So far so good. But what shall we say of this: “All others must follow afar off. For Macer and Lucretius are indeed to |Other epic and didactic poets.| be read, but not for supplying phrase—that is to say, the body of style. Each is elegant in his own subject, but the one is tame and the other difficult.” Now, as to Macer we know little or nothing; he seems to have been a sort of Roman Armstrong or Darwin, who wrote about herbs, drugs, &c.^[413] But Lucretius—a greater master of phrase than Landor himself, nay, a greater, perhaps, than Milton—"not good for supplying" it, and merely “difficult”? One wants, again, some Aristophanic interjection. Varro is damned with faint praise as not indeed despicable (non spernendus quidem), but parum locuples. Ennius is spoken of as some of our own critics used to speak of Chaucer—as a gigantic and aged oak, venerable but not beautiful. Ovid is “wanton,” and too fond of his own conceits. Valerius Flaccus is a great loss. Others—Severus, Bassus, Rabirius, Pedo^[414]—names to us mostly, though we have fragments of at least three, are dismissed, the two first with high praise, the two last with the scarcely enthusiastic remark that the orator may read them, if he has time. Lucan is ardent, eager, and of noble sententiousness, but rather an orator than a poet. Domitian would have been the greatest of poets, if the gods had pleased. But, unluckily, they did not please!

In elegy Tibullus is Quintilian’s choice, but he admits that others prefer Propertius. Ovid is more luxuriant than either; |Elegiac and miscellaneous.| Gallus harsher. Horace receives praise thrice over as terse, pure and just in satire, bitter in iambics (lampoons), and almost the only Roman deserving to be read^[415] in lyric—where he sometimes soars. He is full of pleasant grace, and is agreeably audacious. After this, or rather before it, it is not surprising that Catullus is only mentioned for “bitter” iambics. As older satirists (“Satire is ours!” says Quintilian with a pleasant patriotic exaltation), Lucilius and Varro have praise.

The remarks on Tragedy we are unfortunately unable to check; but it is interesting that Quintilian apparently thought |Drama.| Latin better off here than in Comedy, which we certainly should not have expected. He quotes the traditional praise of the language of Plautus without expressing any opinion on it, but in a fashion pretty clearly intimating that he was unable to agree.^[416] “The ancients extol Cæcilius,”—another phrase which can only be pointed in one way; and Terence, though extremely elegant in his kind, scarcely attains to a faint image of Greek. It seems, however, as if he would have thought better of Afranius had it not been for that foulness of subject which, from the frequency of mention of it in connection with the author, seems to have turned the by no means squeamish stomach even of less moral Romans.

He is much more patriotic in regard to History—in fact, his patriotism rather outruns his discretion. One may have the |History.| highest admiration of Sallust’s masterly sweep (“immortal velocity” Quintilian himself calls it), of his pregnant thought and vivid representation, yet hesitate to match the two miniatures or Kit-cats of the Jugurtha and the Catiline against the mighty grasp and volume, alike in whole and in detail, of the Peloponnesian War. It must, however, be remembered that Sallust wrote a larger History in four books, which is lost except in fragments. Livy with Herodotus, though Quintilian thinks the latter ought not to feel indignant at the match, is only not so impar congressus, because there is here no unequality in scale and range. But, once more, the expression of opinion is a valuable one, and we must come back to it. Of Servilius Nonianus and Aufidius Bassus we know nothing; but the section ends with a high and most interesting panegyric on a certain unnamed living historian, whom we must all hope, though some would identify him with Pliny, to be Tacitus. If he had been equalled with even the greatest of the Greeks, Thucydides might have made room for him with hardly condescending good-humour.

Having thus put himself in the mood of “our country right or wrong” by this time, Quintilian is emboldened to match |Oratory—Cicero.| Cicero against any Greek orator, though he proceeds to explain that this is not meant to depress Demosthenes. Thus minded, he certainly does not go to work “with a dead hand,” as the French say, and endows his favourite not merely with the energy of Demosthenes, but with the flow of Plato and the sweetness of Isocrates. (One may invoke the aid of Echo—courteous nymph—and assent at least to Isocrates.) And then he passes to other Latin orators, praising Pollio for pains, and Messala for an aristocratic elegance. Cæsar (it is noticeable that he says nothing of the Commentaries) has qualities in his speeches which might have made him a rival to Cicero, especially the elegance of his diction. Cælius for wit; Calvus for severe correctness; others for other things, receive homage.

In Latin philosophy he again, with some rashness, advances Cicero as a rival to Plato, and ends with a curious and interesting |Philosophy—Cicero and Seneca.| passage on Seneca, whom he had been supposed to condemn and even hate, whose vitiated taste he still reprehends, but to whose real merits he now makes handsome concessions. This is quite one of Quintilian’s best “diploma-pieces” as a literary critic, in the division of decided but not illiberal censure, qualified by just and not grudging allowance for merits. It is a pity that it is too long to quote.

With the rest of the book, interesting as it is and germane to our subject, we must deal more succinctly. It first handles |Minor counsel of the Tenth Book.| Imitation of the styles just run through, and contains some of the best advice available anywhere on that head. The danger of imitating one style is especially dwelt upon, and Quintilian draws nearer to Greece or England than to Rome, in the simple observation that he has known Ciceronians think themselves quite accomplished when they ended a sentence with esse videatur. Habemus criticum! Another most excellent chapter is devoted to Writing—that is to say, to “exercises in composition,” which, under the dispensation of Rhetoric, were much in use. We know that Cicero wrote theses at the moment, and on the subject, of his sorest trouble. Quintilian’s advice here again is excellent; and if it were worse, it would be saved by the delightful story he tells of Julius Florus, a Gaulish provincial (for literary talent was beginning to be centrifugal), who, to his nephew and Quintilian’s friend Julius Secundus, when he was troubled about his style, observed, “Do you want to write better than you can?” Nor should the subsequent observations on rough copies be passed over. The rough copy is the superstition of those who wish to write better than they can. In some respects, and especially for the urbane, intimate, un-Philistine common-sense of it, this is one of Quintilian’s best chapters. He follows it up by a short one on Correction, wisely observing that we may indulge in that too much; by another on Translation, dedication-writing, and so forth; by yet another on premeditation, and by a last on speaking extempore, which he says (irrefutably from his oratorical point of view, and perhaps not much less so from the point of general literature) is all but a sine qua non. In these later chapters he is, as we may say, pursuing the art of the critic the reverse way—that is to say, he is counselling the author how to anticipate the critic. But it ought to be needless to add that they are not the less important as chapters of a manual of criticism itself.

The Eleventh book is wholly professional, dealing with the manner and general conduct appropriate to the orator, the |Books XI., XII.: The styles of oratory.| cultivation of the memory, delivery, gesture, and so forth. It therefore yields us nothing, while the beginning of the Twelfth, with its respectable paradox that a good orator must be a good man, may not look more promising, nor the subsequent demonstration that he ought to be acquainted with the civil law, and with examples and precedents, that he must have firmness and presence of mind, years of discretion, and also reasonable fees and retainers, that he must study his brief, not lay himself out too much for mere applause, and while preparing carefully, be ready with impromptus and extempore speech when necessary. But when we are beginning to get a little weary of this good-man-of-the-Stoics, called to the bar, an abrupt turn to the style of oratory refreshes us. The sketch of literature in the Tenth Book had been made, it is to be remembered, from a somewhat different point of view; it had been occupied with the authors whom an orator should read, and the qualities which were to be discovered in them. Here the standpoint changes, and the literary quality of what the orator himself is to produce is the question. After a distinctly interesting parallel from painting and sculpture, to illustrate differences of style, Quintilian takes up these differences, in some cases repeating the descriptions of Book X., in reference to Latin orators, and especially renewing his eulogy of Cicero as excellent in every oratorical quality. This, however, he admits, was by no means the universal opinion, either of Cicero’s contemporaries or of succeeding critics. And he hits a distinct blot in too much literary criticism by pointing out that while these earlier critics usually censured the great Arpinate as too flowery, too Asiatic, too fond of jests, his, Quintilian’s, |“Atticism.”| own contemporaries were apt to speak of him as dry and wanting in succulence. Next he turns to the three famous divisions of oratorical style—Attic, Asiatic, and Rhodian: the first chastened, energetic, correct; the second redundant and flowery; the third a mixture of the other two. And then, with his usual unpretending shrewdness, he proceeds to point out that although there certainly is an Attic style, and this style is far the best, yet that there are many, nay, infinite varieties and subdivisions of it—that Lysias is not in the least like Andocides, Isocrates different from either, Hyperides apart from all three. And so, with perfect good sense, he objects to the limitation of the “odour of thyme,” the Attic charm, to those who “flow as a slender stream making its way through pebbles”—that is to say, to those who write in a studiously correct and elegant style, with no magniloquence or turbid rush.

More interesting still, because it is the first and by far the best thing of the kind that we have, is the passage which follows on the oratorical—we may excusably read the |Literary quality of Greek and Latin.| “literary”—qualities of the Latin language as compared with the Greek. There are, it is true, phonetic difficulties here, and probably no wise man will pretend to understand Quintilian’s praise of the “sweetness” of the Greek phi, as compared with the harsh repulsiveness of the Latin f and v. No one but a student of phonetics themselves (that is to say, of a science as arbitrary as the most technical part of the Hermogenean rhetoric) can perceive any difference between phi and f, or the repulsiveness of the latter and of v, or the extra harshness of fr as in frangit. Fr, to a modern English ear, gives a very harmonious sound indeed. He incidentally, however, as far as v is concerned, gives us a “light” by saying that the sound of the digamma was preserved in Servus and Cervus, so that the Romans adopted the Wellerian form in these words; and has a specially interesting observation (because it applies equally to Anglo-Saxon) on the ugliness of terminations in m, “like the lowing of an ox,” as opposed to the clear ringing Greek n. The intonation of Latin he also thinks inferior to Greek, and still more the vocabulary. But sursum corda! after all:—

“Wherefore, if any demand from Latins the grace of Attic speech, let him give us the same sweetness of utterance, and an equal abundance of words. If this be denied, we must match our meaning to the words we have, nor mix a too great subtlety of matter with words too strong, not to say too stout, for it, lest the combination lose either excellence. The less the mere language helps us, the more we must reinforce ourselves by invention of matter. Let us extract sublime and varied meanings. Let us stir all the passions, and illuminate our addresses with gleaming metaphor. We cannot be so graceful; let us be more vigorous. We are conquered in subtlety; let us prevail in weight. They are surer of propriety, let us overcome by numbers. The genius of the Greeks, even in their lesser men, has its own ports; let us spread more ample sail and fill it with a mightier breeze. Nor let us always seek the deep; we must sometimes follow the windings of the shore. They may slip over any shallows; let me find a deeper sea in which my bark may not sink.”

A very little farther^[417] and we find Wordsworth’s paradox in the Preface to the Lyrical Ballads—that there is no natural eloquence but in the speech of ordinary folk—anticipated, stated, and very happily and thoroughly answered, though in reference to prose, not verse; and after this, some interesting further observations on sententiæ—deliberate and ostentatious sententiousnesses.

Later still he returns upon himself, and adopts a fresh threefold division into ἰσχνὸν or plain; ἁδρὸν or grand; and ἀνθηρὸν or florid, examples of each of which, with oratorical adaptations, he proceeds to give, perorating on plain and florid style, in a manner not unworthy of his precepts. He concludes with a sort of postscript on the necessity of the orator’s withdrawing before his natural force is abated, and thus leads, by a not ungraceful parable, to his own Finis.

It may be hoped that the above analysis, however jejune and imperfect, of this remarkable book will at least serve as a basis for some intelligible, if brief, remarks on its position |Quintilian’s critical ethos.| and value in the history of literary criticism. Its status as a document of this is, like that of all other ancient documents without exception (even the Περὶ Ὕψους cannot rank as completely exceptional), an indirect one, one of but partial relevance to the gospel of criticism. The Law of Rhetoric was but a schoolmaster, teaching, like all good schoolmasters, many things which had no absolute bearing on the future life of its pupils. And it is all the more curious that Quintilian should nevertheless give us so much that is of direct importance, because he is not merely a literary critic at intervals, but almost a literary critic malgré lui. Except in the case of Cicero, where his professional feeling comes in, he displays no very great enthusiasm for literature. He is never tempted, as not merely Longinus, but even Dionysius, is, to take a particular author, book, piece, and thoroughly analyse him and it, to grasp it, turn it lovingly inside out, hold it up to the admiration of others, deck it with the ornament, and adore it with the incense, of his own. His interest, though liberal, is just a trifle utilitarian. He holds, like Scott’s counsellor, that “a lawyer without history or literature is a mere mechanic,” and he studies both accordingly; but his study is mainly a means to an end. He may not be exactly insensible to the pure beauty of literature in and by itself; but it may be suspected that, if he spoke of it freely, he would speak in much the same tone that he uses in an odd passage^[418] about working in the country, where he thinks the beauty of tree and flower, the song of birds, the sound of streams, likely to distract rather than to inspire. The prose of the Roman nature, its businesslike character, its matter-of-factness, all betray themselves a little in him.

It is therefore not wonderful that he embodies for us, in a vary edifying fashion, that distrust of the Romantic which appears so often, if not so constantly, in the post-Homeric classical ages, up to his own time, though soon after it was to break down in writers like Apuleius. We saw that if he did not absolutely dislike or despise, he ignored the romantic element in Xenophon, that the “seizing” situation of the Ten Thousand, leaderless though victorious, a handful isolated in the heart of a hostile country, the moving accidents of their journey across the mountain walls and through the warlike clans of Kurdistan, and all the rest, till the sight of the sea, and the rush to the hill-brow to behold it, and the shout of welcome—even though the incident be as rhetorical a thing as history and literature contain—pass entirely unnoticed by him. His astonishing dismissals of Lucretius (though he may have been prejudiced by Cicero^[419] there) as merely “difficult,” of Catullus as merely “bitter,” group themselves with this very well. The grim force of the Lucretian despair, which would so fain persuade itself to be scientific acquiescence in contemplation from the temples of the wise, the throb of the Catullian passion, are not his business. Indeed, what contio, what judices, would pay any attention to the drift of the atoms in the void? what respectable paterfamilias but must highly deprecate verses, not merely immoral but extravagant, to Ipsithilla and Lesbia, attempts to reproduce, in sober Latin, the Greek ravings of a Sappho or about an Attis? Apollonius Rhodius, too, who to us seems a Romantic before Romanticism, touches no chord in Quintilian’s breast. And we may be tolerably certain that the chords which were not responsive in the breast of Quintilian were at least equally mute in other breasts of his time.

But these shortcomings are not only inevitable, they are, for the purpose of the historian, almost welcome. We may protest as lovers, but we register and interpret as students. Moreover, Quintilian, like all the greater men in all periods, and some even of the smaller in some, supplies us with a great deal of matter for registration and interpretation, without any protest at all. In the first place, we see in him the gradual deflection or development (whichever word may be preferred) of Rhetoric into pure Literary Criticism, assisted by the practical disappearance of symbouleutic oratory, by the degradation of epideictic, and by the practical Roman contempt for mere technicalities, unless, as in the case of law, they are intimately and almost inextricably connected with some practical end. It would be possible, as we have seen, by a process of mere “lifting out,” with hardly any important garbling of phrase, to extract from the Institutions a “Treatise on Composition and Critical Reading” which would be of no mean bulk, of no narrow range, and would contain a very large proportion of strictly relevant and valuable detail. And this treatise would be illuminated—for practically the only time, in the range of ancient literature on the subject, to any considerable extent—by that searchlight of criticism, the comparative method; while it would also display, throughout, the other illuminative powers of wide reading, sound judgment, and an excellent and by no means merely pedestrian common-sense.

We may regret, indeed, as we have regretted already, that these good gifts were not turned to the business of direct literary examination of particular books and authors, after the fashion of Dionysius; but it is quite evident why they were not. And their actual use has resulted in passage on passage, in chapter on chapter, of the most precious material. Quintilian can only be despised by those who consider themselves defrauded if critics do not attempt the meteorosophia of the highest æsthetic generalisations. It is, on the other hand, certain that these airy flights, in this particular matter, have too often had the ultimate Icarian fate, and have not often met even with the temporary Icarian success. The “high priori way” has never led to any permanent conquest in literary criticism; and it is never likely to do so, because of the blessed infinity and incalculableness of human genius. It has constantly led that genius into deserts and impasses. Even things that look like generalisations firmly based on actual experience have to be cautiously guarded, and put forth merely as working hypotheses. You make, with the almost superhuman compound of learning and reason belonging to an Aristotle, a general theory of Poetry, and a special one of Tragedy, which require, and command, almost universal agreement. In a few hundred years there drops in a graceless sort of prose tale-tellers, who by establishing, slowly and uncertainly at first, but after a couple of thousand years unmistakably, the kind of prose fiction, sap the very foundations of your theory of poetry. Later still arises a more graceless sort of strolling actors, ne’er-do-weel university men in England, cavaliers or shavelings in Spain, who in the same way bring it about that your theory of tragedy has to acknowledge itself to be only a theory of one kind of tragedy.

The other way is the way of safety; and if it be objected that it is the way of plodders only, one could undertake to make a very striking company of plodders from Longinus to Mr Arnold, who, sometimes not quite wittingly or willingly, have done all their best work in it. It would be but re-summarising our summary to point out once more, in any fulness, what work Quintilian has done. He has given us a history in little of the choicest Greek and Latin literature; he has drawn and placed for us the contrasted styles, not merely of oratorical, but of all prose composition; he has handled the literary side of grammar with singular fairness and sense; and has dealt more satisfactorily—to us at least—than any other ancient writer with the all-important and most difficult question of euphony in written speech. No one among ancient writers has treated the important but delusive subject of the Figures with more sense and skill; no one has contrived to get, out of some of the merest technicalities of the Rhetoric of the Schools, such a solid extract of critical power. The technical observations in Book X., which for want of space we passed over rapidly, form the most invaluable Introduction to Composition to be found in any language; they put our modern books of the kind to shame, at once by the practical character of their suggestions, and by their freedom from mere mechanical arbitrariness of prescription on points where idiom, good usage, and individual ability are really the only arbiters. And lastly, on the all-important and ever-recurring battle of the styles, Plain and Ornate, Attic and Asiatic, or whatever antithesis be preferred, it would be almost impossible to find a more intelligent pronouncement than Quintilian’s.

He can therefore afford to smile at those who say that he chancelle sur le terrain des principes,^[420] and to reply that terrain is exactly the word which does not apply to the principles with which he is reproached for not dealing. The only reproach to which he is perhaps open is one which all antiquity, from Aristotle to Longinus, and including both these great men, shares with him. This is the reproach of never completely clearing up the mind about Rhetoric, and of perpetually confusing it with the Art of Prose Literature, or else leaving prose literature without any “art” at all. We have seen, long ago, how this confusion arose, and how it was maintained by conditions which, though working more feebly in Quintilian’s days, were still working. The matter came to a head (though, oddly enough, the person chiefly concerned seems not quite to have understood it) when Lucian formally renounced Rhetoric and took to essay-writing in dialogue, when Apuleius in the Golden Ass mingled declamation, dialogue, philosophy, and romance in one olla podrida, with a daring sauce of new prose style to make it go down. But the barbarians were then at the gates; and the real recognition and reconstruction was not to take place for ages later, if it has completely taken place even yet.

AULUS GELLIUS: THE ‘NOCTES ATTICÆ’—MACROBIUS: THE ‘SATURNALIA’—SERVIUS ON VIRGIL—OTHER COMMENTATORS—AUSONIUS—THE ‘ANTHOLOGIA LATINA’—THE LATIN RHETORICIANS—RUTILIUS LUPUS, ETC.—CURIUS FORTUNATIANUS, HIS CATECHISM—MARIUS VICTORINUS ON CICERO—OTHERS—MARTIANUS CAPELLA.

The period from Nero to Hadrian is not merely the central and most important period of Latin criticism, but it contains a proportion altogether disproportionate of the bulk as of the value of Latin contributions to the subject. We must, however, complete our view of that subject, before summing up its general characteristics, with another chapter surveying the yield of the second, third, and fourth—perhaps, in view of the uncertainty of date of Martianus, we should add the fifth—centuries. The crop, if not very abundant, or of the very greatest value, is neither very scanty nor very uninteresting. It shall consist, in the specimens of it which we can afford to examine, first of the two famous and by no means unamusing miscellanists, the authors of the Noctes Atticæ and the Saturnalia; then, by an easy transition, of the commentators and scholiasts represented by their prior Servius, himself an interlocutor in the Macrobian symposium; in the third place, of a poetical contingent, much less important indeed than that furnished by the satirists from Horace to Martial, but not quite insignificant; and lastly, of the technical rhetoricians, ending with one of their latest representatives, but perhaps the most interesting of all, Martianus Capella. The chapter will thus, at least, not lack variety.

It would be difficult to have a better example of the indisposition of the Latin mind towards literary criticism proper, |Aulus Gellius: the Noctes Atticæ.| than that which is afforded by the famous Noctes Atticæ of Aulus Gellius.^[421] We know nothing of this good person except that he was probably of more or less pure Roman descent, that he probably lived for the most part of his life at Rome and at Athens in the early second century, that he was a friend of Herodes Atticus, probably knew Plutarch, and was extremely intimate with, and a great admirer of, the rhetorician Favorinus. The well-known miscellany which he has left us, and which, in purporting to give the results of study or conversation in an Attic country-house, has been for seventeen hundred years so fruitful in imitations—mother, indeed, of a family sometimes a great deal fairer than herself—is an amusing book and a valuable, because it preserves for us a great number of quotations from lost authors or books, because its farrago of matter is good pastime, and not least because of a certain Pepysian or Boswellian quality in its author. But though, amid its jumble of things ethical, physical, logical, legal, and, above all, philological, perhaps the larger part is occupied with literature or at least with books, it is quite astonishing how small is the proportion that can be called literary criticism, and how rudimentary and infantine even that small proportion is. Gellius had nearly all the qualities and acquirements of the dictionary-maker; he was interested in etymology, was a most exact and careful purist in the definition and usage of words, and evidently prided himself on his collections of illustrative phrases and passages.^[422] But almost invariably it must be said of him that hæret in litera, or, if he escapes that adhesion, that he gives himself over to the substance and meaning, not to the literary form and art, of what he quotes and studies. In all the nineteen or twenty books of his work there are probably not nineteen or twenty pages of real literary criticism; and where he does give us any it is of the “strawiest” character. Take, for instance, his comparison (ii. 23) of the Greek and Latin comic writers, and especially of some passages of Cæcilius with their originals in Menander. In preferring the Greek he is, of course, quite right; but it is noteworthy that he can hardly render any specific reason for his preference. He says, vaguely if truly, that the Latins seem low and sordid beside the wit and brilliancy of the Greeks, that Cæcilius appears stupid and frigid by Menander. But as to detail he prudently adds, nihil dicam ego quantum differat; and, less prudently transgressing this rule later, confines himself wholly to the matter, accusing the Roman of leaving out a simplex et verum et delectabile remark of the Greek. And if he comes a little nearer in praising (or making his favourite Favorinus praise) the flavum marmor of Ennius, it is still pretty clear that he does this merely or mainly from the side of the dictionary-maker, pleased at getting a light on the exact meaning of flavus. Although to our ears his preference (vi. 20) of “Ora” to “Nola” (in the passage which Virgil is said to have altered from a rather petty spite to the Nolans), “because it makes a sweet hiatus” with Vesevo at the end of the preceding line, may seem all wrong, the principle is æsthetic if the application is not. But, as a rule, we shall find that his critical opinions, where they are not concerned with purely verbal matters, are always decided by moral, philosophical, or in some other way extra-literary considerations. Even in an extremely interesting passage towards the end (xix. 9) where he makes Græci plusculi attack the Spanish-Latin rhetor Antonius Julianus^[423] on the score of the inferiority of Roman to Greek erotic poets, and gives the passages with which Julianus retorted, the chief interest for us is that even the Greeks except Catullus to some extent, and Calvus, from their censure. For there is little or nothing but logomachy to be got out of the condemnation of Hortensius as invenustus and Cinna as illepidus.

This same imputation of logomachy is hard to clear from the dispute in x. 3, whether, though Caius Gracchus is undoubtedly fortis et vehemens, it is or is not intolerable that he should be deemed severior, amplior, acrior, than Cicero. If Gellius had kept to the same words, and had said fortior and vehementior, the observation just made might seem unkind; but as it is, one seems to be dropping into the well-known jargon of our own times, and of all times, to be hearing one reviewer asserting that Johnson is “alert” and another replying that Thompson is “nimble,” or opposing the “poignancy” of Smith to the “swiftness” of Brown. But the attention to words certainly comes in better when the critic objects to the use, in an otherwise non sane incommode adapted version from Euripides by Ennius, of ignobiles and opulenti for ἀδοξοῦντες and δοκοῦντες. XII. 2, however, is a good locus for us in more ways than one. It opens with a sketch of the difference of opinion about Seneca in the age succeeding his own, a difference of which Quintilian had, a little earlier, given us an inkling. “Some,” says Gellius, “think of him as of a most unprofitable writer, one not worth reading, because they hold his style vulgar and hackneyed, his matter and opinions distinguished either by inept and empty haste (impetu) or by frivolous and Old-Bailey (causidicali) wire-drawing, his erudition vernacular and plebeian, and possessing nothing either of the dignity or the grace of the classics. Others, while not denying that he has little grace of phrase, maintain that his matter lacks neither information nor teaching power, and that he has no unhappy gravity and severity in castigating vice.” He himself will give no general censure, but consider Seneca’s opinion of Cicero and Ennius and Virgil. This “consideration,” according to his wont, is rather a string of quotations with objurgatory epithets than a regular criticism. One may not agree with Seneca or one may (there are certainly some who would indorse his confession and avoidance of Cicero’s faults in the words non ejus sed temporis vitium). But the words which Gellius himself uses—insulsissime, homo nugator, inepti et insubidi^[424] hominis joca—surely require some little argument to justify them, and this argument is what Gellius never gives. We may thank him, however, for the criticism as well as for the anecdote preserved (xiii. 2) in the story of the meeting of the tragic poets Pacuvius and Accius at Tarentum, in the extreme old age of the former. Pacuvius had asked his young guest and craftsfellow to read his tragedy of Atreus, and, after the reading, praised it as sonorous and grand, but perhaps a little harsh and austere. “It is so,” said the junior, “but I am not very sorry, for I hope to improve. It is the same in wits as in fruits: the hard and harsh mellow and sweeten, but those that are at first flabby, and soft, and moist, do not ripen but rot. I thought it best to have something in my genius for time and age to mitigate.” A sound principle, though not quite a universal one, as one may see in studying a certain life-work which ranges from “Claribel” to “Crossing the Bar.”

He is in his more meticulous moods when (xiii. 18) he accuses Plato of misquotation and Euripides of plagiarism; but a couple of chapters later a set discourse on euphony, starting from a saying of Valerius Probus, seems to promise well. Some one had asked Probus whether it was better to use the terminations em or im, es or is, for the accusative, where both occur. Aurem tuam interroga, said Probus, which is no doubt the conclusion of the whole matter. But his questioner, either foolish or dogged, asked how he was to do this, and Probus replied, “As Virgil did when he wrote Urbisne invisere Cæsar but Urbes habitant magnas.” Nor are we sorry to hear that when the questioner still bored on, saying that he could not understand why one should be better in one place and another in another, Probus retorted, “You need not trouble yourself; it will do you no harm whatever you use.” Prope inclementer, says Gellius (“Served him right,” most of us will say). But he goes on to accumulate some other instances of this application of the rule of euphony, and perhaps here draws as near to true criticism as he ever does. Nor is he wrong, though he may be fanciful, in deciding in regard to certain almost literal Virgilian imitations of Homer, that the Greek is simplicior et sincerior, Virgil νεωτερικώτερος et quodam quasi ferrumine immisso fucatior.^[425]

He may strain the word again too much, when he bestows a page on the difference of multis hominibus and multis mortalibus (xii. 28), but he recovers esteem when in xiv. 6 we find him rejecting, not without contumely, contributions to his Noctes on the questions “Who was the first grammarian?” and “Why Telemachus did not nudge his bedfellow Pisistratus but kicked him?” &c., &c. Properans reddidi, says he, with the shudder one can fancy, though, to tell the truth, he does himself “something grow to” this kind of disease.

We may close this anthology of the Gellian criticisms with some account of one of the most elaborate—a discourse of Favorinus on Pindar and Virgil.^[426] After quoting the Roman poet’s traditional saying about himself—that he brought forth his verses as a bear does her cubs, licking them slowly and busily into shape—he points out that the facts exactly bear out the description, and that certain verses, not having undergone the process of licking, are very inferior to the others. Among these unlicked cubs, it seems, Favorinus would place the Etna passage. Even Pindar himself, whom Virgil followed, is, the critic thinks, ipso insolentior tumidiorque in the place; but Virgil’s verse is such that Favorinus calls it “begun, not made.” And, the two passages having been cited in full, he indulges in the following drastic verbal censure: “At the very beginning, Pindar, paying more attention to the truth, said what was the fact, and a matter of ocular demonstration, that Etna smoked by day and flamed by night. But Virgil, laboriously seeking noisy-sounding words, confuses the two. The Greek says plainly that fountains of fire are belched forth, and rivers of smoke flow, and yellow, curling volumes of flame are borne down to the shores of the sea like fiery snakes; but this fellow of ours, choosing to interpret ῥόον καπνοῦ αἴθωνα by atram nubem turbine piceo et favilla fumantem, makes a crass and clumsy mixture, and translates the κρουνοὺς of flames, both harshly and inexactly, into ‘globes.’ Again, when he talks of ‘licking the stars,’ he makes an idle and empty exaggeration. Nay, the phrase, ‘emitting a black cloud of smoke full of pitchy whirlwinds and glowing ashes,’ is bad style and almost nonsense.^[427] For glowing things, quoth he, neither smoke nor are black; unless by an improper vulgarism he applies candente, not to glowing but to merely ‘hot’ ash. But when he talks of ‘rocks and cliffs being belched and flung up,’ adding immediately that they are ‘melted, and groan, and are flung in handfuls into the air,’ neither did Pindar write this, nor would anybody else think of saying it, and the thing is the most monstrous of all monstrosities.”

The classical hatred of bombast and the classical propensity to “stick at the word” in criticism are both very well illustrated here; but we should hardly guess, from the sample, that there existed in classical times much power of grasping the literary and poetical merit of a passage as a whole. Virgil, if he had cared to defend himself, would, no doubt, have called attention to the Pindaric words, τέρας and θαῦμα, as justifying even “monstrosity” in his own expanded description, and have urged that this description was at least partly intended to indicate the terror and confusion of mind caused by so portentous a phenomenon.

But this absence of the synoptic grasp of æsthetic means, as applied to produce literary effect, is precisely what we notice most in the ancient criticism which has come down to us. And it may be added that it is also precisely what we should expect to follow from the limitations of the ancient Rhetoric. Grammar provided rules for the arrangement of words, and lexicography provided lists of them, with their authority and their use carefully ticketed; so here criticism was at home. Rhetoric provided lists of Figures with which a man could compare the passage before him. But there was no training in the process of simply “submitting to” this passage, interrogating oneself whether it exercised a charm or not, and then interrogating oneself further whether that charm was genuine, and what was its cause. After all, Gellius has, as we have seen, sometimes come near to the discovery of the true method, and that he loved literature there can be no doubt.^[428]

Nor, much later, shall we find things different with that favourite of the Middle Ages and of Dr Johnson’s youth, |Macrobius: The Saturnalia.| Macrobius,^[429] who, about the beginning of the fifth century, undertook a pendant to the work of Gellius. It is not surprising that the author, qui ot nom Macrobes,^[430] should have been a favourite (for his commentary on the Somnium Scipionis principally, no doubt) with the period between Darkness and Renaissance. He has precisely the “fine confused feeding” in the way of matter and manner that these ages loved; and they would not be likely to quarrel with him for his lack of the criticism which, as we shall see, they themselves hardly, in more than a single instance, relished or understood. But he certainly illustrates, even in a greater degree than Gellius, the small propulsion of the Romans and their vassals towards the proper subjects of this book. Once more we find that etymology, mythology, grammar, the farrago of the antiquary as distinguished from that of the literary enthusiast, of the philologist as opposed to the critic, receive ample attention. And, once more, what we are specially in quest of remains practically, if not entirely, unhandled.

There are few more striking loci in connection with this subject than the end of the first book of the Saturnalia. The guests have been talking mythology and etymology for some stricken hours, till at last a break occurs. Vettius Prætextatus, the host, has just ended a long mythological dissertation, to the admiration of everybody, when Euangelus (the irreverent humourist of the party) breaks in, with some amusement at the practice of citing Virgil as an authority. He supposes that the notion of making Latin poets into philosophers is an imitation of the Greeks, and hints that the process is dangerous, since even Tully himself, who was as formal a professor of philosophising as of oratory, so often as he talks of the nature of the gods, or of fate, or of divination, injures the glory which he has got together through his eloquence, by his desultory handling of things. Symmachus, the scholar-statesman, rebukes this blasphemer gravely, observing that, as for Cicero, he is conviciis impenetrabilis, and may be left aside for the moment, but that he fears Euangelus has learnt his Virgil only as boys do, and thinks him only good for boys, with nothing higher in him. Euangelus is by no means abashed, and takes the offensive. It was all very well, he says, for us as boys to take Virgil at our master’s valuation, but did not he himself pronounce himself far from faultless, inasmuch as he wished them to burn the Æneid? No doubt he was afraid, not merely of ethical blame for such scenes as the request of Venus to her lawful husband in favour of her illegitimate son, but of critical blame for his now Greek, now barbarous, diction, and for the awkward ordonnance of his work. To this, cum omnes exhorruissent, Symmachus, still calm and sententious, makes answer by putting Virgil beside Cicero, and saying of his glory, that as it can grow by no one’s praise, so it is diminished by no one’s abuse. Any grammarian, he continues, can refute these calumnies; and it would be a shame to ask Servius (the famous Virgilian scholiast, who is present) to take the trouble. But he should like to know whether, as Euangelus is dissatisfied with Virgil’s Poetic, he likes his Rhetoric better. “Oh!” says Euangelus, “you have made him a philosopher, and now you are going to make him an orator, are you?”

A conversation of this kind gives us no bad reason to expect something like literary criticism proper, something such as Coleridge has given us in the Biographia Literaria in reference to Wordsworth. But Symmachus for the time contents himself with undertaking to defend the Mantuan’s rhetoric, while the others overwhelm the impenitent Euangelus with a string of affirmations as to the poet’s proficiency in politics, law, augury, astrological and other philosophy, fidelity to the traditions of the Latin language, &c. But the justifications of these praises are deferred by the announcement of dinner, and for a time the conversation turns to lighter subjects—the famous string of stories for which Macrobius is most commonly quoted, including scandal about Princess Julia. Only in the third book, and then, it would seem, after a lacuna, is the detailed criticism of Virgil resumed.

There is no occasion to find fault with the quantity of it, for it fills, with a digression or two of the lighter kind, such as that on the dessert when it appears, four whole books, and some two hundred and forty pages in Eyssenhardt’s text. But the quality is, at any rate from our point of view, not quite so satisfactory. Much simply consists in citation of passages illustrating different “Figures.” A very large part, probably the largest, is mere and sheer quotation from Virgil himself, from Homer, and from other poets, Latin and Greek, with whom he is compared. And the comparison is carried on almost, if not quite entirely, on that most unsatisfying parallel-passage system which, in its abuse, has ever since been the delight of the pedantic criticaster—and the abomination of the true critic.

Of course the parallel passage, rightly handled, is invaluable—is practically indispensable to true literary criticism. The “Truth” passages of the Areopagitica and Halifax’s Character of a Trimmer, the “Death” passages of Raleigh, Marston, and Lee, the different harmonies which the motive “Ask me no more” has suggested to Carew and Tennyson, the accounts of the passing of Arthur or the parting of Lancelot and Guinevere in Malory, and in his probable verse original, are the constant, the inexhaustible, texts and exercises of the critical faculty. But I do not think it unfair to Macrobius to say that hardly in a single occasion does he make any such use of his parallels. And in literary criticism, properly so called, such parallels as