, a thing, there is a recognition of it; but the discovery which especially pertains to the fable and the action is that before mentioned. 4. Feelings excited by revolution and discovery. For a discovery and revolution of this kind will excite either pity or fear; and tragedy is supposed to be an imitation of such actions [as excite fear and pity]. Again, it will happen that infelicity and felicity will be in such like discoveries. 5. Different discoveries. But since discovery is a discovery of certain persons, some [discoveries] are of one person only with reference to another, when it is evident who the other person is, but sometimes it is necessary to discover both persons. Thus Iphigenia was recognised by Orestes through the sending an epistle4; but another discovery was requisite to his being known by Iphigenia. 6. The third requisite is pathos or disaster. [Two parts of the fable, therefore, viz. revolution and discovery, are conversant with these things; but the third part is pathos. And of these, revolution and discovery have been already discussed. Pathos, however, is an action destructive, or lamentable ; such as death when it is obvious, grievous pains, wounds, and such like particulars 6.]

The parts of tragedy according to quantity. [BUT we have before spoken of the parts of tragedy which are requisite to constitute its quality. The parts of tragedy, however, according to quantity, and discoveries of " inanimate things " (rings, bracelets, etc.) as are the means of bringing about the true discovery-that of the persons. For, in what follows, it is implied that these " other sorts of discovery " produce neither terror nor pity, neither happiness nor unhappiness ; which can by no means be said of such discoveries as are instrumental to the personal discovery, and, through that, to the catastrophe of the piece. Of these, he treats afterwards.—Dacier, I think, has mistaken this. Twining. Cf. Eur. Iph. Taur. 759. 92. Ibid. 811. 26. The whole of this paragraph is condemned by Ritter. into which it is separately divided, are as follow: prologue, episode, exode, and chorus, of the parts of which one is the parodos, but the other is the stasimon. These [five] parts, therefore, are common to all [tragedies]; but the peculiar parts are [the songs] from the scene and the kommoi. 2. Prologue. And the prologue, indeed, is the whole part of the tragedy, prior to the entrance of the chorus. Episode. The episode is the whole part of the tragedy between two complete odes of the chorus. Exode. The exode is the whole part of the tragedy, after which there is no further melody of the chorus. Parodos. And of the chorus, the parodos, indeed, is the first speech of the whole chorus; Stasimon but the stasimon is the melody of the chorus, without anapaest and trochee: Commos. and the commos is the common lamentation of the chorus and from the scene. But 3. we have before shown what the parts of tragedy are which must necessarily be used; but the parts of it according to quantity, and into which it is separately divided, are these 7.] Prologue—This may be compared to our first act. See note 40. Twining. Episode—i. e. a part introduced, inserted, etc., as all the dialogue was, originally, between the choral odes. Twining. Exode—i. e. the going out, or exit; the concluding act, as we should term it. The Greek tragedies never finished with a choral ode. Twining. Parode—i. e. entry of the chorus upon the stage: and hence the term was applied to what they first sung, upon their entry. Twining. Stasimon—i. e. stable; because, as it is explained, these odes were sung by the choral troop when fixed on the stage, and at rest: whereas the parode is said to have been sung as they came on. Hence, the trochaic and anapaestic measures, being lively and full of motion, were adapted to the parode, but not to the stasimon. Twining. From a verb signifying to beat or strike; alluding to the gestures of violent grief. Ritter, who has illustrated this whole chapter with great learning and taste, allows its utility, but doubts that it is the work of Aristotle. The reader will find his remarks on the different parts of tragedy very valuable.

IN the next place we must show, as consequent to what has been said, what those who compose fables ought to aim at, and beware of, and whence the purpose of tragedy is effected. 2. Tragedy should not be simple. Proper vicissitudes. Since, therefore, it is necessary that the composition of the most beautiful tragedy should not be simple, but complex, and that it should be imitative of fearful and piteous actions—(for this is the peculiarity of such imitation) -in the first place it is evident, that it is not proper that worthy men should be represented as changed from prosperity to adversity, (for this is neither a subject of terror nor commiseration, but is impious,) nor should depraved characters [be represented as changed] from adversity to prosperity; for this is the most foreign from tragedy of all things, since it possesses nothing which is proper; for it neither appeals to moral sense, nor is piteous, nor fearful. Nor, again, must a very depraved man be represented as having fallen from prosperity into adversity. For such a composition will indeed possess moral tendency, but not pity or fear. For the one is conversant with a character which does not deserve to be unfortunate; but the other, with a character similar [to one's own]. [And pity, indeed, is excited for one who does not deserve to be unfortunate; but fear, for one who resembles oneself2] ; so that the event will neither appear to be commiserable, nor terrible. 3. The character of the hero must be of a moderate excellence or depravity. There remains therefore the character between these. But a character of this kind is one, who neither excels in virtue and justice, nor is changed through vice and depravity, into misfortune, from a state of Ritter considers τὸ φιλάνθρωρπον as a feeling bordering on ἓλεος, " id quidem quod prope ad miserationem accedit." It probably is best expressed by our " humanity," considered in both its senses. Ritter condemns these words as a marginal annotation. great renown and prosperity, but has experienced 4. Simple fable preferable to twofold. this change through some [human] error; such as Œdipus and Thyestes; and other illustrious men of this kind. Hence it is necessary that a plot which is well constructed, should be rather single than twofold, (though some say it should be the latter,) and that the change should not be into prosperity from adversity, but on the contrary into adversity from prosperity, not through depravity, but through some great error, either of such a character [as we have mentioned], or better rather than worse. But the proof of this is what has taken place. 5. Proof from the popular subjects of tragedy. For of old the poets adopted any casual fables; but now the most beautiful tragedies are composed about a few families; as for instance, about Alcmaeon, Oedipus, Orestes, Meleager, Thyestes, and Telephus4, and such other persons as happen either to have suffered or done things of a dreadful nature. The tragedy, therefore, which is most beautiful according to art, is of this construction. Hence they erroneously blame 6. An unhappy end most suitable to tragedy, Euripides, who accuse him of having done this in his tragedies, and for making many of them terminate in misfortune. For this method, as we have said, is right; of which this is the greatest evidence, that in the scenes, and contests of the players, simple fables which terminate unhappily appear to be most tragical, if they are properly acted. And Euripides, though he does not manage other things well, yet appears to be the most tragic of poets5. 7. but least popular, The fable, however, ranks in the second place, though by some it is said to be the first composition, which has a twofold construction, such as the Odyssey, and which terminates in a contrary fortune, both to the better What is here meant by a single fable, will appear presently from the account of its opposite-the double fable. It must not be confounded with the simple fable, though in the original both are expressed by the same word. The simple fable is only a fable without revolution, or discovery. Twining. The same remark applies to the French tragic stage. But below, xv. 5. and Eth. iii. 1, Euripides is justly charged with the improper introduction of comic characters and language. The praise applies only to the catastrophe. and worse characters. It appears, however, to rank in the first place, through the imbecility of the spectators6. 8. and the happy conclusion preferred. For the poets, in composing their plots, accommodate themselves to the wish of the spectators. This pleasure, however, is not [properly] derived from tragedy, but is rather suited to comedy. For there, though the greatest enemies be introduced, as Orestes and AEgisthus, yet in the end they depart friends, and no one falls by the hand of the other7.

Terror and pity, how produced. TERROR and pity, therefore, may be produced from the sight1. But they may also arise from the com- That weakness which cannot bear strong emotions, even from fictitious distress. I have known those who could not look at that admirable picture, the Ugolino of Sir Jos. Reynolds.—To some minds, every thing that is not cheerful is shocking.—But, might not the preference here attributed to weakness, be attributed to better causes-the gratification of philanthropy, the love of justice, order, etc. ?-the same causes which, just before, induced Aristotle himself to condemn, as shocking and disgusting, those fables which involve the virtuous in calamity. Twining. Modern audiences partake of this weakness. Thus, the catastrophe of " Measure for Measure," " Cymbeline," or " Winter's Tale," are more satisfactory than those of " Hamlet" or "Pizarro." Knowles's "Hunchback" is a happy specimen of άναγνώρισις, leading to a happy termination, while his " Wife " displeases. Marston's " Heart and the World," weak as the play is, is more agreeable than his " Patrician's Daughter." But such plays are merely serious comedies. This is excellently applicable to the plays of Vanbrugh and Gibber. Compare the conclusions of the " Relapse," (Sir Tunbelly excepted,) the Provoked Wife," etc. As to the άποθνήσκει ούδεὶςύπ' ούδενος, it is very descriptive of the duels in a modern or ancient comedy, as in " London Assurance," or in the " Beaux Stratagem," according to Scrub's description. See a very pleasant paper of Addison's on this subject, " Spectator," No. 42. We know the effect of the skull and bination of the incidents, which is preferable, and the province of a better poet. For it is necessary that the fable should be so composed, that he who hears the things which are transacted, may be seized with horror, and feel pity, from the events, without the assistance of the sight; and in this manner any one who hears the fable of Oedipus is affected. But to effect this through spectacle is more inartificial, and requires great expense 2. 2. The monstrous not tragic. But they who produce not the terrible, but the monstrous alone, through scenic representation, have nothing in common with tragedy. For it is not proper to expect every kind of pleasure from tragedy, but that which is appropriate. 3. The proper pleasure to be derived from tragedy. Since, however, it is necessary that the poet should procure pleasure from pity and fear through imitation, it is evident that this must be effected by the circumstances. Let us, then, ascertain what kind of events appear to be dreadful or lamentable. 4. What actions are dreadful or lamentable.But it is necessary that actions of this kind should either be those of friends towards each other, or of enemies, or of neither. If, therefore, an enemy kills an enemy, he does not show any thing which is an object of pity, neither while he does the deed, nor when he is about to do it, except what arises from the deed itself. And this will be the case, when one of those who are neither friends nor enemies do the same. But when these things happen in friendships3, as when a brother kills a brother, or a son his father, or a mother her son, or a son his mother, or intends to do it, or does any thing else of the like kind-such subjects are to be sought for. One must not, therefore, [completely] alter the received fables. 5. Further precepts respecting fable. I mean, for instance, such as the fable of Clytemnestra being slain by Orestes, and of black hangings in the " Fair Penitent," the scaffold in " Venice Preserved," the tomb in " Romeo and Juliet," etc. Twining. But Ritter understands ὃψις to mean the countenance of the actor. " Χορηγίαν generatim pro pretioso et operoso apparatu et apte did posse et eo intellectu h. 1. accipiendam esse manifesturn est." Ritter. As in the " Fatal Curiosity " of Lillo. Twining. Eriphyle by Alcmaeon. But it is necessary that the poet should invent the plot, and use in a becoming manner those fables which are handed down. What, however, we mean by [using fables] in a becoming manner, let us explain more clearly. 6. Examples. Now, the action may take place in such a way as the ancients have represented it, viz. knowingly with intent; as Euripides represents Medea killing her children. Men may also do an action, who are ignorant of, and afterwards discover their connexion [with, the injured party,] as in the " Oedipus" of Sophocles. This, therefore, is extraneous to the drama4, but is in the tragedy itself; as in the " Alcmaeon " of Astydamas, or Telegonus in the " Ulysses Wounded 5." 7. Further still, besides these there is a third mode, when some one is about to perpetrate, through ignorance, an atrocious deed, but makes the discovery before he does it6. And besides these there is no other mode. For it is necessary to act, or not; and that knowing, or not knowing. But of these, to intend to perpetrate the deed knowingly, and not to perpetrate it, is the worst; for it is wicked and not tragical; because it is void of pathos. [Hence, no poet introduces a character of this kind except rarely; as in the "Antigone," in which Haemon [endeavours to kill his father] Creon, [but does not The murder of Laius by Œdipus, his son, is supposed to have happened a considerable time before the beginning of the action. Twining. Of these two dramas nothing more is known than the little that Aristotle here tells us. In the first, the poet adhered so far to history, as to make Alcmaeon kill his mother Eriphyle, but with the improvement, (according to Aristotle's idea,) of making him do it ignorantly. The story of Telegonua is, that he was a son of Ulysses by Circe; was sent by her in quest of his father, whom he wounded, without knowing him, in a skirmish relative to some sheep, that he attempted to carry off from the island of Ithaca. It is somewhat singular, that the wound is said to have been given with a kind of Otaheite spear, headed with a sharp fish-bone. See Pope's Odyssey, xi. 167, and the note. Twining. Thus in Talfourd's " Ion," Ion discovers Adrastus to be his father, just as he is on the point of murdering him. effect his purpose.] 7 ] For the action here ranks in the second place. But it is better to perpetrate the 8. deed ignorantly, and having perpetrated to discover; for then it is not attended with wickedness, and the discovery excites horror. The last mode, however, 9. is the best; I mean, as in the "Cresphontes, in which Merope is about to kill her son, but does not, in consequence of discovering that he was her son. Thus, too, in the " Iphigenia in Tauris," in which the sister is going to kill the brother, [but recognises him;] and in the " Helle," the son is about to betray his mother, but is prevented by recognising her. Hence, as has been formerly observed, trage- 10. dies are not conversant with many families; for poets were enabled to discover incident of this kind in fables, not from art, but from fortune 8. They 11. were compelled, therefore, to direct their attention to those families in which calamities of this kind happened.

And thus we have spoken sufficiently concerning the combination of the incidents, and have shown what kind of fables ought to be employed.

Requisites of manners, that they express προαίρε­σις. WITH respect to manners, however, there are four things to which one ought to direct attention: one, indeed, and the first, that they be good. But the tragedy will indeed possess manners, if, as was said, the words or the action render any deliberate intention apparent; containing good manners 1, if the deliberate intention is good. But manners are to be found in each genus; for both a woman and a slave may be good; though perhaps of these, the one is Ritter condemns this passage. See also Donaldson, Introduction to the Antigone, p. xl. i. e. to history or tradition. - The interpolation φαυ̃λον μὲν έὰν φαύληἢ is rightly thrown out by Ritter. less good2, and the other is wholly bad3. 2. that they be adapted to the persons, In the second place, the manners must be adapted to the persons. For there are manners which are characterized by fortitude, but it is not suited to a woman to be either brave or terrible. 3. similar, In the third place, the manners must be similar. For this, as was before observed, differs from making the manners to be good and adapted4. 4. and consistent. In the fourth place, they must be uniform ; for if he is anomalous, who exhibits the imitation, and expresses such like manners, at the same time it is necessary that he should be uniformly unequal. 5. Depraved manners unnecessary. The example, however, of depraved manners is indeed not necessary; such, for instance, as that of Menelaus in the " Orestes," but an example of unbecoming and unappropriate manners is, the lamentation of Ulysses in the tragedy of " Scylla, 5" and the speech of Menalippe; and the example of anomalous manners in the Iphigenia in AnUs. For Iphigenia This is very Euripidean gallantry. Compare Aristoph. 546, sqq. This is observed, to show the consistence of this first precept with the next. The manners must be drawn as good as may be, consistently with the observance of propriety, with respect to the general character of different sexes, ages, conditions, etc. It might have been objected—" You say the character must be good. But suppose the poet has to represent, for instance, a slave ?-the character of slaves in general is notoriously bad"—The answer is,—any thing may be good in its kind. Twining. This is very trivial, compared with Horace's description of the manners suited to different characters. Cf. Ars Poet. 114, sqq. Of the Scylla nothing is known.-Some fragments remain of " Menalippe the Wise," (for this was the title,) a tragedy of Euripides, the subject of which is a curiosity. Menalippe was delivered of two children, the fruits of a stolen amour with Neptune. To conceal her shame, she hid them in her father's cow-house; where he found them, and being less of a philosopher than his daughter, took them for a monstrous production of some of his cows, and ordered them to be burned. His daughter, in order to save them, without exposing herself, enters into a long physical argument, upon the principles of Anaxagoras, to cure her father of his unphilosophical prejudices about monsters and portentous births, and to convince him that these infants might be the natural children of his cows. Twining. supplicating does not at all resemble the Iphigenia in the latter part of the tragedy. 6. Necessity and probability of action. It is requisite, however, in the manners as well as in the combination of the incidents, always to investigate, either the necessary or the probable; so that such a person should say or do such things, either necessarily or probably ; and that it be necessary or probable, that this thing should be done after that. 7. Solution ought to be from the fable, not the machinery. It is evident, therefore, that the solutions of fables ought to happen from the fable itself, and not as in the "Medea, 6" from the machinery, and in the tragedy called the "Iliad," from the particulars respecting the sailing away [from Troy 7]. But we must employ machinery in things which are external to the drama, which either happened before, and which it is not possible for men to know, or which happened afterwards, and require to be previously foretold and announced. For we ascribe to the gods the power of seeing all things, but we do not admit the introduction of any thing absurd in the incidents 8, but if it is introduced it must be external to the tragedy; as in the " Oedipus" of Sophocles. 8. Comparison with painting. Since, however, tragedy is an imitation of better things, it is necessary that we should imitate good painters. For these, in giving an appropriate form to the image, depict the similitude, but increase the beauty 9. Thus, also, it is requisite that the poet, in imitating the wrathful and the indolent, and those who are similarly affected in their manners, should form an example of equity, or asperity: such as Agatho and Homer have represented Achilles. 9. These things, indeed, it is necesary to observe; and besides these, such perceptions of the senses as are attendant upon poetry, besides Of Euripides. Medea is carried off, at the end of the tragedy, in a chariot drawn by flying dragons. Twining. Pope's Iliad, ii. 189, etc. Twining. By incident. of the fable, Aristotle here plainly means all those actions or events which are essential parts of the subject or story, whether previous to the action, and necessary to be known, or included in it, and actually represented in the drama. This seems intended to explain his third precept, of resemblance in the manners; to reconcile it with his first, and to show what sort of likeness the nature of tragic imitation requires. Twining. the necessary ones.10 For in these, errors are fre-. quently committed. But concerning these things, enough has been said in the treatises already published.

On the different recognitions. [1 WHAT discovery, however, is, has been before stated. But with respect to the species of recognition, the first indeed is the most inartificial, and that which most poets use through being at a loss, and is effected 2. through signs. But of these, some are natural, such as the lance with which the earth-born race are marked," or the stars [on the bodies of the sons] in the " Thyestes" of Carcinus. Others are adventitious, and of these some are in the body, as scars; but others are external, such as necklaces; and such as [the discovery] through a small boat, in the 3. Their use. " Tyro3." These signs also may be used in a better or worse manner. Thus Ulysses, through his scar, is in one way known by his nurse, and in another by the swineherds. For the discoveries which are for the sake of credibility, are more in artificial, and all i. e. to the sight, and the hearing; in other words, to actual representation. The reader, who recollects the conclusion of Sect. 14, where the author took a formal leave of the "fable and its requisites," and proceeded to the second essential part of tragedy, the manners, will hardly be of Dacier's opinion, who contends that this section is rightly placed. His reasons are perfectly unsatisfactory. Twining. I have enclosed it in brackets, with Ritter. The descendants of the original Thebans, who, according to the fabulous history, sprung from the earth when Cadmus sowed the dragon's teeth, etc.-This noble race are said to have been distinguished by the natural mark of a lance upon their bodies. Sophocles wrote two tragedies of this name, neither of them preserved.- The story of Tyro leads us to suppose, that Aristotle means the little boat, trough, or, as some render it, cradle, in which Tyro had exposed her children, on, or near, the river: the particular manner of the discovery it would be in vain to guess. See Pope's Odyssey, xix. v. 451, etc., and the note there, on v. 461, and xxi. 226. of them are of this kind; but those which are from revolution, as in the " Washing of Ulysses5," are better. 4. Invented recognitions. And those recognitions rank in the second place, which are invented by the poet, on which account they are inartificial. Thus Orestes in the " Iphigenia" discovers that he is Orestes6. For she indeed recognises her brother through a letter, but Orestes himself speaks what the poet designs, but not what the fable requires ; on which account it is near to the above-mentioned error; since he might have introduced some [of the real things as signs]. Thus, too; in the "Tereus" of Sophocles, the " voice of the shuttle" [produced a recognition7]. 5. By memory. But the third mode of discovery is through memory, from the sensible perception of something by sight, as in the " Cyprii " of Dicaeogenes; for on seeing the picture a certain person weeps. And in the " Tale of Alcinous;" for Ulysses, on hearing the lyrist, and recollecting the story, weeps ; whence also [all these] were recognised. 6. By reasoning. The fourth mode of discovery is derived from syllogism 8, as in the "Choephorae"—a person like The ancients distinguished the different parts of Homer's poems by different titles accommodated to the different subjects, or episodes; and, in referring to him, they made use of these, not of the division into books. Thus, the part of the xixth book of the Odyssey above referred to, was called The Washing. The Tale of Alcinous was another title, which will presently be mentioned. Twining. I follow Ritter, who supplies " to Iphigenia." The older editors interpolated the passage. See Ritter's note on the following passage. The whole disputation is " arguta et obscura," as the learned critic observes. Taylor's note is pre-eminently absurd. Tyrwhitt elegantly explains the passage thus: κερκίδος φωνὴ is a quotation from the play, and denotes the web itself ; by means of which Philomela explained to her sister Procne the injuries she had suffered from Tereus, since, her tongue being cut out, she could not speak. Cf. Ovid, Met. VI. 424 ; Hygin. Fab. 45, quoted by Ritter, whose note deserves the student's attention. As the web is said to speak, which describes, so the shield of Capaneus χρυσοι̃ς φωνει̃ γράμμασιν, Aesch. Sept. C. Th. 434. Occasioned by reasoning;—i. e. by reasoning, (or rather, inference, or conclusion,) in the person discovered. See the note.—It should be remembered, that Aristotle is not, in this chapter, inventing discoveries, nor enumerating all the kinds me is arrived-there is no person like me but Orestes,—Orestes, therefore, is arrived. Thus too in the "Iphigenia9" of Polyides the sophist. For it was probable that Orestes would syllogistically conclude, that because his sister had been immolated, it would likewise happen to him to be sacrificed. Thus also in the "Tydeus 10 " of Theodectes, a certain person comes to discover his son, and himself perishes 11. Another example also is in the "Phinidae." For the women, on seeing the place, inferred what their fate would be, viz. that they must needs perish in this place; for they were exposed in it from their infancy. 7. By the misconception of the audience. There is also a certain compound [discovery], which is produced from the false inference of the spectator, as in the " Ulysses the False Messenger." For he says, he should know the bow, which he had not seen; but the [audience], as if he must be known through this, on this account infer falsely. 8. The best is from the circumstances themselves. The best recognition, however, of all, is that which arises from the things themselves, astonishment being excited through probaBle circumstances; as in the "Œdipus" of Sophocles and the " Iphigenia;" (for it is probable that she would be willing to send letters;) since such things alone are without fictitious signs and necklaces12. But the recognitions which rank in the second place, are those which are derived from syllogism.] possible or practicable; but only classing and examining such as he found in use, or could recollect, in the tragedies and epic poems of his time. Twining. The subject appears to have been the same as that of the Iphigenia in Tauris of Euripides. We are to suppose, that Orestes was discovered to, his sister by this natural exclamation, at the moment when he was led to the altar of Diana to be sacrificed. Twining. Of this and the preceding tragedy, we know nothing but what we learn here : i. e. that in the one, a father, and in the other, the daughters of Phineus, were discovered, and, probably, saved, by those exclamations. Twining. See Ritter. Nothing of this play is known. All this passage is hopelessly corrupt.

Rules for realizing ideas.

IT is necessary, however, that the poet should form the plots, and elaborate his diction, in such a manner that he may as much as possible place the thing before his own eyes1. For thus, the poet perceiving most acutely, as if present with the transactions themselves, will discover what is becoming, and whatever is repugnant will be least concealed from his view. An evidence of this is the fault with which Carcinus is reproached. For Amphiaraus had left the temple, which was concealed from the spectator, who did not perceive it, and the piece was driven from the stage in consequence of the indignation of the spectators. For the poet as much as possible should co-operate with the gestures [of the actor] ; since those ,are naturally most adapted to persuade who are themselves under the influence or passion. 2. The poet should feel what he writes, Hence, also, he agitates others who is himself agitated 3, and he excites others to anger who is himself most truly enraged. Hence, poetry is the province either of one who is naturally clever, or of one who is insane4. i. e. place himself in the position of a spectator. Ritter observes that σόν σννίσταντα must be supplied. νκπίπτειν is properly used of the condemnation of a piece. See Hemst. on Lucian, Nigrin. § 8. In Demosth. de Coron. p. 315, ed. Reisk. it is applied to the actor. Cf. Bud. Corn. L. Gr. p. 536. There is a pun upon the double meaning of the verb. Amphiaraus went (έξέπεσε) away, and the piece was condemned (έξέπεσε) in consequence. But Twining, in a long and learned note, expresses his opinion that χεμρίνει may be used in its proper neuter sense, and that the meaning may be as follows : " The poet should work himself, as far as may be, into the passion he is to represent, by even assuming the countenance and the gestures which are its natural expressions. For they, of course, have most probability and truth in their imitation, who actually feel, in some degree, the passion : and no one expresses agitation of mind (χειμαίνει) so naturally, (άληθινώτατα,) as he who is really agitated, or expresses anger (χαλεπαίνει) so naturally, as he who is really angry (όργιζόμενος)." " In an enthusiasm allied to madness," is Twining's translation, which is all that Aristotle means to say, understanding " eos, qui animo commotiores sunt," as Ritter renders it. On the connexion between poetic enthusiasm and madness, cf. Plato, Ion. p. 145. C. etc. Phaedr. p. 344. B. Laem. with Clemens Alex. Strom. vi. p. 827. Theodoret. θεραπ. II. p. 25. Cicero de Div. I. 37. For of these characters, the one is easily fashioned, but the other is prone to ecstasy. 3. and should form a general sketch of his plot. It is likewise necessary that the poet should in a general way lay down the fables composed by others, and those which he composes himself, and afterwards introduce episodes and lengthen out [the play]. But I say that he should give a general sketch after this manner. Thus, for instance, in the " Iphigenia5," a certain virgin on the point of being sacrificed, and vanishing from the view of those who were to sacrifice her, and being brought to another country in which it was a law to sacrifice strangers to a certain goddess, she is appointed the priestess of these rites. Some time after, it happened that the brother of the priestess came to this place ; [but on what account ? Because some god had ordered him, for a certain reason which does not pertain to the general view of the tragedy,] to come thither, [but why he did so is foreign to the fable6]. The brother, therefore, coming, and being made captive, discovered [his sister], when he is going to be sacrificed ; whether, as Euripides says, [by an epistle,] or, as Poly ides feigns, speaking according to probability, because he said, it was not only requisite that the sister, but that he also should be sacrificed:—and hence safety arises. 4. The episodes must be suitable After these things, the poet having given names to the persons, should insert the episodes; and he must be careful that the episodes be appropriate; as that of the insanity through which Orestes was taken captive, and his being saved through expiation. 5. and short. In dramas, therefore, the episodes are short, but by these the epopee is lengthened. For the fable of the Odyssey is short, viz. a certain man wandering for many years, and persecuted by Neptune, and left alone. And besides this, his domestic affairs being so circumstanced, that his Eur. Iph. Taur. The passage is interpolated. See Ritter. wealth is consumed by suitors, and stratagems are plotted against his son. But driven by a tempest, he returns, and making himself known to certain persons, he attacks the suitors, and is himself saved, but destroys his enemies. This, therefore, is the peculiarity of the fable, but the rest is episode.

Complication and development. [IN every tragedy, however, there is a complication and development 1. And external circumstances indeed, and some of those that are internal, frequently form the complication; but the rest the development. I call, however, the complication, the whole of that which extends from the beginning to the last part, from which there is a transition to good fortune; but I call the development that part which extends from the beginning of the transition to the end. Thus in the Lynceus of Theodectes, the past transactions, and the capture of the son, are the complication ; but the part which extends from the charge of murder to the end, is the development. 2. Four species of tragedy; But of tragedy there are four species; for so many parts of it have also been enumerated. And one species is the complicated, of which the whole is revolution and discovery; another, the pathetic, such as the tragedies of Ajax and Ixion; another, the morale such as the Phthiotides and the Peleus; but the Literally, the tying and untying. With the French, Noeud and Denouement are convenient and established terms. I hope I shall be pardoned for avoiding our awkward expressions of the intrigue and unravelling of a plot, etc. I could find no terms less exceptionable than those I have used. Twining. i. e. in which the delineation of manners or character is predominant. Our language, I think, wants a word to express this sense of the Greek ήθικὸν, and the Latin, moratum. Mannered has, I believe, sometimes been used in this sense; but so seldom, as to sound awkwardly. We know nothing of the subjects here given as examples. Twining. fourth is another such as the " Phorcides " and the "Prometheus," and the tragedies which represent what passes in Hades. 3. all which the poet should understand. It is especially necessary, therefore, that the poet should endeavour to have all these species; or at least that he should have the greatest and most of them, especially since men of the present age calumniate the poets. For as there have been good poets in each part of tragedy, they now expect one poet to excel in all the parts. But it is right to call tragedy different and the same, though not perhaps with any reference to the fable; but this [may be the case with those] of which there is the same plot and solution. But many poets complicate well, and develope badly4. But both these should always be applauded 5. But it is necessary to recollect, as has been often observed, that we must not make tragedy an epic system. 4. On epic tragedy. Now, I call that tragedy an epic system, which consists of many fables; as if some one should compose a tragedy from the whole fable of the Iliad. For in the Iliad, on account of its length, the parts receive an appropriate magnitude. But in dramas, the effect pro- duced would be very contrary to expectation. 5. Illustrations. The truth of this is indicated by such as have represented [in one tragedy] the whole destruction of Troy, and not some part of it, as the "Niobe" or "Medea" of Euripides, and who have not acted like Aeschylus; for these have either been condemned, Aeschylus wrote a tragedy so named. It is difficult to imagine what he could make of these three curious personages, who were born old women, lived under ground, and had but one eye among them, which they used by turns; carrying it, I suppose, in a case, like a pair of spectacles. Such is the tale! Twining. No fault so common : see note 59. It was with the Greek tragedians, probably, as with Shakspeare.—" In many of his plays the latter part is evidently neglected. When he found himself near the end of his work, and in view of his reward, he shortened the labour, to snatch the profit. He therefore remits his efforts where he should most vigorously exert them, and his catastrophe is improbably produced, or imperfectly represented." Johnson's Pref. to Shakspeare. Twining. This passage is contradictory and unintelligible. See Ritter, who condemns the whole as spurious. or contend without success; since Agatho also failed in this alone. But in revolutions, and in simple actions, those poets admirably effect their aim. For this is tragical, and has a moral tendency. This, 6. however, takes place when a wise but a depraved man, such as Sisyphus, is deceived; and a brave but unjust man is vanquished. But this is probable, as Agatho says. For it is probable that many things may take place contrary to probability. It is neces- 7. The business of the chorus. sary likewise to conceive the chorus to be one of the players and a part of the whole, and that it co-operates with the players, not as in Euripides7, but as in Sophocles. But with other tragedians, the choral songs do not more belong to that fable, than to any other tragedy; on which account the chorus sing detached pieces, inserted at pleasure8, of which Agatho was the inventor. What difference, however, Actor;. partes chorus, officiumque virile Defendat: neu quid medios intercinat actus, Quod non proposito conducat et hoereat aptè. Her. A. P. 193. This expression does not, I think, necessarily imply any stronger censure of Euripides, than that the choral odes of his tragedies were, in general, more loosely connected with the subject, than those of Sophocles; which, on examination, would, I believe, be found true. For that this is the fault here meant, not the improper " choice of the persons who compose the chorus," as the ingenious translator of Euripides understands, is, I think, plain from what immediately follows; the connexion being this :—" Sophocles is, in this respect, most perfect; Euripides less so ; as to the others, their choral songs are totally foreign to the subject of their tragedies." See Potter's Euripides—Postscript to the Trojan Dames. Warton's Essay on the Genius, etc., of Pope, vol. i. p. 71. It is curious to trace the gradual extinction of the chorus. At first, it was all; then, relieved by the intermixture of dialogue, but still principal; then, subordinate to the dialogue; then, digressive, and ill connected with the piece; then, borrowed from other pieces at pleasure-and so on, to the fiddles and the act-tunes, at which Dacier is so angry. (See his note, p. 335.) The performers in the orchestra of a modern theatre are little, I believe, aware, that they occupy the place, and may consider themselves as the lineal descendants, of the ancient chorus. Orchestra (όρχηστρα) was the name of that part of the ancient theatre which was appropriated to the chorus. [Jul. Pollux, IV. p. 423.] does it make, to sing inserted pieces, or to adapt the diction of one drama to another, or the whole episode ?

Diction and sentiment. OF the other parts of tragedy enough has now been said. But it remains that we should speak concerning the diction and the sentiments. The particulars, therefore, respecting the sentiments, are unfolded in the treatise on Rhetoric, to which it more properly belongs. But those things pertain to the sentiments, which it is requisite to procure by a reasoning process. 2. Their parts. And the parts of these are, to demonstrate, to refute, and to excite the passions; such as pity, or fear, or anger, and such like; and besides these, to amplify and extenuate. 3. It is evident, however, that in things, also, it is requisite to derive what is useful from the same forms, when it is necessary to procure objects of pity, or things that are dreadful, or great, or probable. Except that there is this difference, that things in tragedy ought to be rendered apparent without teaching, but in an oration they are to be shown by the speaker, and in consequence of the speech. For what employment would there be for the orator, if the things should appear [of themselves] pleasing, and not through the speech ? 4. How far the poet should be acquainted with the different applications of diction. But of things pertaining to diction, there is one species of theory respecting the forms of speech 1, which it is the pro- What are we to understand by these σχήατα λέξεως?— The learned reader will immediately see, that, as Victorius has observed, they are not to be confounded with those σχήματα λέξεως, of which we hear so much from Cicero, Quintilian, Dion. Hal. etc.,-those " figurre verborum," which are opposed to the σχήματα διανοίας, the " figures mentis, sententiarum," etc. Indeed, no such division of σχήματα is, I believe, to be found in Aristotle. It seems to have been the invention of the later rhetoricians; and how little they were agreed, as to the number and the species of these σχήατα, the propriety of the division itself, and even the precise sense of the word σχη̃μα, may be seen in Quintilian ix. 1.—The σχήατα λέξεωςοf Aris- vince of the actor to know, and of him who is a master artist in this profession. Thus, or instance, [it is requisite he should know,] what a mandate is, what a prayer, narration, threats, interrogation and answer are, and whatever else there may be of this kind. 5. For from the knowledge or ignorance of these, the poetic art incurs no blame of any moment. For who would think that Homer errs in what he is reproved for by Protagoras? viz. that while he fancies he prays, he commands, when he says, " The wrath, O goddess, sing." For, says he, to order a thing to be done, or not to be done, is a mandate. Hence, this must be omitted as a theorem pertaining to another art, and not to poetry.

The parts of diction. [1 OF all diction, however, the following are the parts; viz. the letter, the syllable, the conjunction, the noun, the verb, the article, the case, and the sentence. The totle in this place, are plainly such as would have been denominated by later writers, σχήατα σιανοίας—figures of the thought or sense. Indeed we find them actually enumerated among the figures of that class. See Dionys. Halicarn. de Struct. etc. Sect. 8.—So Quintilian; " Figuras quoque mentis, quae σχήματασιανοίας dicuntur, res eadem recipit omnes, in quas nonnulli diviserunt species dictorum, (i. e. of jokes, bons mots). Nam et interrogamus, et dubitamus, et affirmamus, et minamur, et optamus." I see, therefore, not the least reason why the expression σχη̃μα λέξεως should not be rendered here exactly as in the other passages above referred to, "figura orationis "—form, or configuration, of speech. For λέξις, it must be observed, is here used, not in the particular sense of diction, or style and manner of expression, (as it is used Rhet. iii. 8,) but in the general sense of λόγος, speech, as we find it used in the beginning of the next chapter. This chapter is an evident interpolation, and the student will gain but little benefit from its perusal, as the matter of it is incorrect and ill arranged. See Ritter. letter, therefore, indeed, is an indivisible sound2; 2. Letters. yet not every such sound, but that from which an intelligible sound is adapted to be produced. For there are indivisible vocal sounds of brutes, no one of which 3. Their divisions, I call a letter3. But the parts of this indivisible sound are, vowel, semi vowel, and mute. And a vowel, indeed, is that which has an audible sound, without percussion 4; such as a and o. But a semivowel is that which has an audible sound, with percussion ; as s and r. And a mute is that which, even with the concurrence of the tongue, has of itself, indeed, no sound, but becomes audible in conjunction with things which have a certain sound; as g and d. 4. and differences. But these differ by the configurations of the mouth, in the parts [of the mouth 5] by density and tenuity of aspiration, by length and shortness; and further still, they differ by acuteness and gravity, and by the medium between both these; the theory respecting each of which pertains to the metrical art. 5. Syllables. But a syllable is a sound without signification, composed from a mute, and an element which has sound [i. e. from a vowel, or semivowel]. For g r without a is a syllable6, and also with a, as g r a. " Vocal sound," is Taylor's translation of φωνή. " Element of diction." Taylor. As Hermann has ventured to call προσβολὴ allisus, I trust I shall be excused for adopting Twining's quaint, but clear translation. He observes: " Literally, percussion, i. e. of the tongue against the palate, or teeth, the lips against the teeth, or against each other, and all the other modes of consonant articulation. See Hermes, iii. 2. p. 322, where they are called ' contacts.' Dacier makes sad confusion here, both in his version and his notes, by confounding the names of the consonants, when vowels are prefixed, or put after them, to make them separately pronouncible, (Te, eF, eL, etc.,) with their powers in composition—as elements of words. Thus, it is strictly true, that S and R have a sound, without the assistance of a vowel, merely by their mode of articulation. But D, or G, have no sound at all by themselves. The semivowels are 1, m, n, r, s. (Dion. Halicarn. De Struct. Orate sect. 14.) " i. e. the different organs of speech, from which letters are denominated nasal, dental, labial, etc. Taylor. Gr is an instance of a syllable composed of a mute and a semivowel; and gra of a syllable composed of a mute, a vowel, and a semivowel. Taylor. But see Ritter. The study, however, of the differences of these, pertains also to the metrical art. 6. Conjunction. But a conjunction is a sound void of signification, which neither impedes nor produces one significant sound adapted to be composed from many sounds, and which may be placed either at the beginning or the end of the period, unless something requires that it should be placed by itself at the beginning; such as μὲν, ἥτοι, δή. Or it is a sound without signification, composed from more sounds than one, but naturally adapted to produce one significant sound. 7. Article. An article is a sound without signification, which snows the beginning, or end, or distinction of a word 7; as τὸ φημί, and ὸro περί, and others of the like kind. Or it is a sound without signification, which neither impedes nor produces one significant sound naturally adapted to be composed from many sounds, both in the extremes and in the middle. 8. Noun. But a noun is a composite sound, significant without time, of which no part is of itself significant. For in double nouns, we do not use the parts as of themselves significant. Thus, in the word Theodorus, [though Theos signifies God and doron a gift,] yet doron signifies nothing. 9. Verb. A verb is a composite sound, significant with time, 01 which no part is of itself significant, in the same manner also as in nouns. For man or white does not signify time; but he walks, or he did walk, signify, the former indeed the present, and the latter the past time. 10. Case. But case pertains to noun or verb. And one case, indeed, [in nouns] signifies that something is said of this thing, or is attributed to this thing, and the like; This description is most obscure ; but the sense seems to be, that an article is a sound which of itself does not signify any thing definite, but merely serves to indicate a significant sound, before or after which it is placed, or which it distinguishes from other words. Taylor. Ritter denies that there is any sense at all, an opinion in which the reader will probably acquiesce. These only, in modern grammar, are called cases: in Aristotle, number, whether in noun or verb, and the tenses, and modes, (or moods,) of verbs, are comprehended under that term; because cases (πτωσεις—cases) are endings, terminations, inflections, etc., and, in the learned languages, all the above-mentioned differences of meaning are expressed by different terminations. The French use chute, the literal translation of casus, in the sense of termination—" La chute d'une periode," etc. And fall is used, in our poetical language, for a close, or cadence, in music. That strain again—it had a dying FALL. Merch. of Venice. And so Milton in Comus, v. 251. Twining. but another is that which pertains to one thing or many things; as men, or man. And another case pertains to acting9, such as what relates to interrogation or demand. 11. Sentence. For "did he walk" ? Or walk is a case of a verb according to these species. And a sentence is a composite significant sound, of which certain parts of themselves signify something; for not every sentence is composed from nouns and verbs, (since the definition of man10 is a sentence without a verb,) but there may be a sentence without verbs. A sentence, however, will always have some part significant; as in the sentence Cleon walks, the word Cleon is significant. 12. Its unity. But a sentence is one in a twofold respect; for it is either that which signifies one thing, or that which becomes one from many by conjunction 11. Thus the Iliad, indeed, is one by conjunction; but the definition of man is one, because it signifies one thing.]

The species of nouns. WITH respect to the species of a noun, one is simple; and I call the simple noun that which is not These modes are the same which he calls figures of speech, Sect. 23. Twining. The definition alluded to appears to be this, literally rendered: " A terrestrial animal with two feet" (ζῳον δίπουν). Twining. Compare περὶ 'Ερμην. cap. v. p. 38.—Analyt. Post. lib. ii. cap. 10. p. 469, E.—Metaphys. vii. 4. p. 910, D. (where he uses τῳ̃ συνέχει, as equivalent to συνδέσμῳ,) and viii. 931, C. Twining. Ritter remarks, that the compiler of this chapter did not understand the passages he copied. posed from significant parts; but another is twofold. And this either consists of that which is significant, and that which is without signification, or of both parts significant. A noun also may be triple and quadruple, as is the case with many of the nouns of the Megaliotae1; such as Hermocaïcoxanthus2. 2. Their divisions. But every noun is either proper or foreign3, or metaphorical, [or ornamental4,] or invented, or extended, or contracted, or altered. 3. Proper and foreign. But I call that a proper name, which is used by every one; and that a foreign name which is used by other nations. Hence it is evident that the same noun may be both foreign and proper, though not to the same people. For the word Σίγυνον6 is proper to the Cyprians,

I have read, in some ludicrous book, of a country that was " lost by the ignorance of geographers." This seems to have been the case of these Megaliotae, if such a people ever existed. They are no where recorded.—Dacier reads, μεγαρι­ ζόντων—" ceux qui di8ent de grandes choses:" and cites Hesychius—Μεγαρίζοντες—μεγαλα λέγοντες. But this is too distant from the present reading, Μεγαλιωτω̃ν. Winstanley's conjecture—μεγαλείων, ὼς, is somewhat nearer, and, in other respects, preferable: but it is, I think, a strong presumption against its truth, that Aristotle constantly uses οἲον, when he gives an instance; never, as far as I recollect, ὼς.

I have sometimes thought it not very improbable, that the passage might originally have stood thus: τω̃ν μέγαλΑ Διω- KON των : i. e. of those who affect, aim at, are fond of, grandeur and pomp of expression; who love hard words, as we say. Nothing more common than this sense of διωκιν. Twining.

Donaldson, Theatre of the Greeks, p. 27, pt. 2, adopts Tyrwhitt's conjecture, μεγαλείων ὼς, as being confirmed by Xenoph. Mem. II. 1, § 34, and renders it, " the bombastic expressions." Ritter prefers πολλαπλομεγάλωπος, as an example of this kind of word, but Donaldson seems right.

This is a noun composed from the names of the three rivers Hermus, Caicus, and Xanthus. Cf. Rhet. iii. 1and 2. Ritter condemns the addition of κόσμος, as it has no definition. Ritter remarks that γλώτται are both provincialisms and obsolete words. i. e. a spear. but foreign to us. 4. Metaphor. But a metaphor' is the transposition of a noun from its proper signification, either from the genus to the species, or from the species to the genus; or from species to species, or according to the analogous. 5. Different metaphors exemplified. I call, however, a transposition from genus to species, such as, Secure in yonder port ray vessel stands 8. For to be moored is a species of standing. But a transposition from species to genus is such as, Ten thousand valiant deeds Ulysses has achieved9. For ten thousand is a great number, and is now used instead of many. And a transposition from species to species is such as, The brazen falchion drew away his life. And, Cut by the ruthless sword 10. For here to draw away, is used instead of to cut; and to cut is used instead of to draw away: since both imply the taking something away. 6. Analogous. But I call it analogous, when the relation of the second term to the first is similar to that of the fourth to the third; for then the fourth is used instead of the second, or the second instead of the fourth. [And sometimes the proper term is added to the relative terms11.] I say, for instance, a cup has a similar relation to Bacchus that a shield has to Mars. Hence, a shield may be called the cup of Mars, and a cup the shield of Bacchus. Again, evening has the same relation to day that old age has to life. One may therefore say that Aristotle understands metaphor in a more extended sense than we do, for we only consider the third and fourth of the kinds enumerated by him, as metaphors. Our usurpation of the word was in vogue in Cicero's time. See de Orate III. 38, sqq. Ritter. Odyss. A. 185. Il. B. 272. This, and the next species only, answer to what we call metaphor—the metaphor founded on resemblance. The two first species belong to the trope denominated, since Aristotle's time, Synecdoche. Twining. This is perfectly out of place and useless. See Ritter. evening is the old age of day, and that old age is the evening of life ; or as Empedocles calls it, " The setting of life 12." 7. Means of employing them. In some instances, also, where there is no analogous name, this method may be no less similarly employed. Thus, to scatter grain is to sow; but there is no name for the scattering of light from the sun, and yet this has a similar relation to the sun that sowing has to grain. Hence, it is said, Sowing his god-created flame. This mode of metaphor may likewise be used differ- 8. This mode of metaphor may likewise be used differently, when, calling a thing by a foreign name, something belonging to it is denied of it; as if one should call a shield not the cup of Mars, but the wineless cup. 9. Invented words. But an invented noun is that, in short, which, not being adopted by others, is introduced by the poet himself. For it appears that there are certain nouns of this kind ; as substituting ἒρνυγες instead of κέρατα for horns 13, and calling a priest άρητήρ14 instead of ίερεύς. 10. tended and con tracted. And a word is extended or contracted, partly by using a vowel longer than the proper one, or by inserting a syllable ; and partly by taking something away from it. An extended noun, indeed, is such as πόληοςfor πόλεως, and πηληϊάδεω for πηλείδου ; but the contracted, such as κρῖ, and δω̃ 15; and, μία γίνεται ἀμφοτέρων ὄψ16. —The sight of both is one.

11. And a word is changed when the poet leaves part of it, and invents part; as, Changed.

Thy sun is set, thy spring is gone." Gray, Ode on Spring.

" Yet hath my night of life some memory." Shakspeare, Corn. of Errors—last scene. Twining.

" i. e. branches ; which we also use for the horns of a stag. But Aristotle means a new word, not a new application merely of a word already in use. Twining. A supplicator; literally, a prayer, taken in the sense of one who prays; as seer is used for prophet. Twining. κρῖ is used Il. 5. 196. Δω̃, 1l. A. 425. Twining. Part of a verse of Empedocles, quoted by Strabo, p. 3&1, ed. Cantab.

12. On genders. Instead of δεξιόν. [18 Further still, of nouns some are masculine, others feminine, and others between, [ or neuter]. And the masculine, indeed, are such as end in v, and p, and a,, and such as are composed from σ ; but these are two, ψ and ξ. The feminine nouns are such as are composed from vowels, and always end in long vowels; as, for instance, in η and ω, or in a of the doubtful. Hence it happens that the number of terminations for masculine and feminine are equal; for the terminations of ψ and ξ are the same. No noun, however, ends in a mute, or in a short vowel; and only three nouns end in t, viz. μέλι, κόμμι, and πέπερι. But five end in v; viz. πω̃ΰ, να̃πυ, γόνυ, δόρυ, and ἂστυ. And the neuter nouns end in these, and in v and c.]

Two requisites of diction, clearness, and freedom from meanness, How produced. THE virtue of diction, however, consists in being perspicuous, and not mean 1. The diction, therefore, is most perspicuous, which is composed from proper nouns, but then it will be mean. But an example of this is the poetry of Cleophon and Sthenelus. It will, however, be elevated, and remote from the vulgar idiom, by employing unusual words. But I call unusual words, such words as are foreign, the metaphorical, the extended, and every word except the 2. Results and effects of the different kinds of words. proper [name of a thing]. If, however, [a poet wholly employ such words as these, it will be either an enigma, or a barbarism. If, therefore, it were II. E. 393. This following passage is false in its statements, and totally foreign to Aristotle's design. Ritter, therefore, has rightly condemned it. Cf. Rhet. iii. 1, extr. and 2. init. composed from metaphors, it would be an enigma; but if from foreign words, a barbarism. For the essence of an enigma is this, to unite things impossible, yet really true2. Now, from the arrangement of the words, it is not possible to effect this, but it may be effected by a metaphor; as " I saw a man who had glued brass to a man with fire 3;" and others of the like kind. [But from the composition of foreign words a barbarism is produced 4.] Hence language should be moderately varied with these. Foreign, 3. therefore, metaphorical, [and ornamental J words, and the other species that have been mentioned, will cause the diction to be neither vulgar nor mean: but proper words produce perspicuity. But the extend- 4. ing, contracting, and changing of names, contribute in no small degree to the perspicuity of the diction, without vulgarity. For the use of words in a way different from their proper and usual signification, causes the diction to be not vulgar; but the adoption of words in their accustomed meaning, renders it perspicuous. Hence those do not blame rightly, who 5. Objections ill founded. find fault with this mode of speech, and like the ancient Euclid ridicule the poet, [objecting] that verse might easily be composed, if one permit the quantity of syllables to be lengthened at pleasure, making Iambics even in common discourse ; as ἤτοι Χάριν εἶδον Μαραθῶννάδε βαδίζοντα.

And, οὐκ ἂν γ᾽ ἐράμενος τὸν ἐκείνου ἐλλέβορου. τὸ λέγοντα ύπάρχοντα άδύνατα συνάψαι, " to put together things apparently inconsistent and impossible, and at the same time saying nothing but what is true." Twining. Rhet. iii. 2. κόλλησιν εὶπε τὴν τη̃ς σικύας προσβολήν. The operation of cupping is meant. Compare Celsus ii. 11, and Almelov. on Coelius Aurel. Chronic, ii. 1. § 394. Rhod. on Scribon. Larg. Compos. xlvi. The fire alludes to the burnt tow (linamentum) used to exhaust the air in the cupping glass. Apparently a gloss. See Ritter. It is of little use to attempt to settle the reading of these "nonsense verses." Ritter observes that the fault probably lay in pronouncing βαδδίζοντα, έρράμενος. See his notes. 6. It is evident, therefore, that to be detected using this mode of diction is ridiculous. [But measure is common to all the parts of diction 6. J For it would produce the same effect, to make an improper and ridiculous use of metaphors, foreign words, and other 7. forms of diction. But how great a difference is made by the appropriate use of them, may be seen in epic poetry, by putting the words in metre. And he who transfers proper names into foreign words, into metaphors, and the other forms, will see that what we have said is true. Thus, for instance, Aeschylus and Euripides made the same Iambic verse; but by only changing one word from its proper and usual to a foreign signification, the one verse appears beautiful, and the other mean. For Aeschylus indeed, in his Philoctetes, writes, A cancerous ulcer feeds upon my foot. But Euripides, instead of έσθίει, feeds, uses the word θοινāται, banquets on. And, Νῦν δέ μ᾽ ἐὼν ὀλίγος τε καὶ οὐτιδανὸς καὶ ἄκικυς, by inserting proper [and common] words, it will be, Νῦν δέ μ᾽ ἐὼν μικρός τε καὶ ἀσθενικὸς καὶ ἀειδής7. And, Δίφρον ἀεικέλιον καταθεὶς, ὀλίγην τε τράπεζαν. Δίφρον μοχθηρὸν καταθεὶς, μικράν τε τράπεζαν. Twining renders, " in the employment of all the species of unusual words, moderation is necessary." But this meaning can scarcely be elicited from the words. See Ritter, who condemns the passage as an interpolation. In this verse Polyphemus complains that he was deprived of sight by Ulysses, a little, weak, vile man. But Homer, instead of using the word μικρὸς, little, uses όλίγος, which signifies few. Instead of άσθενικὸς, puny, he uses ούτιδανος, which signifies a man of no account; and ἂκικυς, powerless, instead of άειδὴς, obscure. Taylor. Cf. Od. ix. 515. In this verse, which is from the 21st book of the Odyssey, Homer, for the purpose of signifying an ignoble seat, calls it by a foreign word, ἀεικέλιον, and not by the usual word, μοχθηρόν ; and he calls the table, not μικρἀν, small, but όλίγην, few. Taylor. Cf. Od. xxi. 259. Or change, ἠϊόνες βοόωσιν (the shores rebellow9) to ἠϊόνες κράζουσιν (the shores cry out). [10 8. Further objections answered. Again, Ariphrades used to ridicule the tragic poets for employing modes of diction, which no one would use in common conversation; such as δωμάτων ἂπο, and not ,iro δωμάτων, i. e. home from, and not from home; σέθεν [for σου̃]; έγὼ δέ νιν1 1, and 'Ἀχιλλέως πέρι, and not περὶ 'Αχιλλέως, i. e. Achilles about, and not about Achilles; and other expressions of the like kind. For all such forms of language, because they are not in common use, remove vulgarity from the diction. But this he did not know. 9. Proper use of the different words, It is, however, a great thing to use each of the above-mentioned modes in a becoming manner; and also compound and foreign words. But the greatest thing is to employ metaphors well. For this alone cannot be acquired from another, but it is an indication of an excellent genius; since to employ metaphors well, is to discern similitude 12. 10. as suited to each kind of poetry. But of words, the compound are chiefly suited to dithyrambic verse, the foreign to heroic, and metaphors to Iambic verse. And in heroic verse, indeed, all the above-mentioned words are useful; but in Iambics, because they especially imitate common discourse, those words are adapted which may be also used in conversation. And words of this description are, the proper, the metaphorical, [and the ornamental.] And thus much may suffice concerning tragedy, and the imitation in acting 13. 11. P. 265.—Pope's line is, " And distant rocks rebellow to the roar." Twining. Ritter objects to this clause. Twining, however, admires and commends its conclusion. Cf. Soph. Œd. col. 987. More clearly expressed in Rhet. iii. 11. 5, τὸ ὃμοιον καί ἐν πολὺ διέχουσι θεωρεῖν, εὺστόχου. These last words appear to me out of place.

On narrative poetry. Its similarity to tragedy. CONCERNING the poetry, however, which is narrative and imitative in metre, it is evident that it ought to have dramatic fables, in the same manner as tragedy, and should be conversant with one whole and perfect action, which has a beginning, middle, and end, in order that, like one whole animal, it may produce its appropriate pleasure 1; and that it may not be like the custom of histories, in which it is not necessary to treat of one action, but of one time, viz. of such things as have happened in that time, respecting one or more persons, the relation of each of which things to the other is just as it may happen. 2. For as the sea-fight at Salamis, and the battle with the Carthaginians in Sicily, though they happened at the same time, tend nothing to the same end; thus also in successive times, one thing may sometimes be connected with another, from which no one end is produced. 3. Praise of Homer, and comparison with other poets. But nearly all poets do this. Hence, as we have before observed, in this respect also Homer will appear to be divine, when compared with other poets, because he did not attempt to sing of the whole of the Trojan war, though it had a beginning and an end. For if he had, it would have been very great, and not sufficiently conspicuous; or if it bad been of a moderate size, it would have been intricate through the variety of incidents 2. But now, having selected one part of the war, he has made use of many episodes; such as the catalogue of the ships, and other episodes, with which he has adorned his i. e. opposed (as appears from what follows) to that which history gives. Unity of interest is essential to the pleasure we expect from the epic poem; and this cannot exist, at least in the degree required, without unity of action. Twining. Because " the length of the whole would " then " not admit of a proper magnitude in the parts; " and thus an epic poem constructed upon an historical plan, would be exactly in the same case with a tragedy " constructed on an epic plan." Twining. poem. Other poets, however, have composed a fable about one man, and one time, and one action, consisting of many parts; as the authors of the Cypriacs, and the Lesser Iliad. [With respect to the Iliad 4. and Odyssey, therefore, one or two tragedies only could be made from each. But many might be made from the Cypriacs; and from the Lesser Iliad more than eight; such as the Judgment of the Arms 4, Philoctetes5, Neoptolemus, Eurypylus6, the Begging [of Ulysses], the Lacaenae, the Destruction of Troy, the Return of the Greeks, Sinon, and the Troades.

The species and parts of epic poetry. AGAIN, it is requisite that the epopee should have the same species as tragedy. [For it is necessary that it should be either simple, or complex, or ethical, or pathetic 1.] The parts also are the same, except the music and the scenery. For it requires revolutions, discoveries, and disasters; and besides these, the sen- Of this kind seems the poem of Ario8to, the exordium of which not only expresses the miscellaneous variety of his matter, but, also, his principle of unity. Le Donne, i cavalier, 1' arme, gli amori, Le cortesie, l' audaci imprese, io canto, Che furo al TEMPO che passaro i Mori, etc. Ariosto's expedient was, to " intertwist the several actions together, in order to give something like the appearance of one action" to the whole, as has been observed of Spenser [Letters on Chivalry, etc.]: he has given his poem the continuity of basket-work. Or, if I may be indulged in another comparison, his unity is the unity produced between oil and vinegar by shaking them together, which only makes them separate by smaller portions. Twining. So called to distinguish it from the Iliad of Homer, of which it seems to have been a continuation. Twining. i. e. between Ajax and Ulysses. Aeschylus wrote a tragedy on this subject, of which the Ajax of Sophocles is the sequel. Dacier. Of these two plays nothing is known. Condemned by Ritter. timents and the diction should be well formed; all which were first used by Homer, and are used by him fitly. 2. Difference between the Iliad and Odyssey. For of his two poems, the Iliad indeed contains the simple and pathetic; but the Odyssey, the complex; for through the whole of it there is discovery and moral. And besides these things, he excelled all poets in diction and sentiment. 3. Length of the epic poem. The epopee, however, differs from tragedy in the length of the composition, and in the metre. But the proper boundary of its length has been before described ; for it should be such that the beginning and the end may be seen at one view. [And this will be effected if the compositions are shorter than those of the ancient poets, and brought to the same length with the multitude of tragedies that are recited at one hearing3.] 4. Its power of extension. But it is the peculiarity of the epopee to possess abundantly the power of extending its magnitude ; for tragedy is not capable of imitating many actions that are performed at the same time, but that part only which is represented in the scene, and acted by the players. But in the epopee, in consequence of its being a narration, many events may be introduced which have happened at the same time, which are properly connected with the subject, and from which the bulk of the poem is increased. Hence, this contributes to its magnificence, transports the hearer to different places, and adorns the poem with dissimilar episodes. For similitude of events rapidly produces satiety, and causes tragedies to fail. 5. Its proper metre. But heroic metre is established by experience as adapted to the epopee. For if any one should attempt narrative imitation in any other metre, or in many See Pope's translation, xvi. 206, etc., where Ulysses discovers himself to Telemachus—xxi. 212, to the shepherds- xxiii. 211, to Penelope—xxiv. 375, to his father—ix. 17, to Alcinous—iv. 150, etc., Telemachus is discovered to Menelaus by his tears—v. 189, to Helen, by his resemblance to his father—xix. 545, Ulysses is discovered to the old nurse, by the scar. Twining. This is quite contrary to Aristotle's own opinion. See Ritter. Twining's great and tasteful learning cannot bring him to any satisfactory explanation of these words. metres mingled together, the unfitness of it would be apparent. For heroic metre is of all others the most stable and ample. [Hence it especially receives foreign words and metaphors. For narrative imitation excels all others 4. ] But Iambics and tetrametres have more motion ; the one being adapted to dancing, but the other to acting. 6. Obscurity of mixed metres. It would, however, be still more absurd, to mingle them together, as Chaeremon* did. Hence, no one has composed a long poem in any other measure than the heroic; but, as we have said, Nature herself teaches us to distinguish the measure best suited. 7. Praise of Homer. Homer, indeed, deserves to be praised for many other things, and also because he is the only poet who was not ignorant what he ought to do himself. For it is requisite that the poet should speak in his own person as little as possible ; for so far as he does so he is not an imitator. Other poets, therefore, take an active part through the whole poem, and they only imitate a few things, and seldom7. But Homer, after a short preface, immediately introduces a man or a woman, or something else that has manners; for there is nothing in his poem unattended with manners. It is necessary, 8. The wonderful considered in reference to tragedy and epopee. therefore, in tragedies to produce the wonderful; but that which is contrary to reason (whence the wonderful is best produced) is best suited to the epopee, from the agent not being seen9. In the next place, the particulars respecting the pursuit of Hector Condemned by Ritter. Cf. i. 9. Strictly speaking. See Dissertation i. p. 37. Twining. This is remarkably the case with Lucan; of whom Hobbes says, that " no heroic poem raises such admiration of the poet as his hath done, though not so great admiration of the persona he introduceth—[Discourse concerning the Virtues of an Heroic Poem.] Twining. As gods, goddesses, allegorical beings, etc. Twining. The best comment to which I can refer the reader upon all this part of Aristotle, is to be found in the I Oth of the Letters on Chivalry and Romance, in which the Italian poets, and the privileges of genuine poetry, are vindicated, with as much solidity as elegance, against those whom Dryden used to call his " Prose Critics,"—against that sort of criticism " which looks like philosophy, and is not"—Dr. Hurd's Dialogues, etc. vol. iii. Twining. should appear ridiculous in the scene; the Greeks indeed standing still, and not pursuing, and Achilles making signs to them, by the motion of his head, not to engage10. But in the epopee this is concealed. Now, the wonderful phases ; of which this is an indication, that all men. when they wish to gratify their hearers, add something to what they relate. 9. False reasonings respecting poetry. Homer also in the highest degree taught others how to feign in a proper manner. But this is a paralogism. For men fancy that when the consequent follows or results from the antecedent, the consequent may be converted, and that the antecedent will follow from the consequent. This, however, is false. [But why, if the antecedent be false, so long as this other be otherwise, should the consequent necessarily follow ? For through knowing the consequent to be true, our soul paralogizes, and concludes that the antecedent also is true. And there is an example of this in " the 10. Further precepts respecting poetic probability. Washing11."] Again, one should prefer things which are impossible but probable, to such as are possible but improbable. Fables also should not be composed from irrational parts, [but as much as possible, indeed, they should have nothing irrational in them: if, however, this is impossible, care should be taken that the irrational circumstance does not pertain to the fable, as in the case of Œdipus not knowing how Laius died 12. For it must not be brought into the drama, like the narration of the Pythian games in the Electra 13, or him who, in the tragedy of the Pope's Iliad, xxii. 267.—Perhaps the idea of stopping a whole army by a nod, or shake of the head, (a circumstance distinctly mentioned by Homer, but sunk in Mr. Pope's version,) was the absurdity here principally meant. If this whole Homeric scene were represented on our stage, in the best manner possible, there can be no doubt that the effect would justify Aristotle's observation. It would certainly set the audience in a roar. Twining. I follow Ritter's text and version, but both he and Donaldson regard these words as an interpolation. Cf. Soph. Œd. Tyr. 112, sqq. This clause is condemned by Ritter. Id. Electr. 680, sqq. See my Introduction to Sophocles, p. xiii. metres mingled together, the unfitness of it would be apparent. For heroic metre is of all others the derstable and ample. [Hence it especi fables foreign words and metaphors. For if they are tion excels all others4.] But reasonable that they have more motion uity also must be admitted; but there irrational circumstances in the Odyssey, such as Ulysses being left [on the shore of Ithaca by the Phoeacians], would evidently have been intolerable, if they had been fabricated by a bad poet. But now the poet conceals the absurdity, and renders it pleasing by the addition of other beauties. 11. The diction in the respective parts. The diction, likewise, should be laboured in the sluggish parts of the poem, and which exhibit neither manners nor sentiment14. For a very splendid15 diction conceals the manners and the reasoning.

Objections of critics, [WITH respect to critical objections1, and the solutions of them, the number and quality of their species The reader may wonder that Aristotle did not add—" nor passion." But that part of the epic and tragic poem, which he calls the sentiments, includes the expression of passion. Twining. Is " His diction [ Thomson's] is in the bighest degree florid and luxuriant; such as may be said to be to his images and thoughts both their lustre and their shade ; such as invests them with splendour, through which perhaps they are not always easily discerned."—Dr. Johnson's Life of Thomson. Twining. The original is, Problems. This appears to have been a common title of critical works in Aristotle's time. Objections, censures, and the most unreasonable cavils, were conveyed in the civil form of problems and questions. Thus, many criticisms on Hotter were published under the title of Homeric Problems. The scope of this part of Aristotle's work is of more importance to his subject than, at first view, it may appear to be. In teaching how to answer criticisms, it, in fact, teaches, (as far, I mean, as it goes,) what the poet should do to avoid and their solutions. will become apparent from surveying them as follows. Since the poet is an imitator, in the same manner as a painter, or any other person who makes likenesses, it is necessary that he should always imitate one of the three [objects of imitation]. For he must either imitate things such as they were or are, or such as they are said and appear to be2, or such as they ought to be. 2. But these are enunciated either by [common] diction, or by foreign words and metaphors. For there are many modifications of diction; and we concede these to the poets. 3. Error twofold, essential and accidental. Besides this, there is not the same rectitude of politics and poetry, nor of any other art and poetry. But of poetry itself, the error is twofold; the one essential, the other accidental. 4. For the error is essential, when it attempts to imitate that which is beyond its power; but [when it attempts to imitate improperly, as4] if for instance, a horse should be described as moving both its right legs together, or an error in any of the giving occasion to them. It seems, indeed, intended as an apology for Poetry, and a vindication of its privileges upon true poetical principles, at a time when the art and its professors were unfairly attacked on all sides, by the cavils of prosaic philosophers and sophists, such as Ariphrades, Protagoras, Euclid, etc., and by the puritanical objections of PLATO and his followers. Twining. If Ritter's strictures be true, our acquaintance with Aristotle's Poetics is now at an end. In a learned annotation (pp. 263—6) he seems to have completely proved the spurious character of both this and the following chapter. This includes all that is called faery, machinery, ghosts, witches, enchantments, etc. — things, according to Hobbes, " beyond the actual bounds, and only within the " conceived possibility of nature." [See the Letters on Chivalry, as above.] Such a being as Caliban, for example, is impossible. Yet Shakspeare has made the character appear probable; not, certainly, to reason, but to imagination; that is, we make no difficulty about the possibility of it, in reading. Is not the Lovelace of Richardson, in this view, more out of nature, more improbable, than the Caliban of Shakspeare ? The latter is, at least, consistent. I can imagine such a monster as Caliban: I never could imagine such a man as Lovelace. Twining. πάθη, inflections, dialectic variations, etc. These words, τὸ προελέσθαι μὴ όρθῶς, are totally inconsistent with the meaning. See Ritter. arts be committed in poetry, as in medicine, or any other art, when it fabricates things that are impossible, these, therefore, whatever they may be, are not the essential errors of poetry. 5. Whence objections may be removed. Hence, one must refute the objections of critics from surveying these particulars. For in the first place, indeed, the poet errs, if what he fabricates is impossible according to the art itself; but it will be right if the end of poetry is obtained by it. For we have before shown what the end is, viz. if the poet thus renders what he fabricates, or any other part of the poem, more capable of producing a more striking effect5. An example of this is the pursuit of Hector. If, however, this end can be more or less attained, and that according to the art pertaining to these things, then the fault will not be excusable. For it is requisite if possible to be entirely without error. Further still, it should be considered whether the error ranks among things essential to the poetic art, or foreign and incidental. For it is a less fault not to know that a hind has no horns, than to depict a bad copy of one. Besides this, also, if the poet is blamed for 6. not imitating things as they truly are, the reply is, but he imitates them as they should be. Thus Sophocles said, that he described men such as they should be, but Euripides such as they are. If, however, it should be objected, that the poet represents things in neither of these ways, he may say that he represents them as men say they are; as, for instance, in things pertaining to the gods. For per- 7. Laps it is neither better thus to speak, nor true, but it is just as it may happen; as Xenophanes observes, “At any rate they tell us such things6." Perhaps, however, it may be said, that it is not better, indeed, thus to speak, but that the thing was so; as in the passage concerning the arms [of the sleeping soldiers of Diomed]: A false definition. See Ritter. “Tyrwhitt's emendation, άλλ’οὺνφασὶν τάδε, seems to be the best of those which have been proposed." Donaldson.

ἔγχεα δέ σφιν ὄρθ' ἐπὶ σαυρωτῆρος, fixed upright in the earth Their spears stood by For such was the custom at that time, as it is now 8. with the Illyrians. With respect, however, to the inquiry whether a thing is said or done by any one well or ill, we must not only regard the thing itself which is done or said, whether it is good or bad, but we must also [consider] the person by whom it is done or said, viz. concerning whom, or when, or to whom, or on what account, he speaks or acts; as whether it is for the sake of bringing to pass a greater good, or in order to avoid a greater evil. 9. Diction. But it is requisite to remove some objections by looking to the diction; as, for instance, to foreign words: On mules th' infection first began. For perhaps he does not use ούρῆας to signify mules, but guards8. And in what he says of Dolon,

—— of form unhappy It may be said that εἶδος κακὸς does not signify a body without symmetry, but a deformed face. For the Cretans call a man with a good face εὺειδής. And,

For ζωρότερον may not mean undiluted wine, as for 11. X. 152. See Ritter. 11. A. 50. Zoilus thought the pestilence should have begun with the men first. 11. K. 316. The objection of the critics is supposed to have been, that an ill-made man could not be a good racer. See Pope's note. Twining. Iliad ix. 267, 8.—Pope follows the common, and probably the right, acceptation of the word. " Mix purer wine.”—Aristotle's interpretation has not made its fortune with the critics. He seems to have produced it rather as an exemplification of the sort of answer which he is here considering, than as an opinion in which he acquiesced himself. It was, probably, an answer which had been given. The cavil, according to Plutarch, came from Zoilus. [See the Symposiac Probe of Plut. v. 4, where this subject is discussed, and several other conjectural senses of the word Zωρότερον are proposed.] Twining. intemperate drinkers, but wine poured out rapidly. But a thing is said metaphorically, as, 10. Metaphor. The other gods and men " Slept all the night. And at the same time he says : ἤτοι ὅτ᾽ ἐς πέδιον τὸ Τρωικὸν ἀθρήσειεν αὐλῶν συρίγγων θ᾽ ὁμαδόν. For all is said metaphorically, instead of many, all' being a species of many. And thus οἲη δ’ ἂμμορος is said of Orion metaphorically. For that which is most known, is called alone or sole13. Objections also 11. Accent. may be solved from accent, as Hippias the Thasian solved the following passages :

And,

Objections likewise may be solved by punctuation; 12. Punctuation. as in the following instance from Empedocles,

Beginning of II. ii.— What it was that wanted defence in this passage, and that was to be taken metaphorically, we are not told. That it was the representation of the gods as sleeping, is the most probable conjecture. This is somewhat softened by Mr. Pope's" “slumbered.” Homer says-" SLEPT all the night,”—ευδον παννυχιοι. Twining. Tracts is a far-fetched notion from παννύχιοι. The author has blundered between 11. B. 1., and II. I. 1, and 11—13. See Ritter. 11. xxi. 297, has δίδομεν δέ τοι εὖχος άρέσθαι, but the line, as here quoted, is not found in Homer. See Ritter. Taylor observes : "It alludes to the order given by Jupiter to the dream in 11. ii. to deceive Agamemnon. Here, if διδομεν is read with an accent in the antepenult, it will signify damus, and will imply that Jupiter promises Agamemnon glory from the battle; but if it is read with an accent in the penult, διδόμεν, so as to be the infinitive Ionic, it will signify dare. It will therefore imply that Jupiter orders the dream to give the hope of victory to Agamemnon." If this is read with the circumflex on the οὖ, it will signify that the oak became putrid by the rain, which is absurd; but if it is read with an acute accent and spiritus lenis οὖ, it will signify not, and will imply, that the oak was not rotted by the rain. Taylor. The sense here depends on the punctuation. For if the 13. Ambiguity. Or by ambiguous expressions, as [in Iliad, x.] παρῴχηκεν δὲ δὲ πλέων νυξ16. Night of two parts the greater share had waned, But of her empire still a third remain'd. 14. Custom of diction. For the word πλέων is ambiguous. Or objections may be answered from the custom of diction; as when wine is called κεκραμένον, mixed17; whence has been said, Greaves of new-wrought tin. And those that work on iron are called braziers18. Whence Ganymede is said To pour out wine for Jove 19; II. 20. though the gods do not drink wine. But this may also be metaphorically said. 15. Contraries. It is necessary, however, when a word appears to signify something of a contrary nature, to consider how many significations it may have in the passage before us; as, τῇ ῥ᾽ ἔσχετο χάλκεον ἔγχος There stuck the lance20 :

For here the word stuck implies that the lance was impeded. Of how many different senses a word may admit, one may learn thus, by a contrary manner comma is put after ζωρὰ the second line, instead of πρὶν, the sense will be, “Immediately those things were made mortal which before had learnt to be immortal, and pure which before were mixed." But if the comma is put after πρὶν instead of ζωρὰ, the sense will be, " that those things which before were pure, were mixed." Taylor. But see Ritter's learned note. But the ambiguity is occasioned by the word πλέων, which may either signify more than, or the greater part of. Taylor. 11. K. 252. ἄστρα δὲ δὴ προβέβηκε, παρῴχηκεν δὲ πλέων νύξ, τῶν δύο μοιράων. " Ea potio quae ex aqua et vino commixta est tamen vinum nominatur. Nove dictum φασιν εἶναι pro λέγουσιν.” Ritter. “Schol. in Horn. II. T. 283. παλαιὰ ἡ χρῆσις τοῦ χαλκὸν ὀνομάζειν τὸν σίδηρον. ἀμέλει καὶ χαλκέας τοὺς τὸν σίδηρον ἐργαζομένους. Ritter. Il. 20. 234. Il. 20. 272. This is consummate twaddle. See Twining and Ritter. from what Glauco says21, [when he asserts that] " some men presuppose irrationally, and then reason from their own decision: and, having once pronounced their opinion, reprobate whatever is contrary to their [preconceived] opinion." This was the case with respect to Icarius. For the multitude fancy that he was a Laconian. On this supposition, therefore, it is absurd that Telemachus should not meet him, on his arrival at Lacedaemon22. Perhaps, however, the truth is as the Cephalenians say, viz. that Ulysses married among them, and that Icadius, and not Icarius, [was his father-in-law]. 17. It is probable, therefore, that this objection is erroneous. In short, it is necessary to refer the impossible either to the poetry, or to that which is better, or to opinion. For, with respect to poetry, probable impossibility is more eligible, than the improbable and possible, and things should be such as Zeuxis23 painted. And also [we may refer the impossible] to that which is better24: for it is necessary that the pattern should transcend those things which are said to be irrational. The objection, also, that something is irrational may be solved by saying, that sometimes it is not irrational; for it is probable that what is improbable may have happened. 18. But with respect to the solution of apparent contraries, these are to be considered in the same manner as This is most clumsily and indistinctly expressed. See Ritter. " In ancient days, while Greece was flourishing in liberty and arts, a celebrated painter, [Zeuxis,] having drawn many excellent pictures for a certain free state, and been generously rewarded for his labours, at last made an offer to paint them a Helen, as a model and exemplar of the most exquisite beauty. The proposal was readily accepted, when the artist informed them, that in order to draw one Fair, it was necessary he should contemplate many. He demanded, therefore, a sight of all their finest women. The state, to assist the work, assented to his request. They were exhibited before him; he selected the most beautiful; and from these formed his Helen, more beautiful than them all.”—Harris’s Three Treatises, p. 216. Twining. Improved nature, ideal beauty, etc., which, elsewhere, is expressed by what should be. Twining. Ritter rightly sup. plies τὸ άδύνατον δεῐ άνάγειν. elenchi20 in arguments, if the same thing [is affirmed or denied], and with respect to the same thing, and after the same manner, and whether it is the same person [who affirms and denies], and also with what reference he speaks, and what a wise man would understand from his words26. 19. When reprehension is correct. The reprehension [of poets] on the score of improbability27 and vicious manners will be right, through which it is shown, that they have without any necessity devised something irrational. Thus irrationality is used [without any necessity] by Euripides in his Aegeus, and viciousness, in the character of Menelaus, in his 20. Summary. Orestes. The reprehensions, therefore, may be derived from five species. For they are either made because impossibilities are introduced, or absurdities, or things of evil tendency, or contraries, or as errors committed against the rectitude of art. But the solutions may be surveyed from the above-mentioned number; for they are twelve.28]

Respective merits of tragic and epic imitation. ONE may, however, question whether epic or tragic imitation is the more excellent. For if that imitation is the better which is less troublesome to the spectator, and such an imitation pertains to better spectators, that which imitates every thing is evidently attended with molestation. For, as if the spectators will not perceive what is acted without the addition of much movement1, they make great i. e. confutations. I have followed Ritter, who has done much for this awkward heap of tautologies. Surely the words καὶάλογία καὶ μοχθηρία are corrupt or interpolated. The reader who regards his own ease, will, I believe, do well to take this for granted. If, however, he has any desire to try the experiment, he may read my NOTE on this passage ; and I wish it may answer to him. Twining. Though Aristotle instances in gesture only, the objection, no doubt, extended to the whole imitative representation of the theatre, including the stage and scenery, by which place is imitated, and the dresses, which are necessary to complete the imitation of the persons. Twining. gesticulations; just as bad players on the flute turn themselves round, when it is requisite to imitate the action of the discus; or when they sing of Scylla, draw to themselves the coryphaeus, or leader of the band. 2. Complaints respecting actors. Such, then, is tragedy, as the modern actors are in the estimation of their predecessors3. Hence, Myniscus called Callipides an ape, in consequence of carrying his imitation to a great excess. And there was also a similar opinion respecting Pindar [the player]. But as these latter actors are to the former, so is the whole art of tragedy to the epopee. They say, therefore, that the epopee is calculated for hearers of the better sort, on which account it does not require scenery; but that tragedy is calculated for the vulgar. Hence, tragic imitation, which is troublesome to the spectator, will evidently be inferior to epic imitation.

3. But the whole question rather pertains to the actor than the poet. In the first place, however, this accusation does not pertain to the poet, but the actor; since it is possible in reciting epic poetry to overdo action, as Sosistratus did, and singing likewise, as Mnastheus of Opus did. In the next place, neither is all motion to be despised, since neither is every kind of dancing, ἕλκοντες τὸν Κορυφαῐον—To imitate Scylla,—“ naves in saxa trahentem”as Virgil has expressed it. But it is not easy to see how the performer, at least while he was playing, could well spare a hand for this operation.- This was even worse than what we call humouring a catch; when, for instance, a singer who is performing Purcell's " Fie, noy prithee, John," -thinks it necessary to collar his neighbour. Twining. Sheridan has burlesqued this habit of " suiting the action with admirable humour in " the Critic." Sir Christopher Hatton turning out his toes, Lord Burleigh shaking his head, and Tilburnia going mad in white satin, are among the best of his innumerable facetiae on the subject. Compare also " the Rehearsal," Act 1, where Bayes instructs the Thunder and Lightning how to express their noisy and rapid powers by suitable action. The " decline of the drama," then, was a subject of complaint at Athens, as well as in London. but only that which is bad; and hence Callipedes was blamed, as others now are for imitating light women4. Further still, tragedy, in the same manner as the epopee, may fulfil its purpose without gesture ; for by reading, it is manifest what kind of thing it is. If, therefore, it is in other respects better, it is not necessary that it should be accompanied [by motion and gesture]. 4. Reasons for the superiority of tragedy. In the next place, tragedy has every thing which the epopee possesses. For it may use metre, and it has also music and scenery, as no small parts, through which the pleasure it produces is most apparent. To which may be added, that it possesses perspicuity, both when it is read, and when it is acted. 5. The end, too, of its imitation is confined in less extended limits. For being crowded into a narrower compass, it becomes more pleasing than if it were diffused through a long period of time. Thus, for instance, if one were to put the Oedipus of Sophocles into as many verses as the Iliad, Fit would be less pleasing]. 6. Again, the imitation of the epopee has less unity [than tragic imitation]; of which this is an indication, that from any kind of [epic] imitation many tragedies may be produced. Hence, if he who writes an epic poem should choose a fable perfectly one, the poem would necessarily either appear short, as if curtailed, or if it should be accompanied with length of metre, it would seem to be languid5. But if he should compose one fable from many fables, I mean, if the poem should consist of many actions, it would not possess unity. Thus, the Iliad and Odyssey contain many such parts, which of themselves possess magnitude, though these poems are composed, as much as possible, in the most excellent Taylor has gone fearfully astray in his version, " for not imitating free women"! The negative particle belongs to the adjective. Twining observes: " as no actresses were admitted on the Greek stage, their capital actors must frequently have appeared in female parts, such as Electra, Iphigenia, Medea, etc. This is sufficiently proved by many passages of ancient authors; and among others, by a remarkable story of Polus, an eminent Greek Tragic actor, told by Aulus Gellius ύδαρῆ, milk-and-waterish, manner, and are most eminently the imitation of one action. If, therefore, tragedy excels in all these 7. particulars, and besides this, in the work of art, (for neither tragic nor epic imitation ought to produce a casual pleasure, but that which has been stated,) it is evident that it will be more excellent than the epopee, in consequence of attaining its end in a greater degree. And thus much concerning tragedy, 8. and the epopee, as to themselves, their species, and their parts, their number, and their difference, what the causes are of their being good or bad, and also concerning the objections which may be made to them, and the solutions of the objections.