M. Paul ValÉry, a writer for whom I have considerable respect, has placed in his most recent statement upon poetry a paragraph which seems to me of very doubtful validity. I have not seen the complete essay, and know the quotation only as it appears in a critical notice in the AthenÆum, July 23, 1920:

La philosophie, et mÊme la morale tendirent À fuir les oeuvres pour se placer dans les rÉflexions qui les prÉcÈdent.... Parler aujourd’hui de poÉsie philosophique (fÛt-ce en invoquant Alfred de Vigny, Leconte de Lisle, et quelques autres), c’est naivement confondre des conditions et des applications de l’esprit incompatibles entre elles. N’est-ce pas oublier que le but de celui qui spÉcule est de fixer ou de crÉer une notion—c’est-À-dire un pouvoir et un instrument de pouvoir, cependant que le poÈte moderne essaie de produire en nous un État et de porter cet État exceptionnel au point d’une jouissance parfaite....

It may be that I do M. ValÉry an injustice which I must endeavour to repair when I have the pleasure of reading his article entire. But the paragraph gives the impression of more than one error of analysis. In the first place, it suggests that conditions have changed, that “philosophical” poetry may once have been permissible, but that (perhaps owing to the greater specialization of the modern world) it is now intolerable. We are forced to assume that what we do not like in our time was never good art, and that what appears to us good was always so. If any ancient “philosophical” poetry retains its value, a value which we fail to find in modern poetry of the same type, we investigate on the assumption that we shall find some difference to which the mere difference of date is irrelevant. But if it be maintained that the older poetry has a “philosophic” element and a “poetic” element which can be isolated, we have two tasks to perform. We must show first in a particular case—our case is Dante—that the philosophy is essential to the structure and that the structure is essential to the poetic beauty of the parts; and we must show that the philosophy is employed in a different form from that which it takes in admittedly unsuccessful philosophical poems. And if M. ValÉry is in error in his complete exorcism of “philosophy,” perhaps the basis of the error is his apparently commendatory interpretation of the effort of the modern poet, namely, that the latter endeavours “to produce in us a state.”

The early philosophical poets, Parmenides and Empedocles, were apparently persons of an impure philosophical inspiration. Neither their predecessors nor their successors expressed themselves in verse; Parmenides and Empedocles were persons who mingled with genuine philosophical ability a good deal of the emotion of the founder of a second-rate religious system. They were not interested exclusively in philosophy, or religion, or poetry, but in something which was a mixture of all three; hence their reputation as poets is low and as philosophers should be considerably below Heraclitus, Zeno, Anaxagoras, or Democritus. The poem of Lucretius is quite a different matter. For Lucretius was undoubtedly a poet. He endeavours to expound a philosophical system, but with a different motive from Parmenides or Empedocles, for this system is already in existence; he is really endeavouring to find the concrete poetic equivalent for this system—to find its complete equivalent in vision. Only, as he is an innovator in this art, he wavers between philosophical poetry and philosophy. So we find passages such as:

But the velocity of thunderbolts is great and their stroke powerful, and they run through their course with a rapid descent, because the force when aroused first in all cases collects itself in the clouds and.... Let us now sing what causes the motion of the stars.... Of all these different smells then which strike the nostrils one may reach to a much greater distance than another....^[10]

But Lucretius’ true tendency is to express an ordered vision of the life of man, with great vigour of real poetic image and often acute observation.

The philosophy which Lucretius tackled was not rich enough in variety of feeling, applied itself to life too uniformly, to supply the material for a wholly successful poem. It was incapable of complete expansion into pure vision. But I must ask M. ValÉry whether the “aim” of Lucretius’ poem was “to fix or create a notion” or to fashion “an instrument of power.”

Without doubt, the effort of the philosopher proper, the man who is trying to deal with ideas in themselves, and the effort of the poet, who may be trying to realize ideas, cannot be carried on at the same time. But this is not to deny that poetry can be in some sense philosophic. The poet can deal with philosophic ideas, not as matter for argument, but as matter for inspection. The original form of a philosophy cannot be poetic. But poetry can be penetrated by a philosophic idea, it can deal with this idea when it has reached the point of immediate acceptance, when it has become almost a physical modification. If we divorced poetry and philosophy altogether, we should bring a serious impeachment, not only against Dante, but against most of Dante’s contemporaries.

Dante had the benefit of a mythology and a theology which had undergone a more complete absorption into life than those of Lucretius. It is curious that not only Dante’s detractors, like the Petrarch of Landor’s Pentameron (if we may apply so strong a word to so amiable a character), but some of his admirers, insist on the separation of Dante’s “poetry” and Dante’s “teaching.” Sometimes the philosophy is confused with the allegory. The philosophy is an ingredient, it is a part of Dante’s world just as it is a part of life; the allegory is the scaffold on which the poem is built. An American writer of a little primer of Dante, Mr. Henry Dwight Sidgwick, who desires to improve our understanding of Dante as a “spiritual leader,” says:

To Dante this literal Hell was a secondary matter; so it is to us. He and we are concerned with the allegory. That allegory is simple. Hell is the absence of God.... If the reader begins with the consciousness that he is reading about sin, spiritually understood, he never loses the thread, he is never at a loss, never slips back into the literal signification.

Without stopping to question Mr. Sidgwick on the difference between literal and spiritual sin, we may affirm that his remarks are misleading. Undoubtedly the allegory is to be taken seriously, and certainly the Comedy is in some way a “moral education.” The question is to find a formula for the correspondence between the former and the latter, to decide whether the moral value corresponds directly to the allegory. We can easily ascertain what importance Dante assigned to allegorical method. In the Convivio we are seriously informed that

the principal design [of the odes] is to lead men to knowledge and virtue, as will be seen in the progress of the truth of them;

and we are also given the familiar four interpretations of an ode: literal, allegorical, moral, and anagogical. And so distinguished a scholar as M. Hauvette repeats again and again the phrase “didactique d’intention.” We accept the allegory. Accepted, there are two usual ways of dealing with it. One may, with Mr. Sidgwick, dwell upon its significance for the seeker of “spiritual light,” or one may, with Landor, deplore the spiritual mechanics and find the poet only in passages where he frees himself from his divine purposes. With neither of these points of view can we concur. Mr. Sidgwick magnifies the “preacher and prophet,” and presents Dante as a superior Isaiah or Carlyle; Landor reserves the poet, reprehends the scheme, and denounces the politics. Some of Landor’s errors are more palpable than Mr. Sidgwick’s. He errs, in the first place, in judging Dante by the standards of classical epic. Whatever the Comedy is, an epic it is not. M. Hauvette well says:

Rechercher dans quelle mesure le poÈme se rapproche du genre classique de l’ÉpopÉe, et dans quelle mesure il s’en Écarte, est un exÉrcice de rhÉtorique entiÈrement inutile, puisque Dante, À n’en pas douter, n’a jamais eu l’intention de composer une action Épique dans les rÈgles.

But we must define the framework of Dante’s poem from the result as well as from the intention. The poem has not only a framework, but a form; and even if the framework be allegorical, the form may be something else. The examination of any episode in the Comedy ought to show that not merely the allegorical interpretation or the didactic intention, but the emotional significance itself, cannot be isolated from the rest of the poem. Landor appears, for instance, to have misunderstood such a passage as the Paolo and Francesca, by failing to perceive its relations:

In the midst of her punishment, Francesca, when she comes to the tenderest part of her story, tells it with complacency and delight.

This is surely a false simplification. To have lost all recollected delight would have been, for Francesca, either loss of humanity or relief from damnation. The ecstasy, with the present thrill at the remembrance of it, is a part of the torture. Francesca is neither stupefied nor reformed; she is merely damned; and it is a part of damnation to experience desires that we can no longer gratify. For in Dante’s Hell souls are not deadened, as they mostly are in life; they are actually in the greatest torment of which each is capable.

It is curious that Mr. Sidgwick, whose approbation is at the opposite pole from Landor’s, should have fallen into a similar error. He says:

In meeting [Ulysses], as in meeting Pier della Vigna and Brunetto Latini, the preacher and the prophet are lost in the poet.

Here, again, is a false simplification. These passages have no digressive beauty. The case of Brunetto is parallel to that of Francesca. The emotion of the passage resides in Brunetto’s excellence in damnation—so admirable a soul, and so perverse.

And I think that if Mr. Sidgwick had pondered the strange words of Ulysses,

he would not have said that the preacher and prophet are lost in the poet. “Preacher” and “prophet” are odious terms; but what Mr. Sidgwick designates by them is something which is certainly not “lost in the poet,” but is part of the poet.

A variety of passages might illustrate the assertion that no emotion is contemplated by Dante purely in and for itself. The emotion of the person, or the emotion with which our attitude appropriately invests the person, is never lost or diminished, is always preserved entire, but is modified by the position assigned to the person in the eternal scheme, is coloured by the atmosphere of that person’s residence in one of the three worlds. About none of Dante’s characters is there that ambiguity which affects Milton’s Lucifer. The damned preserve any degree of beauty or grandeur that ever rightly pertained to them, and this intensifies and also justifies their damnation. As Jason

The crime of Bertrand becomes more lurid; the vindictive Adamo acquires greater ferocity, and the errors of Arnaut are corrected—

If the artistic emotion presented by any episode of the Comedy is dependent upon the whole, we may proceed to inquire what the whole scheme is. The usefulness of allegory and astronomy is obvious. A mechanical framework, in a poem of so vast an ambit, was a necessity. As the centre of gravity of emotions is more remote from a single human action, or a system of purely human actions, than in drama or epic, so the framework has to be more artificial and apparently more mechanical. It is not essential that the allegory or the almost unintelligible astronomy should be understood—only that its presence should be justified. The emotional structure within this scaffold is what must be understood—the structure made possible by the scaffold. This structure is an ordered scale of human emotions. Not, necessarily, all human emotions; and in any case all the emotions are limited, and also extended in significance by their place in the scheme.

But Dante’s is the most comprehensive, and the most ordered presentation of emotions that has ever been made. Dante’s method of dealing with any emotion may be contrasted, not so appositely with that of other “epic” poets as with that of Shakespeare. Shakespeare takes a character apparently controlled by a simple emotion, and analyses the character and the emotion itself. The emotion is split up into constituents—and perhaps destroyed in the process. The mind of Shakespeare was one of the most critical that has ever existed. Dante, on the other hand, does not analyse the emotion so much as he exhibits its relation to other emotions. You cannot, that is, understand the Inferno without the Purgatorio and the Paradiso. “Dante,” says Landor’s Petrarch, “is the great master of the disgusting.” That is true, though Sophocles at least once approaches him. But a disgust like Dante’s is no hypertrophy of a single reaction: it is completed and explained only by the last canto of the Paradiso.

The contemplation of the horrid or sordid or disgusting, by an artist, is the necessary and negative aspect of the impulse toward the pursuit of beauty. But not all succeed as did Dante in expressing the complete scale from negative to positive. The negative is the more importunate.

The structure of emotions, for which the allegory is the necessary scaffold, is complete from the most sensuous to the most intellectual and the most spiritual. Dante gives a concrete presentation of the most elusive:

or

Again, in the Purgatorio, for instance in Canto XVI and Canto XVIII, occur passages of pure exposition of philosophy, the philosophy of Aristotle strained through the schools.

We are not here studying the philosophy, we see it, as part of the ordered world. The aim of the poet is to state a vision, and no vision of life can be complete which does not include the articulate formulation of life which human minds make.

It is one of the greatest merits of Dante’s poem that the vision is so nearly complete; it is evidence of this greatness that the significance of any single passage, of any of the passages that are selected as “poetry,” is incomplete unless we ourselves apprehend the whole.

And Dante helps us to provide a criticism of M. ValÉry’s “modern poet” who attempts “to produce in us a state.” A state, in itself, is nothing whatever.

M. ValÉry’s account is quite in harmony with pragmatic doctrine, and with the tendencies of such a work as William James’s Varieties of Religious Experience. The mystical experience is supposed to be valuable because it is a pleasant state of unique intensity. But the true mystic is not satisfied merely by feeling, he must pretend at least that he sees, and the absorption into the divine is only the necessary, if paradoxical, limit of this contemplation. The poet does not aim to excite—that is not even a test of his success—but to set something down; the state of the reader is merely that reader’s particular mode of perceiving what the poet has caught in words. Dante, more than any other poet, has succeeded in dealing with his philosophy, not as a theory (in the modern and not the Greek sense of that word) or as his own comment or reflection, but in terms of something perceived. When most of our modern poets confine themselves to what they had perceived, they produce for us, usually, only odds and ends of still life and stage properties; but that does not imply so much that the method of Dante is obsolete, as that our vision is perhaps comparatively restricted.

Note.—My friend the AbbÉ Laban has reproached me for attributing to Landor, in this essay, sentiments which are merely the expression of his dramatic figure Petrarch, and which imply rather Landor’s reproof of the limitations of the historical Petrarch’s view of Dante, than the view of Landor himself. The reader should therefore observe this correction of my use of Landor’s honoured name.