Agamben - END OF THE POEM Essays in Poetics.doc

THE END OF THE POEM
Studies in Poetics

Giorgio Agamben

The End of the Poem: Studies in Poetics was originally published in Italian in 1996 under the title Categorie italiane: Studi di poetica © 1996 by Marsilio Editori for the Italian edition

Stanford University Press Stanford, California

Acknowledgments

"Comedy" first appeared in Paragone 347 ( 1978). "Corn: From Anatomy to Poetics" was published in Le Moyen Âge dans la modernité, Mélanges offerts à Roger Dragonetti, ed. Jean R. Scheidegger, Sabine Girardet, and Eric Hicks ( Geneva: Champion, 1996). "The Dream of Language," originally written for the Fondazione Cini conference "Languages of Dreaming," appeared in Lettere italiane 4 ( 1982). "Pascoli and the Thought of the Voice" was published as a preface to Giovanni Pascoli, Il fanciullino ( Milan: Feltrinelli, 1982). "The Dictation of Poetry" appeared as a preface to Antonio Delfini's Poesie della fine del mondo (Macerata: Quodlibet, 1995). "Expropriated Manner" was published as a preface to Giorgio Caproni , Res amissa ( Milan: Garzanti, 1991). "The Celebration of the Hidden Treasure" was presented at a conference on Elsa Morante in Perugia in January 1993. "The End of the Poem" was presented November 10, 1995, at the University of Geneva during a conference honoring Roger Dragonetti. "An Enigma Concerning the Basque Woman" appeared in Marka 27 ( 1990). "The Hunt for Language" was published in Il Manifesto, January 23, 1990. "The Just Do Not Feed on Light" appeared in Idra 5 ( 1992) as an introduction to Eugenio De Signoribus poems. "Taking Leave of Tragedy" was published in Fine secolo, December 7, 1985.

G.A.

Contents

Preface

Between 1974 and 1976 I met regularly in Paris with Italo Calvino and Claudio Rugafiori to define the program of a review. The project was ambitious, and our conversations, which often were not entirely focused, followed the dominant motifs and muffled echoes of each of our interests. We were, however, in agreement about one thing: one section of the review was to be dedicated to the definition of what we called "Italian categories." It was a matter of identifying nothing less than the categorial structures of Italian culture through a series of conjoined polar concepts. Claudio immediately suggested architecture/vagueness (that is, the domination of the mathematical-architectonic order alongside the perception of beauty as something vague). Italo had already been ordering images and themes along the coordinates of speed/lightness. Working on the essay on the title of the Divine Comedy that opens this collection, I proposed that we explore several oppositions: tragedy/comedy, law/ creature, biography/fable.

For reasons that need not be clarified here, the project was never realized. Once we had returned to Italy, we all--if in different ways--confronted the political change that was already under way and that was to impress the 1980s with its dark seal. It was obviously a time not for programmatic definitions, but for resistance and flight. Echoes of our common project can be found in Italo American Lectures, as well as in a large notebook that has remained among his papers. For my part, I attempted to establish the physiognomy of the project, before it was definitively canceled, in the "program for a review" published in limine in Infancy and History. (Those who are interested may look in those pages for the provisional list of categories in their original, problematic context.)

In their own way, the eight studies collected here (the first of which dates from the time of the project, the last of which was finished in 1995) remain faithful to this program. In the course of time, other categories came to be added to those rudimentary first ones (mother tongue / grammatical language; living language / dead language; style / manner). At the same time, the project of a definition of these categories gradually gave way to a study of the general problems in poetics that they implied. Each of the essays in this book thus seeks to define a general problem of poetics with respect to an exemplary case in the history of literature. The inquiry into the reasons for the title of Dante Comedy makes it possible to cast new light on the comedy/tragedy opposition at the beginning of Romance poetry; a reading of Hypnerotomachia Polifili and Pascoli considers the problem of the relation between living language and dead language as a fundamental internal tension in the poetics of modernity; the introduction to the poetic work of a contemporary Italian writer, Antonio Delfini, functions as an occasion to reformulate the old problem of the relation between life and poetry and to define the principle of narrative in Romance literatures as an invention of lived experience on the basis of poetry; and, finally, an analysis of one of the greatest poets of the twentieth century, Giorgio Caproni, defines the act of writing with respect to the dialectical tension of style and manner.

In "Corn: From Anatomy to Poetics" and "The End of the Poem," the subject of study shifts to the problem of the specific structure of the poem itself. These two essays are thus to be understood as a first contribution to a philosophy and criticism of meter that do not yet exist. The first of these essays, which examines Arnaut Daniel's obscene sirventes, develops Roman Jakobson's problem of the relation between sound and sense; the second, which lends its title to the book as a whole, considers the end of the poem as a point of crisis that is in every sense fundamental to the structure of poetry.

The initial program of a systematic grid of the categories bearing on Italian culture nevertheless remains unfinished, and this book merely offers a torso of the idea of which we once tried to catch sight. It is therefore dedicated to the memory not only of companionship, but also of the one among us who is no longer present to bear witness to it.

THE END OF THE POEM

§ 1 Comedy

1. THE PROBLEM

1. The aim of this essay is the critical assessment of an event that can be chronologically dated at the beginning of the fourteenth century but that, by virtue of its still exerting a profound influence on Italian culture, can be said to have never ceased to take place. This event is the decision of a poet to abandon his own "tragic" poetic project for a "comic" poem. This decision translates into an extremely famous incipit, which one of the author's letters states as follows: "Here begins the Comedy of Dante Alighieri, a Florentine by birth, not by disposition" ( Incipit comoedia Dantis Alagherii florentini natione non moribus). The turn registered by these words is so little a question internal to Dante scholarship that it can even be said that here, for the first time, we find one of the traits that most tenaciously characterizes Italian culture: its essential pertinence to the comic sphere and consequent refutation of tragedy.

The fact that even a few years after the author's death the reasons for the comic title appeared problematic and incoherent to the oldest commentators 1 bears witness to the extent to which this turn hides a historical knot whose repression cannot easily be brought to consciousness. All the more surprising is the poverty of modern critical literature on the subject. That a scholar such as Pio Rajna (who so influenced later studies) could reach such obviously insufficient conclusions as those with which his study of the poem's title ends 2 is something that cannot be explained even by Italian culture's lack of contact with its own origins. Even Erich Auerbach, the author of such penetrating works on Dante's style, does not succeed in explaining the poem's incipit in satisfying terms. "Dante," he writes, referring to the ancient theory of the separation of styles, "never freed himself completely from these views; otherwise he would not have called his great work a comedy in clearest opposition to the term alta tragedia which he applied to Virgil Aeneid." 3 And, concerning Dante's letter to Cangrande, Auerbach writes:

It is not easy to see how Dante, after having found this formula and after having completed the Comedy, could still have expressed himself upon its character with the pedantry exhibited in the passage from the letter to Cangrande just referred to. However, so great was the prestige of the classical tradition, obscured as it still was by pedantic schematization, so strong was the predilection for fixed theoretical classifications of a kind which we can but consider absurd, that such a possibility cannot be gainsaid after all. 4

As far as explanations for Dante's choice of title are concerned, in a certain sense modern criticism has not progressed beyond Benvenuto da Imola's observations or the suggestions with which Boccaccio ends his commentary on the title of the poem. "What," Boccaccio asks,

will we then say of the objections that have been made against it? On the grounds that the author was a most prudent man, I believe that he would have had in mind not the parts contained in comedy but its entirety, and that he named his book on the basis of this entirety, so to speak. And from what one can infer from Plautus and Terence, who were comic poets, the entirety of comedy is this: comedy has a turbulent principle, is full of noise and discord, and ends finally in peace and tranquillity. The present book altogether conforms to this model. Thus the author begins with woes and infernal troubles and ends it in the peace and glory enjoyed by the blessed in their eternal life. And this certainly suffices to explain how the said title suits this book. 5

The methodological principle that we follow in this study is that our ignorance of an author's motivations in no way authorizes the presumption that they are incoherent or faulty. We hold that until proven otherwise, Dante, as "a most prudent man" (oculatissimo uomo), could not have chosen his incipit lightly or superficially. On the contrary, precisely the fact that the comic title appears discordant with respect to what we know of the ideas of the poet and his age brings us to claim that it was carefully considered.

2. A precise study of the passages in which Dante speaks of comedy and tragedy demonstrates that this claim is textually founded.

We thus know that to Dante's eyes, the poetic project that gave birth to the great songs of the Rime seemed eminently tragic. In De vulgari eloquentia, he explicitly states that the tragic style is the highest of all styles and the only one appropriate to the ultimate objects of poetry: "well-being, love and virtue" (salus, amor et virtus). 6 A little later he defines the song [canzone], the supreme poetic genre, as

a connected series of equal stanzas in the tragic style, without a refrain, and focused on a single theme, as I have shown when I wrote "Donne che avete intelletto d'amore." If I say "a connected series in the tragic style," it is because, were the style of the stanza comic, we would use the diminutive and call it a canzonetta.

(iaequalium stantiarum sine responsorio tragica coniugatio, ut nos ostendimus cum dicimus "Donne che avete intelletto d'amore." Quod autem dicimus tragica coniugatio, est quia cum cornice fiat hec coniugatio, cantilenam vocamus per diminutionem.) 7

The poem's comic title therefore above all implies a rupture and a turn with respect to Dante's own past and poetic itinerary, a genuine "categorical revolution" that as such cannot have been decided upon without conscious and vital motivation. In a passage of the letter to Cangrande, Dante seems implicitly to affirm such an awareness of reasons for his choice. With a definition that formally repeats commonplaces of medieval lexicography, 8 Dante here introduces a discussion that cannot be found in any of his known sources. "Now comedy is a certain kind of poetical narration," he writes, "which differs from all others" (Et est comoedia genus quoddam poëtice narrationis ab omnibus aliis differens). 9 This privileged situation of the comic genre, which has no counterpart in either medieval or late ancient sources, presupposes an intention on the poet's part to alter semantically the term "comedy" in a sense that certainly goes beyond what modern criticism believes itself to have ascertained.

From this perspective, the fact that in the Inferno Dante explicitly defines the Aeneid as "high tragedy" 10 is every bit as significant as the fact that he titles his own "sacred poem" a comedy. This is so not only because he thus comes to oppose the Comedy to the work of the poet from whom he considers himself to draw "the beautiful style that has done me honor" (lo bel stile che mi ha fatto onore), but also because the definition of the Aeneid as a tragedy cannot be coherently reconciled with the criteria of the "peaceful beginning" and "foul end" indicated in the letter to Cangrande.

In an attempt to use one half of the problem as an explanation for the other half, it has been said that to Dante's eyes, the Aeneid, as a poetic narration in the high style, could only be a tragedy. In fact, according to a tradition that has its origin in Diomedes and that is still alive in Isidore of Seville, 11 the Aeneid figures in medieval treatises as an example not of tragedy as much as of that genre of poetic narration that was defined as genus commune on account of presenting the speech of both characters and the author. It is curious that, as has been occasionally noted, in medieval treatises the classification of the three styles--whose prototype is to be found in the Rhetorica ad Herennium 12 --does not necessarily coincide with that of the genres of poetic narration. Comedy and tragedy, which never entirely lost their dramatic connotation, were commonly listed alongside satire and mime in the genus activam or dramaticon (in which only characters speak, without the intervention of the author). The enumeration of styles, moreover, always involved a reference at least to the elegy, 13 and could never be exhausted in the comedy/tragedy opposition. The radicality with which the letter to Cangrande transforms this double classification into a tragedy/comedy antinomy--an antinomy that is at once stylistic and substantial, and with respect to which other poetic genres are quickly set aside 14 --is in itself a sufficient sign of a strong, conscious sense of these two terms.

From this perspective, the eclogue to Giovanni di Virgilio constitutes another piece of evidence. Here Dante alludes to his own poem with the expression comica verba. 15 The interpretation of this passage has been led astray by one of Boccaccio's glosses, which explains that "comica, id est vulgaria." The influence of this gloss has been so tenacious that even in the recent Enciclopedia dantesca one reads that, in the first eclogue, Dante resolutely identified "the comic in the vernacular." A text that could have shed light on Dante's choice of title thus became irrelevant, since the identification between comic style and the Italian language is clearly untenable. 16 An attentive reading of Giovanni's verse epistle demonstrates that the reproaches made to Dante by the Bolognese humanist have as their object not simply the use of the vernacular as opposed to Latin but rather the choice of comedy as opposed to tragedy. The expression with which Giovanni characterizes Dante's writing, sermone forensi, does not allude to the vernacular but rather corresponds to the sermone pedestri of the passage in Horace cited by Dante in his letter to Cangrande, as well as to the cotidiano sermone of medieval poetics. 17 Sermone forensi, in other words, refers to a choice of style and not language. This interpretation is confirmed by a further passage in the letter in which Giovanni, specifying his objections, encourages Dante to sing in "prophetic verse" the great facts of the history of his age, that is, the heroic and "public" material of tragedy instead of the "private" matters of comedy.

At the center of the debate with Giovanni di Virgilio, which belongs to the cultural circle from which the first modern tragedy, Mussato tragoedia Ecerinis, was to be born, is not as much the Latin/vernacular opposition as the tragedy/comedy one. This testifies once again to the fact that for Dante, the comic title of his poem is neither contingent nor fragmentary, but rather constitutes the affirmation of a principle.

3. If this is true, then it is all the more dispiriting that the title of the Comedy is not compatible with the set of definitions given by Dante for the tragic/comic opposition, and that these definitions cannot, moreover, be reduced to a unitary system.

As has been noted, these definitions are articulated on two planes: a stylistic-formal one (the modus loquendi), and a materialsubstantial one (the materia or sententia). In De vulgari eloquentia (in which the stylistic aspect is prevalent and whose incompleteness is such that this work gives us no genuine thematic treatment of comedy), the tragic style is defined, according to the principles of the classical tripartition of styles, as the most elevated style (superiorem stilum), in harmony with the height of the material reserved for it (the three great magnalia: salus, amore, and virtus). In the letter to Cangrande, in which the material articulation is prevalent, the tragic/comic opposition is instead characterized on the plane of content and as an opposition of beginning and end: tragedy is marked by an "admirable" and "peaceful" beginning and a "foul" and "horrible" end; comedy by a "horrible" and "foul" beginning and a "prosperous" and "pleasant" end. On the stylistic plane, the tragic/comic opposition is presented as an opposition between what is, in one case, an elevated and sublime modus loquendi and, in the other, a "lowly" and "humble" modus loquendi (tempered, however, by a reference to Horace, who licentiat liquando comicos ut tragicos loqui).

Even a superficial examination of these categories demonstrates that according to the criteria of De vulgari eloquentia, the Comedy cannot justify its title without contradiction, though the Aeneid probably can be coherently defined as a tragedy. According to the criteria of the letter to Cangrande, by contrast, while the tragic justification of the Aeneid appears unfounded, the Comedy sufficiently justifies its title. The only thing that can in fact be affirmed with certainty is that in De v...