Bryn Mawr Classical Review

Bryn Mawr Classical Review 1999.08.07

D. Hershkowitz, The Madness of Epic. Reading Insanity from Homer to Statius.   Oxford:  Oxford University Press, 1998.  Pp. xiii, 346.  ISBN 0-19-815245-0.  $80.00.  

Reviewed by Andrew Zissos, University of Texas at Austin (
Word count: 2313 words

In this book Hershkowitz (henceforth "H.") offers a detailed examination of the treatment of madness in the epics of Virgil, Homer, Ovid, Lucan, and Statius. This is an interesting choice of subject: the centrality of notions of madness to tragedy is well established, but the same could not have been said of epic -- until now. In her wide-ranging and impressive study, H. demonstrates that madness is a crucial element in epic from Apollonius onwards, and is particularly important for Vergilian and post Vergilian epic. All told, this is a well-researched and intelligent volume, which makes valuable contributions to our understanding of each of the poems in question. Throughout the book H. demonstrates philological rigor and theoretical sophistication, along with a thorough mastery of the relevant bibliography.

Following a substantial introduction, a separate chapter is devoted to each poet. H. does not resort to a single theoretical approach: individual texts are examined from various perspectives, running the gamut from Stoic theories of the passions to R. D. Laing's theory of ontological insecurity. There is consequently a certain lack of overall cohesion to the study -- perhaps an inevitable result of the disparate group of chosen texts. In a methodological prolegomenon (14-16) H. offers a justification for her eclectic approach, while identifying a "poetics of madness" as the underlying focus.

The first chapter, an introduction to the entire book, covers a good deal of ground, and does so exceedingly well. After noting the problem of definition (as elusive an issue today as it was in the ancient world), H. identifies three important ancient models of madness -- medical, philosophical, and supernatural. To these, H. draws an analogy to three modern models -- respectively those of neurobiology, psychology, and sociology. As might be expected, the categories fail to correspond exactly, but they do provide helpful orientation for the subsequent analysis on the poetics of madness, which occupies the remainder of the chapter. In an important initial discussion, H. demonstrates that all three ancient models of madness can be found in epic -- though unsurprisingly supernatural models are most common. The use of medical and philosophical models is demonstrated mainly via well-chosen passages from Apollonius Rhodius and Valerius Flaccus. H. then proceeds to investigate the ways in which tragic notions of madness were recontextualized within Roman, and specifically post-Vergilian, epic. This process is exemplified via three categories of imagery: Io, Bacchants and prophetesses, and the Furies. Of particular value here are a fascinating investigation of the differences between the treatment of Io in Apollonius Rhodius and Valerius Flaccus, and an examination of the imagery of the Furies in post-Vergilian epic. Through such analysis, H. establishes that tragedy's profound influence on epic is felt particularly in the representation and imagery of madness; at the same time, epic madness, although derived from tragic madness, is shown to be modified by its recontextualization in a new genre. In the final section of the chapter, H. explores the conventional notion of the mad poet, painstakingly tracing it back to the fifth century BCE, and drawing some important distinctions between Vergil's use of the figure and that of his epic successors.

The second chapter, on Vergil, has two major sections. The madness of Turnus is the topic of the first half of the chapter (68-95). H. takes as her focus the description of Turnus' reaction to the news of the imminent destruction of Latinus' city (Aen. 12.665-71). This is a somewhat narrow topic for almost 30 pages of analysis -- and this may explain the hermeneutic "overkill" which at times obscures the argumentation. If I understand correctly, H.'s principal point is that line 669 (ut primum discussae umbrae et lux reddita menti), captures possibly the only moment in the poem after 7.458 in which Turnus is not mad. The idea that Turnus is sane here is not new -- that thesis was elaborated several years ago in an interesting article by Gaskins.1 What H. has done is to argue that this is a fleeting moment of sanity -- though one experienced within "the cracked mind of the schizophrenic" (94). This may well be an important assertion, but its impact is dissipated by such distractions as the extended comparison of the "discursive contexts" of Turnus' state of mind at 12.665-71 and Aeneas' at 2.604-6 (83-5), and the assertions that the messenger Saces gives Turnus the evil eye (87) and that, what is more, Turnus gives himself the evil eye (90).

The second part of the chapter (95-124) is more clearly successful. Here H. explores in detail the anger of Juno and her resort to underworld elements -- and madness -- in support of her anti-Trojan objectives. This might seem a shopworn topic, but H. builds effectively upon the analysis of previous scholars -- especially Feeney and Hardie -- to produce a subtle and compelling analysis of Juno's metapoetic status. In the second half of the Aeneid, the goddess seeks to usurp Jupiter's functions in an increasingly overt way. As H. shows, the struggle between the two deities takes on a metanarrative dimension, since it involves two competing visions of the narrative. Thus, when Juno resorts to the underworld and its forces of madness, she transforms Jupiter's poem of fata into a poem of furor. Closely related to this idea is Juno's association with closure and Juno with narrative openings. H. further enriches the discussion by locating not only Juno, but also Jupiter at the center of the various instabilities and tensions in the poem. She argues for a pervasive slippage encoded within the actions of Jupiter, who in the end resorts to Juno's expedient, namely, the deployment of an infernal agent. All in all, this is a fascinating analysis and a valuable contribution to Vergilian studies.

Chapter 3, the only one devoted to Greek poetry, addresses Homeric epic and the Iliad in particular. Though not without interest, this chapter is arguably the least essential of the book. The first part of the chapter (125-153) establishes Homer's poems as a "baseline" for the Roman epics, using the kind of linguistic analysis very familiar to Homeric studies. The chapter takes as its initial goal proving "the most important thing about madness in Homer" -- namely, that there is very little of it. This H. demonstrates with little trouble, primarily through an examination of madness terminology in Homeric epic. The initial focus is on the term ἄτη, which, as H. shows, is different in nature from explicit madness terms such as λύσσα and μαίνομαι. Using Agamemnon's early behavior towards Achilles as an example, H. goes on to argue that actions are perceived as driven by ἄτη only according to their outcome. Much of the analysis here seems to achieve little beyond a vague reformulation of Dodds' notion of Homeric society as a "shame culture."2 More significant, however, is H.'s concluding observation that though madness words are found in Homeric epic, they tend to be used by characters rather than by the narrator, and often in a pejorative rather than technical sense.

In the second part of the chapter (153-160) the discussion vaults into the realm of post-modernism, taking as its point of departure the book's hysteron-proteron ordering of chapters 2 and 3, which places Homer after Vergil. What H. advocates here is a reversal of the conventional direction of the intertextual relationship, in order to view Homeric madness as a "response" to Vergilian madness. As strange as it may sound, this move has a solid grounding in recent critical theory, and will be familiar to many Classicists through an important recent article by Don Fowler.3 In essence, H. attempts an explicit "Romanized rereading" of Homer, based on the premise that "in an obvious sense, one cannot read Homeric madness except through Vergilian madness" (126).4 A few of these Romanized rereadings yield mildly interesting results: the episode of Patroclus' death in Iliad 16, for example, does offer stronger hints of madness when read backwards through the stupefaction of Turnus. On the whole, however, H.'s strategy does not live up to its billing. A case in point is the analysis of Achilles' μῆνις, which attempts to "give it a mad dimension by placing it well within the range of passions which are indistinguishable from madness." To achieve this goal, the wrath of Juno (the Vergilian counterpart to Homeric wrath) is read backwards to illuminate aspects of Achilles' μῆνις. The implications of this move are rather ill-defined, though, and more intriguing possibilities are left largely unexplored.5

In the following chapter, H. ventures a psychologizing reading of Ovid's Metamorphoses. The opening section establishes the principal operating assumption of the chapter, namely, that metamorphoses can serve as metaphors for madness: "where someone in another epic would go mad, in Ovid's epic he or she will undergo metamorphosis instead." H. turns to the theories of R. D. Laing, and the concept of ontological insecurity in particular, as a key to reading the Metamorphoses, mapping various transformations in the poem onto the Laingian categories of engulfment, implosion and petrification. An attempt is made to forestall methodological objections by lowering the status of Laing's typology to that of an "interpretive tool." This is an interesting gambit -- and a necessary one since Laing's theories have been discredited, and thus cannot be supposed to serve as a transcultural stratum. All in all, the Laingian approach to Ovid seems to raise almost as many problems as it solves. One obvious inconvenience is that H. can identify no independent cases of "implosion" in the entire poem, and only two instances of overlap with other categories. Still, there is much of value in this chapter, including a useful discussion of erotic depersonalization in the stories of Narcissus and Propoetides. In one of the concluding sections H. explores the "metapoetics of ontological insecurity," offering a fresh perspective on the commonplace that Ovid's most dramatic metamorphosis is the transformation of the epic genre itself and providing a Laingian reading of the sphragis as the product of an ontologically insecure narrator.

The fifth chapter offers a valuable and compelling reading of Lucan's Bellum Civile. In two initial sections (206-218) H. discusses the manner in which the Bellum Civile deconstructs fundamental "Vergilian" oppositions such as furor vs. pietas. Much of the material here is a useful (and fully-acknowledged) rehearsal of John Henderson's important insights into Lucan, which provide the methodological underpinnings for H.'s own analysis.6 The final two sections (218-46) serve in tandem to make H.'s culminating and most important arguments. In essence, H. attempts to problematize the widely-espoused notion of a strict opposition between Lucan's Caesar and Cato by demonstrating a slippage in signification that occurs between crucial terms such as virtus and furor. The use of Henderson's deconstructive approach to refute a position primarily advocated by Henderson's precursors may appear gratuitous at first, but the fundamental importance of H.'s argument quickly becomes apparent. Crucial here is the striking observation that Cato, like Caesar, repeatedly prompts responses in his followers that have an aspect of madness to them. Also significant is the intriguing similarity adduced between the joy that Cato derives from a journey through snake-infested deserts and the joy that Caesar derives from a journey through enemy-infested territory. As H. points out, such affinities problematize the classification of the one as unambiguously embodying virtus and the other furor.

The final chapter combines a pair of very fine articles recently published by H. on Statius' Thebaid.7 The first part of this chapter identifies an inherent pattern of madness in the poem, namely, frenzied activity followed by enervation. H. demonstrates that this pattern is repeatedly echoed on various narrative levels as well as in the imagery of the poem, thereby illuminating an important aspect of Statius' thematic program. Much of the remainder of the chapter is devoted to a detailed discussion of the thematics of sexual perversion in the Thebaid. This analysis is uniformly excellent, and is bolstered by a number of subtle intertextual arguments. In particular, H. shows that the bull similes used to describe Eteocles and Polynices are programmatic, and that through their intertextual models (and Aen. 12.101-6 in particular), they carry a strong erotic charge. It is further demonstrated that Polynices acts not simply out of patriotism, but driven by the underlying sexual attraction that Thebes -- and Jocasta -- hold for him. Having effectively psychoanalyzed the sons of Oedipus, H. now turns her attention to the daughters. Although they, unlike the sons, are models of sexual purity, H. shows that the daughters are similarly ensnared by the problematic sexual equations that plague the family. Thus, in the Thebaid, even the pious love of Antigone and Argia towards Polynices becomes a form of madness. Likewise, the discussion of Ismene's sexual attitude, nicely illuminated via an intertextual relationship with Ovid's Byblis, shows that her loss of sexual innocence is figured as a perversion. In short, H. demonstrates convincingly that sex simply cannot be constructed normally in the house of Laius. Slightly less compelling is the treatment of the end of the poem, where H. argues against a positive view of Theseus' military intervention to permit the burial of the dead. Though H. succeeds in qualifying a wholly optimistic reading of the finale, few are likely to be persuaded by the assertion that Theseus' intervention requires him to exceed the level of madness already found at Thebes (298-9).

The book is written in a fluent and accessible style, with all Greek and Latin fully (and elegantly) translated. The standard of proof-reading is generally high, though occasional misprints do occur.8 All in all, this book represents an extremely valuable contribution to studies of ancient epic, and Roman epic in particular. H. has produced a thought-provoking and highly readable treatment of an important subject, which advances a number of interesting and persuasive readings of the poems in question. Her rich, insightful study deserves and no doubt quickly will find a wide readership.


1.   CQ 40 (1990), 1-15.
2.   E.R. Dodds. The Greeks and the Irrational (Berkeley and Los Angeles, 1951), esp. 1-27.
3.   MD 39 (1997), 13-34.
4.   Here H. might have done better to say (with John Henderson) that the Aeneid forces a specific reading of the Iliad upon its readers -- rather than implying that it precludes non-Vergilian readings.
5.   H. only mentions in passing K.C. King's interesting notion that Roman representations of Achilles' wrath read madness back into the Homeric character by activating the Stoic complex of images which equate anger and insanity (154). More generally, an exploration of the retrospective diagnoses that result from a more collectivist Roman (i.e. Vergilian) perspective on the self-indulgence of the Iliadic hero might have been worth pursuing.
6.   Ramus 17 (1988), 122-164.
7.   MD 33 (1994), 123-147; JRS 85 (1995), 52-64.
8.   I supply the following list, with no claims to exhaustiveness. Doubtful translations -- p. 92: Ares is not raving in the thickets of the mountains in the simile at Il. 15.605-8; pp. 112-3: Aen. 12.849-50 is translated awkwardly as "these attend the throne of Jupiter and in the savage king's threshold." Greek and Latin typos -- p. 10: for ἀνφροπίνων read ἀνθροπίνων; p. 72: for aestuo read aestuat; p. 73: there should not be a period after oculis in line 39 of the Catullus quote; p. 97: there should not be a period after the protasis flectere si nequeo superos; p. 133 for παλλὰ read πολλὰ; p. 202: for parvos ... cruor read parvus ... cruor; p. 229: for tiurbidus read turbidus; p. 262: for nex read nec; p. 289: for saevos pudor read saevus pudor. English typos -- p. 5: H. has "morals" where "mortals" seems to be meant; p. 36: for "self-reflectively" read "self-reflexively;" p. 82: "mixture with Stoic other philosophical ideas" is missing a conjunction; p. 111 "Why have changed your intentions?" either lacks a subject or suffers from poor word order; p. 262: H. has "apposition" where "opposition" seems to be meant; p. 271: for "in poetics terms" read "in poetic terms;" p. 284: period missing at the end of the first sentence.

Read Latest
Index for 1999
Change Greek Display
Books Available for Review

HTML generated at 13:26:26, Friday, 03 April 2009