SOURCE: "Dionysius of Halicarnassus, Plutarch, Lucian, Longinus," in A History of Criticism and Literary Taste in Europe: From the Earliest Texts to the Present Day, William Blackwood and Sons, 1900, pp. 152-72.

[In the following excerpt, Saintsbury discusses elements of the sublime and comments on Longinus's literary and historico-critical importance.]

… It does not fall within the plan of this work to examine at any length the recently much-debated question whether the treatise Peri Hypsos is, as after its first publication by Robortello in 1554 it was for nearly three centuries unquestioningly taken to be, the work of the rhetorician Longinus, who was Queen Zenobia's Prime Minister, and was put to death by Aurelian. It has been the mania of the nineteenth century to prove that everybody's work was written by somebody else, and it will not be the most useless task of the twentieth to betake itself to more profitable inquiries. References which will enable any one who cares to investigate the matter are given in a note.¹ Here it may be sufficient to say two things. The first is, that these questions appertain for settlement, less to the technical expert than to the intelligent judex, the half-juryman, half-judge, who is generally acquainted with the rules of logic and the laws of evidence. The second is, that the verdict of the majority of such judices on this particular question is, until some entirely new documents turn up, likely to be couched in something like the following form:—

1. The positive evidence for the authorship of Longinus is very weak, consisting in MS. attributions, the oldest of which² is irresolute in form, while it certainly does not date earlier than the tenth century.
2. There is absolutely no evidence against the authorship of Longinus, only a set of presumptions, most of which are sheer opinion, and carry no weight except as such. Moreover, no plausible competitor has even been hinted at. I hope it is not illiberal to say that the suggestion of Plutarch, which was made by Vaucher, and has met with some favour, carries with it irresistible evidence that the persons who make it know little about criticism. No two things could possibly be more different than the amiable ethical knack of the author of the Moralia, and the intense literary gift of the author of the Peri Hypsos.

Another of the "Academic questions" connected with the book, however, is of more literary importance, and that is its proper designation in the modern languages. There has been a consensus of the best authorities of late years, even though they may not agree on other points, that "The Sublime" is a far from happy translation of hypsos. Not only has "Sublime" in the modern languages, and especially in English, a signification too much specialised, but the specialisation is partly in the wrong direction. No one, for instance, who uses English correctly, however great his enthusiasm for the magnificent Sapphic ode which Longinus has had the well-deserved good fortune to preserve to us, would call it exactly sublime,³ there being, in the English connotation of that word, an element of calmness, or at any rate (for a storm may be sublime) of mastery, which is absent here. And so in other cases; "Sublime" being more especially unfortunate in bringing out (what no doubt remains to some extent in any case) the inadequateness and tautology of the attempts to define the sources of hypsos. Hall, the seventeenth-century translator, avoided these difficulties by a simple rendering, "the height of eloquence," which is more than literally exact, though it is neither elegant nor handy. Nor is there perhaps any single word that is not open to almost as many objections as Sublime itself. So that (and again this is the common conclusion) it is well to keep it, with a very careful preliminary explanation that the Longinian Sublime is not sublimity in its narrower sense, but all that quality, or combination of qualities, which creates enthusiasm in literature, all that gives consummateness to it, all that deserves the highest critical encomium either in prose or poetry.

Few persons, however, whom the gods have made critical will care to spend much time in limine over the authorship, the date,⁴ the title, and the other beggarly elements in respect to this astonishing treatise. Incomplete as it is—and its incompleteness is as evident as that of the Poetics, and probably not much less substantial—difficult as are some of its terms, deprived as we are in some cases of the power of appreciating its citations fully, through our ignorance of their context, puzzled as we may even be now and then by that radical difference in taste and view-point, that "great gulf fixed," which sometimes, though only sometimes, does interpose itself between modern and ancient,—no student of criticism, hardly one would think any fairly educated and intelligent man, can read a dozen lines of the book without finding himself in a new world, as he compares it with even the best of his earlier critical masters. He is in the presence of a man who has accidentally far greater advantages of field than Aristotle, essentially far more powerful genius, and an intenser appreciation of literature, than Dionysius or Quintilian. And probably the first thought—not of the student, who will be prepared for it, but of the fairly educated man who knows something of Pope and Boileau and the rest of them—will be, "How on earth did this book come to be quoted as an authority by a school like that of the 'classical' critics of the seventeenth-eighteenth century, whose every principle almost, whose general opinions certainly, it seems to have been designedly written to crush, conclude, and quell?" Of this more hereafter. Let us begin, as in former important cases, by a short abstract of the actual contents of the book.

The author commences by addressing a young friend or pupil, a certain Postumius (Terentianus or Florentianus?), on the inefficiency of the Treatise on the Sublime by a certain Cacilius⁵ In endeavouring to provide something more satisfactory, especially as to the sources of Sublimity, he premises little more in the shape of definition than that it is "a certain consummateness and eminence" of words, completing this with the remark (the first epoch-making one of the treatise) that the effect of such things is "not persuasion but transport," not the result of skill, pains, and arrangement, but something which, "opportunely out-flung," carries everything before it. But can it be taught? Is it not innate? The doubt implies a fallacy. Nature is necessary, but it must be guided and helped by art. Then comes a gap, a specially annoying one, since the farther shore lands us in the midst of an unfavourable criticism of a passage supposed to come from the lost Orithyia of/Eschylus, which is succeeded by, or grouped with, other specimens of the false sublime, bombast, tumidity, and theparenthurson.⁶ Next we pass to "frigidity," a term which Longinus uses with a slightly different connotation from Aristotle's, applying it chiefly to what he thinks undue flings and quips and conceits. These particular strictures are, in Chapter V., generalised off into a brief but admirable censure of the quest for mere novelty, of that "horror of the obvious" which bad taste at all times has taken for a virtue. To cure this and other faults, there is nothing for it but to make for the true Sublime, hard as it may be. For (again a memorable and epoch-making saying) "the judgment of words is the latest begotten fruit of many an attempt."⁷

The first canon of sublimity is not unlike the famous Quod Semper, &c. If a thing does not transport at all, it is certainly not Sublime. If its transporting power fails with repetition, with submission to different but still competent judges, it is not sublime. When men different in habits, lives, aims, ages, speech, agree about it, then no mistake is possible.

The sources of Sublimity are next defined as five in number: Command of strong and manly thought; Vehement and enthusiastic passion—these are congenital; Skilfulness with Figures; Nobility of phrase; Dignified and elevated ordonnance.⁸ These, after a rebuke of some length to Cecilius for omitting Passion, he proceeds to discuss seriatim.… [What] he now calls "great-naturedness," holds the first place in value as in order, and examples of it, and of the failure to reach it, are given from many writers, Homer and "the Legislator of the Jews" being specially praised. This laudation leads to one of the best known and most interesting passages of the whole book, a short criticism and comparison of the Iliad and the Odyssey, whereon, as on other things in this abstract, more hereafter. The interest certainly does not sink with the quotation from Sappho, whether we agree or not (again vide post) that the source of its charm is "the selection and composition of her details." Other typical passages are then cited and criticised.

We next come to Amplification,—almost the first evidence in the treatise, and not a fatal one, of the numbing power of "Figures." Longinus takes occasion by it for many illuminative animadversions, not merely on Homer, but on Plato, Herodotus, Demosthenes, and Thucydides, whom (it is very satisfactory to observe) he includes among those who have "sublimity." This handling of Figures, professedly eclectic, is fertile in such animadversions in regard to others besides Amplification—Hyperbata, Polyptota, Antimetathesis, and others still—with especial attention to Periphrasis, to his praise of which the eighteenth century perhaps attended without due attention to his cautions.

Then comes another of the flashes of light. Dismissing the figures, he turns to diction in itself, and has a wonderful passage on it, culminating in the dictum, "For beautiful words are in deed and in fact the very light of the spirit,"—the Declaration of Independence and the "Let there be light" at once of Literary Criticism.

Here the Enemy seems to have thought that he was getting too good, for another and greater gap occurs, and when we are allowed to read again, we are back among the Figures and dealing with Metaphor—the criticism of examples, however, being still illuminative. It leads him, moreover, to another of his nugget-grounds, the discussion on "Faultlessness," which introduces some especially valuable parallels—Apollonius and Homer, Bacchylides and Pindar, Ion and Sophocles, Hyperides and Demosthenes, Lysias and Plato. Then we pass to the figure Hyperbole after a gap, and then to ordonnance and arrangement, with a passage, valuable but, like all similar passages in the ancient critics, difficult, on rhythm. After this a section on …—"littleness," "triviality"—leads abruptly to the close, which is not the close, and which, after some extremely interesting remarks on the ethical and other conditions of the time, ends with an unfulfilled promise of treating the subject of the Passions. The loss of this is perhaps more to be regretted than the loss of any other single tractate of the kind in antiquity. It might have been, and possibly was, only a freshening up of the usual rhetorical commonplaces about the "colours of good and evil," and the probable disposition of the hearer or reader. But it might also, and from Longinus's handling of the other stock subject of the Figures it is much more likely to, have been something mainly, if not wholly, new: in fact, something that to this day we have not got—an analysis of the direct appeals of literature to the primary emotions of the soul.

In considering this inestimable book, it is hardly possible to exaggerate the importance of these early words of it to which attention has been drawn above. The yoke of "persuasion" has at last been broken from the neck of the critic. He does not consider literature as something which will help a man to carry an assembly with him, to persuade a jury, to gain a declamation prize. He does indeed still mention the listener rather than the reader; but that is partly tradition, partly a consequence of the still existing prevalence of recitation or reading aloud. Further, it is sufficiently evident that the critic has come to regard literature as a whole, and is not distracted by supposed requirements of "invention" on the part of the poet, of "persuasion" on the part of the orator, and so forth. He looks at the true and only test of literary greatness—the "transport," the absorption of the reader. And he sees as no one, so far as we know, saw before him (except Dionysius for a moment and "in a glass darkly"), as Dante was the only man after him to see for a millennium and much more, that the beautiful words, the "mots rayonnants," are at least a main means whereby this effect is produced. Instead of style and its criticism being dismissed, or admitted at best with impatience as something fortikon, we have that gravest and truest judgment of the latter as the latest-born offspring of many a painful endeavour. Far is it indeed from him to stick to the word only: his remarks on novelty, his peroration (not intended as such, but so coming to us), and many other things, are proof of that. But in the main his criticism is of the pure asthetic kind, and of the best of that kind. It will not delay us too much to examine it a little more in detail.

The opening passage as to Cacilius, though it has tempted some into perilous hypothetic reconstructions of that critic's possible teaching, really comes to little more than this—that Longinus, like most of us, was not exactly satisfied with another man's handling of his favourite subject. And, curiously enough, the only specific fault that he here finds—namely, that his predecessor, while illustrating the nature of the Sublime amply, neglected to discuss the means of reaching it—rather recoils on himself. For there can be little doubt that the weakest part of the Peri Hypsos is its discussion of "sources." But the great phrase, already more than once referred to, as to transport or ecstasy, not persuasion, lifts us at once—itself transports us—into a region entirely different from that of all preceding Rhetorics, without at the same time giving any reason to fear loss of touch with the common ground and common-sense. For nothing can be saner than the handling, in the second chapter, of that aporia concerning nature and art, genius and painstaking, which has not infrequently been the cause of anything but sane writing.

After the gap, however, we come to one of the passages recently glanced at, and mentioned or to be mentioned so often elsewhere, which warn us as to difference of view. The passage, supposed to be, as we said, Æschylean and from the Orithyia, is no doubt at rather more than "concert-pitch." It is Marlowe rather than Shakespeare; yet Shakespeare himself has come near to it in Lear and elsewhere, and one line at least—

mian pareiras plektamen kheimarron—

is a really splendid piece of metre and phrase, worthy, high-pitched as it is, of the author of the Oresteia and the Prometheus at his very best. So, too, the much-enduring Gorgias would hardly have received very severe reprehension from any but the extremest precisians of modern criticism, at its most starched time, for calling vultures "living tombs." But the horror of the Greeks on the one hand for anything extravagant, bizarre, out of measure, on the other for the slightest approach in serious work to the unbecoming, the unpleasantly suggestive, makes Longinus here a very little prudish. And his general remarks are excellent, especially in reference to to parenthurson, which I have ventured to interpret, not quite in accordance with the general rendering, "the poking in of the thyrsus at the wrong time," the affectation of Bacchanalian fury where no fury need be.

But we still have the same warning in the chapter on Frigidity, coupled with another—that, perhaps, as sometimes happens, Longinus' sense of humour was not quite equal to his sense of sublimity, and yet another—that the historic sense, so late developed everywhere, was, perhaps, not very strong in him. We, at least, should give Timwus the benefit of a doubt, as to the presence of a certain not inexcusable irony in the comparison (in which, for instance, neither Swift nor Carlyle would have hesitated to indulge) of the times taken by Alexander to conquer Asia and by Isocrates to write the Panegyric. On the other hand, he seems to forget the date of Timwus when he finds … the paltrily funny, in the historian's connection of the Athenian Hermocopide and their punishment by Hermocrates, the son of Hermon. There is no reason why Timwus should not have been quite serious, though in the third century after Christ, and even in the first, the allusion might seem either a tasteless freethinking jest or a silly piece of superstition.

But by far the most interesting thing in this context is Longinus' irreconcilable objection to a fanciful metaphor which, as it happens most oddly, was, with a very slight variation, an equal pet of the Greeks of the great age and of our own Elizabethans. Every reader of the latter knows the phrase, "to look babies in the eyes" of the beloved—that is to say, to keep the face so close to hers that the little reflections of the gazer in the pupils of her eyes are discernible. The Greek term for these little images, and the pupils that mirrored them, was slightly different—it was … maidens. And as, from the famous quarrel scene in Homer downwards, the eyes were always, in Greek literature, the seat of modesty or of impudence, the combination suggested, not merely to Timwus but even to Xenophon, a play of words, "more modest than the maidens in their eyes," or conversely, as where Timwus, speaking of the lawless lust of Agathocles, says that he must have had "harlots" …, not "maidens" …, in his eyes. And Longinus is even more angry or sad with Xenophon than with Timwus, as expecting more propriety from him.

But whether we agree with him in detail or not, the inestimable passage, on the mere quest and craze for novelty, which follows, more than reconciles us, as well as the other great saying in cap. vi. as to the "late-born" character of the judgment of style, and that in the next as to the canon of Sublimity being the effect produced unaltered in altered circumstances and cases. When we read these things we feel that literary criticism is at last fully constituted,—that it wants nothing more save greater variety, quantity, and continuance of literary creation, upon which to exercise itself.

No nervous check or chill need be caused by the tolerably certain fact that more than one hole may be picked in the subsequent classification of the sources⁹ of hypsos. These attempts at an over-methodical classification (it has been said before) are always full of snares and pitfalls to the critic. Especially do they tempt him to the sin of arguing in a circle. It cannot be denied that in every one of the five divisions (except, perhaps, the valuable vindication of the quality of Passion) there is some treacherous word or other, which is a mere synonym of "sublime." Thus in the first we have hadrepobolon, mastery of the hadron, a curious word, the nearest equivalent of which in English is, perhaps, "stout" or "full-bodied," as we apply these terms to wine; in the fourth…, "noble," which is only "sublime" in disguise; and in the fifth axioma kai diarsis, of which much the same may be said.

Any suggestion, however, of paralogism which might arise from this and be confirmed by the curious introduction in the third of the Figures, as if they were machines for automatic sublime-coining, must be dispelled by the remarks on Passion of the right kind as tending to sublimity, and by the special stress laid on the primary necessity of megalophrosune, whereof hypsos itself is the mere apekhjma or echo. Unfortunately here, as so often, the gap comes just in the most important place.

When the cloud lifts, however, we find ourselves in one of the most interesting passages of the whole, the selection of "sublime" passages from Homer. A little superfluous matter about Homer's "impiety" (the old, the respectable, Platonic mistake) occurs; but it matters not, especially in face of the two praises of the "Let there be light" of the Jewish legislator, "no chance comer," and of the great en de phei kai olesson of Ajax, the mere juxtaposition of which once more shows what a critic we have got in our hands.

Not quite such a great one perhaps have we—yet one in the circumstances equally fascinating—in the contrasted remarks on the Odyssey. Longinus is not himself impious; he is no Separatist (he is indeed far too good a critic to be that). But he will have the Romance of Ulysses to be "old age, though the old age of Homer." "When a great nature is a little gone under, philomythia is characteristic of its decline." Evidently, he thinks, the Odyssey was Homer's second subject, not his first. He is "a setting sun as mighty as ever, but less intense": he is more unequal: he takes to the fabulous and the incredible. The Wine of Circe, the foodless voyage of Ulysses, the killing of the suitors—nay, the very attention paid to Character and Manners—tell the tale of decadence.

He is wrong, undoubtedly wrong—we may swear it boldly by those who fell in Lyonnesse, and in the palace of Atli, and under the echoes of the horn of Roland. The Odyssey is not less than the Iliad; it is different. But we can hardly quarrel with him for being wrong, because his error is so instructive, so interesting. We see in it first (even side by side with not a little innovation) that clinging to the great doctrines of old, to the skirts of Aristotle and of Plato, which is so often found in noble minds and so seldom in base ones. And we see, moreover, that far as he had advanced—near as he was to an actual peep over the verge of the old world and into the new—he was still a Greek himself at heart, with the foibles and limitations—no despicable foibles and limitations—of the race. Here is the instinctive unreasoning terror of the unknown Romance; the dislike of the vague and the fabulous; even that curious craze about Character being in some way inferior to Action, which we have seen before. By the time of Longinus—if he lived in the third century certainly, if he lived in the first probably—the romance did exist. But it was looked upon askance; it had no regular literary rank; and a sort of resentment was apparently felt at its daring to claim equality with the epic. Now the Odyssey is the first, and not far from the greatest, of romances. It has the Romantic Unity in the endurance and triumph of its hero. It has the Romantic Passion in the episodes of Circe and Calypso and others: above all, it has the great Romantic breadth, the free sweep of scene and subject, the variety, the contrast of fact and fancy, the sparkle and hurry and throb. But these things, to men trained in the admiration of the other Unity, the other Passion, the more formal, regulated, limited, measured detail and incident of the usual tragedy and the usual epic—were at best unfamiliar innovations, and at worst horrible and daring impieties. Longinus will not go this length: he cannot help seeing the beauty of the Odyssey. But he must reconcile his principles to his feelings by inventing a theory of decadence, for which, to speak frankly, there is no critical justification at all.

One may almost equally disagree with the special criticism which serves as setting to the great jewel among the quotations of the treatise, the so-called "Ode to Anactoria." The charm of this wonderful piece consists, according to Longinus, in the skill with which Sappho chooses the accompanying emotions of "erotic mania."¹⁰ To which one may answer, "Hardly so," but in the skill with which she expresses those emotions which she selects, and in the wonderful adaptation of the metre to the expression, in the mastery of the picture of the most favoured lover, drawing close and closer to the beloved to catch the sweet speech,¹¹ and the laughter full of desire. In saying this we should have the support of the Longinus of other parts of the treatise against the Longinus of this. Yet here, too, he is illuminative; here, too, the "noble error" of the Aristotelian conception of poetry distinguishes and acquits him.

With the remarks on … "amplification," as it is traditionally but by no means satisfactorily rendered, another phase of the critical disease of antiquity (which is no doubt balanced by other diseases in the modern critical body) may be thought to appear. Both in the definition of this figure and in the description of its method we may, not too suspiciously, detect evidences of that excessive technicality which gave to Rhetoric itself the exclusive title of techne. Auxesis, it seems, comes in when the business, or the point at issue, admits at its various stages of divers fresh starts and rests, of one great phrase being wheeled upon the stage after another, continually introduced in regular ascent. This, it seems, can be done either by means of … handling of topoi or commonplaces," or by deinōsis, which may perhaps be best rendered tour de force, or by cunning successive disposition … of facts or feelings. For, says he, there are ten thousand kinds of auxesis.

The first description of the method will recall to all comparative students of literature the manner of Burke, though it is not exactly identical with that manner; but the instances of means, besides being admittedly inadequate, savour, with their technicalities of terminology, much too strongly of the cut-and-dried manual. The third article, on a reasonable interpretation of epoikonomia, really includes all that need be said. But one sees here, as later, that even Longinus had not quite outgrown the notion that the teacher of Rhetoric was bound to present his student with a sort of handlist of "tips" and dodges—with the kind of Cabbala wherewith the old-fashioned crammer used to supply his pupils for inscription on wristband or finger-nail. Yet he hastens to give a sign of grace by avowing his dissatisfaction with the usual Rhetorical view, and by distinguishing auxesis and the Sublime itself, in a manner which brings the former still nearer to Burke's "winding into a subject like a serpent," and which might have been more edifying still if one of the usual gaps did not occur. Part, at least, of the lost matter must have been occupied with a contrast or comparison between the methods of Plato and Demosthenes, the end of which we have, and which passes into one between Demosthenes and Cicero. "If we Greeks may be allowed to have an opinion," says Longinus, with demure humility, "Demosthenes shall be compared to a flash of thunder and lightning, Cicero to an ordinary terrestrial conflagration," which is very handsome to Cicero.

Then he returns to Plato, and rightly insists that much of his splendour is derived from imitation, or at least from emulation, of that very Homer whom he so often attacks. The great writers of the past are to be constantly before us, and we are not to be deterred from "letting ourselves go" by any mistaken sense of inferiority, or any dread of posterity's verdict.

Then comes a digression of extreme importance on the subject of… "images." One of the points in which a history of the kind here attempted may prove to be of most service, lies in the opportunity it affords of keeping the changes of certain terms, commonly used in criticism, more clearly before the mind than has always been done. And of these, none requires more care than "Images" and "Imagination." At the first reading, the mere use of such a word as phantasiai may seem to make all over-scrupulousness unnecessary, though if we remember that even Fancy is not quite Imagination, the danger may be lessened. At any rate, it is nearly certain that no ancient writer, and no modern critic before a very recent period (Shakespeare uses it rightly, but then he was Shakespeare and not a critic), attached our full sense to the term. To Aristotle phantasia is merely … a "weakened sensation," a copy furnished by memory from sensation itself. Even animals have it. No idea of Invention seems to have mingled with it, or only of such invention as the artist's is when he faithfully represents natural objects. Of the Imagination, which is in our minds when we call Shelley an imaginative poet, and Pope not one, Sir Edward Burne Jones an imaginative painter, and any contemporary whom it may be least invidious to name not one, there does not seem to have been a trace even in the enthusiastic mind of Longinus, though he expressly includes Enthusiasm—nay, Passion—in his notion of it. You think you see what you say, and you make your hearers see it. Good; but Crabbe does that constantly, and one would hardly, save in the rarest cases, call Crabbe imaginative. In short, öáíôáóßáé here are vivid illustrations drawn from nature—Orestes' hallucination of the Eumenides, Euripides' picture of Phaethon, that in the Seven of the slaying of the bull over the black-bound shield, and many others. No doubt he glances at the fabulous and incredible, the actually "imagined"; but he seems, as in the case of the Odyssey, to be a little doubtful of these even in poetry, while in oratory he bars them altogether. You must at one and the same time reason and illustrate—again the very method of Burke.

In the rest of the illustrations of the use of Figures—for the central part of the treatise expressly disclaims being a formal discussion of these idols—the positive literary criticisms scattered in them—the actual "reviewing"—will give most of the interest. The great Oath of Demosthenes, "By those who fell at Marathon!" with its possible suggestion by a passage of Eupolis, supplies a whole chapter and part of another. And now we find the curious expression (showing how even Longinus was juggled by terms) that Figures "fight on the side of the Sublime, and in turn draw a wonderful reinforcement from it," wherein a mighty if vague reality like the Sublime, and mere shadows (though neatly cut-out shadows) like the Figures, are most quaintly yoked together.

Though still harassed by gaps, we find plenty of good pasture in the remarks, the handling of Periphrasis being especially attractive. For the eighteenth century—the time which honoured Longinus most in theory, and went against him most in practice—undoubtedly took part of his advice as to this figure. It had no doubt that Periphrasis contributed to the Sublime, was hypselopoion: unluckily it paid less attention to his subsequent caution, that it is a risky affair, and that it smells of triviality.¹² In fact, it is extremely noticeable that in the examples of Periphrasis which he praises we should hardly apply that name to it, but should call it "Allusion" or "Metaphor," while the examples that he condemns are actually of the character of Armstrong's "gelid cistern" and Delille's "game which Palamede invented."

At no time perhaps has the tricksy, if not (as one is almost driven to suspect) deliberately malignant, mutilator played such a trick as in abstracting four leaves from the MS. between caps. xxx. and xxxi. Here Longinus has begun to speak of diction generally; here he has made that admirable descant on "beautiful words" which, though almost all the book deserves to be written in letters of gold, would tempt one to indulge here in precious stones, so as to mimic, in jacinth and sapphire and chrysoprase, the effect which it celebrates. When we are permitted another glimpse we are back in particular criticism, interesting but less valuable save indirectly, and in criticisms, too, of Cacilius, criticisms which we could do without. No great good can ever come of inquiries, at least general inquiries, into the permissible number and the permissible strength of Metaphors. Once more we may fall back on the Master, though perhaps rather in opposition to some of the Master's dicta in this very field. "As the intelligent man shall decide" is the decision here, and the intelligent man will never decide till the case is before him. One bad metaphor is too much: twenty good ones are not too many. Nor is "the multitudinous seas incarnadine" an "excess," though no doubt there have been bad critics who thought so.

Longinus himself, though he had not had the happiness to read Macbeth, was clearly not far out of agreement with the concluding sentiment of the last paragraph, and he makes this certain by the disquisition on Faultlessness which follows. As a general question this is probably, for the present time at any rate, past argument, not so much because the possibility of a "faultless" great poem is denied, as because under the leaden rule of the best modern criticism—leaden not from dulness but from adaptability—few things are recognised as "faults" in se and per se. A pun may be a gross fault in one place and a grace beyond the reach of art in another: an aposiopesis may be either a proof of clumsy inequality to the situation or a stroke of genius. But the declaration of Longinus that he is not on the side of Faultlessness is of infinitely greater importance than any such declaration from an equally great critic ("Where is he? Show him to me," as Rabelais would say) could possess to-day. The general Greek theory undoubtedly did make for excessive severity to faultfulness, just as our general theory makes perhaps for undue leniency to it. That Longinus could withstand this tendency—could point out the faults of the faultless—was a very great thing.

As always, too, his individual remarks frequently give us, not merely the satisfaction of agreement, but that of piquant difference or curiosity. We may agree with him about Bacchylides and Pindar—though, by the way, the man who had the taste and the courage to admire a girl … as possessing that yellow ivory tint of skin which lights so magnificently¹³—was certainly one to dare to challenge convention with what its lilies-androses standard must have thought a "fault." But we cannot help astonishment at being told that both Pindar and Sophocles "often have their light quenched without any obvious reason, and stumble in the most unfortunate manner." For those of us who are less, as well as those who are more, enthusiastic about Sophocles would probably agree in asking, "Where does he 'go out in snuff,' where does he 'fall prostrate' in this fashion?" Surely all the faults cannot be in the lost plays! We want a rather fuller text of Hyperides than we possess to enable us quite to appreciate the justice of the comparison of him with Demosthenes, but that justice is striking even on what we have. On the other hand, we are rather thrown out by the contrast of Plato and Lysias—it may be owing to the same cause. Even if the comparison were one of style only, we should think it odd to make one between Burke and Berkeley, though the Sublime and Beautiful would help us a little here.

But all this is a digression,¹⁴ and the author seems to have returned to his Metaphors (in a gap where the demon has interfered with less malice than usual), and to Hyperboles, under the head of which we get a useful touch of contempt for Isocrates. We are in deeper and more living waters when we come to the handling, alas! too brief (though nothing seems here to be lost), of ordonnance, "composition," selection and arrangement of words. Here is yet another of those great law-making phrases which are the charter of a new criticism. "Harmony is to men not only physically connected with persuasion and pleasure, but a wonderful instrument of magniloquence and passion." It may be difficult for us, with our very slight knowledge (it would, perhaps, be wiser to say almost absolute ignorance) of Greek pronunciation, to appreciate his illustrations here in detail. But we can appreciate the principle of them exactly, and apply that principle, in any language of which we do know the pronunciation, with perfect ease and the completest success. The silly critics (they exist at the present day) who pooh-pooh, as niceties and fiddle-faddle, the order of words, the application of rhythmical tests to prose, and the like, are answered here beforehand with convincing force by a critic whom no one can possibly charge with preferring sound to sense.

This refers to prose, but the following chapter carries out the same principle as to poetry with equal acuteness. Longinus, great as his name is, probably is but little in the hands of those who object (sometimes almost with foam at the mouth) to the practice of analysing the mere harmonic effect of poetry. But it is pleasant to think of these passages when one reads the outcries, nor is the pleasantness rendered less pleasant by the subsequent cautions against that over-rhythmical fashion of writing which falls to the level of mere dance-music.

The caution against over-conciseness and over-prolixity is rather more of a matter of course, and the strictures on the microtis, occasionally to be found in Herodotus, like some in the earlier parts of the treatise, sometimes elude us, as is the case with similar verbal criticisms even in languages with which we are colloquially familiar.

And then there is the curious Conclusion which, as we have said, is no conclusion at all, as it would seem, and which yet has an unmistakable air of "peroration, with [much] circumstance," on the everlasting question, "Why is the Sublime so rare in our time?" In that day, as in this, we learn (the fact being, as in King Charles II.'s fish-experiment, taken for granted), divers explanations, chiefly political, were given for the fact. Democracy was a good nurse of greatness: aristocracy was not. But Longinus did not agree. It was money-getting and money-seeking, pleasure-loving and pleasure-hunting, he thought. Plain living and high thinking must be returned to if the Heights were to be once more scaled. A noble conclusion, if perhaps only a generous fallacy. Had Longinus had our illegitimate prerogative-postrogative of experience, he would have known that the blowing of the wind of the spirit admits of no such explanations as these. Ages of Liberty and Ages of Servitude, Ages of Luxury and Ages of Simplicity, Ages of Faith and Ages of Freethought—all give us the Sublime if the right man is there: none will give it us if he is not. But our critic had not the full premisses before him, and we could not expect the adequate conclusion.

Yet how great a book have we here! Of the partly otiose disputes about its date and origin and authorship one or two things are worth recalling, though for other purposes than those of the disputants. Let it be remembered that it is not quoted, or even referred to, by a single writer of antiquity.¹⁵ There is absolutely no evidence for it, except its own internal character, before the date of its oldest manuscript, which is assigned to the tenth century. Even if, assuming it to be the work of Longinus, we suppose it to have been part of one of the works which are ascribed to him (a possible assumption, see note), there is still the absence of quotation, still the absence even of reference to views so clearly formulated, so eloquently enforced, and in some ways so remarkably different from those of the usual Greek and Roman rhetorician. That the book can be of very late date—much later, that is to say, than that of Longinus himself—is almost impossible. One of its features, the lack of any reference to even a single writer later than the first century, has indeed been relied upon to prove that it is not later itself than that date. This is inconclusive for that purpose. But it makes every succeeding century less and less probable, while the style, though in some respects peculiar, is not in the least Byzantine.

This detachment from any particular age—nay, more, this vita fallens, this unrecognised existence of a book so remarkable—stands in no merely fanciful relation to the characteristics of the book itself. It abides alone in thought as well as in history. That it is a genuine, if a late, production of the classical or semi-classical age we cannot reasonably doubt, for a multitude of reasons, small in themselves but strong in a bundle,—its style, its diction, its limitations of material, and even occasionally of literary view, its standards, all sorts of little touches like the remark about Cicero, and so forth. Yet it has, in the most important points, almost more difference from than resemblance to the views of classical critics generally. The much greater antiquity of Aristotle may be thought to make comparison with him infructuous, if not unfair. But we have seen already how far Longinus is from Dionysius, how much further from Plutarch; and we shall see in the next Book how far he is from Quintilian. Let us look where we will, to critics by profession or to critics by chance, to the Alexandrians as far as we know them, to the professional writers on Rhetoric, to Aristophanes earlier and Lucian later, always we see Longinus apart—among them by dispensation and time, but not of them by tone, by tendency, by temper.

For though he himself was almost certainly unconscious of it, and might even have denied the fact with some warmth if it had been put to him, Longinus has marked out grounds of criticism very far from those of the ancient period generally, further still from those which were occupied by any critic (except Dante) of the Middle Ages and the Classical revival, and close to, if not in all cases overlapping the territory of, the modern Romantic criticism itself. As we have seen, the ancient critic was wont either to neglect the effect of a work of art altogether, and to judge it by its supposed agreement with certain antecedent requirements, or else, if effects were considered at all, to consider them from the merely practical point of view, as in the supposed persuasive effect of Rhetoric, or from the ethical, as in the purging, the elevating, and so forth, assigned to Tragedy, and to Poetry generally. Longinus has changed all this. It is the enjoyment, the transport, the carrying away of the reader or auditor, that, whether expressedly or not, is always at bottom the chief consideration with him. He has not lowered the ethical standard one jot, but he has silently refused to give it precedence of the aesthetic; he is in no way for lawlessness, but he makes it clear, again and again, that mere compliance with law, mere fulfilment of the requirements of the stop-watch and the hundredth-of-an-inch rule, will not suffice. Aristotle had been forced, equally by his system and his sense, to admit that pleasure was an end—perhaps the end—of art; but he blenches and swerves from the consequences. Longinus faces them and follows them out.

In his attention to rhythm, especially of prose, Longinus is much less unique, for this point (as we have seen and shall see) was never neglected by the best ancient critics. But there is again something particularly distinguishing in his attempt to trace the sources of the literary pleasure in specimen passages. The ancient tendency is, though not universally, yet too generally, the other way, to select specimen passages merely as illustrations of general rules.

And this brings us to his greatest claim of all—that is to say, his attitude towards his subject as a whole. Although he nowhere says as much in so many words, no one can read his book with attention—above all, no one can read it again and again critically—without seeing that to him literature was not a schedule of forms, departments, kinds, with candidates presenting themselves for the critic to admit them to one or the other, on and during their good behaviour; but a body of matter to be examined according to its fruits, according to its provision of the literary pleasure. When it has been examined it is still for the critic to explain and justify (according to those unwritten laws which govern him) his decision that this was good, this not so good, this bad,—to point out the reasons of success and failure, to arrange the symptoms, classify the methods, and so forth. Where Longinus fell short it was almost always because ancient literature had not provided him with enough material of certain kinds, not because he ruled these kinds out a priori. Longinus was no Rymer. We could submit even Shakespeare to him with very little fear, and be perfectly certain that he would not, with Rapin, pronounce Dantes Aligerus wanting in fire.¹⁶ Nay, with a sufficient body of material to set before him, we could trust him with very much more dangerous cases than Shakespeare and Dantes Aligerus.

Yet, as we have said, he stands alone. We must skip fifteen hundred years and come to Coleridge before we meet any critic entirely of his class, yet free from some of his limitations. The hand of the author of the Peri Hypsos is not subdued, but raised to what he deals in. And his work remains towering among all other work of the class, the work of a critic at once Promethean and Epimethean in his kind, learning by the mistakes of all that had gone before, and presaging, with instinctive genius, much that was not to come for centuries after.

Notes

¹ The most elaborate discussion of the whole matter still is that of Vaucher (Geneva, 1854). The editions I myself use are those of Toup (Oxford, 1778); Egger (Paris, 1837), a particularly handy little volume, with the fragments; and Prof. Rhys Roberts (Cambridge, 1899), with translation and full editorial apparatus. Those who do not read the Greek lose much: but they will find a good (though somewhat too free) translation, with an excellent introduction by Mr Andrew Lang, in the work of Mr H. L. Havell (London, 1890).

²Dionysiou e Longinous of the Paris MS. 2036. (Others even have íùíýìïõ.) Robortello intentionally or unintentionally dropped the e [or], thereby putting students off the scent.

³ Blair saw this, but, with the ill-luck of his century, regarded the work as merely "elegant."

⁴ Longinus (? 213-273) represents the middle of the third century. Nobody puts it later than this, and nobody earlier than the first.

⁵ A Sicilian rhetor, probably of Calacte, said by Suidas to have been of Greek, or at any rate non-Roman, birth, and a Jew in religion. Dionysius knew him, and he lived in the time of Augustus. There was another (confused by Suidas) in that of Hadrain. This may be our C.

⁶ A phrase of the rhetor Theodorus, meaning "the thyrsus poked in at the wrong time," "enthusiasm out of place."

⁷Logon krisis pollis esti peiras teleutalon epigenema. Dionysius (v. supra, pp. 130, 131) had said as much in sense, but less magisterially in phrase. I have translated logon in its narrowest equivalent, instead of "style" or "literature," which it doubtless also means, in order to bring out the antithesis better. I have small doubt that Longinus meant, here as elsewhere, to fling back the old contempt of the opposition of "words" and "things."

⁸ This word, which has the stamp of Dryden, is often preferable to "composition."

⁹ It may, however, be plausibly argued that the circle is more apparent than real, resulting from a kind of ambiguity in the word pegai. If Longinus had slightly altered his expression, so as to make it something of this kind, "There are five points [or ways, or aspects] in which hypsos may be attained, thought, feeling, 'figure,' diction, and composition," he would be much less vulnerable. And, after all, this is probably what he meant.

¹⁰ Literal.

¹¹ Fond and foolish fancy as it may be, there seems to me something miraculous in the mere juxtaposition of plesion and adu—the silent adoring lover, jealous, as it were, of the very air robbing him of a portion of the sweetness.

¹² … [This] means literally "perishable," "apt to go off," to get stale or flat.

¹³ Simonides had used the word literally of the nightingale, and there are those who hold that Bacchylides merely meant to compliment the lady's voice. But let us think more nobly of him.

¹⁴ I must be allowed to say that it contains one of the most ambitious and successful passages of Longinus as an original writer—the vindication of Nature's command to man to admire the magnificent—in cap. xxxv. It is a temptation to quote it.

¹⁵ "John of Sicily" (Walz, vi. 225), who in the thirteenth century cites the lost philologoi homiliai almost as if he was citing the Peri Hypsous, is certainly no exception. The undated Byzantine (Cramer, Aneed. Oxon., iii. 159, quoted by Professor Roberts after Usener), who couples logginou kriseis with those of Dionysius, may come nearer, as may the anonymous scholiast on Hermogenes (Walz, vii. 963), who cites the druiliai on to stomphōdes, "mouthing."

¹⁶ Sir Thomas Pope Blount, Characters and Censures of the most Considerable Poets. London, 1694. P. 58. "Rapin tells us that Dantes Aligerus wants fire, and that he has not heat enough."