AUTHORSHIP OF THE CRITICISM ATTRIBUTED TO ARISTOTLE—ITS SUBJECT-MATTER—ABSTRACT OF THE ‘POETICS’—CHARACTERISTICS, GENERAL—LIMITATIONS OF RANGE—ETHICAL TWIST—DRAWBACKS RESULTING—OVERBALANCE OF MERIT—THE DOCTRINE OF ἁμαρτία—THE ‘RHETORIC’—MEANING AND RANGE OF “RHETORIC”—THE CONTENTS OF THE BOOK—ATTITUDE TO “LEXIS”—VOCABULARY: “FIGURES”—A DIFFICULTY—“FRIGIDITY”—ARCHAISM—STOCK EPITHET AND PERIPHRASIS—FALSE METAPHOR—SIMILE—“PURITY”—“ELEVATION”—PROPRIETY—PROSE RHYTHM—LOOSE AND PERIODIC STYLE, ETC.—GENERAL EFFECT OF THE ‘RHETORIC’—THE “HOMERIC PROBLEMS”—VALUE OF THE TWO MAIN TREATISES—DEFECTS AND DRAWBACKS IN THE ‘POETICS’—AND IN THE ‘RHETORIC’—MERITS OF BOTH—“IMITATION”—THE END OF ART: THE οἰκεία ἡδονή—THEORY OF ACTION—AND OF ἁμαρτία—OF POETIC DICTION.

The uncomfortable conditions which have prevailed during the examination of Greek criticism during the Pre-Aristotelian |Authorship of the criticism attributed to Aristotle.| age disappear almost entirely when we come to Aristotle himself. Hitherto we have had either no texts at all, mere fragments and titles, or else documents fairly voluminous and infinitely interesting as literature, but as criticism indirect, accidental, and destitute of professional and methodical character. With the Rhetoric and the Poetics in our hands, no such complaints are any longer possible. It is true that in both cases certain other drawbacks, already glanced at, still exist, and that the Poetics, if not the Rhetoric, is obviously incomplete. But both, and especially the shorter and more fragmentary book, give us so much that it is almost unreasonable to demand more—nay, that we can very fairly, and with no rashness, divine what the “more” would have been like if we had it. In these two books the characteristics of Greek criticism, such as it was and such as probably in any case it must have been, are revealed as clearly as by a whole library.

In dealing with them we are happily, here as elsewhere, freed from a troublesome preliminary examination as to genuineness. There is no reasonable doubt on this head as far as the Rhetoric goes, and I should myself be disposed to say that there is no reasonable doubt as to the Poetics, but others have thought differently. It so happens, however, that for our special purpose it really does not matter so very much whether the book is genuine or not. For it can hardly by any possibility be much later than Aristotle, and that being so it gives us what we want—the critical views of Greek literature when the first great age of that literature was pretty well closed. It is by Aristotle, probably, by X or Z possibly, but in any case by a man of wide knowledge, clear intellect, and methodical habits.

Before we examine in detail what these views were, let us clearly understand what was the literature which this person (whom in both cases we shall call, and who in both pretty certainly was, Aristotle) had before him. The bulk of it was in verse, and though unfortunately a large proportion of that bulk is now lost, we have specimens, and (it would seem) many, if not most, of the best specimens, of all its kinds. Of a great body of epic or quasi-epic verse, only Homer and Hesiod survive; but Homer was admittedly the greatest epic, and Hesiod the greatest didactic, poet of this class. In the course of less than a century an enormous body of tragic drama had been accumulated, |Its subject-matter.| by far the greatest part of which has perished; but we possess ample specimens of the (admittedly) first Three in this kind also. Of the great old comic dramatists, Aristophanes survives alone—a mere volume, so to speak, of the library which Aristotle had before him: yet it is pretty certain that if we had it all, the quantity rather than the degree and kind of literary pleasure given by the series from the Acharnians to the Plutus would be increased. We are worst off in regard to lyric: it is here that Aristotle has the greatest advantage over his modern readers. Yet, by accident or not (it may be strongly suspected not), it is the advantage of which he avails himself least. On the other hand, some kinds—the pastoral, the very miscellaneous kind called epigram, and others—were scarcely yet full grown; and, much of them as is lost, we have more advantage of him.

In prose he had (or at least so it would seem likely) a lesser bulk of material, and what he had was subject to a curious condition, of which more hereafter. But he had nearly all the best things that we have—Plato and Xenophon, Herodotus and Thucydides, all the greatest of the orators. Here, however, his date again subjected him to disadvantages, the greatest of which—one felt in every page of the Poetics, and not insensible in the Rhetoric—was the absence, entire or all but entire, of any body of prose fiction. The existence, the date, the subjects, the very verse or prose character of the “Milesian tales,” so often talked of, are all shadows of shades, and whatever they were, Aristotle takes no count of them. It seems to be with him a matter of course that “fiction” and “poetry” are coextensive and synonymous.^[37]

Of the enormous and, to speak frankly at once, the very disastrous, influence which this limitation of his subject-matter has on him, it will be time to speak fully later. Let us first see what this famous little treatise^[38]—than which perhaps no other document in the world, not religious or political, has been the occasion of fuller discussion—does actually contain.

He first defines his scheme as dealing with poetry itself and its various kinds, with their essential parts, with the |Abstract of the Poetics.| structure of the plot, the number and nature of the parts, and the rest of poetic method. Then he lays it down that Epic, Tragedy, Comedy, Dithyrambic, as well as auletice and kitharistice generally, are mimesis—"imitation," as it is generally translated—but that they differ in the medium, the objects, and the manner of that imitation. And after glancing at music and dancing as non-literary mimetic arts, he turns to the art which imitates by language alone. Here he meets a difficulty: there is, he thinks, no common name which will suit the mimes of Sophron and Xenarchus, the “Socratic dialogues,” and iambic or elegiac mimesis. He objects strongly to the idea that metre makes the poet, and produces instances, among which the most striking is his refusal of the name poet to Empedocles. Having disposed of the medium—rhythm, metre, &c.—he turns to the objects. Here he has no doubt: the objects of mimesis are men in action, and we must represent them as “better than life” (heroic or idealising representation), as they are (realistic), or worse (caricature or satire). The manner does not seem to suggest to him much greater diversity than that of epic (or direct narrative), and dramatic, as to the latter of which he has a slight historical excursus.

Then he philosophises. Poetry, he says, has two causes: one the instinct of imitation, with the pleasure attached to it; the other, the instinct for harmony. And then he again becomes historical, and reviews briefly Homer, Æschylus, Sophocles, and the progress of poetry under them.

Comedy he dismisses very briefly. He thinks that it ἔλαθε διὰ τὸ μὴ σπουδάζεσθαι—little attention was paid to it, as not being taken seriously. Epic and tragedy must be treated first—tragedy first of all. And then he plunges straight into the famous definition of tragedy, discussion of which had best be reserved. The definition itself is this: “An imitation of an action, serious, complete, and possessing magnitude, in language sweetened with each kind of sweetening in the several parts, conveyed by action and not recital, possessing pity and terror, accomplishing the purgation of such^[39] emotions.” Tragedy will require scenic arrangements, musical accompaniments, and “words,” as modern actors say; his own term, “lexis,” is not so very different. But it will also require character, “thought,” and plot or story. The most important of all is the last, which he also describes by another name, the “setting together of incidents,” the Action—to which he thinks character quite subsidiary, and indeed facultative. There cannot, he says categorically, be tragedy without action; there may be without character. “The most powerful elements of emotional interest,” as Professor Butcher translates οἷς ψυχαγωγεῖ, “the things with which tragedy leads souls,” are revolutions and discoveries, and these are parts of action. Novices can do good things in diction and in character, not in plot. Still Character is second. Thought is third, Diction apparently a bad fourth. Song is only a chief embellishment or “sweetening,” and Scenery is the last of all, because, though influencing the soul, it is inartistic and outside poetry. So he turns once more as to the principal or chief thing, to the plot or action. This is to be a complete whole, and of a certain magnitude, with a beginning, middle, and end. A very small animal organism^[40] cannot be beautiful, as neither can one “ten thousand stadia long.” Then he comes to the great question of Unity—or, since that word is much blurred by usage, let us say “what makes the story one.” It is not enough to have a single hero; life, even a part of a life, is too complicated for that. We must have just so much and just so little that the action shall present neither gaps nor redundancies. Nor need the poet by any means stick to historical or prescribed fact—the probable, not the actual, is his game. He may invent wholly (subject to this law of probability) if he likes. Plots with episodes are bad.

We have, however, to go further. Not only must the action of tragedy be complete and probable, but it must deal with terrible and pitiful things: if these surprise us, so much the better. After distinguishing between simple plots (without Revolution and Discovery) or complex (with them), and describing these two elements at more length, he attacks, in a rather suspected passage, the Parts—Prologue, Episode, Exodus,^[41] the choric part, &c.—and then, preferring the complex scheme, shows how it is to be managed. The hero must not blamelessly pass from prosperity to adversity, nor blamefully in the opposite direction. He must be a person of considerable position, who by some error or weakness (ἁμαρτία) comes to misfortune. Also the special kind of pity and terror which is to be employed to make him interesting, the oikeia hedone of tragedy, is most important, not a few examples being taken in illustration from the great tragedians.

Then we pass to Character. It must be good—even a woman is good sometimes—it must be appropriate, true to life, and consistent. Probability is here as important as in Action; the Deus ex machina is to be used with extreme caution. After turning to the details of Discovery, and dealing with Gesture, Scene, &c., he goes to the two main stages of Tragedy, desis and lusis, Twisting and Unravelling, and to its four kinds (an extension of his former classification)—Simple, Complex, Pathetic, and Ethical. And the tragic poet is especially warned against Tragedy with an Epic structure—that is to say, a variety of plots. The Chorus must bear part in the action, and not give mere interludes.

“Thought” is somewhat briefly referred to Rhetoric (vide infra), and then we come to Diction. This is treated rather oddly, though the oddity will not seem so odd to those who have carefully studied the contents, still more the texts, of the foregoing chapter. Much of the handling is purely grammatical. The “Figures,” especially metaphor, make some appearance: and of style proper we hear little more than that it is to be clear without being mean, though we have some illuminative examples of this difference.

Then Aristotle passes briefly to Epic, his prescription for which is an application of that already given—the single action, with its beginning, middle, and end. The organism, with its oikeia hedone, the parts, the kinds are the same, with the exception of song and scenery. The only differences are scale (Epic being much larger) and metre, with a fuller allowance for the improbable, the irrational. Some rather desultory remarks on difficulties of criticism or interpretation follow, and the piece ends abruptly with a consideration of the purely academic question whether Epic or Tragedy ranks higher. Some had given the primacy to Epic: Aristotle votes for Tragedy, and gives his reasons.

This summary has been cut down purposely to the lowest point consistent with sufficiency and clearness; but I trust it is neither insufficient nor obscure. We may now see what can be observed in it.

We observe, in the first place, not merely a far fuller dose of criticism than in anything studied hitherto, but also a great |Characteristics, general.| advance of critical theory. Not only has the writer got beyond the obscure, fragmentary, often irrelevant, utterances of the early philosophers: but he is neither conducting a particular polemic, as was Aristophanes, nor speaking to the previous question, like Plato. An a posteriori proof of the depth and solidity of the inquiry may be found in the fact that it is still, after more than two thousand years, hardly in the least obsolete. But we are not driven to this: its intrinsic merit is quite sufficient.

At the same time, there are certain defects and drawbacks in it which are of almost as much importance as its merits, and which perhaps require prior treatment. That it is incomplete admits of no doubt; that part of it shows signs of corruption, that there are possible garblings and spurious insertions, does not admit of very much. But the view throughout is so firm and consistent; the incidental remarks tally so well with what we should expect; and, above all, the exclusions or belittlings are so significant, that if the treatise were very much more complete, it would probably not tell us very much more than we know or can reasonably infer already.

In the first place, we can see, partly as a merit and partly as a drawback, that Aristotle has not merely confined himself with philosophical exactness to the Greek literature actually before him, but has committed the not unnatural, though unfortunate, mistake of taking that literature as if it were final and exhaustive. He generalises from his materials, especially from Homer and the three Tragedians, as if they provided not merely admirable examples of poetic art, but a Catholic body of literary practice to go outside of which were sin. It is impossible not to feel, at every moment, that had he had the Divina Commedia and Shakespeare side by side with the Iliad and Æschylus, his views as to both Epic and Tragedy might have been modified in the most important manner. And I at least find it still |Limitations of range.| more impossible not to be certain that if there had been a Greek Scott or a Greek Thackeray, a Greek Dumas or a Greek Balzac before him, his views as to the constitutive part of poetry being not subjective form but “imitative” substance would have undergone such a modification that they might even have contradicted these now expressed. If tragedy, partly from its religious connection, partly from its overwhelming vogue, but most of all from the flood of genius which had been poured into the form for two or three generations past, had not occupied the position which it did occupy in fact, it would probably not have held anything like its present place in the Poetics. And so in other ways. It may be consciously, it may be unconsciously, Aristotle took the Greek, and especially the Attic, literature, which constituted his library, and treated this as if it were all literature. What he has executed is in reality an induction from certain notable but by no means all-embracing phenomena; it has too much of the appearance, and has too often been taken as having more than the appearance, of being an authoritative and inclusive description of what universally is, and universally ought to be.

We have also to take into account the Greek fancy for generalising and philosophising, especially with a strong ethical |Ethical twist.| preoccupation. Aristotle does not show this in the fantastic directions of the earlier allegorising critics, but he is doubly and trebly ethical. He has none of the Platonic doubts about Imitation as being a bad thing in itself, but he is quite as rigid in his prescription of good subjects. Although we have no full treatment of Comedy, his distaste—almost his contempt—for it is clear; and debatable as the famous “pity and terror” clause of the definition of tragedy may be, its ethical drift is unmistakable.

Thus his criticism, consciously or unconsciously, is warped and twisted by two unnecessary controlments. On the one |Drawbacks resulting.| hand, he looks too much at the actual occupants of his bookcase, without considering whether there may not be another bookcase filled with other things, as good but different. On the other, he is too prone, not merely to generalise from his facts as if they were the only possible facts, but to “overstep the genus” a little in his generalisation, and to merge Poetics in Ethics. That others went further than he did, that they said later that a hero must not only be good but white, and superadded to his Unity of Action a Unity of Time and a Unity of Place, which his documents do not admit, and which his doctrines by no means justify, are matters for which, no doubt, he is not to be blamed. But of the things for which he is legitimately responsible, some are not quite praiseworthy.

In the first place, “Imitation” is an awkward word, though no doubt it is more awkward in the English than in the Greek, and “Representation” or “Fiction” will get us out of part of the difficulty. Not only does this term for the secondary creation proper to art belittle it too much, but it suggests awkward and mischievous limitations: it ties the poet’s hands and circumscribes his aims.^[42] Indirectly it is perhaps responsible for Aristotle’s worst critical slip—his depreciation of Character in comparison with Action. This very depreciation is, however, a serious shortcoming; and so is the failure to recognise, despite some not indistinct examples of it in the matter before him from the Odyssey downwards, what has been called “Romantic Unity,” that is to say, the Unity given by Character itself, though the action may be linear and progressive rather than by way of desis and lusis. The attempt to extend (save in respect of scale only) the limitations of Tragedy to Epic is another fault; and so perhaps is the great complexity and the at least not inconsiderable obscurity of the definition of Tragedy itself. In such a treatise as this it is possible merely to allude to the famous clause, “through pity and terror effecting the katharsis of such emotions.” Volumes have been written on these few words,^[43] the chief crux being, of course, the word katharsis. It cannot be said that any of the numerous solutions is by itself and to demonstration correct, but it is clear that the addition is out of keeping with the rest of the definition. Hitherto Aristotle, whether we agree with him or not, has been purely literary, but he now shifts to ethics. You might almost as well define fire in terms strictly appropriate to physics, and then add, “effecting the cooking of sirloins in a manner suitable to such objects.”

Yet the advantages of this criticism far exceed its drawbacks. In the first place it is, not merely so far as we positively know, |Overbalance of merit.| but by all legitimate inference, the earliest formal treatise on the art in European literature. In the second place, even if it sticks rather too close to its individual subject, that individual subject was, as it happens, so marvellously rich and perfect that no such great harm is done. A man will always be handicapped by attempting to base criticism on a single literature, yet he who knows Greek only will be in far better case than he who only knows any one other, except in so far as the knowledge of any later literature inevitably conveys an indirect dose of knowledge of Greek.

Then, too, Aristotle’s use of his material is quite astonishingly judicious. In almost every single instance we might |The doctrine of ἁμαρτία.| expect his limitations to do him more harm than they have done. He might, for instance, with far more excuse than Wordsworth, have fallen into Wordsworth’s error of considering metre not merely as not essential to poetry, but as only accidentally connected with it. And it is also extremely remarkable how little, on the whole, his ethical preoccupation carries him away. He exhibits it; but it does not blind him (as it had blinded even Plato) to the fact that the special end of Art is pleasure, that the perfection of literature is not an end in itself but a means to an end. Even more surprising is the acuteness, the sufficiency, and the far-reaching character of his doctrine of the Tragic ἁμαρτία. For there can be no question that he has here hit on the real differentia of tragedy—a differentia existing as well in the tragedy of Character, which he rather pooh-poohs, and in the Romantic tragedy which he did not know, and on his actual principles was bound to disapprove if he had known it, as in the Classical. Shakespeare joins hands with Æschylus (and both stand thus more sharply contrasted with inferior tragedians than in any other point) in making their chief tragic engine “the pity of it,” the sense that there is infinite excuse, but no positive justification, for the acts which bring their heroes and heroines to misfortune. Wherever the tragedian, of whatever style and time, has hit this ἁμαρτία, this human and not disgusting “fault,” he has triumphed; wherever he has missed it, he has failed, in proportion to the breadth of his miss.

With respect to the minor and verbal points of the Poetics there is less to say, because there is very much less of them: |The Rhetoric.| and what there is to say had better be said when we have considered the contents of the other great critical book, the Rhetoric, which may be taken as holding, if not intentionally yet actually, something of the same position towards Prose as that which the Poetics holds towards verse.

Before giving an analysis of this book,^[44] to match that given above of the Poetics, a few words may properly be said to justify |Meaning and range of “Rhetoric.”| what may seen to be the rather arbitrary proceeding of, on the one hand, attaching to Rhetoric a sense avowedly somewhat different from Aristotle’s, and on the other dropping consideration of the major part of what he has actually written in it.

It is a mistake to force too much the bare meanings of words; but I suppose one may, without much danger of controversy, take the bare meaning of Rhetoric to be “speechcraft.” Now, it is not difficult to prove that, in Aristotle’s time, speechcraft practically included the whole of prose literature, if not the whole of literature. Poems were recited; histories were read out; the entire course of scientific and philosophic education and study went on by lecture or by dialogue. Nay, it is perhaps not fanciful to point out that the very words for reading, ἀναγιγνώσκω and ἐπιλέγομαι seem to represent it as at best a secondary and parasitic process, a “going over again” of something previously said and heard.^[45]

Yet though this is an important point, and has been rather too commonly overlooked, it is no doubt inferior in gravity to the universally recognised fact that the importance of speechcraft proper, of oratory, was in Greece such as it is now only possible dimly to realise. Every public and private right of the citizen depended upon his power to speak or the power of somebody else to speak for him; a tongue-tied person not only had no chance of rising in the State, but was liable to be insulted, and plundered, and outraged in every way. To some it has seemed that the great and almost fatal drawback to that Athenian life, which in not a few ways was life in a sort of Earthly Paradise, was the incessant necessity of either talking or being talked to. It was therefore not in the least wonderful that the first efforts—those of the Sicilian sophists (or others)—to reduce to something like theory the art of composition, of arranging words effectively, should be directed to spoken words, and to spoken words more particularly under the all-important conditions of the public meeting and the law court—by no means neglecting the art of persuasion, as practicable in the Porch, or the Garden, or the private supper-room. That prose literature—that all literature—has for its object to give pleasure dawned later upon men. Aristotle and persons much earlier than Aristotle—Corax and Tisias themselves—would probably have acknowledged that prose, like poetry, ought to please, but only as a further means to a further end, persuasion. Its object was to make men do something—pass or negative such a law, bring in such a verdict, appoint such an officer, or (in the minor cases) believe or disbelieve such a tenet, adopt or shun such a course of conduct. Even in poetry, as we have seen, the ethical preoccupation partly obscured the clear æsthetic doctrine—you were to be purged as well as pleased, and pleased in order that you might be purged. But in prose the pleasure became still more subsidiary, ancillary, facultative. You were first of all to be “persuaded.”

Now, if this be taken as granted, and if, further, we keep in mind Aristotle’s habit of sticking to the facts before him, we |The contents of the book.| shall not be in the least surprised to find that the Rhetoric contains a great deal of matter which has either the faintest connection with literary criticism, or else no connection with it at all. It is true that of the three subjects which the Rhetoric treats, pistis (means of persuasion), lexis (style), and taxis (arrangement), the second belongs wholly and the third very mainly to our subject, while it would be by no means impossible for an ingenious arguer to make good the position that pistis, with no extraordinary violence of transition, may be laid at least under contribution for that attractive quality which all literature as a pleasure-giving art must have. But in actual handling Pistis has two out of the three books, and is treated, as a rule, from a point of view which leaves matters purely literary out of consideration altogether—"The Characteristics of Audiences," “The Colours of Good and Evil,” “The Passions as likely to exist in an Audience,” “The Material of Enthymeme” (the special rhetorical syllogism), and so forth.

Only in the third book (which, by the way, is shorter than either of the other two) do we get beyond these counsels to the advocate and the public speaker, into the Higher Rhetoric which concerns all prose literature and even some poetry. And even then we meet with a sort of douche of cold water which may not a little dash those who have not given careful heed to the circumstances of the case.

Inquiry into the sources and means of persuasion (our author admits graciously, but with a touch of superiority which, |Attitude to lexis.| as we shall see, accentuates itself later) does not quite exhaust Rhetoric. It must also discuss style and arrangement. But style is a modern thing, and, rightly considered, something ad captandum.^[46] Indeed Aristotle never seems to keep it quite clear from mere elocution or delivery—from the art of the actor as contradistinguished from that of the writer. He remarks that he has dealt with style fully in his Poetics; and as he has certainly not done so in the Poetics which we have, this is an argument that they are incomplete, though by no means that they are spurious. But it is almost impossible to mistake the touch of patronage, not to say of scorn, with which he deals with it here, and we need not doubt that, if we had the other handling to which he refers, something of the same sort would appear there. The fact is, that the Greeks of this period were what we may call High-fliers; anything that had the appearance of being “mechanical,” anything that seemed to subject the things of the spirit to something not wholly of the spirit, they regarded with suspicion and impatience, which rather suggests the objection of some theologians to good works. Words, like colours, materials of sculpture and architecture, and the like, were “filthy rags”; and if Aristotle’s common-sense carried him a little less far in this direction than his master Plato’s philosophical enthusiasm, it certainly carried him some way.

This same common-sense, however, seldom deserted him, and it makes sometimes wholly for good, sometimes a little less |Vocabulary—"Figures".| so, throughout the treatise. At the very outset he commits himself to that definition of style as being first of all clear—as giving the meaning of the writer—which has so often captivated noble wits down to Coleridge’s time, and even since, but which yet is clearly wrong, for “two and two make four” is the a per se of clearness, and there is uncommonly little style in it notwithstanding. That he himself saw this objection cannot be doubted, for he hastens to add^[47] that it must be not only clear, but neither too low nor too far above the subject, thus producing a useful and perfectly just distinction between the styles of poetry and prose. And then he gives us, as he had done in the Poetics, one of those distinctions of his which are so valuable—the distinction of vocabulary into what is κύριον or current (which conduces to clearness), and what is ξένον or unfamiliar (which conduces to elevation). Let us note that this, like the ἁμαρτία theory in the Poetics, is one of Aristotle’s great critical achievements. But the note of greatness may perhaps be discovered less in the attention which from this point he begins to pay to Metaphor. Not of course that metaphor is not a very important thing; but that the example of ticking it off in this fashion with a name spread rapidly in Rhetoric, and became a mere nuisance. Even Quintilian, who spoke words of wit and sense about the Greek mania for baptising new Figures, submitted to them to some extent: and any one who wishes to appreciate the need of Butler’s jest to the effect that

has no farther to look than to the portentous list at the end of Puttenham’s Art of Poetry.

Yet his cautions as to metaphors themselves, which he regards as the chief means of embellishment in prose, are perfectly just and sound. They must, he says, be selected with careful reference to the particular effect intended to be produced, be euphonious, not far-fetched, and drawn from beautiful objects.

Here, perhaps as well as in reference to any single passage in the Poetics, we have an opportunity of considering for the |A difficulty.| first time a difficulty, not unexpected, not uninteresting, which meets us, and which will recur frequently, in ancient (and sometimes in the most modern) criticism. It is the difficulty which so did please Locke and his followers in the attack on the doctrine of Innate Ideas,—in other words, the difficulty of an apparently hopeless difference of standard on points of taste—the difference between Greek and modern love, between English and Hottentot beauty. One should, says the philosopher, say ῥοδοδάκτυλος rather than φοινικοδάκτυλος, while ἐρυθροδάκτυλος is the worst of all. The commentators have tried to get out of the difficulty by suggesting that the last suggests the redness of frost-bitten or domestically disfigured fingers. φοινικοδάκτυλος would in the same way, I suppose, be considered as objectionable because the colour is overcharged in the epithet, and might even suggest “red-handed” in the sense of “bloodstained.” Yet one may doubt whether Aristotle’s objection is based on anything but the fact that Homer uses the one epithet, not the others. The verb ἐρυθριάω, at any rate, is invariably used for blushing, not an unattractive or unbeautiful proceeding by any means. And we shall find very much stronger instances of this difficulty later.

The explanation is partly supplied by the very next section, which deals with ψυχρότης and is one of the most valuable |“Frigidity.”| keys existing to the whole tone of Greek, indeed of classical, criticism. It is rather unlucky that “frigidity,” our only equivalent, is not quite clear to English ears. In fact, “fustian” comes nearest to what is meant, though it is not completely adequate and coextensive. The idea is not difficult to follow—it is that of something which is intended to excite and inflame the auditor or reader, while in fact it leaves him cold, if it does not actually lower his spiritual temperature. Aristotle gives four cases, or (which is nearly the same thing) four kinds of it—words excessively compounded, foreign terms, too emphatic or minute epithets, and improper metaphors. To these, as generalities, few would object, but the instances are sometimes decidedly puzzling. Lycophron (the sophist, not the poet) is blamed for calling the heavens πολυπρόσωπον (“many-visaged”), the earth μεγαλοκόρυφον (“mightily mountain-topped”), and the shore στενοπόρον (“leaving a narrow passage between cliff and sea”). Now, perhaps these terms are too poetical, yet we should hardly call them frigid, for they are not untrue to nature, and they not only show thought and imagination in the writer, but excite both in the reader. Still, they are all slightly excessive; they pass measure, as do other things blamed in Alcidamas and Gorgias still more.

The second objection is of still greater interest, because it has practically supplied a shibboleth in the Classic-Romantic debate |Archaism.| up to the present moment. It is the objection to archaic, foreign, and otherwise inusitate words, which Aristotle seems to apply even to Homeric terms, not as poetic but as obsolete, just as other good persons in times nearer our own have applied the same to Chaucerisms and the like. The sounder doctrine, of course, is that nullum tempus occurrit regi in this transferred sense also—that what the old kings of literature have stamped remains current for ever, and what the new kings of literature stamp takes currency at once.

Almost as interesting is the third punishment-cell, in which epithets too long, too many, or out of place are bestowed. The |Stock epithet and periphrasis.| two habits which seem to be mainly aimed at here (Alcidamas is still the chief awful example) are the use in prose of the poetical perpetual epithet (“white milk” is the example chosen) and the undue tendency to periphrasis, which, curiously enough, reminds one of the besetting sin of the extreme “Classical” school of the last century.

Most puzzling of all are the examples pilloried for impropriety in the fourth class, the unfortunate Alcidamas being rebuked for calling philosophy “the intrenchment of law,” and |False metaphor.| the Odyssey a “mirror of human life.” The most, thoroughgoing Aristotelians have given up this last criticism with an acknowledgment that ancient and modern tastes differ; while Mr Cope even suggests that Aristotle “winked,” not nodded, when he wrote the whole passage. I do not so easily figure to myself a winking Stagirite.

In the chapter on Simile which follows there is much that is sensible, but nothing that is surprising—the relation of simile |Simile.| and metaphor being the main point. One’s expectations are more raised in coming to the great subject of “purity” of style—"Hellenising," “writing Greek.” This phrase, in our author, is directed against something corresponding rather to the French “fautes de Français” than to our “not English,” having regard to the syntax, the sentence-building, |“Purity.”| rather than to the actual diction. But it differs from both in having, like so much of his criticism, more to do with matter than form. In fact, it has been well observed that “Perspicuity” rather than “Purity” is really the subject of the chapter. It is, however, of great importance, and the next, on Elevation, or Grandeur, or Dignity, is |“Elevation.”| of greater still. Some slight difficulty may occur at starting with the word thus variously rendered in English, ὄγκος. In its non-rhetorical use, the word (which strictly means “bulk,” with the added notion of weight) inclines rather to an unfavourable signification, often signifying “pretentiousness,” “pomposity”: it is sometimes used later in Rhetoric itself with such a meaning; and I think those who compare the earlier passage on Frigidity will be inclined to suspect that Aristotle himself was not using it entirely honoris causa. He gives, however, some hints for its attainment, and a bundle of instances, where our ignorance of the context makes the illustrative power somewhat small.

Next we come to that quality of τὸ πρέπον, “the becoming,” “propriety,” which is commonly and not wrongly taken to be |Propriety.| the special note of “classical” writing. And we have rules for its attainment, some ethical rather than æsthetic, some æsthetic enough but curiously arbitrary, as that unusual words are not appropriate except to a person in a state of excitement. At the close there is an interesting glance at the irony of Gorgias and of Socrates.

The next division is one of the very apices of the whole. It deals with that subject of the rhythm of prose which, though |Prose rhythm.| (as we see from Quintilian as well as from Aristotle) never neglected by the ancients, is one of the most difficult parts of their critical Rhetoric for us to understand, and (perhaps for that reason) has been, till the last hundred years or so, strangely neglected in the criticism of modern languages.

We see from its very opening words that the great distinctions between verse and prose literature on the one hand, and between literary and non-literary composition on the other, had been already hit upon. Prose style, says he, must be neither emmetron nor arrhythmon—that is to say, it must not have metre nor lack rhythm. But he does not very accurately define the difference between these things; and it cannot be said that any of his commentators and successors have supplied this defect, though it is easy enough to do so.^[48] He, however, allows feet if not metre in prose, and proceeds to inquire what feet will do, making observations on the subject which are in the three degrees of obscurity to all who are not fond of guessing. Dactylic, iambic, and trochaic rhythms are dismissed for various reasons, rather bad than good—it not having apparently struck the critic that all these arrange themselves too easily, certainly, and definitely into metre. He pitches finally on the pæan, a foot which, though admissible in those Greek choric measures which are a sort of compromise between prose and poetry, at once reveals its suitableness for prose in modern languages by the fact that it is unsuitable for modern verse. The pæan or pæon is a tetrasyllabic foot, consisting of three short syllables and a long one, of which in strictness there may be four varieties, the long syllable being admissible in any of the four places. But Aristotle only admits two, with the long syllable in the first and fourth place respectively. And here, most tantalisingly, he breaks off.

The distinction between loose and periodic style,^[49] which the modern composition-books have run so tiresomely to death, |Loose and periodic style, &c.| and which is really a very unimportant technical detail, follows; and then we return to those Delilahs of the ancient rhetorician, Figures—Metaphor once more, Antithesis, Personification, Hyperbole, &c. Yet even this is more to our purpose than the demonstration that follows, showing that each kind of Rhetoric, judicial, deliberative, and declamatory, should have its particular style. And with this the handling of lexis proper closes, the rather brief remainder of the book being devoted partly to taxis (ordonnance, as Dryden would say), but with special reference to the needs of the pleader, and partly to a fresh handling of the old questions of enthymeme, the dispositions of the audience, and the like.

It will be seen from this that the Rhetoric, like the Poetics, is invaluable to the historian of literary criticism, but that, in |General effect of the Rhetoric.| this case as in that, literary criticism was only partly the object in the writer’s eye, while even so far as he had it before him, his views were very largely limited, and were even in some cases distorted, coloured, and positively spoilt by certain accidents of place, time, and circumstance. As our poetical criticism was injuriously affected by the non-existence of the novelist, so our prose criticism is injuriously affected by the omnipresence of the orator. As our Poetics were adulterated with ethic and other things, so our Rhetoric is warped by poetical, jurisprudential, and other preoccupations. In the first, poetry itself is not indeed itself a secondary consideration, but divers secondary considerations ride it, like a company of old men of the sea. In the second, prose as prose is merely and avowedly a secondary consideration: it is always in the main, and sometimes wholly, a mere necessary instrument of divers practical purposes.

To supplement these two general treatises, we could wish for more particular applications, but we have not got them. We have indeed some vestiges of work of the kind which are not altogether encouraging. M. Egger^[50] has endeavoured to extract some references to literary criticism from the general Problems; but these deal at best with the remotest fringes of the topic—why melancholy is so often apparent in persons of genius, and the like,—questions indeed of the very first interest, but not of the kind which we are here pursuing. In the extant fragments, however, which belong or may have belonged to the lost Homeric Problems^[51] (or aporems or zetems) |The Homeric Problems.| we have metal more attractive. It may be said that the scholiasts, through whom we have most of these excerpts, were likely to select them according to the principles which, as we shall see,^[52] governed themselves; but they do not all come through scholiasts, and yet the complexion of all is more or less uniform. It is that “ethical-dramatic” complexion, as we may call it, which we have noticed and shall notice as being the Greek critical “colour”—sometimes to the utter exclusion, and almost always to the effacement, of actual criticism. “Why did Agamemnon try experiments on the Greeks? Why did Odysseus take his coat off? Why is Menelaus represented as having no female companion? Why [a curial instance of that commentatorial lues which infects the greatest commentators as the least, the most ancient as the most modern] is Lampetie represented as carrying to the Sun the news of the slaughter of his oxen, when the Sun sees everything? Why did the poet make Paris a wretch who was not only beaten in duel, who not only ran away, but who was specially excited by love immediately afterwards?”

These are mainly moral questions; but the great philosopher appears to have carried his solicitude so far as to meddle with military matters. “Why [somebody had asked], in Il. iv. 67-69, are the cavalry represented as marshalled in front, the cowards in the middle, and the best infantry behind?” If Aristotle had heard of the “cavalry screen” he would no doubt have used this lusis: as it is, it appears, he suggested that prota means not “in front” but “on the wings.” And there is all the quality which endeared Aristotle to the idler side (which was not the only side by any means) of Scholasticism, in his condescension to the aporia—"If the gods drank nothing but nectar, why is Calypso spoken of as ‘mixing’ for Hermes? For any ‘mixture,’ even with water, is something different from nectar; and, therefore, as the gods do not drink their nectar neat, they do not drink that only." Quoth the great master (in reply, or at least “Schol. T.” says so), “The word does not only mean to ‘mix,’ but also simply to ‘pour,’ and this is what Calypso did.” But why should Calypso herself and Circe and Ino, alone of goddesses, have the epithet αὐδήεσσα? Even he could not answer that, and was driven ignominiously to suggest a change of reading.

It is not, I hope, necessary to say that I have no intention of raising an inept laugh at the Great One. As has been already said, the attitude of the Greeks to Homer was the attitude of a seventeenth-century Puritan to the English Scriptures. Every word, almost every letter, had its reason and its meaning—often many more than one—which had to be reverently sought out. The analogy, however, itself establishes and makes clear my point, which is to show that an attitude of this kind practically excludes pure literary criticism on the one hand, and is exceedingly unlikely on the other to be taken up by any one who is strongly bent towards such criticism. We know how Milton, who must have had an exquisite critical gusto originally, and who never wholly lost it, was by the cultivation of such an attitude so stunted and checked in his taste that he could throw the reading of Shakespeare in his dead king’s face,^[53] dismiss the delightful work (hardly inferior to the best of his own) of the Cavalier poets as “vulgar amorism” and “trencher fury,” and even when he was not thinking of matter, sink all critical perspective in his blind craze against rhyme itself. The Homer-worship of the Greeks on the one hand, and their philosophical preoccupations on the other, had almost unavoidably a similar effect, though not so bad a one.

Yet the value of the two main documents is so inestimable, that if the incompleteness and the shortcomings of the Poetics, the unavoidable irrelevance of much of the Rhetoric, |Value of the two main treatises.| were far greater than they are, our gratitude for both would still be hard to exaggerate. We have here not merely the first constituting documents, the earliest charters at once and discussions of European criticism, but we have them from the hand of a master whose very weaknesses make him, as compared with some other masters, specially fit for the office of critic. For the magnificent but almost always a priori and unpractical metaphysics of Plato, for the shrewd but personal and rather unfair polemic of Aristophanes, we have a patient examination of a subject in itself so rich and varied, that one regrets having to point out that its riches and its variety are not quite exhaustive. Nowhere, perhaps, does Aristotle sketch the actual Wesen of the man of letters with the dæmonic completeness of the author of the extraordinary passage attributed to Simylus and quoted formerly; but that might be, and probably is, a mere flash. His own conclusions, only sometimes inadequate, very seldom positively erroneous, exhibit the true modes of criticism as perhaps they have never been exhibited since—with an equal combination of patience and of power. It is impossible for Aristotle to do harm, unless his principles are not merely taken too literally, but augmented and falsified, as was done by the “classical” criticism of the seventeenth and eighteenth centuries. It is impossible for any one who undertakes the office of a critic to omit the study of him without very great harm. Let us first review briefly what seem to be the shortcomings, accidental or essential, of his performance, and then set down what its better parts establish for us as the state of Literary Criticism at the close of the first and greatest age of Greek literature, at the close of the first age of the literature of Europe as a whole.

Partly by mere induction from actual Greek practice, and partly no doubt also as a genuine result of Greek |Defects and drawbacks in the Poetics.| taste and literary philosophy, we find the importance and the character of certain kinds of literature treated with some extravagance. The importance of Tragedy (as we are enabled to see clearly by the invaluable though rather unfair aid of the historic estimate) is altogether exaggerated. It never, as a matter of fact, has held anything like the position here assigned to it, save twice in two thousand years and more, on each occasion for a generation or two only. And there is no reason, in the order and logic of thought, why it should hold such a position. It is again clearly evident (though we owe the clearness again not to our own wits, but to time and chance) that part of this importance is attained by an illegitimate sacrifice, or an accidental ignoring, of the just claims of other branches of literature—by making lyric a mere playhouse handmaid, by converting the stage into a pulpit, and by blocking out, not merely the existence, but the very possibility, of the prose novel. We can see further that the glorious achievements of the three great tragedians whom we in part possess, and of others, probably not much inferior, whom we have almost wholly lost, seduced their critic into taking what he found in them too hastily for what ought to be found in all—induced him (aided no doubt by the Greek taste generally) to exalt Plot, to depress Character, to put quite undue stress on artificial Unity. Lastly (to keep to the Poetics), we perceive a most unfortunate, though by no means inexplicable, tendency to give insufficient weight to Metre, and a decided inclination, on the one hand not to give quite enough importance to Diction, and on the other to lay down arbitrary rules about it.

Something of the same general tendency manifests itself in the Rhetoric, reinforced by the necessary results of the Persuasion-theory, |And in the Rhetoric.| and the inordinate importance given to Oratory. With every possible allowance for the undoubtedly true plea that Aristotle had no intention of writing a treatise on Prose Composition generally, but only one on such Prose Composition as suited the purposes of the Orator, we can see that if he had written Prosaics, to match the Poetics, the same limitations would have appeared. He cannot free himself from the notion that there is, after all, something derogatory in paying great attention to style: and it is clear that he does not wish to consider a piece of prose as a work of art destined, first of all, if not finally, to fulfil its own laws on the one hand, and to give pleasure on the other. The salutary but easily exaggerated difference between prose and poetic style is actually exaggerated here. Above all, the germ of mischief, if not exactly the mischief itself, is clearly discernible in his account of the Figures of Speech. It was the drawback, not merely (as is sometimes said unjustly) of the Platonic philosophy only, but of all Greek philosophy, to “multiply entities”—to take for granted that because names are given to things, things must necessarily exist behind names. And so, instead of regarding these Figures as merely rather loose, sometimes not inconvenient, but in reality often superfluous, tickets for certain literary devices and characteristics, there grew up, if not in Aristotle himself, at any rate in his followers, a tendency to regard the Figures (which were soon enormously multiplied) as drugs or simples, existing independently, acting automatically, and to be “thrown in,” as the physician exhibits his pharmacopœia, to produce this or that effect.

But enough of this. It is the pleasanter, and, though not in kind, yet in degree, the more important, business of the historian, to call attention to the enormous positive |Merits of both.| advance which we make with these two books. It is almost the advance from chaos to cosmos; and we shall find nothing in all the rest of the history quite to match it, though the resurrection of Criticism with the revival of learning, and the reformation of it at the Romantic era, come nearest.

In the first place, we find the great kinds of literature, if not finally and exhaustively, yet in nearly all their most important points, discerned, marked off, and as far as possible furnished with definitions. The most important of all demarcations, that between poetry and prose, is rather taken for granted than definitely argued out; but we see that, with whatever hesitations and reservations, it is taken for granted. So, too, with the kinds of poetry itself. If prose is inadequately treated, both in general and in its departments, we have been able to assign something like a reason for that; and a good deal is actually done in this direction. In other words, the field, the “claim,” of literary criticism is pretty fairly pegged out.

In the second place, the only sound plan—that of taking actually accomplished works of art and endeavouring to ascertain how it is that they give the artistic pleasure—is, with whatever falterings, pretty steadily pursued. The critic, as Simylus, Aristotle’s own contemporary, has it, consistently endeavours to “grasp” his subject; and he does grasp it over and over again.

Let us review our positive gains from this grasp.

That the “Imitation” doctrine of the Poetics is in some respects disputable need not be denied; and that it lends itself rather easily to serious misconstruction is certain. |“Imitation.”| But let us remember also that it is an attempt—probably the first attempt, and one which has not been much bettered in all the improvements upon it—to adjust those proportions of nature and art which actually do exist in poetry. For by Imitation, whatever Aristotle did mean exactly, he most certainly did not mean mere copying, mere tracing or plaster-of-Paris moulding from nature. It is not quite impossible that his at first sight puzzling objection to Alcidamas' use of the “mirror” as a description of the Odyssey had something to do with this.^[54] A mirror, he would or might have said, reproduces passively, slavishly, and without selection or alteration: the artist selects, adapts, adjusts, and if necessary alters. Now this is the true doctrine, and all deviations from, it, whether in the shape of realism, impressionism,^[55] and the like, in the one direction, or of adherence to generalised convention on the other, have always led to mischief soon or late. The artist must be the mime, not the mirror: the reasonable, discreet, free-willed agent, not the passive medium. The single dictum that poetry does not necessarily deal with the actual but with the possible—that it is therefore “more philosophic,” higher, more universal, than history, though it requires both extension and limitation, will put us more in the true critical position than any dictum that we find earlier, or (it may be very frankly added) than most that we shall find later.^[56]

So, too, the all-important law that the end of art is pleasure appears solidly laid down.^[57] True, it is not laid down so explicitly as it is in the Metaphysics and the Politics, |The end of art: the οἰκεία ἡδονή.| but it is assumed throughout, and such assumption practically more valuable than argument. We have left behind us the noble wrongheadedness of the Platonic depreciation of pleasure; we are even past the stage when it might seem necessary to plead humbly and with bated breath for its locus standi. Moreover, the doctrine of the oikeia hedone not only by implication lays down the end of all art, but guards (in a fashion which should have been sovereign, though the haste and heedlessness of men have too often robbed it of its virtues) against one of the greatest dangers and mistakes of criticism in time to come. That what we have to demand of a work of literature is pleasure, and its own pleasure—how simple this seems, how much a matter of course! Alas! Aristotle himself is not entirely free from the charge of having sometimes overlooked it, while since his time the great majority of critical errors are traceable to this very overlooking. The obstinate ignoring or the captious depreciation of Latin literature by the later Greeks; the wooden “Arts of Poetry” of the Latins themselves; the scorn of Chaucer for “rim-ram-ruffing”; of the Renaissance for mediæval literature; of Du Bellay for Marot; of Harvey for the Faerie Queene; of Restoration criticism for the times before Mr Waller improved our numbers; of our Romantic critics for Dryden and Johnson; of Mr Matthew Arnold for French poetry,—all these things, and many others of the same class, come from the ignoring of the oikeia hedone, from the obstinate insistence that this thing shall be other than it is, that this poet shall be not himself but somebody else.