The Poetics

Aristotle

Wor

k re

prod

uced

with

no

edito

rial r

espo

nsib

ility

Notice by Luarna Ediciones

This book is in the public domain becausethe copyrights have expired under Spanish law.

Luarna presents it here as a gift to its cus-tomers, while clarifying the following:

1) Because this edition has not been super-vised by our editorial deparment, wedisclaim responsibility for the fidelity ofits content.

2) Luarna has only adapted the work tomake it easily viewable on common six-inch readers.

3) To all effects, this book must not be con-sidered to have been published byLuarna.

www.luarna.com

1

Our subject being Poetry, I propose to speaknot only of the art in general but also of its spe-cies and their respective capacities; of the struc-ture of plot required for a good poem; of thenumber and nature of the constituent parts of apoem; and likewise of any other matters in thesame line of inquiry. Let us follow the naturalorder and begin with the primary facts.

Epic poetry and Tragedy, as also Comedy, Dit-hyrambic poetry, and most flute-playing andlyre-playing, are all, viewed as a whole, modesof imitation. But at the same time they differfrom one another in three ways, either by adifference of kind in their means, or by differ-ences in the objects, or in the manner of theirimitations.

I. Just as form and colour are used as means bysome, who (whether by art or constant practice)

imitate and portray many things by their aid,and the voice is used by others; so also in theabove-mentioned group of arts, the means withthem as a whole are rhythm, language, andharmony—used, however, either singly or incertain combinations. A combination of rhythmand harmony alone is the means in flute-playing and lyre-playing, and any other artsthere may be of the same description, e.g. imita-tive piping. Rhythm alone, without harmony, isthe means in the dancer's imitations; for evenhe, by the rhythms of his attitudes, may repre-sent men's characters, as well as what they doand suffer. There is further an art which imi-tates by language alone, without harmony, inprose or in verse, and if in verse, either in someone or in a plurality of metres. This form ofimitation is to this day without a name. Wehave no common name for a mime of Sophronor Xenarchus and a Socratic Conversation; andwe should still be without one even if the imita-tion in the two instances were in trimeters or

elegiacs or some other kind of verse—though itis the way with people to tack on 'poet' to thename of a metre, and talk of elegiac-poets andepic-poets, thinking that they call them poetsnot by reason of the imitative nature of theirwork, but indiscriminately by reason of themetre they write in. Even if a theory of medi-cine or physical philosophy be put forth in ametrical form, it is usual to describe the writerin this way; Homer and Empedocles, however,have really nothing in common apart from theirmetre; so that, if the one is to be called a poet,the other should be termed a physicist ratherthan a poet. We should be in the same positionalso, if the imitation in these instances were inall the metres, like the Centaur (a rhapsody in amedley of all metres) of Chaeremon; and Chae-remon one has to recognize as a poet. So much,then, as to these arts. There are, lastly, certainother arts, which combine all the means enu-merated, rhythm, melody, and verse, e.g.Dithyrambic and Nomic poetry, Tragedy and

Comedy; with this difference, however, that thethree kinds of means are in some of them allemployed together, and in others brought inseparately, one after the other. These elementsof difference in the above arts I term the meansof their imitation.

2

II. The objects the imitator represents are ac-tions, with agents who are necessarily eithergood men or bad—the diversities of humancharacter being nearly always derivative fromthis primary distinction, since the line betweenvirtue and vice is one dividing the whole ofmankind. It follows, therefore, that the agentsrepresented must be either above our own levelof goodness, or beneath it, or just such as we

are in the same way as, with the painters, thepersonages of Polygnotus are better than weare, those of Pauson worse, and those of Diony-sius just like ourselves. It is clear that each ofthe above-mentioned arts will admit of thesedifferences, and that it will become a separateart by representing objects with this point ofdifference. Even in dancing, flute-playing, andlyre-playing such diversities are possible; andthey are also possible in the nameless art thatuses language, prose or verse without har-mony, as its means; Homer's personages, forinstance, are better than we are; Cleophon's areon our own level; and those of Hegemon ofThasos, the first writer of parodies, and Nico-chares, the author of the Diliad, are beneath it.The same is true of the Dithyramb and theNome: the personages may be presented inthem with the difference exemplified in the...of... and Argas, and in the Cyclopses ofTimotheus and Philoxenus. This difference it isthat distinguishes Tragedy and Comedy also;

the one would make its personages worse, andthe other better, than the men of the presentday.

3

III. A third difference in these arts is in themanner in which each kind of object is repre-sented. Given both the same means and thesame kind of object for imitation, one may ei-ther (1) speak at one moment in narrative andat another in an assumed character, as Homerdoes; or (2) one may remain the same through-out, without any such change; or (3) the imita-tors may represent the whole story dramati-cally, as though they were actually doing thethings described.

As we said at the beginning, therefore, the dif-ferences in the imitation of these arts come un-der three heads, their means, their objects, andtheir manner.

So that as an imitator Sophocles will be on oneside akin to Homer, both portraying good men;and on another to Aristophanes, since bothpresent their personages as acting and doing.This in fact, according to some, is the reason forplays being termed dramas, because in a playthe personages act the story. Hence too bothTragedy and Comedy are claimed by the Dori-ans as their discoveries; Comedy by the Mega-rians—by those in Greece as having arisenwhen Megara became a democracy, and by theSicilian Megarians on the ground that the poetEpicharmus was of their country, and a gooddeal earlier than Chionides and Magnes; evenTragedy also is claimed by certain of the Pelo-ponnesian Dorians. In support of this claimthey point to the words 'comedy' and 'drama'.

Their word for the outlying hamlets, they say,is comae, whereas Athenians call themdemes—thus assuming that comedians got thename not from their comoe or revels, but fromtheir strolling from hamlet to hamlet, lack ofappreciation keeping them out of the city. Theirword also for 'to act', they say, is dran, whereasAthenians use prattein.

So much, then, as to the number and nature ofthe points of difference in the imitation of thesearts.

4

It is clear that the general origin of poetry wasdue to two causes, each of them part of humannature. Imitation is natural to man from child-

hood, one of his advantages over the loweranimals being this, that he is the most imitativecreature in the world, and learns at first by imi-tation. And it is also natural for all to delight inworks of imitation. The truth of this secondpoint is shown by experience: though the ob-jects themselves may be painful to see, we de-light to view the most realistic representationsof them in art, the forms for example of the lo-west animals and of dead bodies. The explana-tion is to be found in a further fact: to be learn-ing something is the greatest of pleasures notonly to the philosopher but also to the rest ofmankind, however small their capacity for it;the reason of the delight in seeing the picture isthat one is at the same time learning—gathering the meaning of things, e.g. that theman there is so-and-so; for if one has not seenthe thing before, one's pleasure will not be inthe picture as an imitation of it, but will be dueto the execution or colouring or some similarcause. Imitation, then, being natural to us—as

also the sense of harmony and rhythm, the me-tres being obviously species of rhythms—it wasthrough their original aptitude, and by a seriesof improvements for the most part gradual ontheir first efforts, that they created poetry out oftheir improvisations.

Poetry, however, soon broke up into two kindsaccording to the differences of character in theindividual poets; for the graver among themwould represent noble actions, and those ofnoble personages; and the meaner sort the ac-tions of the ignoble. The latter class producedinvectives at first, just as others did hymns andpanegyrics. We know of no such poem by anyof the pre-Homeric poets, though there wereprobably many such writers among them; in-stances, however, may be found from Homerdownwards, e.g. his Margites, and the similarpoems of others. In this poetry of invective itsnatural fitness brought an iambic metre intouse; hence our present term 'iambic', because it

was the metre of their 'iambs' or invectivesagainst one another. The result was that the oldpoets became some of them writers of heroicand others of iambic verse. Homer's position,however, is peculiar: just as he was in the seri-ous style the poet of poets, standing alone notonly through the literary excellence, but alsothrough the dramatic character of his imita-tions, so too he was the first to outline for usthe general forms of Comedy by producing nota dramatic invective, but a dramatic picture ofthe Ridiculous; his Margites in fact stands in thesame relation to our comedies as the Iliad andOdyssey to our tragedies. As soon, however, asTragedy and Comedy appeared in the field,those naturally drawn to the one line of poetrybecame writers of comedies instead of iambs,and those naturally drawn to the other, writersof tragedies instead of epics, because these newmodes of art were grander and of more esteemthan the old.

If it be asked whether Tragedy is now all that itneed be in its formative elements, to considerthat, and decide it theoretically and in relationto the theatres, is a matter for another inquiry.

It certainly began in improvisations—as didalso Comedy; the one originating with the au-thors of the Dithyramb, the other with those ofthe phallic songs, which still survive as institu-tions in many of our cities. And its advanceafter that was little by little, through their im-proving on whatever they had before them ateach stage. It was in fact only after a long seriesof changes that the movement of Tragedy stop-ped on its attaining to its natural form. (1) Thenumber of actors was first increased to two byAeschylus, who curtailed the business of theChorus, and made the dialogue, or spoken por-tion, take the leading part in the play. (2) Athird actor and scenery were due to Sophocles.(3) Tragedy acquired also its magnitude. Dis-carding short stories and a ludicrous diction,

through its passing out of its satyric stage, itassumed, though only at a late point in its pro-gress, a tone of dignity; and its metre changedthen from trochaic to iambic. The reason fortheir original use of the trochaic tetrameter wasthat their poetry was satyric and more con-nected with dancing than it now is. As soon,however, as a spoken part came in, nature her-self found the appropriate metre. The iambic,we know, is the most speakable of metres, as isshown by the fact that we very often fall into itin conversation, whereas we rarely talk hex-ameters, and only when we depart from thespeaking tone of voice. (4) Another change wasa plurality of episodes or acts. As for the re-maining matters, the superadded embellish-ments and the account of their introduction,these must be taken as said, as it would proba-bly be a long piece of work to go through thedetails.

5

As for Comedy, it is (as has been observed) animitation of men worse than the average;worse, however, not as regards any and everysort of fault, but only as regards one particularkind, the Ridiculous, which is a species of theUgly. The Ridiculous may be defined as a mis-take or deformity not productive of pain orharm to others; the mask, for instance, that ex-cites laughter, is something ugly and distortedwithout causing pain.

Though the successive changes in Tragedy andtheir authors are not unknown, we cannot saythe same of Comedy; its early stages passedunnoticed, because it was not as yet taken up in

a serious way. It was only at a late point in itsprogress that a chorus of comedians was offi-cially granted by the archon; they used to bemere volunteers. It had also already certaindefinite forms at the time when the record ofthose termed comic poets begins. Who it waswho supplied it with masks, or prologues, or aplurality of actors and the like, has remainedunknown. The invented Fable, or Plot, how-ever, originated in Sicily, with Epicharmus andPhormis; of Athenian poets Crates was the firstto drop the Comedy of invective and framestories of a general and non-personal nature, inother words, Fables or Plots.

Epic poetry, then, has been seen to agree withTragedy to this extent, that of being an imita-tion of serious subjects in a grand kind of verse.It differs from it, however, (1) in that it is in onekind of verse and in narrative form; and (2) inits length—which is due to its action having nofixed limit of time, whereas Tragedy endeav-

ours to keep as far as possible within a singlecircuit of the sun, or something near that. This,I say, is another point of difference betweenthem, though at first the practice in this respectwas just the same in tragedies as in epic poems.They differ also (3) in their constituents, somebeing common to both and others peculiar toTragedy—hence a judge of good and bad inTragedy is a judge of that in epic poetry also.All the parts of an epic are included in Tragedy;but those of Tragedy are not all of them to befound in the Epic.

6

Reserving hexameter poetry and Comedy forconsideration hereafter, let us proceed now tothe discussion of Tragedy; before doing so,

however, we must gather up the definition re-sulting from what has been said. A tragedy,then, is the imitation of an action that is seriousand also, as having magnitude, complete initself; in language with pleasurable accessories,each kind brought in separately in the parts ofthe work; in a dramatic, not in a narrative form;with incidents arousing pity and fear, where-with to accomplish its catharsis of such emo-tions. Here by 'language with pleasurable ac-cessories' I mean that with rhythm and har-mony or song superadded; and by 'the kindsseparately' I mean that some portions areworked out with verse only, and others in turnwith song.

I. As they act the stories, it follows that in thefirst place the Spectacle (or stage-appearance ofthe actors) must be some part of the whole; andin the second Melody and Diction, these twobeing the means of their imitation. Here by'Diction' I mean merely this, the composition of

the verses; and by 'Melody', what is too com-pletely understood to require explanation. Butfurther: the subject represented also is an ac-tion; and the action involves agents, who mustnecessarily have their distinctive qualities bothof character and thought, since it is from thesethat we ascribe certain qualities to their actions.There are in the natural order of things, there-fore, two causes, Character and Thought, oftheir actions, and consequently of their successor failure in their lives. Now the action (thatwhich was done) is represented in the play bythe Fable or Plot. The Fable, in our presentsense of the term, is simply this, the combina-tion of the incidents, or things done in thestory; whereas Character is what makes us as-cribe certain moral qualities to the agents; andThought is shown in all they say when provinga particular point or, it may be, enunciating ageneral truth. There are six parts consequentlyof every tragedy, as a whole, that is, of such orsuch quality, viz. a Fable or Plot, Characters,

Diction, Thought, Spectacle and Melody; two ofthem arising from the means, one from themanner, and three from the objects of the dra-matic imitation; and there is nothing else be-sides these six. Of these, its formative elements,then, not a few of the dramatists have madedue use, as every play, one may say, admits ofSpectacle, Character, Fable, Diction, Melody,and Thought.

II. The most important of the six is the combi-nation of the incidents of the story.

Tragedy is essentially an imitation not of per-sons but of action and life, of happiness andmisery. All human happiness or misery takesthe form of action; the end for which we live isa certain kind of activity, not a quality. Charac-ter gives us qualities, but it is in our actions—what we do—that we are happy or the reverse.In a play accordingly they do not act in order toportray the Characters; they include the Char-acters for the sake of the action. So that it is the

action in it, i.e. its Fable or Plot, that is the endand purpose of the tragedy; and the end is eve-rywhere the chief thing. Besides this, a tragedyis impossible without action, but there may beone without Character. The tragedies of most ofthe moderns are characterless—a defect com-mon among poets of all kinds, and with itscounterpart in painting in Zeuxis as comparedwith Polygnotus; for whereas the latter isstrong in character, the work of Zeuxis is de-void of it. And again: one may string together aseries of characteristic speeches of the utmostfinish as regards Diction and Thought, and yetfail to produce the true tragic effect; but onewill have much better success with a tragedywhich, however inferior in these respects, has aPlot, a combination of incidents, in it. Andagain: the most powerful elements of attractionin Tragedy, the Peripeties and Discoveries, areparts of the Plot. A further proof is in the factthat beginners succeed earlier with the Dictionand Characters than with the construction of a

story; and the same may be said of nearly allthe early dramatists. We maintain, therefore,that the first essential, the life and soul, so tospeak, of Tragedy is the Plot; and that theCharacters come second—compare the parallelin painting, where the most beautiful colourslaid on without order will not give one thesame pleasure as a simple black-and-whitesketch of a portrait. We maintain that Tragedyis primarily an imitation of action, and that it ismainly for the sake of the action that it imitatesthe personal agents. Third comes the element ofThought, i.e. the power of saying whatever canbe said, or what is appropriate to the occasion.This is what, in the speeches in Tragedy, fallsunder the arts of Politics and Rhetoric; for theolder poets make their personages discourselike statesmen, and the moderns like rhetori-cians. One must not confuse it with Character.Character in a play is that which reveals themoral purpose of the agents, i.e. the sort ofthing they seek or avoid, where that is not ob-

vious—hence there is no room for Character ina speech on a purely indifferent subject.Thought, on the other hand, is shown in allthey say when proving or disproving someparticular point, or enunciating some universalproposition. Fourth among the literary ele-ments is the Diction of the personages, i.e. asbefore explained, the expression of theirthoughts in words, which is practically thesame thing with verse as with prose. As for thetwo remaining parts, the Melody is the greatestof the pleasurable accessories of Tragedy. TheSpectacle, though an attraction, is the least ar-tistic of all the parts, and has least to do withthe art of poetry. The tragic effect is quite pos-sible without a public performance and actors;and besides, the getting-up of the Spectacle ismore a matter for the costumier than the poet.

7

Having thus distinguished the parts, let us nowconsider the proper construction of the Fable orPlot, as that is at once the first and the mostimportant thing in Tragedy. We have laid itdown that a tragedy is an imitation of an actionthat is complete in itself, as a whole of somemagnitude; for a whole may be of no magni-tude to speak of. Now a whole is that whichhas beginning, middle, and end. A beginning isthat which is not itself necessarily after any-thing else, and which has naturally somethingelse after it; an end is that which is naturallyafter something itself, either as its necessary orusual consequent, and with nothing else afterit; and a middle, that which is by nature afterone thing and has also another after it. A well-constructed Plot, therefore, cannot either beginor end at any point one likes; beginning andend in it must be of the forms just described.

Again: to be beautiful, a living creature, andevery whole made up of parts, must not onlypresent a certain order in its arrangement ofparts, but also be of a certain definite magni-tude. Beauty is a matter of size and order, andtherefore impossible either (1) in a very minutecreature, since our perception becomes indis-tinct as it approaches instantaneity; or (2) in acreature of vast size—one, say, 1,000 mileslong—as in that case, instead of the object beingseen all at once, the unity and wholeness of it islost to the beholder.

Just in the same way, then, as a beautiful wholemade up of parts, or a beautiful living creature,must be of some size, a size to be taken in bythe eye, so a story or Plot must be of somelength, but of a length to be taken in by thememory. As for the limit of its length, so far asthat is relative to public performances andspectators, it does not fall within the theory ofpoetry. If they had to perform a hundred

tragedies, they would be timed by water-clocks, as they are said to have been at one pe-riod. The limit, however, set by the actual na-ture of the thing is this: the longer the story,consistently with its being comprehensible as awhole, the finer it is by reason of its magnitude.As a rough general formula, 'a length whichallows of the hero passing by a series of prob-able or necessary stages from misfortune tohappiness, or from happiness to misfortune',may suffice as a limit for the magnitude of thestory.

8

The Unity of a Plot does not consist, as somesuppose, in its having one man as its subject.An infinity of things befall that one man, some

of which it is impossible to reduce to unity; andin like manner there are many actions of oneman which cannot be made to form one action.One sees, therefore, the mistake of all the poetswho have written a Heracleid, a Theseid, or simi-lar poems; they suppose that, because Heracleswas one man, the story also of Heracles mustbe one story. Homer, however, evidently un-derstood this point quite well, whether by artor instinct, just in the same way as he excels therest in every other respect. In writing an Odys-sey, he did not make the poem cover all thatever befell his hero—it befell him, for instance,to get wounded on Parnassus and also to feignmadness at the time of the call to arms, but thetwo incidents had no probable or necessaryconnexion with one another—instead of doingthat, he took an action with a Unity of the kindwe are describing as the subject of the Odyssey,as also of the Iliad. The truth is that, just as inthe other imitative arts one imitation is alwaysof one thing, so in poetry the story, as an imita-

tion of action, must represent one action, acomplete whole, with its several incidents soclosely connected that the transposal or with-drawal of any one of them will disjoin and dis-locate the whole. For that which makes no per-ceptible difference by its presence or absence isno real part of the whole.

9

From what we have said it will be seen that thepoet's function is to describe, not the thing thathas happened, but a kind of thing that mighthappen, i.e. what is possible as being probableor necessary. The distinction between historianand poet is not in the one writing prose and theother verse—you might put the work of He-rodotus into verse, and it would still be a spe-

cies of history; it consists really in this, that theone describes the thing that has been, and theother a kind of thing that might be. Hence po-etry is something more philosophic and ofgraver import than history, since its statementsare of the nature rather of universals, whereasthose of history are singulars. By a universalstatement I mean one as to what such or such akind of man will probably or necessarily say ordo—which is the aim of poetry, though it af-fixes proper names to the characters; by a sin-gular statement, one as to what, say, Alcibiadesdid or had done to him. In Comedy this hasbecome clear by this time; it is only when theirplot is already made up of probable incidentsthat they give it a basis of proper names, choos-ing for the purpose any names that may occurto them, instead of writing like the old iambicpoets about particular persons. In Tragedy,however, they still adhere to the historic names;and for this reason: what convinces is the pos-sible; now whereas we are not yet sure as to the

possibility of that which has not happened, thatwhich has happened is manifestly possible, elseit would not have come to pass. Neverthelesseven in Tragedy there are some plays with butone or two known names in them, the rest be-ing inventions; and there are some without asingle known name, e.g. Agathon's Anthens, inwhich both incidents and names are of thepoet's invention; and it is no less delightful onthat account. So that one must not aim at a rigidadherence to the traditional stories on whichtragedies are based. It would be absurd, in fact,to do so, as even the known stories are onlyknown to a few, though they are a delight nonethe less to all.

It is evident from the above that, the poet mustbe more the poet of his stories or Plots than ofhis verses, inasmuch as he is a poet by virtue ofthe imitative element in his work, and it is ac-tions that he imitates. And if he should come totake a subject from actual history, he is none

the less a poet for that; since some historic oc-currences may very well be in the probable andpossible order of things; and it is in that aspectof them that he is their poet.

Of simple Plots and actions the episodic are theworst. I call a Plot episodic when there is nei-ther probability nor necessity in the sequence ofepisodes. Actions of this sort bad poets con-struct through their own fault, and good oneson account of the players. His work being forpublic performance, a good poet often stretchesout a Plot beyond its capabilities, and is thusobliged to twist the sequence of incident.

Tragedy, however, is an imitation not only of acomplete action, but also of incidents arousingpity and fear. Such incidents have the verygreatest effect on the mind when they occurunexpectedly and at the same time in conse-quence of one another; there is more of themarvellous in them then than if they happenedof themselves or by mere chance. Even matters

of chance seem most marvellous if there is anappearance of design as it were in them; as forinstance the statue of Mitys at Argos killed theauthor of Mitys' death by falling down on himwhen a looker-on at a public spectacle; for inci-dents like that we think to be not without ameaning. A Plot, therefore, of this sort is neces-sarily finer than others.

10

Plots are either simple or complex, since theactions they represent are naturally of this two-fold description. The action, proceeding in theway defined, as one continuous whole, I callsimple, when the change in the hero's fortunestakes place without Peripety or Discovery; andcomplex, when it involves one or the other, or

both. These should each of them arise out of thestructure of the Plot itself, so as to be the conse-quence, necessary or probable, of the antece-dents. There is a great difference between athing happening propter hoc and post hoc.

11

A Peripety is the change from one state ofthings within the play to its opposite of thekind described, and that too in the way we aresaying, in the probable or necessary sequenceof events; as it is for instance in Oedipus: herethe opposite state of things is produced by theMessenger, who, coming to gladden Oedipusand to remove his fears as to his mother, re-veals the secret of his birth. And in Lynceus: justas he is being led off for execution, with

Danaus at his side to put him to death, the in-cidents preceding this bring it about that he issaved and Danaus put to death. A Discovery is,as the very word implies, a change from igno-rance to knowledge, and thus to either love orhate, in the personages marked for good or evilfortune. The finest form of Discovery is oneattended by Peripeties, like that which goeswith the Discovery in Oedipus. There are nodoubt other forms of it; what we have said mayhappen in a way in reference to inanimatethings, even things of a very casual kind; and itis also possible to discover whether some onehas done or not done something. But the formmost directly connected with the Plot and theaction of the piece is the first-mentioned. This,with a Peripety, will arouse either pity or fear—actions of that nature being what Tragedy isassumed to represent; and it will also serve tobring about the happy or unhappy ending. TheDiscovery, then, being of persons, it may bethat of one party only to the other, the latter

being already known; or both the parties mayhave to discover themselves. Iphigenia, for in-stance, was discovered to Orestes by sendingthe letter; and another Discovery was requiredto reveal him to Iphigenia.

Two parts of the Plot, then, Peripety and Dis-covery, are on matters of this sort. A third partis Suffering; which we may define as an actionof a destructive or painful nature, such as mur-ders on the stage, tortures, woundings, and thelike. The other two have been already ex-plained.

12

The parts of Tragedy to be treated as formativeelements in the whole were mentioned in a

previous Chapter. From the point of view,however, of its quantity, i.e. the separate sec-tions into which it is divided, a tragedy has thefollowing parts: Prologue, Episode, Exode, anda choral portion, distinguished into Parode andStasimon; these two are common to all trage-dies, whereas songs from the stage and Com-moe are only found in some. The Prologue is allthat precedes the Parode of the chorus; an Epi-sode all that comes in between two whole cho-ral songs; the Exode all that follows after thelast choral song. In the choral portion theParode is the whole first statement of the cho-rus; a Stasimon, a song of the chorus withoutanapaests or trochees; a Commas, a lamenta-tion sung by chorus and actor in concert. Theparts of Tragedy to be used as formative ele-ments in the whole we have already men-tioned; the above are its parts from the point ofview of its quantity, or the separate sectionsinto which it is divided.

13

The next points after what we have said abovewill be these: (1) What is the poet to aim at, andwhat is he to avoid, in constructing his Plots?and (2) What are the conditions on which thetragic effect depends?

We assume that, for the finest form of Tragedy,the Plot must be not simple but complex; andfurther, that it must imitate actions arousingpity and fear, since that is the distinctive func-tion of this kind of imitation. It follows, there-fore, that there are three forms of Plot to beavoided. (1) A good man must not be seenpassing from happiness to misery, or (2) a badman from misery to happiness.

The first situation is not fear-inspiring or pite-ous, but simply odious to us. The second is themost untragic that can be; it has no one of therequisites of Tragedy; it does not appeal eitherto the human feeling in us, or to our pity, or toour fears. Nor, on the other hand, should (3) anextremely bad man be seen falling from happi-ness into misery. Such a story may arouse thehuman feeling in us, but it will not move us toeither pity or fear; pity is occasioned by unde-served misfortune, and fear by that of one likeourselves; so that there will be nothing eitherpiteous or fear-inspiring in the situation. Thereremains, then, the intermediate kind of person-age, a man not pre-eminently virtuous and just,whose misfortune, however, is brought uponhim not by vice and depravity but by someerror of judgement, of the number of those inthe enjoyment of great reputation and prosper-ity; e.g. Oedipus, Thyestes, and the men of noteof similar families. The perfect Plot, accord-ingly, must have a single, and not (as some tell

us) a double issue; the change in the hero's for-tunes must be not from misery to happiness,but on the contrary from happiness to misery;and the cause of it must lie not in any deprav-ity, but in some great error on his part; the manhimself being either such as we have described,or better, not worse, than that. Fact also con-firms our theory. Though the poets began byaccepting any tragic story that came to hand, inthese days the finest tragedies are always onthe story of some few houses, on that of Ale-meon, Oedipus, Orestes, Meleager, Thyestes,Telephus, or any others that may have beeninvolved, as either agents or sufferers, in somedeed of horror. The theoretically best tragedy,then, has a Plot of this description. The critics,therefore, are wrong who blame Euripides fortaking this line in his tragedies, and giving ma-ny of them an unhappy ending. It is, as wehave said, the right line to take. The best proofis this: on the stage, and in the public perform-ances, such plays, properly worked out, are

seen to be the most truly tragic; and Euripides,even if his elecution be faulty in every otherpoint, is seen to be nevertheless the most tragiccertainly of the dramatists. After this comes theconstruction of Plot which some rank first, onewith a double story (like the Odyssey) and anopposite issue for the good and the bad per-sonages. It is ranked as first only through theweakness of the audiences; the poets merelyfollow their public, writing as its wishes dic-tate. But the pleasure here is not that of Trag-edy. It belongs rather to Comedy, where thebitterest enemies in the piece (e.g. Orestes andAegisthus) walk off good friends at the end,with no slaying of any one by any one.

14

The tragic fear and pity may be aroused by theSpectacle; but they may also be aroused by thevery structure and incidents of the play—whichis the better way and shows the better poet. ThePlot in fact should be so framed that, even wit-hout seeing the things take place, he who sim-ply hears the account of them shall be filledwith horror and pity at the incidents; which isjust the effect that the mere recital of the storyin Oedipus would have on one. To produce thissame effect by means of the Spectacle is lessartistic, and requires extraneous aid. Those,however, who make use of the Spectacle to putbefore us that which is merely monstrous andnot productive of fear, are wholly out of touchwith Tragedy; not every kind of pleasureshould be required of a tragedy, but only itsown proper pleasure.

The tragic pleasure is that of pity and fear, andthe poet has to produce it by a work of imita-tion; it is clear, therefore, that the causes shouldbe included in the incidents of his story. Let ussee, then, what kinds of incident strike one ashorrible, or rather as piteous. In a deed of thisdescription the parties must necessarily be ei-ther friends, or enemies, or indifferent to oneanother. Now when enemy does it on enemy,there is nothing to move us to pity either in hisdoing or in his meditating the deed, except sofar as the actual pain of the sufferer is con-cerned; and the same is true when the partiesare indifferent to one another. Whenever thetragic deed, however, is done within the fam-ily—when murder or the like is done or medi-tated by brother on brother, by son on father,by mother on son, or son on mother—these arethe situations the poet should seek after. Thetraditional stories, accordingly, must be kept asthey are, e.g. the murder of Clytaemnestra byOrestes and of Eriphyle by Alcmeon. At the

same time even with these there is somethingleft to the poet himself; it is for him to devisethe right way of treating them. Let us explainmore clearly what we mean by 'the right way'.The deed of horror may be done by the doerknowingly and consciously, as in the old poets,and in Medea's murder of her children in Eu-ripides. Or he may do it, but in ignorance of hisrelationship, and discover that afterwards, asdoes the Oedipus in Sophocles. Here the deed isoutside the play; but it may be within it, likethe act of the Alcmeon in Astydamas, or that ofthe Telegonus in Ulysses Wounded. A third pos-sibility is for one meditating some deadly in-jury to another, in ignorance of his relationship,to make the discovery in time to draw back.These exhaust the possibilities, since the deedmust necessarily be either done or not done,and either knowingly or unknowingly.

The worst situation is when the personage iswith full knowledge on the point of doing the

deed, and leaves it undone. It is odious andalso (through the absence of suffering) un-tragic; hence it is that no one is made to act thusexcept in some few instances, e.g. Haemon andCreon in Antigone. Next after this comes theactual perpetration of the deed meditated. Abetter situation than that, however, is for thedeed to be done in ignorance, and the relation-ship discovered afterwards, since there is noth-ing odious in it, and the Discovery will serve toastound us. But the best of all is the last; whatwe have in Cresphontes, for example, whereMerope, on the point of slaying her son, recog-nizes him in time; in Iphigenia, where sister andbrother are in a like position; and in Helle, whe-re the son recognizes his mother, when on thepoint of giving her up to her enemy.

This will explain why our tragedies are re-stricted (as we said just now) to such a smallnumber of families. It was accident rather thanart that led the poets in quest of subjects to em-

body this kind of incident in their Plots. Theyare still obliged, accordingly, to have recourseto the families in which such horrors have oc-curred.

On the construction of the Plot, and the kind ofPlot required for Tragedy, enough has nowbeen said.

15

In the Characters there are four points to aimat. First and foremost, that they shall be good.There will be an element of character in theplay, if (as has been observed) what a person-age says or does reveals a certain moral pur-pose; and a good element of character, if thepurpose so revealed is good. Such goodness is

possible in every type of personage, even in awoman or a slave, though the one is perhaps aninferior, and the other a wholly worthless be-ing. The second point is to make them appro-priate. The Character before us may be, say,manly; but it is not appropriate in a femaleCharacter to be manly, or clever. The third is tomake them like the reality, which is not thesame as their being good and appropriate, inour sense of the term. The fourth is to makethem consistent and the same throughout; evenif inconsistency be part of the man before onefor imitation as presenting that form of charac-ter, he should still be consistently inconsistent.We have an instance of baseness of character,not required for the story, in the Menelaus inOrestes; of the incongruous and unbefitting inthe lamentation of Ulysses in Scylla, and in the(clever) speech of Melanippe; and of inconsis-tency in Iphigenia at Aulis, where Iphigenia thesuppliant is utterly unlike the later Iphigenia.The right thing, however, is in the Characters

just as in the incidents of the play to endeavouralways after the necessary or the probable; sothat whenever such-and-such a personage saysor does such-and-such a thing, it shall be theprobable or necessary outcome of his character;and whenever this incident follows on that, itshall be either the necessary or the probableconsequence of it. From this one sees (to digressfor a moment) that the Denouement alsoshould arise out of the plot itself, arid not de-pend on a stage-artifice, as in Medea, or in thestory of the (arrested) departure of the Greeksin the Iliad. The artifice must be reserved formatters outside the play—for past events be-yond human knowledge, or events yet to come,which require to be foretold or announced;since it is the privilege of the Gods to knoweverything. There should be nothing improb-able among the actual incidents. If it be un-avoidable, however, it should be outside thetragedy, like the improbability in the Oedipus ofSophocles. But to return to the Characters. As

Tragedy is an imitation of personages betterthan the ordinary man, we in our way shouldfollow the example of good portrait-painters,who reproduce the distinctive features of aman, and at the same time, without losing thelikeness, make him handsomer than he is. Thepoet in like manner, in portraying men quick orslow to anger, or with similar infirmities ofcharacter, must know how to represent them assuch, and at the same time as good men, asAgathon and Homer have represented Achilles.

All these rules one must keep in mind through-out, and further, those also for such points ofstage-effect as directly depend on the art of thepoet, since in these too one may often makemistakes. Enough, however, has been said onthe subject in one of our published writings.

16

Discovery in general has been explained al-ready. As for the species of Discovery, the firstto be noted is (1) the least artistic form of it, ofwhich the poets make most use through merelack of invention, Discovery by signs or marks.Of these signs some are congenital, like the'lance-head which the Earth-born have onthem', or 'stars', such as Carcinus brings in inhis Thyestes; others acquired after birth—theselatter being either marks on the body, e.g. scars,or external tokens, like necklaces, or to takeanother sort of instance, the ark in the Discov-ery in Tyro. Even these, however, admit of twouses, a better and a worse; the scar of Ulysses isan instance; the Discovery of him through it ismade in one way by the nurse and in anotherby the swineherds. A Discovery using signs asa means of assurance is less artistic, as indeedare all such as imply reflection; whereas one

bringing them in all of a sudden, as in the Bath-story, is of a better order. Next after these are (2)Discoveries made directly by the poet; whichare inartistic for that very reason; e.g. Orestes'Discovery of himself in Iphigenia: whereas hissister reveals who she is by the letter, Orestes ismade to say himself what the poet rather thanthe story demands. This, therefore, is not farremoved from the first-mentioned fault, sincehe might have presented certain tokens as well.Another instance is the 'shuttle's voice' in theTereus of Sophocles. (3) A third species is Dis-covery through memory, from a man's con-sciousness being awakened by something seenor heard. Thus in The Cyprioe of Dicaeogenes,the sight of the picture makes the man burstinto tears; and in the Tale of Alcinous, hearingthe harper Ulysses is reminded of the past andweeps; the Discovery of them being the result.(4) A fourth kind is Discovery through reason-ing; e.g. in The Choephoroe: 'One like me is here;there is no one like me but Orestes; he, there-

fore, must be here.' Or that which Polyidus theSophist suggested for Iphigenia; since it wasnatural for Orestes to reflect: 'My sister wassacrificed, and I am to be sacrificed like her.' Orthat in the Tydeus of Theodectes: 'I came to finda son, and am to die myself.' Or that in ThePhinidae: on seeing the place the women in-ferred their fate, that they were to die there,since they had also been exposed there. (5) The-re is, too, a composite Discovery arising frombad reasoning on the side of the other party. Aninstance of it is in Ulysses the False Messenger: hesaid he should know the bow—which he hadnot seen; but to suppose from that that hewould know it again (as though he had onceseen it) was bad reasoning. (6) The best of allDiscoveries, however, is that arising from theincidents themselves, when the great surprisecomes about through a probable incident, likethat in the Oedipus of Sophocles; and also inIphigenia; for it was not improbable that sheshould wish to have a letter taken home. These

last are the only Discoveries independent of theartifice of signs and necklaces. Next after themcome Discoveries through reasoning.

17

At the time when he is constructing his Plots,and engaged on the Diction in which they areworked out, the poet should remember (1) toput the actual scenes as far as possible beforehis eyes. In this way, seeing everything withthe vividness of an eye-witness as it were, hewill devise what is appropriate, and be leastlikely to overlook incongruities. This is shownby what was censured in Carcinus, the returnof Amphiaraus from the sanctuary; it wouldhave passed unnoticed, if it had not been actu-ally seen by the audience; but on the stage his

play failed, the incongruity of the incident of-fending the spectators. (2) As far as may be,too, the poet should even act his story with thevery gestures of his personages. Given the sa-me natural qualifications, he who feels theemotions to be described will be the most con-vincing; distress and anger, for instance, areportrayed most truthfully by one who is feelingthem at the moment. Hence it is that poetrydemands a man with special gift for it, or elseone with a touch of madness in him; the formercan easily assume the required mood, and thelatter may be actually beside himself with emo-tion. (3) His story, again, whether already madeor of his own making, he should first simplifyand reduce to a universal form, before proceed-ing to lengthen it out by the insertion of epi-sodes. The following will show how the uni-versal element in Iphigenia, for instance, may beviewed: A certain maiden having been offeredin sacrifice, and spirited away from her sacrifi-cers into another land, where the custom was to

sacrifice all strangers to the Goddess, she wasmade there the priestess of this rite. Long afterthat the brother of the priestess happened tocome; the fact, however, of the oracle havingfor a certain reason bidden him go thither, andhis object in going, are outside the Plot of theplay. On his coming he was arrested, and aboutto be sacrificed, when he revealed who hewas—either as Euripides puts it, or (as sug-gested by Polyidus) by the not improbable ex-clamation, 'So I too am doomed to be sacrificed,as my sister was'; and the disclosure led to hissalvation. This done, the next thing, after theproper names have been fixed as a basis for thestory, is to work in episodes or accessory inci-dents. One must mind, however, that the epi-sodes are appropriate, like the fit of madness inOrestes, which led to his arrest, and the purify-ing, which brought about his salvation. Inplays, then, the episodes are short; in epic po-etry they serve to lengthen out the poem. Theargument of the Odyssey is not a long one.

A certain man has been abroad many years;Poseidon is ever on the watch for him, and he isall alone. Matters at home too have come tothis, that his substance is being wasted and hisson's death plotted by suitors to his wife. Thenhe arrives there himself after his grievous suf-ferings; reveals himself, and falls on his ene-mies; and the end is his salvation and theirdeath. This being all that is proper to the Odys-sey, everything else in it is episode.

18

(4) There is a further point to be borne in mind.Every tragedy is in part Complication and inpart Denouement; the incidents before the ope-ning scene, and often certain also of those wit-hin the play, forming the Complication; and the

rest the Denouement. By Complication I meanall from the beginning of the story to the pointjust before the change in the hero's fortunes; byDenouement, all from the beginning of thechange to the end. In the Lynceus of Theodectes,for instance, the Complication includes, to-gether with the presupposed incidents, the sei-zure of the child and that in turn of the parents;and the Denouement all from the indictmentfor the murder to the end. Now it is right, whenone speaks of a tragedy as the same or not thesame as another, to do so on the ground beforeall else of their Plot, i.e. as having the same ornot the same Complication and Denouement.Yet there are many dramatists who, after agood Complication, fail in the Denouement.But it is necessary for both points of construc-tion to be always duly mastered. (5) There arefour distinct species of Tragedy—that being thenumber of the constituents also that have beenmentioned: first, the complex Tragedy, which isall Peripety and Discovery; second, the Tragedy

of suffering, e.g. the Ajaxes and Ixions; third, theTragedy of character, e.g. The Phthiotides andPeleus. The fourth constituent is that of 'Specta-cle', exemplified in The Phorcides, in Prometheus,and in all plays with the scene laid in the netherworld. The poet's aim, then, should be to com-bine every element of interest, if possible, orelse the more important and the major part ofthem. This is now especially necessary owing tothe unfair criticism to which the poet is sub-jected in these days. Just because there havebeen poets before him strong in the several spe-cies of tragedy, the critics now expect the oneman to surpass that which was the strong pointof each one of his predecessors. (6) One shouldalso remember what has been said more thanonce, and not write a tragedy on an epic bodyof incident (i.e. one with a plurality of stories init), by attempting to dramatize, for instance, theentire story of the Iliad. In the epic owing to itsscale every part is treated at proper length;with a drama, however, on the same story the

result is very disappointing. This is shown bythe fact that all who have dramatized the fall ofIlium in its entirety, and not part by part, likeEuripides, or the whole of the Niobe story, in-stead of a portion, like Aeschylus, either failutterly or have but ill success on the stage; forthat and that alone was enough to ruin a playby Agathon. Yet in their Peripeties, as also intheir simple plots, the poets I mean show won-derful skill in aiming at the kind of effect theydesire—a tragic situation that arouses the hu-man feeling in one, like the clever villain (e.g.Sisyphus) deceived, or the brave wrongdoerworsted. This is probable, however, only inAgathon's sense, when he speaks of the prob-ability of even improbabilities coming to pass.(7) The Chorus too should be regarded as oneof the actors; it should be an integral part of thewhole, and take a share in the action—thatwhich it has in Sophocles rather than in Eurip-ides. With the later poets, however, the songs ina play of theirs have no more to do with the

Plot of that than of any other tragedy. Hence itis that they are now singing intercalary pieces,a practice first introduced by Agathon. And yetwhat real difference is there between singingsuch intercalary pieces, and attempting to fit ina speech, or even a whole act, from one playinto another?

19

The Plot and Characters having been discussed,it remains to consider the Diction and Thought.As for the Thought, we may assume what issaid of it in our Art of Rhetoric, as it belongsmore properly to that department of inquiry.The Thought of the personages is shown ineverything to be effected by their language—inevery effort to prove or disprove, to arouse

emotion (pity, fear, anger, and the like), or tomaximize or minimize things. It is clear, also,that their mental procedure must be on the sa-me lines in their actions likewise, wheneverthey wish them to arouse pity or horror, orhave a look of importance or probability. Theonly difference is that with the act the impres-sion has to be made without explanation; whe-reas with the spoken word it has to be pro-duced by the speaker, and result from his lan-guage. What, indeed, would be the good of thespeaker, if things appeared in the required lighteven apart from anything he says?

As regards the Diction, one subject for inquiryunder this head is the turns given to the lan-guage when spoken; e.g. the difference be-tween command and prayer, simple statementand threat, question and answer, and so forth.The theory of such matters, however, belongsto Elocution and the professors of that art.Whether the poet knows these things or not, his

art as a poet is never seriously criticized on thataccount. What fault can one see in Homer's'Sing of the wrath, Goddess'?—which Prota-goras has criticized as being a command wherea prayer was meant, since to bid one do or notdo, he tells us, is a command. Let us pass overthis, then, as appertaining to another art, andnot to that of poetry.

20

The Diction viewed as a whole is made up ofthe following parts: the Letter (or ultimate ele-ment), the Syllable, the Conjunction, the Arti-cle, the Noun, the Verb, the Case, and theSpeech. (1) The Letter is an indivisible sound ofa particular kind, one that may become a factorin an intelligible sound. Indivisible sounds are

uttered by the brutes also, but no one of these isa Letter in our sense of the term. These elemen-tary sounds are either vowels, semivowels, ormutes. A vowel is a Letter having an audiblesound without the addition of another Letter. Asemivowel, one having an audible sound by theaddition of another Letter; e.g. S and R. A mute,one having no sound at all by itself, but becom-ing audible by an addition, that of one of theLetters which have a sound of some sort oftheir own; e.g. D and G. The Letters differ invarious ways: as produced by different con-formations or in different regions of the mouth;as aspirated, not aspirated, or sometimes oneand sometimes the other; as long, short, or ofvariable quantity; and further as having an acu-te grave, or intermediate accent.

The details of these matters we must leave tothe metricians. (2) A Syllable is a nonsignificantcomposite sound, made up of a mute and aLetter having a sound (a vowel or semivowel);

for GR, without an A, is just as much a Syllableas GRA, with an A. The various forms of theSyllable also belong to the theory of metre. (3)A Conjunction is (a) a non-significant soundwhich, when one significant sound is formableout of several, neither hinders nor aids the un-ion, and which, if the Speech thus formedstands by itself (apart from other Speeches)must not be inserted at the beginning of it; e.g.men, de, toi, de. Or (b) a non-significant soundcapable of combining two or more significantsounds into one; e.g. amphi, peri, etc. (4) An Ar-ticle is a non-significant sound marking thebeginning, end, or dividing-point of a Speech,its natural place being either at the extremitiesor in the middle. (5) A Noun or name is a com-posite significant sound not involving the ideaof time, with parts which have no significanceby themselves in it. It is to be remembered thatin a compound we do not think of the parts ashaving a significance also by themselves; in the

name 'Theodorus', for instance, the doron meansnothing to us.

(6) A Verb is a composite significant sound in-volving the idea of time, with parts which (justas in the Noun) have no significance by them-selves in it. Whereas the word 'man' or 'white'does not imply when, 'walks' and 'has walked'involve in addition to the idea of walking thatof time present or time past.

(7) A Case of a Noun or Verb is when the wordmeans 'of or 'to' a thing, and so forth, or for oneor many (e.g. 'man' and 'men'); or it may consistmerely in the mode of utterance, e.g. in ques-tion, command, etc. 'Walked?' and 'Walk!' areCases of the verb 'to walk' of this last kind. (8)A Speech is a composite significant sound, so-me of the parts of which have a certain signifi-cance by themselves. It may be observed that aSpeech is not always made up of Noun andVerb; it may be without a Verb, like the defini-tion of man; but it will always have some part

with a certain significance by itself. In theSpeech 'Cleon walks', 'Cleon' is an instance ofsuch a part. A Speech is said to be one in twoways, either as signifying one thing, or as aunion of several Speeches made into one byconjunction. Thus the Iliad is one Speech byconjunction of several; and the definition ofman is one through its signifying one thing.

21

Nouns are of two kinds, either (1) simple, i.e.made up of non-significant parts, like the wordge, or (2) double; in the latter case the wordmay be made up either of a significant and anon-significant part (a distinction which disap-pears in the compound), or of two significantparts. It is possible also to have triple, quadru-

ple or higher compounds, like most of our am-plified names; e.g.' Hermocaicoxanthus' andthe like.

Whatever its structure, a Noun must always beeither (1) the ordinary word for the thing, or (2)a strange word, or (3) a metaphor, or (4) anornamental word, or (5) a coined word, or (6) aword lengthened out, or (7) curtailed, or (8)altered in form. By the ordinary word I meanthat in general use in a country; and by a stran-ge word, one in use elsewhere. So that the sameword may obviously be at once strange andordinary, though not in reference to the samepeople; sigunos, for instance, is an ordinaryword in Cyprus, and a strange word with us.Metaphor consists in giving the thing a namethat belongs to something else; the transferencebeing either from genus to species, or from spe-cies to genus, or from species to species, or ongrounds of analogy. That from genus to speciesis eXemplified in 'Here stands my ship'; for

lying at anchor is the 'standing' of a particularkind of thing. That from species to genus in'Truly ten thousand good deeds has Ulysseswrought', where 'ten thousand', which is a par-ticular large number, is put in place of the ge-neric 'a large number'. That from species tospecies in 'Drawing the life with the bronze',and in 'Severing with the enduring bronze';where the poet uses 'draw' in the sense of 'se-ver' and 'sever' in that of 'draw', both wordsmeaning to 'take away' something. That fromanalogy is possible whenever there are fourterms so related that the second (B) is to thefirst (A), as the fourth (D) to the third (C); forone may then metaphorically put B in lieu of D,and D in lieu of B. Now and then, too, theyqualify the metaphor by adding on to it that towhich the word it supplants is relative. Thus acup (B) is in relation to Dionysus (A) what ashield (D) is to Ares (C). The cup accordinglywill be metaphorically described as the 'shieldof Dionysus' (D + A), and the shield as the 'cup

of Ares' (B + C). Or to take another instance: Asold age (D) is to life (C), so is evening (B) to day(A). One will accordingly describe evening (B)as the 'old age of the day' (D + A)—or by theEmpedoclean equivalent; and old age (D) as the'evening' or 'sunset of life'' (B + C). It may bethat some of the terms thus related have nospecial name of their own, but for all that theywill be metaphorically described in just thesame way. Thus to cast forth seed-corn is called'sowing'; but to cast forth its flame, as said ofthe sun, has no special name. This nameless act(B), however, stands in just the same relation toits object, sunlight (A), as sowing (D) to theseed-corn (C). Hence the expression in the poet,'sowing around a god-created flame' (D + A).There is also another form of qualified meta-phor. Having given the thing the alien name,one may by a negative addition deny of it oneof the attributes naturally associated with itsnew name. An instance of this would be to callthe shield not the 'cup of Ares,' as in the former

case, but a 'cup that holds no wine'. * * * A coinedword is a name which, being quite unknownamong a people, is given by the poet himself;e.g. (for there are some words that seem to beof this origin) hernyges for horns, and areter forpriest. A word is said to be lengthened out,when it has a short vowel made long, or anextra syllable inserted; e. g. polleos for poleos,Peleiadeo for Peleidon. It is said to be curtailed,when it has lost a part; e.g. kri, do, and ops inmia ginetai amphoteron ops. It is an altered word,when part is left as it was and part is of thepoet's making; e.g. dexiteron for dexion, in dex-iteron kata maxon.

The Nouns themselves (to whatever class theymay belong) are either masculines, feminines,or intermediates (neuter). All ending in N, P, S,or in the two compounds of this last, PS and X,are masculines. All ending in the invariablylong vowels, H and O, and in A among the vo-wels that may be long, are feminines. So that

there is an equal number of masculine and fe-minine terminations, as PS and X are the sameas S, and need not be counted. There is noNoun, however, ending in a mute or in eitherof the two short vowels, E and O. Only three(meli, kommi, peperi) end in I, and five in T. Theintermediates, or neuters, end in the variablevowels or in N, P, X.

22

The perfection of Diction is for it to be at onceclear and not mean. The clearest indeed is thatmade up of the ordinary words for things, butit is mean, as is shown by the poetry of Cleo-phon and Sthenelus. On the other hand theDiction becomes distinguished and non-prosaicby the use of unfamiliar terms, i.e. strange

words, metaphors, lengthened forms, and eve-rything that deviates from the ordinary modesof speech.—But a whole statement in suchterms will be either a riddle or a barbarism, ariddle, if made up of metaphors, a barbarism, ifmade up of strange words. The very natureindeed of a riddle is this, to describe a fact in animpossible combination of words (which can-not be done with the real names for things, butcan be with their metaphorical substitutes); e.g.'I saw a man glue brass on another with fire',and the like. The corresponding use of strangewords results in a barbarism.—A certain ad-mixture, accordingly, of unfamiliar terms isnecessary. These, the strange word, the meta-phor, the ornamental equivalent, etc.. will savethe language from seeming mean and prosaic,while the ordinary words in it will secure therequisite clearness. What helps most, however,to render the Diction at once clear and non-prosaic is the use of the lengthened, curtailed,and altered forms of words. Their deviation

from the ordinary words will, by making thelanguage unlike that in general use give it anon-prosaic appearance; and their having muchin common with the words in general use willgive it the quality of clearness. It is not right,then, to condemn these modes of speech, andridicule the poet for using them, as some havedone; e.g. the elder Euclid, who said it was easyto make poetry if one were to be allowed tolengthen the words in the statement itself asmuch as one likes—a procedure he caricaturedby reading 'Epixarhon eidon Marathonade Badi—gonta, and ouk han g' eramenos ton ekeinou helleboron as verses. A too apparent use of theselicences has certainly a ludicrous effect, butthey are not alone in that; the rule of modera-tion applies to all the constituents of the poeticvocabulary; even with metaphors, strangewords, and the rest, the effect will be the same,if one uses them improperly and with a view toprovoking laughter. The proper use of them is avery different thing. To realize the difference

one should take an epic verse and see how itreads when the normal words are introduced.The same should be done too with the strangeword, the metaphor, and the rest; for one hasonly to put the ordinary words in their place tosee the truth of what we are saying. The sameiambic, for instance, is found in Aeschylus andEuripides, and as it stands in the former it is apoor line; whereas Euripides, by the change ofa single word, the substitution of a strange forwhat is by usage the ordinary word, has madeit seem a fine one. Aeschylus having said in hisPhiloctetes:phagedaina he mon sarkas hesthiei podos

Euripides has merely altered the hesthiei hereinto thoinatai. Or supposenun de m' heon holigos te kai outidanos kai haeikos

to be altered by the substitution of the ordinarywords into

nun de m' heon mikros te kai hasthenikos kai haei-dos

Or the linediphron haeikelion katatheis olingen te trapexan

intodiphron moxtheron katatheis mikran te trapexan

Or heiones boosin into heiones kraxousin. Addto this that Ariphrades used to ridicule the tra-gedians for introducing expressions unknownin the language of common life, doeaton hapo(for apo domaton), sethen, hego de nin, Achilleosperi (for peri Achilleos), and the like. The merefact of their not being in ordinary speech givesthe Diction a non-prosaic character; but Ariph-rades was unaware of that. It is a great thing,indeed, to make a proper use of these poeticalforms, as also of compounds and strangewords. But the greatest thing by far is to be amaster of metaphor. It is the one thing thatcannot be learnt from others; and it is also a

sign of genius, since a good metaphor impliesan intuitive perception of the similarity in dis-similars.

Of the kinds of words we have enumerated itmay be observed that compounds are most inplace in the dithyramb, strange words in he-roic, and metaphors in iambic poetry. Heroicpoetry, indeed, may avail itself of them all. Butin iambic verse, which models itself as far aspossible on the spoken language, only thosekinds of words are in place which are allowablealso in an oration, i.e. the ordinary word, themetaphor, and the ornamental equivalent.

Let this, then, suffice as an account of Tragedy,the art imitating by means of action on the sta-ge.

23

As for the poetry which merely narrates, orimitates by means of versified language (with-out action), it is evident that it has severalpoints in common with Tragedy.

I. The construction of its stories should clearlybe like that in a drama; they should be based ona single action, one that is a complete whole initself, with a beginning, middle, and end, so asto enable the work to produce its own properpleasure with all the organic unity of a livingcreature. Nor should one suppose that there isanything like them in our usual histories. Ahistory has to deal not with one action, but withone period and all that happened in that to oneor more persons, however disconnected theseveral events may have been. Just as twoevents may take place at the same time, e.g. thesea-fight off Salamis and the battle with the

Carthaginians in Sicily, without converging tothe same end, so too of two consecutive eventsone may sometimes come after the other withno one end as their common issue. Neverthe-less most of our epic poets, one may say, ignorethe distinction.

Herein, then, to repeat what we have said be-fore, we have a further proof of Homer's mar-vellous superiority to the rest. He did not at-tempt to deal even with the Trojan war in itsentirety, though it was a whole with a definitebeginning and end—through a feeling appar-ently that it was too long a story to be taken inin one view, or if not that, too complicated fromthe variety of incident in it. As it is, he has sin-gled out one section of the whole; many of theother incidents, however, he brings in as epi-sodes, using the Catalogue of the Ships, forinstance, and other episodes to relieve the uni-formity of his narrative. As for the other epicpoets, they treat of one man, or one period; or

else of an action which, although one, has amultiplicity of parts in it. This last is what theauthors of the Cypria and Little Iliad have done.And the result is that, whereas the Iliad or Od-yssey supplies materials for only one, or at mosttwo tragedies, the Cypria does that for several,and the Little Iliad for more than eight: for anAdjudgment of Arms, a Philoctetes, a Neoptolemus,a Eurypylus, a Ulysses as Beggar, a LaconianWomen, a Fall of Ilium, and a Departure of theFleet; as also a Sinon, and Women of Troy.

24

II. Besides this, Epic poetry must divide into thesame species as Tragedy; it must be either sim-ple or complex, a story of character or one ofsuffering. Its parts, too, with the exception of

Song and Spectacle, must be the same, as it re-quires Peripeties, Discoveries, and scenes ofsuffering just like Tragedy. Lastly, the Thoughtand Diction in it must be good in their way. Allthese elements appear in Homer first; and hehas made due use of them. His two poems areeach examples of construction, the Iliad simpleand a story of suffering, the Odyssey complex(there is Discovery throughout it) and a story ofcharacter. And they are more than this, since inDiction and Thought too they surpass all otherpoems.

There is, however, a difference in the Epic ascompared with Tragedy, (1) in its length, and(2) in its metre. (1) As to its length, the limitalready suggested will suffice: it must be possi-ble for the beginning and end of the work to betaken in in one view—a condition which will befulfilled if the poem be shorter than the oldepics, and about as long as the series of trage-dies offered for one hearing. For the extension

of its length epic poetry has a special advan-tage, of which it makes large use. In a play onecannot represent an action with a number ofparts going on simultaneously; one is limited tothe part on the stage and connected with theactors. Whereas in epic poetry the narrativeform makes it possible for one to describe anumber of simultaneous incidents; and these, ifgermane to the subject, increase the body of thepoem. This then is a gain to the Epic, tending togive it grandeur, and also variety of interestand room for episodes of diverse kinds. Uni-formity of incident by the satiety it soon createsis apt to ruin tragedies on the stage. (2) As forits metre, the heroic has been assigned it fromexperience; were any one to attempt a narrativepoem in some one, or in several, of the othermetres, the incongruity of the thing would beapparent. The heroic; in fact is the gravest andweightiest of metres—which is what makes itmore tolerant than the rest of strange wordsand metaphors, that also being a point in which

the narrative form of poetry goes beyond allothers. The iambic and trochaic, on the otherhand, are metres of movement, the one repre-senting that of life and action, the other that ofthe dance. Still more unnatural would it ap-pear, it one were to write an epic in a medley ofmetres, as Chaeremon did. Hence it is that noone has ever written a long story in any butheroic verse; nature herself, as we have said,teaches us to select the metre appropriate tosuch a story.

Homer, admirable as he is in every other re-spect, is especially so in this, that he aloneamong epic poets is not unaware of the part tobe played by the poet himself in the poem. Thepoet should say very little in propria persona,as he is no imitator when doing that. Whereasthe other poets are perpetually coming forwardin person, and say but little, and that only hereand there, as imitators, Homer after a brief pre-face brings in forthwith a man, a woman, or

some other Character—no one of them charac-terless, but each with distinctive characteristics.

The marvellous is certainly required in Trag-edy. The Epic, however, affords more openingfor the improbable, the chief factor in the mar-vellous, because in it the agents are not visiblybefore one. The scene of the pursuit of Hectorwould be ridiculous on the stage—the Greekshalting instead of pursuing him, and Achillesshaking his head to stop them; but in the poemthe absurdity is overlooked. The marvellous,however, is a cause of pleasure, as is shown bythe fact that we all tell a story with additions, inthe belief that we are doing our hearers a plea-sure.

Homer more than any other has taught the restof us the art of framing lies in the right way. Imean the use of paralogism. Whenever, if A isor happens, a consequent, B, is or happens,men's notion is that, if the B is, the A also is—but that is a false conclusion. Accordingly, if A

is untrue, but there is something else, B, that onthe assumption of its truth follows as its conse-quent, the right thing then is to add on the B.Just because we know the truth of the conse-quent, we are in our own minds led on to theerroneous inference of the truth of the antece-dent. Here is an instance, from the Bath-story inthe Odyssey.

A likely impossibility is always preferable to anunconvincing possibility. The story should ne-ver be made up of improbable incidents; thereshould be nothing of the sort in it. If, however,such incidents are unavoidable, they should beoutside the piece, like the hero's ignorance inOedipus of the circumstances of Lams' death;not within it, like the report of the Pythian ga-mes in Electra, or the man's having come to My-sia from Tegea without uttering a word on theway, in The Mysians. So that it is ridiculous tosay that one's Plot would have been spoilt wit-hout them, since it is fundamentally wrong to

make up such Plots. If the poet has taken such aPlot, however, and one sees that he might haveput it in a more probable form, he is guilty ofabsurdity as well as a fault of art. Even in theOdyssey the improbabilities in the setting-ashore of Ulysses would be clearly intolerablein the hands of an inferior poet. As it is, thepoet conceals them, his other excellences veil-ing their absurdity. Elaborate Diction, however,is required only in places where there is no ac-tion, and no Character or Thought to be re-vealed. Where there is Character or Thought,on the other hand, an over-ornate Diction tendsto obscure them.

25

As regards Problems and their Solutions, onemay see the number and nature of the assump-tions on which they proceed by viewing thematter in the following way. (1) The poet beingan imitator just like the painter or other makerof likenesses, he must necessarily in all in-stances represent things in one or other of threeaspects, either as they were or are, or as theyare said or thought to be or to have been, or asthey ought to be. (2) All this he does in lan-guage, with an admixture, it may be, of strangewords and metaphors, as also of the variousmodified forms of words, since the use of theseis conceded in poetry. (3) It is to be remem-bered, too, that there is not the same kind ofcorrectness in poetry as in politics, or indeedany other art. There is, however, within thelimits of poetry itself a possibility of two kindsof error, the one directly, the other only acci-

dentally connected with the art. If the poetmeant to describe the thing correctly, and failedthrough lack of power of expression, his artitself is at fault. But if it was through his havingmeant to describe it in some incorrect way (e.g.to make the horse in movement have both rightlegs thrown forward) that the technical error(one in a matter of, say, medicine or some otherspecial science), or impossibilities of whateverkind they may be, have got into his description,his error in that case is not in the essentials ofthe poetic art. These, therefore, must be thepremisses of the Solutions in answer to thecriticisms involved in the Problems.

I. As to the criticisms relating to the poet's artitself. Any impossibilities there may be in hisdescriptions of things are faults. But from an-other point of view they are justifiable, if theyserve the end of poetry itself—if (to assumewhat we have said of that end) they make theeffect of some portion of the work more as-

tounding. The Pursuit of Hector is an instancein point. If, however, the poetic end might havebeen as well or better attained without sacrificeof technical correctness in such matters, theimpossibility is not to be justified, since thedescription should be, if it can, entirely freefrom error. One may ask, too, whether the erroris in a matter directly or only accidentally con-nected with the poetic art; since it is a lessererror in an artist not to know, for instance, thatthe hind has no horns, than to produce an un-recognizable picture of one.

II. If the poet's description be criticized as nottrue to fact, one may urge perhaps that the ob-ject ought to be as described—an answer likethat of Sophocles, who said that he drew menas they ought to be, and Euripides as they we-re. If the description, however, be neither truenor of the thing as it ought to be, the answermust be then, that it is in accordance with opin-ion. The tales about Gods, for instance, may be

as wrong as Xenophanes thinks, neither truenor the better thing to say; but they are cer-tainly in accordance with opinion. Of otherstatements in poetry one may perhaps say, notthat they are better than the truth, but that thefact was so at the time; e.g. the description ofthe arms: 'their spears stood upright, butt-endupon the ground'; for that was the usual way offixing them then, as it is still with the Illyrians.As for the question whether something said ordone in a poem is morally right or not, in deal-ing with that one should consider not only theintrinsic quality of the actual word or deed, butalso the person who says or does it, the personto whom he says or does it, the time, the means,and the motive of the agent—whether he doesit to attain a greater good, or to avoid a greaterevil.

III. Other criticisms one must meet by consider-ing the language of the poet: (1) by the assump-tion of a strange word in a passage like oureas

men proton, where by oureas Homer may per-haps mean not mules but sentinels. And in say-ing of Dolon, hos p e toi eidos men heen kakos, hismeaning may perhaps be, not that Dolon'sbody was deformed, but that his face was ugly,as eneidos is the Cretan word for handsome-faced. So, too, goroteron de keraie may mean not'mix the wine stronger', as though for topers,but 'mix it quicker'. (2) Other expressions inHomer may be explained as metaphorical; e.g.in halloi men ra theoi te kai aneres eudon (hapantes)pannux as compared with what he tells us at thesame time, e toi hot hes pedion to Troikon hathre-seien, aulon suriggon *te homadon* the word ha-pantes 'all', is metaphorically put for 'many',since 'all' is a species of 'many '. So also his oied' ammoros is metaphorical, the best knownstanding 'alone'. (3) A change, as Hippias sug-gested, in the mode of reading a word will sol-ve the difficulty in didomen de oi, and to men oukataputhetai hombro. (4) Other difficulties maybe solved by another punctuation; e.g. in Em-

pedocles, aipsa de thnet ephyonto, ta prin mathonathanata xora te prin kekreto. Or (5) by the as-sumption of an equivocal term, as in parochekende pleo nux, where pleo in equivocal. Or (6) by anappeal to the custom of language. Wine-and-water we call 'wine'; and it is on the same prin-ciple that Homer speaks of a knemis neoteuktoukassiteroio, a 'greave of new-wrought tin.' Aworker in iron we call a 'brazier'; and it is onthe same principle that Ganymede is describedas the 'wine-server' of Zeus, though the Godsdo not drink wine. This latter, however, may bean instance of metaphor. But whenever also aword seems to imply some contradiction, it isnecessary to reflect how many ways there maybe of understanding it in the passage in ques-tion; e.g. in Homer's te r' hesxeto xalkeon hegxosone should consider the possible senses of 'wasstopped there'—whether by taking it in thissense or in that one will best avoid the fault ofwhich Glaucon speaks: 'They start with someimprobable presumption; and having so de-

creed it themselves, proceed to draw infer-ences, and censure the poet as though he hadactually said whatever they happen to believe,if his statement conflicts with their own notionof things.' This is how Homer's silence aboutIcarius has been treated. Starting with, the no-tion of his having been a Lacedaemonian, thecritics think it strange for Telemachus not tohave met him when he went to Lacedaemon.Whereas the fact may have been as the Cephal-lenians say, that the wife of Ulysses was of aCephallenian family, and that her father's namewas Icadius, not Icarius. So that it is probably amistake of the critics that has given rise to theProblem.

Speaking generally, one has to justify (1) theImpossible by reference to the requirements ofpoetry, or to the better, or to opinion. For thepurposes of poetry a convincing impossibilityis preferable to an unconvincing possibility;and if men such as Zeuxis depicted be impossi-

ble, the answer is that it is better they should belike that, as the artist ought to improve on hismodel. (2) The Improbable one has to justifyeither by showing it to be in accordance withopinion, or by urging that at times it is not im-probable; for there is a probability of thingshappening also against probability. (3) The con-tradictions found in the poet's language oneshould first test as one does an opponent's con-futation in a dialectical argument, so as to seewhether he means the same thing, in the samerelation, and in the same sense, before admit-ting that he has contradicted either somethinghe has said himself or what a man of soundsense assumes as true. But there is no possibleapology for improbability of Plot or depravityof character, when they are not necessary andno use is made of them, like the improbabilityin the appearance of Aegeus in Medea and thebaseness of Menelaus in Orestes.

The objections, then, of critics start with faultsof five kinds: the allegation is always that so-mething in either (1) impossible, (2) improb-able, (3) corrupting, (4) contradictory, or (5)against technical correctness. The answers tothese objections must be sought under one orother of the above-mentioned heads, which aretwelve in number.

26

The question may be raised whether the epic orthe tragic is the higher form of imitation. It maybe argued that, if the less vulgar is the higher,and the less vulgar is always that which ad-dresses the better public, an art addressing anyand every one is of a very vulgar order. It is abelief that their public cannot see the meaning,

unless they add something themselves, thatcauses the perpetual movements of the per-formers—bad flute-players, for instance, rollingabout, if quoit-throwing is to be represented,and pulling at the conductor, if Scylla is thesubject of the piece. Tragedy, then, is said to bean art of this order—to be in fact just what thelater actors were in the eyes of their predeces-sors; for Myrmiscus used to call Callippides'the ape', because he thought he so overactedhis parts; and a similar view was taken of Pin-darus also. All Tragedy, however, is said tostand to the Epic as the newer to the olderschool of actors. The one, accordingly, is said toaddress a cultivated 'audience, which does notneed the accompaniment of gesture; the other,an uncultivated one. If, therefore, Tragedy is avulgar art, it must clearly be lower than theEpic.

The answer to this is twofold. In the first place,one may urge (1) that the censure does not

touch the art of the dramatic poet, but only thatof his interpreter; for it is quite possible to over-do the gesturing even in an epic recital, as didSosistratus, and in a singing contest, as didMnasitheus of Opus. (2) That one should notcondemn all movement, unless one means tocondemn even the dance, but only that of igno-ble people—which is the point of the criticismpassed on Callippides and in the present dayon others, that their women are not like gen-tlewomen. (3) That Tragedy may produce itseffect even without movement or action in justthe same way as Epic poetry; for from the merereading of a play its quality may be seen. Sothat, if it be superior in all other respects, thiselement of inferiority is not a necessary part ofit.

In the second place, one must remember (1)that Tragedy has everything that the Epic has(even the epic metre being admissible), togetherwith a not inconsiderable addition in the shape

of the Music (a very real factor in the pleasureof the drama) and the Spectacle. (2) That itsreality of presentation is felt in the play as read,as well as in the play as acted. (3) That thetragic imitation requires less space for the at-tainment of its end; which is a great advantage,since the more concentrated effect is morepleasurable than one with a large admixture oftime to dilute it—consider the Oedipus ofSophocles, for instance, and the effect of ex-panding it into the number of lines of the Iliad.(4) That there is less unity in the imitation ofthe epic poets, as is proved by the fact that anyone work of theirs supplies matter for severaltragedies; the result being that, if they takewhat is really a single story, it seems curt whenbriefly told, and thin and waterish when on thescale of length usual with their verse. In sayingthat there is less unity in an epic, I mean an epicmade up of a plurality of actions, in the sameway as the Iliad and Odyssey have many suchparts, each one of them in itself of some magni-

tude; yet the structure of the two Homeric po-ems is as perfect as can be, and the action inthem is as nearly as possible one action. If, then,Tragedy is superior in these respects, and alsobesides these, in its poetic effect (since the twoforms of poetry should give us, not any orevery pleasure, but the very special kind wehave mentioned), it is clear that, as attaining thepoetic effect better than the Epic, it will be thehigher form of art.

So much for Tragedy and Epic poetry—for the-se two arts in general and their species; thenumber and nature of their constituent parts;the causes of success and failure in them; theObjections of the critics, and the Solutions inanswer to them.