Nouns are of two kinds, either (1) simple, i.e. made up of non-significant parts, like the word ge, or (2) double; in the latter case the word may be made up either of a significant and a non-significant part (a distinction which disappears in the compound), or of two significant parts. It is possible also to have triple, quadruple or higher compounds, like most of our amplified names; e.g.' Hermocaicoxanthus' and the like.

Whatever its structure, a Noun must always be either (1) the ordinary word for the thing, or (2) a strange word, or (3) a metaphor, or (4) an ornamental word, or (5) a coined word, or (6) a word lengthened out, or (7) curtailed, or (8) altered in form. By the ordinary word I mean that in general use in a country; and by a strange word, one in use elsewhere. So that the same word may obviously be at once strange and ordinary, though not in reference to the same people; sigunos, for instance, is an ordinary word in Cyprus, and a strange word with us. Metaphor consists in giving the thing a name that belongs to something else; the transference being either from genus to species, or from species to genus, or from species to species, or on grounds of analogy. That from genus to species i.e.emplified in 'Here stands my ship'; for lying at anchor is the 'standing' of a particular kind of thing. That from species to genus in 'Truly ten thousand good deeds has Ulysses wrought', where 'ten thousand', which is a particular large number, is put in place of the generic 'a large number'. That from species to species in 'Drawing the life with the bronze', and in 'Severing with the enduring bronze'; where the poet uses 'draw' in the sense of 'sever' and 'sever' in that of 'draw', both words meaning to 'take away' something. That from analogy is possible whenever there are four terms so related that the second (B) is to the first (A), as the fourth (D) to the third (C); for one may then metaphorically put B in lieu of D, and D in lieu of B. Now and then, too, they qualify the metaphor by adding on to it that to which the word it supplants is relative. Thus a cup (B) is in relation to Dionysus (A) what a shield (D) is to Ares (C). The cup accordingly will be metaphorically described as the 'shield of Dionysus' (D + A), and the shield as the 'cup of Ares' (B + C). Or to take another instance: As old age (D) is to life (C), so i.e.ening (B) to day (A). One will accordingly describe evening (B) as the 'old age of the day' (D + A)—or by the Empedoclean equivalent; and old age (D) as the 'evening' or 'sunset of life' ' (B + C). It may be that some of the terms thus related have no special name of their own, but for all that they will be metaphorically described in just the same way. Thus to cast forth seed-corn is called 'sowing'; but to cast forth its flame, as said of the sun, has no special name. This nameless act (B), however, stands in just the same relation to its object, sunlight (A), as sowing (D) to the seed-corn (C). Hence the expression in the poet, 'sowing around a god-created flame' (D + A). There is also another form of qualified metaphor. Having given the thing the alien name, one may by a negative addition deny of it one of the attributes naturally associated with its new name. An instance of this would be to call the shield not the 'cup of Ares,' as in the former case, but a 'cup that holds no wine'. * * * A coined word is a name which, being quite unknown among a people, is given by the poet himself; e.g. (for there are some words that seem to be of this origin) hernyges for horns, and areter for priest. A word is said to be lengthened out, when it has a short vowel made long, or an extra 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 in mia ginetai amphoteron ops. It is an altered word, when part is left as it was and part is of the poet's making; e.g. dexiteron for dexion, in dexiteron kata maxon.

The Nouns themselves (to whatever class they may 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 invariably long vowels, H and O, and in A among the vowels that may be long, are feminines. So that there is an equal number of masculine and feminine terminations, as PS and X are the same as S, and need not be counted. There is no Noun, however, ending in a mute or i.e.ther of the two short vowels, E and O. Only three (meli, kommi, peperi) end in I, and five in T. The intermediates, or neuters, end in the variable vowels or in N, P, X.


The perfection of Diction is for it to be at once clear and not mean. The clearest indeed is that made up of the ordinary words for things, but it is mean, as is shown by the poetry of Cleophon and Sthenelus. On the other hand the Diction becomes distinguished and non-prosaic by the use of unfamiliar terms, i.e. strange words, metaphors, lengthened forms, and everything that deviates from the ordinary modes of speech.—But a whole statement in such terms will be either a riddle or a barbarism, a riddle, if made up of metaphors, a barbarism, if made up of strange words. The very nature indeed of a riddle is this, to describe a fact in an impossible combination of words (which cannot be done with the real names for things, but can be with their metaphorical substitutes); e.g. 'I saw a man glue brass on another with fire', and the like. The corresponding use of strange words results in a barbarism.—A certain admixture, accordingly, of unfamiliar terms is necessary. These, the strange word, the metaphor, the ornamental equivalent, etc.. will save the language from seeming mean and prosaic, while the ordinary words in it will secure the requisite 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 the language unlike that in general use.g.ve it a non-prosaic appearance; and their having much in common with the words in general use will give it the quality of clearness. It is not right, then, to condemn these modes of speech, and ridicule the poet for using them, as some have done; e.g. the elder Euclid, who said it was easy to make poetry if one were to be allowed to lengthen the words in the statement itself as much as one likes—a procedure he caricatured by reading 'Epixarhon eidon Marathonade Badi—gonta, and ouk han g' eramenos ton ekeinou helle boron as verses. A too apparent use of these licences has certainly a ludicrous effect, but they are not alone in that; the rule of moderation applies to all the constituents of the poetic vocabulary; even with metaphors, strange words, and the rest, the effect will be the same, if one uses them improperly and with a view to provoking laughter. The proper use of them is a very different thing. To realize the difference one should take an epic verse and see how it reads when the normal words are introduced. The same should be done too with the strange word, the metaphor, and the rest; for one has only to put the ordinary words in their place to see the truth of what we are saying. The same iambic, for instance, is found in Aeschylus and Euripides, and as it stands in the former it is a poor line; whereas Euripides, by the change of a single word, the substitution of a strange for what is by usage the ordinary word, has made it seem a fine one. Aeschylus having said in his Philoctetes:

phagedaina he mon sarkas hesthiei podos

Euripides has merely altered the hesthiei here into thoinatai. Or suppose

nun de m' heon holigos te kai outidanos kai haeikos

to be altered by the substitution of the ordinary words into

 nun de m' heon mikros te kai hasthenikos kai haeidos

Or the line

diphron haeikelion katatheis olingen te trapexan


diphron moxtheron katatheis mikran te trapexan

Or heiones boosin into heiones kraxousin. Add to this that Ariphrades used to ridicule the tragedians for introducing expressions unknown in the language of common life, doeaton hapo (for apo domaton), sethen, hego de nin, Achilleos peri (for peri Achilleos), and the like. The mere fact of their not being in ordinary speech gives the Diction a non-prosaic character; but Ariphrades was unaware of that. It is a great thing, indeed, to make a proper use of these poetical forms, as also of compounds and strange words. But the greatest thing by far is to be a master of metaphor. It is the one thing that cannot be learnt from others; and it is also a sign of genius, since a good metaphor implies an intuitive perception of the similarity in dissimilars.

Of the kinds of words we have enumerated it may be observed that compounds are most in place in the dithyramb, strange words in heroic, and metaphors in iambic poetry. Heroic poetry, indeed, may avail itself of them all. But in iambic verse, which models itself as far as possible on the spoken language, only those kinds of words are in place which are allowable also in an oration, i.e. the ordinary word, the metaphor, and the ornamental equivalent.

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


As for the poetry which merely narrates, or imitates by means of versified language (without action), it i.e.ident that it has several points in common with Tragedy.

I. The construction of its stories should clearly be like that in a drama; they should be based on a single action, one that is a complete whole in itself, with a beginning, middle, and end, so as to enable the work to produce its own proper pleasure with all the organic unity of a living creature. Nor should one suppose that there is anything like them in our usual histories. A history has to deal not with one action, but with one period and all that happened in that to one or more persons, however disconnected the several events may have been. Just as two events may take place at the same time, e.g. the sea-fight off Salamis and the battle with the Carthaginians in Sicily, without converging to the same end, so too of two consecutive events one may sometimes come after the other with no one end as their common issue. Nevertheless most of our epic poets, one may say, ignore the distinction.

Herein, then, to repeat what we have said before, we have a further proof of Homer's marvellous superiority to the rest. He did not attempt to deal even with the Trojan war in its entirety, though it was a whole with a definite beginning and end—through a feeling apparently that it was too long a story to be taken in in one view, or if not that, too complicated from the variety of incident in it. As it is, he has singled out one section of the whole; many of the other incidents, however, he brings in as episodes, using the Catalogue of the Ships, for instance, and other episodes to relieve the uniformity of his narrative. As for the other epic poets, they treat of one man, or one period; or else of an action which, although one, has a multiplicity of parts in it. This last is what the authors of the Cypria and Little Iliad have done. And the result is that, whereas the Iliad or Odyssey supplies materials for only one, or at most two tragedies, the Cypria does that for several, and the Little Iliad for more than eight: for an Adjudgment of Arms, a Philoctetes, a Neoptolemus, a Eurypylus, a Ulysses as Beggar, a Laconian Women, a Fall of Ilium, and a Departure of the Fleet; as also a Sinon, and Women of Troy.


II. Besides this, Epic poetry must divide into the same species as Tragedy; it must be either simple or complex, a story of character or one of suffering. Its parts, too, with the exception of Song and Spectacle, must be the same, as it requires Peripeties, Discoveries, and scenes of suffering just like Tragedy. Lastly, the Thought and Diction in it must be good in their way. All these elements appear in Homer first; and he has made due use of them. His two poems are each examples of construction, the Iliad simple and a story of suffering, the Odyssey complex (there is Discovery throughout it) and a story of character. And they are more than this, since in Diction and Thought too they surpass all other poems.

There is, however, a difference in the Epic as compared with Tragedy, (1) in its length, and (2) in its metre. (1) As to its length, the limit already suggested will suffice: it must be possible for the beginning and end of the work to be taken in in one view—a condition which will be fulfilled if the poem be shorter than the old epics, and about as long as the series of tragedies offered for one hearing. For the extension of its length epic poetry has a special advantage, of which it makes large use. In a play one cannot represent an action with a number of parts going on simultaneously; one is limited to the part on the stage and connected with the actors. Whereas i.e.ic poetry the narrative form makes it possible for one to describe a number of simultaneous incidents; and these, if germane to the subject, increase the body of the poem. This then is a gain to the Epic, tending to give it grandeur, and also variety of interest and room for episodes of diverse kinds. Uniformity of incident by the satiety it soon creates is apt to ruin tragedies on the stage. (2) As for its metre, the heroic has been assigned it from experience; were any one to attempt a narrative poem in some one, or in several, of the other metres, the incongruity of the thing would be apparent. The heroic; in fact is the gravest and weightiest of metres—which is what makes it more tolerant than the rest of strange words and metaphors, that also being a point in which the narrative form of poetry goes beyond all others. The iambic and trochaic, on the other hand, are metres of movement, the one representing that of life and action, the other that of the dance. Still more unnatural would it appear, it one were to write an epic in a medley of metres, as Chaeremon did. Hence it is that no one has ever written a long story in any but heroic verse; nature herself, as we have said, teaches us to select the metre appropriate to such a story.

Homer, admirable as he is i.e.ery other respect, i.e.pecially so in this, that he alone among epic poets is not unaware of the part to be played by the poet himself in the poem. The poet should say very little in propria persona, as he is no imitator when doing that. Whereas the other poets are perpetually coming forward in person, and say but little, and that only here and there, as imitators, Homer after a brief preface brings in forthwith a man, a woman, or some other Character—no one of them characterless, but each with distinctive characteristics.

The marvellous is certainly required in Tragedy. The Epic, however, affords more opening for the improbable, the chief factor in the marvellous, because in it the agents are not visibly before one. The scene of the pursuit of Hector would be ridiculous on the stage—the Greeks halting instead of pursuing him, and Achilles shaking his head to stop them; but in the poem the absurdity is overlooked. The marvellous, however, is a cause of pleasure, as is shown by the fact that we all tell a story with additions, in the belief that we are doing our hearers a pleasure.

Homer more than any other has taught the rest of us the art of framing lies in the right way. I mean the use of paralogism. Whenever, if A is or 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 on the assumption of its truth follows as its consequent, the right thing then is to add on the B. Just because we know the truth of the consequent, we are in our own minds led on to the erroneous inference of the truth of the antecedent. Here is an instance, from the Bath-story in the Odyssey.

A likely impossibility is always preferable to an unconvincing possibility. The story should never be made up of improbable incidents; there should be nothing of the sort in it. If, however, such incidents are unavoidable, they should be outside the piece, like the hero's ignorance in Oedipus of the circumstances of Lams' death; not within it, like the report of the Pythian games in Electra, or the man's having come to Mysia from Tegea without uttering a word on the way, in The Mysians. So that it is ridiculous to say that one's Plot would have been spoilt without them, since it is fundamentally wrong to make up such Plots. If the poet has taken such a Plot, however, and one sees that he might have put it in a more probable form, he is guilty of absurdity as well as a fault of art. Even in the Odyssey the improbabilities in the setting-ashore of Ulysses would be clearly intolerable in the hands of an inferior poet. As it is, the poet conceals them, his other excellences veiling their absurdity. Elaborate Diction, however, is required only in places where there is no action, and no Character or Thought to be revealed. Where there is Character or Thought, on the other hand, an over-ornate Diction tends to obscure them.

As regards Problems and their Solutions, one may see the number and nature of the assumptions on which they proceed by viewing the matter in the following way. (1) The poet being an imitator just like the painter or other maker of likenesses, he must necessarily in all instances represent things in one or other of three aspects, either as they were or are, or as they are said or thought to be or to have been, or as they ought to be. (2) All this he does in language, with an admixture, it may be, of strange words and metaphors, as also of the various modified forms of words, since the use of these is conceded in poetry. (3) It is to be remembered, too, that there is not the same kind of correctness in poetry as in politics, or indeed any other art. There is, however, within the limits of poetry itself a possibility of two kinds of error, the one directly, the other only accidentally connected with the art. If the poet meant to describe the thing correctly, and failed through lack of power of expression, his art itself is at fault. But if it was through his having meant to describe it in some incorrect way (e.g. to make the horse in movement have both right legs thrown forward) that the technical error (one in a matter of, say, medicine or some other special science), or impossibilities of whatever kind they may be, have got into his description, hi.e.ror in that case is not in the essentials of the poetic art. These, therefore, must be the premisses of the Solutions in answer to the criticisms involved in the Problems.

I. As to the criticisms relating to the poet's art itself. Any impossibilities there may be in his descriptions of things are faults. But from another point of view they are justifiable, if they serve the end of poetry itself—if (to assume what we have said of that end) they make the effect of some portion of the work more astounding. The Pursuit of Hector is an instance in point. If, however, the poetic end might have been as well or better attained without sacrifice of technical correctness in such matters, the impossibility is not to be justified, since the description should be, if it can, entirely free from error. One may ask, too, whether the error is in a matter directly or only accidentally connected with the poetic art; since it is a lesser error in an artist not to know, for instance, that the hind has no horns, than to produce an unrecognizable picture of one.

II. If the poet's description be criticized as not true to fact, one may urge perhaps that the object ought to be as described—an answer like that of Sophocles, who said that he drew men as they ought to be, and Euripides as they were. If the description, however, be neither true nor of the thing as it ought to be, the answer must be then, that it is in accordance with opinion. The tales about Gods, for instance, may be as wrong as Xenophanes thinks, neither true nor the better thing to say; but they are certainly in accordance with opinion. Of other statements in poetry one may perhaps say, not that they are better than the truth, but that the fact was so at the time; e.g. the description of the arms: 'their spears stood upright, butt-end upon the ground'; for that was the usual way of fixing them then, as it is still with the Illyrians. As for the question whether something said or done in a poem is morally right or not, in dealing with that one should consider not only the intrinsic quality of the actual word or deed, but also the person who says or does it, the person to whom he says or does it, the time, the means, and the motive of the agent—whether he does it to attain a greate.g.od, or to avoid a greater evil.)

III. Other criticisms one must meet by considering the language of the poet: (1) by the assumption of a strange word in a passage like oureas men proton, where by oureas Homer may perhaps mean not mules but sentinels. And in saying of Dolon, hos p e toi eidos men heen kakos, his meaning may perhaps be, not that Dolon's body 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 in Homer 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 the same time, e toi hot hes pedion to Troikon hathreseien, aulon suriggon *te homadon* the word hapantes 'all', is metaphorically put for 'many', since 'all' is a species of 'many '. So also his oie d' ammoros is metaphorical, the best known standing 'alone'. (3) A change, as Hippias suggested, in the mode of reading a word will solve the difficulty in didomen de oi, and to men ou kataputhetai hombro. (4) Other difficulties may be solved by another punctuation; e.g. in Empedocles, aipsa de thnet ephyonto, ta prin mathon athanata xora te prin kekreto. Or (5) by the assumption of an equivocal term, as in parocheken de pleo nux, where pleo i.e.uivocal. Or (6) by an appeal to the custom of language. Wine-and-water we call 'wine'; and it is on the same principle that Homer speaks of a knemis neoteuktou kassiteroio, a 'greave of new-wrought tin.' A worker in iron we call a 'brazier'; and it is on the same principle that Ganymede is described as the 'wine-server' of Zeus, though the Gods do not drink wine. This latter, however, may be an instance of metaphor. But whenever also a word seems to imply some contradiction, it is necessary to reflect how many ways there may be of understanding it in the passage in question; e.g. in Homer's te r' hesxeto xalkeon hegxos one should consider the possible senses of 'was stopped there'—whether by taking it in this sense or in that one will best avoid the fault of which Glaucon speaks: 'They start with some improbable presumption; and having so decreed it themselves, proceed to draw inferences, and censure the poet as though he had actually said whatever they happen to believe, if his statement conflicts with their own notion of things.' This is how Homer's silence about Icarius has been treated. Starting with, the notion of his having been a Lacedaemonian, the critics think it strange for Telemachus not to have met him when he went to Lacedaemon. Whereas the fact may have been as the Cephallenians say, that the wife of Ulysses was of a Cephallenian family, and that her father's name was Icadius, not Icarius. So that it is probably a mistake of the critics that has given rise to the Problem.

Speaking generally, one has to justify (1) the Impossible by reference to the requirements of poetry, or to the better, or to opinion. For the purposes of poetry a convincing impossibility is preferable to an unconvincing possibility; and if men such as Zeuxis depicted be impossible, the answer is that it is better they should be like that, as the artist ought to improve on his model. (2) The Improbable one has to justify either by showing it to be in accordance with opinion, or by urging that at times it is not improbable; for there is a probability of things happening also against probability. (3) The contradictions found in the poet's language one should first test as one does an opponent's confutation in a dialectical argument, so as to see whether he means the same thing, in the same relation, and in the same sense, before admitting that he has contradicted either something he has said himself or what a man of sound sense assumes as true. But there is no possible apology for improbability of Plot or depravity of character, when they are not necessary and no use is made of them, like the improbability in the appearance of Aegeus in Medea and the baseness of Menelaus in Orestes.

The objections, then, of critics start with faults of five kinds: the allegation is always that something i.e.ther (1) impossible, (2) improbable, (3) corrupting, (4) contradictory, or (5) against technical correctness. The answers to these objections must be sought under one or other of the above-mentioned heads, which are twelve in number.


The question may be raised whether the epic or the tragic is the higher form of imitation. It may be argued that, if the less vulgar is the higher, and the less vulgar is always that which addresses the better public, an art addressing any and every one is of a very vulgar order. It is a belief that their public cannot see the meaning, unless they add something themselves, that causes the perpetual movements of the performers—bad flute-players, for instance, rolling about, if quoit-throwing is to be represented, and pulling at the conductor, if Scylla is the subject of the piece. Tragedy, then, is said to be an art of this order—to be in fact just what the later actors were in the eyes of their predecessors; for Myrmiscus used to call Callippides 'the ape', because he thought he so overacted his parts; and a similar view was taken of Pindarus also. All Tragedy, however, is said to stand to the Epic as the newer to the older school of actors. The one, accordingly, is said to address a cultivated 'audience, which does not need the accompaniment of gesture; the other, an uncultivated one. If, therefore, Tragedy is a vulgar art, it must clearly be lower than the Epic.

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 that of his interpreter; for it is quite possible to overdo the gesturing even in an epic recital, as did Sosistratus, and in a singing contest, as did Mnasitheus of Opus. (2) That one should not condemn all movement, unless one means to condemn even the dance, but only that of ignoble people—which is the point of the criticism passed on Callippides and in the present day on others, that their women are not like gentlewomen. (3) That Tragedy may produce its effect even without movement or action in just the same way as Epic poetry; for from the mere reading of a play its quality may be seen. So that, if it be superior in all other respects, thi.e.ement of inferiority is not a necessary part of it.

In the second place, one must remember (1) that Tragedy has everything that the Epic has (even the epic metre being admissible), together with a not inconsiderable addition in the shape of the Music (a very real factor in the pleasure of the drama) and the Spectacle. (2) That its reality of presentation is felt in the play as read, as well as in the play as acted. (3) That the tragic imitation requires less space for the attainment of its end; which is a great advantage, since the more concentrated effect is more pleasurable than one with a large admixture of time to dilute it—consider the Oedipus of Sophocles, for instance, and the effect of expanding it into the number of lines of the Iliad. (4) That there is less unity in the imitation of the epic poets, as is proved by the fact that any one work of theirs supplies matter for several tragedies; the result being that, if they take what is really a single story, it seems curt when briefly told, and thin and waterish when on the scale of length usual with their verse. In saying that there is less unity in an epic, I mean an epic made up of a plurality of actions, in the same way as the Iliad and Odyssey have many such parts, each one of them in itself of some magnitude; yet the structure of the two Homeric poems is as perfect as can be, and the action in them is as nearly as possible one action. If, then, Tragedy is superior in these respects, and also besides these, in its poeti.e.fect (since the two forms of poetry should give us, not any or every pleasure, but the very special kind we have mentioned), it is clear that, as attaining the poeti.e.fect better than the Epic, it will be the higher form of art.

So much for Tragedy and Epic poetry—for these two arts in general and their species; the number and nature of their constituent parts; the causes of success and failure in them; the Objections of the critics, and the Solutions in answer to them.