Homeric Effects in Vergil's Narrative: Updated Edition [Updated edition with a New Foreword and New Afterword] 9781400852482

Citation preview

HOMERIC EFFECTS IN VERGIL’S NARRATIVE

HOMERIC EFFECTS IN VERGIL’S NARRATIVE ALESSANDRO BARCHIESI T r a n s l at ed b y ILARIA MARCHESI & MATT FOX W it h a n ew for ewor d b y PHILIP HARDIE and a new afterword by the author

PRINCETON UNIVERSITY PRESS Princeton & Oxford

Copyright © 2015 by Princeton University Press Published by Princeton University Press 41 William Street, Princeton, New Jersey 08540 In the United Kingdom: Princeton University Press 6 Oxford Street, Woodstock, Oxfordshire OX20 1TW Originally published in Italian as La traccia del modello © 1984 by Giardini editori e stampatori in Pisa press.princeton.edu Jacket photograph: detail of a silver cup from Hoby's Tomb © Jens Vermeersch. Cropped from original. Licensed under Creative Commons 2.0 license (https://creativecommons.org/licenses/by-nc-nd/2.0/) All Rights Reserved Library of Congress Cataloging-in-Publication Data Barchiesi, Alessandro, author. [Traccia del modello. English] Homeric effects in Vergil's narrative / Alessandro Barchiesi ; translated by Ilaria Marchesi and Matt Fox with a new foreword by Philip Hardie and a new afterword by the author. pages cm Includes bibliographical references and index. ISBN 978-0-691-16181-5 (hardcover) 1. Virgil. Aeneis.  2. Epic poetry, Latin—History and criticism.  3. Narration (Rhetoric)—History—To 1500.  4. Latin poetry—Greek influences.  5. Homer— Appreciation—Rome.  6. Imitation in literature.  7. Rome—In literature. 8. Homer—Influence.  9. Rhetoric, Ancient. PA6931.B3413 2015 873'.01—dc23 2014012449 British Library Cataloging-­in-­Publication Data is available This book has been composed in Garamond Premier Pro and League Gothic Printed on acid-­free paper. ∞ Printed in the United States of America 1 3 5 7 9 10 8 6 4 2

CONTENTS FOREWORD by Ph i l i p H a r di e vii INTRODUCTORY NOTE xv 1 T h e De at h of Pa l l a s Intertextuality and Transformation of the Epic Model 1 2 T h e St ruc t u r e of A e n e i d 10 35 3 T h e A r ms i n t h e Sk y Diffraction of a Narrative Theme 53 4 T h e De at h of T u r n us Genre Model and Example Model 69 V

APPENDIX T h e L a m e n t of J u t u r na 95 AFTERWORD by A l e ss a n dro Ba rch i e si 115 NOTES 135 WORKS CITED 175 SELECT INDEX 185 SELECT INDEX LOCORUM 188 INDEX OF MODERN AUTHORS 190

VI

FOREWORD Alessandro Barchiesi began La traccia del modello. Effetti omerici nella narrazione virgiliana by noting that twenty years had passed since the publication of Georg Knauer’s monumental study of Vergil’s use of Homer in the Aeneid, Die Aeneis und Homer. Studien zur poetischen Technik Vergils (1964), a revision of Knauer’s 1961 Habilitationsschrift. It is now (2014) thirty years since the Italian publication of Homeric Effects in Virgil’s Narrative. It is astonishing to think that La traccia was based substantially on its author’s honors thesis, the work of a brilliant scholar in his early twenties. Knauer’s book, building on centuries of Vergilian commentary and drawing on the riches of German Vergilian scholarship of the previous seventy years, was the first full and integrated study both of the complex architecture of Vergil’s imitation of Homer and of the filigree detail of Vergil’s allusion to particular episodes, lines, and phrases of the Homeric epics. Fifty years on, Die Aeneis und Homer remains the starting point for any study of Vergil and Homer. La traccia was the product of a watershed in the history of Latin literary studies in the 1970s and early 1980s, a period that has largely determined the way Latinists have been doing things down to the present day. The young Barchiesi was at the very forefront of this nouvelle vague; the book is already completely at home with a number of terms that have defined major approaches to the study of ancient literature over the last three decades but that in the early 1980s were only beginning to make their way into the consciousness of Anglophone Latinists: “narratology,” “intertextuality,” “reception.” VII

VIII

F oreword

The book also makes a significant contribution to another highly productive area of Latin studies, the dialogue of genres, or generic polyphony, with its perceptive analyses of Vergil’s refraction of Homeric epic models through the lens of the outlooks and structures of Attic tragedy. Barchiesi’s early intellectual formation was in continental and Russian literary theory, but the major achievement of La traccia is to be located in the area of what has been perhaps the most influential Italian contribution to modern Latin literary scholarship, the study of allusion and intertextuality. Prior to La traccia the major landmarks are Giorgio Pasquali’s 1951 essay “L’arte allusiva,” and the essays by one of Barchiesi’s own teachers at Pisa, Gian Biagio Conte, collected in Memoria dei poeti e sistema letterario: Catullo, Virgilio, Ovidio, a book published in 1974 but whose full impact outside Italy had to await its English translation as the first part of G. B. Conte, The Rhetoric of Imitation: Genre and Poetic Memory in Virgil and Other Latin Poets (1986). La traccia advances the discussion in a number of ways. The Italian title puns in a manner not easily conveyed in English. Traccia can mean “trace,” “trail,” or “track,” holding together two ways of thinking about the relationship between a model text (in this case, the Homeric epics) and an alluding text (the Aeneid): a reader may either focus on fragmentary traces of the model text subsumed within the structures of the alluding text or look for systematic Homeric trails inscribed in the Vergilian text.1 Barchiesi has a particular interest in a systematic approach to Vergil’s Homeric allusion, which he contrasts with Pasquali’s attention to the “microcontext of poetic memory” (chapter 4). La traccia draws out the importance of the Homeric model for reading the plot of the Aeneid, demonstrating the vital contribution of an awareness of Homeric intertexts to the “legibility” of the Aeneid. Vergil’s reader is called upon to be attentive both to the incorporation of Homeric stereotypes within the Aeneid (for example, patterns of heroic behavior on the battlefield) and to Vergil’s swerves from these stereotypes, examples of the oppositio in imitando (“opposition in imitation”) that earlier students of allusivity had tended to examine at the level of the small-­scale detail. The way that meaning is thus generated through divergences within a horizon of expectations may also be compared to Francis Cairns’s

F oreword

IX

use of the schemata of situational genres as a starting point for reading ancient poetry, in his Generic Composition in Greek and Roman Poetry (1972), to name an important work in the Anglophone tradition that was developing independently of continental work in the late 1960s and early 1970s. Barchiesi develops his approach through sustained and multilayered readings of a limited number of passages in books 8, 10, and 12 of the Aeneid, together with a revelatory quasi-­structuralist analysis of the whole sequence of encounters on the battlefield in Aeneid 10. In the several chapters he deploys the full armory of philological and literary-­historical weaponry in the service of a powerful intertextual method. Not the least important service rendered by the book is the demonstration of the inseparability of, on the one hand, formalist readings, and, on the other, cultural-­historical and ideological readings (for programmatic statements of principle on the need to combine the formalist and the historical, see chapter 4). Together with Gian Biagio Conte, the second of Barchiesi’s thesis supervisors was Antonio La Penna, the leading Italian Latinist in a marxisant tradition of historical and political readings of poetry. In this respect the book achieves an unusually satisfying coincidentia oppositorum, steering a course between the Scylla of an inward-­looking formalism that is the fate of some intertextual and generic studies and the Charybdis of an exclusive set toward the “culturalism” or “ideology critique” against which Charles Martindale (2005) has taken up cudgels. La traccia generates complex, but never confusing, readings of the Aeneid through two strategies: first the demonstration of the presence within a single Vergilian episode of separate narrative strands in the Homeric text (akin to what Knauer labels “Kontaminierung,” and to what I have labeled “combinatorial imitation,” with reference to post-­Vergilian epic imitations of the Aeneid);2 and second, through an excavation of the different stratifications in the ancient reception of Homer, thus activating contrasts between the ideology of Homer’s original audiences and the critical readings of the philosophical and literary culture of later centuries. An outstanding example of the first is the discussion of the climactic encounter of Aeneas and Turnus at the end of the poem, which alludes to the duel of Achilles and Hector in Iliad 22, to the general pattern of Homeric

X

F oreword

scenes of supplication (never successful) on the battlefield, and to Priam’s successful supplication of Achilles for the body of Hector in Iliad 24. In Aeneas’s famous hesitation as to whether or not to kill Turnus we are to read “two contrasting traces—­two tracks mixed—­ both signed by Homer” (chapter 4). This remains one of the most penetrating studies of the problematic end of the Aeneid. An example of the use of the later reception of Homer is the juxtaposition of Juturna’s complaint, in her lament for her brother Turnus, that as an immortal she cannot escape her grief through death, with the statement by the Epicurean Philodemus (On Piety 2) that the gods are more wretched than mortals, since they “endure woes that last for all eternity, because they live forever” (see appendix). Once again the young Barchiesi was ahead of the game, anticipating the increased pace over the last couple of decades of the use of Philodemus in the interpretation of Augustan poetry. In his emphasis on the inseparability of Vergil as reader and Vergil as writer, and on the importance of using ancient Homeric scholarship and criticism to reconstruct Vergil’s reading of Homer, Barchiesi was converging with other scholars of that time. R. R. Schlunk’s The Homeric Scholia and the Aeneid: A Study of the Influence of Ancient Homeric Literary Criticism on Vergil came out in 1974, and in my Virgil’s Aeneid: Cosmos and Imperium (1986) I tested the possibility that Vergil drew on a Pergamene allegorical reading of Homer in his own Homeric imitation. These days we are very comfortable (perhaps too comfortable) with the idea that Roman poets, as well as Hellenistic Greek poets, read their texts with scholarly and exegetical aids. It was not always so: I remember an eminent Oxford classicist in 1980 responding to my own suggestion that Homer came to Vergil bearing the freight of the intervening history of scholarship and interpretation with his own conviction that Vergil, of course, just went straight to the text of Homer. In the matter of “reception” Barchiesi 1984 may seem dated. The reader’s response to Vergil’s imitation of Homer, the “legibility” of the text, is seen as being controlled by generic frameworks and by the history of reading, leading to an allusive—­and controlled—­ polyphony. Barchiesi is not shy of the language of intentionality, for example, at the start of chapter 3, “the multitude of models becomes fodder for narrative strategies that aim to produce and control

F oreword

XI

determined effects of reading.” Or the argument in chapter 4 that the “difficulty” of Aeneas’s decision whether or not to kill Turnus is “intentionally produced by the text.” We may here detect the influence of his teacher Gian Biagio Conte, who similarly avoids the pitfalls of appeal to an authorial intentionality by hypostasizing an intentio operis. The model of a controlled polyphony is also comparable to the “intentional polysemy” posited for Augustan art and literature by Karl Galinsky (who, however, is also prepared to speak of “authorial centres”).3 La traccia emerges from the intellectual climate of structuralism and semiotics, as yet untouched by poststructuralist indeterminacy and the relativism of Charles Martindale’s influential brand of reception.4 The book does, however, position itself with regard to what at the time was the new wave of Anglophone Latin literary criticism, a form of close reading generally in line with the methods of New Criticism, and which often pushed textual ambiguity in the direction of a moral ambivalence, or even hostility, toward the political goals of the principate. Barchiesi is happier with a model of ideological complexity and contradiction, which finds expression through the superimposition within the Vergilian text of the different readings of Homeric epic at different times in history. Barchiesi’s resistance to the kind of associative close reading practiced by American critics such as Michael Putnam sometimes leads to an excessive skepticism—­for example, in the summary dismissal of the idea that the (conventionally) Punic lion in the simile at A. 12.4–­9 establishes a symbolic link between Turnus and Dido; the language of line 5, saucius ille graui uenantum uolnere pectus, forcibly reminds an unbiased reader of Aeneid 4.1–­2, At regina graui iamududum saucia cura | uulnus alit uenis . . . (the connection is sympathetically entertained by Richard Tarrant in his recent commentary, on A. 12.4). La traccia is also very much the product of a “Pisan moment.” It was published as the first monograph in the “Biblioteca di Materiali e Discussioni,” and “The Lament of Juturna” appeared as an article in the first issue (1978) of the journal itself, which also included articles by Gian Biagio Conte, Maurizio Bettini, Alessandro Perutelli, and others. Within a few years links began to be forged between the “Pisan school” of Latinists and scholars in the United States and Great Britain—­in particular, a second wave of theoretically aware

XII

F oreword

Anglophone Latinists, many of whom were students of, or otherwise close to, John Henderson in Cambridge, a loose group to which Don Fowler pinned the label of the “New Latin.”5 Barchiesi’s early works, including La traccia, found a receptive audience among the Anglo-­ American New Latinists, with whom Barchiesi himself, as well as other products of the Pisa school, have established close working relationships. One might indeed speak of an Anglo-­American-­Pisan tradition of New Latin, although the adjective “New” becomes less and less appropriate as the years pass and as once young Turks come up to retirement! It will be interesting to see in what new directions the current generation of young Latinists, whose reading and writing practices have been significantly shaped by Don Fowler’s “New Latin,” will take our discipline. Coming back to Homeric Effects a quarter of a century after I first read it, I am struck by its continuing freshness and ability to stimulate, a mark of the prescience and fertility of its agenda. The conceptual breadth is combined with the highest standards of scholarship, an object lesson in the ideal symbiosis of theory and philology, and a hallmark of all of Barchiesi’s work. Homeric Effects has long been a part of the mental furniture of professional Latinists. Its publication in English is to be welcomed as making it accessible to a wider range of readers and students. Ph i l i p H a r di e Trinity College, Cambridge

F oreword

XIII

Wor k s Ci t e d i n t h e For e wor d Conte, G. B. 1974. Memoria dei poeti e sistema letterario. Torino. Fowler, D. 1995. “Modern Literary Theory and Latin Poetry: Some Anglo-­ American Perspectives,” Arachnion. A Journal of Ancient Literature and History on the Web. http://www.cisi.unito.it/arachne/num2/fowler .html (accessed May 4, 2014). Galinsky, G. K. 1996. Augustan Culture: An Interpretive Introduction. Princeton, NJ. Hardie, P. 1990. “Flavian epicists on Virgil’s epic technique,” in A. J. Boyle, ed., The Imperial Muse: Ramus Essays on Roman Literature of the Empire, 3–­20. Berwick, Victoria, Australia. Hinds, S. E. 1998. Allusion and Intertext: Dynamics of Appropriation in Roman Poetry. Cambridge. Knauer, G. N. 1964. Die Aeneis und Homer. Studien zur poetischen Technik Vergils mit Listen der Homerzitate in der Aeneis (Hypomnemata, 7). Göttingen. Martindale, C. 2005. Latin Poetry and the Judgement of Taste: An Essay in Aesthetics. Oxford. Pasquali, G. 1951. “Arte allusive,” in Stravaganze quarte e supreme, 11–­20. Venice. Tarrant, R. J. 2012. Virgil: Aeneid Book XII. Cambridge.

INTRODUCTORY NOTE Twenty years have passed since Knauer published his comprehensive collection of comparisons between the Aeneid and Homer: one glimpses in this monumental work the closure—­and the definitive balance sheet—­of a scholarly activity stretching from ancient inquiries into furta Vergilii right down to the great commentaries of the modern period. But one can also see in this inventory of comparative materials an opening for new research. Space remains, I think, for one who wants to investigate the functions the Homeric model assumes in the composition of the Vergilian text. For me this involves taking up again the analysis of intertextual echoes by attempting to interpret them as moments where the narrative legibility of the ­Aeneid is exposed to view. This means, on one hand, considering Homeric echoes for the overall meaning they acquire through continuous reading; on the other, it means considering them as elements in a strategy of storytelling: the narrative character of the Vergilian text is what requires specific analytical tools. And if we nearly always begin with comparison of single passages, that favored terrain for studies of allusion, we will often find that we have trespassed into the disciplinary field that some call poetics, others narratology.1 As this research has unfolded, I have found it ever more necessary to differentiate the many possible meanings that coexist in our usual notion of a literary model. Thus it is useful, for example, to introduce two working distinctions. The first contrasts direct imitation of the Homeric texts (recuperated in all their distant fixity) with XV

XVI

I ntroductory N ote

the reappearance of Homer through the filter of a long, continuous transcoding process (diverse reinterpretations, inscribed within successive cultural systems). The second, from another angle, opposes a Homer that serves as example model, a source of authoritative memories, and a Homer that functions as a genre model, a matrix capable of generating a new text that is meant to take that model’s place. This multiplying of intertextual functions has proven useful not just as a theoretical refinement but also as a critical response to complex qualities of Vergil’s poetic practice. It is my aim to explain the effects of reading that the text of the Aeneid intends. The interaction of diverse models does not remain, from this perspective, solely a cultural fact or the sedimentation of a tradition: each model, in becoming language, accepts coexistence with others and adapts itself to constructing the sense of the narrative. And it is precisely the action of underlying, implicit models to set us on the path from the extended meaning of the narrative text to the problematic dimension of its ideology. I have collected here not all of my Vergilian papers but only those that seem to me part of a unified project, one that is still similar today even if it has become more specific along the way. For example, the first chapter insists most of all on the reception of Homer that conditioned the Vergilian transformation; the fourth and last, connected to my more recent interests, concerns instead the link between the complexity of meaning in Vergil’s narrative and the plurality of uses to which the Homeric text is twisted.2 A l e ss a n dro Ba rch i e si Pisa 1984

HOMERIC EFFECTS IN VERGIL’S NARRATIVE

Ch a p t e r 1

THE DEATH OF PALLAS Intertextuality and Transformation of the Epic Model

1.   T r a nsfor m at ions of t h e Pat rok l e i a In the series of slaughters that occupies all of Aeneid 10, a compositional project gradually takes shape. Enough material is gathered here to fill several Iliadic books and nearly all the poem’s heroes are given their own lengthy aristeia punctuated by minor episodes. But still the reader is guided through this chaotic chain of events by a clarifying thread, since a familiar and already assimilated model gives light to navigate it. The tenth book “corresponds to the victorious events of Patroclus, to his death at the hand of Hector and the beginning of Achilles’ revenge.”1 Pallas is based on Patroclus, Turnus on Hector, Aeneas on Achilles. Just like Achilles, Aeneas, after the death of his young friend, is driven by overwhelming desire for final vengeance.2 As with Hector, Turnus’s victory is at once the peak of his success and, without his realizing it, the basis for his defeat. Like Patroclus, Pallas is dragged by desire for valor into a tragic combat against a stronger enemy—­and the narrator’s passionate participation intentionally balances here the famous qualities of pathos found in Patroclus’s episode. But more than analyzing the mere content of the Vergilian debts, which in any case have already been sufficiently traced many times,3 what interests me here now is how the Latin poet goes about transforming a paradigmatic event of heroic epic like the Patrokleia. 1

2

C hapter 1

If a reader finds in the web of the Homeric narrative a “guide” for deciphering the new narrative, this is not just confirmation of her role as epic addressee and witness to the renewal of a tradition: rather it is the first step toward a new, more sophisticated and critical, relationship with the epic action. The two texts present themselves to readers as though in immediate continuity, which makes the system of differences far more conspicuous. This dynamic of transformation—­the intertextuality that generates the sense of the text—­is not established without the reader’s active cooperation; indeed, it nearly coincides with the process of comprehension that the author envisioned. A first example may be useful at this point, to sketch in outline the kind of analysis I will develop later. When young Lausus is killed, the poet develops Aeneas’s reaction with unusual psychological depth (et mentem subiit patriae pietatis imago [“an image of paternal care struck his mind”], 10.824): the victor groans in sorrow and extends his hand to address the dead as a sign of mourning;4 his words pay homage to the young man’s courage; he leaves to him the armor that made him so happy and, finally, not only grants him burial but also takes up the lifeless body in his own arms to return it to his comrades, encouraging them to approach without fear. This scene of great pathos constitutes a systematic reversal of a literary stereotype that the Vergilian text has already established in readers’ memory, both by its insistent repetition and its (“vertical”) appeal to the established typology of Homeric tradition. Epic scenes of death in battle normally entail, with fixed formulae: a triumphal euchos over the fallen body (the victor boasts of his courageous feat, addresses with derision the dead and his companions5), the stripping of the armor, sometimes accompanied by the theft and mistreatment of the corpse, and its abandonment to dogs and birds. More specifically, the image of Aeneas lifting the enemy’s body from the ground, unique both in Vergil’s poem and in the Homeric model itself,6 vividly symbolizes the concept of pietas and misericordia as integral parts of a new heroic ideal, one further defined by its very difference from the Homeric predecessor. And the model stands out all the more clearly thanks to the immediate context. A narrative parallel, in fact, clearly links Lausus to a warrior of the enemy camp, Pallas: both are young, valorous, of outstanding

T he D eath of P allas

3

beauty, sons of fathers who love them as their sole comfort, and destined to die young from blows of a stronger and older adversary. The narrator explicitly juxtaposes them, in lines that also serve to foreshadow their fates: Hinc Pallas instat et urget, hinc contra Lausus, nec multum discrepat aetas, egregii forma, sed quis fortuna negarat in patriam reditus. Ipsos concurrere passus haud tamen inter se magni regnator Olympi; mox illos sua fata manent maiore sub hoste (10.433–­38). [Here Pallas hunts and harries, against him Lausus, both about the same age, each fine in form, but fortune denied them their homecoming. Yet the ruler of great Olympus did not suffer them to confront each other; their fates awaited them soon at the hand of a greater foe.] It is clear now that the parallelism becomes an instrument of contrast: by killing Pallas, Turnus acts as a typical Homeric warrior, a negative counterpoint to Aeneas’s humanity.7 The juxtaposition of the two fates has a certain affinity with the thematic structure of the Patrokleia, where the consecutive deaths of Sarpedon and Patroclus seem to form a single complex marked by their shared tone of pathos. But most decisive is the transformation Vergil achieves by multiplying the cultural values that the text puts in play. If the functional opposition of these two scenes is readily discernible, and even emphasized by the recurring polarity between the two chief antagonists Aeneas and Turnus, still perhaps a couple of less obvious observations can be made. First, it does not seem that we are meant simply to contrast the attitudes of the two main heroes and trace them back to distinct, opposing paradigms. I take it as given that the correct analysis of characters in a narrative involves comparing each character not with one other character more or less arbitrarily determined, but instead with the entire articulated complex that makes up the system of characters. From this angle, it becomes clear that Turnus’s behavior toward the enemy is shared by all the characters involved in the epic action and falls within the

4

C hapter 1

horizon of normal expectations that the text establishes through parallelism and repetitions of events (any battle scene of the Aeneid could be cited as example of this), whereas Aeneas’s behavior toward Lausus is entirely unique and thus “abnormal.” (Naturally it would be easy to jump to conclusions here: for Vergil’s Roman readers—­ the audience Vergil had in mind—­if the concept of normality were suddenly overturned by its opposite, Aeneas’s abnormality could not but represent the arrival of authentic civility; but we will see momentarily that this may be too hasty a reading.) This observation may be confirmed by the fact that Aeneas’s gesture of pity not only contradicts but even lays waste to what Aeneas himself does on the battlefield, even in this very book. Indeed, after Pallas’s death Aeneas gives up any self-­control and his actions recall, through explicit allusions, Achilles’ raging slaughter in the Iliad to avenge Patroclus (cf. books 20 and 21). Deaf to supplications, sarcastic and cruelly dismissive even of family bonds (cf. especially 10.595–­601), even ready to ritually sacrifice prisoners (517ff.), pius Aeneas strides the battlefield like Aegaeon, who dared to challenge Jupiter’s thunderbolts, brandishing spears in his hundred arms. It thus seems arbitrary to emphasize unilaterally the negative value that the synkrisis assumes with regard to Turnus alone.8 What matters is the construction of a complex system of perspectives that awaken the audience’s critical understanding through a continuous adjustment of “narrative distance.” In the prospective vision that the Vergilian text makes its readers adopt, the Homeric text acts not only as literary model (both as a complex of elements and structures to imitate and as a formal matrix for epic), but it is present also as a cultural model that has by now become relativized. (Intertextual significance involves precisely holding these two modes together in a single communicative act.) The Homeric world is not configured as the only possible one (nor, on the other hand, as a place of “uncivilized” values to be rejected), but instead assumes a “precivilized” quality. And indeed it is this intertextual mode that allows Vergil convincingly to evoke the heroes of distant, “primitive” Italy. It is not lost on anyone that Turnus, Mezentius, and Camilla are much more “Homeric” characters than Aeneas. Let us ask the question from the side of the positive protagonist. Contrary to the Homeric heroes—­and this is quite a substantial difference—­Aeneas is

T he D eath of P allas

5

essentially a founder. If we consider the values of the protagonist and those of the world around him in the poem, we must conclude that Aeneas is a hero of a humanity that does not yet exist. For this reason he pays the price for his ambiguous relation to “barbarity”: in the name of constructive values that are firm but still far away in the future, his duty is to destroy; the darkest and most difficult moment in the process of foundation is reserved for him—­the man of peace will instead be, as we can clearly gather, his son Ascanius/Iulus. After all, the nature of Italic barbarity also recalls two different ages at once, the Age of Saturn and the Iron Age, and thus includes both many outmoded values and, at the very same time, many germs of positive virtues. The literary representation of this conflict could not help but foreshadow a system of disparities between characters in action such that it threatens the integrity of the values typical of archaic epic. In this sense, the orientation of the Aeneid toward the Augustan present risks not finding a point of synthesis with the Homeric tradition, since the juggling of values and ideologies was not easily accommodated by the epic form of horrida bella. Clearly, we are speaking here of cultural expectations that tend to push to a breaking point the communicative capacity of epic narration, already codified by Homer in the service of other values and quite different expectations. With these premises, my argument aims to be immediately substantiated by textual evidence. Its main advantage with respect to other possible critical angles (which share a certain initial structural resemblance) seems to consist in being able to get directly into the work’s linguistic texture, in search of a functional nexus between imitation of Homer and poetic signification. In short, my inquiry pays attention to the relation between the transformation of the Homeric model (the moment of origin and construction of the new epic work)9 and what the Aeneid seeks to convey to its readers; and we might add that it may also involve Vergil’s attempt to construct a radically new “epic addressee.” An area particularly suited to this inquiry is a portion of the text where a maximum of Homeric redundancy occurs with a maximum of purposeful information: the episode of Pallas’s death in book 10, a complex synthesis of Homeric imitation and at the same time a turning point in the plot of the Aeneid.

6

C hapter 1

2 .   H e rcu l e s’ T e a r s: T h e Hom e r ic Mode l a s St r at i f ic at ion (10. 4 64– ­7 3) By combining a scene of divine counsel with the narration of a duel, the epic poet both pauses the action and—­as Homeric scholiasts already observed10—­generates a particular state of expectation in the reader. Using this technique Vergil spotlights and foreshadows the episode of Pallas’s death. The youth’s prayer to his protector Hercules prior to battle is heard in heaven, but it cannot be fulfilled: hence this Olympian intermission, where Hercules sits silent, suppressing his grief and shedding useless tears, and Jupiter intervenes—­father to son—­to explain the great laws of fate governing human affairs. Homer had treated with a similar technique two important episodes: the duels between Hector and Achilles and, before that, in Iliad 16, between Sarpedon and Patroclus. The two scenes are built on parallel lines: twice Zeus is moved to pity and about to rescue from certain death a warrior (Sarpedon, Hector) to whom he is joined by a special bond. This intervention, analogous to so many others that dot the Trojan landscape, would put the god in conflict not only with the inevitability of moira (we know, after all, that in the Iliad these fated deaths are not the result of an iron-­clad universal necessity, as is the case for Vergil’s Fates) but even with his own boulé. The conflict would be irreconcilable, since Zeus himself had already predicted the entire series of killings by inserting them into his overall plan: Patroclus will have to kill Sarpedon, son of Zeus, so that Hector will kill Patroclus and finally Achilles kill Hector. Twice, ­ thena) interfacing this crisis of the divine order, a goddess (Hera, A venes and forces Zeus to accept the necessary unfolding of events. The two scenes thus appear to be marked by a theologically scandalous motif: the inferiority of Zeus.11 The conflict is particularly acute in the dialogue preceding Sarpedon’s death. Twice Zeus has already saved his son from death in battle (Il. 5.662; 12.402ff.) by visibly revealing the intensity of his protective love. When in book 16 the god “sees and feels pity for” Sarpedon (431), we are prepared—­thanks to the rigidity of the formulaic language—­for a repetition of these events. In the Iliad, divine help is consistently mediated by ἐλεεῖν (“pity”), which generally involves a wholly private interest and active participation in

T he D eath of P allas

7

the affairs of a mortal motivated by a blood-­relation and a special devotion.12 For an omnipotent being to feel pity normally entails acting in ways that change the course of mortal affairs. So Zeus’s words have a surprise in store for us. He pities Sarpedon and suffers for him, but his heart is torn between two irreconcilable alternatives: save his life or respect a predestined “plan.” Hera replies with a rhetorically sophisticated argument that sounds like a threat: if he spares Sarpedon’s life contrary to Fate, Zeus will be setting an example of insubordination for all the other gods who have seen and will see their own children die on the fields of Troy. It is a moment in which we witness the relative stability of Zeus’s power on Olympus and discern his dependence on a divine consensus that could even, in theory, suffer collapse. Thus the god must obey (οὐδ’ ἀπίθησε, 16.458): the help denied Sarpedon while alive will be somehow made up for with intervention by Apollo, who will at least prevent any violation to Sarpedon’s body. Zeus then rains down blood, bathing the battlefield on which Fate is about to unfold. The event reveals in an exemplary fashion a strong limit to divine power: the inability to completely overcome the death of human beings.13 The scene is justly famous, and Vergil’s imitation of it is indisputable, blatant even, because of a reflexive sort of allusion (. . . quin occidit una / Sarpedon, mea progenies, 10.470–­71):14 the poet thus establishes an intertextual relation that provokes comparison between two distinct and opposed entities, without caring in the least bit to hide the sutures binding the “old” text and its new contextualization. But in order to capture precisely this interaction between the two texts, it is necessary to clarify the specific meaning of each one, and so, first, we must reckon with our reading of the Homeric model. Let’s open up one of the standard commentaries on Vergil: “The scene is of course suggested by the celebrated passage  .  .  . where Zeus weeps tears of blood for his son Sarpedon” (Nettleship ad A. 10.464ff.). Zeus’s “tears of blood” would indeed offer a satisfying parallel to Hercules’ lacrimae inanes (though amounting to a sort of Dämpfung, their reduction to a more human level). But is there a legitimate basis for this connection? In Homer, Zeus sheds on the earth “bloody drops” (αἱματοέσσας . . . ψιάδας, 16.459) to honor his son about to die. Surely we know that Zeus suffers (his speech begins with ὤ μοι ἐγών, 433), that his heart is weeping for Sarpedon’s

8

C hapter 1

lot (τεὸν δ’ ὀλοφύρεται ἦτορ, 450), but the motif of the bloody rain cannot at all be interpreted, along these lines, as a symbol of lament. It is clear, instead, that it is meant as a portent, like the one described at Il. 2.45, when Hera and Athena make the heavens rumble to honor their protégé Agamemnon, just like, also, the “bloody dew” that Zeus spreads on the earth “because he was about to hurl to Hades many stalwart heads” (11.53–­55). We are dealing with a prodigy that has a precise, honorific function,15 and at the same time serves to anticipate earthly events; according to the same logic Zeus causes dreadful carnage over the body of the son (cf. Il. 16.567ff. and 662) as though to compensate for and memorialize his death with a literal bloodbath. “Zeus’s weeping,” which modern critics so often mention,16 is thus an interpretation almost certainly foreign to the intentions of the Homeric text. And yet, one can show with equal ease that this reading is not extraneous to Vergil, since it is rooted already in ancient culture along a line of reception that runs from Plato to the Christians.17 The representation of extreme grief is harmful—­explains Plato in the Republic—­and thus a certain type of mimetic poetry should be banned from the education of children. It is therefore problematic that Homer describes not only human beings in the throes of this kind of emotion but even the gods. He shows them as ὀδυρομένους, as when Thetis laments for Achilles, or even the father of gods who tortures himself over the death of his beloved Sarpedon (Rep. 388B–­C). To emphasize the bad effects of such mimesis, Plato tendentiously exaggerates the Homeric scene’s content: even the quotation itself seems manipulated in this direction, producing the impression that Zeus is allowed to fall into a sort of threnos.18 Thus Plato’s criticism has probably radicalized the scandalous effect of the Homeric episode for later readers. It is difficult to think it a coincidence when we learn that Xenodotus deleted from his edition of the Iliad the whole dialogue between Zeus and Hera (this is how καθόλου περιγράφει of schol. II. 16.432 [A], 257ff. E. is interpreted).19 So lodged even in philological practice and in the material constitution of the Homeric text is a widespread critical attitude toward the incoherence of Homeric theology. We can understand then why the quasi-­Stoic attempt to rehabilitate this episode through allegory came about: the bloody rain cannot mean that Zeus sheds

T he D eath of P allas

9

tears—­for Zeus is ἄκλαυστος—­but it is instead an act of sympatheia, a general cosmic compassion manifested in the “tears of the ether.”20 The image arises (only to be immediately negated because exegetically inappropriate) of tears of blood: this reading conceals and then keeps hardening over the original text—­until it becomes the only valid reading for Christians, who use it polemically to discredit the mythological tradition.21 So the model for Hercules’ lament must be sought not on the objective surface of the Homeric text but in the complex history of its “stratified” reception. But another, more important consideration serves to clarify the relation between the Homeric episode and the new ideological orientation of Vergil’s imitation. In the Iliadic scene it is clear that Zeus could in theory halt the destiny, the aisa, of Sarpedon. We are not explicitly told that a mortal’s fate is immutable; Hera’s only objection is that by doing so Zeus will open himself up to criticism from the other gods (v. 443) and will induce them to act likewise in similar cases (v. 445ff.; for an analogous episode, see Il. 5.110ff.). The situation changes drastically if this personal and somewhat flexible destiny is replaced by a totalizing understanding of fate as a universal necessity superior to the individual gods’ wills. Subjecting the Homeric theme to this new cultural filter, Cicero illustrates it in this way: si enim nihil fit extra fatum, nihil levari re divina potest. Hoc sentit Homerus [!], cum querentem Iovem inducit quod Sarpedonem filium a morte contra fatum eripere non posset (“If nothing happens outside fate, nothing can be alleviated through divine help. This is what Homer realizes, when he has Jupiter lament that he cannot rescue his own son Sarpedon from death against the will of fate.”) (Div. 2.25). The dialogue between Zeus and Hera, then, becomes a paradigmatic demonstration of one of the most widespread gnomic sayings: fate is so inescapable that not even gods are free from its laws.22 This interpretation finds its way even into Homeric exegesis, which manages to neutralize the critiques deriving from philosophical sources: “We must not blame the poet. Actually we need either to abandon the idea that gods are bound to humankind, or make statements coherent with this. Furthermore, Zeus’s lament has an educational force, since the poet teaches that even the gods abide by the heimarmene: all the more will human beings need to bear the weight of their destinies with nobility” (schol. Il. 16.433–­38, 258 E).

10

C hapter 1

This is exactly the “didactic” content that Vergil’s imitation strives to make explicit: stat sua cuique dies, but men must in any case exercise their ennobling virtus; even the gods, even Jupiter, are subject to this law of fate. Hera’s reasoning: πολλοὶ γὰρ περὶ ἄστυ μέγα Πριάμοιο μάχονται υἱέες ἀθανάτων, τοῖσιν κότον αἰνὸν ἐνήσεις. (16.448–­49) [For many fighting around the great city of Priam are sons of gods, whose dread anger you will arouse.] is presented as upside down according to this new direction. No longer is it a matter of what the other gods could do for their sons, but of what Zeus (in the “past” of the Iliad) has already agreed to submit to. In this way a dangerous contradiction between divine pronoia and heirmarmene—­one that could endanger the ideological framework of Vergil’s epos—­is a priori reconciled.23 The role of the Homeric Zeus is unloaded onto a character of a different status, closer to the human level and open to emotion; Jupiter, on the other hand, can only speak the language of Fate.24 Through a remarkable exchange of roles Jupiter has landed a role somewhat opposed to the one he used to play in the Homeric model: the Zeus admonished by Hera now admonishes Hercules in turn. In the literary operation I have sought to describe, the transformation of Homer cannot be understood without taking into account its reception, not only because Vergil, necessarily, started from his personal reading of the model (this aspect, frankly, situates itself in a sort of prehistory of the text, and the analysis would run the risk of devolving into a merely conjectural genetic construction), but most of all because this reception already represents the system of cultural expectations with which the work must come to terms in order to attempt its communicative venture. From this point of view, which effectively privileges the addressee as a function inscribed and prefigured in the text, it becomes productive not to limit ourselves to focusing on connections with the Iliad but instead to consider the entire system of intertextual meanings that accumulate in the “Homeric” narrative fabric. When, as here, Vergil takes into his text Homeric material—­the narrative opposition

T he D eath of P allas

11

between two divinities debating the destiny of a character—­the multiplication of cultural codes takes on an interpretive function: its role is to mediate the reader’s reactions and to guarantee, by so steering them, a certain type of understanding of the events. To recognize these “additive” meanings now will be the second point of our analysis. The opposition between Hercules and Jupiter, voice of compassion and voice of Fate, is made concrete in the text through a play of contrasting registers. From Hercules comes a gesture of pity, where suffering is mixed with self-­control: Audiit Alcides iuvenem magnumque sub imo corde premit gemitum lacrimasque effundit inanis. (10.465–­66)25 [Hercules heard the young man and suppressed a deep groan in his heart and shed tears in vain.] The deified hero is still susceptible to human suffering, but must master it within the bounds of decency unavoidable for gods according to a well-­known cultural norm: κατ’ ὄσσων δ’ οὐ θέμις βαλεῖν δάκρυ, Artemis says in a rather similar tragic situation. The gods are not allowed to weep for human affairs.26 This image of Heracles afflicted is all the more effective when we consider that in ancient tradition he is the exemplar of long-­suffering par excellence (ἀστένακτος αἰὲν in Soph. Trach. 1074; αἰὲν ἄδακρυν in Theocr. 24.31). For this reason the idea of Heracles weeping underwent literary developments that exploit its surprising and almost paradoxical impact. In Bacchylides’ fifth epinician the hero, having returned from Hades, hears the story of the premature death of Meleager and is so moved that, for the very first time in his life (says the poet), a tear came to his eye (151ff.). In turn, Sophocles and Euripides repeat the motif in their dramas on Heracles: “I moan and weep like a girl, something nobody has ever seen happen to this man; I’ve always yielded to misfortunes without a groan”; “I have never shirked any trial, never had tears pour down my eyes, nor ever imagined it coming to this: that a tear would fall from my eyes.”27 This singular instance of Heracles’ lament (a

12

C hapter 1

unicum that the poets so often repeat!) is so popular that it even finds its way onto funerary monuments; an inscription commemorating Polystratos reads: “for a friend of Heracles, whose eyes never wept, Heracles has shed a tear and let loose a groan.”28 So Vergil, by deliberately evoking a common traditional motif, strikes a note of sorrowful compassion that serves a narrative purpose for the premature death of Pallas. Jupiter’s speech has a much greater burden of information that it needs to convey. He has to communicate, with all the authority that the work’s ideological system confers upon him, the fundamental teaching on the need to adapt oneself to Fate—­in accord with the standard reading of the Homeric model, as we have seen. But here we have ideological themes that are also essential to narrative development, since by announcing these higher truths Jupiter anticipates and motivates what happens first to Pallas and then to Turnus. For this commenting function to come off, the “didactic” content must be expressed in a proverbial, commanding form, and the written word itself must gain the audience’s consent, by presenting a detached and almost atemporal vision of the narrated world. ( Jupiter cannot possibly allow himself to be tainted by human passions like the Homeric gods: he has to set rules.) The highly topical structure of the entire speech thus becomes clear: the programmatic use of predictable and persuasive conceits that permit the audience to decode the epic action in ideological terms. Stat sua cuique dies, breve et inreparabile tempus / omnibus est vitae; sed famam extendere factis, / hoc virtutis opus (“Everyone has their day, and a lifetime is short and cannot be turned back; but to spread fame with deeds, this is virtue’s work.”).29 A sententia of the most generic sort, expressed in a steady and monumental rhythm, opens the speech, with an opposition between the brevity of life and the endurance of glory gained through deeds of valor. One might be tempted here to go back to the heroic ideology of the Iliad: Achilles too, as we know, is faced with the problem of kleos, and chooses a short life in exchange for it (cf. Il. 9.410–­16). But it is, after all, a conscious choice, based on a single specific prophecy. The Vergilian sententia is distinct and casts a wider net: it exploits the ideological opposition between immortal glory—­gained through virtus, which is its necessary complement—­ and the inevitable brevity of human life, defining it with reference

T he D eath of P allas

13

to Pallas’s specific condition of being ἄωρος. The sense of compensation between death and fame is fully achieved in the original stylistic figure famam extendere factis: the idea of stretching implied by extendere conflicts with the breve . . . tempus of Pallas’s life, and the tension so created discharges onto fama. This novel phrase reveals, even at a level of rhetorical construction, how fame (time that is “spoken” by others) is clearly the only possible surrogate for lived time.30 The theme of transience is therefore no isolated element (just one among countless commonplace reflections on the fleeting nature of time),31 but seems instead tied to a more complex motif that is, as it were, already culturally inflected. We must remember that it is essentially a rhetorical motif—­not by chance often found in Cicero’s work—­one articulated around a consistent metaphorical scheme: etenim, Quirites, exiguum nobis vitae curriculum natura circumscripsit, immensum gloriae (“Indeed, citizens, nature has circumscribed a narrow course of life for us, but one unbounded in glory.”) (Rab. perd. 30); nullam enim virtus aliam mercedem laborum periculorumque desiderat praeter hanc laudis et gloriae; qua quidem detracta, iudices, quid est quod in hoc tam exiguo vitae curriculo et tam brevi tantis nos in laboribus exerceamus? (“For virtue desires no other reward for its labors and dangers besides praise and glory, which if it is taken away, judges, what is there in this such short and narrow span of life for us to exert ourselves in such labors for?”) (Arch. 28); nesciebam vitae brevem esse, cursum, gloriae sempiternum (“Did I not know that life’s span is short but of glory everlasting?”) (Sest. 47). It is no less suitable also for the didactic and philosophizing prologue of a historiographical work: quo mihi rectius videtur ingeni quam virium opibus gloriam quaerere et, quoniam vita ipsa, qua fruimur, brevis est, memoriam nostri quam maxume longam efficere (“For which it seems to me more proper to seek glory with the resources of spirit more than of strength, and since the life we have available is short, to make our memory last as long as possible.”) (Sall. Cat. 1.3).32 In the Vergilian context this recognizably didactic tone is meant also to console: it shows Pallas’s death not only as part of the general human condition—­as ancient theorists of consolations prescribed upon the death of ἄωροι: breve et inreparabile tempus / omnibus est vitae33—­but as specifically compensated by glory won in battle. It is glory alone, in Cicero’s words again, quae brevitatem vitae posteritatis

C hapter 1

14

memoria consolaretur (“that can make consolation for life’s brevity with the memory of posterity”) (Mil. 97). This anchoring in positive values derived from official Roman ideology balances the intrinsic tragic nature of the events, by introducing a criterion of judgment foreign to the original epic code. Another great consolatory scheme, and very common, is reminding the grieving person that their loss is not exceptional since others have died and will die;34 similarly, he or she is hardly the first or last person ever to suffer such a hardship.35 Obviously, this rhetoric does not work as well in cases of untimely death: generalizing then becomes problematic, since this specific case does not follow any rule of nature but represents a painful and exceptional violation. One can only resort to a “weak” argument less directly tied to universal laws, the so-­called consolatio per exemplum. The consoler names great characters carefully chosen from among those who had to succumb to death;36 better still, he or she finds mythic correspondence with the present situation among the sons of gods who had to suffer premature death despite the divine power of their parents: καὶ θεῶν σκότιοι φθίνουσι / παῖδες ἐν θανάτῳ (Eur. Alc. 989 f.); καὶ μακάρων παῖδες ἔνερθεν ἔβαν (GV 2006, 8). It is no accident that we find this melancholy motif less in true attempts to console than in the sort of formulaic expression of grief meant for epitaphs of ἄωροι.37 Sometimes Sarpedon himself serves as the exemplar of untimely death: οὐδ᾿αὖ βροτῶν τε καὶ θεῶν πάντων ἄναξ Σαρπηδόν᾿οὐκ ἔκλαυσεν, οὐκ ἐκώκυσεν . . . ;38 [Did not the lord of men and gods weep, did he not moan for Sarpedon?] In Vergil the sequence Troiae sub moenibus altis tot gnati cecidere deum, quin occidit una Sarpedon, mea progenies; etiam sua Turnum fata vocant metasque dati pervenit ad aevi

T he D eath of P allas

15

[Under Troy’s high walls so many sons of gods fell, even Sarpedon among them, my offspring; his fates call Turnus too, and he has reached the goal of the stretch allotted to him] combines, in the wake of Homeric precedent, the more generic commonplace “others have died and will die” with the specific angle “even the sons of gods, like mortals, must bow to untimely death.” But one immediately notes that the use of such stiff clichés does not cause any inertia on the level of contextual meaning. The example of Sarpedon is chosen to produce an allusive recall that is at once useful for comprehending the whole episode, as we have seen, and rhetorically functional in the narrower context as a persuasive comparison. The future example, Turnus, is required by the economy of the narrative, since the reader needs to be able to anticipate the events that unfold from the death of Pallas—­just as the reader of the Patrokleia is made to expect the Death of Hector that follows. We are now aware that Turnus too is near his end,39 and the following episode will provide new means of anticipation and commentary. Meanwhile Jupiter, his task as mouthpiece of Fate accomplished, can avert his gaze from the forces on the field:40 the heroes will clash in full “epic liberty” beneath an empty sky. 3.   E f f ec ts of R e a di ng: T h e T h e m e of t h e Ba l dr ic (B a lt eu s, 10. 474– ­5 0 0) No device can remain unexploited in the story [ fabula], no episode can be without consequence for the situation of the story. It was to compositional motivations that Chekhov was referring when he insisted that if at the start of a story . . . we find that a nail has been started into a wall, at the end of the story the hero has to hang himself by that nail. —­B . Tom a še vsk i j (1965 , 2 82) The duel between Turnus and Pallas cannot be conflated with the many similar episodes crowding the poem’s final books with a more than “Iliadic” density: the narrative economy demands that this scene emerge decisively from its immediate context in order then

16

C hapter 1

to be recalled in the final scene. The very legibility of the narrative’s resolution is in play, Aeneas’s vengeance against Turnus now beaten down and suppliant. So Vergil’s textual strategy makes two significant interventions. The autonomy and narrative emphasis that govern the episode—­fundamental means to mobilize the reader’s attention—­result from the “prologue in heaven” just analyzed (the importance of an epic event is often proportional to the Olympian world’s involvement in the action), and from a particular way of describing combat. The linguistic treatment of the duel makes it stand out from its context (naturally the minor duels in book 10 are treated less completely, their narrative time far less dilated), and at the same time it appeals to the continuity of epic tradition with unusual force, through the most typical form Homer uses to narrate the strife among his heroes. The details are readily recalled: after exchanging threats and boasts, one fighter—­usually the one doomed to die—­takes the first shot and misses or only wounds his opponent lightly; then the second warrior strikes back and lays him low. The fall of the defeated to the ground is then described—­his death, the victor’s reaction (if any), and the stripping of armor. In reprising this entire “strong performance” of a Homeric duel, Vergil abandons his usual narrative practice: usually his sequences are abbreviated, which leaves it up to the reader’s epic competence to integrate; he omits formulaic passages (for example, the slain man’s fall to the ground is very often disregarded); or on the contrary, he intensifies unusual particulars, by playing on variation more than continuity. In this duel it seems instead that he presents a montage of fixed or “formulaic” elements, genuine epic brands, where the author’s freedom is limited to arranging in context the traditional material, conflating diverse sources, and inserting strictly measured, “subjective” notes within the rigid lattice of epic narration. The overall vision of this literary operation (which condenses into a unified register both Homer and archaic Latin epics) warrants at least an endnote.41 As always, of course, a sufficiently fine-­grained analysis would also demonstrate here Vergil’s absolute artistic independence and assert his forceful expressive originality.42 But for our purposes a different level of analysis is needed, one that emphasizes not the striking innovation of single lexical or syntactic traits but the overall

T he D eath of P allas

17

attempt to create a “type scene,” which functions as one strong signal of allusion to the epic tradition. The duel stands out as a unified figurative topos with which, in terms of expression, strongly codified and immediately recognizable verbal-­rhythmic sequences are associated: Homer, Ennius, and the other Latin epic poets constitute, from this point of view, a single literary system that can be cited collectively. Moreover, here as elsewhere in literature, to down­play the importance and utility of stereotypes would be a dangerous critical error. Traditional clichés are valid at least as genre markers and thus as a necessary premise for the new text’s legibility (whatever the author’s adherence to or respect for the tradition); they produce in the reader not arbitrary and uncontrollable mental associations but processes of constrained understanding that, though transcending the letter of the text, have still been prefigured by the author. They may, for example, serve to reintegrate the cliché within its original boundaries, or else set the cliché in “horizontal” opposition to the context in which it has now been transplanted. (In practice, poets presuppose a literary genre as a communication program, determining the response to the solicitations contained in each single text.) But most importantly—­to come to our specific problem—­the citation of stereotypes like these Homeric ones has the advantage of creating by itself a system of coordinates that brings into view every possible divergence, however small, from the tradition. It is therefore a highly productive mechanism for generating narrative meaning: all the more effective when the text being cited is strongly codified and even repetitive, and when it is deeply stamped in the cultural memory of the audience. This is not an act of allusion in the “Alexandrian” sense: the Vergilian recasting does not presuppose recognition of any particular expression or theme, but involves on a global level a group of typical, formulaic elements spread widely through the Homeric text (as can easily be seen from the comparisons collected in note 41 earlier). More than a single episode, we find here the general skeleton of the epic duel—­and recognizing this is surely easier for readers of “average” competence than tracing back a precise allusion to its source. The system of expectations carefully implanted in this way must be fulfilled by the end of the duel: after the fall of the vanquished comes, in the normative syntax of Homeric narration, the victor’s gesture of placing

18

C hapter 1

his foot on his enemy’s corpse, ripping out his spear, and despoiling his armor. Here is what one prominent critic has found in Vergil’s text. Comparing Il. 16.503–­5, where Patroclus tears his spear from Sarpedon’s body and his phrenes depart from the hero along with his life, Klingner 1967 observes: “the victorious adversary rips his spear from Sarpedon and from Pallas . . . in a similar manner” (578 n. 1). It seems then that Turnus, like a typical Homeric victor, puts his foot on the body of his enemy, laid low and dying, to recover his own weapon (Il. 16.503–­4: ὃ δὲ λὰξ ἐν στήθεσι βαίνων / ἐκ χροὸς ἕλκε δόρυ; cf. Hector over Patroclus’s corpse at Il. 16.862–­63). But there is a sort of optical illusion here. In Vergil’s story it is not Turnus but Pallas, the defeated, who tears the spear from his own wound; Turnus does place his foot on his conquered enemy’s corpse, but to do something else: et laevo pressit pede  .  .  . exanimem rapiens immania pondera baltei. It is curious to note that the critic, distracted by the flagrant typicality in Vergil’s scene, failed to remember the text by illusorily extending the Homeric structure right where it has ceased to apply; he has fallen prey to a mechanical association, while the text, on the contrary, has programmed a deviation that represents a significant innovation. The two gestures just specified are in fact foreign to the iconographic typology of Homeric tradition43 and cannot be assimilated with the stereotyped duel: Pallas’s gesture occurs at the moment of his death, while Turnus’s belongs in a different context. It is to this that we now need to turn attention. Turnus treads upon his enemy’s body already dead (exanimem is emphasized through enjambement)44 and strips his baldric from him, a fateful object on which the narrator places decisive focalization. What in Homer was a neutral gesture lacking implications since it was tied to an eminently practical situation like a victorious, warrior’s need to recover his weapons and return to combat, has now been detached from its normal context and synchronized with an emphatic turning point in the action. The recurrence of fixed elements deriving from the Homeric battle scene, apparently aimed at a neutral repetition of the epic code, in reality manages to produce a genuine effect of reading that does not result from some conspicuous modification but is based on a simple mismatch in the recomposition of events.45 One might object that the figurative stereotype

T he D eath of P allas

19

must have a constant significance, guaranteed by the persistence of the epic code as a system of general reference. But it is precisely the code’s univocal fixity that is thrown into crisis, for Vergil as for his public. At points like these, the story overpowers the communicative structure of epic, causing slippage in its code of signs. “And shouldn’t you trample on Ajax, even now that he’s dead?” Agamemnon asks Odysseus in Sophocles’ Ajax. The speaker is a tyrannical character conveying obsolete values; Odysseus, expressing the new common moral, answers: “ignoble gain, son of Atreus, brings me no cheer” (1348–­49). To trample a conquered enemy, ἐπεμβαίνειν, now denoted, by metaphorical extension, arrogance and violation of rights: Scipio after Zama is praised that “he did not trample the fortune of Hannibal” (Plut. Flam. 21.2). Even propaganda and figurative arts in the Hellenistic-­Roman world shied away from this image of triumph (which in other civilizations, the Hebraic and Egyptian, is entirely normal): in fact, it becomes a negative model, a reversal of Roman clementia and an antithetical image in opposition to Horace’s iacentem lenis in hostem.46 Turnus’s gesture thus activates a double intertextual recall, at once maintaining its surface hold on the Homeric model and also producing a new connotative significance, whose plane of expression is constituted by the narrative denotation originally given in the epic code. It is clear that this multiplication of codes of reference yields more information than is transmitted with the poetic text. From this the audience derives an implicit but clear element of judgment about Turnus’s behavior, and hence its comprehension of future events is oriented, in accordance with the entire system of cultural expectations. The brief scene that follows the death of Pallas plays a decisive role in the plot of the Aeneid. Just as Hector seals his own fate by killing Patroclus and despoiling his corpse, so Turnus will be killed by ­Aeneas to avenge Pallas. Hector dons, as a tangible sign of his victory, the divine armor of Achilles he stripped from Patroclus; Turnus stripped Pallas’s baldric and wears it as a symbol of triumph. This basic parallelism takes shape in a grand narrative sweep that joins books 10 and 12 of the Aeneid, which is made to resemble the relation that ties books 16 (“Patrokleia”) and 22 (“Death of Hector”)

20

C hapter 1

in Homer. One victim, one killer doomed in his turn, one avenger: but how is this thematic continuity realized concretely between the deed of Hector/Turnus and the vengeance of Achilles/Aeneas? In Vergil the theme of the baldric plays a fundamental role. By stripping it from Pallas’s corpse, Turnus unwittingly provokes his own end: its depicted figures flash into Aeneas’s eyes like “reminders of wild grief ” (12.945) and unleash his fatal vengeance. For this important connection to be read with sufficient clarity, the theme of the baldric needed to be elaborated in advance, from right back at the moment when Turnus, obviously unaware, snatches it as a material symbol of his victory. And the scene, as in the part already seen earlier, is charged with signals that prefigure the narrative development: Turnus’s boasting insult haud illi stabunt Aeneia parvo / hospitia makes salient the theme of the bond of hospitality and friendship between Aeneas and Pallas and thereby prepares the need for Aeneas’s reaction;47 the divine scene with Hercules and Jupiter functions as prolepsis to announce the necessary and fatal bond between the ends of Pallas and of his slayer; even the “type scene” of slaying and the euchos, as mentioned earlier, conveys an iconic model negatively connotated.48 Behind all these individual connections looms a general recall of the Iliad, an intertextual script that in itself puts tension on the reader’s interpretive freedom (unlimited, if you will, but only in the abstract); in this sense the Iliad appears as a single, gigantic exemplum that with its constraining force helps readers choose relevant meanings in Vergil’s text and even motivates the very destinies of its characters. What is more, even the description of the baldric is not simply a moment of deferral in the epic action; it produces a tragic consonance with the narrative situation. As Conte has observed, the slaughter of young men on their marriage night alludes, through cultural metonymy, to the pitiful end of youths like Pallas:49 the specific condition of ἄγαμοι is only an emotionally charged instance in the general category of ἄωροι, youths who die before their time. At the same time, if we look at the connection between the object and Turnus’s fate, in addition to the baldric’s figurative content, even the language in which it is first mentioned seems relevant and full of implications:

T he D eath of P allas

21

. . . rapiens immania pondera baltei impressumque nefas . . . [snatching the baldric’s immense weight and the crime inscribed on it] The object’s introduction is broken up into a neutral, denotative aspect that is nevertheless stylistically solemn (immania pondera baltei), and an aspect that valorizes one particular quality of the baldric: the presence on it of a crime, the representation of the Danaids’ wedding-­night slaughter. With this, Turnus’s deed is immediately imbued with a meaning that transcends the context and ironically eludes his own conscious awareness: in stripping the baldric he takes ownership of the nefas inscribed upon it—­just as Aeneas, by taking upon his shoulders the shield telling Rome’s destiny, unknowingly dons famamque et fata nepotum. But the nefas cannot but call down, as in king Latinus’s dark prediction, a corresponding punishment: te Turne, nefas, te triste manebit / supplicium (7.596–­97). Pallas’s baldric, now that it has become a part of the scenery, is predetermined to resurface as a tragic reversal of knowledge, as an anagnorismos: . . . infelix umero apparuit alto balteus et notis fulserunt cingula bullis Pallantis pueri, victum quem volnere Turnus straverat atque umeris inimicum insigne gerebat. Ille, oculis postquam saevi monumenta doloris exuviasque hausit . . . (12.941–­46) [The unhappy baldric appeared high on his shoulder—­the belt of young Pallas gleamed with its familiar ensigns, the one whom Turnus conquered, wounded, and laid low—­he wore his enemy’s emblem on his shoulders. When his eyes drank deep the spoils, reminders of savage grief . . .] Aeneas will see at the crucial moment saevi monumenta doloris / exuviasque—­that is, one more time a single object is split linguistically, in the usual binary mode,50 into a flat designation (exuviae)

22

C hapter 1

and an accompanying expression characterizing the object according to its function in the story, in relation to the subjectivity of characters. Infelix balteus, inimicum insigne are polysemic indications that seal Turnus’s fate, by defining it as a recoiling upon himself of Pallas’s enmity and wretchedness.51 The effectiveness in dramatic terms of the “theme of the baldric” offers an obvious analogy with the Iliadic theme of Achilles’ armor. But we are not dealing with any simple textual repetition, since Vergil is not drawing on something already fully realized but instead brings to light a potentiality in the Iliad’s plot that had remained latent and uncertain. It is useful here then to reconsider how the well-­known Homeric events are articulated: a. Patroclus’s armor is stripped by Apollo (16.793–­804), leaving him γυμνός (815) when he is struck in the back by Euphorbus: this gives Hector his chance to finish him off as he falls back, wounded. Patroclus himself before dying insists that “Zeus and Apollo” took the armor off his shoulders. By this point what is obviously lacking is the “type scene” of stripping the corpse—­inevitably, since exceptionally the fallen is already disarmed!—­and we still are not aware of what has happened to Achilles’ armor.52 b. The following episode, the deeds of Menelaus, is a struggle to defend Patroclus’s body and his armor (17.91ff.): we suddenly learn afterward, however, that Hector has “taken” the arms of Achilles from Patroclus (17.125), and now Hector boasts that he stripped them “rightfully” from his fallen foe and retires from the fray momentarily in order to don them (184ff.). c. Now comes an important point of emphasis: the narrator reiterates that these are the divine arms Peleus gave Achilles and, even more, adds a remark by Zeus, who reveals Hector’s fate and deplores his impulsive and ill-­considered exultation (17.198–­208). Poor Hector does not know his own death is near, as he is donning the mortal armor of a great hero and has slain his valiant comrade and taken the arms οὐ κατὰ κόσμον. Zeus’s intervention declares, in the most authoritative way within the Homeric code, the motivation for the events

T he D eath of P allas

23

that occupy the later parts of the Iliad, exploiting the “theme of arms” as a means of contrast. d. Now we know, after an elaborate preparation not without contradictions, that Achilles will see his own armor turned against him: but then this theme ceases to play any appreciable role in the action.53 The loss Achilles has suffered becomes, if anything, occasion for a delaying episode (“the forging of the shield”), but the lost armor is spoken of only in incidental and unspecific ways (as when his horse, Xanthus, says “the Trojans have taken the armor from Patroclus’s shoulders,” 19.412). Then comes the fatal moment when Achilles is taking aim for the blow that will kill Hector: the desired spot at the base of his neck has been left uncovered by the fine armor “that Hector seized when killing brave Patroclus” (22.323). The expression is reproduced from 17.187, in the scene foretelling Hector’s fate, but it is not enough to provoke an effect of dramatic reversal. Just a single, brief focalization, captured in Achilles’ boast Ἕκτορ ἀτάρ που ἔφης Πατροκλῆ’ ἐξεναρίζων σῶς ἔσσεσθ’ [“Hector, surely you claimed while stripping Patroclus that you’d be safe”] serves, not by chance, as the track for the opening of a crucial boast in Vergil’s poem, the last word of the protagonist Aeneas before the ending: Tune hinc spoliis indute meorum / eripiare mihi? I was referring to this link when I mentioned the “potentiality” of the model that Vergil is putting forward and transforming. The resemblances are evident: tune recalls the emphatic interrogative Ἕκτορ ἀτάρ που (in the same opening position), and spoliis indute meorum is a sort of gloss on ἐξεναρίζων; but the simple Homeric reference, compressed in the common and technical ἐξεναρίζειν (“to strip away the armor / slay”), has become the emotional core of the entire Vergilian scene. On the contrary, the Homeric expression puts no “subjective” emphasis on whose particular arms were stolen. Even later—­for instance, when Hector’s body is being stripped—­no stress is placed on the fact that the armor belongs to Achilles.

24

C hapter 1

To sum up, the story of Achilles’ armor seems strongly rooted in the narrative continuum of the Iliad (even the most rigorous of Analysts have never considered it separable from the poem54), but it is realized in a very disjointed manner. The only time the fact that Hector dons his enemy’s armor is pointedly thematized and tied to the plot development is in Zeus’s remark in book 17, contained in a brief episode that seems constructed ad hoc: before this scene there is no description whatsoever of how Hector has obtained the arms, and the type scene that should follow the killing is omitted; later, when after a rather extended narrative the armor theme might be brought to the foreground (in book 22), it has been reduced to a weak echo. The Vergilian reinvention has begun from this fragmented sequence, and concentrates into a single narrative unity, aptly highlighted in context, what in Homer emerges episodically in the long flux of the story: the killing, the stripping of the body (omitted in the Iliad, as we have seen), the victor’s triumphant exulting as he dons the spoils,55 the comment that reveals his own ruinous blindness. In Homer it is Zeus who, surely by virtue of being the “spirit of the story,” unmasks Hector’s fate and renders predictable, even necessary, his final destruction. This episode’s narrative function is clear enough, but what does not emerge with equal clarity is the moral judgment, the commenting function implicit in Zeus’s speech. In the god’s words Hector has slain and despoiled Patroclus οὐ κατὰ κόσμον (17.205); this is certainly a negative evaluation, but this completely isolated expression, from which too much has often been gleaned,56 does not shed much light on the events. (Perhaps we might take this evaluation along with the rather anomalous circumstances in which Patroclus is slain, as we have seen: a divinity and two warriors collaborate in the deed. But the internal reference, if so, is not among the most explicit.) Unlike the original audience envisioned for Homeric poetry, later readers no longer have access to an anthropological code that makes Zeus’s words fully comprehensible. What is more, the context of the Iliad does not cooperate in our reading: not only does it lack anything like the “modern” notion of guilt, but in particular a warrior’s death and the appropriation of his armor might have been considered as normal heroic behavior, an action that can naturally arouse vengeful reactions, but that no one in the name of heroic morality would dream of condemning. This

T he D eath of P allas

25

will be (residually) the ethic of Mezentius: nullum in caede nefas (“there is no crime in slaughter”). 4 .   Hom e r ic Com m e n ta ry (10.501–­5 05) Τὸν δ’ ὡς οὖν ἀπάνευθεν ἴδεν νεφεληγερέτα Ζεὺς τεύχεσι Πηλεΐδαο κορυσσόμενον θείοιο, κινήσας ῥα κάρη προτὶ ὃν μυθήσατο θυμόν· “ἆ δείλ’ οὐδέ τί τοι θάνατος καταθύμιός ἐστιν ὃς δή τοι σχεδὸν εἶσι· σὺ δ’ ἄμβροτα τεύχεα δύνεις ἀνδρὸς ἀριστῆος, τόν τε τρομέουσι καὶ ἄλλοι· τοῦ δὴ ἑταῖρον ἔπεφνες ἐνηέα τε κρατερόν τε, τεύχεα δ’ οὐ κατὰ κόσμον ἀπὸ κρατός τε καὶ ὤμων εἵλευ· ἀτάρ τοι νῦν γε μέγα κράτος ἐγγυαλίξω, τῶν ποινὴν ὅ τοι οὔ τι μάχης ἐκνοστήσαντι δέξεται Ἀνδρομάχη κλυτὰ τεύχεα Πηλεΐωνος.” (Il. 17.198–­208) [But when Zeus who gathers clouds saw from afar {Hector} donning the armor of the godlike son of Peleus, he shook his head and spoke to himself: “Ah wretch, you have no idea of the death that is right upon you, putting on the immortal armor of the best man, who makes others tremble. And you killed his comrade, kind and strong, and stripped the armor from his head and shoulders not according to order; and for now I grant you great strength, a compensation for these things, that you will not return from battle and Andromache will not receive the glorious arms of the son of Peleus.”] Zeus’s apostrophe from a distance establishes and sanctions the narrative bond between the deaths of Patroclus and Hector; the corresponding element in Vergil is an intervention made directly in the voice of the narrator: Nescia mens hominum fati sortisque futurae et servare modum rebus sublata secundis! Turno tempus erit, magno cum optaverit emptum intactum Pallanta et cum spolia ista diemque oderit.

26

C hapter 1

[The mind of men is ignorant of fate and coming events, unable to preserve the mean when inflated by favorable affairs! A time will come when Turnus will wish he had paid a great price for Pallas to be untouched, when he will hate that day and those spoils.] This freedom to offer commentary on events from the outside certainly does not fall within the communicative rules of epos; perhaps for this reason criticism has been more concerned to explain the insertion’s contextual function than to track down any possible literary matrix for it. Heinze (1915, 371–­72) focused on authorial interventions in Homeric epic with the aim of showing that Vergil fully adheres to and does not depart from epic objectivity except in cases where the Homeric model already offered some “authorization.” This theory, perhaps attempting to exonerate Vergil from Romantic criticism (according to which the epic author can only disappear behind his work “like God in the universe”),57 does not find secure support in the Iliad itself.58 Elsewhere Heinze admits (1915, 226) that this statement ex persona poetae is needed to give the episode of Pallas’s death the required importance, but at the same time he accuses it of “violating the poetic rule” (der künstlerischen Regel entgegen) because it lacks immediate consequences for the action. Here too a normative vision seems to prevail, according to which Homer’s paratactic narration is a perfect and unattainable model for all epic. On the contrary, one might observe instead that narrative prolepsis in a wider sense, founded on the narrator’s ironic superiority to the characters within the action, is not lacking even in the tight objectivity of Homeric storytelling. If in the Patrokleia these foreshadowings are limited to brief pathetic apostrophes fully internal to the development of action, a case of greater narrative distance is nevertheless met with in the Odyssey. Turnus dons Pallas’s spoils unaware that it is bringing his own death nearer, just as the suitor Antinous hopes in his heart to succeed at stringing Odysseus’s bow and so win Penelope in the trial of the axes; “instead he would be first to taste the arrow” from Odysseus’s hands, the owner of that weapon that will prove fatal to the Suitors (Od. 21.96–­99). In both cases a fateful object—­the bow, the baldric—­is put to use by the poet to swiftly turn the narrative perspective toward a telos still

T he D eath of P allas

27

unpredictable—­even if in Homer the relation with the object will not be a specific cause, as in Vergil, of the character’s death. More generally, it is clear that the dimension of prophecy and Nachgeschichte is not in itself incompatible with epic: the Iliad even foretells the death of Achilles, which falls outside the poem’s narrated time. But Vergilian prolepsis is not confined to anticipating Turnus’s future; it comments on it, making him into an exemplum for the human condition. This is why other interpreters have labored to clarify the particular subjectivity of this intervention,59 and have abandoned the comparison with epic tradition and resort instead to tragedy, a genre that continually interweaves moral commentary and generalizations into the action with no concern for the limits of “objective” narration. It is said, with some generality, that it is a “choral”60 intervention or an “aside.”61 Somewhat greater resemblance exists, I think, with the type of deviation in dramatic discourse that jumps from the concrete situation in order to speak as though the listener is not a character in the scene but a collective, if imagined, entity;62 something, that is, falling somewhere between normal stage speech and jokes ad spectatores, though this is of course only a lexical analogy. Often such moments of reflection open with an epiphonema, which serves to signal a shift in tone and movement to a generic interlocutor, and appeals to universal deficiencies of human nature: ἀλλ’ ὦ μάταιοι (Eur. Suppl. 549); ὦ κενοὶ βροτῶν (ibid., 744); ὦ ταλαίπωροι βροτῶν (ibid., 949; cf., for example, El. 383–­84, Ino fr. 419 N.2). This is perhaps one root of the sententious opening nescia mens hominum in Vergil: it is an intradiegetic intrusion stronger than the usual epic address to characters, but still not such as to break fully the narrative illusion with a different communicative relation that trespasses into a metaliterary dimension (as indeed happens with genuine appeals to the reader). This narrative prolepsis foretelling Turnus’s end by commenting on the lack of moderation and foresight that characterizes mortals could have been achieved through more traditional, Homeric models. In the Iliad, Hector’s fate is foretold in the voice of other characters both human (like Patroclus as he dies) and divine (the apostrophe of Zeus, which served as Vergil’s point of departure): anticipation is given without outside intrusion, without the story ceasing to tell itself on its own. In Vergil’s text, instead, the story (for

28

C hapter 1

a moment) becomes discourse: as opposed to the ironclad Tempuserwartung that the epic genre cultivates in its readers, we encounter here future events—­Turno tempus erit  .  .  . optaverit63—­managed directly in the narrator’s own voice. This exception has gone unobserved by commentators on the Aeneid but is well known in manuals of syntax and in studies of the Latin verb, which tend to see in this isolated instance the first poetic example of a distinct temporal category in its own right (though in truth one fairly heterogeneous)64 called the “historic future.” And on closer look, it does not lack the other constituting element that defines the discourse with respect to history: the one speaking—­the one in control of the epic narration—­refers to the fatal spoils Turnus has gathered up with a demonstrative that automatically evokes a dialogic dimension, ista, where we might expect a less discursively marked form of anaphora. It is an ostensive gesture from narrator to audience that founds a direct communication by momentarily halting the course of narrated events. (As one might suppose, iste recurs elsewhere in the poem only in the direct discourse of characters.65) As such, this incursion of discourse into the narrative is realized with specific language that signals the passage from normal narrative writing to an immediate connection (even if not expressed on the surface) between narrator and audience. Not only does it address the need to reveal the origin of a fundamental line running through the overall narrative plot; more importantly, it also establishes for the audience a cultural norm that enables them to examine critically the characters’ actions. The text’s adoption of this norm is secured by a powerfully evocative means: the surge of direct speech from the epic narrator, who possesses an elevated, albeit hidden, level of awareness within the sphere of literary communication. The narrator’s intervention therefore appears as an explicit comment on and motivation for the various forms of implicit commentary already examined: the montage of events, the superimposition of diverse codes of reference, the relay of intertexts. This sort of deterministic relation does not need to be repeated on every occasion (indeed, continual commentary of this sort would be uneconomical and would ill-­suit the workings of epic action). What matters is that once, and once and for all, the interpretative process desired by the author has been activated: readers will thereby know how to reproduce it in their reading and

T he D eath of P allas

29

when it occurs will enrich their own referential understanding of the events. This mode of signifying, however, is subjected to an important constraint: the motivation that the text proposes needs to be plausible—­that is, it must resonate fully with models that the culture itself, at a certain phase, puts forward. For this, we need to look for evidence outside the text of the Aeneid. Clearly, the value of “servare modum in victory” is what the commentary activates and for which it aims to obtain immediate consent from its Roman readers. (It is easy to detect here, beyond a general need to overdetermine the narrative, the imprint of a poetics of plausible realism, one that can no longer settle for bare predestination but rather needs to infuse the structures of ancient epos with novel expectations.) Here it is not necessary to recall the whole philosophical tradition (Aristotle, Panaetius, and so on) that treats moderation in success as a fundamental virtue of the megalopsychos, and makes metriotes a basic value of human society. More interesting for our purposes are the “average” expressions of this concept, its translation in pragmatic situations at Rome—­like the famous beginning of Cato’s pro Rhodiensibus (fr. 163 Malc.4): scio solere plerisque hominibus rebus secundis atque prolixis atque prosperis animum excellere atque superbiam atque ferociam augescere atque crescere . . . secundae res laetitia transvorsum trudere solent a recte consulendo atque intellegendo . . . (“I know most men in success and prosperity become exalted in spirit and grow excessive in pride and haughtiness . . . prosperity is apt to turn men in their joy aside from the path of deliberating and thinking rightly”), where the ethical-­ political concept of superbia is fixed in relation to the dangers of res secundae—­and in interpretations of contemporary history (on the theme of res secundae the proem to Sallust’s Historiae readily comes to mind). It is no accident that in Livy (30.42) this virtue is used to explain, and naturally also to celebrate, Rome’s historical victories: populum Romanum eo invictum esse, quod in secundis rebus sapere et consulere meminerit (“the reason why the Roman people are unconquered is that they have the power to stay sober and wise in favorable times”). The abstract, synchronic model, being highly recognizable and adaptable to any use, was also transformed into a “narrativized” evolutionary scheme: the peril of not knowing how to master one’s own good luck, and thus undergoing moral degeneration, offered

30

C hapter 1

an interpretation of well-­known historical events and explained the fate of great personalities that did not in some degree seem to embody its example. I am thinking of the literature on Alexander the Great, the man who, at a very young age, achieved boundless triumphs sed ignarus quemadmodum rebus secundis uti conveniret (Cic. Tusc. 3.21) and τετυφωμένος ταῖς τοσαύταις εὐτυχίαις (Str. 15.1.5); a man who, at last, super mensuram iam humanae superbiae tumens (Sen. Ben. 5.6.1), revealed himself to be nothing but tumidissimum animal (ibid., 2.16.2);66 a most effective paradigm for any future “mirror for princes.” Moving along analogous lines is the figure of Turnus, a captain who appears tumidus secundo Marte (10.21–­ 22) and superbus caede nova (10.514–­15): his description draws on ethical-­political language67 rather than the warrior ethos and the mainesthai of Homeric heroes, thus serving that need for exemplarity so deeply inscribed in the history of Roman epic. This characteristic trait does not stand in isolation. Excessive elation in success has as its necessary complement another fundamental defect, unrestrained prostration at the onset of the first reversal of fortune: quem res plus nimio delectavere secundae, / mutatae quatient (“He who enjoys favorable circumstances too much is crushed when things change”) (Hor. Ep. 1.10.30–­31). And again, “the base soul exalts itself in good fortune, abases itself in disgrace.”68 The megalopsychos, on the contrary, is neither too happy in success nor too afflicted by misfortune, but shows his greatness by holding up equally well in the two opposing conditions.69 This paradigmatic association, inscribed in the audience’s system of expectations, resurfaces at intervals in the epic construction of Turnus: after his great successes in battle, the first adverse turn drives him to contemplate suicide (10.677ff.), but new hopes of victory overwhelm him again (subita spe fervidus ardet, 12.325) with momentary contentment but then frustration (iam minus atque minus successu laetus . . . , 12.616) that leaves him amens (12.622) and finally—­here the negative model is surpassed, however, in the direction of at least an embryonic form of psychological character development—­lucidly determined to die. The reverse of this model is the Roman ideal of the victorious commander who refuses to imitate the superbia of his own enemies: Aemilius Paulus at Pydna, who taught not to μεγαλαυχεῖν ἐπὶ τοῖς κατορθώμασι (Plb. 29.20) since in secundis rebus nihil in quemquam

T he D eath of P allas

31

superbe ac violenter consulere decet . . . is demum vir erit, cuius animum neque prospera [sc. fortuna] flatu suo efferet, nec adversa infringet (“it is right in favorable circumstances not to treat anyone with pride or violence . . . he will be a man whose mind is neither puffed up in prosperity nor broken in adversity”) (Liv. 45.8.6–­7).70 In these terms the ideal of “knowing how to conquer” was developed by ­Caesar after his seizure of power, and then renewed in Augustus’s propaganda about clementia.71 Vergil is therefore able to cite in summary form, through the simple opposition between res secundae and servare modum, a universally recognized and accepted cultural model: from this an ongoing commentary is cast that allows the reader to ideologize according to his own value system the poem’s “Iliadic” content and participate in foreseeing its developments. Thus commented upon, the action takes on a plausible realism. The narrative produces its post hoc ergo propter hoc by construing itself in reading as a necessary concatenation. But this commentary must remain completely incorporated in the narration if it is not to “kill” (as will happen, for example, in Lucan) the free unfolding of epic events by superimposing itself on them. So the Vergilian text cannot help but permit a host of “voices” to coexist within it (voices linking up in various way with the Homeric texts, of which they represent diverse readings and transformations): it accepts the risk of being multivocal and even exploits, for precise communicative ends, the polyphony of its own cultural codes. Perhaps this was the only way to remake Homer without ignoring the distance that inevitably separated it from the complex, civilized world in which the poet lived. Yet this distance is less insurmountable if, once again, we do not consider the Homeric text as a fixed and immutable object but as a layering of historically diverse readings. Recall now the basic homology linking the two passages quoted at the beginning of this section—­the reason why, despite all the differences between them just examined, commentators have always seen that particular Iliadic passage as Vergil’s model.72 The act of triumphing is what will unwittingly cost the victor his life,73 but still this death is not foreseeable by the human mind: οὐδέ τί τοι θάνατος καταθύμιός ἐστιν—­nescia mens hominum fati sortisque futurae. But this imposes a meaning on the Homeric text that it could not have intended or implied. Homer treats the theme (so dear to archaic

32

C hapter 1

Greek poetry) of the inevitability of fate, but he does not say why Hector is doomed to fall, nor does he offer any positive value as a counterpoint to the hero’s blindness. The event remains objectively under the sign of Ate. Vergil instead speaks of the incapacity to servare modum. This reflection, as modern interpreters of the Aeneid flatly note, “is not in the spirit of Homer.”74 To be sure, the norm of moderation as a rule of life had no place in the ideological world of archaic epos, but Vergil found exactly this concept in current interpretations of his Homeric model. According to Alexandrian exegesis, as far as we can reconstruct it from the scholia,75 Zeus’s interjection on Hector’s fate has a specific moral/didactic content: σχετλιάζων ἐπὶ τῷ ἐπάρματι τῆς εὐτυχίας (rebus sublata secundis) διδάσκει μετριάζειν (servare modum) “by deploring the excess caused by good fortune, he teaches moderation.”76 5.  V e rgi l’s Hom e r So we come back again, as in the former analysis, to the requirement not to separate the Homeric model “interpreted” by Vergil and the history of its reception (which is useful to represent as a stratification of successive readings). In this way we are able to grasp the link between Vergilian writing and the reading of Homer that this presupposes: two different productions of meaning but inscribable within the same cultural system. At base this way of representing the intertextual relation between Vergil and Homer can be deduced from a more general critical principle, certainly not new, but one found more often stated and put to concrete use: the necessity never to separate, in a given cultural period, the modes of reading and of writing. And hazarding an inference of wider scope, we can add that Vergil’s attitude toward his model is not conceivable outside the cultural background in which practices of exegesis and critical commentary had become the sole “authorized” way to gain access to the poetry of the past—­which had by then been turned, in the hands of the philologist-­poets, into a literary object.77 This fact becomes particularly apparent, and could be further illustrated with specific examples, when one observes the coincidence between certain adjustments Vergil made to the Homeric contents and the critical categories through which Homer’s reception was filtered, categories

T he D eath of P allas

33

that originated in Alexandria and Pergamum (one need only think of the rise of the “probable” and “decorum”). From this perspective the transformation of the model coincides without rupture with the genesis of the new text: it no longer appears isolated in a sort of prehistory for Vergil’s work (a distant avant-­texte where the critic is attempting some complicated alchemy)78 but instead acts directly as a communicative strategy. In this way one recovers the necessary connection between the intertextual relation—­the condensation of literary codes in the text—­and the poetry’s orientation toward its audience, real readers who begin from their given context to construct meaning out of it. Indeed, the interpretation the narrator requires of the audience is based on the competence that it already possesses; part of this competence is, by necessity, not an absolute Homer but rather certain particular readings of Homer that have been subjected to the cultural constraints of the period. This way, the project of the poem is to overcome the partial extinction of the anthropological code that had supported the epic genre in its archaic phase (the original sense, as we know, is very often more fluid and transient than some surface structures); replenishing Homer’s gaps with a new culture means that the play of literary signification has been reopened. Because the audience has already received the events of Hector and Patroclus according to certain mental categories—­having very little to do with the genuine intention of the Homeric text, but that is a secondary matter—­the textual allusion serves not just to project a “script,” a known field of reference, but also to activate a process of ideological interpretation of the events narrated. From the moment the prodigy of bloody rain had been read as “tears of Zeus,” and received as an extreme symbol of (“Stoic”) subordination of divinity to Fate, Vergil can start from this interpretation, instead of from a literal reading of the model, and propose to his readers a god who has already learned this lesson. (Note then that, paradoxically, the inversion of the model can be its most historically faithful reading.) If, finally, a typical and formulaic gesture in Homeric narrative has long since been given an unavoidable transvalued connotation—­ancient culture stigmatizes trampling conquered enemies as a sign of inhumanity, and so this image can no longer be received in some mechanical association with its

34

C hapter 1

original context—­the narrative economy of the new text will rely on this very additional meaning to engender wide-­ranging implications. (Besides motivating Turnus’s death, it reminds readers of the jarring effect of Aeneas’s “non-­Homeric” behavior toward Lausus). The competence of the audience is therefore a constituent part of the meaning: reconstructing it is useful for understanding the text. For us this involves frequent resort to evidence and clues about ancient Homeric interpretation, but not with the goal of painting the improbable image of a bookish Vergil laboriously tracking down in commentaries and scholia anything he did not already find in Homer. It just means looking for evidence and clues that point toward something else: the general reception of the Homeric text in Roman culture. This is where the system of presuppositions is rooted, and with it the intertextual relation that ensures Vergilian epic its legibility.

Ch a p t e r 2

THE STRUCTURE OF AENEID 10 1. Intertextual relation, legibility: this question has arisen several times already and calls for fuller examination. What function does a literary model play in the legibility of a narrative text? The task is to discern to what extent the presence of a model cooperates in making legible the narrative implicating it, how it serves to guide the reader, and aids in making sense of the text. This takes us well beyond the realm of the “art of allusion” as it is generally understood. We are concerned with not just occasional contacts between single speech acts but also with entire implicit scripts and plot lines that the model projects upon the new text and its reading. Up to now our examination focused on a rather short textual passage and sought examples of how Homer was “read” and transformed by Vergil. Now I want to show to what degree Homer comes into play for reading the plot. This time we will engage a sufficiently long stretch of text, and by focusing on a larger narrative arc, we will be able to discover how the complex process unfolds and at the same time measure (not in terms of single allusions but in whole narrative movements) the presence of the Homeric model as a trace of sense and aid to narration. I will devote this morosa intentio to reading an entire book: the tenth. In reading a story, as we know, everything takes place—­on the surface or, better, at first pass—­as a linear passage during which 35

36

C hapter 2

meaning is progressively accrued according to the text’s apparent organization. Meanwhile, what happens as reading continues is that what has been recounted rearranges in memory and the scroll transforms into a field of significations and organized expectations. The syntagmatic disposition (the expansive series of meanings) gives way to a new paradigmatic ordering of contents assimilated to one another through reading. In this passage between syntagmatic decipherment and paradigmatic comprehension a literary text reveals its own legibility. We will first observe how this process happens in the concrete, through what procedures (as well as what conventions) the narrated text—­in presenting itself as narrative, an ordered succession of events—­acquires the status of full-­fledged model, endowed with particular and specific sense. We will pay particular attention to two functions: the first is a certain relation of the text “to itself,” in the sense that a narrative text can generate paradigms through repetition, parallelism, contrast—­in short, by precipitating in reading a principle of recursive regularity: this is the text’s visible organization. The second concerns the intervention of intertextual models underlying the text that orient reading and spread out a horizon of expectations. 2. A summary of the narrated events will be useful. Book 10 begins, according to a specific epic ritual inherited from the Iliad,1 with a long council of the gods in which first Juno and then Venus argue their opposing cases. The council ends with Jupiter resolving that men can defend their own fortunes without divine intervention for the duration of the day of battle; what Fate wills to occur will coincide with the unfolding of human deeds shifting from one side to the other.2 118.  The siege of the Trojan camp continues, but Aeneas is on his way back, bringing along the thirty ships of the Etruscan coalition. Vergil provides a brief catalogue of them, opened by an invocation of the Muses; he reminds us of the friendship between Aeneas and Pallas, a brief mention whose function will be evident later. While Aeneas is leading by sea the Etruscan army, he is met by sea nymphs who used to be his fleet and inform him of

THE STRUCTURE OF AENEID 10

37

the dangers the Trojan people are facing. Aeneas quickly approaches the battlefield and fighting ensues with the Rutulians. 362.  After some successes by Aeneas (who thus inaugurates his participation in the war), Pallas appears at the head of the Arcadians and excels in combat; in contrast to Pallas’s aristeia we have a very brief aristeia of Lausus, young son of the Etruscan tyrant Mezentius. But the two young men are not destined to fight each other. Shortly thereafter, in fact, Turnus, replacing Lausus, confronts Pallas and challenges him. Pallas is too inferior, in age and strength, and his divine protector, Hercules, cannot intervene (because of the law established by Jupiter at the beginning of the book). Thus the young warrior falls, killed in his first day of battle, and Turnus strips him of the decorated sword-­belt. 510.  Learning of this, Aeneas unleashes true carnage, with much more wild violence than in his previous aristeia.3 The scale of the war thus tips in the Trojans’ favor. 606.  Meanwhile Juno secures from Jupiter a brief respite for Turnus. Using a ruse, an apparition in the form of Aeneas, she draws Turnus away from the fray and onto a ship, which carries him away from the battlefield for a temporary period of inactivity. 689.  At just this moment Mezentius takes the field and makes a counteroffensive; his aristeia, described with great emphasis, balances Aeneas’s and levels out the fortunes of war. Aeneas then attempts to eliminate Mezentius, but young Lausus intervenes and takes his father’s place as he retreats wounded from the fighting. Aeneas would like to avoid the unequal match, but when rashly assailed he cannot help but react and kills Lausus. After Lausus dies, Mezentius feels vengeance as his only reason to live. Determined to die, he mounts his horse and charges, then rides around Aeneas with a hail of volleys; but Aeneas unhorses and kills him. The book closes with the dying breath of one of Aeneas’s great adversaries, and in this way casts its gaze forward, through compositional parallelism, toward the poem’s end, the death of Turnus. At the same time, through a device not uncommon in Vergil, the closing image—­the arms of Mezentius drenched in blood—­seems to recapitulate and synthesize the dominant tone of the whole book.

38

C hapter 2

3. In this complex series of narrative events we can in fact distinguish—­at the descriptive level usually called “plot”—­a unified thread that is fairly clear and simple. Leaving aside the opening divine council (which of course is not without purpose, since it sanctions the conditions in which the human characters will act), all the main events narrated in book 10 are directly tied to the protagonist Aeneas, and to his relations with antagonists and allies. From this perspective, two episodes especially catch the reader’s attention and memory: the duel between Turnus and Pallas, where Aeneas’s antagonist kills his closest ally, and the final clash where Aeneas kills first Lausus, son of Mezentius, then Mezentius himself, Turnus’s chief comrade-­in-­arms. These two moments are placed in relief by a set of rhetorical and narrative procedures that deserve careful analysis. But for now we prefer to take their importance in terms of plot as a given and examine instead the narrative structuring that connects them. We may begin by noting that the entire narrated action is articulated as a series of duels, hand-­to-­hand combats in Homeric fashion, standing out in a continuum of “collective” actions: alternating androktasiai, scenes of mass slaughter, and aristeiai, where a strong warrior, a main character in the story, kills or wounds less important foes, often characters brought in ad hoc and given only a “passive” role. (Even if Vergil famously tends to affectively qualify many of these “extras,” so that only rarely can we speak of these minor figures as a mere Funktionswert. After all, Homeric poetics—­though far from coherent on this point—­already supplied various meaningful antecedents for pathetic stylization of deaths in battle.) It is clear that much of this material constitutes a sort of “free variation” around the main duels, the plot’s real articulation points. There is then, even among the main hand-­to-­hand combats that oppose pairs of important heroes, a certain hierarchical function. The whole plot of Vergil’s Iliadic books is constructed in such a way as to render the duel between Aeneas and his chief antagonist, Turnus, predictable and necessary—­analogous to how the Iliad prepares and loads with significance the final combat between Hector and Achilles. So too within book 10, as in the two following books, a duel between Aeneas and Turnus seems like the most predictable and

THE STRUCTURE OF AENEID 10

39

anticipated resolution: Aeneas and Turnus constitute a “proper” pair of adversaries. Yet, in another obvious recall of the Iliad, this resolution is long delayed and the course of action takes continuous detours: first Aeneas’s absence from the Trojan camp, then ­Juno’s ruse that removes Turnus from battle, later the appearance of “substitutes” for Turnus (like Mezentius himself ),4 and finally the broken pacts in book 12. In book 10, therefore, the expectation of a confrontation between the two proper adversaries and rivals is created only in order to be postponed and deferred. What happens instead is that both Turnus and Aeneas separately engage in important duels, each of which is emphatically marked in the narrative fabric.5 It is easy to see that the prominence of the two episodes—­the duel between Turnus and Pallas and Aeneas’s clash with the pair Lausus-­Mezentius—­is guaranteed first and foremost by the order of composition. The episode where Aeneas plays protagonist has the obviously preferential role of closing the whole book; the duel between Turnus and Pallas occupies roughly the center of a narrative section that can be thought of as opening when Aeneas comes on the scene (260ff.) and closes with Turnus’s disappearance—­up to 10.688, patris antiquam Dauni defertur ad urbem—­where Turnus is returned (not by chance in his father’s name) to a space of relative safety; the following, final section will commence with the name of Mezentius, the character who, named emphatically in the first verse (689), becomes the protagonist down to the final verses that describe his end. At the same time, the two episodes are set in high relief because they are narrated in extenso, with a slower rhythm and attention to detail that other battle scenes do not earn. (However, they are not developed in identical registers. The duel between Turnus and Pallas is one of the most typically “Homeric” scenes in the entire poem, dominated by a pseudo-­formularity that Vergil surely felt was adequate to the heroic material;6 the clash between Aeneas and the two Etruscan warriors abounds instead with nontraditional motifs, or at least ones that we might call extra-­Homeric. But what interests us here, for the moment, is that both episodes derive emphasis from their respective stylistic registers.) This emphasis contributes to the strong bond between the two scenes, with immediate thematic consequences for the story.

40

C hapter 2

At the end of the book Aeneas kills Lausus and therefore must contend, right afterward, with Mezentius; Mezentius, after his son’s death, has no other reason to live than for immediate vengeance (and his intention is expressed in a way rather like the language of Achilles as avenger of Patroclus); but along this line the central episode, the death of Pallas, remains rather incomplete, deprived of its “natural” development, the intervention of an avenger to confront Turnus. Such a possible development is frustrated, as noted before, by a supernatural intervention. In this way the bond between the two episodes is marked by a sort of privative opposition. On the other hand, in the entire succession of events the poet invites us to grasp a fatal collision of fates. The “fair” fight between Aeneas and Turnus does not, as we come to see, materialize; but neither does another possible fair and balanced confrontation, between Pallas and Lausus: Hinc Pallas . . . Hinc contra Lausus, nec multum discrepat aetas, egregii forma, sed quis fortuna negarat in patriam reditus. Ipsos concurrere passus haud tamen inter se magni regnator Olympi; mox illos sua fata manent maiore sub hoste. (10.433–­38) [Here Pallas . . . against him Lausus, neither one much different in age, each fine in form, but fortune denied them their homecoming. Yet the ruler of great Olympus did not suffer them to confront each other; their fates awaited them soon at the hand of a greater foe.] For a moment the narrator nearly reveals the constructive artifice that aims to make both characters mirror a single model: youths of courage and beauty (a trait not lacking in aestheticism but connected especially with sorrow for the ephemeral, given that beauty is transvalued, after the fact, by premature death) marked by an adverse fate; opposed only because they belong to two enemy armies and (implicitly) have very different father figures. This latter trait recalls another: Pallas and Lausus are beloved only sons of two fathers who want to but cannot protect them. So this parallelism between

THE STRUCTURE OF AENEID 10

41

the two youths introduces a significant displacement. Both characters are diverted into struggle with a stronger adversary, and in both cases the text emphasizes the unequal character of the conflict:7 Pallas fights viribus imparibus (v. 459) and Aeneas calls to Lausus: quo moriture ruis maioraque viribus audes? (811).8 This meticulous list of narrative parallels would be of little use if we did not add that parallelism is the most efficient means to make the differences between two narrative situations systematic and meaningful. Examined in this light, the two scenes—­the deaths of Pallas and of Lausus—­can be broken down into a set of contrasts: a. Turnus goes looking for the “unfair” fight with Pallas; he replaces Lausus (succedere Lauso, 439) and drives away his comrades, saying: solus ego in Pallanta feror, soli mihi Pallas / debetur (442–­43). While Turnus attacks and challenges Pallas, Aeneas, conversely, is attacked by Lausus; Lausus assaults Aeneas (ruis, 811) and is even warned not to persist in this folly. Aeneas suffers, while Turnus provokes, the “imbalanced” combat. b. Turnus wishes that Evander were there to witness his own son’s death (cuperem ipse parens spectator adesset, 443); Aeneas instead is touched by filial love for Lausus (fallit te incautum pietas tua, 812) and moved by the boy’s dead body to reflect on paternal love (et mentem subiit patriae pietatis imago, 824). c. Turnus performs all the typical actions of the “Homeric” victor over the body of his beaten enemy: he speaks a euchos, a boasting speech to provoke his enemies; he sets his foot on his slain foe; he strips his armor—­more specifically, the baldric that will become his downfall. Aeneas’s behavior toward Lausus is the exact reverse of this epic stereotype. Instead of vaunting over or mocking his foe, Aeneas speaks a short eulogy of Lausus; instead of stripping them, he leaves the arms that had made Lausus happy (arma quibus laetatus habe tua, 827). d. Aside from this difference, both victors avoid “uncivilized” treatment of the bodies of the defeated9 and instead return them, though with attitudes only superficially parallel: Qualem meruit,

42

C hapter 2

Pallanta remitto. / Quisquis honos tumuli, quidquid solamen humandi est / largior (“I am sending Pallas back to him, the way he deserves to have him. Still, I bestow such honor as comes with a tomb, and what solace burial brings.”) (Turnus); . . . teque parentum / manibus et cineri, si qua est ea cura, remitto (“I am sending you back, if this care is still felt, to the Manes of your parents and to the ashes.”) (Aeneas). Without making Turnus do anything truly “barbaric,” Vergil does not fail to differentiate him here as well from the protagonist Aeneas.10 As we see, parallelism and repetition—­through the “type scenes” that the epic repertoire provides—­end up depositing a system of differences that individualize the characters in action, expressing their nature through specific behavior. But insisting on the characterizing function of these methods is not enough. Parallelism also has an immediate relevance for deciphering the plot: at the risk of a somewhat schematic view, it is worth noting that the successive duels are not just strung together in a linear extension but also gather together to form a unique under­ lying paradigm: • 

 e opposition between Turnus and Aeneas is deferred by a kind Th of initial disruption, the interposition of the pair (which is also fair and balanced) composed of Pallas and Lausus.

• 

 en we have a pair of “unequal” duels, Turnus with Pallas, Th ­Aeneas with Lausus—­in both cases, predictably, the stronger and more experienced combatant prevails.11

• 

 ut the conflict between Aeneas and Lausus provokes an immeB diate vengeful reaction on the part of Mezentius, who cannot allow his son’s death to go unpunished.

• 

 t this point the parallelism ends. The conclusion of the duel A with Mezentius leaves a void, so to speak, a blank that reader expectations are invited to fill with meaning; the death of Pallas remains unavenged, without the compensation of a “fair” fight; but vengeance is predictable and necessary, reclaimed by the contextual system.

THE STRUCTURE OF AENEID 10

43

Repetition and parallelism trace out a narrative scheme that defines an empty, asymmetrical space: the very space into which the story’s definitive resolution will plummet, the duel between Turnus and Aeneas in book 12. “IMPAR PUGNA” “AEQUA PUGNA” Turnus = Aeneas Turnus against Pallas Ž —————————————­ Lausus = Pallas Aeneas against Lausus Ž Aeneas against Mezentius

4. Meanwhile, finally, book 10 closes with a balanced and “necessary” confrontation, the one between Aeneas and Mezentius. The story presents them as adversaries worthy of one another, and moreover it is clear that their opposition is rooted in certain specific and antithetical characteristics. In this way book 10 can, on one hand, “close” its own formal autonomy neatly, while on the other hand “open” itself toward the text’s continuation. Indeed, the story’s saturation at this point is only relative and partial; it must still accomplish the events that Pallas’s death opens up. Let us pause briefly on an interesting requirement of this narrative structuring: the need for the Etruscan tyrant to appear in the course of book 10 as a worthy substitute for Turnus and a ­worthy antagonist to Aeneas. This requirement is of a piece with the problem of reference to tradition; we can safely assume that in a long line of tradition before Vergil Mezentius played the role of Aeneas’s main antagonist. The Aeneid’s plot envisions instead a marked reduction of Mezentius’s importance. And indeed, in the course of book 10, the amount of time this character holds center stage is relatively short. Vergil solved this dilemma by making Mezentius the dominant focus throughout his time in the story, centering all the reader’s interest on him alone. His aristeia is no less weighty than Aeneas’s (in narrative time and number of victories),12 and the poet mobilizes no less than four consecutive similes to describe it, all linked to the lone figure of the Etruscan warrior and marked by a clear dynamics of crescendo. Mezentius first appears as a rock that stands unassailable against the fury of a stormy sea (693ff.); then as a boar surrounded by hunters, but still fearsome and keeping them at bay (707ff.); when he goes on the attack, he is a ravenous lion,

44

C hapter 2

jaws drenched in blood (723ff.); at the climax of his aristeia, when he is about to confront Aeneas, he is as huge as Orion, the giant who wades the depths of the sea with water up to his shoulders and treads on mountains touching clouds with his head: Quam magnus Orion cum pedes incedit medii per maxima Nerei stagna viam scindens, umero supereminet undas, aut summis referens annosam montibus ornum ingrediturque solo et caput inter nubila condit, talis se vastis infert Mezentius armis. (763–­68) [As great as Orion when he strides through the midst of the deepest pools of the Ocean cleaving a path, shoulders above the waves, or bringing an ancient ash tree down from the mountains, he treads the earth with his head hidden in clouds, so Mezentius advanced in his huge armor.] This crescendo of images makes us perceive the expansion of Mezentius as a character, and the final grandiose comparison perfectly balances a simile devoted to Aeneas in the course of his preceding aristeia: Aegaeon qualis, centum cui bracchia dicunt centenasque manus, quinquaginta oribus ignem pectoribusque arsisse, Iovis cum fulmina contra tot paribus streperet clupeis, tot stringeret ensis: sic toto Aeneas desaevit in aequore victor . . . (565–­69) [Like Aegaeon, who they say had a hundred arms and a hundred hands, who breathed fire from fifty mouths and chests, when he crashed as many shields against Jupiter’s thunder and drew as many swords; so Aeneas raged victorious over the entire plain . . .] These two forceful similes, which among other things lack (unusually) any Homeric precedent, give the impression of and prepare us for a Titanic conflict.13 So the duel between Aeneas and Mezentius

THE STRUCTURE OF AENEID 10

45

will carry out on an extended scale the idea of dynamic balance that runs through all of book 10 and is virtually its unifying Gestalt: magno discordes aethere venti / proelia ceu tollunt animis et viribus aequis . . . anceps pugna diu, stant obnixa omnia contra: / haud aliter Troianae acies aciesque Latinae / concurrunt; haeret pede pes densusque viro vir (356–­61); agmina concurrunt (ducibusque et viribus aequis (431); iam gravis aequabat luctus et mutua Mavors / funera; caedebant pariter pariterque ruebant / victores victique; neque his fuga nota neque illis (755–­57). The overall “blow for blow” effect arises not only on the level of plot—­where as we have seen single episodes are balanced through repeated counterstrikes—­but also permeates the individual descriptions at the expressive level of rhythm and diction. 5. This equilibrium between Aeneas and Mezentius gives strong narrative motivation for the final duel: but the fulfillment of the story and of the reader’s expectations is only partial and relative, just as the closure that the book’s ending gives is only partial. Indeed, the knot presented by the death of Pallas still needs to be resolved, and that resolution will only come at the work’s definitive conclusion, in the struggle that closes the twelfth book. The relationship between Lausus and Mezentius is so naturally close that the reader perceives the events that follow as natural and almost automatic. But the narrative is also constructed to motivate, along the same lines but through more elaborate means, the bond between Aeneas and Pallas, perhaps the most delicate and most carefully worked part of the plot. It is true that the characters only meet in book 8, when the young man is entrusted to Aeneas as comrade and “apprentice.” But book 10 contains a brief yet conspicuous recall of this situation. In a vignette inserted ad hoc, Pallas is seen adfixus to Aeneas’s side, eager to learn from him the positions of stars (they are navigating by night), and asking the hero to relate his adventures, quae passus terraque marique; what shines through is a very personal bond (within the poem Aeneas gives a similar account only to Dido), one that surely imitates the heroic pair Patroclus-­Achilles, though it has a very different connotation and is more focused on Pallas’s apprenticeship

46

C hapter 2

as a hero and the consequent protective role of Aeneas. It is hardly arbitrary, in other words, to connect this bond to the sphere of paternal feeling. The Aeneid’s readers already recognize how greatly this emotional situation pervades the entire poem.14 And it is all the more interesting to notice how, through a typically Vergilian filter, the relationship between two ἑταῖροι like Achilles and Patroclus is transformed—­while preserving the same structural properties within the plot—­to possess very different qualities and is harmonized with a set of affective relations that the Aeneid tends continually to repeat. It should not go unnoticed that, in sketching the bond between Aeneas and Pallas, the text clearly highlights an anthropologically situated juncture, something that goes beyond affect and sentiment. When Aeneas is informed of Pallas’s defeat, a clear image strikes his mind in shock: Pallas, Evander, in ipsis omnia sunt oculis, mensae quas advena primas tunc adiit, dextraeque datae . . . (10.515–­17) [Pallas, Evander, all was before his eyes, the bread they broke as he arrived at first as a stranger, the right hands given . . .] The dextrae datae here are the sign, contractual as it were, of Roman hospitium, the institution that regulates relations between strangers and obliges one through a private foedus or pact of mutual aid, which the parties are bound in honor to respect.15 (The expression, little noticed by Vergilian commentators, is paralleled exactly in Livy, 30.13.8: recordatio hospitii dextraeque datae). On the other hand, the reader has learned in book 8 that Aeneas has an important predecessor as hospes chez Evander: Hercules (and the two heroic figures have other significant connections in Vergil). Hercules and the values of hospitality are precisely what Pallas invokes before his ill-­fated duel with Turnus:16 per patris hospitium et mensas quas advena adisti. (10.460) [By my father’s hospitality and the bread we broke when you arrived as stranger.]

THE STRUCTURE OF AENEID 10

47

From the verbal echoes (mensae, advena adire) the reader recognizes the transference of obligation from Hercules—­who for all his suffering is unable to intervene due to Jupiter’s commands—­to Aeneas, who, after failing in his duty to protect, must now assume the task of vengeance. Furthermore, in his reckless speech of triumph over Pallas’s corpse, Turnus says that Evander will have paid dear for his bond of hospitality with Aeneas, his Aeneia hospitia (cf. 10.494–­95). But this retaliatory barb recoils on Turnus himself: in a speech pivotal to the story, in fact, Evander reminds the Trojans of their bond of hospitality (quas / iunximus hospitio dextras, 11.164–­65) and at the same time declares that Aeneas cannot fail to wreak vengeance on Turnus for Pallas’s death. So the cultural model of hospitium acts in reading as an isotopy, a thread that orients the reading according to cultural expectations: Evander should be the one to avenge Pallas, but, as we hear many times, the father is too old for combat—­and is consigned to the role of a Homeric Priam; otherwise presumably he would act like Mezentius at the end of book 10. So a “substitute” is required, and according to the text’s network of motivations this can be none other than ­Aeneas. It is interesting that this thread of the story concerns Aeneas in a private and personal way. In fact the death of Turnus at the end of the poem does not seem directly motivated by “political” relations; the conflict between the two peoples is reconciled through fiat on Olympus and a collective intermingling is already imminent. Aeneas will kill the enemy commander out of a personal obligation and, what is more, driven by a surge of uncontrollable rage at the sight of Pallas’s baldric. This seems like a complicated narrative device: on one hand, Aeneas predictably fulfills the bond of hospitality, but on the other hand, he is not animated by a prolonged and ongoing μῆνις—­like Achilles, who instead pursues his vengeance without hesitation—­but acts on a sudden lightning-­like impulse. The reminder of the “baldric theme” thus has a decisive function. We can see that Turnus’s death is amply motivated on the narrative level, but this narrative motivation does not amount automatically to a full ideological justification. Indeed, to respect the bond of hospitality Aeneas goes against part of his own values, since he has to ignore the clementia toward one beaten down and suppliant (in less subjective terms, he has to reject a hiketéia in the name of xenia).

48

C hapter 2

6. This brings us back to the narrative texture of book 10, which in this aspect as well prefigures developments in the action. At 521ff., when Aeneas has just learned of Pallas’s death and is searching for Turnus to kill him, he comes upon an enemy named Magus who implores him as a supplex to spare his life; Anchises and Iulus figure in his appeal: Per patrios manis et spes surgentis Iuli te precor, hanc animam serves gnatoque patrique. (524–­25) [By your father’s shades and the hope in growing Iulus, I pray you, spare my life for my son and father.] This appeal to familial pietas has no effect on Aeneas (quite unlike the later episode with Lausus) because the events of war press in another direction.17 This very situation recurs much later in Turnus’s supplication: miseri te si qua parentis tangere cura potest, oro (fuit et tibi talis Anchises genitor), Dauni miserere senectae. (12.932–­34) [If any care for your sad father can touch you, I pray—­and your father Anchises was the same to you—­pity Daunus in his old age.] Two influential motives collide: on one side Aeneas feels the call of paternal love (and allusion also creates an impression of possible reconciliation: the scene between Priam and Achilles in the Iliad is resolved with the moving words “remember your father”);18 on the other side, he is brought back to vengeance by memory and situation, a link evident to anyone who connects the dots of Aeneas’s actions across the narrative arc that unites books 10 and 12. The sight of the baldric “reproduces” behavior in Aeneas similar to what already occurred in battle in book 10, in the episodes after the defeat of Pallas. So the theme of the baldric—­superficially suggested by the

THE STRUCTURE OF AENEID 10

49

Iliad’s theme of Patroclus’s armor (see chapter 1)—­takes on a psychological charge in Vergil that is altogether original. 7. We have been occupied so far with how a narrative is organized, in the act of reading, into paradigms that become models of interpretation for the reader. In light of this it has been possible to identify certain typical methods: recursion and parallelism, which not only produce formally marked schemes but also establish systems of oppositions; the mention of cultural models, which the narrated action implies as its referent. But Vergil’s textual strategy makes use also of specifically literary models, above all the Homeric model. And this intertextual background also needs to be brought into consideration. It should be readily apparent that the intertextual relations between Vergil and Homer do not boil down to a single, simple typology of allusion. We can, almost at first glance, make a rough distinction between at least two major intertextual directions that look to two different aspects of the Homeric text. (Of course, the special status the Homeric model enjoys is what accounts for the highly peculiar character of Vergilian imitation.) For one thing, the Iliad offers a story to imitate, pure and simple; an individual story that has the potential to be the quintessential epic plot for future imitators. Vergil takes up, in order to transform it, this story of two great heroes whose lives intersect in a scheme of vengeance (Aeneas-­ Turnus-­Pallas as Achilles-­Hector-­Patroclus). But this exploitation (the model as “subject matter”) is not the only way the Homeric text functions as a model. Homer also offers a repertoire of codified materials and his text becomes almost a code for new epic. So Vergil does not draw only on a single exemplary story that can be cited largely through allusion but also on a genuine repertoire—­of epic formulas strongly codified in rhythmic language; of topoi—­that is, figurative types—­less encoded in specific terms; of combinatory rules for grouping typical materials. Vergil visibly adheres to these epic constraints, but at the same time he deviates from them toward radically new “reading effects.” His combining of duel scenes with aristeiai does not on the surface

50

C hapter 2

differ much from the typology of battle scenes in the Iliad. But one important novelty, as we have tried to show, resides in the contrastive interweaving that underlies the “type scenes” of combat and slaying. Through this very repetition of codified elements (rituals of challenging, posturing, boasting) a complex of behaviors comes to light that supplies a system of characters, and the reader thus moves without break from the iconographic repertoire to the poem’s ideological world. To get to this point, Vergil gives fresh meaning and function even to traits in Homeric epic that had become grammaticalized, so to speak, and had lost autonomous significance. Recall, for example, a detail already touched upon elsewhere (see chapter 1). The gesture of planting one’s foot on a conquered enemy’s corpse is entirely habitual in Homeric duels; the formulaic nature of the gesture precludes it being marked strongly and given special significance in any once instance where it occurs. In Vergil, on the other hand, the detail of Turnus who tramples the lifeless body of Pallas is not felt as a “formulaic” gesture. In fact, the action of Turnus who laevo pressit pede . . . exanimem rapiens immania pondera baltei (10.495–­96) [Held the corpse down with his left foot, snatching the baldric’s heavy mass] is rendered visible and attracts attention through contrast with Aeneas’s behavior when—­in the context of a parallel episode where readers will have expectations of encountering a “type scene”—­he takes up in his arms the body of his slain foe;19 it is also highlighted by its synchrony with another marked gesture, the act of seizing a prize spoil that will turn out to be fatal (rapiens immania pondera baltei). Besides this connotation of “exchange,” based on a link at a distance between the two episodes, Vergil can also count on another kind of connotation. In Homer there is no indication that planting one’s foot on a conquered foe is an iconic behavior, something that in that culture would be, to some degree, liable to symbolic translation. Conversely, in Roman culture “trampling a beaten foe” has a clear negative charge. It is the anti-­model of knowing how to conquer, of restraining oneself in success. In the Aeneid, therefore,

THE STRUCTURE OF AENEID 10

51

it both repeats a fixed and formulaic epic trait and at the same time, thanks to a process of estrangement, produces a new connotative meaning: the iconographic model, already a complex sign in itself, becomes as a whole the signifier of a new sign, whose content is something like: “not knowing how to be moderate in victory.” 8. By working on the wide cultural margin that separates his work from the Homeric model, Vergil is able to give new functions to some fixed and repetitive elements of the traditional epic repertoire. The problem book 10 poses lends itself to several perspectives: from the point of view of reception, we may note that the Vergilian text does not at all refuse to exploit an aesthetics of regularity of the sort that managed relations between epic singers and their original public;20 epic would not know how to get away from its need for repetition, which was satisfying on formal as much as on thematic-­ideological levels. On the other hand, it is obvious that Vergil’s text wants and needs to address an audience that has developed more complex cultural needs as well. Even more important, the shift from a poetics of regularity to a largely “open” poetics is accomplished precisely by going along with the trend toward repetition, by taking the institutionalized narrative language “at its word.” The audience that a poetics of this sort presupposes has a natural inclination to accept this institutional language (now entirely severed from its original basis that was marked by traces of oral composition and poetics) while scrutinizing it for multilayered contextual motivations. The story adapts itself to this sensibility because it tends to motivate the action, even in many of its peripheral details. The scene where Turnus kills Pallas is conceived in such a way that almost any descriptive trait might, by choosing a specific explanatory model, be linked to the progression of the plot (rather like how the choice to continue a linguistic structure in a definite way orients an utterance at the outset). The same cannot be said of the killing of Patroclus in Iliad 16. Not that the scene is conceived on its own, as a heroic song that stands as an isolated unit with respect to the continuous story. The fact is that the narrator’s poetics and the aesthetic horizon of his audience are not trained to valorize the contextualized selection of

52

C hapter 2

certain descriptive traits (for example, those “blocked out” in the register of formulaic style) by ordering them on the principal axis of narration. 9. Pushing this kind of comparative observation further, we can read the transformation of the epic model not as a reworking and sifting of sources (something that happens in some kind of avant-­texte, before the process of literary communication) but rather as the construction of a new narrative strategy: the variations on the Homeric model can be better interpreted as “transcodifications” that obey the laws of a coherent narrative model underlying the unfolding story. In the programmatic proem of book 7 Vergil had alerted his readers to be ready for the renewal of a tradition, while deliberately selecting as a topic the loftiest thematic core of Homeric epic: actos animis in funera reges. Book 10 is a dense realization of this poetic program. Those who wish to read it as a simple account of heroic adventures are not spared any of the expected traditional imaginary elements: massive slaughters and duels, but also supernatural transformations, divine interventions, phantoms that deceive fighters, catalogues of troops, and invocations of the Muses. After insisting so strongly on the deep plot structure, on the riches of implicit organization, it is fair to pay respects to this dimension of reading that, although “super­ficial,” is no less important than others. The figurative expansion, the exuberance of imaginative elements, the interstitial enjoyment of redundant details, are all fundamental aspects of epic. While he controls and organizes them in a comprehensive form, the narrator also knows how to respect them. His job is to “take apart” but also to reconstruct, continually, an instrument of traditional communication.

Ch a p t e r 3

THE ARMS IN THE SKY Diffraction of a Narrative Theme

The two preceding chapters aimed to show how Vergil makes the Homeric model “work” in the text of the Aeneid (by transforming it according to definite cultural filters and mobilizing it in plot constructions). But it would be wrong to think that an analysis so framed intends in some way to reduce and simplify the meaning of a text so rich in cultural referents and so complicated in construction. (The error would be in the prejudice that the fact that a text refers to already known models necessarily implies reduction and impoverishment, as though these references were a surrogate for meaning.) What we know for sure—­since it was the common reaction of all the Aeneid’s readers, no matter how cultured or knowledgeable—­is that Vergil is writing a complex epic, a text that opens contradictions and multiplies questions and reflections. On the other hand, this time thanks to philological tools, we know equally well that this poetry lives on a multitude of models, a stratification of cultural codes. These two properties must in some way be reconnected with each other. What matters then is grasping the specific way the complexity of cultural presuppositions (and their interactions) acts in reading as a source of complex meanings—­that is, tightening the link between the density of literary signification and the multitude of implied models. In becoming discourse, the multitude of ­models becomes fodder for narrative strategies that aim to produce and control determined effects of reading. To recognize and interpret 53

54

C hapter 3

some of these narrative strategies and their relative effects of sense will be the goal of this and the following chapter. Let me begin with a scene where a familiar Homeric theme—­the arms of Achilles—­rubs up against models from another genre. The relevant passage comes from Aeneid book 8:1 Vix ea fatus erat, defixique ora tenebant 520 Aeneas Anchisiades et fidus Achates multaque dura suo tristi cum corde putabant, ni signum caelo Cytherea dedisset aperto. Namque inproviso vibratus ab aethere fulgor cum sonitu venit at ruere omnia visa repente 525 Tyrrhenusque tubae mugire per aethera clangor. Suspiciunt, iterum atque iterum fragor increpat ingens: arma inter nubem caeli in regione serena per sudum rutilare vident et pulsa tonare. Obstipuere animis alii, sed Troïus heros 530 adgnovit sonitum et divae promissa parentis. Tum memorat ‘Ne vero, hospes, ne quaere profecto, quem casum portenta ferant: ego poscor Olympo. Hoc signum cecinit missuram diva creatrix, si bellum ingrueret, Volcaniaque arma per auras 535 laturam auxilio. Heu quantae miseris caedes Laurentibus instant! quas poenas mihi, Turne, dabis! quam multa sub undas scuta virum galeasque et fortia corpora volves, Thybri pater! poscant acies et foedera rumpant.’ (8.520–­39) 540 [He had just spoken these words and they held their faces downcast, Aeneas Anchises’ son and trusty Achates, thinking of many hardships in their sad hearts, when Cytherea gave a sign in clear sky. For suddenly unexpected in the heavens lightning flashed with thunder and everything seemed to crash down and a Tyrrhenian trumpet blast moaned through the air. They look up and over and over came the great crashing sound. Arms in the clouds, they saw them redden in a bright clear quarter of heaven and their clashing roared. The others

T he A rms in the S ky

55

were stunned in mind but the Trojan hero recognized the sound and his divine mother’s promises. Then he tells them: “My host, don’t ask by any means what misfortune these portents bring. I am called by Olympus. My goddess mother foretold she would send this sign if war impended, and arms from Vulcan she would bring me from the sky as aid. How much slaughter is waiting for the wretched Laurentines! What penalties you’ll pay me, Turnus! How many shields and helmets and strong bodies of men you’ll roll under your waves, father Tiber! Let them muster ranks and break the treaties.”] Troubled in mind, Aeneas and Achates reflect quietly on their cares in the royal halls of Evander. The elderly king of the Palatine has promised concrete aid for the coming war and, more importantly, has indicated that Aeneas is the man called by fate to lead the Etruscan troops into battle: according to a prophecy by the haruspices, Aeneas may be the foreign leader quem numina poscunt (8.512). Nevertheless Aeneas would have sunk into bitter thoughts if Venus had not suddenly sent a heavenly sign: lightning and thunder in a clear sky, as though the universe were crumbling, then the gloomy sound of an Etruscan tuba. In the cloudless sky above they see armor flashing and resounding as though a hand were shaking and making them crash together. All are frozen except Aeneas, who recognizes his mother’s promise—­something the reader of the Aeneid learns about only at this moment—­and can explain the prodigy in a positive light: ego poscor Olympo; this sign from the heavens is meant for him and it points to divine protection for the looming war. And in fact the armor that Vulcan forged during the night is being brought to Aeneas as a tangible sign of divine aid and an omen of victory in the war with Turnus. Roman readers will perhaps have thought of the legend of the ancilia, the sacred shield that rained down from heaven amid ominous bolts of lightning (Ov. Fast. 3.67–­74) and had been preserved as a pledge of Roman power since the earliest times. But surely they could also not ignore, thanks to a path of allusion laid down by Vergil, the presence of a much closer model. As with any summary, this one amounts to only a partial and reductive reading of Vergil’s scene, selecting as pertinent certain meanings over others. The coherence of this particular reading, which we

56

C hapter 3

might from here on treat as an isotopy, is ensured not only by the semantic compatibility of the whole sequence but also, more generally, by the orienting action of a paradigm that the text accepts and assumes is known. To identify this model it is enough to situate the scene in its context: while Aeneas is sleeping in Evander’s house, Venus has paid a visit to Vulcan. This domestic scene recalls, overall and in particulars, the encounter between Thetis and Hephaestus in Iliad 18—­which Venus cites explicitly as precedent (8, 83–­84) in her conversation with Vulcan, but above all the two scenes have a parallel function in the plot. Vergil’s reader can hardly avoid as a framework the intertextual continuity here between the Aeneid and the Homeric poem. The plausibility of Vergil’s text is clearly based on repeating the theme of armor; the son of a goddess, called into combat, receives as a gift from heaven armor of divine handiwork. There are also exact analogues in situation: Dawn with saffron robes rose from the streams of Ocean bearing light to immortals and mortals, and Thetis came to the ships bringing the god’s gifts. She found her son prostrate on Patroclus’s corpse, weeping intensely; around him many comrades were lamenting. She stopped in their midst, the great goddess, took Achilles’ hand, and spoke and called him by name: “My son, so tormented, let us leave this man to lie here, since he was slain by the will of the gods. But you, take these gleaming arms of Hephaestus, beautiful work, which no man has ever borne on his shoulders!” So she spoke and the goddess set the armor down before Achilles. And they resounded with a crash and none had the courage to look directly on them, they trembled so with fear. But Achilles, when he saw them, was filled with wrath, and his eyes shone terribly under his brow, like a flame: and he rejoiced, holding in his hands the splendid gifts of the god. (Il. 19.1–­18) The armor’s arrival is to the hero the sign for war that pulls him out of his sorrowful state and fills him with violent courage and a sense of elation, but the arms are also a superhuman appearance that strikes fear in all the others. Naturally, this divergent reaction expresses the heroic figure’s complete individualism. Add to this also

T he A rms in the S ky

57

that the physical delivery of the armor is mediated in Vergil by a prodigious sign; the hero is responsible for its correct interpretation, thanks to his superior level of awareness. The scene is therefore constructed on relations of unequal knowledge, a significant difference from the Homeric style of narration. From this point of view, the form of Vergil’s scene might also be defined as the dramatization of an epic theme: a helpful example is the structurally parallel scene in Sophocles’ Oedipus at Colonus: CHORUS: I cannot say any heavenly command is vain. Time sees, sees all these events . . . God! The sky is thundering . . . OE DI PUS: Daughters, my daughters! If anyone is here nearby, lead me to the courageous Theseus. A N T IGON E: Oh father, why are you calling? OE DI PUS: A command from Zeus, this winged thunder will drag me down to Hades . . . CHORUS: Behold, the greatest, unspeakable ruin is crashing down, hurled by Zeus. Fear passes through me to the tips of my hair. My spirit trembles, another light flashes in the sky . . . I am afraid. It never cracks in vain, without some grave outcome. Oh boundless aether . . . Oh, Zeus! OE DI PUS: Daughters, the end the gods predicted is coming. Nothing can stop it . . . This is my hour of decision . . . The gods in person, their own messengers, are announcing it to me: they have neglected not one of the appointed signs. (1451–­1512) Standing opposite on the stage are a terrified chorus that does not know how to interpret the heavenly sign and a knowing protagonist who recognizes the prodigy as the fulfillment of a prearranged signal. This agreement sanctions the bond between man and god: of Oedipus it is said ἐκ θεοῦ καλούμενος (OC 1629), and Aeneas says of himself: ego poscor Olympo (8.533).

58

C hapter 3

Following the trail Vergil has laid out, it is natural to see in this scene the consecration of Aeneas as a combatant. This happens on the subjective level, since the hero overcomes the mysterious indecision that is still holding him back, and on the objective level, since the fulmen consiliarium and the promise of divine aid are the most obvious legitimations for beginning a war. (It is surely no surprise in this context that Venus has a warlike role; if the goddess is neglecting her usual duties to have arms made it is because she is already thinking about the divine pedigree of the Julian clan; and Venus Victrix was the password of the victors at Pharsalus.) This overall interpretation, which to me seems to arise directly from the lines of the text, nevertheless tends to cancel out another important aspect: the different attitudes of the characters toward the war. Despite the divine assurances, Aeneas and Achates spontaneously sink into sadness; the immediate reaction for those who witness the heavenly sign is troubled bewilderment.2 Aeneas tries to reassure Evander (ne vero, hospes, ne quaere profecto / quem casum portent ferant; 532–­33),3 but he himself, at the very moment he recognizes the divine aid, also lingers quite unexpectedly in a prophetic lamentation on the miseries of war (heu quantae miseris caedes Laurentibus instant! / quas poenas mihi, Turne, dabis! quam multa sub undas / scuta virum galeasque et fortia corpora volves, / Thybri pater!; 537–­40). The style of these lines denotes dismay and almost a sense of helplessness; it has the same exclamatory force as Anchises’ prediction of the coming civil war disasters in book 6.4 Despite the promise of aid from Evander, and his sending of young Pallas, the fate of this expedition is already surrounded with an aura of tragedy. This is why some scholars, unsatisfied with the “Homeric” interpretation, have proposed a key to reading this scene that I would call symbolic. In this interpretation the reader should probe more deeply into ­Aeneas’s psychology: we should see the hero’s sadness as an ill omen, in particular regarding Pallas’s fate. The clue to this reading is the patronymic Anchisiades (521): Aeneas, as son of Anchises, shares with Evander his feelings of paternal love and concern at this moment.5 But despite the merit of enriching analysis and doing justice to certain points of significance, I would hasten to make clear that this interpretation is basically unacceptable, on account of the excessive importance it lays on the unexpressed, its privileging of psychology over action, which

T he A rms in the S ky

59

seems an inadequate way to deal with reading an epic text regulated by definite conventions. Instead, we should be examining the legibility of the Vergilian text, its explicit and implicit codes of reading, and abandon the search for hidden intentions. There is a danger of making too sharp a distinction between those who pursue a unified reading, one that is totalizing and quickly recognizable (the reductive reading mentioned at the beginning), and those who opt instead for explications of a “hidden” sense—­as though we are forced to choose between reductive reading and cryptology.6 Some interpreters, obdurate exponents of Römertum, want to impose on Vergil the filter of their own simplifying ideology; others, raised in different aesthetic climates, dismiss the most obvious sense in the name of symbolic suggestions, following intuitions that elude concrete verification. It is possible that these two analytical styles, opposed but in some sense complementary, should both be rejected. Let us pause momentarily on a point of exegesis where an interesting rift has opened: the (apparent) problem over identifying the arms. Major Vergilian scholars have been divided, with sound arguments on both sides, about whether or not the arms in Venus’s sign are exactly the same arms that Aeneas will don and use. Briefly put, one side can maintain that they are the same because this is substantially Aeneas’s own interpretation (Volcaniaque arma per auras / laturam auxilio; 535–­36) and because the blatant Homeric allusion, examined just earlier, points to the same conclusion. But against this it can be argued that the arms are not the same because their appearance is a constituent in a discrete paradigm: the whole supernatural vision, along with the trumpet blast and thunder and lightning in a calm sky, is typical of Roman prodigy.7 It is no idle question since, even if poorly stated, it isolates a critical point for reading the scene. From our point of view, surely this is not a problem that concerns the referent and it will not be solved by straining our eyes at some specific features of the apparition. Perhaps it is true, as Warde Fowler concludes, that the reader is left free to interpret the image outside of any unifying logic,8 but rather than accepting a non liquet this should be a point of departure for new inquiries. A good explanation will also take account of and encapsulate other apparently irreconcilable explanations. If we are facing

60

C hapter 3

a Leerstelle here, an empty space, it will not do to refill it with arbitrary fantasies; we can regard it as a site of polyphony constrained by the text’s structure. In that case, we need to determine (in terms of structural semantics) the point of disjunction that gives rise to, and justifies the presence of, this interpretive ambivalence. We have already discussed the clues that lead to a “Homeric” solution to the scene as a whole. Now let us follow the track created by the prodigy—­Vergil’s innovation on the narrative fabric of his epic model. Comparable texts quickly reveal that arms in the sky and a trumpet blast are a well-­known vision:9 they are belli mala signa, as Tibullus will call them (2.5.71), and for Ovid they are signa . . . luctus haud incerta futuri (Met. 15.782). To be sure, generally speaking a prodigy is a flexible framework open to varying interpretations; but it also happens that a sign can become firmly anchored to a definite event. So we can say that arms in the sky “represent” the civil wars, since that is what they were consistently associated with. (It is an example of a metaphorical use of metonymy—­that is, the prodigious sign of arms, through a particular act of narrative characterization, turns into a symbol.) It is not a matter of insisting on either specific pieces of evidence or on their possible lines of influence. Tibullus can easily summarize an entire epoch with this image: it announced and attended upon the war between Marius and Sulla, the Sullan proscriptions, the strife between Caesar and Pompey, Pharsalus, Munda, the assassination of Caesar, Philippi, the proscriptions of the triumvirs, and so on. The very extent of the motif and its character as a “collective representation” argue against identifying a single archetype. Yet it is interesting that the earliest attestation we have is a text of Vergil, the end of Georgics 1 devoted to the civil wars; here already the motif of heavenly arms is centrally situated in a significant dialectic of opposing images. In the past the sound of arms in heaven signaled imminent disaster (1.474); now, when discord reigns on earth, the peaceful arms of farmers are being turned into swords (1.506–­8); only in a distant future might the fateful apparition of arms in heaven be in some sense overturned: one day a peasant working the soil will unearth with surprise buried armor, the wreckage of a miserable past.10 The important thing is to establish that in Vergil, and for his anticipated readers, the prodigy evoked a specific negative connotation:

T he A rms in the S ky

61

not divine aid by any means but rather the wrath of the gods and the onset of a catastrophe. This is all the more notable when we consider that the episode has all the air of being a free Vergilian innovation; while prodigies in the Aeneid are generally suggested by a specific tradition (of epic or antiquarian origin), this scene derives its meaning only from the cultural models that underlie it. We can now summarize it this way: the image of the armor is the textual locus of a disjunction into which two distinct isotopies have been grafted. The prodigy’s inauspiciousness blends perfectly with other contextual elements and especially works to motivate Aeneas’s mournful reaction and the terror of those who witness it. In turn, the characters’ reactions serve as a model for the correct reception of the heavenly sign, inviting the reader’s collaboration in tracing it to a disturbing correspondence in their own cultural memory. At this point we may ask whether other aspects of Vergilian narration confirm this “negative” reading of the war that Aeneas must conduct in Latium. The war in Latium is particularly brutal because it divides peoples who are too similar. Vergil reworks the saga to give the Trojans and Italians a common origin, projecting their unity into the distant past by giving Dardanus an Italic ancestry and imagining the two nations as having split from a single race. In this way the war between Aeneas and Turnus subverts the plan of fate and is more a rupture of a prior unity than a clash of divergent opposites. The conflict is driven by furor (a term frequently applicable to both sides), by scelerata insania belli; the adjectives used to describe it, besides those sanctioned by the epic tradition, are genuine curses like infandum (12.804), dirum (11.217), importunum (11.305). Instead of being an inevitable phase in the founding of a new equilibrium, the war seems like the desecration of a possible and longed-­for concord. The breaking of pacts (foedera rumpant, 8.540) is a typical trait of civil war (vicinae ruptis inter se legibus urbes / arma ferunt, G. 1.510–­11; also finitimas in bella feram rumoribus urbes, A. 7.549). This verdict is confirmed by the system of narrative perspectives, since almost every character in the poem is either detached from or abhors the war. Viewing it from Olympus, Jupiter condemns the hostilities for all the wasted energies that ought to be conserved for future common action that will be required against foreign foes, when Rome has to defend itself against the Carthaginians; and the reader would

62

C hapter 3

also surely have recognized the typical pattern of condemning civil war, echoed in many writers of the Augustan period and simply transposed here by Vergil into the time of origins. The poet’s interest in belli causae, more than merely traditional epic curiosity, suggests instead the problematic of causae civilium armorum (cf. 7.40, 481, 552). The origins of the conflict are attributed to a demon of discord, Allecto: her usual activity is to break pacts and subvert peace and even unanimos armare in proelia fratres (7.335). On several occasions the poet seems interested to portray Italy as a primitive land of populi feroces naturally given to violence and in need of civilization; but when he goes on to narrate the genesis of the war this concept—­ which would serve as a strong justification—­is surprisingly abandoned. The aged Latinus has ruled his city through a long era of peace (7.45ff.), just as before him Saturn had governed the peoples of Latium in peace (8.325); the golden age still shines and at the outbreak of the war the poet can call Italy inexcita Ausonia atque immobilis ante (7.613). Recalling the framework of civil wars traced before in the Georgics, Vergil portrays farmers leaving their fields (8.8; G. 1.507) and wickedly exchanging their work tools for arms of battle (7.635–­36; G. 1.506–­8). The narrator does not hesitate to open up contradictions and disrupt literary habits or norms when it serves to give the events the desired connotation. In some cases a great deal is left up to the reader’s free collaboration. When Aeneas and Latinus are called, proleptically, “son-­in-­law” and “father-­in-­law,” this might appear perfectly natural, and it is understandable that Juno says “the son-­in-­law and father-­in-­law will unite at the price of their own people’s blood” (7.317). But as Fraenkel, among others, has observed, any Roman reader would associate this slogan with a bitter memory, and even within the poem, in Anchises’ prophecy, Caesar and Pompey are described as father-­in-­law against son-­in-­law (see A. 6.830–­31).11 A similar case is when Aeneas prepares a human sacrifice for the funeral of Pallas (10.517ff.). This act, so evidently barbaric in Roman thinking at the time, makes satisfactory sense in relation to the heroic age, according to the model of Achilles’ vengeance (which was already, moreover, explicitly blamed by Homer, in a way that is exeptional for his narrative practice; Il. 23.175–­76, according to schol. Hom. ad loc., 398 E.). But it is interesting to recall the human victims Octavian sacrificed, as Caesaris ultor, in the course of the bloody

T he A rms in the S ky

63

war at Perusia. According to a slippage well known to the civil war veterans, the entire war tended to be presented as a vendetta pursued against the machinations and plots of those who provoked the war. It is no accident that Aeneas’s words poscant acies et foedera rumpant seem like the antithesis of the sarcastic vaunt hurled by the demon of discord at 7.546: dic in amicitiam coeant et foedera iungant! The most interesting (and perhaps the only certain) cases, however, occur when the implied meaning is made visible by a strained appeal to epic convention. In book 12 Vergil repeats the old epic custom where the narrator invokes the Muse or some other deity for aid. Because the poet has to narrate a particularly long stretch of androktasia, one would expect him to summon divine omniscience or memory, or the traditional ten tongues and voice of tireless bronze (after all, Vergil has already paid homage to this archaic method several times). But instead the narrator subverts this expectation and asks the god not to narrate but to explain the events: “Which god can explain to me now so much brutality, the various killings and deaths of chiefs that Turnus and Aeneas now caused? Did it seem just to you, Jupiter, that peoples destined to eternal peace should clash in so much turmoil?” (see 12.500–­504). Clearly the (problematic) answer is left to the reader, disrupting the normal autonomy of epic narrative. This aspect of Vergil’s textual strategy deserves further scrutiny. Describing the war between Trojans and Latins as civil war creates a sort of narrative swerve; the war is so strongly condemned that the model of civil war takes over by metonymy, a model of war at its most intrinsically negative. The existence of this swerve is enough to expose the flaws in the communicative script of epos. Where the theme of civil war presses most obsessively on the narrative’s epic form, it creates “open” zones, though ones that are, so to speak, controlled. These spaces for reflection are made possible by the multiplication of cultural codes and the estrangement of narrative conventions—­but without the work failing to convey its basic epic message. Characterizing the war as civil war does not play any obvious functional role in the plot; what it does instead is add a set of qualifying elements that take on a meaning that should not be underestimated.

64

C hapter 3

Aeneas the founder is also the destroyer of an ancient harmony; we stand at the beginning of an inevitable process that will drag Rome and Latium out of the pastoral Arcadian myth into confrontation with the turmoil of history. The arms of Aeneas signify divine aid, and the golden shield bears upon it in concrete detail—­thanks to Vulcan’s prescience—­the surety of a distant future. But the arms in heaven are also a coded sign of an Iron Age that will bring to an end the primitive Golden Age of Saturn, and of a war that will transform, through hard sacrifices, both the victors and the conquered. Even the primitive Rome to which this prodigy has been given must go through a bloodbath (one thinks of the sacrifice of young Pallas), because it is now the era in which plowshares become swords. At the moment that Aeneas becomes the knowing executor of the Olympian project, this double-­edged omen comes to recall both the necessity and the price of this mission. (In this light even the description of Evander’s precarious idyllic world, of this Rome that is not yet there, takes on a sense of impossible nostalgia, of a utopia overturned in the past. A parallel theme is found in Tibullus’s elegy for Messalinus [2.5],12 where Rome’s history is divided into two Sybilline prophecies, both of them true but also completely antithetical; it foresees ex eventu, on one hand, the greatness of eternal Rome, which beginning with Aeneas’s triumphs puts to an end for good its idyllic origins, but on the other hand, the horror of the civil wars.) If a single prodigy serves at the same time as a sign of divine aid to the victors and as a sanction of a catastrophic war (the arms of heaven, the arms in heaven), this inevitably raises a more general ideological reflection that turns on the nature of power. An avenger has used the arms of heaven to found his power, thus bringing to an end an epoch of terrible violence by restoring the peace; but this power is founded on civil war, and its actual constitution cannot be separated from the “great terror” in which it arose. Retracing this tortured genesis can be, at the same time, both a justification and a call to critical reflection. On one hand, this look at the past serves to legitimize the acts of a man in the name of a divine mission; on the other hand, the new power surely does not want to reflect so much on the past that it calls into question its own violent foundation. (One thinks of the ambigous role that the archetype of Romulus, the fratricidal founder, played in the culture of this period.)

T he A rms in the S ky

65

One cannot stress enough how mediated, how strongly controlled and repressed, these contradictions are in their appearance in Vergil’s epic poetry. Against every historicist or even allegorical straining we must balance the cautious activity of philology and exegesis. But we can at the very least suspect that the polyphony of Vergil’s text is also a means to allude to this dialectic, to transform it into language without making it entirely explicit. The ambiguity of the text would then be like the fixation of an irresolvable ideological conflict: not the passive reflection of a triumphant ideology but the portrayal of a foreclosed dialectic, “a dialectic in stillness.”

We have marked out the elusive pathway that unites two discursive manifestations quite distinct from one another: an ideological contradiction and a narrative “polyphony.” But our initial project, it will be recalled, was centered more on the second term of this pair. We will therefore conclude with the form and workmanship of a “pluri-­ isotopic” narrative text. From the point of view of reception, multiplying codes creates an epic narrative that lays down an overabundance of reading conventions and requires a more complex attitude from the audience. Comprehending the text is slowed down by an estrangement from established cultural habits, and readers’ hesitation between heterogeneous codes of reference is productive because it prevents them from trusting uncritically to stock, conventional meanings. The recognition of two distinct cultural codes is justified in our case by the assumption of an audience, a model reader who possesses a necessary competence: obviously this also implies detailed analysis through which, thanks to philological methods, the relevance of the individual cultural paradigms is clarified (such as the model of Achilles’ armor, or of Roman prodigy). Our passage then appears as the intersection of two diverse systems, as though the product of a double synchrony; the necessity of this double reading leads us to accept that this ambivalence is constitutive. Note finally that the two isotopies thus identified in the narrative occupy different levels in terms of narratological function. As we have seen, the narrative isotopy that derives from Homer—­the motif of the warrior who receives the gift of divine armor to go into

66

C hapter 3

battle and exact revenge—­is firmly rooted in the Aeneid’s plot. It could be called the main isotopy, not so much for its extensiveness or the number of semantic traits, the coded elements that compose it, but rather because it mobilizes the greatest number of significant connections within the context. In this way the episode is motivated by prior events and motivates others in turn which have a direct bearing on the poem’s outcome. Through its degree of productivity this type of reading clearly leads us to the level of distributive relations: those connections founded on a homogeneous level of the text and governed by a dominant realism. The second type of reading we have developed offers us instead a case of integrative relation: we are faced with a series of semantic traits not directly necessary to understand the events “linearly,” but which the reader nonetheless perceives and retrieves at another, more general interpretive level. We have seen that this reading is a sum of qualitative elements, scattered across the whole of the text and not bound by relations of motivation. As such their position in the text does not seem determined and what counts instead is the general model to which the individual traces point. (The model of civil war would thus be the “vanishing point” of this semantic perspective: an abstract model unrealized in the text, but presupposed as a reference point, since the reader can use it to triangulate to their own interpretation. Hence the analogy to the imaginary point that orients the lines of perspective in a visual composition.) Undoubtedly, this sort of isotopy is not conspicuously laid out because it plays no role in the plot. The fact that the war is described in a certain way does not, in fact, alter the direction of the narration, nor does it impinge on the structure of actor roles at play in the story. One may well ask then what objective proofs support a reading of this sort, which explicates descriptive aspects that are incidental, disconnected, and even to some degree conflicting with the narrative’s dominant isotopy. It is clear that it cannot be sufficiently proven by simply decomposing the text into sequences of more or less identical semantic features.13 It seems necessary therefore to reevaluate the role of cultural discourse—­of the interpretive constraints on reading that arise from the system of expectations. (In our case, to be explicit, the audience

T he A rms in the S ky

67

had to see in the Vergilian prodigy a familiar and definite paradigm and then seek to integrate this with the other meanings at play in the text). In this sense the critic needs to aim at reconstructing not so much the dictionary (in Greimas’s theoretical sense) as the encyclopedia: the competence the text expects and that serves to integrate the referential comprehension of the events narrated. To reevaluate the activity of this cultural intertext, moreover, would also be a way to open new spaces for philological practice (which often runs the risk of a formalistic view of intertextuality), as well as to bring back into narratological analysis the pleasures of historical-­ cultural reconstruction.

Ch a p t e r 4

THE DEATH OF TURNUS Genre Model and Example Model

In this concluding chapter I will begin from a less traditional question in the study of Vergil and Homer. So far we have treated the Homeric model as a functional element, among others, that contributes to the production of “reading effects.” We saw that, in practice, the action of the Homeric model has to be traced to a complex, polyphonic intertextuality. In short, to analyze individual narrative sequences entailed reckoning with a plurality of models. We could speak of “models in contact,” in as much as we are clearly dealing with models that are heterogeneous and only in communication with one another thanks to a synthesizing intervention. We recognize, however, that the Homeric model—­that is, the way in which Vergil receives and actively models the text of the Homeric poems—­ does not exercise just one, unified sort of intertextual function. Perhaps this plurality of functions can be better defined by tackling it in a more systematic way. The composition of a text is a process (just like its decipherment, which proceeds mirror-­like by recognizing programmed effects that arise from the various components of the project that constructed the text). But it is a generative process, not merely a summative or aggregating one. To be more specific, the composition of a narrative involves the creation of an autonomous space, with a beginning and ending, and the formulation of a continuous substantive discourse. The only perspective that takes into account the way these effects of 69

70

C hapter 4

meaning objectively perceivable in reading are produced is to think of the text as a process that generates the overall sense. This process, once carried out, also creates a kind of thickness: a stratification of cultural codes made up of a multiplicity of models. We regard these models as presuppositions common to the poet and the implied audience. The narrative text employs the strategy of eliciting these assumptions as mutually cooperating implications. Seen from this angle, the first striking aspect is the sheer extent and volume of the relationship between Vergil and Homer. Just think of the endless comparative material that commentators on Vergil have accumulated throughout its history. This rich tradition of scholarship tends by its nature to fragment the texts; mostly it documents the contact points between specific speech-­acts and progressively reframes them as minute and static particles; often it juxtaposes without interpreting anything. Naturally, any current analysis along these lines cannot forget its debt to the tradition of furta Vergilii. But it is equally clear that no adequate account of the Vergil-­Homer relationship can be achieved through a mere stockpiling of individual references, especially if the overall process, the composition of the narrative text, is left out of the picture. The unusual extent of intertextual contacts itself seems to require tools that go beyond the dimension of the arte allusiva, an analytical method (it is well to recall) that was first developed by Pasquali 1951 to deal with the specific microcontext of poetic memory. After all, this concept of allusion was based—­in those first examples that Pasquali originally selected—­on imitative relations found in the narrow compass of a single context through which the poet establishes, in a specifically Alexandrian sense, his own literary culture. But this by no means exhausts the experience of readers who compare Vergilian and Homeric narration, which has to do not just with single allusions but also with plots, story lines, narrative conventions, type scenes, and the construction of episodes. The imitation of Homer, as readers actually experience it, concerns every level of the text, from forms of expression at the microlevel to larger forms of content. What is needed therefore is to bring some order to these various types of intersection. As a start, let us try to consider Vergil’s relationship with the Homeric texts at the most general level. To structure the plot as well as the discourse Vergil is already working

T he D eath of T urnus

71

on the fabric of the Homeric story; it serves him as a kind of script or screenplay, but one he works with according to a number of strategies. From this perspective, the Aeneid obviously performs a suite of functions on the Homeric models: a. First and most notably, Vergil attempts a “combined montage” of the Odyssean and Iliadic stories; in this, as we know, he had Apollonius and Naevius as precedents.1 b. But the Aeneid also aims to be a continuation of the Homeric poems. This needs some clarification. Even if the Vergilian narrative has certain “Cyclic,” “post-­Homeric” qualities—­and despite the fact that specific imitations of the Cyclic texts have been detected—­this is not exactly what I mean by continuation. I am thinking instead of how the Aeneid, even as it narrates its own story, does not neglect to “fill” the gaps left by the Greek model; the reader follows Aeneas and encounters the post-­Homeric fortunes of Antenor and Neoptolemus, Priam and Diomedes. The new narrative also justifies itself, at least partly, as a sequel to Homer’s stories, particularly the Iliad. The protagonist ­Aeneas begins his own journey with an apparition of the shade of Hector, the hero buried and honored in the last verse of Homer’s poem. To tell how Andromache lives without Hector, or what Diomedes thinks about when recalling the war at Troy, is a small but significant part of the literary mandate Vergil claims for himself. Are there precedents or analogies for such a peculiar sort of intertextual filling in the gaps? Seeing the Aeneid, along with some recent critics, as a kind of New Testament has the merit of emphasizing the uniqueness of this phenomenon, though the ideological subtext has its perils. Instead, notably, a closer parallel and precedent was at hand, once again, in Homer himself. The complex texture of relations that bind the Aeneid to the two Homeric poems seems analogous to what binds the Odyssey to the Iliad. This is a familiar enough phenomenon. The Iliad is a poem nearly without a past; we learn very little about events prior to the few days directly narrated and even important premises are omitted. By contrast, the Odyssey is full of references to the past,

72

C hapter 4

and this past tends either to meld and match with, or else to complete the story of the Iliad by filling in intermediate gaps. Interestingly enough, Alexandrian critics already observed, using a fine image for their interrelations, that the Odyssey “fills in” the gaps left by the Iliad: it tells the stories of Achilles and the two Ajaxes, Agamemnon and Menelaus, of the wooden horse and the return voyage with its great storms; entire episodes, like the accounts in the Telemachy, the songs of Demodocus among the Phaeacians, and the two underworld scenes, serve the specific function of resuming, developing, and completing matters already treated in the Iliad. “What [the poet] had left off midway in the Iliad he returns to in the Odyssey, as though he were treating it all as one whole”; “the Odyssey may justly be described as a completion of the Iliad.”2 Anaplerosis, “filling in” of gaps left in the model. This is perhaps the description we are looking for. It’s quite possible that Vergil learned how to continue Homer from “Homer” himself, who offered a key to this technique in the way he continued “himself ” in the transition from Iliad to Odyssey. c. Then there is the schema of repetition. In certain episodes the Vergilian story is not so much a continuation or completion of Homer but rather at least potentially a repetition. For instance, the second half of the poem is marked at times by the idea of a second Trojan War. The idea that the Trojan War could recur, and that Turnus will figure as a new Achilles, has obvious implications of tragic irony.3 d. But these traces of repetition go on to sustain yet another intertextual figure that we can describe as inversion and overcoming. The wandering Aeneas slips the clutches of Odyssean marvels and avoids the traps that ensnare Odysseus; his journey is one of leaving a known land in search of an unknown world, the exact opposite of the return to Ithaca,4 and the Trojan War will not repeat because the new poem is not about the destruction of a city but the construction of new walls. In this sense the fabulae of the Iliad and Odyssey also play a role in the narrative by reappearing in a reversed direction: a search for a new land and the foundation of a city instead of destruction of a city and homecoming.

T he D eath of T urnus

73

We have listed several modes of contact between the fabulae, the “scripts,” of Vergil and Homer. But note that this was possible to do without ever mentioning another aspect of literary imitation. In fact, Vergil could very well have restricted himself to reproducing, citing, “copying” Homer, without ever in the least bit writing like Homer. Obviously, to write like Homer does not mean replicating him, that is, treating his texts as the only ones, as exemplars to be reproduced, and reproducible and quotable because they are unique and singular; the problem was rather to remake and “replace” Homer by extracting from his texts a sort of generative matrix. In other words, one can begin in this way to distinguish between two rather different functions under the single concept of imitation. One may add also that generally speaking there is no imitation that limits itself to the text (otherwise it would be a case of sheer reproduction). What one really imitates are styles, conventions, norms, genres. To imitate a text, then, presupposes at the outset that it has been constructed as a model—­a terribly ambiguous word, of course. It can be applied to a text, viewed as a concrete totality, but also to a set of distinctive features, a generative structure. Between imitating text and imitated text this stands as an authoritative model that inevitably generalizes certain traits of the text for imitation; it selects them, identifies them as typical, and constitutes a matrix of imitation. “To write in the manner of,” to remake a past text, is only possible if that text is treated as a “generic” model, possessing generalizable traits.5 (From this perspective there is no great difference between psychologically quite distinct activities like aemulatio, pastiche, and parody. It is clear that what is conspicuous in these activities are not the original, intended qualities of the exemplar text but those generic qualities of that text that the imitator or parodist takes as salient, selects over others, makes into a system and puts it to work. (One thinks of how the tradition of epic parody has varied over time.) To summarize, we can say that Vergil uses Homer in two noticeably different ways: as a script and as a repertoire of genre. On one hand, the Homeric text serves Vergil as an individual exemplar to copy; it offers, for instance, a series of individual events that can be reproduced or endowed with an allusive dimension. Ancient culture made these into the very hallmark of heroic epic, which was full of exceptional scenes and images: the marvelous and supernatural

74

C hapter 4

encounters of Odysseus; the wrath of Achilles, the fatal spoils of Patroclus, the divine armor that descends from heaven to Achilles; Hector’s words to his son; Andromache’s lament; Priam’s appeal to Achilles. On the other hand, the Homeric model is not for Vergil simply a weighty collection of famous, exemplary commonplaces. It also constitutes a veritable repertoire, which is the very matrix of imitation. Here come into play all those distinctive features of Homeric poetics that cannot be reduced to marked, individual occurrences. First and foremost, of course, is the repertory of epic formulas, with all its verbal associations strongly codified in rhythmic and lexical terms; then all the topoi, the figurative types that are less lexicalized but still recurring; as well as the most general combinatorial rules that show by their repetition how narrative material can be organized and given form. For the sake of clarity we can call the first aspect of the Homeric model example model and the second genre model. Two intertextual tendencies correspond to these two modes. In the first case the Vergilian text refers by implication to a specific site in the Homeric text. In the second it takes advantage of repetitive elements that might be lifted indifferently from various contexts and easily assembled into typical paradigms; these are readily inflected and put to use in producing the new text. Some important questions of poetics are connected to this second aspect and merit brief attention here. The codified, formulaic character of Homer with its repetitions and type-­scenes is connected with a series of diachronic changes that affect the sensibility of readers of Homer. This characteristic is so constitutive of Homeric epic that it cannot be ignored, but it is also a major problem for those who receive the Homeric texts outside the cultural context that engendered it. The problem as Vergil would have seen it did not differ much from the perspective that Homer’s Alexandrian readers would have had—­nor for that matter from the position of many modern readers. The generic and repetitive qualities of the Homeric text makes Homer both easy to imitate but also difficult to “remake.” Indeed, these qualities were generated within a specific cultural setting to which “post-­Homeric” readers would no longer possess the coordinates. The most visible aspect of the genre model is of course the hefty amount of formulaic diction. All the phenomena that we

T he D eath of T urnus

75

summarize under this label are clearly tied to a poetics or, if you will, to a system of attitudes and approaches that link together the modes of production and reception of the text. The poet manages the repetitions with technical skill; his intended audience knows how to recognize in this skill specific communicative and aesthetic affects—­ just as, and this is important, they do not ask of formulaic diction more than it is able and meant to express. In changing the modes of composition and use, the Homeric text is then exposed to needs and concerns for which it was not originally intended. Homer’s new readers cannot help but interpret the formulaic diction as a stylistic choice; the system of epic repetitions, formed as token and trace of a mode of “oral” composition and bound to a kind of communicative pact between author and audience, now seems out of place and in need of new motivations. This tension gives rise to interesting compromise solutions. Readers and imitators of Homer fervently wish to assimilate the great model of the epic genre into their own poetic horizons, but then they cannot avoid the problems that arise from the clash of conflicting poetics. Seen as stylistic elements or stylemes, formulaic expressions can seem unmotivated within their context. Of course not all formulae create problems. Readers of Homer can accept, for example, that common and repeated actions are expressed in regular and repetitive ways; the formulae of speaking, of eating and drinking, of dying and of sunrise, are repetitive but not unmotivated in their respective contexts. (Even Vergil the imitator, as we will see, deigns to create para-­formulaic connections for these situations.) Still, these readers have difficulty accepting the system of ornamental epithets; not that the epithets lack meaning (except for a handful that already in the archaic period sounded mysterious), but because many epithets are indeed unmotivated within their context. Strangely, the ornamental epithets turn out to be now, in the framework of a naturalistic poetics, the most arbitrary and least manageable part of the epic repertoire, when originally they were the most organic and predictable tool of formulaic diction. The fact is that Alexandrian readers of Homer preferred epithets that suited the situation and were inclined to trade the generalizing nature of epithets for a surplus of meaning that fit the occasion. If the Scamander River says πλήθει γὰρ δή μοι νεκύων ἐρατεινὰ ῥέεθρα (“my lovely

76

C hapter 4

streams are choking full of corpses”; Il. 21.218), ancient commentators observed that the epithet ἐρατεινά [lovely] suits the psychology of the context: it expresses the river’s indignation at seeing its own waters polluted. Naturally, not every noun-­epithet pair could pass such a “test for motivation.” There was thus no alternative than to accept, and theorize, the existence of ornamental epithets, chosen κόσμου χάριν (“for adornment’s sake”), and suited to the nature of the thing rather than to how things stand in individual cases in light of particular contexts. But this category does not overcome every difficulty either. The same professional exegetes that theorized this stylistic practice were bothered that Nausicaa’s dresses were called, when they were being taken out to wash, “splendid dresses” (Od. 6.58), or that Priam, old and unfit for war, is described as “good with the spear” (Il. 4.47; putting him to the test, Vergil’s Priam will only be able to hurl a telum . . . imbelle sine ictu [“an unwarlike spear without force,” 2.544]). They liked even less that the characters speak the same formulaic language used by the poet in the diegetic portions (where they were slightly more tolerant of it). At Iliad 23.582, Menelaus, raging against Antilochus over a sporting event, calls Antilochus διoτρεφές (“Zeus-­ nurtured”); at 3.352, Menelaus again prays that Zeus give him victory against the “divine” Paris; at 11.123, a character who is bribed to vote against returning Helen to the Greeks is called δαίφρων, which perhaps means “wise, prudent”: some philologists want to emend this to κακόφρων (“malevolent”). Alexandrian readers of Homer thus wanted a narrative where the characters’ speeches would normally be governed by the immediate situation and by psychological realism, and where the poet would be responsible for every lexical choice not in relation to his tradition but to his own story, bound to an ideal of organic naturalism. Of course readers of Homer understand, however, that Homeric narrative is not built to fully satisfy these new desires for verisimilitude. In this way conflict zones arise between the original poetics, the set of conventions that enabled the production of the Homeric text, and the poetics of the new audience, who would prefer somehow to assimilate the Homeric text to themselves in order to better decipher its meaning. The questions of poetics posed by his own genre model therefore gave Vergil responsibility for the form of his own story. The generic

T he D eath of T urnus

77

and repetitive aspect of epic diction was by then a fundamental mark of the literary genre, but it also carried baggage that limited its narrative functionality and jostled against a poetics governed by a different idea of verisimilitude and contextual motivation. Ways to begin and end a character’s speech are not very flexible, and no one expects them to be adapted to the psychological content of the discourse and situation. The rigidity of epithets is geared to a style undoubtedly heroic but too stylized and overworked (of roads always wide, lords divine, and ships swift). The paratactic piling up of the narrative pacing, conditioned in turn by the syntax, ends up affecting the level of compositional structure, discouraging, for example, long-­range links between episodes. Vergil’s solution, as we might expect, consists of a compromise between regeneration and transformation of the epic repertoire. Ornamental epithets once again give a good clue to describe this better. In Vergil’s text they appear not as mere residual traits but as a true mark of elevated style (and this way of writing corresponds, as we have seen, to a certain way of reading Homer: interpreting the traces of the formulaic system as a set of stylemes). We can imagine, with some inevitable overstatement, what function epithets played in the communicative system of archaic epos: they afforded the audience the pleasure of repetitions seen and felt beforehand, and of the skillful artistic redistribution of traditional paradigms; perhaps they even offered assurance of ideological stability and a vital connection between the audience’s experience and the world recounted in the text. By absorbing ornamental epithets into his own poetics Vergil could not help but estrange them from their original function. Homeric verse had attained the status of a sublime model of poetic language; much of its defamiliarized character derives precisely from this cultural discontinuity, where a text from an oral tradition is made into the absolute exemplary model for written poetry, obscuring thereby its own inner laws of composition and the vitality of the code that generated it. Thus the presence of rigid descriptions unfitted to their context suggests to imitators of Homer a sort of metaphorical swerve; it then becomes typical of elevated poetic language to contain a degree of resistance to functional communication. In other words, epic epithets inspire a sort of figurative redundancy that in effect becomes a stylistic figure. By preserving it, even

78

C hapter 4

if in a highly selective manner, Vergil could not possibly avoid genuinely estranging effects. In Homeric diction, a phrase like οὐρανὸν ἀστερόεντα is so fixed that sometimes a character in full daylight could raise a prayer “to the starry heaven” (Il. 15.371); in turn Vergil uses a pathos-­laden phrase like ferit aurea sidera clamor (“clamor strikes the golden stars”) not only in a nighttime scene—­where we would expect it due to his narrative poetics that is so attentive to coherence and the motivated choice of details—­but also in a daytime scene. Ferit aurea sidera clamor is used at 2.488 of the tragic night in Troy; at 11.832 the same image, ferit aurea clamor / sidera, describes the violent (daytime!) battle that follows Camilla’s death. Here Homeric clichés are sampled and reused as such, in a way that their fixity becomes a mark of epic, a figure of elevated style that resonates with its new context. Another, more subtle strategy is opposite this one on the axis of what we have called transformation. These are instances of adaptation where the reader is led to seek new motivations for the use of a traditional expression. The Homeric epithet ἐραννός, in the sense of “fair, lovely,” is always used of places and appears in formulaic pairings with place-­names like Καλυδῶνος ἐραννῆς (“lovely Calydon”; Il. 9.531 and 577). At A. 11.270 this fixed phrase is literally translated pulchram Calydona. But the translation does not just convey a typical quality (and thus serve as a recall of the epic repertoire); “lovely Calydon” is now a subjective perception well-­suited to its context. Diomedes says: “if only I could return to my fatherland, coniugium optatum et pulchram Calydona viderem!” The fatherland seems fair and lovely because of its aura of nostalgia, similar to the dulcia . . . arva at Ecl. 1.3. So the epic cliché Καλυδῶνος ἐραννῆς is not only translated (and with precision: the rare ἐραννή is glossed as καλή in the Homeric scholia ad loc.) but remotivated as well, expressing the point of view of an exile. Again at A. 10.862ff., Mezentius, speaking to his warhorse, expresses his eagerness to avenge Lausus: aut hodie victor spolia illa cruenta / et caput Aeneae referes Lausique dolorum / ultor eris mecum . . . (“Today you will either bring back, in victory, the head and the blood-­stained spoils of Aeneas, and you will be the avenger of Lausus, together with me”); Vergilian critics have not noticed that spolia cruenta is an exact equivalent of one of the most common Homeric battle formulas, ἔναρα βροτόεντα or αἱματόεντα, a generalized

T he D eath of T urnus

79

expression used in contexts of description (for example, Il. 17.540 “he took up the bloody spoils and laid them in his chariot”). Vergil here reworks it into the subjective view of a character in action, thus recharging the cliché “bloody spoils” by the context, where Mezentius wants to behead Aeneas and already sees, literally, his blood flowing (instead, at 10.908, it will be Mezentius’s blood that soaks the armor, undantique animam defundit in arma cruore).6 Broadly speaking, then, the reuse of the genre model involves two basic strategies. The cliché can be recuperated “as such” and inserted like a mosaic tile colored by contrast; there is then a play on the arbitrariness of the image, which acts to recall an antique code of expression. Or else the cliché can find new motivation in context, which contrasts with the greater rigidity of the Homeric usage—­ here Vergil’s writing style appears parallel to how the Alexandrians read Homer because it responds to the same needs and the same discomforts. Yet the Homeric text could also function—­as already implied in our “generative” conception of the Homeric model—­as a stimulus to produce “new” epic clichés. In other words, not all of Vergil’s expressions that display a repetitive, para-­formulaic quality are directly evoked by an exact Homeric counterpart. The Homeric matrix also works to stimulate novel para-­formulaic expressions. Here as well, we can distinguish between connotative markers of epicness that lack contextual motivation7 and para-­formulaic structures that evoke particular effects of remotivation. Compare the following two passages: ubi tot Simois correpta sub undis scuta virum galeasque et fortia corpora volvit (A. 1.100–­101) [where the Simois swept down under its waves rolling shields and helmets and strong bodies of so many men] quam multa sub undas scuta virum galeasque et fortia corpora volves Thybri pater (8.538–­40) [how many shields and helmets and strong bodies of men you will roll beneath your waves, father Tiber]

80

C hapter 4

The description, from Homeric tradition, of a river clogged with bodies (cf. Il. 12.22–­23; 21.214ff.), at first seems simply reproduced, with only the minimum syntactic variation required by epic repetition, first about the Simois and then again about the Tiber. But in these two contexts, from books 1 and 8, the lines serve as a precise marker of recollection and of opposition. In book 1, Aeneas trembles to hear the blasts of the storm sent by Aeolus and laments that he did not die along with so many other great heroes in the Simois at Troy. In book 8, Aeneas hears the thunder of Venus and presages the great victory over the Italians, which will cost so much blood. In the first episode the Simois represents the bloody past of the exiled hero longing to die; in the second, the Tiber points toward the bloody victory over new enemies. A strong sense of narrative construction gives both sequences not just a sense of repetition but also a mirroring effect.8 The repetitive, generic, and “serial” quality of Homeric verse constitutes an unavoidable reference point for anyone wanting to write in the Homeric manner and follow in the epic tradition. On the other hand, this abundance of codified elements also constitutes a supporting framework for original creations that stand apart from linguistic and narrative conventions; or better, they receive light from them, since the Homeric works that Vergil took as his example are criss-­crossed by a dialectic between convention and originality. This puts Vergil in a particularly delicate position. Indeed, the two aspects of imitation, so far for analytical purposes considered as separate and contrasting, are really intertwined with one another. And this not only in the Homeric texts. Vergil himself, in composing his own work, reactivates both the generic and serial matrices and the individual exempla offered by Homer. Vergil’s literary skill exhibits a marked sensitivity to both the typical aspects and the technical aspects of Homeric narrative art. This can be demonstrated by analyzing discrete narrative sequences, and this double sensitivity also reveals itself in a synchronic manner. It is common in epic for a warrior, when accomplishing feats in battle or about to have his valor tested, to be compared to a lion on the prowl. The epic code, as we know, has certain fixed traits for comparison, such as courage and recklesness, his solitary nature, his

T he D eath of T urnus

81

passion for blood. Clearly these points of contact between lion and hero are linked to the conception of the warrior in an aristocratic ideology. In line with this epic convention Vergil describes Turnus setting out for vengeance at the start of book 12 as a raging lion: Poenorum qualis in arvis saucius ille gravi venantum volnere pectus tum demum movet arma leo gaudetque comantis excutiens cervice toros fixumque latronis impavidus frangit telum et fremit ore cruento: haud secus accenso gliscit violentia Turno. (12.4–­9) [As in Punic fields, gravely wounded in the chest by hunters, a lion then will take up arms and thrills to puff up the mane on his neck, fearless he snaps the shaft the hunter has pierced him with and roars with bloody mouth; so the violence swells in Turnus burning with rage.] The comparison immediately strikes us as Homeric in the sense that, although not traceable to any single model passage more closely than another,9 making it difficult for readers to trace it to a specific Homeric locus over other possible parallels, it is based instead on a paradigm, a typology of lion similes such as we have described earlier in schematic outline. This epic typology combines structural formulas, verbal recurrences, common narrative events, and assumptions (some ideological, some iconographic). Use of this generative paradigm fully harmonizes with the narrative’s explanandum, where Turnus, beaten down by the course of the war, only strengthens his violent charge and lust for combat. But one trait in the simile is left unexplained in this setting and, for another thing, I hasten to add, is not part of the typology of the Homeric comparison: the idea that the lion has been gravely wounded in the chest, saucius  . . . gravi venantum vulnere pectus (5). This detail is of some interest since it has no correlate in the immediate context of the image. It becomes even more significant, however, if we are inclined to regard the comparison also as the πρόσωπον τηλαυγές (“far-­gleaming facade”) for all of book 12. Situated in the first verses, the simile would serve to spotlight the protagonist of the tragic action that will occupy the

82

C hapter 4

whole book, anticipating both the character and his fate. But can an epic simile be positioned to discharge its narrative valence over such a long arc of narrative? (The lion’s chest wound at the start of the book anticipates Turnus’s reckless courage and indeed already looks forward to Turnus’s chest pierced by the enemy’s sword in the book’s last lines: 12.950, ferrum adverso sub pectore).10 This brings us back to the many-­sided virtues of the Homeric model. We have said that the lion similes (with others of the same sort) generally occur within aristeiai; in such contexts it is not clear what function the idea that the animal is (already) gravely wounded might serve. Yet there is one significant exception to the common iconography of the lion in Homer:11 when Patroclus pounces on the dead Cebriones to despoil his armor he is compared to a lion “that has been wounded in the chest and his own strength destroyed him” (16.753). Ancient commentators already looked closely at this detail;12 the lion is gravely wounded because the poet is narrating the last phase of Patroclus’s aristeia. Right after killing Cebriones, Patroclus is swept by his courage into a fray with no way out. Patroclus is about to die like a lion that ἔβλητο πρὸς στῆθος, a lion wounded in the chest like Turnus heading into his final battle. Here, as I have said, Homer supplies on one hand a grammar of narration, according to which Vergil declines the image of the lion; on the other hand, he also suggests a pathetic lexicon, a unique example that can be resumed and cited through allusion. Imitation has both grammatical and lexical dimensions. This distinction will be useful for interpreting an episode of great importance, both for its density of Homeric traces and for the crucial role it plays in the poem’s plot, a scene so controversial that it always causes some embarrassment whenever it comes up: the killing of Turnus at the end of the Aeneid. A reasonable starting point to study this brief episode is to review the copious bibliography that this scene has generated in recent decades. Some of the proposed interpretations are so partial and reductive as to be almost caricatures. They oversimplify because they fail to respect the objective complexity of Vergil’s text. Aeneas is viewed either as a priest solemnly processing toward the execution of an enemy of the Roman state or else as a war criminal who, at the decisive moment, betrays

T he D eath of T urnus

83

the very values he is trying to establish. These solutions say a lot about the ideology of those who offer them and little about how the Vergilian text is made. Of course, a text is not polysemic simply because historically it has been interpreted in many contradictory ways. Still, a text can be made in such a way that it generates a problematic interpretation. This is just what seems to emerge from an analysis that, discarding all false oversimplifications, tries to interpret the meaning of the text with philological acumen. I believe many scholars of Vergil could agree with the following summary: “What the poet has done with this ending has been to make as difficult as possible, in moral terms, the problem of whether to kill Turnus or spare him.”13 This “difficulty” intentionally produced by the text—­which readers of Vergil objectively perceive even if with understandable differences of emphasis—­has very little to do with the Homeric examplar that Vergil presupposes: the death of Hector in Iliad 22. This model is so clearly present in Vergil’s text that it directs our expectations in a precise and univocal direction. In the previous chapter we analyzed how this allusive tracing is developed; the two episodes function in the plot analogously and are developed in the narrative along parallel lines. Vergil, as we know, has allusively readapted countless points of detailed correspondence.14 It is now time to focus attention on the differences between the deaths of Hector and of Turnus. An initial example may help bring into view the different trend of events that Vergil inscribes on his story. When Achilles hurls the spear that will strike Hector’s final blow, the poet pauses to spotlight this decisive moment with a simile. Consistent with Homeric poetics the simile, occurring at the crucial point in an aristeia, highlights the conquering hero’s triumphant force in terms of fiery splendor. Achilles’ spear is like the star that outshines the rest in the night sky, Hesperus, the fairest star of all: so shone the sharpened point of the spear Achilles was launching (Il. 22.317ff.). The comparison is entirely focused on Achilles’ glory; no part of the image looks toward the imminent fate of Hector, who is about to die. (A little before, at Il. 22.25ff., Priam spots Achilles from afar and he shines like a star, the Dog Star of Orion, “the brightest star of all, but an ill omen, / and it brings great fever upon wretched mortals.” Seeing him, Priam groans, a reaction fully prepared for by the foregoing simile.) In

84

C hapter 4

Vergil the spear of Aeneas that is about to bring down Turnus is instead compared to destructive phenomena: a stone hurled by a ballista, a thunderbolt, a whirlwind. Here too the narrator emphasizes the hero’s strength as a warrior, but the emotional center determining the choice of images lies elsewhere. Achilles’ spear is like the most beautiful evening star, while Aeneas’s is like a dark whirlwind: volat atri turbinis instar / exitium dirum hasta ferens (923–­24). The choice of epithets (ater, dirum) tends to absorb the reader into Turnus’s perspective and shifts the narrative focus from victor to vanquished. This shifted perspective is important; it leads the reader to participate to some degree in the subjective point of view of Turnus. Not surprisingly, after Aeneas hurls his spear the text devotes significant space to Turnus’s behavior and the reasons for it. The first point to be considered apparently concerns what we might call a technical aspect of combat. Hector, mortally wounded, can do nothing besides pronounce a series of mandata morituri; he knows his end has come and he can do nothing but plead for the fate of his body. In addition, because of his circumstances, his speech does not measure up to a true and formal supplication.15 The only hint in this direction is the brief, conventional appeal (cf. Il. 15.660) “by your life, your knees, and your parents” (22.338). Herein lies a Vergilian transformation that fundamentally alters the coding of the scene. Indeed, Turnus is not wounded in a vital organ; his knees have buckled to the ground only because his thigh, pierced by Aeneas’s spear, cannot hold his body’s weight. Turnus thus finds himself, just before starting his appeal, duplicato poplite (“on bended knee”):16 his pose vividly displays his condition as supplex.17 Humilis supplex (“on the ground [ / humbly] and suppliant”),18 Turnus accompanies his speech with the ritualized gestures that Homer’s Hector was “unable” to offer Achilles: oculos dextramque precantem protendens. Extending up and outward the hands and arms (and eyes) often corresponds in Vergil to a request for pity. In another case marked by pathos, the eyes appear as a substitute for the hands. At 2.405–­6, as she is dragged off by her conquerors, Cassandra is shown ad caelum tendens ardentia lumina frustra / lumina, nam teneras arcebant vincula palmas (“holding her burning eyes to heaven in vain, her eyes, for chains bound her tender hands”). What is more, here the gesture of supplication already implied in the regular phrase dextram

T he D eath of T urnus

85

(palmas) (pro)tendere is amplified and strengthened by the epexegetical precantem.19 It is thus a conspicuous choice on Vergil’s part to characterize the last episode of his narrative as the slaying of a suppliant warrior, a choice (as we will see even more clearly in a bit) with no analogue in the Iliad’s model of the death of Hector. The meaning of this choice should undoubtedly be examined in the particular context of Vergilian narration; yet a scene so unique and memorable deserves to be considered not just iuxta propria principia, but also—­in light of what we have called the genre model for Vergil—­as a typical situation for epic storytelling. Examining the typology of ancient epic reveals, in fact, that this situation, a defeated warrior supplicating his victor, recurs very often and always follows an underlying pattern. The consistency of this pattern lies in the fact that (as far as I know, in all of ancient epic before Lucan)20 in the many cases of attempted supplication the outcome is fixed and constant. The suppliant warrior is never spared. In other words, the problem Aeneas faces with the suppliant Turnus is hardly unique, and it would be wrong to read this epic situation without measuring it against the narrative typology that it involves. In Homeric and, in evident parallelism, Vergilian epos, no prisoners are taken. This is obviously no mere statistical fact, and even less does it attest to practices of warfare that the poet implies as universally valid. Homer’s characters themselves are perfectly aware of the custom of sparing the conquered, and Achilles himself recalls having done so in the past. The possibility of clemency is thus always present as an alternative choice. For example, at Iliad 6.55ff., when Menelaus seems inclined to spare an unarmed Trojan who has grasped his knees and promised ransom, Agamemnon upbraids him. No Trojan is to be saved. Troy is to be destroyed, down to the last baby in its mother’s womb. The suppliant is summarily slain. And so, on every similar occasion, to kill is a contingent choice, but it is the only one ever made, a restriction motivated by the fact that Homer is not narrating an episode of war among many others, but the bloodiest and most ruthless stage of conflict without quarter that will end in Troy’s destruction. The same, and more so, holds for Vergil. For him also, the theme of epic is (dicam horrida bella) a heightened war carried to extremes, heroic warfare where one can

86

C hapter 4

only kill or be killed. It turns out then that Vergil’s genre model—­ the typological coherence of Homeric narrative—­predicts and suggests only one kind of solution. And if Vergil’s readers turn their attention to this Homeric basis for the narrative, they see there a univocal trace: Turnus cannot but be slain. But as more sensitive readers of Vergil have always correctly perceived, Turnus has good reasons to expect success and the text gives ample space to Aeneas’s reaction to his appeal.21 How then does the Vergilian text make Aeneas’s choice “difficult”? We have seen that the directive force of the models seems to create a one-­sided expectation: anyone who writes “the reader expects Aeneas to show mercy”22 tends to tear the situation a bit too far out of its natural frame of reference, the epic tradition. The difficulty lies in the last words Turnus addresses to Aeneas. This speech contains no relevant point of contact with Hector’s words. And yet, Turnus’s appeal: miseri te si qua parentis tangere cura potest, oro (fuit et tibi talis Anchises genitor), Dauni miserere senectae. (12, 932–­34) [If any care for your sad father can touch you, I pray—­and your father Anchises was the same to you—­pity Daunus in his old age.] has an undoubtedly traditional ring, but not because it repeats a commonplace motif. On the contrary, it reprises a unique motif within the Iliad, a memorable and easily identifiable occurrence even in its allomorphic form here. As every reader of Homer knows, the appeal “have pity, think of your aged father” is the line that moves Achilles to pity in Iliad 24: the only episode in the entire poem where the supplication of an enemy is heeded.23 Literal citations of this Homeric example are familiar to readers of Vergil: at 1.459ff., Aeneas sees before anything else in the temple at Carthage the image of Priam (en Priamus, 461), the pathetic Priam supplicating fierce Achilles; this is the only detail Aeneas points out to Achates and the only view that evokes a deep groan of pity (tum vero ingentem gemitum dat pectore ab imo, 485). At 2.535ff., Priam reminds Pyrrhus (who soon will pay for his brutality with his

T he D eath of T urnus

87

life) of Achilles’ treatment of the elderly supplex.24 This memorable Iliadic scene is tied directly to the means by which Priam moved and persuaded Achilles. But before turning to the “rhetorical” content of this speech, we can already identify a certain homology in situation between Vergil and his model. It has been noted that Vergil could not emphasize any more clearly Turnus’s status as a suppliant of Aeneas. These signs should not be taken as pointers to the main model (the death scene in the “Slaying of Hector”) so much as to the iconography of another episode, Priam’s supplication of Achilles in the “Ransoming of Hector.” It is important to go back to the narrative construction of the Homeric model, which repeatedly focuses on specific gestures: χερσὶν Ἀχιλλῆος λάβε γούνατα καὶ κύσε χεῖρας ἀνδρὸς παιδοφόνοιο ποτὶ στόμα χεῖρ’ ὀρέγεσθαι ἁψάμενος δ’ ἄρα χειρὸς ἀπώσατο ἦκα γέροντα γέροντα δὲ χειρὸς ἀνίστη οἰκτίρων

478 506 508 515

[He grasped Achilles’ knees with his hands and kissed his hands Raising to my lips the hand of the man who killed my sons Taking his hand he set the old man gently away He raised the old man by the hand, pitying him.] The dynamic of Vergil’s scene similarly plays on the gestural correspondence between suppliant and supplicated: ille humilis supplex oculos dextramque precantem protendens “. . . victum tendere palmas” Aeneas, volvens oculos dextramque repressit.

930 936 939

[He lowered his eyes in supplication and stretched out a pleading right hand . . . Defeated, stretching out my hands . . . Aeneas turning his eyes stayed his right hand.] It is clear that gestures—­for example, the repeated focus on the characters’ hands—­function as indicators of emotion, or as signals of choices and intentions that point to one another. In particular,

88

C hapter 4

Turnus’s tendere palmas or dextram protendere may well be a recall of Priam’s χεῖρ’ ὀρέγεσθαι. Indeed, ancient and modern interpreters of Homer waver between two possible interpretations for this gesture: “extend the hand” (toward the supplicated, to touch his chin) or “bring the hand (of the supplicated) toward the mouth (of the suppliant)” (the latter interpretation is by far the more likely).25 What is of interest here is only the “reading” of the model that Vergil adopted. This for once is possible for us to determine thanks to an internal parallel. At A. 1.453ff., Aeneas is moved by the view of Priam figured in the act of ransoming his son’s body; the character is fixed in the pathetic gesture of extending unarmed hands (tendentemque manus Priamum conspexit inermes, 486). Finally, we may emphasize that in Vergil just as in Homer the gesture of supplication is first “spoken” by the poet, then emphatically echoed by the suppliant in his speech, in a way that results in an extraordinary act of self-­abasement (“He kissed that hand,” Il. 24.478; “I have suffered what no other mortal has, / kissing the hand of him who killed my children,” 24.505–­6; dextramque precantem / protendens, A. 12.930–­31; vicisti et victum tendere palmas / Ausonii videre, 12.936–­37). The similarities between Turnus’s and Priam’s supplications are even more poignant when we consider the content of their speeches. miseri te si qua parentis tangere cura potest, oro (fuit et tibi talis Anchises genitor), Dauni miserere senectae et me, seu corpus spoliatum lumine mavis, redde meis. (12, 932–­36) [If any care for your sad father can touch you, I pray—­and your father Anchises was the same to you—­pity Daunus in his old age, and return me, or if you prefer, my body despoiled of life, to my people.] The thrust of Turnus’s appeal—­the only one that might engage the enemy and evoke his pity—­recalls the great pathos-­laden motif that opens and closes Priam’s speech in the tent of Achilles:

T he D eath of T urnus

89

τὸν καὶ λισσόμενος Πρίαμος πρὸς μῦθον ἔειπε. “Mνῆσαι πατρὸς σοῖο, θεοῖς ἐπιείκελ’ Ἀχιλλεῦ, τηλίκου ὥς περ ἐγών, ὀλοῷ ἐπὶ γήραος οὐδῷ . . .  . . .  . . .  . . .  . . . ἀλλ’ αἰδεῖο θεοὺς Ἀχιλεῦ, αὐτόν τ’ ἐλέησον, μνησάμενος σοῦ πατρός· ἐγὼ δ’ ἐλεεινότερός περ. (Il. 24.485–­ 87; 503–­4) [And beseeching him Priam spoke a word: “Remember your father, Achilles who looks like the gods, the same as me, on the sad threshold of old age . . . but respect the gods, Achilles, and pity me, remembering your father, for I am indeed more to be pitied.”] We are dealing here not just with a famous passage of great poetry but also a recognized model of effective persuasion.26 Ancient rhetorical theorists who discuss the emotion of eleos and the means to elicit it see in this scene the prime example of how effective the power of the word can be even in extremely unfavorable situations.27 “Speaking with an enemy, the slayer of one’s sons, who hates you and yours intensely for the loss of his friend, he could not have found any exhortation more apt than to remind him of his father”; καὶ σχεδὸν ἀντὶ Πριάμου γέγονε Πηλεύς (“and we almost have Peleus instead of Priam”).28 Tracing the lines of Priam’s speech, Turnus emphasizes the miserable nature of old age, which Daunus and Anchises would share in common: Dauni miserere senectae picks up ὀλοῷ ἐπὶ γήραος οὐδῷ (24.487), and the phrase fuit et tibi talis / Anchises genitor comes straight from 24.487 τηλίκου ὥς περ ἐγών, though not without influence from a parallel passage. At Il. 22.420–­21, already intent on facing Achilles to seek his pity and having pondered what tack to take, Priam says to Achilles: καὶ δέ νυ τῷ γε πατὴρ τοιόσδε τέτυκται / Πηλεύς (note also the metrical and syntactical resemblance).29 Again, the presence of the “model of Priam” is suggested also by an expressive refrain that runs through Turnus’s appeal: miseri te si qua parentis tangere cura potest . . . . . . Dauni miserere senectae

90

C hapter 4

ἀλλ’ αἰδεῖο θεοὺς, Ἀχιλεῦ, αὐτόν τ’ ἐλέησον μνησάμενος σοῦ πατρός· ἐγὼ δ’ ἐλεεινότερός περ (24.503–­4) [But respect the gods, Achilles, and pity me, remembering your father; and I am even more to be pitied.] The repetition miseri / miserere is built on the formal model ἐλέησον / ἐλεεινότερος because it repeats its moving pathos and, what is more, its rhetorical effectiveness, as though the figura etymologica contained in itself the entire persuasive force of the whole speech. Homophony aims to produce in the listener homoiopatheia, and the repetition of sense already anticipates the mechanism of compassion in the one toward whom the words are directed. The redundancy of the sign does not act simply as accretion (one thinks of cases like Pl. Ps. 4 quae miseriae te tam misere macerent), but is the motivated sign of a “rhetorical” intention.30 But Vergil’s rhetorical narrative also knows how to bring to light, in Turnus’s appeal, motivated aspects that subvert its intentions: the μεταβολὴ εἰς ἔχθραν (“change into hatred”) that the sight of the baldric provokes reveals within Turnus’s supplication a sort of negative predestination. On seeing the baldric, Aeneas’s indignatio instantly reverses the miseratio that Turnus had sought. The warrior who supplicates in the father’s name wears on his own body the spoils of a puer (Pallantis pueri, 943); the one seeking moderation from his conqueror was himself in turn a conqueror incapable of acting with measure (vicisti et victum . . . videre, 936; victum quem vulnere Turnus / straverat, 943–­44; even the alliterations confirm the recall). In a moment of nearly convulsive frenzy Aeneas seems bound to overturn every word spoken by his enemy (et ME, seu corpus SPOLIATUM lumine mavis, / REDDE MEIS, 935–­36; TUne hinc SPOLIIS INDUTE MEORUM / ERIPIARE MIHI?, 947–­48). The distribution of Homeric models in the short compass of the scene displays in its very making, so to speak, the project that governs Vergilian narration.31 On the intertextual foundation that he extracted from Homer, Vergil does not plot out any unitary direction to channel the flow of events in a way that would render them predictable. On the contrary, his use of the Homeric model corresponds to a problematic tension that stretches across the whole scene’s

T he D eath of T urnus

91

composition—­without impairing at all its coherent linear unfolding. Above all, the reuse of the Homeric model reveals Vergil’s profound “epic skill”: his ability to assimilate and master the institutionalized narrative language, taking from it both its typical aspects and its potentials for originality. With a genuine imitative tour de force Vergil invites us to read, in Aeneas’s hesitation, two contrasting traces—­two tracks mixed—­both signed by Homer. In Aeneas’s hesitation coexists, before the narrative cuts out, the track of Achilles the avenger from Iliad 22 and, less predictably but clearly presupposed by the text, the track of merciful Achilles from Iliad 24. Indeed, Turnus’s words summon for the reader Priam’s appeal, which in ancient culture was the universal paradigm of effective supplication. One might say that while he marks out his own ending on the traces of Iliad 22, Vergil does not fail to glance for a moment at the true ending of Homer: the sorrowful, but also somewhat reconciling, atmosphere of Iliad 24. The “alternative” ending—­which shows itself in Turnus’s supplication and Aeneas’s hesitation—­thus recalls the true conclusion of the Homeric text: that final stretch of the Iliadic plot that Vergil drastically cut short in his transformation of Homer.32 It is clear that this kind of narrative possibility, to be credible, at least so long as the course of action does not move beyond the bifurcation and bring about a different outcome, cannot rely solely on an intertextual connection with Homer. It requires something more: that the reader finds a possible correspondence—­that is, one coherent with the plot—­between Aeneas and the merciful Achilles of Il. 24, the Achilles who manages to overcome his rage as he allows a grieving Priam to transport him in thought to a grieving Peleus. But working along these lines are not just the intertextual traces Vergil evokes but also a whole series of connections internal to the text of the Aeneid. Faced with the deaths of Priam, Pallas, and Lausus, the most immediate thought for Aeneas is an image of paternity that binds in strong emotional bonds the fates of various characters.33 The allusion to the Ἕκτορος λύτρα is therefore nothing but the trigger of a narrative isotopy that makes us believe, at least for the space that it appears, that an outcome different from the “real” one is possible. In this sense, the intertextual references and the narrative isotopies “inside” the text cannot but work together to become mutually relevant.

92

C hapter 4

We are thus brought back to the dimension of intertextuality outlined in chapter 3. In a space of disjunction and uncertainty the text collapses the tension between conflicting models, and the tension is signaled to the reader through a specific intertextual effect that arises from the interaction of diverse models (it makes no difference whether, as in this chapter, all are lifted from the same model-­text). It is in this way that the Vergilian text makes its readers “work,” by opening contradictory spaces where a conflict of ideas is evoked through an extreme case. Without the narrated action losing its coherent development, the confrontation of differing possibilities maintains the need for critical reflection. It would be inconsistent at this point to claim that only readers who appreciate the complexity of the intertextual background can achieve this kind of understanding. I would say that the reference to Homer is certainly constitutive of the text—­since it is deeply ingrained in its composition, as we have tried to show—­but it is by no means indispensible for its correct decipherment. Not by chance, we began by citing a set of interpretations substantially adequate and focused but reached along divergent paths and often almost completely disregarding the Homeric background. This multiplicity of pathways corresponds to an objective and verifiable fact—­that is, the redundant and “hypercodified” character of a narrative text that, as though distrustful of overly arbitrary readings, multiplies its levels of coherence and communicative strategies, never trusting its decipherment to a single line of reading. The advantage of an “intertextual” reading like the one pursued here is not that it gives access to sophisticated hidden meanings, but that it helps explain and justify some simple reading effects that the text regularly emits, which other types of interpretation tend rather to paraphrase. For reasons I cannot explain, intertextual analyses are often accused of reading the text reductively, in formalistic and mechanical ways that deprive the text of its real and vital problematic history. In the case of the Aeneid’s closing scene, troubling conclusions of this sort have been more often reached through studies that are ideological and vigorously historicizing, with everything projected onto the political and social context. Indeed, these often begin with the most wide-­ranging research: it has been argued, for example, that Aeneas cannot help but kill Turnus because this corresponds to the idea of

T he D eath of T urnus

93

vengeance as Vergil’s public understood it. Aeneas cannot give way to mercy because that would privilege a value foreign to “Roman mentality.”34 This type of reasoning is wrong twice over: it is reductive both regarding the ideological problem and regarding the text. Regarding the text, it is a perilous assumption that a text has to depend fundamentally on the ideological horizons of its first readers, or some average of them. On the contrary, the best examples of “reception criticism” have shown us that there are innovating texts capable of challenging the dominant norms of ideology and morals. Regarding ideology, the problem of clementia toward the losing side was perhaps the central problem of Roman politics between Caesar and Augustus and thus can hardly be dispensed with in a schematic way and treated as a premise for interpreting the Aeneid.35 On the contrary, when we analyze the literary models that Vergil presupposes we begin to get a more adequate idea of the difficult dialectic between misericordia and ultio. To speak of “mentality” and “values” is rather weak when it comes to the concrete political urgency of the question. How to treat the defeated had been, for generations of Romans, a crucial decision that determined the fabric of civil society and affected the fate of Rome. There is no doubt that the final scene of the Aeneid is speaking to this as well.36 Literature, great literature, puts ideology on trial only as language in actual practice: in the ­Aeneid ideology must resurface as narrative language, structures of a story that accumulate a meaning. Models are integral to language, to the rhetoric of language—­in this case, of epic language.37 It is not simply the usual matter of paying respect to the specific nature of literature; what models help us grasp is precisely the historically determined meaning. Narrative disjunction (as noted in the previous chapter) is the very mechanism that enables—­in specifically literary terms—­the production of ideological conflict. But what controls these processes are the poet’s ability to give it form, and our own ability to interpret it.

APPENDIX The Lament of Juturna

At once sympathetic and blinded by love, compelled to act but conscious of her hopelessness, Juturna in Aeneid 12 embodies all the contradictions of Vergil’s heroic world. Before her intervention in the action that will bring only fruitless delay, this “goddess of ponds and sonorous streams” makes her appearance marked by passivity and grief: . . . lacrimas oculis Iuturna profundit terque quaterque manu pectus percussit honestum. “Non lacrumis hoc tempus,” ait Saturnia Iuno; “adcelera et fratrem, si quis modus, eripe morti . . . ” (12.154–­57) [ Juturna’s eyes pour tears as she three and four times beat her noble breast with her hand. “This is no time for tears,” said ­Saturnian Juno. “Hurry and, if there’s any way, save your brother from death . . . ”] Her silent weeping may recall another figure intermediate between the divine and human, Hercules in book 10,1 but the gesture that accompanies her weeping is typical of feminine goos (“lament”). Juturna knows full well that her brother’s end is nearing, but she tries to postpone it many times: after she abandons him, obeying heaven’s 95

96

A ppendix

mandate, Turnus’s last link to life is broken—­not unlike how, in the Iliad, the departure of his protector seals Hector’s fate (“. . . Hector’s final day sank / and finished down in Hades; Phoebus Apollo left him then”; 22.212–­13). Between these two decisive focal points of action, the divine scene of reconciliation and the concluding duel, is an episode that has never particularly stirred the interest of Vergilian scholars: Juturna receiving the dark warning of Jupiter—­his sending of the Dira—­and raising a last protest against her brother’s fate. Reasons for this scene’s critical “misfortune” are not hard to divine. If epic must above all be action and choice, a plot involving human, heroic, and divine exploits, Juturna’s monologue is left outside this definition. Suspended between impotent protest and agonizing surrender, the expression of a character divinely omniscient but limited and overwhelmed by human emotions, this rhesis (“speech”) might seem like a simple moment of epic delay to critics determined to follow uncompromisingly the main thread of the Sendung (“mission”) of Rome. From this perspective, which privileges the “Olympian” mediation between Jupiter and Juno as the poem’s ideological axis, Juturna’s protest can only appear as a kind of unresolved remainder; more clearly still, Juturna’s sacrifice quickly sums up a price Juno has already decided to pay, one with its own important compensations in the renunciation Jupiter imposes on the Trojans. The perspective is not much different if we choose to begin with comparison to the Iliad: here too the episode is reduced to an ancillary function, petering out in the Apollo-­Juturna parallelism already sketched out earlier. From a morphological perspective, it is not even clear how to label this speech that no one hears and leading to no choice; Heinze’s definition (“lament-­monody used in an irregular situation”) is above all an admission of embarrassment, even if the terms lament and monody provide some concrete comparative concepts. So it seems the only way to proceed is to take at face value, without preconceptions or implicit judgment, the monologic characteristics of this episode—­that is, to accept its isolation, its lack of integration with the epic context, as the result of conscious artistic choice (as a fact of writing) and, from this point of departure, to address anew all of the problems the episode has traditionally posed (function, motivation, literary sources). It will be a matter then of verifying this characteristic as it is realized at each level of the text, in the style,

T he L ament of J uturna

97

in the compositional placement, in the form and substance of its contents. In this exegetical plan it will often be helpful to recall that a text of this sort resists interpretation, precisely because it opposes a norm—­and this very thing, as we know, helps render it visible.2 This norm, it seems, may be found not so much in an implicit poetics as in the same epic context of Juturna’s monologue: in the association between “Homeric” narrative (the great narrative schema of the Hektoros anairesis) and ideological orientation (the victory of the new order desired by the fates through divine consensus). At the very moment when the norm coincides with the providential sense of history, the Vergilian speech finds in this episode a final and limited space of liberty and tragic reflection. Roman epic, while exploring with Juturna new themes and forms, comes close to a breaking point, and runs a latent risk of implosion. 1. At procul ut Dirae stridorem agnovit et alas, infelix crinis scindit Iuturna solutos unguibus ora soror foedans et pectora pugnis: ‘Quid nunc te tua, Turne, potest germana iuvare? aut quid iam durae superat mihi? qua tibi lucem arte morer? talin possum me opponere monstro? iam iam linquo acies. Ne me terrete timentem, obscenae volucres: alarum verbera nosco letalemque sonum, nec fallunt iussa superba magnanimi Iovis. Haec pro virginitate reponit? quo vitam dedit aeternam? cur mortis ademptast condicio? possem tantos finire dolores nunc certe et misero fratri comes ire per umbras! Immortalis ego? aut quicquam mihi dulce meorum te sine, frater, erit? o quae satis ima dehiscat terra mihi Manisque deam demittat ad imos?’ Tantum effata caput glauco contexit amictu multa gemens et se fluvio dea condidit alto. (12.869–­86) [But when from afar she recognized the shriek of the Dira’s wings, unhappy Juturna tore her flowing hair, his sister clawed

98

A ppendix

her cheeks and beat her breast: “What now, Turnus, can a sibling do to help you? What remains for me who’ve had it so hard? What art can I use to prolong your life? Can I oppose such an omen? Now I quit the ranks, now—­I’m afraid, don’t scare me more, dread birds! I know that wing beat, its deadly sound, and don’t mistake the proud commands of great-­hearted Jupiter. Is this how he repays my virginity? For what did he give me eternal life? Why am I deprived of the mortal condition? I could put an end to these great sorrows and go with my wretched brother to the shadows. Me, immortal? Will anything ever be sweet without you, brother? What earth can gape deep enough to send me, a goddess, down to the deepest Shades?” So she spoke and covered her head in a gray mantle, and groaning much the goddess hid herself deep in the stream.] Stylistically this is clearly a mixture of pathetic discourse: the emotion is concentrated in clusters of recurring questions that allow no continuous, linear development of thought. The whole of Juturna’s speech is organized in three groups of interrogatives (872–­74; 878b–­80a; 882–­84), which in turn are subdivided into very short bursts, four in the first group and three in the other two: the last of these sequences, which in fact functions as a clause, presents a clear tendency toward ascending cola. This fragmented articulation is a general characteristic of pathetic discourse and, in fact, a lack of ornament and periodos was prescribed by rhetorical theory in similar cases: ἴδιον δὲ καὶ χαιρόντων καὶ θρηνούντων τὸ συντόμως καὶ διὰ βραχέων ἕτερα ἐφ’ ἑτέροις διάγειν (“it is typical of cheerful reactions and of lamentation that the style proceeds by accumulation of short and concise cola”) (Nicol. progymn. 66,12 f. F.). The accumulation of monosyllables (three initial monosyllables for verses 871–­72), lending excitement to the start of the speech, obeys the same expressive tendency. This does not mean Vergil was “applying” a particular rhetorical recipe, as is sometimes tempting to say.3 We are dealing with a specific gesture of pathetic discourse whose literary development was already complete in good part before Vergil—­for example, the protagonist’s brief monologue at the beginning of Moschus’s ­Europa is entirely composed of interrogative phrases (21–­27). I believe

T he L ament of J uturna

99

analogous stylistic samples from older Alexandrian poetry would be hard to find. Apollonius of Rhodes works instead on the style and form of contents of “deliberative” Homeric monologue (though making it express new themes). But even in Moschus the frequent interrogative breaks do not attain the same vibrato effect, the mimetic tension that animates Vergil’s monologue. Most notable here is the relation between phrase and metrical unit: at least nine of these thirteen verses have a strongly punctuating caesura (or dieresis) that coincides with the end of a colon. Naturally the frequency of internal pauses in the hexameter is accompanied by enjambment, which pushes the emotional thrust of phrases beyond the limits of the metrical unit. These features of the verse clearly define Vergilian pathetic style in contrast to Catullus, who is more comparable in terms of themes and content. A strict coincidence of hexameter and unit of sense, and the elimination of strong internal pauses within the line, were formal achievements of the neoteroi in opposition to the liberty of archaic Roman poetry; Vergil renounces this smooth and polished uniformity since it would prove inadequate to express new dramatic contents. It thus happens in situations like this that enjambment becomes almost a rule and the verse breaks fall outside canonical positions (cf. 878, 882). Juturna’s speech possesses no simple symmetry but a dynamic balance, achieved at the extremes with two framing gestures of mourning (the planctus and the velatio capitis). Within this frame of threnody two strong alliterative and assonant sequences highlight the beginning and end: te tua, Turne, potest (872) and terra mihi Manisque deam demittat ad imos (884). The first alliteration hammers the pathos of the situation, Juturna’s attachment to her brother;4 the second drives home an almost paradoxical conclusion, the desire to no longer be a goddess and be able to die (Manisque deam demittat . . .).5 Word order is here and there disrupted by requirements of expression, such as alliteration, but it never tends toward arrangements too neatly “artistic” that might dampen the emotional ambience;6 the vocabulary is altogether simple enough and even displays some prosaic dissonance, certainly intentional and in some ways parallel to the “harsh” metrical choices already indicated. The use of an intensely unpoetic term like condicio (880), an absolute hapax in Vergil, is indicative of this protesting tone that progresses almost like a debate.7

100

A ppendix

The convergence of these expressive traits orients Juturna’s rhesis toward other contexts in the poem, ones analogously dominated by the pathetic outburst of an isolated individual; and naturally we recall the great monologues of book 4, which offer parallels in form and situation. The most frequently recalled seems to be the lament of Anna, the other devoted sister in a struggle with destiny. The introduction of the speech, unguibus ora soror foedans et pectora pugnis, is repeated verbatim (= 4.673). The difficult nexus aut quid iam durae superat mihi? is clarified by the echo of 4.681, sic te ut posita crudelis abessem (“so that I would be absent as you lay here, so cruel”; the sense of survivor’s guilt).8 Misero fratri comes ire per umbras (881) brings us back to 4.677–­78, comitemne sororem/ sprevisti moriens? (“did you die scorning your sister’s company?”)—­even if the use of comes is a latent cliché of mourning.9 But there is also no lack of coincidences with other passages in book 4 with Dido as central figure: aut quicquam mihi dulce meorum . . . (882) is modeled on the sound of fuit aut tibi quicquam / dulce meum (“or anything of me was sweet to you”; 4.317–­18); Manisque . . . ad imos indeed goes back to 4.387 (cfr. 11.181); the topos of being swallowed by the earth, suggested by the Homeric formula τότε μοι χάνοι εὐρεῖα χθών (“then the wide earth would gape for me”; Il. 4.182 and so on), is varied already in Dido’s speech at the book’s opening: tellus optem . . . ima dehiscat (“I wish the depths of earth would gape open”; 24). In these two cases the topical expression refers to a genuine desire to die, whereas the Homeric line expresses more a fear of being branded by communal judgment.10 The metrical texture is livened by other kinds of repetition. For example, a conspicuously anomalous verse like magnanimi Iovis. haec pro virginitate reponit?—­with the period ending in the second foot and the monosyllable haec before the penthemimeral position—­ recalls a similar protest by Dido against Jupiter: nec Saturnius haec oculis pater aspicit aequis (4.372).11 If I insist on such recurrences too pedantically, it is not indeed to endorse a simplistic theory of symbolic patterning according to which any verbal repetition in Vergil implies a symbolic-­allegorical reference to the preceding context and the creation of a system of long-­range correspondences; an operation all the more debatable in that it cares not the least to distinguish between divergent levels of intention present in poetic

T he L ament of J uturna

101

memory (simple phonic recursion, the “inner ear,” formularity generated by emulation of Homer or Ennius, a precise and intentional echo of a determined context). More interesting is to identify with some precision a stylistic register that recurs in the poem by associating with specific situations—­which I will provisionally call “monologic,” in order to examine the question deeper in what ­follows—­a series of stylistic and metrical traits marked by their opposition to the context. In other words, the frequent verbal resumptions are not all generated by an immediate, specific desire to superimpose in filigree—­”symbolically”—­the character of Juturna on those, for example, of Anna and Dido; taken in their totality, they activate a stylistic register that distinguishes itself from the epic-­warrior context of book 12 and calls on the reader’s memory to integrate itself with the Stimmung (“mood”) of the “tragic” book of Dido.12 This preliminary conclusion needs to be verified at the level of the form of the contents; we need to analyze the narrative context in order to reveal what kind of motivation and function corresponds to the system of stylistic differences so far identified. 2. The immediate justification for our scene is the overarching necessity, foreseen by Jupiter, to distance Juturna from her brother in danger: this way, freed from divine interference, the duel can take its natural and fatal course (see 12.844). Thus Jupiter, after reconciling with Juno, mobilizes the Dirae, supernatural apparitions described with a violent deinosis:13 one of them is sent to earth as a sign aimed at Juturna. We may well ask, of course, about the religious status and origins of these beings that put their hellish nature at Jupiter’s service; but those questions have the disadvantage of projecting real and localized difficulties and ambiguities, immanent to our context, onto a cultural backdrop too varied and deep to be clarifying. An exhaustive study of the matter concluded that we should see the Dirae as a syncretistic fusion of Roman aves prodigiales and Greek Vogeldämonen,14 and that functions of diverse origins are inextricably combined and made subservient, in Vergil, to the execution of Jupiter’s ordaining plans. Such an attempt to legitimize the Dirae as a “regular” component of the Greco-­Roman pantheon, by delineating

102

A ppendix

their historical development, is a worthwhile enough endeavor, but it should not distract us from realizing that such a peculiar fusion is first attested in this narrative context and so, above all, ought to be studied and understood as a special function of it. There remains an unexpected duplicity in the role of the Dira that appeals to religious tradition cannot explain. Though sent by Jupiter as a warning addressed to Juturna alone, the infernal creature in fact achieves two quite distinct results. On one hand, predictably, Juturna immediately recognizes (v. 876, nosco) the omen’s negative import (obscenae volucres is almost a technical definition)15 and though suffering and protesting she acknowledges it in an entirely rational way: she “reads” in its menacing display the iussa superba of the strongest god and in fact addresses her accusation to Jupiter. But in parallel—­and this is unexpected—­the fluttering of the little nocturnal bird produces an entirely instinctive and irrational effect on another character. Turnus is stricken with paralyzing anguish: Illi membra novus solvit formidine torpor arrectaequc horrore comae et vox faucibus haesit. (867–­68) [His limbs went slack with a strange numbing dread, his hair bristled up straight, and his voice caught in his throat.] The symptoms are conspicuously like those provoked by Allecto’s epiphany in book 7 (At iuveni oranti subitus tremor occupat artus, / deriguere oculi . . . [“but as the young man spoke his body started trembling, his eyes stood fixed . . . “]; 7.446–­47). Indeed this distinction of effects, establishing a definite hierarchy between the divine and human levels, introduces to us the episode’s fundamental narrative trait: the isolation that abruptly divides Turnus and his sister, when their psychological states, powerless awareness for one and paralysis of the senses for the other, are forced into a profound mutual estrangement. The reader quickly discovers the stakes in this subtle play that strains the limits of narrative clarity. Distracted by his agony, Turnus is unaware of Juturna’s disappearance; this discovery is “saved” by the poet for the critical moment of the duel. It is the moment that Turnus, again persecuted and blocked in his attempts by the Dira (cfr. 914), hesitates and concedes the advantage of the strike to his adversary:

T he L ament of J uturna

103

nec currus usquam videt aurigamve sororem. Cunctanti TELUM Aeneas FATALE coruscat . . . (918–­19)16 [He can’t see his chariot anywhere or his charioteer, his sister. As he wavers, Aeneas shakes his fatal spear . . .] Interestingly, the narrative perspective of Turnus reminds us of the role another ominous bird played in a famous duel of Roman history: the intervention of the raven in the single combat between the Gaulish champion and M. Valerius Corvus (cf. the epic passage in Liv. 7.25). Like the enemy of Valerius, Turnus is disturbed by a bird that is perceived as a supernatural event; but Juturna’s reading of the epiphany is on an entirely different level of self-­consciousness. Juturna’s speech, while formally addressed to her brother, has in fact opened in the text a disparity of awareness that constitutes the “tragic” backdrop—­the tragic irony—­to the epic and Homeric action of Turnus. This narrative mechanism that isolates its protagonists under the weight of destiny seems to me to justify the apparent confusion in the scene of the omen: understanding it prepares us to interpret Juturna’s speech as a monologue in the full sense, as an act of solitude. The expressive correlate of this isolation is the stylistic tendency already analyzed, an appeal to a linguistic code (or better, to a way of writing) based elsewhere, in the grand pathetic speeches of book 4, but here impossibly out of context. The point of view we find developed in Juturna’s monologue, to consider it now in its thematic articulation, is obviously partial and marginalized with respect to the panoramic vision that dominates the epic action and unveils its global meaning, but it draws its power from being made completely self-­sufficient: closed on itself, it does not enter into dialogue with any of the other modes in which the action of the text is perceived. On the other hand, Juturna’s protest, though isolated, is not entirely unmotivated since it summons up a rather ironic connotation that the “irregular” appearance of the Dira suggests. It is surely no casual choice by the poet, who as we have seen intervenes rather creatively here in the tradition of aves prodigiales, to present in this episode a discrepancy between ends and means. Whereas the superior, celestial order embodied by Jupiter employs a “dark” divinity, the anti-­Roman principle, Juno, appeals to the devoted fraternal love of a benevolent nymph. The Dira

104

A ppendix

repeats, in short, many typical characteristics of the fury Allecto but reverses her aims and direction: a hellish creature operating from above rather than below. As happens so often in Vergil, this parallel does not translate into a merely symbolic architecture but instead remains open and problematic. The existence of a demonic principle in Jupiter’s service is certainly not a direct contradiction of the poem’s theodicy, and yet the tortured atmosphere of this scene does not allow the principles of synthesis and conciliation, enforced with difficulty on Olympus, to extend to terrestrial events. From this, as we will see, we get a sense that Juturna’s protest adds to the poem’s ideology a fragment of lucid clarity: as though in this fold of the text nests an “unhappy consciousness” of Vergilian epic. 3. Lament for the dead, as Homer shows very clearly, is not really monologue, since it is intended to be heard: the original involuntary loud outbursts of grief have become a standard part of the θρῆνος [lament]. Thus in Vergil the laments of Evander (9.152), Aeneas (11.42), Euryalus’ mother (9.481) and of Anna (4.675):[93] but whereas Aeneas speaks his words of farewell in a composed fashion, and Evander, after lengthy, silent weeping, voices a conventional lament, the words of the two women are intended to portray a genuine outburst of emotion: that, too, is the manner of drama, not of ancient epic. Vergil did the same with the lament of Juturna (12.872), who has to leave her brother struggling with death; she does address Turnus, having stood by him until this moment in the shape of Metiscus, but Vergil cannot have meant that her lament really reached his ears; this is the established standard form of monodic lament used in the wrong place. (Heinze 1993, 430–­31) Although a brief and arbitrary excerpt, the quotation reveals the approach of a great critic: I do not mean so much the incisive, poignant style (here reading almost like a verdict) as the application of a general method that nowadays no one who studies the Aeneid can afford to ignore. As we know, Homer is for Heinze the standard

T he L ament of J uturna

105

and very nature of epic; but already in Vergil’s eyes too, as Alexander Pope famously put it, Nature and Homer were the same. The critic must therefore elucidate Vergil’s individuality through deviation and variance, by reopening the historical gap that separates him from Homer and pinpointing there the intermediate literary influences (tragedy, lyric, epyllion) that orient how Vergil intrudes upon the “natural” material of the heroic epic. In our case we often begin by defining the Homeric threnos as an originary form, one suited (we might say) to a living anthropological code, and then derive the Vergilian use of lament from this model, as a literary reflex not always motivated by the context. These are themes Heinze will develop subsequently (in a less categorical spirit) in the famous monograph on Ovid as narrator,17 where the interval from Homer to Vergil is alive with cultural influences and the history of the monologue finds its touchstones in Euripides and Alexandrian epyllion. I am not convinced, however, that in our case the concrete results measure up to the insightful methodology. The monologic character of Juturna’s lament seems almost the product of error, as though Vergil were actually working with the typological prescriptions and abstractions of literary historians; more generally, the use of funerary lament in the Aeneid seems like an inert renewal of a historically obsolete form.18 In reality, as we saw earlier, the entire narrative device is constructed around the formal isolation of this speech: the only frame suitable for the thematics of frustration and surrender. At the same time, Heinze defines even the “natural” form of the threnos in an overly rigid way. It is of course true that the funerary laments in the Iliad present quite pronounced anthropological and linguistic constants tied to the collective participation in ritualized mourning,19 but the Iliad itself contains one peculiar deviation that is relevant for its affinity to our episode. As in the case we have been discussing, in one scene the poet needs to express grief for a character who is still alive. To deal with this unique situation, following a principle of economy that seems to dominate the literary system (all the more so in an oral poetic context), he has to tap into motifs and gestures for ritual lament for the dead. In Iliad 18, Thetis hears the complaints of her suffering son; without yet knowing the reasons for them, she decides to come to his aid. The goddess’s sadness, although she does not yet know

106

A ppendix

Patroclus is dead, is expressed as grief and mourning; she is accompanied by the planctus of her Nereids as though a proper dirge were about to begin: αἳ δ’ ἅμα πᾶσαι στήθεα πεπλήγοντο, Θέτις δ’ ἐξῆρχε γόοιο (18.50–­51) [All of them together beat their breasts, and Thetis began the lament.] Those familiar with the Homeric code will read in these verses the goos for a dead person. But Thetis very quickly makes clear that the object of her lament is Achilles—­still alive, but condemned to a looming premature death. This bold, inspired twist presupposes a horizon of expectations for the routinized association between threntic gestures and dirge; but it briskly transcends this to become an image of individualized grief removed from the pattern of r­ itual.20 Beginning from this first rift it is possible to understand how the “normal” form of lament can transform so as to express, in our case, themes absolutely unrelated to the collective grief tradition. 4 .1 This evolution of the monologic lament, which in Homer we find in embryonic state, still blocked by cultural and psychological constraints,21 will be liberally deployed on the tragic stage. Elements of the Klage tradition can be combined with new literary conventions centering on the forms of dramatic monologue and monody. This whole evolution is important for understanding Vergilian monologues. Here it will be possible only to give some comparative taste, which will perhaps allow us to verify diachronically the consistency of the tragic register that we have perceived “horizontally,” measuring it in terms of its divergence from the context and its recurrences within the poem. Now let us reconsider the extreme flexibility of Juturna’s speech, which suits it to express successive psychological states—­sorrow, rebellion, surrender—­coinciding with an articulation based on the apostrophes: from the initial apostrophe (“threnetic”) to her

T he L ament of J uturna

107

brother, Juturna withdraws inwardly with addubitatio (“doubtful uncertainty”) (aut quid iam durae superat mihi?) until the final invocation of death (879ff.). Sophocles’ Electra provides a model of fraternal love from tragedy: here the protagonist, believing her brother dead, points out to herself the consequences of this loss. This same sequence emerges: an apostrophe to her brother ( Ὀρέστα φίλταθ’ . . . [“dearest Orestes”]; 808); uncertainty and solitude (νῦν δὲ ποῖ με χρὴ μολεῖν; /μόνη γάρ εἰμι [“where can I go now? I am alone”]; 812–­13); a concluding rejection of life (τοῦ βίου δ’ οὐδεὶς πόθος [“I have no desire for life”]; 822).22 But in Euripidean tragedy especially these forms of monologic expression, born from secularized lament and prayer, find regular place and coalesce into an autonomous dramatic form: monody.23 In a theater that rarely employs the “aside,” monody is among the most remarkable forms of scenic isolation: set under the sign of individual lament (schol. Eur. Andr. 103, μονῳδία ἐστιν ᾠδὴ ἐνὸς προσώπου θρηνοῦντος [“monody is a song of one character lamenting”]), it is mainly performed by a character marginalized from the action, who often does not seem to notice the presence or interruption of others. The “monodic” person, not by chance often a woman, tends to reflect not so much to intervene in the situation as to sanction his/ her own agonizing helplessness and passivity, or to protest in vain against enslavement and injustice. In this theater of crisis we see the return, with a different valence, of the Homeric experiment of the threnos for a living person. In Euripides’ Orestes, Electra intones a lament “in advance” for her brother and herself; knowing that after her death, which seems inevitable, there will be no one to cry for them, she describes with rigorous precision (obviously also functional as stage directions) the traditional gestures of grief: (κατάρχομαι στεναγμόν . . . τιθεῖσα λευκὸν ὄνυχα διὰ παρηΐδων . . . κτύπον τε κρατός [“I lead the lament . . . putting a white nail through my cheeks . . . and beating my head”]; 960–­63).24 In this dialectic of forms the traditional gesture assumes importance precisely because it is estranged from its natural and expected place and used to express an individual grief in conflict with the surrounding world. Juturna’s threnetic gesture, inasmuch as it is serves as narrative anticipation of her brother’s death, reveals itself as a “dramatization” of a Homeric cue, which

108

A ppendix

becomes understandable only through the whole literary tradition of epic and drama. It would not be difficult to anthologize the work of Euripides to extract other thematic currents that feed into the Vergilian monologue form. For example, the theme of the unhappy union with a god (cfr. 878ff. and 12.144, Iovis ingratum  .  .  . cubile [“the unwelcome bed of Jupiter”]), which without doubt constitutes a Vergilian innovation on the specific tradition concerning Juturna,25 had been in Euripides the central motif of an entire play, the Ion. As the tragedy unfolds we hear in dialogue and soliloquy bitter meditations and protests against the god who raped a mortal girl, not only an injustice but a source of later misfortunes (cfr. 384ff.; 429ff.; 1312ff.; especially 876ff.). Another comparable tragic situation, producing pathos, is that of goddesses who give birth to mortal children: characters like Thetis, who accuses Apollo of her son’s death (the scene occurred in one of Aeschylus’s lost plays) or the Muse who weeps for her son in the Rhesus. In this lament one notices, among other things, the sarcastic motif of the “reward” voiced by Juturna in haec pro virginitate reponit? Holding her slaughtered son in her arms, the Muse reviews her merits and concludes bitterly: καὶ τῶνδε μισθὸν παῖδ’ ἔχουσ’ ἐν ἀγκάλαις θρηνῶ (“as a reward for this, I am lamenting my child, holding him in my arms”; Rhes. 948–­49). Before attempting a general assessment of Vergil’s reuse of these tragic motifs, it seems necessary to examine similarly the reflection that constitutes a true thematic crux of the monologue. It is the point where a typical emotion of funeral lament, the desire to accompany the dead person to the underworld, confronts Juturna’s particular situation—­her divine condition—­resulting in this trenchant protest: quo vitam dedit aeternam? cur mortis ademptast condicio? possem tantos finire dolores nunc certe et misero fratri comes ire per umbras. Immortalis ego? (12.879–­82) [For what did he give me eternal life? Why am I deprived of the mortal condition? I could put an end to these great sorrows and go with my wretched brother to the shadows. Me, immortal?]

T he L ament of J uturna

109

4 .2 In Aeschylus’s Prometheus Bound the protagonist hurls the harshest accusations against divine justice (see, for instance, 92, ἴδεσθέ μ’ οἷα πρὸς θεῶν πάσχω θεός [“do you see how I, a god, suffer at the hands of gods?”]) and emphasizes the enormous duration of his suffering (99–­100, πῇ ποτε μόχθων / χρὴ τέρματα τῶνδ’ ἐπιτεῖλαι [“when will an end be set for these toils?”]) although he knows he is not to be condemned for eternity (511–­12; cf. 1043–­53). Once he admits to Io that death would be a liberation from pain, πημάτων ἀπαλλαγή (754);26 but the theme of death, in relation to Prometheus’s divine nature, does not seem dominant in the tragedy. It is perhaps not risky to suppose that more space was given to him in the second installment of the trilogy, the Unbound, because a fragment preserved in Cicero speaks precisely of death as a liberation denied (for our purposes it is not crucial to determine the highly uncertain authorship of this translation): sic me ipse viduus pestis excipio anxias amore mortis terminum anquirens mali sed longe a leto numine aspellor Iovis. (Cicero, Tusc. 2.25. Cf. μόχθων / . . . τέρματα at Prom. 99–­100) [“Abandoned by myself, I must accept plague and suffering; in love with death I seek an end of evil—­but I am driven far from death by the will of Jove.”] But a reflection of this sort in Aeschylus always needs to be read in relation to the general certainty of Zeus’s justice; if doubts can arise, or better still dramatic conflicts, they are always destined to be reconciled within the work’s dialectic. These words of Prometheus signify the momentary outburst of a dramatic character whose destiny is to find at last a moment when he will overcome his struggles and reconcile himself to divine justice. Quite differently, Juturna’s amor mortis is characterized by the lack of a secure horizon. Not only does the nymph criticize divine justice with “Promethean” accents, sarcastically calling Jupiter

110

A ppendix

magnanimus27 and in reality the author of tyrannical commands; in addition to Jupiter’s justice, she also questions the traditional representation of the divine—­that is, the combination of immortality and the mutability of human feeling. Juturna highlights this aspect of traditional representations that is accepted as normal, almost making it the main characteristic of divinity: and for this very reason she criticizes the “anthropomorphic” model of the divine with tools only a rationalistic culture could furnish. It is as though she were saying: If being a god means suffering the same as mortals, then this condition is worse than that of humans. It is just undying unhappiness. Clearly this kind of approach can surface only when we place ourselves, even if for a moment, outside the theology proper to poets in order to reveal its contradictions. Let us see now what influence opened this path for Vergil.28 4 .3 In a famous excursus, the author of On the Sublime discusses the greatness and misery of the Homeric deities. The poetic appeal of images like the battle of the gods goes along with their aprépeia (“impropriety”) and is inseparable from it: if it is not interpreted allegorically (a solution from which Anonymous fortunately detaches himself ), scenes like the divine battle are athea (“ungodly, godless”; 9.7). The fact is that Homer has humanized the gods, creating a morally dangerous result: fingebat haec Homerus et humana ad deos transferebat: divina mallem ad nos (“Homer fashioned these and transferred human things to the gods; I’d prefer divine things transferred to us”; Cic. Tusc. 1.65). The anthropomorphism of the representation is the cause of every imaginable aprepés (“unseemliness”): τραύματα θεῶν στάσεις τιμωρίας δάκρυα δεσμὰ πάθη πάμφυρτα (“woundings of gods, civil unrest, vengeance, tears, bondage, emotions of every sort”). If every human weakness and vulnerability and suffering is shared between us and them, what distinction is left between men and gods? Evidently only eternity. Therefore, Anonymous concludes, Homer renders the divine condition worse than that of man: ἡμῖν μὲν δυσδαιμονοῦσιν ἀπόκειται λιμὴν κακῶν ὁ θάνατος, τῶν θεῶν δ’ οὐ τὴν φύσιν, ἀλλὰ τὴν ἀτυχίαν ἐποίησεν αἰώνιον (“For us in our wretchedness death is reserved as a refuge from evils;

T he L ament of J uturna

111

but for the gods he made not their nature eternal but only their misfortune”). The gods are not eternal in physis, but in unhappiness. One might suspect that this lucid criticism of Homeric theology is part of the original Gedankengut of περὶ ὕψους, and the coincidence with Vergil must therefore seem fortuitous.29 On the contrary, the same concept is attested in a philosophical circle not only prior to Vergil, but also closely linked to his intellectual formation. Discussing the difference between mortals and immortals in Homeric and mythic representation, Philodemus arrived at the same conclusion in book II of perì eusebeias (On Piety): οἱ μὲν γὰρ ὄντες θνητοὶ παραγράπτους ἔχουσι τὰς κακοπαθίας, οἱ δ’ ἀεὶ ζῶντες αἰωνίους ἀναδέχονται τὰς συμφοράς (text: Gomperz 1866, 36l = Philippson 1920, 232; [cf. the translation of the whole context by D. Obbink, based on his revision of the MS source, P.HERC. 433 FR.2 COL.2 + 1088 FR. 1: “. . . Hera {several words missing}. Having inflicted suffering and mental illness upon them (i.e., the gods), Homer and Pisander have introduced them (together with debased ideas about them) as being even more wretched still than those (i.e., humans) who really do have such troubles. For the latter group have sufferings which must reach a conclusion, since they are mortal. But the former group, because they live forever, endure woes that last for all eternity.”]) The Homeric norm revealed itself as an image torn by incurable contradictions; it was easy to contrast the Epicurean vision of the divine as happy, incorruptible, free of passions.30 Juturna’s epic biography­—­a life that owes much to Vergilian innovation, to judge from the state of the tradition—­is dominated by the passion of fraternal love, and now mourning: she is a living example of the contradiction discussed by Philodemus. (From this Vergilian perspective, the character is accursed by her own name, not only “quae iuuat Turnum” [who aids Turnus] but also “Diuturna,” the Lasting One, since her grief will be perennial.) But according to the same tradition, not only is it a scandal to imagine the suffering of immortals; it is no less shocking to imagine gods as dominated by lust and agents of rape. The other main aspect of Juturna’s biography, her rape by Jupiter compensated with immortal life, is thus part of the same complex of ideas. The motif of unhappy immortality, born on the ground of philosophical criticism of Homer, thus inserts itself, thanks to the narrow gap in the epic narrative made possible by the monologic structure

112

A ppendix

that we have examined, in a text that wants to situate itself as a direct continuation of the Homeric tradition: scholars who study (with good reason) the Aeneid as an “open” work, characterized by ideological tensions, in the context of the “crisis of the ancient world,” will find this contradiction significant.31 5. The undeniable evocation of tragedy that deeply marks Vergil’s work establishes, in scenes such as this one, a style that makes an immediate gesture of pathos; but we also must note, at the same time, that this expressive register (the aspect of Vergilian style where we can identify and study philologically credible borrowings from tragedy) exhausts and consumes the impulse toward the tragic. From tragedy Vergil carries over a certain violent immediacy that imposes the ego of characters as a total perspective on the world. But on these originally irreducible personalities the tragic form knows how to enact a principle of synthetic recomposition: the character “learns” from action, changing and being changed by others. All this does not seem possible in the Aeneid. This “dialogic” dimension is lacking in minor characters like Juturna, but also in Dido and Turnus—­precisely where critics have rightly recognized (I am thinking of certain definitive pages in Heinze and Pöschl) the emergence of theatrical accents. Not that the poem lacks potentially tragic material or the required maturity of a dramatic style: its limitation should be located rather in the form of the content, which does not take up the task of recomposing the various worldviews by making them dialectical, but instead brusquely superimposes a dominant perspective—­that of Fate. Rather than attempting to transcend tragedy, the Aeneid manifests concretely its impossibility. As a narrative parenthesis within the epic (Iliadic) action, ­Juturna’s monologue finds an original thematic center in its final protest, which is entirely cut off from the poem’s positive ideological axis. In the suspension of action, achieved by the unusual narrative mechanism and the counterpoint of stylistic registers, a principle of self-­consciousness is liberated, one with the potential to precipitate a crisis for the epic’s system of meaning by laying bare its conventions. It is no accident that her reflection lines up (if the

T he L ament of J uturna

113

derivation proposed here is convincing) with the radical Epicurean criticism leveled at the religious world of the entire Homeric tradition. No dramatic development intervenes to reduce this variance, reabsorbing Juturna’s protest in some theodicy or, at the least, interweaving it with other points of view. Her solitary speech wagers everything in the brief gap that divides the two great traditional planes of the action, the divine level, in which the synthesis willed by fate has already prevailed, and the human one, in which characters remain free to act epically in that they do not attain a level of knowing (or what amounts to the same, not recognizing) the necessity of such a synthesis. For this reason perhaps the two interlocutors Juturna evokes at the margins of her monologue, Jupiter and Turnus, cannot help but ignore or not perceive her words. This way the tragic, freezing itself, is reduced to the mournful. Her disappearance illustrates this stalemate; in this gesture appears, for almost the last time in the poem, the profoundly Vergilian motif of exile from a brutal world: Tantum effata caput glauco contexit amictu multa gemens et se fluvio dea condidit alto.32 (12.885–­86) [So she spoke and covered her head in a gray mantle, and groaning much the goddess hid herself deep in the stream.] [Translated and slightly revised from the Italian, Md 1, 1978, 99–­ 121. Important recent discussions of this passage include Obbink 2002 (building on his improved text of Philodemus, On Piety 2, and with important broadening observations on divine action in the Aeneid and Epicurean intertexts); Perkell 1997 (combining the approach of my paper with “Harvard-­style” pessimistic readings of Vergil); and Reed 2007, 49–­54 (on the significance of the intertext in Bion’s Adonis). The whole discussion of epic’s responsibility and self-­consciousness in representing divine action has been raised to a new level by Feeney 1991, and the long neglected book 12 now has two major commentaries (Traina 1997 and Tarrant 2012). My paper was not intended as a complete discussion of Juturna, and I plan to approach on another occasion her important associations with Roman topography and history.]

AFTERWORD I will survey in retrospect some of the implications and loose ends of this book of mine from three different viewpoints: (1) the function of the study of Homeric imitation in the development of Vergilian criticism, (2) the relationship between work on intertextuality in classical studies and modern literary theory, and (3) the rewards of integrating criticism of the Aeneid with the growing area of Homeric reception studies. 1. The project of La traccia (I use this tag as a shorthand for my La traccia del modello. Effetti omerici nella narrazione virgiliana, published in 1984, while I am writing this retrospective note in 2014) starts from Servius’s attempt to put the Aeneid in a nutshell. Servius, incidentally, is not a canonical author, and not even an “author” in any modern sense; it is even more remarkable, then, that his commentary, together with Macrobius, is still one of the main impulses behind research on intertextuality and its tropes.1 According to the introduction to the Servian commentary, the intention of the poem is twofold: to imitate Homer and to praise Augustus (Intentio Vergilii haec est, Homerum imitari et Augustum laudare a parentibus, Serv. ad Aen. praef.).2 This influential approach positions the imitation of Homer as the most central concern of every reader of Vergil, on a par with the epic’s ideological agenda and its historical context (both represented 115

116

A fterword

for Servius, himself, a summa of a late antique reading community, by the simple name “Augustus” as a summa and zenith of Empire). My initial set of questions starts from this simple one: If imitating Homer was “the intention,” was it also a strategy of communication? And how does communication through allusion work, both in specific contexts and in the overall narrative frame? In my early book, therefore, I am still mostly busy with questions of form and intertextuality, although in some instances, especially at the end of chapters 3 and 4, I end up asking questions about “praising Augustus” as well as “imitating Homer.” Incidentally, the way I am working on Latin texts more recently is typically concerned with the interface between form and content, and between literary texts and historical environments. In fact, my next project on Vergil will be quite different from La traccia: it will be a study of Italy in the Aeneid, oriented toward the relationship between the narrative and the memory of Rome’s Social War with the Italic communities. I do not need to explain that I would write a different book today, but there is some need to explain the different situation of this kind of study thirty years ago.3 Back in 1984 my formalist work was using as a starting point the impressive book by G. K., Die Aeneis und Homer (Knauer 1964). This volume will not, I hope, be forgotten, because it is a striking recapitulation of virtues and vices of “source criticism.” Knauer offers an unparalleled broad and rich vision of intertextual contacts between the Aeneid and the Homeric poems. There are some problems with his views, as I explain earlier (chapter 1, note 9), and my book tries to complement and revise his whole approach. On the other hand, my self-­avowed limitation is that I work only on case studies and selected examples: in any event, Knauer has not in the meantime been superseded (or translated) and certainly not extended, perhaps not by chance. A complete revision in the style of La traccia would be a map almost as big as the territory, if one wanted to describe the “intertextual effects” and their multiple links with the context. Over the last thirty years we have had two comprehensive books on Vergil that improve on Knauer in ways similar to what La traccia was hoping to achieve but in much more systematic and thorough ways: significantly, one is on the Aeneid, but the model discussed is of a scope inferior to Homer (although

A fterword

117

by no means minor)—­namely, Apollonius’s Argonautica; and one is not on the Aeneid at all but on the rich variety of models in the ­ nauer’s forte remains the ability to go beyond indiGeorgics.4 K vidual and localized allusions: in my study I follow him in paying attention to large-­scale structures of imitation, while most of the study of Roman imitation in recent years has been oriented toward a pointillistic reading of momentary effects (although these are, of course, interesting as well for my own agenda).5 Before La traccia there had been comparatively little interest in issues of Homeric reception (on which, see my final paragraph), and most scholars had been happy with comparing Vergil’s Odyssey and Iliad (basically, the first and second half of the poem) with their Homeric counterparts. The most significant research from this perspective, aside from Knauer, included Eduard Fraenkel’s case study of Aeneid VII (Fraenkel 1945) and Bill Anderson’s insight (already important in my own argument in La traccia: Anderson 1969) that the model of the Iliad is a dynamic one in Aeneid 7–­12, since readers are oriented by contrasting clues: the Roman text invites us to see Turnus as a new Achilles but also as a new Hector, and vice versa for Aeneas. The unfolding of the plot plays those twin contrasting roles one against the other, creating effects of surprise, foreshadowing, and dramatic reversal. This approach has been strengthened after the publication of my reading by Alfonso Traina’s illuminating insight (still not as famous as it should be in English-­speaking studies of the Aeneid: Traina 1989) that the prophecy of the Sibyl at Aen. 6.89, alius Latio iam partus Achilles, includes a jarring note, since this use of alius is not, at face value, idiomatic Latin. If she wanted to say “a second, new Achilles is born against the Trojans,” she should use alter not alius. Latin alter is the regular expression for the meaning “a second x” viewed in a binary parallel (a second Alexander, a new Rome, another Homer: alter Homerus, not alius Homerus). Alius is ironic (as typical in Sibylline style) and is the seed of future reversal and backfiring: on the surface it is a threat to Aeneas, who has high hopes in a peaceful future in Latium, but since alius carries an idea of “alterity,” multiplication and difference, not as in alter of perfect replica or substitution, the ironic implication is that Turnus will be a different Achilles, and a loser, a Latin Hector versus Aeneas as a new, Trojan, Achilles.

118

A fterword

On the other hand, the schematic division between Aen. 1–­6 as Odyssey and Aen. 7–­12 as Iliad was already ripe for deconstruction.6 Knauer in the early 1960s had already stressed the continuing importance of the Odyssey in Aen. 7–­12, but there had been no Cairns 1989 to argue that the poem as a whole was primarily an Odyssey in which significant Iliadic episodes were to be found. After this helpful reframing of the discussion, Dekel 2012 has argued that Vergil was looking at the Iliad through an Odyssean lens. This latter approach is in fact very close to my short assessment (earlier, chapter 4), and it has the advantage of helping us to see the Nostos aspect of the Aeneid very clearly. The Aeneid is basically a Nostos story that at some point decides (not without surprise effect) to become a new (different) Iliad.7 At the same time, as a Nostos story (and like many tragedies of the Trojan cycle), it cannot avoid being a recollection and a conflicted memory of the Iliad. At this point, however, it is helpful to stop and consider a significant difference. True, the Aeneid is in a very general sense a Nostos kind of poem, parallel to the other stories of homecomings from the Trojan war, although with the underlying paradox that a Trojan hero needs to invent a home, not find, or retake, his old one.8 But here comes a significant difference. One thing we seem to know (in spite of the fragmentary tradition) about Nostos epic poems is that they are not interested at all in etiology and ktisis. This limitation—­a paradoxical one, considering that in later tradition the wanderings of Greek and Trojan heroes are perceived as “proto­colonial” narratives—­is noted for example by West 2013, 245: “There is no evidence that any of the returning heroes founded new towns, as in many of the legends that proliferated later.” This is not the world, and the generic approach, of Vergil, and of Lycophron before him.9 In any case, if I have to summarize the trajectory of this kind of study after La traccia, the growing area (one of two growing areas, more exactly) has been a richer and more nuanced vision of the interplay between Odyssey and Iliad. The forthcoming book by Farrell (Juno’s “Aeneid”) will address this issue and also promote an approach that has been influential elsewhere, in narratology and in drama studies: the idea of taking the characters seriously as “authors” of their own version of the text they inhabit. If we combine this idea with Knauer’s approach to the poem, we can now scrutinize, with

A fterword

119

Farrell, the reworking of Homer together with the subjective initiative of characters who desire to write their own plot and rewrite the story as it unfolds. This way characters in the story can be seen to compete for their own appropriation of the Homeric narrative capital. (I will mention what I think should be another growth area in section 3). 2. La traccia came out in 1984 as the first item in a series flanking the recently founded Pisan journal MD—­Materiali e discussioni per l’analisi dei testi classici, most of the material for the book deriving from my honors thesis of 1977, dubbed “Criticism and Homeric Exegesis in the Aeneid.” (Even the Juturna paper, published in 1978 in the pilot number of MD, had been a part of the thesis, and is now, in this revised edition, reunited with its siblings). The setting for both the thesis and the journal was Pisa, and this is important for me. In 1978 I was (I gratefully recall) the youngest member of an enthusiastic group of people, where the (relatively) senior members were Gian Biagio Conte and Maurizio Bettini. The collective project was, at least according to the program of the journal in 1978, to bridge the gap between classical studies and the evolution of the humanities at large. Most of us were at the time and remained literary scholars, so the “gap” being addressed was mostly about poetics, structuralism, narratology, critical theory, and even methods of close reading. Anthropology, however, was Bettini’s main target area, and he went on to develop a special fusion of anthropology and Latin studies, as a companion piece to the French school of Hellenic anthropology, already flourishing in the 1970s and popular in Pisa and other Italian campuses: as I write he is my colleague in Siena and runs the Center for Anthropology and the Ancient World. The reason for my reminiscence is that the original conditions of my project are different from most of my later research, in which I have been constantly writing on Roman texts of poetry and narrative while looking at the work of a substantial number of fellow-­ travelers in the English-­speaking world.10 Back in my student years this was simply not the case. I had become interested in literary

120

A fterword

imitation, allusion, narratology, and intertextuality when reading modern work by Genette, Todorov, Kristeva, the Russian formalists, Umberto Eco, and others. The “gap” that the MD group wanted to address was the striking difference in ways of doing literary history and close reading between classics and modern literature. When we were looking at classical studies in English at the time, the gap was actually larger, not smaller. Something was beginning to happen, we had noticed, in a couple of collections published by Tony Woodman and David West in the UK, but frankly those books were improving the quality of close reading more than opening a dialogue with literary theory. So in practice La traccia is influenced simultaneously by traditional classics—­and that for my generation meant initially “produced in the German tradition”—­and by literary theory and its innovative tools, but is unaffected by mainstream scholarship produced at the time in the UK and the United States, not only in classics but even in modern literature (with few notable exceptions—­for example, Jonathan Culler and Michael Riffaterre). The U.S.-­based criticism of Vergil in particular (or so I used to think) was dominated by New Criticism, and it tended to produce fine formal readings of the text with an almost exclusive emphasis on internal correspondences and self-­quotations—­especially in the Vergilian Lebenswerk of Michael Putnam. For me, it was interesting to note that the typical, distinctive reading strategy used by Putnam, based on internal echoes and long-­distance similarities, was very similar to that used in the 1950s by Viktor Pöschl, although the two critics had totally opposed views on Vergil’s coherence, aims, and ideology. (In some cases, La traccia exaggerates in the opposite direction. I regret being negative on the significance of Turnus as a Punic lion in the simile at 12.4–­9 and its associations with Dido: see, for example, Putnam 2014.) So I decided to focus on intertextuality and allusion rather than on internal echoes and recurring structures: I had no intention of explaining Vergil with Vergil, I wanted to explain Vergil with Homer. Only later, although not much later, I became aware of many interesting developments in the growing area of Anglophone classics, but they were initially more dynamic in Greek studies, in fields such as Greek tragedy, Homer, and later in Hellenistic poetry. A few years later I found myself working on Latin texts side by side

A fterword

121

with a generation of English-­speaking scholars who were, independently, very close to my own intentions and outlook—­critics like Don Fowler, Philip Hardie, Denis Feeney, Joe Farrell, and Stephen Hinds. If one wanted to play one of those games of “who cites whom” now made possible by the digital media, and so useful to epidemiology-­style histories of criticism, it is relevant to note that they—­but not a lot of people besides them outside Italy—­ noticed and quoted La traccia early in their careers. Without that kind of early attention, La traccia was bound to disappear without a trace, and it almost did. The combination of not being in English and being distributed very badly often slows down or sinks Italian books, and mine was no exception. I had been hoping that at least the scholarly points I was making about connections between Vergil and Homeric scholarship (cf. section 3) would be noted, but even a very scholarly Oxford commentary like Stephen Harrison’s on Aeneid book 10 (Harrison 1991) does not take on board my Quellenforschung. La traccia was being quoted by a limited number of people, but not in an entirely random way. Users (few) of La traccia, to quote a cliché of electronic commerce, were also likely to be interested in various ways of interfacing classical literature and literary theory, in the style typical for example of Don Fowler’s post-­modern Roman Constructions (Fowler 2000) or of the indispensable series launched by Stephen Hinds and Denis Feeney with Cambridge University Press. This book also attempts to make a couple of contributions toward a general discussion of practices such as arte allusiva and intertextuality, or appropriation. Those points are in the same vein as the discussion of allusion in Conte 1974, which attempted to identify two kinds of dynamics in the uncharted territory of allusion, defined as “integrative” and “reflexive.” This kind of approach did not have a long pedigree in classical studies back then. I also had the ambition of generating more dialogue between formal and historical approaches to ancient poetry. My dissertation committee was a stimulus in this direction. The thesis was directed by two very different scholars, Conte and La Penna, who would not (at least up to now) cooperate again in the future. My intention was to exploit what I thought were the best qualities of the two, and trying to combine them, drawing some energy from their evident

122

A fterword

frictions. At the time Conte was enthusiastic about literary theory, was a committed formalist, and had a tendency toward lapidary and abstract formulations, clearly as a corrective to the traditional distrust of literary theory in Italian classics. Antonio La Penna was, and remained, a sanguine historicist, with a powerful ability to bring out political and social aspects in a text, but also with a tendency to produce interminable paraphrases of texts and to separate formal and historical aspects. His contempt for literary theory and formalism verged on political indignation. My hope in this first book, and also later on, was to be able to combine formal and historical aspects in a more supple kind of reading, and with more attention to readerly reception. With this book I had hoped to make a mark in two ways—­first in infusing the Homer-­Vergil synkrisis with a diachronic sense of an evolving model, Vergil’s Homer so to speak (I return to this topic later, in section 3), and second in differentiating genre model and example model in the study of Homeric imitations (see now Hinds 1998, 41–­42). This second approach was part of a general trend in classical scholarship, a growing attention to issues of self-­reflexivity and meta-­poetry. A first influential move in this direction was the attempt in Conte 1974 to differentiate two functions of imitation, the “integrative” and the “reflexive.” The individual examples Conte chose for “reflexive” allusions (often quoted later as representative of his approach) had been around for a long time in German scholarship (for example, Kroll 1924, 150; Haupt 1876, 71; Ehwald 1894, discussing inter alia the Sarpedon/Pallas example that is so prominent in my own book; Ziehen 1896, 313, with the surprisingly modern formulation eine art von Zitat unter dem Gesichtspunkt der Poetik [“a kind of citation viewed from the vantage point of poetics”]), but the vision was new because the familiarity with modern semiotics was encouraging a more systematic reference to the interaction between text and reader. Kroll’s approach to imitation in 1924 had been author-­oriented—­not that this is unimportant, as Hinds 1998 shows in his case studies of “appropriation” in a specifically Roman cultural context, and note also the strong culture-­based framing of imitation in Hutchinson 2013—­nor was he interested in the possibility of reordering the practice of imitation into categories not

A fterword

123

identical with the concepts of ancient scholarship. In fact, scholar­ ship from the 1970s onward exhibits a surprisingly unanimous interest in enhancing the aspect of self-­reflexivity in ancient literature. This is perhaps worth some discussion in historical context, and typically one would need to look at prior developments in the areas of modern literary studies and art history, including the whole “postmodern” turn. The huge success of this kind of reading depends on its degree of immediate gratification: simply by putting some interpretive pressure on individual passages, critics gain what looks like immediate access to the poetics of the author, and de facto expand the empire of criticism over the text. While previous generations had kept discussing endlessly the same purple passages in search of metaliterary formulations and poetic programs, scrutinizing proems, middle proems, dramatic prologues, and addresses to the reader, the new self-­reflexive approach was able to squeeze out hidden acts of self-­consciousness from every corner and detail of the text, often from apparently innocent narrative and descriptive passages. Scholars who favor a more historically rigorous approach often claim that this kind of additional or parasitical semiotics has no warrant in ancient theories of literature or in ancient scholarship. But this is not a valid criticism because classical scholarship always stands to learn much precisely from conceptual tools that cannot easily be translated into Greek or Latin—­ideas that are not emic, like “gender,” “the unconscious,” “point of view,” or “performance,” or indeed “emic/etic.” (If one decides to use only instruments and concepts that can be fully historicized, it would be logical to do away also with “literature” and “religion”). True, there have been excesses: it is quite possible that this approach has created its own tendentious literary history, overrating the importance and centrality of some authors and texts (for example, Callimachus, Vergil’s Bucolics), and excessively insulating poetry from prose. In any case, we have learned how to read the metaliterary, metatheatrical, poetological, and programmatic aspect of ancient texts without limiting ourselves to the obvious declarations by the poets and playwrights. Unavoidably, some of those interpretive strategies have become clichés. Favorite moves have been shuttling swiftly between formal and thematic aspects, or, even easier, looking for critical metaphors that become literalized in narrative or descriptive

124

A fterword

texts: no ancient forest has remained undisturbed, no muddy pond or tiny boat has been left alone by metaliterary r­ eaders. Yet arguably the excesses and the routinizations of this kind of reading have not obscured the main contribution: the idea that the text dynamically develops tropes for its own composition, poetics, and position in literary history. More recently, this trend has been reoriented in two new directions. The first is taking more into account interpretive communities and their responsibility for interpretation. The other, the realization that the different modes are not mutually exclusive and tidily separated categories (Hinds 1998, 1–­51). Hinds has also recuperated and clarified, among much else, the distinction outlined in La Traccia between code model and exemplar(y) model (earlier, chapter 4), a distinction based on Genette 1982 and also discussed by Conte. One aspect of my position (and Conte’s) that has been justly criticized by Hinds is the attempt to deemphasize the author in intertextual studies. Hinds has argued very well that the intention of the author should be recuperated in reader-­response criticism as the object of a “guesswork” made by all kinds of readers. This is true of human interactions of every kind: we don’t normally respond to acts of communication without making guesses about intentions and even the psychology of the sender (cf. the admirable summing up in Hinds 1998, 20–­21 and 49–­51). This position is, I agree, more balanced than resorting to the idea of the “intention of the text” (Eco 1992), an approach that represents a defensive compromise, justified only by the fear of “wild” appropriations. Here, however, it is helpful to remember that this kind of discussion tends to be (so to speak) positional, and depends on which alternative positions are already on the ground (especially in a given community or tradition, sometimes a national tradition) at the same moment, waiting to be criticized. In his discussion of intertextuality, Shawcross 1991 (quoted by Hinds 1998, 48 n. 62) calls the criticism of author-­centered approaches a rear-­g uard discussion. He must be right about the 1990s, and about modern criticism in English in the 1990s, and it is true that Barthes’s essay on the Death of the Author in fact has nothing to do with this discussion and has often been quoted out of context. Yet for someone like me, working in the early 1980s after studying classics in Italy in the 1970s,

A fterword

125

the polemics against author-­centered studies of intertextuality did have a different value and urgency. In literary history as practiced in Italy and other European traditions at the time, the author was very much the biographical individual, and it was still widely believed that studies of intertextuality should basically be used to test and flesh out literary biographies and personal agendas of the authors (Did Propertius like Tibullus or not? Was Pindar an admirer of Aeschylus?).11 3. My own outlook had been based on the idea that the true object of this kind of study is “modes of reading” literature more than authorial strategies. In fact, it is time to mention it, my initial project had been about the reading of Homer and its history. Up to now I have discussed the plan of La traccia back in 1984, but the book also has a prehistory that may be worth mentioning as a conclusion. The title of my 1978 thesis was “Criticism and Exegesis of Homer in the Aeneid.” Back then it had sounded like a good plan and I want to explain why I was not able to bring it to completion. In fact my original dissertation plan had been to write what I thought would be the first book ever about the Homeric scholia, or ancient Greek commentaries on Homer more generally, and Vergil. I remember being influenced initially by the developments for example of Dante criticism, where nobody was looking at “classical models” anymore without taking into account the diachrony of reception and the existence of modes of reading and responding to the classical text that were surprising for us as modern readers and also not identical with ancient approaches to reading and textuality.12 The publication of Schlunk 1974, made available in Pisa when I was already mid-­dissertation, rendered my project unnecessary. I regret being too negative about this book in my review (Barchiesi 1977, my first publication). It was clear that some of the comparisons in Schlunk, and some of my own attempts, were significant, and their range could be increased: a number of ancient zetemata, connected to the A-­scholia tradition and to the debates of Alexandrian editors, could also have interested a reader like Vergil. In my review, for example, I noted how Juturna enters the complex human

126

A fterword

action on the ground before the duel (in medias dat sese acies haud nescia rerum, 12.227) in disguise, but also emphatically aware and in full control of the situation. Haud nescia rerum had been suspected of being an empty supplementary interpolation by Ribbeck and Wagner (see now Tarrant 2012, 149), especially after 12.222–­23; but Vergil may be answering the zetema created, or occasioned, by his Homeric model for this moment of the plot, at Iliad 4.86–­89, where Athena in human shape enters the situation of another duel, and is described “looking for Pandaros, if she could find him” (88) until she “finds him” (89). The idea of a goddess who needs to “search for” a mortal was perceived as unworthy of a divine being: Zenodotus had resorted to a textual modification, while Aristarchus had countered that the goddess is not in full control of the details because she is using a mortal, fallible alias (see Kirk 1985, 339). We cannot of course be precise and should not speculate too much about what precise kind of information had an influence on Vergil: it could have been an Alexandrian edition, accompanied by a diacritical mark and by a hypomnema, or a more complex subgenre of Homeric learned text, or, for instance, a dissertation on gods and mortals in Homer (a topic familiar to Philodemus), or a distant predecessor of our “interpretive” bT scholia, in which case we don’t even need to presuppose a regular “text” but could even invoke the “live” practice of Homeric interpretation in class (“from Vergil’s schooldays”?). I have no space for a full series of examples, but I think we are beginning to see that our target is not a specific kind of written source, but rather a “mode of reading” Homer in a historically situated context. Not a precise model, but still a valuable goal for a Vergilian scholar, although less so for a textual critic who wants to reconstruct the history of Homeric scholarship. The gain would be that in some cases we could appreciate how what look like slightly overdone rhetorical emphases in the Aeneid have something to do with interpretive dilemmas and debates stimulated by the Homeric text: they could compensate, correct, or answer the “traccia del modello.” Schlunk’s work was based on what had become by the mid-­1970s a prescientific text of the Homeric scholia, and the great multi­ volume edition of Erbse (1969–­88) was beginning to make a difference. What kind of difference for Vergilian scholarship, however, is still an open question, because the new text is more reliable and

A fterword

127

in some aspects richer, but less rich in other aspects. Both Schlunk and I, working on a very wide range of different scholia, had come to the conclusion that the materials of what became known after Erbse as the bT recension, the main pool of exegetical scholia, were for the time being the most potentially interesting for Vergilians. It was no more than a hunch, but it seemed just possible that the areas of morality and character, exemplarity, coherence, decorum, theology, and more were not only characteristic of the exegesis found in the bT recension, but also relevant to the cultural horizon of an ­Augustan poet. This is not to say that we have any hope of recovering the precise format, let alone the wording, of a particular commented edition that Vergil would have used. Even the idea of “format” and “edition” is a mirage, since for the first century BCE we are not able to be very precise about the transition from separate learned books or “companions” about the Homeric text, hypomnemata, toward annotated texts of Homer. The interest lies in what looks like a series of shared approaches to the Homeric text, in Vergil and in other readers of his age: questions of decorum, sexuality, historicity, realism, theology, moral value and exemplarity, coherence, rhetorical effect. One crucial point must be the recurring habit to see “problems,” “questions,” interpretive bifurcations, even “aporias” (Hexter 2010, 33) in the poetic text of Homer: this perception, this mental habit is the true comparandum we need to import into studies of the Aeneid. When Euryalus and Nisus wander in the dark woods of Latium, we are meant to think of the Homeric Doloneia. The landscape there is different, but in both texts it is the dead of night, unusually for epic battle action. One problem is raised by the B scholion on Iliad 10.399: how could the spy Odysseus discern the spy Dolon at night? “Probably,” the scholiast argues, “because towards dawn the moon arose.” In Vergil the moonlight betrays Euryalus at a crucial moment (9.373–­74): haud temere est uisum is the narrator’s arch comment (9.375, “not for nothing”). Then Nisus looks up to the moon (9.403) and prays to the Moon, the moon goddess, to guide his arrows, right before his desperate attack. The moon, as Hexter puts it (2010, 35; cf. Schlunk 1974, 71), “rose out of a reader’s imagination.” If the idea came from ancient interpretation of Homer, this

128

A fterword

does not diminish the dramatic and emotional force the moon has in the Vergilian context. The scholiastic comment (or the schoolteacher’s marginale) had created an absent presence in Homer’s text. The Aeneid makes the presence come alive and shine: it will not do to talk about “moonshine.” We may think that asking questions about light and darkness in the Doloneia is pedantic, but we have to remember that the cultural authority of Homer enforces a reading of the text where every detail matters, and so every detail can create a “Homeric problem.” In Vergil, where the text is constantly shaped by empathy, sympathy, and imaginative participation, those problems can become an active part of what we perceive as poetic and aesthetic effects. If we think of the available information on the history of exegesis, a number of developments now allow a different picture to emerge.13 Erbse’s edition includes what we can reconstruct of the Alexandrian work on the Homeric poems (certainly interesting and accessible material for someone of Vergil’s learning, including textual discussions related to specific modes of reading poetry) plus the bT material that can be assessed as a “conglomerate” of the early Byzantine period: of this second source, nobody can safely exclude earlier predecessors, since those traditions are remarkably slow in forming their sediments. Future researchers on Vergil and Homer will, in addition, have access to a growing number of scholia, hypomnemata, and other scholarly formats in papyri, to the tradition of the D scholia, and much else, and to a reliable edition of the Odyssey scholia and of Eustathius—­not to mention all of the recent guides and handbooks listed in my note 13. And finally, at the risk of creating bewilderment, I’d like even to expand the area of Homeric reading in antiquity by mentioning outliers who have rarely been taken into account by Vergilian scholars, and the contrasting destiny of two of them: Plato from Athens and Agallis from Corcyra. It is an understatement to say that Homeric reception is a large area. Almost every individual in the Greco-­Italic world had at least heard or “seen” Homer (that is, knew about the name and the myths, for example, through vase painting, hearsay, parody, proverbs, recitals, tabulae, spectacles) and many of the ones able to read had read some Homeric verse or paraphrase in school, or (if appropriate) made use of Latin versions and glossaries. Every literary text, and many a

A fterword

129

subliterary, is potentially under the influence of Homer. Under the Roman empire, this situation applies to large areas of the Mediterranean, Greek-­speaking but also Romanized. Besides the very visible example of Trimalchio, Homeric scholar, parodist, and impresario, we have the famous anecdote in Synesius (epist. 128) about simple people who have some dim perception about the Roman empire as the space they inhabit, but think it is ruled by Agamemnon. The story is often quoted as a revelation of how distant the emperor was to many residents of the empire, but it is no less striking that Synesius has no trouble attributing to simple people some knowledge of a Homeric character, in fact the quintessential Homeric king. In this stratified environment, our contemporary terminology can be a source of bias. We have nearly settled on two main critical terms for the operations described in my book: reception and appropriation. I could reasonably claim that my book is a study of Vergilian appropriation of Homer, or of Homeric reception in Augustan Rome. But what happens if we foreground Homer’s readers, instead of Homer, as producers of the meaning Vergil is appropriating and transforming? In common critical parlance, reception is either neutral or implies a superior authority of the received voice, while appropriation is a bit more assertive, and the appropriating subject tends to be more active and enterprising than the “receiving” one. This difference is relevant to studies of Homeric influence/ appropriation because the literati who borrow from Homer, besides being a hugely varied and rich sample of ancient society, belong to very different levels of authority and ranking. They include, and this is my final example, Plato and Agallis. Now if reception implies a “weaker” receiving end, we are less ready to discuss Plato as “Homeric reception” because Plato is a strong reader, an appropriating reader, and, to put it simply, a classic:14 his agenda tends to be foregrounded when we find him discussing Homer or alluding to Homer—­similarly, Pindar tends to be viewed more easily as an “active” imitator of Homer than as a testimony to the history of reception and transmission. Yet there is no discussion that Plato was inter alia a reader of Homer, and that he was an important author to Vergil, perhaps no less important than many poets who are usually foregrounded in modern discussions of the ideal “library of Vergil.” We can safely conclude that what Plato has to say

130

A fterword

on Homeric myths and heroes is potentially relevant to Vergil, no less than the words of scholars and schoolmasters. Except that the function of “Homeric reader” is a limiting one for someone who has such visibility and such a complex intellectual agenda: this is perhaps why potential Platonic influence on Vergil’s reading of Homer has frequently been neglected. At the other end of the spectrum, some readers escape attention because they are not famous enough. I think of Agallis of Corcyra: not a celebrity, surviving in tiny indirect fragments, and almost the only woman who made herself a name in Greco-­Roman antiquity as a Homeric scholar. Agallis was never a big name among Homeric scholars, let alone students of the Aeneid. In my original thesis I did have a couple of pages on her because her striking interpretation of the Homeric shield of Achilles looked potentially relevant to the transformation of Achilles’ shield into the clipeus of Aeneas in Aeneid 8. Apparently, Agallis had argued that the two cities on the shield of the Iliad, the city at war and the city in peace, were in fact two historical places, two historical Hellenic cities endowed with great symbolic authority: Athens and Eleusis. This makes the Vergilian choice of Rome and its battles for survival as the images of the new shield less isolated, although still very original. Both Agallis and Vergil wanted the shield to be about historical cities, their origin, their place in the cosmos, their stories of strife and concord, and their prestigious holiness. I stopped asking myself further questions when I saw that the new book by Philip Hardie (1986) was making significant use of her interpretation, and also implicitly vindicating Agallis as a thoughtful professional interpreter of Homer, not just an eccentricity. In his whole book, Hardie had noticed many other points of contact between the poetics of the Aeneid and what one may define as a “Pergamene” culture, or even atmosphere; a cultural atmosphere hospitable to allegory, to images of Gigantomachy, and to attention to the natural cosmos and the space of heroic epic. Some traces of the interpretation by Agallis, in fact, point toward a powerful combination of historical and natural reference in the shield, and this is remarkably germane to Hardie’s own overall interpretation of the Roman poet (Hardie 1986, 343–­46). It is not unavoidable to say that Vergil was a reader of Agallis, or of the scholia accessible to us

A fterword

131

that mention her work, but we do need more attention to the rich variability of Homeric readings in antiquity, including voices that are not mainstream nor endowed with particular prestige. This is a viable project if we accept the idea that we are not only looking for overlaps and frictions and interferences of texts, but also for modes and tropes of reading,15 and for the mouvance and variability of ­Homer’s message throughout antiquity.16 A l e ss a n dro Ba rch i e si Arezzo 2014

132

A fterword

Wor k s Ci t e d i n t h e A f t e rwor d Anderson, W. S. 1969. The Art of the Aeneid. Englewood Cliffs, NJ. Barchiesi, A. 1977. “Rev. of R. R. Schlunk, The Homeric scholia and the ­Aeneid,” Rivista di Filologia e Istruzione Classica 105, 349–­55. ———. 2006. “Le sofferenze dell’impero,” in Publio Virgilio Marone, ­Eneide, trans. and notes by R. Scarcia. Milan. Broggiato, M. 2001. Cratete di Mallo. I frammenti. La Spezia. Cairns, F. 1989. Vergil’s Augustan Epic. Cambridge. Casali, S. 1998. “Ovidio e la preconoscenza della critica: qualche generalizzazione a partire da Heroides 14,” Philologus 142, 94–­113. Conte, G. B. 1974. Memoria dei poeti e sistema letterario. Turin Dekel, E. 2012. Vergil’s Homeric Lens, Oxford. Dickey, E. 2007. Ancient Greek Scholarship. New York–­Oxford. Eco, U. 1992. Interpretation and Overinterpretation. Cambridge. Ehwald, R. 1894. “Vergilische Vergleiche,” Philologus 53, 729–­44. Elliott, J. 2013. Ennius and the Architecture of the Annales. Cambridge. Erbse, H. 1969–­88. Scholia Graeca in Homeri Iliadem, I–­VII. Berlin. Farrell, J. 1991. Virgil’s Georgics and the Traditions of Ancient Epic: The Art of Allusion in Literary History. Oxford–­New York. ———. forthcoming. Juno’s Aeneid: Narrative, Metapoetics, Dissent. Prince­ton, NJ. Farrell, J., and M. Putnam, eds. 2010. Vergil’s Aeneid and Its Tradition. London. Fowler, D. 1991. “Subject Review,” Greece & Rome 38, 90. ———. 2000. Roman Constructions. Oxford. Fraenkel, E. 1945. “Some Aspects of the Structure of Aeneid VII,” Journal of Roman Studies 35, 1–­14. ———. 1994. Kleine Beiträge, I–­II. Rome. Genette, G. 1982. Palimpsestes. Paris. Goldschmidt, N. 2013. Shaggy Crowns: Ennius’ Annales and Virgil’s Aeneid. Oxford. Hardie, P. 1986. Virgil’s Aeneid: Cosmos and Imperium. Oxford. Harrison, S. J. 1991. Aeneid 10: A Commentary. Oxford. Haupt, M. 1876. Opuscula II. Leipzig. Hexter, R. 2010. “On First Looking into Vergil’s Homer,” in J. Farrell and M. Putnam, eds., Vergil’s Aeneid and Its Tradition, 26–­36. London. Hinds, S. 1998. Allusion and Intertext. Cambridge.

A fterword

133

Horsfall, N. 1989. “Aeneas the Colonist,” Vergilius 35, 8–­27. Hutchinson, G. O. 2013. Greek to Latin: Frameworks and Contexts for Intertextuality. Oxford. Kirk, G. S., ed. 1985. The Iliad: A Commentary. Volume 1. Cambridge. Knauer, G. N. 1964. Die Aeneis und Homer. Studien zur poetischen Technik Vergils mit Listen der Homerzitate in der Aeneis (Hypomnemata 7). Göttingen. Kroll, W. 1924. Studien zum Verständnis der römischen Literatur. Stuttgart. Lamberton, R., and J. L. Keaney, eds. 1992. Homer’s Ancient Readers. Princeton, NJ. Martindale, C., ed. 1997. The Cambridge Companion to Virgil. Cambridge. McNelis, C., and A. Sens, 2011. “Trojan Glory: The Survival of Troy in Lyco­phron, Alexandra,” Trends in Classics 3, 54–­82. Montanari, F., ed. 1994. La philologie grecque à l’epoque hellénistique et ­Romaine (Entretiens Hardt, 40). Vandoeuvres–­Geneva. ———. 2002. Omero tremila anni dopo. Rome. Mühmelt, F. 1965. Griechische Grammatik in der Vergilerklärung. Munich. Nelis, D. P. 2001. Vergil’s Aeneid and the Argonautica of Apollonius Rhodius. Leeds. ———. 2010. “Vergil’s Library,” in J. Farrell and M. Putnam, eds., Vergil’s Aeneid and Its Tradition, 13–­25. London. Nünlist, R. 2009. The Ancient Critic at Work. Cambridge. Pontani, F. 2005. Sguardi su Ulisse, La tradizione esegetica greca all’Odissea. Rome. Pontani, F., ed. 2007. Scholia Graeca in Homeri Odysseam, 1. Rome. ———. 2010. Scholia Graeca in Homeri Odysseam, 2. Rome. Porter, J. I., ed. 2006. Classical Pasts. Princeton, NJ. Putnam, M. 2014. “Dido’s Long Dying,” Daedalus 143, 99–­106. Quint, D. 1993. Epic and Empire. Princeton, NJ. Rengakos, A. 1994. Apollonios Rhodios und die antike Homererklärung. Munich. Richardson, N. 1980. “Literary Criticism in the Exegetical Scholia to the Iliad,” Classical Quarterly 30, 265–­87. Scaffai, M., ed. 1991. J. Tolkiehn, Omero e la poesia latina. Bologna. ———. 2006. La presenza di Omero nei commenti antichi a Virgilio. Bologna. Schlunk, R. R. 1974. The Homeric Scholia and the Aeneid. Ann Arbor, MI. Schmit-­Neuerburg, T. 1999. Vergils Aeneis und die antike Homerexegese. Berlin–­New York.

134

A fterword

Shawcross, J. W. 1991. Intentionality and the New Traditionalism. University Park, PA. Tarrant, R. 2012. Virgil. Aeneid Book XII. Cambridge. Traina, A. 1989. “Ambiguità virgiliana. Monstrum infelix (Aen. 2, 245) e alius Achilles (Aen. 6, 89),” in Mnemosynum. Studi in onore di Alfredo Ghiselli, 547–­55. Bologna. West, M. L. 2013. The Epic Cycle: A Commentary on the Lost Troy Epics, Oxford. Wigodsky, M. 1972. Virgil and Early Latin Poetry. Wiesbaden. Ziehen, J. 1896. “Epenzitate bei Statius,” Hermes 31, 313–­17.

NOTES For e wor d 1. For a fine exposition of this aspect of La traccia see Hinds 1998, 101, referring to Don Fowler’s rendering of Barchiesi’s title as “the trace/trail/ track of the model” (to trace a further path of scholarly influence). 2. Hardie 1990. 3. Galinsky 1996, index s. vv. “multiplicity of meanings.” 4. Martindale 1993. 5. See Fowler 1995, http://www.cisi.unito.it/arachne/num2/fowler.html (accessed May 4, 2014).

I n t roduc tory Not e 1. The functional, contextualized, rhetorical, and linguistic treatment of allusions is the chief contribution of Conte 1974, a study that has been particularly influential for my research. 2. The first three chapters are modified and larger versions of three already published essays (chapter 1 in 1980; chapter 2 in 1982; chapter 3 in 1983).

Ch a p t e r 1 1. Klingner 1967, 592. 2. Probably the first to stress the metamorphosis of Aeneas throughout book 10 was Conway 1928, 135ff. 135

136

N O T E S T O C hapter 1

3. Obviously essential, and with previous bibliography, is Knauer 1964, 298ff. Of great interest are the methods and comments of Otis 1969. 4. For similar gestures during funerary proceedings, see Sittl 1890, 66 and 74. 5. Instead, Aeneas stresses, paradoxically, how Lausus can obtain glory from falling in such a duel (10.829–­30). The pathetic motif of 10.831–­32 terra sublevat ipsum / sanguine turpantem comptos de more capillos may have been partly suggested by Il. 17.51–­52 αἵματί οἱ δεύοντο κόμαι Χαρίτεσσιν ὁμοῖαι / πλοχμοί θ’, οἳ χρυσῷ τε καὶ ἀργύρῳ ἐσφήκωντο (“his locks, similar to the Charites, were drenched in blood, and the braids, kept in place by spirals of gold and silver”), a reference absent from Knauer’s lists. 6. However, the scene would have a clear Homeric antecedent (one hardly fit for our analysis) if we were to follow the reference suggested by Pöschl 1977, 139 n. 199 (and already in the previous editions of the famous essay): Menelaus lifting Patroclus’s body in Il. 17.587ff. In the Homeric passage, Apollo in disguise tells Hector that Menelaus now has singlehandedly stolen a corpse away from the Trojans (οἴχεται οἶος ἀείρας νεκρὸν ὑπ’ ἐκ Τρώων); the context, however, clearly shows that the corpse is not Patroclus’s, but that of the Trojan Podes, just killed by Menelaus, who is now dragging it away (ἔρυσεν, 581) toward his own army, certainly not with the same pious intentions Aeneas has toward Lausus. After all, even if we were to consider it the corpse of Patroclus (cf., indeed, Il. 17.722–­23), the alleged source would lack exactly what makes the Vergilian theme unique: the practice of compassion toward the enemy. Similarly, investigations in visual art have borne little fruit (Maréchal 1940 mentions representations of Penthesilea’s death, but without convincing arguments). The possibility of non-­Homeric epic models should also be taken into account (cf. Fraenkel’s well-­known hypothesis, 1964, 177ff.). See also later, note 20. 7. See later, chapter 2, for more details. 8. We will see again in the following pages how the excessive undervaluing of Turnus as a character is often associated with an ideological reading of the entire poem that is untenable (for more on this issue, see Barchiesi 1978, 1546ff.). 9. Knauer’s work, with its abundant philological documentation, would seem to have given the final word on the constitutive character of Homeric imitation in the Aeneid. It is progress, of course (and it is not the only advance that this extraordinary tool has made possible), but the

N O T E S T O C hapter 1

10. 11.

12. 13. 14.

15.

16.

17.

137

price to pay would be too high if we wholly accepted Knauer’s method and his critical ideology. I am thinking especially of the reification of tacit methodological assumptions imposed before the basic sense of the text has been determined, and which are therefore impossible to verify; for this reason we see a proliferation of hazardous categories such as Ausgangssituationen, “coincidences e silentio,” coincidences in “structure” and in “position,” and an overall evaluation of the relationship between Vergil and Homer based on biblical paradigms (see Knauer 1964, 353ff.). Schol. Il. 16.431–­61 (bT), 257 E. Or at least so they appeared in the classical period, when Homer was considered the repository for a coherent theological paradigm. The never-­ending debate about the relationship between Zeus and Fate among modern Homeric scholars is of no immediate relevance to this study, even if Il. 16 is one of the most debated and variously interpreted books when it comes to this issue. See, at least, the clear summary by Chantraine 1952, esp. 70–­73, who stresses how Zeus, in spite of not being able to govern it, is not subject to Fate—­even if, de facto, the results always coincide with the god’s will. More cursorily Lloyd-­Jones 1971, 5, insists even in this case on the identity between human moira and Zeus’s will. On the typical ways that divine intervention is represented, see Kullmann 1956, 42ff.; Paul 1964; Griffin 1978. Cf. Od. 3.236ff. I summarize here Conte’s important general definition (1974, 43–­45). The allusive quality of the Vergilian passage was noticed, though not sufficiently analyzed, by Ehwald 1894, 733. Ovid reminds us of the falling of guttae . . . cruentae among the omens foretelling Caesar’s death (Met. 15.788). A more ample documentation is in Cook 1940, III.1, 478ff. Obviously this is not a rare misunderstanding, especially among Vergilian scholars (for example, “tears of blood” are mentioned by Williams 1977, 352, and Forman 1973, 200). Good observations on the epic motif of the rain of blood are found in Whitman 1958, 123. There is still need for a careful reconstruction of this interesting motif; collections of passages (albeit not complete) can be found in Geffcken 1907, 203, and in Pease’s commentary to Cicero’s De divinatione, Darmstadt 1963 (at 2.25).

138

N O T E S T O C hapter 1

18. Plato’s quote begins Zeus’s speech (16, 433) with αἲ αἴ instead of the commonly transmitted ὤ μοι. Since αἲ αἴ is an expression absent from Homer, referring elsewhere to genuine crying scenes (cf., for example, Bacchyl. 5.153; Bion Adon. 28 and so on; Ov. Met. 10.215ff.), it might indeed have been inserted in the text by Plato himself, who thus—­with or without realizing it—­made Zeus’s words all the more sensational for his readers (the equivalent, one may say, of replacing “alas” with “howl! howl!”). On the other hand, the preexistence of the corruption cannot be ruled out. Labarbe 1949, 183ff., maintains—­with little supporting argument—­that it is an ancient variant introduced by an “Athenian rhapsode.” 19. Following hypotheses proposed by others in similar cases, Merkelbach 1951, 376 n. 4, postulates a Platonic influence on Zenodotus’s athetizing here. That Zenodotus had eliminated those lines in his recensio, because absent in his best witnesses, is an idea proposed by Schwartz (and later taken up by Pfeiffer 1968, 113ff. and n. 7) that seems to have been abandoned. Nickau 1977, 140ff. (and cf. RE s.v. Zenodotos, X A c. 35), explains the deletion differently but also admits (cf. 219) that Zenodotus could not have completely ignored Plato’s criticisms, even if he did not transform them into a strict philological criterion. 20. Schol. Hom. II. 16, 459, 261 E.; Ps. Heracl. All. Hom., 51–­52. Buffière = Schol. Hom. II. 16, 459 (B2), 131, Dindorf. For other allegorical variants see Ps.-­Plut. De vit. Hom. 111 (the rain derives from the blood of the fallen evaporating off the ground and condensing in the air, by anathymiasis, a typical “physical” allegory); Procl. In Remp. 1, 123–­25 Kr. 21. I list the other ancient passages known to me, in the hope that they may contribute to reconstructing a complete history of these polemical interpretations: Min. Fel. Oct. 24.4; Tert. Apol. 14.3 and Nat. 1.10.39; Athenag. Suppl. 21.3; Athan, c. Gent. 11; Firm. Mat. 12.9; Joseph c. Ap. 2.34, 102, Reinach; Clem. Alex. Protr. 4.55 and Orat. ad gr. 2; Max. Tyr. 5.5, 59, Hobein. Cf. Auson., 78, Peiper. See also Lenaz 1980; Tandoi 1984. 22. This is a ubiquitous concept in the ancient world, found in the most varied literary genres (see, for instance, Aesch. Prom. 518; Plat. Leg. 5.741 A; Philem. fr. 31.4–­5 K.; Liv. 9.4, 16; Ov. Met. 9.434 and especially 15.780–­82), but it obviously becomes more meaningful when tied to the Stoic worldview: Comm. Lucani 2.306, p. 69, Us. Secundum Stoicos qui omnia dicunt fato regi et semel constituta nec a numinibus posse mutari (“according to the Stoics, who maintain that everything is

N O T E S T O C hapter 1

139

ruled by Fate, and that after Fate’s decision the gods have no power to modify”); 240ff., even though the richest evidence is in Pease’s note to Cic. Div. 2.25. 23. The agreement between Jupiter and Fate is proclaimed in the council of the gods with which book 10 opens (see, for this, Büchner 1947, 17, and, more fully, Kühn 1971, 131ff.). For the general issue of the relationship between divinity and (Stoic-­like) conceptions of fate in the epic action, Heinze 1915, 293–­94, is still essential. A careful analysis of the contradictions that stem from it can be found in La Penna’s introduction to Vergil’s works (1966, LXXVI–­L XXVII). There is, indeed, an inevitable contradiction (one felt in all the epics in the Western tradition): the epic code cannot dispense with the divine apparatus or do without the existence of a destiny that motivates and makes plausible the sequence of events. If, however, Jupiter is used as the “speaker of fate” (as often happens in the Aeneid), this superior will needs to be contradicted and circumvented to guarantee epic action. It is not by chance that Jupiter reappears in book 10 after a long absence. 24. The transformation of the concept of destiny ends up deeply conditioning the ideology of Roman epos. In the context of a different genre, Seneca similarly “reread” the Oedipus Rex through the filter of Stoic culture, which considered it an exemplum of fate’s ineluctability, a δρᾶμα τῆς εἱμαρμένης [SVF II 941, 271, 33], and composed his Oedipus accordingly (cf. Schetter 1972, 404). 25. Hercules’ tears are at once “useless” and humanly inevitable. Even the Stoics allowed crying when faced with the news of a premature death: cum primum nos nuntium acerbi funeris perculit . . . lacrimas naturalis necessitas exprimit  .  .  . hae lacrimae per elisionem cadunt nolentibus nobis (“as soon as we are struck by the news of an early death . . . a natural necessity makes us cry . . . the tears fall as if pressed out against our will,”), Sen. Ep. 99.18–­19. The reader of the Aeneid is reminded by assonance of 4.449 mens immota manet, lacrimae volvontur inanes, where an analogous model for controlling emotions is presented, if indeed, as I tend to believe, the tears are those of Aeneas. This line’s interpretation is always much debated; among more recent contributions it is worth mentioning West 1969, 44–­45, and Hudson-­Williams 1978, which serve also as convenient entry points to the extensive prior bibliography. On the crying theme in the Aeneid one can see in general Rieks 1970, 183ff.

140

N O T E S T O C hapter 1

26. Eur. Hipp. 1396 (cf. Ov. Fast. 4.521ff.). Again in Ovid one finds an interesting allusion to our Vergilian passage: tum vero gemitus (neque enim caelestia tingui / ora licet lacrimis) alto de corde petitos / edidit (Met. 2.621ff.). The Ovidian divinity behaves in the opposite way: he lets loose the groan that Hercules represses (sub imo / corde premit—­alto de corde petitos), while instead holding back the tears (as the Euripidean Artemis prescribed). It is as if Ovid were ironically criticizing Vergil for misapplying the paradigm. 27. Cfr. Soph. Trach. 1071ff.; Eur. Her. 1353ff. (Sophocles’ passage was also known in Rome through Cic. Tusc. 2.20ff.). On the presence of these two tragedies in Vergil see in general König 1970. 28. Cf. EG 790, 6 f. Kaibel. 29. For the interpretation of extendere, cf. A. 6.806 virtutem extendere factis (here too the phrase is used in a didactic, stylistically solemn context), for which see Norden 1957 ad loc., and especially Henry 1883 (1972), 421ff. 30. La Penna 1966, LXXVIIIff., points out that by adhering to the traditional understanding this very passage reveals the limits of the idea of the soul’s survival that Vergil himself had clearly articulated in book 6. We would expect “Elysian” immortality to be valor’s reward, not earthly glory (accordingly devalued, for example, in the Somnium Scipionis). On the other hand, the epic-­heroic parts of the poem naturally offer greater resistance to the inroads of such speculative philosophical themes. 31. This proverbial theme is amply documented and categorized by Hommel 1957. 32. Many Ciceronian passages are discussed by Buchheit 1969. Determining the original source of this concept is not relevant here, especially since, as a commentator on Sallust cautions, its wide distribution discourages attempts of that sort (cf. Vretska 1976, 40). Still useful in general is Knoche 1934. 33. Conrad 1922, 62ff., cites, in this sense, [Plut.] cons. Ap. 31 and other similar passages. Generally useful, Kassel 1958. 34. Cf., for example, already in Il. 15.139ff. 35. Cf., for example, Eur. Alc. 416–­17. 36. Good general comments focused on this topic can be found in Nisbet-­ Hubbard 1970, 325–­26, and Monaco 1972. See also Lattimore 1942, 254–­55; and, for developments in the direction of diatribe, Conte 1965, 119ff.; Stork 1970, 127ff.

N O T E S T O C hapter 1

141

37. Astutely noted by Griessmair 1966, 26. 38. Anth. Pal. Didot. Ill (1890), 123, and n. 217, 25–­26 (cited by Monaco 1972). 39. Etiam sua Turnum / fata vocant metasque [cfr. ὀλέθρου . . . πείρατα in Il. 6.143] dati pervenit ad aevi. The phrase fata vocant immediately recalls Homeric expressions in which the corresponding subject is the gods (δή σε θεοὶ θάνατόνδε κάλεσσαν, Il. 16.693; cf. 22.297). But the use of the subject fata, however, is a gesture in a different direction, one best identified by Socrates’ words: ἐμὲ δὲ νῦν ἤδη καλεῖ, φαίη ἂν ἀνὴρ τραγικός, ἡ εἱμαρμένη (“as a tragic poet would say, destiny is now summoning me”), Plat. Phaed. 115 A. 40. After what has been said there should be little doubt about how to read sic ait, atque oculos Rutulorum reicit arvis (10.473). For a god, “averting the gaze” means to renounce intervening in human matters (cfr. Il. 13.1–­ 4), just as, conversely, watching means to intervene (cf., for example, A. 1.223ff.; 4.220ff., and also earlier, note 12). The implication is, therefore, that Jupiter chooses not to be attentive because, despite being aware of and emotionally invested in the human tragedy unfolding, he intends to let destiny run its course as he programmatically stated at 10.113: fata viam invenient. Commentators (see esp. Conington and Nettleship 1963 ad loc.) who see in this gesture only an expression of his disturbed state of mind and compassion are probably mistaken: we have seen that the episode is set up so as to free Jupiter from any excessively anthropomorphic element. Oculos reicere is an isolated phrase in Vergil, but interpretation does not appear problematic if we consider idioms of opposite meaning such as oculos traicere (Lucr. 4.424) or conicere (cf., for example, Cic. Verr. 5.35; Varr. Rust. 2.2.2) and the possible epic derivation from expressions such as ὀφθαλμῶν βολαὶ (Hom. Od. 4.150; cfr. also Ap. Rh. 3.288 and Malten 1961, p. 24 and n. 7). In any case, Servius’s interpretation reicit ad arva needs to be rejected. 41. Consider these close correspondences: At Pallas magnis emittit viribus hastam (Il. 16.284 and so on: Πάτροκλος δὲ πρῶτος ἀκόντισε δουρὶ φαεινῷ) / vaginaque cava fulgentem deripit ensem (for example, 22.306: εἰρύσσατο φάσγανον ὀξύ) / Illa volans . . . tandem etiam magno strinxit de pectore Turni (Il. 4.139: ἀκρότατον δ’ ἄρ’ ὀϊστὸς ἐπέγραψε χρόα φωτός) / Hic Turnus ferro praefixum robur acuto / in Pallanta diu librans iacit (Il. 10.135 and so on: εἵλετο δ’ ἄλκιμον ἔγχος ἀκαχμένον ὀξέϊ χαλκῷ) atque ita fatur: / “Aspice num mage sit nostrum penetrabile

142

N O T E S T O C hapter 1

telum” (Il. 11.391–­92: ἦ τ’ ἄλλως ὑπ’ ἐμεῖο, καὶ εἴ κ’ ὀλίγον περ ἐπαύρῃ, / ὀξὺ βέλος πέλεται) / . . . cuspis . . . pectus perforat ingens ([Liv. Andr.] 35 Mor. hasta volans perrumpit pectora ferro) /. . . Una eademque via sanguis animusque secuntur (Il. 16.505: τοῖο δ’ ἅμα ψυχήν τε καὶ ἔγχεος ἐξέρυσ’ αἰχμήν) / Corruit in vulnus, sonitum super arma dedere (Il. 4.504 and so on: δούπησεν δὲ πεσών, ἀράβησε δὲ τεύχε’ ἐπ’ αὐτῷ; Il.14.420: ἀμφὶ δέ οἱ βράχε τεύχεα ποικίλα χαλκῷ; Enn. Ann. 415 V2) concidit, et sonitum simul insuper arma dederunt (cf. Lucr. 4.1049) omnes plerumque cadunt in volnus / et terram hostilem moriens petit ore cruento (Il. 2.418: ὀδὰξ λαζοίατο γαῖαν; Il. 22.17: γαῖαν ὀδὰξ εἷλον; Il. 11.749 and so on: ὀδὰξ ἕλον οὖδας [not as close is Od. 22.296: ἤριπε δὲ πρηνής, χθόνα δ’ ἤλασε παντὶ μετώπῳ, though this is perfectly rendered at A. 10.349 with fronte ferit terram]; Mat. 6 Mor.: ille hietans herbam moribundo continit ore). For the history of this last trope (and of the text of Matius’s fragment), see Traina 1974. Other Ennian derivations, in addition to those already mentioned, have been suspected on stylometric grounds (for line 475, cf. Norden 1957, 207; for line 480, Norden again, 137). The strongly Homeric aspect of the entire strategy of the duel is well emphasized by Metger 1957, 210. 42. Turnus’s spear flies in slow motion for four lines (482–­85), and the arrangement of the objects, heaped up in polysyndeton (the shield, its various layers, the breast plate, Pallas’s breast) suggests the unfolding of an inescapable process, culminating in line 484 with a framing hyperbaton (vibranti cuspis medium transverberat ictu). Una eademque via sanguis animusque secuntur (487) evokes Pallas’s agony with the harsh, rough sonority created by the unique combination of synalepha and synizesis in the first foot. 43. A useful overview of “typical gestures” in the Homeric battle scenes can be found in Kurz 1966, 35ff. The case of Il. 13.618, cited by many Vergilian commentators, is unique in the Homeric typology and has no stylistic analogies with our passage in the Aeneid. 44. In a not dissimilar way Hector addresses Patroclus καὶ τεθνηῶτα (16.858). 45. For this “misleading” technique, cf. the scene in which Mezentius defeats Orodes and scoffs at his death prophecies; during the entire exchange (10.736–­46) Mezentius leans, with his foot and spear, on the body of his prostrate enemy, and only after ending his blasphemous speech rips out the weapon, thus causing Orodes’ death. Cf. also

N O T E S T O C hapter 1

143

12.338–­39 (Turnus): miserabile caesis / hostibus insultans; 12.356–­57: semianimi lapsoque supervenit et pede collo / impresso. 46. On the negative connotation of this iconographic scheme, see the important observations by Fraenkel 1957, 160–­61 (already in 1950, 412). It is well known that the representation of the victor placing his foot on the defeated enemy is usually avoided in official Greek and Roman art; it will appear in Rome only at the height of the Imperial period, perhaps on account of Near Eastern influences (see, for example, Schweit­zer 1931, 216). The corresponding literary paradigm is foreign to the Roman ideology of triumph and is even contested by Augustan propaganda (the famous poetic passages on the winner’s moderation are collected, for example, by La Cerda and then Norden in the note to A. 6.853 parcere subiectis). More generally, the penetration of this model appears limited to allegorical uses, such as the image of the “triumph of the spirit” that tramples superstitious fears (Lucr. 1.78; cf. Verg. G. 2.490ff.) or the elegiac image of the “triumph of love” (Prop. 1.1.4; cf. Meleager, A.P. 12.48.1 and Panaenius fr. 9 Mart.). The first to explicitly represent the victories of Augustus and Rome according to the paradigm of the imponere pedem seems to be Ovid, on this point an extreme interpreter of the imperialistic ideology (cf., for example, Tr. 3.12.48; 4.2.44; Pont. 2.2.78; more material, if a bit disorganized, in Bömer’s note to Ov. Fast. 4.858). For the narrator’s comment on the necessity of moderation in victory, see chapter 4. 47. The commercial metaphor is suitably interwoven at 10.503–­4: Turno tempus erit, magno cum optaverit emptum / intactum Pallanta, and in Aeneas’s reaction from afar: belli commercia Turnus / sustulit ista prior (10.532–­33; cf. Achilles in Il. 21.99–­100). And even Turnus’s wish that Evander be eyewitness to his son’s death (10.443) invites the reader to make meaningful connections. In fact, Pyrrhus’s behavior comes to mind—­a paradigm of savage cruelty within the poem—­gnatum ante ora patris, patrem qui obtruncat ad aras. Vergil did not miss the opportunity to point out that Pyrrhus was later punished, indeed in a manner that mirrored his own crime (A. 3.332: Orestes assaults him patriasque obtruncat ad aras). In contrast with Turnus, obviously, Aeneas is the hero moved by the patriae pietatis imago (10.824), who honors all blood relations. But the importance of this motif of paternal love materializes especially in the poem’s final scene. Turnus begs Aeneas and

144

N O T E S T O C hapter 1

almost succeeds in placating him by explicitly juxtaposing Anchises to his aging father Daunus. We are confronted with a genuine “false ending” to the poem (which will be followed by the dramatic reversal provoked by the baldric): indeed, the reader expects a sudden pacification, owing to the fact that the Iliad’s last episode, with the final reconciliation between Priam and Achilles, culminates in the moving words “remember your father” (Il. 24.486). Connected with this, it is noteworthy that in Vergil the narrative overdetermination of Turnus’s death is not equivalent to a whole-­ hearted ideological justification of Aeneas’s actions. Turnus is defeated and he admits it. Aeneas should therefore be bound by his father’s ethical call to parcere subiectis; after suddenly seeing the baldric he adopts a sacralized language (Pallas te . . . immolat . . . poenam scelerato ex sanguine sumit), but his behavior is full of unrestrained fury, underscored by the fiery metaphors usually reserved for Turnus (furiis accensus et ira, 945; fervidus, 951). Among the treatments of the episode Thome’s analysis (1979, 274–­ 347) is still balanced and informative. 48. To seize the spoils of an enemy is not in itself reprehensible: we not only have examples of this practice in Roman history (Camps 1969, 39), but the Vergilian context itself shows that it is in fact a “normal” behavior: Turnus acts as Pallas himself would have had he been the victor (10.462–­63: cernat semineci sibi me rapere arma cruenta / victoremque ferant morientia lumina Turni). Vergil does not make Turnus do anything barbaric, which would have destroyed the tragic dimension of his character (see Pöschl 1952, 140): although with some condescension, Turnus returns the young man’s body to his companion (whereas in the Iliad, we should not forget, Hector threatens to behead Patroclus, throw him to the dogs, and expose his severed head on a palisade, Achilles, not content with defiling Hector’s corpse, even wishes to devour his flesh). It is thus a question of evaluating not practical actions but rather “effects of reading.” We should move from the level of direct reference (to which the single elements belong) to that in which context-­triggered nuances generate indirect meanings for the story. 49. Conte 1980a, 96ff. Hornsby 1966 shows, with several Vergilian examples, how widespread is the literary topos according to which wearing somebody else’s armor leads to one’s own death (although the individual interpretations are not fully convincing). One might easily

N O T E S T O C hapter 1

145

generalize that the same deep underlying structure can be found in those cases where the transfer of the armor happens via exchange between enemies (Hector is dragged around Troy with the baldric Ajax had given him, Ajax kills himself with Hector’s sword) and those where possession is obtained through killing (Hector, Turnus, Euryalus, and Messapus’s helmet). 50. It is not strictly a hendiadys, since the two terms are neither overlapping nor complementary. On the other hand, definitions of these figures of speech are always problematic and fluctuating (for similar cases in Vergil, a starting point might be Williams 1960 at A. 5.410–­11; to orient oneself in this delicate matter, see Hofmann-­Szantyr 1965, 782–­83, and Kroll 1980, 30ff., with invaluable addenda by the editor, A. Lunelli). 51. The scene loses its tragic intensity if one interprets infelix as “bad luck” (Quinn 1968, 276). And even Quinn’s description of Aeneas’s relationship to the baldric is unacceptable: supposedly Aeneas, like a true psychopath, kills because the scene of violence portrayed on the belt triggers in him some sort of conditioned response. 52. A brief anticipation at lines 799ff.—­that Zeus will give Hector the helmet of Patroclus that falls to the ground, but Hector’s death is also near—­for now remains unfulfilled. 53. In Aeschylus’s Myrmidones this narrative reversal, which the Homeric text did not foreground, was readily translated into a dramatic reversal. Achilles illustrates his situation with the “Libyan fable” about the eagle who, struck with a poisoned arrow, realizes that the notch was made with his own feathers (fr. 231 Mette). 54. Cf., for example, von der Mühll 1952, 259. 55. Among them, as we saw, Vergil selects a single object and uses the ekphrasis to highlight it. Turnus’s loot did not consist only of the baldric, of course: from A. 11.91–­92 hastam alii galeamque ferunt, nam cetera Turnus / victor habet, we gather (as Servius already noted ad loc., 487 Th.-­H.) that Pallas was stripped of all his arms. This is not inconsistency in the narrative, though: the poet abridges the events to bring into focus only the details that are “dramatically” productive, at the expense of others. 56. Emblematic, already from the title, is Bassett’s 1923 article “Hector’s Fault in Honor,” with which Knauer (1964, 303 n. 2) unfortunately agrees, and uses when he writes: “The ominous οὐ κατὰ κόσμον (17.205) from Zeus assigns Hector’s fate and ‘blame’ [Schuld] just as firmly as

146

N O T E S T O C hapter 1

the ‘nescia mens . . . ‘ of Turnus’s fate in Vergil” (304). The use of Schuld, even if lightened by quotation marks, is typical of this critical tendency to flatten two culturally remote texts into one another when one should take exactly the opposite approach. Apart from the absurdity of applying the concept of guilt to Homer, one should see, for Vergil, the harsh objection by Klingner 1967, 579 n. 1. For the profound difference between Zeus’s address and the Vergilian comment, cf. chapter 4: when we speak of characters’ “guilt,” we make the problem abstract and hence unsolvable. Otis 1964, 56, leans toward the same approach, only superficially historicizing his moral terminology (culpa, inhumanitas); similarly Williams 1977, 350. 57. I am thinking of A. W. Schlegel’s and W. von Humboldt’s famous distaste for the “subjective” tendencies of Vergilian epos (cf., for example, Wellek 1955, 51), an attitude that later had decisive influence on nineteenth-­century German classicism. Prophecy and “objective” anticipation of events are, on the contrary, essential traits of Homeric diegesis (Goethe’s letter to Schiller from April 22, 1797). After Heinze, Duckworth 1933 has contributed valuable insights to the issue. 58. Heinze’s direct model for Vergilian prolepsis is an extremely vague hint to Patroclus’s destiny (Il. 16.46–­47). Perhaps closer are the anticipations at 16.794–­800: Apollo made Patroclus’s Achillean helmet roll onto the ground: Zeus gave it to Hector to wear, but Hector too is approaching death (see, in fact, the scene in which the Trojan hero is wearing Patroclus’s spoils in book 17). One can see here that the narrative focuses on a fated object; but the discourse of simple predestination does not in any way become commentary. While Hector dons these fatal arms the poet intervenes (17.194–­97), but curiously enough signals that it is Achilles who is destined to die young (“he did not grew old in his father’s armor”). Of course, the theme of Achilles’ death is not irrelevant to the Homeric narrative, but the prolepsis still points to an event not fulfilled within the poem. 59. Anderson 1969, 83: “perhaps the most subjective passage in the whole Aeneid.” 60. Feder 1954, 208. 61. Duckworth 1956, 312: “one of his rare ‘asides’ ” (as for the content, Duckworth sees a certain affinity with the Horatian ode to Dellius). 62. According to a sharp analysis by Fraenkel 1967 (these examples are taken from the sequel that appeared in 1968, Museum Helveticum 25, 179–­80).

N O T E S T O C hapter 1

147

63. Some regard forms such as optaverit as dictated simply by metrical reasons (for example, Norden 1957, 151), but the future perfect possibly functions to express intensified “certainty” (cf. Ladewig et al. at A. 2.77). It is more difficult to consider (with Ley 1877, 21) optaverit as expressing anteriority over oderit. 64. Current treatments of this grammatical category suffer from a certain lack of differentiation and nuance, without which it is impossible to focus on the peculiarity of Vergilian usage. Hofmann and Szantyr (1963, 310) specify this as the first occurrence of historical future in poetry, but connect it to examples that, as far as epic style is concerned, have a distinct status, like apostrophes (cf. Ov. Fast. 2.386). We need to be careful when considering comparison with phrases such as annus hic erit insignis in Liv. 7.1 (established by Samuelsson 1905–­6, 36ff.), since anticipations of that kind are largely codified in the genre of historiography, especially in the annalistic “register” (“the year I am starting now to narrate—­from day one—will be particularly noteworthy  .  .  .”), and they are usually placed at the beginning; in Vergil, instead, the proleptic comment is a force in contrast with epic illusion, and disrupts the normal linearity of the narration. As opposed to historiography, epic usually tends to conceal the presence of a subject who organizes the events. Wackernagel 1926–­28, 207, cites, as typical example of historical future, V. Fl. 1.451–­52, Canthus, in Aeaeo volvet quem barbara cuspis / pulvere (the fulfillment of which will occur in book 7): this is, however, part of a catalogue of the Argonauts, in which the poet gives us a comprehensive glimpse of the fate of his characters before setting into motion the action itself. The use is legitimized, for example, by Hom. Il. 2.694; 2.724. 65. Iste never loses its natural deictic force, grounded in the daily communicative use of language and exploited particularly in theater: it is from here that it later develops its strong anaphoric value—­as opposed to hic and ille—­liable to be colored by various context-­dependent emotions (cf. Keller 1946). It is perhaps avoided in high style (Norden 1957, 120), and in any case for several reasons it occurs rarely in Augustan poetry (Axelson 1945, 71–­72). 66. On this topic see Stroux 1933. Also notable is Plut. Arat. 49 (on Philip V of Macedon): Ἐπεὶ δὲ τῆς τύχης εὐροούσης ἐπαιρόμενος [for this term, see later, note 76] τοῖς πράγμασι πολλὰς μὲν ἀνέφυε καὶ μεγάλας ἐπιθυμίας, ἡ δ’ ἔμφυτος κακία . . . κατὰ μικρὸν ἀπεγύμνου καὶ διέφαινεν αὐτοῦ τὸ ἦθος.

148

N O T E S T O C hapter 1

67. It is one of the first instances in poetry where tumidus = puffed up with arrogance, vainglorious (noted in Tränkle 1960, 82; but cf. Hor. S. 1.7.7; A. 9.596). For superbus and relevant bibliography, see Traina 1981, 117ff.: I also agree with his overall evaluation of Turnus’s character. The meaning of the epithet furens is more layered, since it encompasses divergent semantic paradigms: the epic-­archaic code, in which warlike rage is a normal trait of heroes and their god Ares, the “tragic” one (as in Euripides’ Hercules), and the “Roman” one, in which furor has a specific political sense (see Weische 1966, 23ff.; Jal 1963, 421ff.). 68. Epic. fr. 488 Us. 69. For instance, Arist. EN 4.1124 a 12ff.; Pseud. Ar. de virt. 1250 b 34; Cic. Off. 1.90. 70. See also Plut. Aem. Paul. 27 (on all these passages, see Fuchs 1947, 167–­ 68 and n. 44). Such considerations fit well with what we know of Latin historical epic—­in particular, its tendency to comment upon events and turn them into occasions for moral and political teaching (as in the hypothetical contexts proposed for Enn. Ann. 355 V2: quippe solent reges omnes in rebus secundis). 71. See, especially, Cic. Marc. 3. 72. The passage was first identified in the commentary of Fulvio Orsini (1568). 73. Quo nunc Turnus ovat spolio gaudetque potitus (v. 500) is a Romanized version of Il. 17.472–­73 (cf. 18.131–­32) τεύχεα δ’ Ἕκτωρ / αὐτὸς ἔχων ὤμοισιν ἀγάλλεται Αἰακίδαο (“now Hector is triumphant, having on his own shoulders the arms of the grandson of Aeacus”). 74. Conington and Nettleship 1963 ad loc. 75. Schol. Hom. Il. 17.201 (bT), 367 E. 76. For the scholium’s language, in addition, obviously, to μετριάζειν, what is interesting is the term ἔπαρμα (/ἐπαίρομαι), which occurs in the gnomic tradition (Sotad. 9.4 Powell; cf. ἐπαίρεται in Comp. Men. Et Phil. I 297, 101 Jäkel) and in proverbial sayings about fortune (see, for instance, Lys. 2.10 ὑπὸ τῆς τύχης ἐπαρθέντες; Plut. Arat. 49). The semantic sphere (lifted up/elation) is perfectly analogous to that of Latin sublatus. 77. Criticism, commentary, and exegesis are all terms that define, not only as metaphors, Callimachus’s narrative practice; see Perutelli’s good points (1979, 87–­88).

N O T E S T O C hapter 2

149

78. This tendency to absolutize Homeric imitation in Vergil, to study epic tradition as source rather than as means of expression, is, as already noted (see note 9), a limitation of Knauer’s essay on the Aeneid.

Ch a p t e r 2 1. The comparison is especially with the council that opens Iliad 8 (in addition, we cannot exclude, obviously, the codification of similar episodes in archaic Latin epos). The situations are parallel in structure and in their placement at the beginning; noteworthy too is the partial similarity of the opening line (see the exegetical controversies on 10.1 panditur interea; as for Il. 8.1, Zenodotus puts it after 8.52, for reasons not readily apparent). 2. Beyond their indisputable value in the immediate context, Jupiter’s words have fairly complex implications, especially for the idea of fate they presuppose. In the vast bibliography on this question, see especially Wilson 1979. 3. The difference in Aeneas’s behavior before and after Pallas’s death is a plot element that invites focalization. 4. At Iovis interea monitis Mezentius ardens / succedit pugnae, 10.689–­90 (cf. Interea soror alma monet succedere Lauso / Turnum, 10.439–­40). 5. It will be ever more evident as we proceed that this formal structure becomes meaningful only if we are ready to recognize as important what at first seems ancillary to the action: for instance, character traits, which escape notice at this level of description and are not bound to the narrative grammar of the supporting structures. 6. For the Homeric correspondences, see earlier, chapter 1, note 41. On the issue of the Cycle’s influence (Pallas and Lausus; Antilochus and Troilus), see Kopff 1981, 938ff. 7. McDermott 1979–­80 notes the importance of the motif but does not elaborate. For the different treatment in Homer of the theme of inferiority in battle, Krischer 1979, 147–­49 proves useful. 8. Here too one can discern the agreement between the voice of the narrator and of Aeneas the character. On this consonance see Conte’s insightful remarks (1980b, 89). 9. Mutilating an enemy’s corpse is behavior already rejected in Homeric culture: in the Iliad it shows up here and there as a regressive threat to be exorcised (see Segal 1971).

150

N O T E S T O C hapter 2

10. This is an important point worth the risk of digression. Not everything in Turnus’s behavior is to be condemned wholesale: we may label as “uncivilized” his wish for Evander (a new Priam) to witness the death of his own son, but it seems to be within his right to despoil Pallas’s corpse. It cannot be by chance, in fact, that just a little earlier Pallas has declared the same intention toward Turnus: cernat semineci sibi me rapere arma cruenta (10.462). More generally, when one considers the entire system of values embedded in the text, it is not Turnus’s but Aeneas’s behavior that stands out. It is Aeneas who, in some circumstances, goes beyond the Homeric framework and appears as the hero of a different kind of humanity, which in the poem cannot yet fully emerge, and whose values cannot be fully recognized. Turnus, on the contrary, is fully immersed in the heroic code that many characters share with him (and is therefore far from being a “dark” hero, archetype of enemies of the Roman order). 11. For the contrast with Homeric usage, see again Krischer 1979, n. 7. It is noteworthy also that the clash between Aeneas and Turnus will be felt at a certain point as unbalanced (see 12.216ff. at vero Rutulis impar ea pugna videri . . .). It is an example of subjective style (see especially the splendid image of Turnus at 219–­21), not only because it foreshadows the outcome of the duel, but especially because adopting the point of view of the Rutulians motivates psychologically (unlike what happens in a parallel Homeric episode) their participation in the breaking of the truce that Juturna plots. 12. For details, see Perret (1980, 207; 210), who looks for balancing acts even in the number of killings. 13. It is important that no part of the simile emphasizes the impious nature of Orion sometimes stressed in the mythological tradition; any attempt to extract from the simile an evaluative characterization (as, for instance, Leach 1971) is therefore unfounded. (For the mythological tradition on Orion, see Fontenrose 1981). If anything, traits of impiety seem present in the simile devoted to Aeneas: here Aegaeon, unlike in much of the poetic tradition, is explicitly portrayed as a θεομάχος, one who fights against the gods. The idea that Vergilian similes are always meant to express judgment thus has clear limitations. For the narrative technique of “counterbalancing” similes one can look at Aen. 11.721–­24; 751–­56. At the end of her aristeia, Camilla is likened to a hawk that seizes and mangles a dove in flight. Soon thereafter, Tarchon intervenes and strikes

N O T E S T O C hapter 2

151

back; while the balance of power is reestablished the Etruscan leader is likened to an eagle catching a snake and bearing it aloft. The two similes share, among other things, vaguely assonant diction (volsae labuntur ab aethere plumae, 724; simul aethera verberat alis, 756). 14. For the recurring theme of paternal love, see Owen Lee 1979, who, however, does not provide the kind of philological analysis its importance deserves. As for Homer, the friendship between Achilles and Patroclus also seems at times likened, at least metaphorically, to a father-­son bond (the clues are examined by Moulton 1977, 101ff.). This “paternal” isotopy, however, does not seem as important as the one presupposed by the Vergilian text. 15. On the principles of ancient hospitality, see the assessment by Bolchazy 1978; for the handshake, the passages collected by Marquardt 1886, I, 192, n. 10. Another element strengthening the bond among Aeneas, Evander, and Pallas is the reference to the Roman practice of contubernium; see the incisive observations by Williams 1983, 103–­5. 16. In Hercules’ case it is also evident that the hospitable relationship is the next best thing to kinship. After all there were some who made Pallas a son of Hercules (see D. H. Ant. Rom. 1.32; 1.43): we cannot rule out that Vergil was somehow influenced by this association (cf. the parallel situations of Zeus/Sarpedon: Hercules/Pallas). 17. In the same context Aeneas immolat a priest (537ff.) and rejoices about denying a burial (557ff.). In general, all of Aeneas’s opponents in this (and no other) phase of the plot are given the chance to supplicate him (see oranti 10.536; orantis nequiquam 554; miserere precantis 598; oranti 559). The model usually assumed behind these scenes is that of the “vengeance” books of the Iliad (20 and 21, introductions to the “Killing of Hector”). See also chapter 4. 18. The juxtaposition I proposed in 1980 (38, n. 45) is suggested now also by Putnam 1981, 152 (in the context of a different argument). But see chapter 4. 19. A Homeric image, memorable for its rarity in the Iliadic context, is likely influencing this scene: Achilles lifting Hector’s body in his arms and placing it himself on the bier to be returned to the Trojans (Il. 24.589); see also my earlier comment on this Vergilian passage (note 6). 20. This kind of phrasing is quite common in the scholarship on archaic epic, but the radical boldness with which Russo’s work (1981, 51ff.) synthesizes this matter is noteworthy.

152

N O T E S T O C hapter 3

Ch a p t e r 3 1. On this scene, several contributions deserve preliminary mention: Bömer 1944; Pöschl 1952; Liebing 1953, 134ff.; Kühn 1957 (and 1971, 119ff.); Wlosok 1967, 123ff.; Knauer 1964, 259–­80; Wigodsky 1965; Grassmann-­Fischer 1966, 29ff.; Klingner 1967, 535ff.; Galinsky 1968; Wlosok 1973. Decisive stimulus to study this scene in greater depth came from Pöschl’s elegant interpretation, along with the more technical discussion by Grassmann-­Fischer and Wlosok’s significant clarifications. Other studies on more specific points will be mentioned later. 2. In the description of the omen as it is perceived by the characters, in fact, no element clarifies its good or inauspicious nature; notice the absence of spatial indicators such as left and right, which could solve the ambiguity. See, in contrast, the style of an account such as Enn. Ann. 527 V2: tum tonuit laevom bene tempestate serena; Verg. A. 9.627–­28. 3. For the tone, see Luc. 6.812: tu fatum ne quaere tuum cognoscere (in a context of omens inauspicious for the interlocutor); notable also is the force of profecto, only here in Vergil (for the emotional value, in “high” dramatic versions of everyday speech, see Acc. trag. 143 R2). 4. Heu quantum inter se bellum, si lumina vitae / attigerint, quantas acies stragemque ciebunt, / aggeribus socer Alpinis atque arce Monoeci / descendens, gener adversis instructus Eois! (“Oh, what a great war, what battles, what massive slaughter they will start between them, if they make it to the light of the living: father-­in-­law dashing down from the Alps and the citadel of Monaco, and son-­in-­law lining up the ranks from the opposing land of Dawn”; A. 6.828–­31). I thank my friend Mario Labate for suggesting this parallel. That the tone is one of grief here is clarified also by certain parallels with the epicedium for Marcellus: o nate, ingentem luctum ne quaere tuorum . . . quantos ille virum magnam Mavortis ad urbem / campus aget gemitus! vel quae, Tiberine, videbis / funera . . . heu pietas . . . (“son, don’t ask about the deep mourning of your people . . . oh, how the Campus Martius will invade the great city with strong laments of warriors! What a funeral you will see, Lord of the Tiber . . . Alas, what pious spirit!”; 6.868–­78). The exclamations and the image of the bloody river (on whose use in prophecies, see Norden 1957: ad Aen. 6.87) will return, joined together, in the prophecy on Pharsalus in Lucan 7.114–­16:

N O T E S T O C hapter 3

5.

6.

7.

8.

153

quantum scelerum quantumque malorum / in populos lux ista feret! quot regna iacebunt! / sanguine Romano quam turbidus ibit Enipeus (“what a flood of crime and calamity will this day bring to mankind! How many kingdoms will lie low! What color the Enipeus, swirling with Roman blood”; see Narducci 1979, 127). Another significant parallel (which might have influenced Vergil) is Hor. Carm. 15.9–­11: HEU, HEU, QUANTUS equis, QUANTUS adest viris / sudor! QUANTA moves FUNERA Dardanae / genti. It is a threatening prophecy of the Trojan War (cf. Hom. Il. 2.388ff.; Alc. fr. 283 LP; the Vergilian and Horatian images are deftly combined in Stat. Theb. 3.210–­11). “I can never forget the thrill that went through me when the meaning of Anchisiades was first made plain to me by my tutor, H. Nettleship” (Warde Fowler 1918, 88 n. 1). But the idea, whose critical fortunes will last throughout the century, could already be found in La Cerda’s brief note at Aen. 10.822. Forced interpretations can often be found in both camps: for instance, those who interpret the scene omitting all the “pessimistic” signals (see especially Kühn 1971, 120 and n. 7), emphasize the meaning of 8.543–­ 44 larem parvosque penatis / laetus adit. But laetus here has little to do with a state of mind, and does not reveal, as the modern reader may suppose, the “interiority” of the character. As Conington well remarks at 7.430, “laetus is constantly used of the spirit in which a person is bidden to obey a command.” Conversely, there has been great exaggeration in explaining, as has often been done, Aeneas’s entire speech as an answer to a melancholy foreboding of Pallas’s destiny. For instance, see on the one hand Henry 1972, 735–­39, and Pöschl 1952, n. 1; on the other Serv. Dan. ad Aen. 8.529, 275 Th.; Conington on 8.523 and 535; Knauer 1964, 259, n. 1. “Whether the actual material armour was up there in a cloud, or only a semblance and promise of it, does not much matter. The reader may take his own view of the spirit of the passage” (Warde Fowler 1918, 93). It may be relevant that Achilles’ own divine arms are often regarded, in ancient exegesis, as prodigious objects, somehow “animated” (on Hephestus’s Zauberwerke see, for instance, Schrade 1950, 47ff.); cf., for example, schol. Hom. Il. 19.13 (bT) p. 575 E. ἀνέβραχε (in the scene in which Thetis hands the armor over to Achilles) ἐψύχωσε δὲ τὰ ὅπλα διὰ τοῦ ἔβραχεν . . . ἐντεῦθέν τινες τὰ ὅπλα ἔμψυχα ἐνόμισαν εἶναι. διὰ τοῦτο εἰκότως καὶ οἱ Μυρμιδόνες δεδοίκασιν (“the use of ἔβραχεν personifies

154

N O T E S T O C hapter 3

the arms . . . whence some have argued that the arms are endowed with life. It stands to reason, then, that the Myrmidons are terrified”). 9. Ιul. Obs. 56a: Cinna et Mario per bella civilia crudeliter saevientibus Romae in castris Gnaei Pompei caelum ruere visum (“when Marius and Cinna were wreaking havoc in Rome through civil war, in the camp of Pompey the sky seemed to come crashing down”) (cfr. A. 8.525: ruere omnia visa repente); ibid., 57; per Syllana tempora inter Capuam et Volturnum ingens signorum sonus armorumque horrendo clamore auditus  .  .  . Lucius Sylla post quintum annum victor in Italiam reversus magno terrori fuit inimicis (“in the age of Sulla between Capua and Volturnus a huge sound of trumpets and arms was heard, with frightening screams. . . . After four years, returning to Italy as a conqueror, Lucius Sulla brought great terror to his adversaries”) (cf. August. C.D. 2.25; Joh. Lyd. de ostentis p. 13 W.); D.S. 38/39.5 (from Johannes of Antiochia; cf. Liv. fr. 18 W.-­M., and Walton 1965, 240–­44): “Hence the age of civil wars started . . . Livy and Diodorus, among the many signs revealing the setting into motion of looming misfortunes, mention that from a clear, cloudless sky came loud a trumpet sound  .  .  . and that those who heard it were dumbfounded in fear” (cf. also Varro in Serv. Dan. ad Aen. 8.526; Plut. Sulla 7.6 f.); Luc. 1.569ff.; Petr. 122, 134 f. (with V. Max. 1.6.12; D.C. 41.61.3); Verg. G. 1.474 f.; Ov. Met. 15.782ff.; App. BC 4.4.14; D.C. 47.2.1; Iul. Obs. 69; D.C. 51.17.14; and, in general, Tib. 2.5.71–­80. 10. The weapon imagery hinges on two main oppositions (arms in the sky / buried weapons; warrior weapons / farming arma) that embody the peace-­war dichotomy. To these models we should probably add the one (still traditional, but less frequently flagged by Vergil’s commentators) according to which the useless and rusty weapons (G. 1.495 exesa . . . scabra robigine pila) are the symbol of a peaceful age (see, for instance, Bacch. fr. 4.31ff. Snell-­Maehler; Hor. S. 2.1.43; Tib. 1.10.49–­50; Ov. Fast. 4.927–­28). Still, this image of peace carries a sense of distance, a quasi-­utopian quality; the giantlike bones of the dead (grandiaque . . . ossa, G. 1.497) imply a temporal displacement of generations, like the one that separates Homer from the heroic age of the Trojan War. That Vergil left a passage like this one at the end of Georgics 1 unchanged even in the post-­Actium climate should suggest less simplistic and schematic ways of understanding Augustan ideology: see Fraenkel 1957, 288; Hubbard 1974, 94.

N O T E S T O C hapter 4

155

11. Fraenkel 1964, 151. 12. I am thinking in particular of the attractive interpretation of Merklin 1970. 13. The task, in other words, is to shift the analysis of the narrative text from a strict focus on structural semantics (from which we have borrowed useful tools up to this point) toward the study of implied cultural models. For indeed it is these models that are actually responsible for guiding the production of meaning and orienting the reading.

Ch a p t e r 4 1. See the important analysis by Mariotti 1966, 13ff. 2. See schol. Q Od. 4.187; schol. E Od. 3.248, respectively; several other passages in Bühler 1964, 46–­47. 3. Many have discussed such implications; see especially Callen King 1983. 4. On these aspects interesting observations and points of departure for further study can be found in Conte 1980b, 94–­95, n. 35 (Aeneas’s pietas as anti-­metis); Otis, 1969, 30ff. (the Aeneid as “reversed” nostos). 5. See the concise overview of this aspect of imitation in Conte 1981, 148. On imitation as “generalization” Genette 1982, 89ff., is also useful. 6. The Homeric model and the reference to 10.908 are clearly in favor of cruenta against cruenti of P. Cruenti (Aeneae) (of bloodthirsty Aeneas) is a grammatically harsh expression and yields a sense that does not seem to fit Mezentius’s ethos. 7. Which does not necessarily mean lack of narrative effectiveness. One could say, for example, that the epithets Vergil reserves for Aeneas create a para-­formulaic system (see Parry’s invaluable comments, 1971, 171 and 36): they are, in other words, repetitions that intentionally “mimic” the effect of Homeric formularity (without, to be sure, aiming at or achieving its generative functions). Most of the time the use of these epithets eludes a test of contextual pertinence or motivation, but taken together these repetitions are indeed helpful in deciphering the narrative texture. First they guarantee a frequent reference to certain characteristics of Aeneas that are fundamental to the plot structure (for example, religious and familial pietas, his sense of leadership); second, they help the reader to mark Aeneas as the protagonist. In fact, Vergil almost never uses para-­formulaic tags when referring to

156

8.

9.

10.

11.

12. 13.

N O T E S T O C hapter 4

other characters in the story. The antagonist Turnus is also given pronounced, recurring, and thematically important traits, and yet he is never marked by epithets of this sort. The use of regular epithets becomes, therefore, a distinctive trait that within the story isolates the status of the protagonist and distinguishes him from everyone else: it is a stylistic feature, but also a narrative signal. The example is taken from the brief and enlightening assessment of “quasi-­formularity” in Conte 1974, 42 n. 26. For similar phenomena in the Argonautica, see Ciani 1975, 197–­98. In fact, no model offered by modern commentators of the simile is any more pertinent than others: see especially Pöschl 1952, 141ff.; Knauer 1964, 58 n. 2 (Il. 5.136–­43; 20.164–­73; understandable reservations on the relevance of these passages in Williams 1983, 171). Pöschl elegantly notes the simile’s foreboding quality (p. 142, “Turnus aber ist wirklich—­schicksalmässig—­ein zu Tode getroffener, ein dem Untergang Preisgegebener,” [“But Turnus is truly—­almost by destiny—­one marked for death and consigned to his downfall”]), but insists on a Homeric model (Il. 20.164ff.) that does not clarify this specific point, and so ends up positing too sharp a contrast between Vergil’s and ­Homer’s poetics. Not convincing either is the proposal of a symbolic juxtaposition between Turnus (“Punic lion”) and Dido; we should not overrate the designation Poenorum . . . in arvis, since Vergilian lions are “Punic” already in Ecl. 5.27, and above all because geographic markers are a regular and rather unmarked trait of epic similes in Vergil. For Vergil’s tendency to “engrave” the beginning and end of a book with images that recall one another (whether parallel or complementary) see, for example, book 6 (landing at Cumae / landing at Gaeta); 8 (Turnus brandishes his arms as signal of war / Aeneas wears Vulcan’s arms); 9 (Turnus purifying himself near the Tiber / he dives into the river which cleanses him of the slaughter); 11 (dawn / sunset). On the uniqueness of the image, see for example, Schnapp-­Gourbeillon 1981, 42–­43; Baltes 1983, 41. In Vergil, the detail of the wounded lion is isolated as well (Hornsby 1970, 120, speaks of “wounded lion” a propos of A. 9.791–­96, but the detail is not there). Schol. Hom. Il. 16.753a (bT), p. 297 E.; an observation in Servius at 12.8, p. 576 Th.-­H. converges on this point. Williams 1983, 223. Among studies that help clarify this aspect, especially useful are Thome 1979, 274ff.; Lyne 1983.

N O T E S T O C hapter 4

157

14. Detailed comparative analyses can be found, for instance, in West 1974; Pöschl 1980. In spite (or perhaps because) of the traditional character of this synkrisis, mistakes are still made. In A. 12.897 Turnus gets a glimpse of a saxum antiquum ingens; this is the λίθον . . . μέλανα τρηχύν τε μέγαν τε Athena uses as weapon in Il. 21.403–­4; several commentators have antiquum derive from the Homeric μέλανα, interpreted as a sign of antiquity. But Homer clearly states that the rock was placed there by “men of former times,” τόν ῥ’ ἄνδρες πρότεροι θέσαν (21.405). 15. On the importance of gestures (particularly of physical contact) in supplications, see Gould 1973: this study, in the context of a mainly literary analysis, provides a healthy counterpoint thanks to its strictly anthropological approach. The material gathered by Gould clarifies what cultural model underlies what we would otherwise treat as the “typical situation” of the Greco-­Roman epos. 16. Cf. Catul. 64.370 proiciet truncum summisso poplite corpus (in which lofty style is made to match pathetic context). 17. The semantic value of supplex is therefore probably intensified by the context (see Traina 1979, 178 = 1981, 116); the lexical choice of duplico in the rare sense of “bend in half ” would activate it. For the problems connected with the etymology and use of supplex, Heinze 1960, 28ff., is still useful. Together with Traina (1979, 176 n. 5), I see humilis supplex as a pair of attributes in asyndeton. The nexus has perhaps a Steigerung effect (cf. Kroll 1923 on Catull. 1.1, recalling A. 12.888 [telum . . .] ingens arboreum). See in general La Cerda on 12.930 (“membris omnibus precabundum Turnum . . . non solum ore, sed manibus, oculis, toto corpore”). 18. On the value of humilis, see again Traina’s exemplary reading (1979, 176–­180 = 1981, 114ff.). For the terminology, see also the recommendations περὶ έλέου given by Cicero, Inv. 1.109: orantur . . . illi qui audiunt humili et supplici oratione ut misereantur (“one should beg the listeners with a humble and supplicating style to provoke compassion”; cf. in Vergil humilis supplex, 930; oro, 933; miserere, 934; for deprecor at 931 see Cic. Inv. 2.104: deprecatio est, in qua . . . ignoscendi postulatio continetur [“ ‘deprecatio’ is an utterance requesting forgiveness”]). 19. Aligned with oro (933) and intentionally reverberating nec deprecor (931). There is a clear reference to the verb λίσσομαι that introduces Hector’s final request (Il. 22.338). 20. For examples of this tradition, see Lebek 1976, 155 and n. 67, and the observations by Homeric scholars (see, from different point of views,

158

N O T E S T O C hapter 4

Marg 1976, 13–­14; Gould 1979; Pedrick 1982; Fenik 1968, 83: “it is interesting that a rather large number of persons plead for their lives in the Iliad [. . .], but every one of them is slain”). 21. Every detail in the description invites us to see Aeneas as deeply tormented, as is customary before a decision for clemency is made. On volvens oculos (939), see Anderson’s incisive comments (1971). Aeneas’s tendency to flectere his mind and accept the supplication is implicit also in the words he pronounces while killing Turnus, 12.948–­49: Pallas te hoc volnere, Pallas (cf. 12.943 victum quem volnere Turnus, with clear assonance in the clause) / immolat. The insistence on Pallas’s name implies a marked antithesis. Its actual premise is the reversal of the opposite determination (“it is not I—­the one whom you are supplicating—­who is killing you: it is Pallas”). Turnus’s name, in line 939, acquires the same relevance; the killer’s name is the only non-­ alliterating term within the entire phonic sequence of the verse (on the mise en relief of nonalliterating terms, especially proper names, see the cases listed by Traina 1981, 128 n. 19). 22. Williams 1977, 503. 23. On the peculiar and isolated nature of the scene in the context of the Iliad, see, for instance, Burkert 1955, 103ff. 24. Just like Priam’s words, but much more directly and with more irony, Pyrrhus’s cruel reply, referes ergo haec et nuntius ibis / Pelidae genitori. Illi mea tristia facta / degeneremque Neoptolemum narrare memento (“You shall bring that news back and serve as a messenger to my father, the son of Peleus. Don’t forget to tell him my dark actions and that Neoptolemus is a degenerate!” 2.548–­49), presupposes Homeric inspiration. Relegated to Hades, Achilles cannot wait to get news of his son and deeply rejoices when Odysseus tells him that Neoptolemus is a glorious warrior, just like his father (see Macleod 1983, 2), who excelled in the capture of Troy: Od. 11.492–­540. One may add that at A. 2.540ff. the elements in Priam and Achilles’ encounter are established, even formalized, along more markedly “juridical” lines than they were in the Iliad (At non ille . . . Achilles / talis in hoste fuit Priamo; sed iura fidemque / supplicis erubuit). Priam insists on his status as hostis and on the iura fidemque connected to the requests of a supplex. One is tempted, for this thorough “Romanization” of the context, to go back to Ennius’s Hectoris Lytra. The fragments offer some hint in this direction: scen. 186 Vahl2, per vos et vostrum imperium

N O T E S T O C hapter 4

159

et fidem, Myrmidonum vigiles, commiserescite, where, in spite of serious textual issues ( Jocelyn 1967, 297ff.), one can detect the use of formulas foreign to the literary Greek tradition of hiketeia, but closer to the protocols of the Roman deditio in fidem (see Jocelyn 1967, 299); scen. 188–­89, melius est virtute ius . . . ius atque aecum se a malis spernit procul (“right is better than courage . . . right and equity are segregated from evil people”), where most likely the issue of reclaiming Hector’s body was addressed in terms of fairness and law—­a setting not typical of either Homer or Aeschylus. See also Deichgräber 1972, 124–­25. This kind of codification is notably absent from the supplication in Aeneid 12, where Vergil’s readers are, without adaptations, directed to the pathos typical of the Homeric model. 25. See Peppmüller 1876, 241–­43 (still useful for the analysis of the scholia ad loc.); Macleod 1982, 129–­30. Macleod has convincing arguments for χεῖρ’ ὀρέγεσθαι (“bring one’s hand to the mouth”), and believes A. 1.486 to be a misunderstanding on Vergil’s part. Such a hypothesis is obviously not necessary; what matters is that Vergil is representing Priam’s and Turnus’s supplication according to the same iconographic system. At any rate, an intermediary step between the two scenes may be found in visual artifacts. There are several examples of a Priam who, kneeling in front of Achilles, stretches his hand or hands toward him (cf. RE XXII. 2 coll. 1871–­1977; LIMC 1.1 [Zürich and München 1981], 148–­ 61). Here too we must keep in mind the existence of tragic models now lost—­see also earlier, note 24. 26. On the success of this motif, see the remarks by Rutherford 1982, 158–­60 (the importance of Il. 24 for Greek tragedy). On the topoi connected to pity in general, but with useful references to our archetype, see Stevens 1944 and 1941, who attests to the widespread presence of similarly pathos-­charged situations in Roman theater. Knauer 1964, 283 n. 2, remarks that the scene between Achilles and Priam is not structurally repeated in the Aeneid, and adds: “One may wonder whether Vergil did not imitate specific Homeric scenes perhaps for the reason that they were simply too famous.” It is, however, exactly on account of its notoriety that the scene was fit to be recalled by implication. Here, as in other cases, Knauer concerns himself excessively with those imitations that he calls (somewhat incorrectly) “structural.” 27. Language is rendered completely powerless, instead, in Vergil’s narrative. The failure of Turnus’s sermo can be grouped with other aspects

160

N O T E S T O C hapter 4

of the Aeneid that show—­in contrast with Homer himself—­great distrust of the power of speech (see also Feeney 1983). Thus the Vergilian story ends with the manifestation of an anger triggered by memory (saevi monimenta doloris . . . furiis accensus et ira / terribilis, 945–­47) that structurally recalls the wrath of Achilles; similarly, for this narrative function, the memor ira of saeva Iuno at the beginning of the story (A. 1.4; cf. also 1.25–­26, necdum etiam causae irarum saevique dolores / exciderant animo) evokes Poseidon’s anger that serves the same function at the beginning of the Odyssey (1.20, ὁ δ’ ἀσπερχὲς μενέαινεν). 28. Cf., respectively: schol. Hom. Il. 24.486a (bT) p. 600 E.; Ps. ­Heracl. All. Hom. 59.7; in addition, on the episode as example of how to create homoiopatheia in one’s audience, Apsin. Rhet. Gr. 1, 318.4 Spengel-­Hammer. 2 9. The fact that the same motif returns in 12.43–­45 is an intriguing example of Vergil’s “epic competence.” In this passage, Latinus begs Turnus to stop fighting using arguments similar to those Turnus will employ to move Aeneas. Commentators (see also Giancotti 1983, 442) compare the lines miserere parentis / longaevi, quem nunc maestum patria Ardea longe / dividit with Il. 22.59–­61, where old Priam begs Hector not to fight Achilles with similar words: πρὸς δ’ ἐμὲ τὸν δύστηνον ἔτι φρονέοντ’ ἐλέησον, / δύσμορον, ὅν ῥα πατὴρ Κρονίδης ἐπὶ γήραος οὐδῷ / αἴσῃ ἐν ἀργαλέῃ φθίσει (miserere—­ἐλέησον; maestum—­δύστηνον . . . δύσμορον; longaevi—­ἐπὶ γήραος οὐδῷ; one should notice also how longaevus, a solemn word first attested in Vergil, well reflects the rarity of the epic expression ἐπὶ γήραος οὐδῷ; see also aequaevus at A. 2.561, perhaps suggested by τηλίκος at Il. 4.487). One may add that the motif of the ­father’s distance, which for obvious reasons is absent from P ­ riam’s words (and is, in fact, a bit odd in the Vergilian context, given the proximity between Ardea and Larentum), could be integrated with another Homeric passage: not by chance this is the dialogue between Achilles and Priam in Iliad 24 (in turn strictly connected to the dialogue between Hector and Priam; cf. Macleod 1982, 22, and in his notes on 24.487 and 516). Priam reminds Achilles that Peleus is far away, helpless, and Achilles regrets the fate of his distant father: οὐδέ νυ τόν γε / γηράσκοντα κομίζω, ἐπεὶ μάλα τηλόθι πάτρης / ἧμαι ἐνὶ Τροίῃ (“I am not able to look after him, since I am staying in Troy, so far away from home”; 24.540–­42). This contamination proves, once more, Vergil’s skill in capturing thematic connections between elements

N O T E S T O C hapter 4

161

appearing at different points in the Homeric text and using them to generate imitations. 30. Miseri is made prominent both by placement at the beginning of the sentence as well as syntax and meter. The term, placed between two possessives recalling each other (tua/te) and two marked caesuras, is emphasized by Vergil no less than ἐλεεινότερός περ is in the Homeric text. For a more functional than lexical analogy, cf. the role etymology plays in the phrase miseram . . . miserias in Enn. scen. 256–­57 V2. Cf. also Hom. Od. 8.530–­31 ἐλεεινοτάτῳ ἄχεϊ . . . ἐλεεινὸν . . . δάκρυον. The verbal correspondence allows us to consider the suggestive sympathetic juxtaposition between Odysseus, who recalls the sack of Troy, and the captive woman, widow of a warrior who fell while defending his own city. 31. This redistribution of pathetic contents is so effective that after some time it serves as auctoritas, influencing the very reception of the Homeric model. In this regard it is useful to look at the work of a Roman literary author, who for evident reasons should have “frontally” transcribed the Homeric model, but understandably was unable to forget his Vergil. In the Latin Iliad, Turnus’s supplication rebounds upon both Homeric models and transforms them; to be precise, the Priam of Iliad 24 learns from Turnus how to acknowledge defeat: vicisti Priamum (1041); the Hector of Iliad 22 performs a real supplication: Afflicti miserere patris, moveat tua Peleus / pectora pro Priamo, pro nostro corpore Pyrrhus (“Please have pity of a grieving father, may Peleus touch your soul in favor of Priam, and Pyrrhus in favor of my body”; 986–­87), in terms foreign to the Homeric model, and instead directly suggested by Turnus—­”instructed,” as is now evident, by the Priam of Iliad 24. (For the reference to Achilles’ son, we may notice that in Homer Hector does not talk about Pyrrhus, and in Vergil Turnus does not refer in his appeal to Ascanius’s existence. The reference to Ascanius is present, however, in a parallel passage, at A. 10.524. Further­more, for 12.933 Servius intriguingly suggests placing a comma between Anchises and genitor and to take genitor as vocative, as if to mean: “your father is similar to mine, and you are a father yourself.”) 32. Not content with this solution, Statius—­never a poet to give up—­ adds to his most Vergilian epic a conclusion that, while exploiting multiple epic and tragic suggestions, in practice recalls the Homeric ending that Vergil “dismissed.” After the decisive battle, yet another

162

N O T E S T O C hapter 4

entire book of the Thebaid is dedicated to themes emulating Iliad 24: burial of bodies, funeral rites, repeated supplications by relatives, burials denied and then granted, threnodies, and even an altar dedicated to the goddess Clementia. By making Vergil and Homer overlap, as he is wont to do, Statius almost seems to denounce the “cut” Vergil made into the contents of the Homeric model, and reacts with a peculiarly personal horror vacui. 33. As seen earlier (chapter 2) the alternative isotopy that foreshadows the killing of a suppliant Turnus is rather supported by the description of Aeneas’s behavior in the aristeia immediately following Pallas’s death. 34. As an example of this trend, see Schrijvers 1978, who nevertheless treats the matter with balance, and without those excesses common in German scholarship between the two wars. 35. Clemency toward the defeated asking for pardon is indeed a topos of the official ideology and Augustan propaganda: Vell. Pat. 2.86.2 (after Actium): victoria vero fuit clementissima nec quisquam interemptus est nisi paucissimi et qui ne deprecari quidem pro se sustinerent (“it was a most merciful victory, and nobody was killed except very few, the ones who could not even endure to plead for clemency”); Res Gestae 3.1: victorque omnibus veniam petentibus civibus peperci (“in victory, I spared all citizens who were asking for mercy”). 36. From this vantage, it is worth returning to some of Conte’s conclusions (1978, 48 = 1980b, 83): “contradictory registers enrich the epos when reason, and with it language, is divided: when an age is divided. . . . The reasons of others, voiced in all their force, not only increase the artistry of the poem, but are a reminder that no victory is everlasting. Even the dead may return, if those who won have not been able also to vindicate their reasons, voicing them with the utmost strength.” Conte’s reasoning provides a useful term of comparison, not only because it gives a balanced representation of Vergil’s ideological world but also, and most importantly, because he locates the ideology of the text in the very modality of its composition, the structuring of the contents. When perspectives directing the narrative multiply, “relations of truth” get complicated; the narrative word is disputed between the subjectivities of various characters and the all-­knowing objectivity of the authorial voice. In short, ideology emerges in the complex form given to it by epic communication.

N O T E S T O A ppendix

163

37. This choice of genre allows Vergil to speak “from the remote legendary center of his poem” (as R. Sabbadini nicely put it). For him, going back to Homer is the main resource available for reconstructing this imaginary world, separated from his present (in Vergil’s own chronology) by more than a millennium.

A ppe n di x 1. Cf. 10.464 f.: as noted in chapter 2, this characterization of Heracles (exceptionally) crying builds on a widespread Greek tradition (­Bacchyl. 5.156ff.; Soph. Trach. 1070ff.; Eur. H. F. 1352ff.). By contrast, Juturna seems to be a character without literary roots: Vergil’s auffal­ lende Erdichtung (“conspicuous fiction”; Heinze 1915, 245). Perhaps not by chance (I am anticipating here a guiding principle of my reading) she is about the only figure in the Vergilian pantheon who finds an autonomous space to reflect “critically” on her own condition; it would have been difficult to employ similarly a divine figure who for Vergil’s audience was already somehow defined and characterized by cultural or literary traditions. 2. On epic norm and epic code, cf. the approach of Conte 1980b. 3. Kroll 1908, 526, rightly notes the importance of this style of asyndetic cola, but explains it as a mere consequence of Vergil’s rhetorical education. 4. The combination Turne . . . iuvare is particularly significant if Vergil had in mind the etymology of Juturna from iuvare (attested in Varro, L. 5.71; cf. Serv. Dan. ad Aen. 12.139, p. 591 Th.-­H.). The name would then encapsulate the entire narrative function of the character—­ almost a quae iuvat Turnum! For modern etymologies and historical attestations, see Radke 1965, 280ff. Ancient etymologies also connect to iuvare the names of Jupiter (Enn. Var. 58 V2, and so on) and Juno (Varr. L. 5.67). 5. For syntactical antithesis, cf. the normal, “balanced” Homeric type θνητὰς ἀθανάτῃσι (Od. 5.213; Greek material in Fehling 1969, 280ff.). 6. For example, vitam dedit aeternam does not seem to be a very marked and poetic hyperbaton: cf. Lucr. 5.1175 aeternamque dabant ritam; Philodem. de piet. 361 Gomp. αἰωνίους ἀναδέχονται τὰς συμφοράς (“they suffer eternal misfortunes”); Cic. Tusc. 1.9 in miseriam nascimur sempiternam (“we are born for endless misery”).

164

N O T E S T O A ppendix

7. An analogous effect is in Hor. Carm. 3.5.14, dissentientis condicionibus: the shrieking of the two long prosaic words evokes the heroic toughness of Attilius Regulus (an unconvincing assessment of this prosaic style in Axelson 1945, 103). For the collocation mortis condicio, see ­Publil. D 12; Prop. 3.2.22. Conceivably reponit (878), in the sense of “giving in exchange,” is also prosaic: cf. Cic. Fam. 1.9.19 and the passages cited by Mayor 1881 in his commentary on Juvenal 1.1. 8. Crudelis must in fact be nominative (the arguments favored by Conington, Cartault, and Henry are conveniently assembled by Pease 1935 ad loc.) rather than vocative (but Austin 1955 ad loc. refuses to decide, claiming we should retain the poet’s desired ambiguity). 9. There are probably variations of usual formulas from funerary language; cf. Epiced. Drusi 298: Quor sine me, quor sic incomitatus abis? 10. Aut quae iam satis ima dehiscat / terra mihi? (10.675–­76) occurs in circumstances closer to the Homeric spirit. For this kind of exclusively poetic question expressing desire, see Hofmann-­Szantyr 1965, 331. 11. The type would be archaic in itself (Enn. Ann. 90, 565 V2 and so on) if the monosyllable cannot make a metrical phrase with what comes before. Here, however, the precedent of Catullus’s pathetic style helps: at non haec quondam blanda promissa dedisti voce mihi, non haec miserae sperare iubebas (“but this isn’t what you once promised me, with your seductive voice. You didn’t urge me to hope for this !”; 64.139–­40). Indeed the presence of strong punctuation before haec makes it much more problematic to accept the penthemimeral caesura (for an analogous case, see Drexler 1967, 107). The clash between syntax and metrical expectations may create turbulence that, by making meaning difficult to ascertain, serves to emphasize the word. 12. The thesis of a symbolic value in Vergilian repetitions is developed in extreme form by Putnam 1964. He imposes a system of parallels and recurrences on the text working from suggestions by Pöschl 1977, 165, who insisted specifically on the parallel between Turnus and Dido (and thus, in a subordinate way, between Juturna and Anna). The particular echoings marked here can be added to the dossier opened by La Penna 1967, who rightly moves beyond the “typological” parallels and recurrences noted by Pöschl: the homology concerns books 4 and 12 in their entirety (not by chance, many of the analogies fall under the rubric “tragic register,” as discussed earlier), and the priority of book 4 suggests itself in terms of chronology of composition.

N O T E S T O A ppendix

165

13. For a stylistic reading of the passage, cf. Williams and Carter 1974. I would emphasize the frequency of elevated stylistic features: nox intempesta, 846 (in the same metrical place in Enn. Ann. 102, 167 V2); mortalibus aegris, 850; horrificum, 851 (Lucr. 3.906; -­fer Pac. trag. 82); rex, 851, at hexameter end (traditional since Enn. Ann. 175, 177 V2; this renders Carter’s point on the whole improbable, when he claims that rex would connote Jupiter’s “despotism”: Williams and Carter 1974, 173, n. 5). 14. Hübner 1970, esp. 12–­42; 74ff. (the reference to Roman tragedy on p. 76 is interesting); see also Grassmann-­Fischer 1966, 100ff. The distinction between Hellenic “winged daemons” and Roman ominous birds is probably too clear-­cut, if one considers the popularity of flying daemons in Etruscan culture. 15. See, in general, Thierfelder 1956, 98ff. A traditional question is why Juturna uses the plural. It is difficult to believe that the plural evokes the confused state of Juturna’s soul (so Nettleship in the note ad loc., where “Turnus” is evidently a slip for “Iuturna”); unlike her brother, the nymph is perfectly compos sui. Heyne and Wagner ad loc. speak of a generalizing plural, with a vague reference to tragic diction. But remember that Juturna identifies the Dira by its inauspicious sound (At procul ut Dirae stridorem agnovit et alas . . . , 869; alarum verbera nosco . . . , 876) and thus she may easily be referring to the category of this apparition—­birds of bad omen. 16. The entire finale of the Aeneid is regulated by an economy of details that reduces and suppresses the usual Homeric richness. For example, only a rationalist critic would ask when exactly Juturna has left the guise of the charioteer Metiscus (taken on in 12.784) in order to appear, at the end of the monologue, now in the form and dress of a river nymph (cf. 855). The poet makes space only for dramatically significant details. 17. On the Heinzian typology of the monologue, see Offerman 1968. 18. Furthermore, it is not even correct to claim that Vergil adheres passively to the Homeric pattern of the threnos in “regular” situations such as Evander’s lament (11.152ff.). Following Heinze, Knauer 1964 maintains: später will er [sc. Evander] klagen, wie die Frauen um Hektor klagen (“later on he will lament in the way the women lament over Hector’s body”) (343). On the contrary Vergil indicates clearly the ritual lament of women (quae postquam matres and so on, 11.146ff.)

166

19. 20.

21. 22.

N O T E S T O A ppendix

precisely to disconnect Evander’s following speech from the ritual function of ἐξάρχειν γόοιο (lead the lament). Evander’s speech initially foregrounds the apostrophe to the dead, following typical practice (152), but then immediately removes itself from the pattern of laudatio funebris (praise for the dead). The originality of the monologic structure is evident if one considers the articulation of the apostrophes; Evander addresses successively his dead wife (158), the Trojans present at the funeral rites (164), then again his deceased son Pallas (169), and finally the absent Turnus and Aeneas (173–­75; 177–­79). With this final turn Evander’s lament assumes a narrative function in the economy of the poem; not simply a manifestation of grief as an end in itself but also a message (mandata, 176) that makes Aeneas confront the necessity to exact vengeance from Turnus. In the same way Aeneas’s “threnos” for Pallas, though conceding something to ritual (an initial ­apostrophe to the dead, 11.42ff.), turns quickly to address Evander: Non haec Evandro de te promissa parenti . . . Infelix, nati funus crudele videbis! (“these were not the promises I made your father Evander . . . miserable man, to see your son’s cruel burial!”) (Evander will say non haec, o Palla, dederas promissa parenti [“Pallas, this was not the promise you gave your ­father”], 152). In this way the two so-­called threnoi are reciprocally integrated and strategic, weaving a sort of dialogue from afar that prepares and makes necessary the concluding vendetta. Vergil has reworked the literary form of the funeral lament into an instrument of narrative anticipation; in this lies its absolute independence from the Iliad, where the laments for Hector are rather effective means of conclusion and closure (a compositional solution reprised in the last book of Statius’s Thebaid). For the typology, popular and literary, of the threnos, consult Reiner 1938; see also Maas 1936. To explain this exceptional use of the motifs of goos it has been argued that the poet of this episode is reworking epic models where Thetis grieves for Achilles in death (Kakridis 1949, 65ff.; cf. Kullmann 1960, 36–­37). On the pathetic importance of weeping for a living person, see schol. Il. 24.85, 536f. E.; cf. Il. 6.499–­500; Aesch. Ag. 1322; Suppl. 113ff., with Fraenkel 1950, III, 617. Cf. Russo and Simon, 1968. For the frequent passage from second to first person in dramatic soliloquy, cf. Schadewaldt 1966, 56ff.; in the discourse of Electra “die

N O T E S T O A ppendix

167

Du-­Form ist . . . keine Anrede, sondern ein Ausdruck des Ethos” (“the second person is not a form of address, but an expression of ethos”; 57). Apostrophe then becomes a real means for structuring the pathos of discourse in the monologues of Alexandrian verse. The three parts of the discourse of Europa in Mosch. 2.135ff. hinge on three different apostrophes (to the bull, 135; to herself, 146; to Poseidon, 149); Ariadne’s monologue in Catullus signals her changing states of mind, marked by the “tercet” of 164–­66, by a switch from addressing Theseus to referring to him in the third person (tibi, 161, but ille autem, 167). 23. Useful is the synopsis by Barner 1971, 277ff. 24. For the reasons to attribute these verses to Electra, see Di Benedetto in his commentary ad loc. and already in Di Benedetto 1961, 138–­39. 25. Bömer 1957, 30, rightly underlines this in relation to the Ovidian version of the saga. 26. The tragedy also has an instance of the motif of Jupiter’s ingratum cubile, represented by the sad fate of Io. Prometheus comments: πικροῦ δ’ ἔκυρσας, ὦ κόρη, τῶν σῶν γάμων / μνηστῆρος (“Girl, you’ve got a bitter suitor for marriage”) (739–­40). 27. Fowler 1919, 152, claims that magnanimus cannot be read as ironic, but one needs to compare first the tone of Juno’s words in the parallel occurrence at 12.144 (magnanimi Iovis ingratum . . . cubile). It is no less misleading to think that magnanimus is a regular epithet for Jupiter (see Williams 1977, 448 and 502); on the contrary, in Latin this adjective always refers to human beings and is applied to a god only in these two passages in Vergil (cf. ThlL s.v., col. 103). The same seems true for the Greek term μεγάθυμος (“great-­hearted”), used of a divinity (Athena) just twice in Homer (Od. 8.520; 13.121), an exceptional usage that has been noted by Homeric critics (cf. Marzullo 1952, 161). Also, the combination with iussa superba is unmistakable; for superbus in the sense of “tyrannical,” cf. A. 8.481–­82 (the superbum imperium of Mezentius); 11.539. The union between the two concepts in rather diverse situations is found already in Plaut. Amph. 212–­13; Catull. 64.85. 28. The less remote poetic parallel was cited by Heyne ad loc., and disappeared from later commentaries. In Adonis’s epitaph, attributed to Bion, Venus laments not being able to follow her beloved to the under­ world: ἁ δὲ τάλαινα ζώω καὶ θεός ἐμμι καὶ οὐ δύναμαί σε διώκειν (“O wretched life I live, I am a goddess and cannot follow you”) (52–­53). The lament of a god for his immortality is also found in a late version of

168

29.

30.

31.

32.

N O T E S T O A ppendix

the myth of Glaucus, narrated in the scholia of Plato’s Republic (10.611, 275 Greene). This passage of On the Sublime is connected by Pohlenz 1911, 109–­10 (who does not seem to be aware of the idea in Philodemus) to fragments of the Demeter of Philitas. This attempt to backdate the concept to the early Hellenistic era is not persuasive: in Philitas there is absolutely no talk of eternity as a source of eternal affliction, and in fact one of the few extant fragments refers to a consolation of the goddess. On the whole, Demeter’s unhappiness seems expressed in a totally conventional way (as rightly noted by Webster 1964, 40–­41). It seems to me better to cite rather Plin. Nat. 2.27 nec (deus) sibi potest mortem consciscere, si velit, quod homini dedit optimum in tantis vitae poenis (“god cannot decree death for himself if he wanted to, which is the best thing he gave man in a life full of so many torments”) (a passage that commentators do not seem to have fully understood: cf. the edition of Beaujeu 1950, 131 n. 3). The influence of Philodemus on the environment of the young Vergil seems undeniable (see for example Frank 1967, 48–­49) as well as the direct circulation of his teaching (more probable than any written diffusion: Rostagni 1959, 380ff.). The most convincing coincidences are obviously those, as indeed in our case, that do not concern already established literary motifs. A specialist in Epicureanism like Gigante has insisted on the need to enrich the ties between Augustan poetry and the teaching of Philodemus (1969; for Horace especially, 65ff.; 123ff.). As for the salience of Epicurean motifs in the Aeneid, one must say in this area things are often exaggerated, by overinterpreting verbal coincidences with Lucretius or through strained readings of Epicurean ideology (for example, Frank 1920, 115ff.); but there are undisputable instances, often mediated by the voice of a character outside the action of fate (as is the case with Dido’s famous invective at 4.379–­80). The reference is in general to La Penna 1966, and in particular to the claim that there is “a gap between the Aeneid as an Augustan poem and the human drama that gives it substance” (LXXXII). The picture, perhaps distantly suggested by the Homeric episode of the Nereids quoted earlier, is inscribed formally in the usual representation of a fluvial divinity: cf. the Tiber of 8.33–­34 (eum tenuis glauco velabat amictu / carbasus [“fine linen veiled him in a gray mantle”]; see the

N O T E S T O A fterword

169

commentaries for cross-­references to the figurative tradition). For the clause condidit alto, cf. 8.66 and Hom. Od. 4.425; 11.253. Much more interesting is the problem posed by the gesture of covering the head, a gesture that notoriously receives in the Roman world various connotations inside and outside the context of ritual: see the thorough collection of evidence in Freier 1963; Freier seeks to connect all aspects (177) to the original idea of a consecration to underworld gods. Juturna’s action (fixed by the rare collocation caput contegere: cf. Catull. 64.64) could refer in a generic way to a mournful reaction (documentation in ThIl s. v. caput, col. 387, 21ff., by B. Maurenbrecher). A painting by the Greek artist Timanthes caused a great sensation in antiquity: to suggest Agamemnon’s mourning at the sacrifice of Iphigenia the painter depicted him, alone among those present, with veiled head (see Cic. Or. 22.74; Val. Max. 7.11 ext. 6; Plin. Nat. 25.73; Quint. 2.13.13). Other famous images of suffering could be tracked down in Greek tragedy. But if Juturna’s desire for death is taken as the keynote in the whole episode (with the help of its Epicurean contrastive underpinning), we may be tempted to offer an even stronger suggestion: the fact that the ancients also used to cover their heads before suicide (the passages in Waltz 1939, 292ff., and Freier 1963, 153–­57).

A f t e rwor d 1. For example, some of the metaphors used in Macrobius’s Saturnalia for the intertextual relationship between Vergil and Homer are still influential as ways of representing imitation: Vergil is imagined as composing de Homeri speculo “from the mirror of Homer,” 5.12.3 and 6.2.1; contrast Alciphron, fr. 34 Avezzù on the Odyssey “beautiful mirror of human life”), and “with Homeric threads” (5.2.9 Homericis filis). On actual Homeric citations in Roman commentaries on Vergil, note Scaffai 2006; on the importance of comparing Servius with Greek scholiastic interpretations of Homer, see for example Fraenkel 1964, II, 339–­90; Mühmelt 1965. 2. The same starting point, for a different project, in Dekel 2012; my 1984 book already argues, but very briefly (earlier, chapter 4), that the intertextual relationship with both Homeric poems is influenced by the relationship between the Odyssey and the Iliad. In this respect I accept the idea that the entire Aeneid is an Odyssean rewriting of the Iliad,

170

3.

4. 5.

6.

N O T E S T O A fterword

but the later revisionist approach (inaugurated by the important book of Cairns 1989) that there is no “Iliadic half ” in the poem seems to me to be an overreaction. My comments later are restricted to specialized studies of Homeric imitation: in a more general perspective, studies that have changed the research agenda certainly include Hardie 1987 and Quint 1993; for updated introductions I recommend Martindale 1997 and Farrell and Putnam 2010 (with rich bibliography), and also the illuminating critical survey of intertextuality in Roman poetry by Hinds 1998. My own current approach to the Aeneid is represented by Barchiesi 2006 and by my essay in progress mentioned earlier (The War for Italia, California University Press, based on the 2011 Sather Classical Lectures). Nelis 2001 on Apollonius; Farrell 1991 on the Georgics. The study of early Roman epic and tragedy as an influence on the Aeneid, for example, has long been limited to micro-­contexts because of the fragmentary nature of the sources (in a good number of cases known to us from source indications transmitted by late antique Vergilian scholars and grammarians), and also because early twentieth-­ century attempts to reconstruct large-­scale patterns of imitations have encountered skepticism. A good example of this limitation is the solid and helpful book by Wigodsky 1972, who clearly states that authors like Livius Andronicus, Naevius, and Ennius should not be expected to contribute models that work on the scale and narrative extension of Homeric imitations. The question is difficult but worth re-­examining, as shown for example by the recent interventions of Goldschmidt 2013 and Elliott 2013; it also involves the important question of how far “Roman history” (the main subject in the epics of Naevius and Ennius) is a macro-­model for the Aeneid, just as important as the Homeric poems, and of how we decide to deal with “history” as a quasi-­textual narrative model. In fact the last generation of work on Livy provides important clues for the study of the Aeneid. There is no space here for a real discussion, but in my view this does not prove that the traditional concept of Odyssean and Iliadic half is irrelevant and should be forgotten: the bipartition is clearly thematized within the plot, but the poem creates a dynamic approach to Homeric intertextuality, so that structures that seem definitive can be redefined “in corso d’opera.”

N O T E S T O A fterword

171

7. This is not to say that warfare with the local people is unnatural in ancient colonization and travel stories: the point is rather that the scale and moral intensity of the conflict are not justified by the existing traditions about the Aeneas legend, and owe a great deal to the Iliad: the poem itself thematizes this appropriation as a surprise effect. 8. Although recent discussions of Nostos as a word and a cultural construct often argue that the meaning “home-­coming” is not the core value of this idea in archaic Greek culture. 9. On the importance of keeping in mind Lycophron and on colonization in the Aeneid cf. the fundamental insights provided by McNelis and Sens 2011 (importance of Trojan glory and survival in the Alexandra; to be followed by other studies by the same authors, and by a major reappraisal of Lycophron by S. Hornblower); Horsfall 1989 (important on Vergil and Greek colonial culture, but there is need for more attention to the different approach to colonization in Roman society and culture). 10. This stems not only from generational affinity but also from the fact that, like me, those fellow-­travelers tend to draw their impulse from developments in modern literary criticism and critical theory—­for example, the illuminating discussion of self-­reflexivity in ancient poetry by Hinds 1998 acknowledges an early stimulus in the lectures of Christopher Ricks on English poetry (Hinds 1998, 3 and n. 7). 11. There is, however, a good reason to keep this perspective in mind, since what we have of classical literature is, inter alia, a product of self-­canonization and networking: authors have a vested interest in building credentials through acts of reciprocal or unilateral acknowledgment, and textual interaction creates cultural capital for them. 12. More generally, this kind of interest must have something to do with the increasing awareness, typical of my generation of scholars, of the need for a historical approach to the act of reading. It is also significant that in the same period (early 1970s to 1980s) studies of Hellenistic poetry had started emphasizing the connections between poetic choices and Homeric interpretation as expressed in the Alexandrian scholarly tradition (on this approach, see for example Rengakos 1994): there are however significant differences, since for historical, linguistic, and cultural reasons this approach tends to focus rather on the “glossographic” tradition represented by the D scholia, or on the leading scholars of the Alexandrian library, represented chiefly in

172

N O T E S T O A fterword

the A tradition. Issues of style, dialect, and glosses appear in a different light when the imitation is in another language, although the question of how far Roman authors were using glossaries, scholia, and learned tools remains a valid one. 13. New developments made available after my book include not just additional comparanda between Homeric scholia and Virgil’s imitation of Homer (for example, Schmit-­Neuerburg 1999), but also a number of invaluable instruments that should probably encourage new work on the Aeneid as a part of Homeric reception history: the first reliable edition of the Odyssey scholia (Pontani 2007 and 2010, to be continued); overviews of Homeric reception that include allegory and philosophical discussions (Lamberton and Keaney 1992; Pontani 2005, useful for his broad perspective); a comprehensive introduction to Greek scholarship (Dickey 2007, with huge bibliography); stimulating work on the rhetoric, poetics, and aesthetics embedded in the Homeric scholia (Richardson 1980; Nünlist 2009); a web-­based project for the exploration of the scholia, as a part of the Homer Multitext Project designed by G. Nagy; an updated revision of the only comprehensive dissertation on Homer in Latin poetry (Scaffai 1991); discussion of the fragments of Crates (Broggiato 2001); see also Nelis 2010, 13–­24, and Hexter 2010, 26–­36 (both with excellent bibliography); nuanced analysis of the evolution of Homeric scholarship such as Montanari 1994 and 2002 (note particularly the paper by M. Schmidt in Montanari 2002 on the strategic difference between A and bT traditions). For reason of space I cannot even begin to list important work on the transformation of the Homeric text, such as the studies of Gregory Nagy and the discussions generated by his interventions: Vergilian criticism is constantly influenced by turns and sea-­changes in Homeric research. It would be interesting to see new research on the Aeneid in this area, considering that the traditions of Homeric interpretation (and its impact on Alexandrian or Hellenistic poetry and prose) are perhaps not yet sufficiently integrated into contemporary commentaries on the Aeneid. 14. On the importance of keeping an eye on the making and unmaking of classics in antiquity and in the modern age, see Porter 2006, in particular the introduction by the editor. 15. The observations of Hexter 2010, 26–­36, on “Vergil’s Homer” are important and should be combined with Casali 1998, an innovative paper

N O T E S T O A fterword

173

arguing that some aspects of Ovid’s poetry express the anticipation, and playful manipulation (“foreknowledge of criticism”), of reading modes now familiar from later scholiastic interpretations of ancient poetry, but already active in the Augustan age—­for example, in the form of teaching and discussion of classical poets. 16. Cf. the programmatic statements of the Homer Multitext Project at the Center for Hellenic Studies (website), and its bibliography. For advice and help with this paper and reactions to my old book I am grateful to Joe Farrell, Denis Feeney, Philip Hardie, Stephen Hinds, Filippomaria Pontani, David Quint, and the late Don Fowler.

WORKS CITED

Anderson, W. S. 1969. The Art of the Aeneid. Englewood Cliffs, NJ. ———. 1971. “Two Passages from Book 12 of the Aeneid,” California Studies in Classical Antiquity 4, 49–­65. Axelson, B. 1945. Unpoetische Wörter. Lund. Baltes, M. 1983. “Zur Eigenart und Funktion von Gleichnissen im 16. Buch der Ilias,” Antike und Abendland 29, 36–­48. Barchiesi, A. 1978. “Review of Présence de Vergile. R. Chevallier ed., Paris, 1978,” Atti della Scuola Normale Superiore di Pisa 3, 8, 1543–­47. ———. 1980. “Le molte voci di Omero. Intertestualità e trasformazione del modello epico nel decimo dell’Eneide,” Materiali e Discussioni per l’analisi dei testi classici 4, 9–­58. ———. 1982. “Le armi nel cielo: diffrazione di un tema narrativo in Virgilio,” in Atti del Convegno Internazionale: Letterature classiche e narratologia, 117–­30. Rome. ———. 1983. “Lettura del decimo libro dell’Eneide,” in M. Gigante, ed., Lecturae Vergilianae III. L’Eneide, 343–­64. Naples. Barner, W. 1971. “Die Monodie,” in W. Jens, ed., Die Bauformen der greichischen Tragödie, 277–­320. Munich. Bassett, S. E. 1923. “Hector’s Fault in Honor,” Transactions of the American Philological Association 54, 117–­27. Beaujeu, J. 1950. Pline l’Ancien. Histoire Naturelle. Livre I.–­Livre II. Paris. Bolchazy, J. 1978. “From Xenophobia to Altruism: Categories of Homeric and Roman Hospitality,” Ancient World 1, 45–­64. Bömer, F. 1944. “Studien zum 8. Buch der Aeneis,” Rheinische Museum 92, 319–­69. 175

176

W orks C ited

———. 1957. P. Ovidius Naso. Die Fasten, I. Heidelberg. Buchheit, V. 1969. “Ciceros Triumph des Geistes,” Gymnasium 76, 232–­53. Büchner, K. 1947. Der Schicksalsgedanke bei Vergil. Freiburg. Bühler, W. 1964. Beiträge zur Erklärung der Schrift vom Erhabenen. Göttingen. Burkert, W. 1955. Zum altgriechischen Mitleidsbegriff. Diss. Erlangen–­ Nuremburg. Callen King, K. 1983. “Foil and Fusion: Homer’s Achilles in Vergil’s Aeneid,” Materiali e Discussioni per l’analisi dei testi classici 9, 31–­57. Camps, W. A. 1969. An Introduction to Virgil’s Aeneid. Oxford. Chantraine, P. 1952. “Le divin et les dieux chez Homère,” in H. J. Rose et al., La notion du divin depuis Homère jusqu’a Platon, Entretiens sur l’Antiquité Classique, 1, 47–­94. Fondation Hardt, Vandoeuvres-­Geneve. Ciani, M. G. 1975. “Ripetizione formulare in Apollonio Rodio,” Bollettino dell’Istituto di Filologia Greca 2, 191–­208. Conington, J., and H. Nettleship. 1963. The Works of Virgil. Hildesheim. Conrad, E. 1922. Untersuchungen über die Technik der Reden in Vergils Aeneis. Diss. Tübingen. Conte, G. B. 1965. “Il Trionfo della Morte e la galleria dei grandi trapassati in Lucrezio III 1024–­1053,” Studi Italiani di Filologia Classica 37, 114–­32. ———. 1974. Memoria dei poeti e sistema letterario. Torino. ———. 1980a. II genere e i suoi confini. Cinque studi sulla poesia di Virgilio. Torino. (Reworking of “Il balteo di Pallante,” Rivista di filologia e di istruzione classica 98 [1970], 291–­300.) ———. 1980b. “Saggio d’interpretazione dell’Eneide. Ideologia e forma del contenuto,” in ll genere e i suoi confini, 55–­107. Turin. ———. 1981. “A proposito dei modelli in letteratura,” Materiali e Discussioni per l’analisi dei testi classici 6, 147–­57. Conway, R. S. 1928. Harvard Lectures on the Virgilian Age. Cambridge, MA. Cook, A. B. 1940. Zeus: A Study in Ancient Religion, 3 vols. Cambridge. Deichgräber, K. 1972.5. Der letzte Gesang der Ilias (Abhandlungen der Akademie der Wissenschaften und der Literatur). Mainz. Di Benedetto, V. 1961. “Note critico-­testuali all’Oreste di Euripide,” Studi Classici e Orientali 10, 122–­55. ———. 1965. Euripidis Orestes. Florence. Drexler, H. 1967. Einführung in die römische Metrik. Darmstadt.

W orks C ited

177

Duckworth, G. E. 1933. Foreshadowing and Suspense in the Epics of Homer, Apollonius, and Vergil. Diss. Princeton University. ———. 1956. “Animae Dimidium Meae: Two Poets of Rome,” Transactions and Proceedings of the American Philological Association 87, 281–­316. Ehwald, R. 1894. “Vergilische Vergleiche,” Philologus 53, 729–­44. Feder, L. 1954.”Vergil’s Tragic Theme,” Classical Journal 49, 197–­209. Feeney, D. 1983. “The Taciturnity of Aeneas,” Classical Quarterly 33, 204–­19. ———. 1991. The Gods in Epic. Oxford. Fehling, D. 1969. Die Wiederholungsfiguren und ihr Gebrauch bei den Griechen vor Gorgias. Berlin. Fenik. B. 1968. Typical Battle Scenes in the Iliad: Studies in the Narrative Techniques of Homeric Battle Description. Wiesbaden. Fontenrose, J. 1981. Orion: The Myth of the Hunter and the Huntress. Berkeley–­Los Angeles. Forman, R. J. 1973. A Commentary on Vergil, Aeneid 10. Diss. New York University. Fowler, W. W. 1919. The Death of Turnus. Oxford. Fraenkel, E. 1950. Aeschylus. Agamemnon, II. Oxford. ———. 1957. Horace. Oxford. ———. 1964. Kleine Beiträge II. Rome. ———. 1967. “Anreden an nur gedachte Zuhörer,” Museum Helveticum 24, 190–­93. Frank, T. 1920. “Epicurean Determinism in the Aeneid,” American Journal of Philology 41, 115–­26. ———. 1967. Vergil: A Biography, 2nd ed. New York. Freier, H. 1963. Caput velare. Diss. Tübingen. Fuchs, H. 1947. “Rückschau und Ausblick im Arbeitsbereich der lateinischen Philologie,” Museum Helveticum 4, 147–­98. Galinsky, G. K. 1968. “Aeneid 5 and the Aeneid,” American Journal of Philology 89, 157–­85. Geffcken, J. 1907. Zwei griechische Apologeten. Leipzig–­Berlin. Genette, G. 1982. Palimpsestes. Paris. Giancotti, F. 1983. “Lettura del dodicesimo libro dell’Eneide,” in M. Gigante, ed., Lecturae Vergilianae 3. L’Eneide, 389–­524. Naples. Gigante, M. 1969. Ricerche filodemee. Naples. Gomperz, T. 1866. Philodem über die Frommigkeit (Herculanische Studien, 2). Leipzig. Gould, J. 1973. “Hiketeia,” Journal of Hellenic Studies 93, 74–­103.

178

W orks C ited

Grassmann-­Fischer, B. 1966. Die Prodigien in Vergils Aeneis. Munich. Griessmair, E. 1966. Das Motiv der “mors immatura” in den griechischen metrischen Grabinschriften. Innsbruck. Griffin, J. 1978. “The Divine Audience and the Religion of the Iliad,” Classical Quarterly 28, 1–­22. Heinze, R. 1906. “Supplicium,” Archiv fur lateinische Lexikographie und Grammatik 15, 89–­105. (Now in Vom Geist des Römertum, Stuttgart, 1960, 28ff.) ———. 1915. Vergils epische Technik, 3rd ed. Leipzig-­Berlin. ———. 1993. Virgil’s Epic Technique. Berkeley, CA. Henry, J. 1883. Aeneidea. New York. (= 1972, 2nd ed.) Heyne, C. G., and G.P.E. Wagner. 1830–­41. P. Vergili Maronis Opera, 4th ed., 5 vol. Leipzig. Hofmann, J. B., and A. Szantyr. 1965. Lateinische Syntax und Stilistik. Munich. Hommel, H. 1957. “Euripides in Ostia,” Epigraphica 19, 109–­64. (= 1976, Symbola I, Hildesheim–­New York, 147ff.) Hornsby, R. A. 1966. “The Armor of the Slain,” Philological Quarterly 45, 347–­59. ———. 1970. Patterns of Action in the Aeneid. Iowa City. Hubbard, M. 1974. Propertius. London. Hudson-­Williams, A. 1978. “Lacrimae Illae Inanes,” Greece and Rome, n.s. 25, 16–­23. Hübner, W. 1970. Dirae im römischen Epos. Hildesheim. Jal, P. 1963. La guerre civile à Rome. Paris. Jocelyn, H. D. 1967. The Tragedies of Ennius. Cambridge. Kakridis, J. T. 1949. Homeric Researches. Lund. Kassel, R. 1958. Untersuchungen zur griechischen und römischen Konsolationsliteratur (Zetemata, 18). Munich. Keller, R. M. 1946. “Iste Deiktikon in Early Roman Dramatists,” Transactions and Proceedings of the American Philological Association 77, 261–­316. Klingner, F. 1967. Virgil. Bucolica Georgica Aeneis. Zürich–­Stuttgart. Knauer, G. N. 1964. Die Aeneis und Homer. Studien zur poetischen Technik Vergils mit Listen der Homerzitate in der Aeneis (Hypomnemata, 7). Göttingen. Knoche, U. 1934. “Der Römische Ruhmesgedanke,” Philologus 89, 102–­24. (Reprinted in H. Oppermann, ed., Römische Wertbegriffe, Darmstadt, 1967, 420–­45.)

W orks C ited

179

König, A. 1970. Die Aeneis und die griechische Tragödie. Diss. Berlin. Kopff, E. C. 1981.”Virgil and the Cyclic Epics,” Aufstieg und Niedergang der römischen Welt, 31, 2, 919–­47. Krischer, T. 1979. “Unhomeric Scene—­Patterns in Vergil,” in F. Cairns, ed., Papers of the Liverpool Latin-­Seminar II, 143–­54. Kroll, W. 1908. “Die Originalität Vergils,” Neue Jahrbücher für das klassische Altertum 21, 513–­31. ———. 1923. Catull. Leipzig and Berlin. Kühn, W. 1957. “Rüstungsszenen bei Homer und Vergil,” Gymnasium 64, 28–­59. ———. 1971. Götterszenen bei Vergil. Heidelberg. Kullmann, W. 1956. Das Wirken der Götter in der Ilias. Berlin. ———. 1960. Die Quellen der Ilias (Troischer Sagenkreis). Wiesbaden. Kurz, G. 1966. Darstellungsformen menschlicher Bewegung in der Ilias. Heidelberg. Labarbe, J. 1949. L’Homère de Platon. Liège. La Cerda, J. L. de. 1612. P. Vergili Maronis Opera. Leiden. La Penna, A. 1966. “Introduction,” in Virgilio. Tutte le opera. Florence. ———. 1967. “Amata e Didone,” Maia 19, 309–­18. Lattimore, R. 1942. Themes in Greek and Latin Epitaphs. Urbana, IL. Leach, E. W. 1971. “The Blindness of Mezentius (Aeneid 10.762–­768),” Arethusa 4, 83–­89. Lebek, W. D. 1976. Lucans Pharsalia: Dichtungsstruktur und Zeitbezug (Hypomnemata, 44). Göttingen. Lenaz, L. 1980. “Regitur fata si Iuppiter ipse . . . ,” in Perennitas. Studi in onore di A. Brelich, 293–­309. Rome. Ley, J. 1877. Vergilianarum Quaestionum Specimen Prius. De Temporum Usu. Saarbrucken. Liebing, H. 1953. Die Aeneasgestalt bei Vergil. Diss. Kiel. Lloyd-­Jones, H. 1971. The Justice of Zeus. Berkeley–­Los Angeles–­London. (= Quaderni Urbinati di Cultura Classica 12, 1971.) Lyne, R.O.A.M. 1983. “Vergil and the Politics of War,” Classical Quarterly 33, 188–­203. Maas, P. 1936. θρῆνος. RE VI A 1, col. 596ff. Macleod, C. 1982. Homer: Iliad Book XXIV. Cambridge. ———. 1983. Collected Essays. Oxford. Malten, L. 1961. Die Sprache des menschlichen Antlitzes im frühen Griechentum. Berlin.

180

W orks C ited

Maréchal, A. 1940. “Sur la mort de Lausus: Virgile, Aen. X, 811–­832,” in Mélanges de philologie, de littérature et d’histoire anciennes offerts à Alfred Ernout, 251–­57. Paris. Marg, W. 1976. “Kampf und Tod in der Ilias,” Würzburger Jahrbücher für die Altertumswissenschaft 2, 7–­19. Mariotti, S. 1966. Il “Bellum Poenicum” e l’arte di Nevio, 2nd ed. Rome. Marquardt, J. 1886. Das Privatleben der Römer, I, 2nd ed. Leipzig. Martindale, C. 1993. Redeeming the Text: Latin Poetry and the Hermeneutics of Reception. Cambridge. Marzullo, B. 1952. Il problema omerico. Florence. McDermott, E. A. 1979–­80. “The ‘Unfair Fight’: A Significant Motif in the Aeneid,” Classical Journal 75, 153–­55. Merkelbach, R. 1951. “Review of J. Labarbe, L’Homère de Platon, 1949,” Gnomon 23, 375–­79. Merklin, H. 1970. “Zu Aubhau und Absicht der Messalinus-­Elegie Tibulls,” in W. Wimmel, ed., Forschungen zur römischen Literatur (Festschrift Büchner), 301–­14. Wiesbaden. Metger, W. 1957. Kampf und Tod in Lucans Pharsalia. Diss. Kiel. Monaco, G. 1972. “Bonus Ancus,” Giornale Italiano di Filologia n.s. 3, 244–­50. Moulton, C. 1977. Similes in the Homeric Poems. Göttingen. Mühll, P. von der. 1952. Kritisches Hypomnema zur Ilias. Basel. Narducci, E. 1979. La provvidenza crudele. Pisa. Nickau, K. 1977. Untersuchungen zur textkritischen Methode des Zenodotos von Ephesos. Berlin–­New York. Nisbet, R.G.M., and M. Hubbard. 1970. A Commentary on Horace Odes, Book 1. Oxford. Norden, E., ed. 1957. Aeneis, Buch VI, 3rd ed. Leipzig. Obbink, D. 2002. “Virgil, Philodemus, and the Lament of Iuturna,” in J. F. Miller et al., eds., Vertis in usum: Studies in Honor of Edward Courtney, 90–­113. Munich–­Leipzig. Offerman, H. W. 1968. Monologe im antiken Epos. Diss Munich. Otis, B. 1964. Virgil: A Study in Civilized Poetry. Oxford. ———. 1969. “The Originality of the Aeneid,” in D. R. Dudley, ed., Virgil, 27–­66. London. Otto, A. 1962. Die Sprichwörter und sprichwörtliche Redensarten der Römer. Hildesheim. (= Leipzig 1890.) Owen Lee, M. 1979. Fathers and Sons in Virgil’s Aeneid. Albany, NY.

W orks C ited

181

Parry, M. 1971. The Making of Homeric Verse. Oxford. Pasquali, G. 1951. “Arte allusive,” in Stravaganze quarte e supreme, 11–­20. Venice. Paul, A. 1964. Die Barmherzigkeit der Götter im griechischen Epos, 32. Diss. University of Vienna. Pease, A. 1963. Cicero’s De divinatione, 2nd ed. Darmstadt. Pedrick, V. 1982. “Supplication in the Iliad and the Odyssey,” Transactions of the American Philological Association 112, 125–­40. Peppmüller, R. 1876. Commentar des vierundzwanzigsten Buches der Ilias. Berlin. Perkell, C. 1997. “The Lament of Juturna: Pathos and Interpretation in the Aeneid,” Transactions of the American Philological Association 127, 257–­86. Perret, J., ed. 1980. Virgile. Enéide, Livres IX–­XII. Paris. Perutelli, A. 1979. La narrazione commentata. Studi sull’epillio latino. Pisa. Pfeiffer, R. 1968. History of Classical Scholarship, I. Oxford. Philippson, R. 1920. “Zu Philodems Schrift über die Frömmigkeit,” Hermes 55, 225–­78. Pohlenz, M. 1911. “Die hellenistische Poesie und die Philosophie,” in Charities für Leo, 76–­113. Berlin. (= now in Kleine Schriften, Hildesheim, 1965.) Pöschl, V. 1952. “Das Zeichen der Venus und die Gestalt des Aeneas,” in Festschrift, Otto Regenbogen zum 60, 135–­43. Heidelberg. (Reprinted, in different form and with new material, in Die Dichtkunst Virgils, 70ff.) ———. 1977. Die Dichtkunst Virgils, 3rd ed. Berlin–­New York. ———. 1980. “Der Zweikampf zwischen Aeneas und Turnus,” in F. Krin­ zinger, ed., Forschungen und Funde, Festschrift Bernard Neutsch, 349–­55. Innsbruck. Putnam, M. 1964. The Poetry of the Aeneid. Cambridge, MA. ———. 1981. “Pius Aeneas and the Metamorphosis of Lausus,” Arethusa 14, 139–­56. Quinn, K. 1968. Virgil’s Aeneid: A Critical Description. London. Radke, G. 1965. Die Götter Altitaliens. Münster. Reed, J. D. 1997. Virgil’s Gaze: Nation and Poetry in the Aeneid. Princeton, NJ. Reiner, E. 1938. Die rituelle Totenklage der Griechen. Stuttgart. Rieks, R. 1970. “Die Tränen des Helden,” in M. von Albrecht, ed., Silvae (Festschrift für E. Zinn). Tübingen.

182

W orks C ited

Rostagni, A. 1959. “Il De Morte di L. Vario Rufo,” Rivista di filologia e di istruzione classica 37, 380–­94. Russo, J. 1981. “Cosa comunica Omero, e in che modo? Il verso omerico come messaggio e come strumento di comunicazione,” in E. A. Havelock and J. P. Hershbell, eds., Arte e comunicazione nel mondo antico, 51–­71. Rome–­Bari. (Already in Classical Journal 71, 1976, 289–­99.) Russo, J., and B. Simon. 1968. “Homeric Psychology and the Oral Epic Tradition,” Journal of the History of Ideas 29, 483–­98. Rutherford, R. B. 1982. “Tragic Form and Feeling in the Iliad,” Journal of Hellenic Studies 102, 145–­60. Samuelsson, J. 1905–­6. “Futurum historicum im Latein,” Eranos 6, 29–­44. Schadewaldt, W. 1966. Monolog und Selbstgespräch, 2nd ed. Berlin– ­Zürich–­Dublin. Schetter, W. 1972. “Senecas Oidipus-­Tragödie,” in E. Lefèvre, ed., Senecas Tragödien. Darmstadt. Schnapp-­Gourbeillon, A. 1981. Lions, héros, masques. Paris. Schrade, H. 1950. “Der Homerische Hephaistos,” Gymnasium 57, 38–­55 and 94–­122. Schrijvers, P. H. 1978. La valeur de la pitié chez Virgile, in R. Chevallier, ed., Présence de Virgile, 483–­95. Paris. Schweitzer, B. 1931. “Dea Nemesis Regina,” Jahrbuch des deutschen archäologischen Instituts 46, 175–­246. Segal, C. 1971. The Theme of the Mutilation of the Corpse in the Iliad. Leiden. Sittl, C. 1890. Die Gebärden der Griechen und Römer. Leipzig. Stevens, E. B. 1941. “Topics of Pity in the Poetry of the Roman Republic,” American Journal of Philology 62, 426–­40. ———. 1944. “Some Attic Commonplaces of Pity,” American Journal of Philology 65, 1–­25. Stork, T. 1970. Nil igitur mors est ad nos. Diss. Saarland. Stroux, J. 1933. “Die stoische Beurteilung Alexanders des Grossen,” Philologus 88, 222–­40. Tandoi, V. 1984. “Intorno a una citazione poetica latente in Tertulliano, Apol. 25, 8 Fato stat Iuppiter ipse,” in V. Tandoi, ed., Disiecti membra poetae. Studi di poesia latina in frammenti, I, 175–­99. Foggia. Tarrant, R. J. 2012. Virgil: Aeneid Book XII. Cambridge. Thierfelder, A. 1956. “Obscaenus,” in Navicula Chiloniensis: Studia filologica F. Jacoby oblata, 98–­107. Leiden.

W orks C ited

183

Thome, G. 1979. Gestalt und Funktion des Mezentius bei Vergil, mit einem Ausblick auf die Schluβszene der Aeneis. Frankfurt am Main–­Bern–­Las Vegas. Tomaševskij, B. 1965. “Thématique,” in T. Todorov, ed., Théorie de la littérature, 263–­307. Paris. Traina, A. 1974. “Mazio fr. 6 Mor. (Storia d’un verso),” in G. Puccioni et al., eds., Poesia latina in frammenti, 35–­42. Genova. (= Poeti latini [e neolatini] II, 2 ed., Bologna 1986, 47–­59.) ———. 1979. “Da Virgilio a D’Annunzio: ambiguità di un predicativo,” Materiali e Discussioni per l’analisi dei testi classici 2, 175–­81. ———. 1981. Poeti latini (e neolatini) II. Bologna. ———. 1997. Virgilio. L’utopia e la storia. Il libro XII dell’Eneide e antologia delle opere. Turin. Tränkle, H. 1960. Die Sprachkunst des Properz und die Tradition der lateinischen Dichtersprache. Wiesbaden. Vretska, K. 1976. C. Sallustius Crispus. De Catilinae coniuratione, 1. Heidelberg. Wackernagel, J. 1926–­28. Vorlesungen über Syntax, I. Basel. Walton, F. R. 1965. “A Neglected Historical Text,” Historia 14, 236–­51. Waltz, R. 1939. “Autour d’un texte de Sénèque (‘Nat. Quaest.,’ IV, praef. 17),” Revue des études latines 17, 292–­308. Warde Fowler, W. 1918. Aeneas at the Site of Rome. Oxford. Webster, T.B.L. 1964. Hellenistic Poetry and Art. London. Weische, A. 1966. Studien zur politischen Sprache der römischen Republik. Münster. Wellek, R., 1955. A History of Modem Criticism: 1750–­1950, II. New Haven, CT. West, D. 1969. “Multiple Correspondence-­Similes in the Aeneid,” Journal of Roman Studies 59, 40–­49. ———. 1974. “The Deaths of Hector and Turnus,” Greece and Rome 21, 21–­31. Whitman, C. H. 1958. Homer and the Heroic Tradition. Cambridge, MA. Wigodsky, M. “The Arming of Aeneas,” Classica et Mediaevalia 26, 192–­221. Williams, G. 1983. Technique and Ideas in the Aeneid. New Haven, CT–­London. Williams, R. D. 1960. P. Vergili Maronis Aeneidos liber quintus. Oxford. ———. 1977. The Aeneid of Virgil: Books 7–­12, 2nd ed. London.

184

W orks C ited

Williams, R. D., and C. J. Carter, 1974 “Critical Appreciations II: Virgil, Aeneid XII.843–­86,” Greece and Rome n.s. 21, 165–­77. Wilson, C. H. 1979. “Jupiter and the Fates in the Aeneid,” Classical Quarterly 29, 361–­71. Wlosok, A. 1967. Die Göttin Venus in Vergils Aeneis. Heidelberg. ———. 1973. “Review of Grassmann-­Fischer, Die Prodigien in Vergils Aeneis, 1966,” Gnomon 45, 245–­49.

SELECT INDEX

Achilles: and Patroclus 40, 45–­46, 56, 151 n 14; arms of 19–­20, 22–­24, 54, 56, 65, 83–­84, 130, 153 n 8; compared to Turnus 117; revenge of 4, 40, 62; supplication of x, 48, 85–­91, 144 n 47, 159 nn 25–­26, 160 n 29; see also supplex Alexander the Great 30, 117 Alexandrian scholarship: and Pergameme school 10, 33, 130; response to Homer 17, 32–­33, 70–­72, 74–­79, 126–­128, 172 n 13 Apollonius of Rhodes 71, 99, 117, 170 n4 aristeia 37–­38, 43–­44, 49, 82–­83, 150 n 13, 162 n 33 authorial intention 124 Bacchylides 11, 138 n 18, 154 n 10, 163 n 1 baldric theme: in Homer 18, 144 n 47, 145 n 49; in Vergil 15, 18–­22, 41, 47–­50, 90, 144 n 47, 145 n 51, 145 n 55 Cato (the Elder) 29 Catullus 99, 157 nn 16–­17, 164 n 11, 167 n 22, 167 n 27, 169 n 32 Cicero 9, 13, 30, 109–­110, 137 n 17, 139 n 22, 140 n 27, 141 n 40, 148 n 69,

148 n 71, 157 n 18, 163 n 6, 164 n 7, 169 n 32 civil war 58, 60–­66, 154 n 9 consolatio 13–­14, 168 n 29 cultural models: in Homer 3–­4, 11, 49, 74, 77, 106, 128; in Vergil 28, 31–­33, 46–­47, 51–­53, 61–­63, 65, 70, 145 n 56, 155 n 13, 157 n 15; see also hospitium, servare modum, supplex Dira(e) 96, 101–­103, 165 n 15 Ennius 17, 101, 142 n 41, 148 n 70, 158 n 24, 161 n 30, 163 n 4, 164 n 10, 165 n 13, 170 n 5 epic formulae: in Homer 2, 6, 14, 33, 49–­50, 74–­76, 81, 100, 159 n 24; in Vergil: 16–­17, 39, 50–­52, 77–­79, 155 n 7, 156 n 8, 164 n 9; see also genre model Epicureanism 111, 113, 168 n. 30 euchos 2, 20, 41 Euripides 11, 14, 27, 105–­108, 125, 140 nn 26–­27, 140 n 35, 148 n 67, 163 n 1 example model: xvi, 73–­74, 122; and epic typology 81–­82; and grammatical imitation 82; and lexical imitation 82; and Vergilian intertextuality 91–­92 185

186

select index

fama 12–­13, 21 fata: inevitability of in Homer: 5–­7, 9, 24, 27, 31–­32, 83–­84,137 n 11; in Vergil 6, 10–­12, 14–­15, 21–­22, 25–­ 26, 33, 36, 46, 58, 61, 82, 97, 112–­113, 138 n 22, 139 n 23, 141 n 40, 149 n 2, 168 n 30 genre model: xvi, 74, 122; and epic typology 86, 105; and formulaic expression 75–­77, 79, 85–­86, 122 gloria: in Homer 25, 83; in Vergil 12–­13, 136 n 5, 140 n 30; see also fama heroic death: and the Patrokleia: 3, 15, 26; conventions of in Homer 2, 5–­6, 7–­9, 14–­15, 24, 38, 83, 85–­87; in Vergil: 4–­6, 12–­14, 16, 18–­19, 41–­43; related to premature death: 11–­12, 14, 40, 106, 139 n 25 heroic duel: and androktasiai 38, 63; conventions of in Homer15–­18, 52, 142 n 41; in Vergil 18, 38–­39, 42–­ 45, 49–­50, 101, 136 n 5, 150 n 11 Homeric model: Aeneid as a continuation of 4, 16, 20, 52, 56, 69–­72, 112; as a functional element in Vergil 3, 5, 15, 77; Vergilian innovation on 6–­18, 60–­61, 161 n 31 Horace 30, 146 n 61, 148 n 67, 152 n 4, 154 n 10, 164 n 7, 168 n 30 hospitium 46–­47 isotopy 47, 56, 61, 65–­66, 91, 151 n 14, 162 n 33 Juvenal 164 n 7 lament: in tragedy 11–­12, 107–­108; symbols of in Homer 8–­9, 56, 104–­ 106; in Vergil 58, 95–­96, 100, 104–­ 105, 108, 152 n 4, 165 n 18; threnos 8, 105, 107, 165 n 18, 166 n 19 lion motif xi, 43, 80–­82, 120, 156 nn 9–­11

Livius Andronicus 142 n 41, 170 n 5 Livy 29, 31, 46, 103, 138 n 22, 147 n 64, 154 n 9, 170 n 5 Lucretius 141 nn 40–­41, 143 n 46, 163 n 6, 165 n 13; 168 n 30 Meleager 11, 143 n 46 metaphor 13, 19, 60, 77, 143 n 47, 169 n1 metonymy 20, 60, 63 monologue: in epic 98–­100, 108, 111, 165 n 16, 166 n 18; in Juturna’s speech 96–­97, 101–­104, 112–­113; in tragedy 107; in verse 167 n 22 Moschus 98–­99, 167 n 22 Naevius 71, 170 n 5 narrative disjunction 60–­61, 92–­93 narrative legibility viii–­x, xv, 16, 35–­ 36, 59 Odyssey: of Vergil 117; relation to ­Homer’s Iliad 71–­72, 118, 169 n 2; relation to Vergil’s Aeneid 26, 71, 118, 158 n 24, 160 n 27, 161 n 30, 170 n 6 Ovid 55, 60, 105, 137 n 15, 138 n 22, 140 n 26, 143 n 46, 147 n 64, 154 nn 9–­ 10, 173 n 15 parallelism: between characters 40–­41, 96; internal parallelism 58, 88–­89, 100–­104, 156 n 10,161 n 31, 164 n 12, 167 n 27; narrative parallelism between texts 36–­37, 42–­43, 49–­50, 56–­57, 71, 81, 149 n 1, 150 n 11, 151 n 16, 152 n 4, 167 n 28 Philodemus x, 111, 113, 126, 163 n 6, 168 nn 29–­30 Plato 8, 128–­130, 138 nn 18–­19, 138 n 22, 141 n 39, 167 n 28 Plautus 167 n 27 Plutarch 19, 140 n 33, 147 n 66, 148 n 70, 148 n 76, 154 n 9; pseudo-­ Plutarch 138 n 20

select index Polybius 30 polyphony viii, x–­xi, 31, 60, 65, 69 Polystratos 12 prodigy in Homer 8, 33; in Vergil 55–­61, 64–­67, 101, 103, 153 n 8 prolepsis: 20, 26–­28, 62, 146 n 58, 147 n 64; “historic future” 28 rhesis 96, 100 Sallust 13, 29, 140 n 32 self-­reflexivity 122–­123 Seneca (the Younger) 30, 139 n 25 servare modum 25, 29–­32

187

Sophocles 11, 19, 57, 107, 140 n 27, 163 n 1 Statius 152 n 4, 161 n 32, 165 n 18 Strabo 30 supplex 48, 84, 87, 157 nn 17–­18, 158 n 24; see also Achilles Theocritus 11 tears: of Hercules 6, 11, 139 n 25, 140 n 26; of Juturna 85; of Zeus 7–­9, 33, 137 n 16 Tibullus 60, 64, 154 n 9 tragedy: relation to epic viii, 27, 58, 105, 107–­109,112, 120, 159 n 26, 167 n 26, 170 n 5

SELECT INDEX LOCORUM

Aeschylus Pr. 92, 109; 99–100, 109; 511–512, 109; 739–740, 167 n 26 Anthologia Palatina Didot. Ill (1890), 123, and n. 217, 25–26, 14, 141 n 38 Cato pro Rhod. fr.163 Malc., 29 Catullus 64.139–140, 164 n 11 Cicero Arch. 28, 13 Div. 2.25, 9, 137 n 17, 139 n 22 Mil. 97, 13–14 Rab. Perd. 30, 13 Sest. 47, 13 Tusc. 1.65, 110 ; 2.25, 109; 3.21, 30 Epigrammatum graecorum 790, 6 f. Kaibel, 12, 140 n 28 Euripides Her. 1353, 11, 140 n 27 Or. 960–963, 107 Rh. 948–949, 108 Homer Il. 2.45, 8; 2.418, 142 n 41; 4.139, 141 n 41; 4.182, 100; 4.504, 142 n 41; 6.55ff, 85; 6.143, 8, 141 n 39; 9.531, 78; 188

9.577, 78; 10.135, 141 n 41; 11.53–55, 8 ; 11.391–392, 142 n 41; 11.749, 142 n 41; 13.618, 18, 142 n 43; 14.420, 142 n 41; 15.371, 78; 16.284, 141 n 41; 16.459, 7; 16.503–505, 18, 142 n 41; 16.693, 141 n 39; 16.794–800, 146 n 58; 17.51–52, 136 n 5; 17.187, 23; 17.194–197, 146 n 58; 17.198–208, 22, 25; 17.472–473, 148 n 73; 17.540, 78–79; 17.587ff, 136 n 6; 18.50–51, 106; 19.1–18, 56; 21.218, 75–76; 21.403–404, 157 n 14; 22.17, 142 n 41; 22.25, 83; 22.59–61, 160 n 29; 22.212–213, 96; 22.306, 141 n 41; 22.317, 83; 22.323, 23; 22.338, 84; 22.420–421, 89; 23.175–176, 62; 24.478, 87, 88; 24.485–487, 89; 24.486, 144 n 47; 24.503–504, 89, 90; 24.506, 87, 88; 24.508, 87, 88; 24.515, 87; 24.540–542, 160 n 29; 24.589, 151 n 19 Od. 4.150, 141 n 40; 8.520, 167 n 27; 8.530–531, 161 n 30; 13.121, 167 n 27; 21.96–99, 26; 22.296, 142 n 41 Horace Carm. 15.9–11, 153 n 4 Ep. 1.10.30–31, 30 Livy 30.13.8, 46; 30.42, 29; 45.8.6–7, 31

select index locorum Lucan 6.812, 152 n 3; 7.114–116, 152 n 4 Nicolaus Myrensis Progymn. 66.12 f. F, 98 Ovid Fast. 3.67–74, 55 Met. 2.621, 140 n 26; 15.788, 137 n 15 Tr. 3.12.48, 143 n 46; 4.2.44, 143 n 46 Pont. 2.2.78, 143 n 46 On The Sublime 9.7–8, 110–111 Philodemus de Piet. 2 (Gomperz 1866, 36l = ­Philippson 1920, 232), 111, 163 n 6 Plato Phaed. 115A, 141 n 39 Rep. 388B-C, 8, 138 n 18 Pliny Nat. 2.27, 168 n 29 Plutarch Arat. 49, 147 n 66 Sallust Cat. 1.3, 13 Seneca (Minor) Ben.2.16.2, 30; 5.6.1, 30 Ep. 99.18–19, 139 n 25 Servius ad Aen. praef. 115 schol. Il. 16.432 [A], 257ff. E., 8; 16.433– 38, 258 E., 9; 24.85, 536f. E., 166 n 20 schol. Hom. Il. 16, 459, 261 E., 9, 138 n 20 ; 17.201 (bT), 367 E., 32, 148 n 75; 19.13 (bT) p. 575 E., 153 n 8; 23.175–76. 398 E., 62 Sophocles Aj.1348–1349, 19

189

Elec. 808, 107; 812–813, 107; 822, 107 OC. 1451–1512, 57 Trach. 1074, 11 Strabo 15.1.5, 30 Vergil A. 1.100–101, 79; 2.548–549, 158 n 24; 4.673, 100; 6.89, 117; 6.828–831, 58, 152 n 4; 7.549, 61; 8.520–539, 54; 8.537–540, 58, 79; 10.356–361, 45; 10.431, 45 ; 10.433–438, 3, 40; 10.443, 41, 143 n 47; 10.460, 46; 10.464ff, 95, 163 n 1; 10.465–466 11; 10.467–469 12; 10.469–472, 14; 10.470–471, 7; 10.471–472, 15, 141 n 39; 10.473, 141 n 40; 10.495–496, 18, 50; 10.496–497, 21; 10.490–494, 42; 10.500, 148 n 73; 10.501–505, 26; 10.503–504, 143 n 47; 10.515–517, 46; 10.524–525, 48, 161 n 31; 10.565–569, 44; 10.675–676, 100, 164 n 10; 10.689–90, 39, 149 n 4; 10.755–757, 45; 10.763–768, 44; 10.824, 2, 41, 143 n 47; 10.825–828, 42; 10.862, 78;12.4–9, xi,81, 120; 12.43–45, 160 n 29; 12.154–157, 95; 12.867–868, 102; 12.869–886, 97, 165 n 15; 12.879–882, 107, 108; 12.880, 99; 12.885–886, 113; 12.918–919, 103; 12.923–924, 84; 12.930–931, 85, 87, 88; 12.932–934, 48, 86; 12.932–936, 88, 90; 12.947– 948 23, 90, 158 n 21 Ecl. 1.3, 78; 5.27, 156 n 9 G. 1.495–497, 154 n 10; 1.506–508, 60, 62; 1.510–511, 61

INDEX OF MODERN AUTHORS

Anderson, W. S. 117, 146 n 59, 158 n 21 Axelson, B. 147 n 65, 164 n 7 Baltes, M. 156 n 11 Barchiesi, A. vii–xii, 125, 135 n 1, 136 n 8, 170 n 3 Barner, W. 167 n 23 Bassett, S. E. 145 n 56 Beaujeu, J. 168 n 29 Bolchazy, J. 151 n 15 Bömer, F. 143 n 46, 152 n 1, 167 n 25, Broggiato, M. 172 n 13 Buchheit, V. 140 n 32 Büchner, K. 139 n 23 Bühler, W. 155 n 2 Burkert, W. 158 n 23 Callen King, K. 155 n 3 Cairns, F. viii, 118, 170 n 2 Camps, W. A. 144 n 48 Carter, C. J. 165 n 13 Casali, S. 172 n 15 Chantraine, P. 137 n 11 Ciani, M. G. 156 n 8 Conington, J. 141 n 40, 148 n 74, 153 nn 6–7, 164 n 8 Conrad, E. 140 n 33 Conte, G. B. viii–ix, xi, 20, 119, 121–122, 124, 135 n 1, 137 n 14, 140 n 36, 144 190

n 49, 149 n 8, 155 nn 4–5; 156 n 8, 162 n 36, 163 n 2 Conway, R. S. 135 n 2, Cook, A. B. 137 n 15 Dekel, E. 118, 169 n 2 Di Benedetto, V. 167 n 24 Dickey, E. 172 n 13 Drexler, H. 164 n 11 Duckworth, G. E. 146 n 57, 146 n 61, Eco, U. 120, 124 Ehwald, R. 122, 137 n 14, Elliott, J. 170 n 5 Erbse, H. 126–128 Farrell, J. 118–119, 121, 170 nn 3–4, 173 n 16 Feder, L. 146 n 60 Feeney, D. 113, 121, 160 n 27, 173 n 16 Fehling, D. 163 n 5 Fenik. B. 158 n 20 Fontenrose, J. 150 n 13 Forman, R. J. 137 n 16 Fowler, D. xii, 121,135 n 1, 135 n 5,173 n 16 Fowler, W. W. 167 n 27 Fraenkel, E. 62, 117, 136 n 6, 143 n 46, 146 n 62, 154 n 10, 155 n 11, 166 n 20, 169 n 1

index of modern authors Frank, T. 168 n 30 Freier, H. 169 n 32 Fuchs, H. 148 n 70 Galinsky, G. K. xi, 135 n 3, 152 n 1 Geffcken, J. 137 n 17 Genette, G. 120, 124, 155 n 5 Giancotti, F. 160 n 29 Gigante, M. 168 n 30 Goldschmidt, N. 170 n 5 Gomperz, T. 111 Gould, J. 157 n 15, 158 n 20 Grassmann-Fischer, B. 152 n 1, 165 n 14 Griessmair, E. 141 n 37 Griffin, J. 137 n 12 Hardie, P. v, 121, 130, 135 n 2, 170 n 3, 173 n 16 Harrison, S. J. 121 Haupt, M. 122 Heinze, R. 26, 96, 104–105, 112, 139 n 23, 146 nn 57–58, 157 n 17, 163 n 1, 165 n 18 Henry, J. 140 n 29, 153 n 7, 164 n 8 Hexter, R. 127, 172 n 13, 172 n 15 Heyne, C. G. 165 n 15, 167 n 28 Hinds, S.E. 121–122, 124, 135 n 1, 170 n 3, 171 n 10, 173 n 16 Hofmann, J. B. 145 n 50, 147 n 64, 164 n 10 Hommel, H. 140 n 31 Hornsby, R. A. 144 n 49, 156 n 11 Horsfall, N. 171 n 9 Hubbard, M. 140 n 36, 154 n 10 Hudson-Williams, A. 139 n 25 Hübner, W. 165 n 14 Hutchinson, G. O. 122 Jal, P. 148 n 67 Jocelyn, H. D. 159 n 24 Kakridis, J. T. 166 n 20 Kassel, R. 140 n 33 Keaney, J. L. 172 n 13 Keller, R. M. 147 n 65

191

Kirk, G. S. 126 Klingner, F. 18, 135 n 1, 146 n 56, 152 n 1 Knauer, G. N. vii, ix, xv, 116–118, 136 n 3, 136 n 5, 136–137 n 9, 145 n 56, 149 n 78, 152 n 1, 153 n 7, 156 n 9, 159 n 26, 165 n 18 Knoche, U. 140 n 32 Konig, A. 140 n 27 Kopff, E. C. 149 n 6 Krischer, T. 149 n 7, 150 n 11 Kroll, W. 122, 145 n 50, 157 n 17, 163 n 3 Kühn, W. 139 n 23, 152 n 1, 153 n 6 Kullmann, W. 137 n 12 Kurz, G. 142 n 43 Labarbe, J. 138 n 18 Labate, M. 152 n 4 La Cerda, J. L. de. 143 n 46, 153 n 5, 157 n 17 Lamberton, R. 172 n 13 La Penna, A. ix, 121–122, 139 n 22, 140 n 30, 164 n 12, 168 n 31 Lattimore, R. 140 n 36 Leach, E. W. 150 n 13 Lebek, W. D. 157 n 20 Lenaz, L. 138 n 21 Ley, J. 147 n 63 Liebing, H. 152 n 1 Lloyd-Jones, H. 137 n 11 Lyne, R.O.A.M. 156 n 13 Maas, P. 166 n 19 Macleod, C. 158 n 24, 159 n 25, 160 n 29 Malten, L. 141 n 40 Maréchal, A. 136 n 6 Marg, W. 158 n 20 Mariotti, S. 155 n 1 Marquardt, J. 151 n 15 Martindale, C. ix, xi, 135 n 4, 170 n 3 Marzullo, B. 167 n 27 McDermott, E. A. 149 n 7 McNelis, C. 171 n 9 Merkelbach, R. 138 n 19 Merklin, H. 155 n 12 Metger, W. 142 n 41

192

index of modern authors

Monaco, G. 140 n 36, 141 n 38, 152 n 4 Montanari, F. 172 n 13 Moulton, C. 151 n 14 Mühll, P. von der. 145 n 54 Mühmelt, F. 169 n 1 Narducci, E. 153 n 4 Nelis, D. P. 170 n 4, 172 n 13 Nettleship, H. 7, 141 n 40, 148 n 74, 153 n 5, 165 n 15 Nickau, K. 138 n 19 Nisbet, R.G. M. 140 n 36 Norden, E. 140 n 29, 142 n 41, 143 n 46, 147 n 63, 147 n 65, 152 n 4 Nünlist, R. 172 n 13 Obbink, D. 111, 113 Offerman, H. W. 165 n 17 Otis, B. 136 n 3, 146 n 56, 155 n 4 Owen Lee, M. 151 n 14 Parry, M. 155 n 7 Pasquali, G. viii, 70 Paul, A. 137 n 12 Pease, A. 137 n 17, 139 n 22, 164 n 8 Peppmüller, R. 159 n 25 Perkell, C. 113 Perret, J. 150 n 12 Perutelli, A. xi, 148 n 77 Pfeiffer, R. 138 n 19 Philippson, R. 111 Pohlenz, M. 168 n 29 Pontani, F. 172 n 13, 173 n 16 Porter, J. I. 172 n 14 Pöschl, V. 112, 120, 136 n 6, 144 n 48, 152 n 1, 153 n 7, 156 n 9, 157 n 14, 164 n 12 Putnam, M. xi, 120, 151 n 18, 164 n 12, 170 n 3 Quinn, K. 145 n 51 Quint, D. 170 n 3, 173 n 16 Radke, G. 163 n 4 Reed, J. D. 113

Reiner, E. 166 n 19 Rengakos, A. 171 n 12 Richardson, N. 172 n 13 Rieks, R. 139 n 25 Rostagni, A. 168 n 30 Russo, J. 151 n 20, 166 n 21 Rutherford, R. B. 159 n 26 Samuelsson, J. 147 n 64 Scaffai, M. 169 n 1, 172 n 13 Schadewaldt, W. 166 n 22 Schetter, W. 139 n 24 Schlunk, R. R. x, 125–127 Schmit-Neuerburg,T. 172 n 13 Schnapp-Gourbeillon, A. 156 n 11 Schrade, H. 153 n 8 Schrijvers, P. H. 162 n 34 Schweitzer, B. 143 n 46 Segal, C. 149 n 9 Sens, A. 171 n 9 Shawcross, J. W. 124 Simon, B. 166 n 21 Sittl, C. 136 n 4 Stevens, E. B. 159 n 26 Stork, T. 140 n 36 Stroux, J. 147 n 66 Szantyr, A. 145 n 50, 147 n 64, 164 n 10 Tandoi, V. 138 n 21 Tarrant, R. J. xi, 113, 126 Thierfelder, A. 165 n 15 Thome, G. 144 n 47, 156 n 13 Todorov, T. 120 Tomaševskij, B. 15 Traina, A. 113, 117, 142 n 41, 148 n 67, 157 nn 17–18, 158 n 21 Tränkle, H. 148 n 67 Vretska, K. 140 n 32 Wackernagel, J. 147 n 64 Wagner, G. P. E. 126, 165 n 15 Walton, F. R. 154 n 9 Waltz, R. 169 n 32 Warde Fowler, W. 59, 153 n 5, 153 n 8

index of modern authors Webster, T. B. L. 168 n 29 Weische, A. 148 n 67 Wellek, R. 146 n 57 West, D. 120, 139 n 25, 157 n 14 West, M. L. 118 Whitman, C. H. 137 n 16 Wigodsky, M. 152 n 1, 170 n 5

193

Williams, G. 151 n 15, 156 n 9, 156 n 13 Williams, R. D. 137 n 16, 145 n 50, 146 n 56, 158 n 22, 165 n 13, 167 n 27 Wilson, C. H. 149 n 2 Wlosok, A. 152 n 1 Ziehen, J. 122