Book Search:  


Google full text of our books:


Subjecting Verses:
Latin Love Elegy and the Emergence of the Real
Paul Allen Miller

Book Description | Endorsements | Table of Contents

COPYRIGHT NOTICE: Published by Princeton University Press and copyrighted, © 2003, by Princeton University Press. All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher, except for reading and browsing via the World Wide Web. Users are not permitted to mount this file on any network servers. Follow links for Class Use and other Permissions. For more information, send e-mail to

This file is also available in Adobe Acrobat PDF format

Chapter 1


It seems to me rather that we have to look at failures of form, the impossibility of certain kinds of representation in a certain context, the flaws, limits, obstacles, which become the clue to the social truth or social meaning.
     --Jameson 1998: 361
It can often be the emphasis on the impossibility of representation that gives the clue and organizes things.
     --Jameson 1998: 369

THE PURPOSE of this book is to provide a history or genealogy of the Latin love elegy. That history is problematic and demands a more comprehensive explanation in part because it is so short. Many books have already treated the form, and in recent years several have offered exciting and sophisticated readings of its rhetoric and modes of characterization (Greene 1998; Kennedy 1993; Veyne 1988), but none has offered a convincing exegesis of this subgenre's sudden flaring into existence and its just as sudden extinction. Indeed, most treatments have largely eschewed historical modes of reading, except for now outdated forays into the uncertain terrain of biographical criticism.

Latin love elegy first comes to light in the last years of Catullus's life, around 56 B.C.E. It effectively disappears with Ovid's death in exile in 17 C.E.1 Seventy-three years may not seem short by the standards of popular culture, but it is only a blink of the eye compared with the life-spans enjoyed by genres such as epic (Ennius to Statius)2 and verse satire (Lucilius to Juvenal)3 in the Roman world, or the sonnet sequence (Petrarch to Shakespeare and beyond)4 and the novel (Cervantes to the present)5 in modern times. Nonetheless, even this rather limited chronology is over-generous. Catullus is generally considered a precursor of elegy rather than an elegist in his own right. Ovid's exilic poetry shares with erotic elegy only meter, subject position, and allusions to the conventions defining the form. His beloved in exile is Rome, not some coy mistress. If we limit ourselves to the period between the appearance of Gallus's first book of elegies, generally considered the first complete exemplar of the genre, circa 50 B.C.E., and Ovid's publication of the definitive edition of the Amores, circa 7 B.C.E. to 1 C.E.,6 the last collection of love elegy to have any observable influence on subsequent literary history, then the entire genre, as an effective and authentic form of literary expression, can be said to have bloomed and died in a mere fifty years (Lee-Stecum 1998: 16-18; Albrecht 1997: 744; Elia 1981: 74-75; Boucher 1980: 34).

Of course, this does not mean that elegies ceased to be written. We continue to have references to occasional practitioners of elegiac verse later in the imperial period, but none of them merits appearance in Quintilian's canonical list of the Roman elegists (10.1) or Diomedes the Grammarian's fifth-century compilation (1.484) or had any recognizable influence on the literature of his day (Ross 1975: 101; Boucher 1980: 164). Thus, Pliny the younger mentions a nephew of Propertius (Epistles 6.15.1, 9.22.1-2),7 and Statius a certain Stella (Silvae 1.2),8 but neither of them left either any substantial record of his work. The genre's moment, it seems, had passed. The extreme tensions, which I argue constitute the elegiac subject position, no longer assumed the same forms. Erotic elegy's extraordinary public dramatization of a private sphere that both engages socially constituted norms of individual conduct and insistently calls them into question--the vital contradictions at the heart of its being--were no longer able to find a place from which they could be directly spoken. Those tensions were not so much resolved as displaced. The Ovid of the Tristia, as we shall see in chapter 8, could only continue the discourse of elegy by speaking from the realm of the dead. What we see in him and those who come after is the specter of elegy rather than elegy proper. No longer possible is the overtly contradictory position of a Tibullus, who accepts a life of traditional martial virtue for his patron Messalla but rejects it for himself (1.1; see chapter 4); of a Propertius, who casts his love for Cynthia in terms recalling Antony's for Cleopatra while praising Caesar's victory at Actium (2.15, 2.16; see chapter 5), or of an Ovid, who simultaneously invokes the power of ius while proposing stratagems of adultery (Amores 1.4; see chapter 6). The ideological space required for this type of openly split subject is no longer available (Boucher 1980: 34-35).

Instead, we see a new model emerge in which the subject is always already absent from view, speaking from nowhere, from a place beyond the contingencies of the here and now (Newman 1989: 1501). Persius under Nero digs a grave (scrobis) in which to tell the truth and bury it (1.119-20).9 Juvenal under Trajan dares only speak of (and thus effectively from) the dead (1.170-71). The position of the speaking subject has changed in a fundamental way (Auerbach 1965: 247-48; Foucault 1984: 105; Henderson 1993: 130; Edwards 1993: 32). As Shadi Bartsch demonstrates in her reading of Tacitus's Dialogus de Oratoribus, the place that was once the republican orator's, addressing the people in propria persona, could now only be occupied by the poet Maternus whose tragedy, Cato, by virtue of the dramatic form's remove from the first-person speaking subject, paradoxically became the heir to republican eloquence. Yet even this distance was not sufficient: for, as Tacitus intimates, Maternus paid with his life for the boldness of writing a historical drama implicitly critical of the regime (Bartsch 1994: 101-25). Only silence, it seems, could with candor address the present. Thus, the imperial period, as Pliny himself notes (Panegyricus 2.1-2), was one in which the seamless whole of public and private life that had constituted the arena of traditional republican virtus--a space whose increasingly contradictory and tense nature was the place of elegiac passion--had been definitively rent asunder (Bartsch 1994: 149-50). Propertius's nephew, Lucius Arruntius Stella, and others may have continued to produce elegies, and even to write about love, though we shall never know exactly what they said. But the subject position that constituted Roman erotic elegy, as the vital and authentic genre that is object of this study, had closed, and Ovid's Tristia was its obituary. This book offers not so much a chronology of names and dates related to the genre as a genealogy of that lack.

The question this study answers then is, Why? Why elegy? How can we explain both the sudden appearance and rapid disappearance of this subgenre that today constitutes one of antiquity's most widely studied forms of poetry and that exercised such a profound influence on the history of Western lyric verse?10 Indeed, as Stapleton (1996) has recently shown, neither medieval nor renaissance love poetry is comprehensible outside the light of Ovid's Amores, the last and fullest flowering of the genre.11 The Roman state lasted more than a thousand years, but one of its most conspicuous contributions to European culture survived less than fifty. What characterized this unique moment in time in which both the Roman state underwent the most profound reorganization of its recorded existence, the violent transformation from an oligarchic republic into a multinational empire, and in which one of the unique and lasting contributions to European culture was created?

To begin to answer to this question, this chapter examines two areas. First, how is genre history possible? How can we conceive of the relation between a given literary type and the world that produced it, without succumbing to the illusion that one of these terms possesses an ontological priority over the other so that either literature is an epiphenomenon of history or history the illusion projected by literature? To respond to these demands requires us to advance a theory of the relation of text to context that, while owing much to its predecessors, presents a new synthesis of Marxist, historicist, dialogic, and psychoanalytic ideas and relates them to first-century Rome. Second, why have previous studies of love elegy failed to come to terms with the history of the genre? Here, it is argued that this failure can be attributed to an inadequate conceptualization of literary history itself, thereby demonstrating the necessity of a theoretical model such as we are elaborating.

Finally, in the following chapters, we move on to a series of case studies, starting in chapter 2 with the earliest precursor of the distinctively Roman genre of first-person erotic elegy, Catullus, in whose carmina we find for the first time a collection of poems devoted to a single mistress (Kennedy 1993: 88-89; Hinds 1998: 29).12 The Greek models for elegy, while important, are all partial at best. Although Antimachus's Lyde possessed an erotic subjective frame for a series of traditional narrative elegies, and Callimachus had presented an interventionist and opinionated narrator for the mythological tales told in the Aitia,13 no precedent for Catullus 68's combination of "autobiographical" narrative and mythological exempla, nor for 76's agonized internal dialogue can be found in Hellenistic poetry. Likewise, the poet's complete subjection (servitium amoris) to a single mistress (domina, era) is unprecedented in ancient Greek poetry of any era.14 Indeed, as I (1994) have argued before, Catullus represents the beginning of that uniquely interiorized voice that I term lyric consciousness and of which erotic elegy can be seen as a subgenre. What Hellenistic literature offered at the end of the first century B.C.E. was not a model to be slavishly copied but an alternative value system to the Roman Republic's traditional mos maiorum (Gutzwiller and Michelini 1991: 75).15 Catullus and the elegists would exploit this resource, but from the unique subject positions offered by a Roman ideological system in the process of collapse and restructuration (Fredrick 1997: 179; Albrecht 1997: 751). The themes and dramatic situations exploited in the elegist's relation to his domina may have already been explored by the figure of comedy's iuvenis in his subjection to the meretrix and his conflict with the senex. Nonetheless, the unique position of the first-person speaking subject in elegy, its constitution as the site of contradiction and aporia, of temporal complexity and personal depth, remain unprecedented. What this book wants to explore is not the thematics or plot of elegy--which differs substantially from comedy with its emphasis on final reconciliation and consummation (Cutolo 1995: 93)--but the historical conditions of possibility for elegy's specific instantiation of what I (1994) call lyric consciousness.

After this second chapter, in the next four we examine the subgenre's full flowering in Tibullus and the early works of Propertius and Ovid and their elaboration of the contradictory and schizoid discourse that lies at the heart of erotic elegy. The final chapters examine the afterlife of Roman erotic elegy in Propertius's book 4 and Ovid's exilic poetry. In each case, we ask what is the nature of the elegiac subject projected in a given collection and how can we understand its relation to history in both its immediate and broadest sense. In short, we shall pose the question of what is the relation of elegy and its subject to the protean nature of the Real.

Before beginning our detailed theoretical argument, it is helpful to introduce three terms originating from Lacanian psychoanalysis but now used by a range of critics. They are the Imaginary, the Symbolic, and the Real. These terms are conventionally capitalized in English usage to indicate that they are being used in a specialized sense, although Lacan himself never followed this practice. The Imaginary is a term derived from the Freudian concept of the Imago. Although there are many subtleties to Lacan's theorization of it, on the most basic level it refers to the image of ourselves that we project upon the world. The Symbolic in contrast is the world of rules and codes. It includes language and all other shared semiotic systems. It is based on the concept of langue in Saussurean linguistics and represents the shared communal grid of intelligibility that defines a community. We are recognized as subjects by others only through assuming a position in the Symbolic. That position may correspond either more or less well to the vision we have formed of ourselves in the Imaginary. Poetry as a linguistic practice takes place in the Symbolic, but clearly--especially in a genre such as elegy--it always includes substantial material from the Imaginary. The Real is that which falls outside of either of the two preceding categories. It is not "reality" because it is precisely what escapes linguistic expression and Imaginary appropriation. Thus the Real cannot be the object of conscious experience. It is important to note that the Imaginary, the Symbolic, and the Real do not denote mutually exclusive realms but intepenetrating registers of existence. Our Imaginary projections are bathed in the codes and norms of the Symbolic, and the Symbolic offers rules, codes, and forms for processing the work of the Imaginary. The Real represents not so much a world outside these two as their necessary limitations.

The advantages of this triad are many. It allows us to theorize the relation of self to community without ever permitting either one to collapse into a simple opposition to the other, and without rendering one term a mere secondary reflection of the other. This triadic structure itself cannot be reduced to an essentialism, because each of the terms always relativizes and recontextualizes the others. There is no one Imaginary, no one Symbolic, but only specific examples that relate to one another in different ways. There is even no one Real. The Real marks the limits of any self-projection and any communal system of norms and codes. In doing so, it also marks the space of historical change, the space where any given ideological system or personal projection comes up against its own finitude and hence the necessity of change.

What follows in the first section is a detailed theoretical argument for the advantages of this intellectual framework when confronting problems of literary history. For those who find such abstract discussions distasteful or opaque, they may leave the theoretical subtleties aside and move directly to the more historically grounded argument that begins in the second section. There the case is made that the changes taking place in the Roman Real that led to the collapse of the republic created a crisis in the Symbolic that also led to the emergence of the subject position we recognize as that of the erotic elegists. The thumbnail definitions just offered are sufficient to grasp the arguments found there and later in the book, although a full understanding of the power of this intellectual framework and of the reasons for using it can only be had by engaging the theoretical issues on their own terms.

When gravitas and levitas, excess and deprivation, equality and status distinctions cease to abut one another, a chasm opens, a No Man's Land out of which the monster and the gladiator arise. They are the mirrors of the absolute, inhabiting the territory at, and outside, the limits within which a compensatory system could be maintained, the territory in which the categories employed by Roman culture to define and create itself are threatened with dissolution.
     --Barton 1993: 187
History means that there must be some unresolved traumatic exclusion which pushes the process forward.
     --Zizek in Hanlon 2001: 16

To begin the labor of solving the riddle of the history of elegy, we could do worse than to look to Jameson's concept of "the semantic conditions of possibility" (1981: 57), which argues that a given text's historicity is best revealed, not by inserting it

into a genetic process in which it is understood as emerging from this or that prior moment of form or style; nor [by] "extrinsically" [relating it] to some ground or context which is at least initially given as something lying beyond it . . . [but through] the hypothetical reconstruction of the materials which had to be given in advance in order for that particular text to be produced in its unique historic specificity. (1981: 57)

This formulation is important. It provides a way of talking about the relation of literature to history that avoids the two chief temptations awaiting any such endeavor: the reduction of literature to a mere reflection of external events; and the invocation of an evolutionist teleology16 in which the a priori logic of the unfolding of form denies the openness of time (Morson and Emerson 1990: 47).

In this latter case, history is reduced to a logical pattern that effectively eliminates real historical change. An example of this kind of argumentation is the contention that elegy ends with Ovid's Amores because the genre had then reached a point of formal exhaustion allowing no possibility of further development.17 Yet, such an explanation is not only denied by the facts--many genres continue to write variations on the same basic themes for hundreds of years without losing their vitality (e.g., epic or the love sonnet)--but it is also fundamentally ahistorical. It views the possibilities of the form as given from the outset. The evolutionist view sees genre as an ideal form whose relation to time is closed because its end is contained in its beginning (Cohen 1986b: 207; Jauss 1986: 42-49; Frow 1986: 292-93; Morson and Emerson 1990: 292-93).18

By contrast, the logical extension of the naive reflectionist model is the very negation of the literary. In the past, it has led to the crudest excesses of the Stalinist concept of the material base's relationship to the cultural superstructure and, in turn, to that concept's aesthetic corollary, socialist realism. More recently, this sort of reductive model has led to the censure of both the homoerotic photography of Robert Mapplethorpe and the constructivist sculpture of Richard Serra, each of whom has been criticized in the name of art's ideal correspondence to a "nature" that is at once culturally normative and untouched by the artifice of abstraction. In both cases, there is a belief that art somehow represents a world that is at once simply "out there" and yet a reflection of the viewer's understanding of and desires for that world. In the case that concerns us, this view is most fruitfully represented by Gordon Williams's (1968: 540) claim that the elegiac genre finds its termination with Ovid's Amores because the lex Iulia of 18 B.C.E. had made adultery a legally prosecutable offense, thus limiting the scope of permissible erotic intrigue to dalliances with meretrices as opposed to upper class matronae. Such a claim, however, is not only reductive, but it also ignores the ten-to-twenty year gap between the passage of the law and the final edition of the Amores, and it assumes against much evidence that the dominae or "mistresses" of the other elegists were both real women and married.19 The task, however, is not to reduce the text to something exterior to it, to see it as a reflection of a reality that is in some sense primary, but to understand the relation of the discursive form to historical change.

Jameson's concept of the "semantic conditions of possibility" avoids these two extremes of a formalist teleology or a historical reductionism, allowing us to establish "the relative autonomy" of literature, to borrow Althusser's carefully balanced phrase (Althusser 1971a: 120-36; Jameson 1981: 23-46; Frow 1986: 110; Kang 1995: 11).20 Through this concept, the text is tied to specific conditions of possibility, but it is never reducible to the mere expression of those conditions. It both retains a real connection to the vital mess of the world in which it is produced and consumed, a connection that cannot be reduced to the mere unfolding of formalist categories, and it remains irreducible to a theory of correspondence that inevitably is beholden, in the last analysis, to an ideology of the culturally normative. The concept of the conditions of possibility is not a causal theory, which reduces the text to some prior moment, but a way of situating a given literary form that recognizes its temporal boundedness.

What more precisely, then, is the nature of the relationship of text to world, if literature--and, more specifically, elegy--never simply reflects a reality that is somehow external to it (Bakhtin/Medvedev 1985: 18; Frow 1986: 123)? Indeed, the whole concept of externality in this context is highly problematic,21 for, as much modernist and postmodernist philosophy, theory, and literary criticism have shown, language not only makes literature and communication possible but also constructs human reality as an intelligible, but necessarily ideological, entity.22 Reality thus comes to us as always already socially constructed by the intersubjective world of language and its affective connotations (Parkhurst 1995: 45). In Althusser's formulation, "what is 'experienced' is always shot through with ideology" (1996b: 76). There is, thus, no neutral realm of fact or experience against which our various ideological constructions can be confidently measured,23 no valuation that is not always already an evaluation (Morson and Emerson 1990: 134; Gardiner 1992: 85). This is a realization that, while banal in the modern languages, has only recently gained currency in the study of the ancient world and one whose implications have yet to be fully realized (Wyke 1989: 27; Joshel 1992: 12).

Another way of putting this is that fiction--in its original sense of something made or fashioned, fictum--makes the reality that we experience possible; for linguistic fictions ground the world's intelligibility (Eco 1976: 61-66). This observation, however, leads traditional literary study into a conundrum. If reality's existence is predicated on fictions, then on what ontological ground might it be said to be reflected by them? A mirror cannot reflect itself. This constructivist position, however, can be easily misunderstood, and we must make a distinction, if we are not to fall into the postmodernist trap of saying all is discourse (Ragland-Sullivan 1986: 98, 194, 255). Let us be clear. I am not here advocating the position that there is nothing outside the text. Rather I am contending that there is a split in our world. On the one hand, there are those necessary fictions that we have termed "reality," while, on the other, there is the world we posit beyond any historically or culturally specific construction of reality. What Nabokov called the mysterious X factor, Lacan labels the Real.24

Reality is a coherent system of meanings that allows us to function. It is a picture or an understanding of the world and, as such, is finite and historicizeable. Different societies and different groups within societies produce different symbolic systems, different realities (Cle ´ment 1975: 16; Ragland-Sullivan 1986: 183, 305). In Bloch's memorable aphorism, "Not all people exist in the same Now" (1977: 22). This is precisely the sense of the term "reality" that is used when we speak of friends or colleagues "not being in touch with reality"--that is, they do not share the same set of symbolic norms and interpretations that we do.25 Anyone who has sat through a contentious university tenure and promotion meeting should understand exactly what this means. The world "they" live in is not the same as "ours." It does not mean that they have less access to the world in all its brute facticity, or that matter impinges on them less than on us. This mute realm, beyond meaning, is the Real, the silent domain that supports our various realities, the multiple nows in which we all exist (Lacan 1975: 50; Zizek 1991: 29-32).

Language is an intervention in the Real, not its reflection; nor does the efficacy of its intervention necessarily correlate with its capacity to represent the nature of things. Language is at least as performative as it is constative, as productive as it is descriptive.26 A few quick examples make this clearer. Chinese medicine is based on a theory of entities and energies that Western thought does not recognize as existing, yet it produces visible, predictable effects within the world. Thus, the discourse of Chinese medicine produces a representational "reality," but its ability to have effects in the Real are not necessarily related to its creation of that reality. Likewise, it is now commonly acknowledged that the Romans understood sexual life in very different terms than we do. Words like cinaedus and homosexual, though covering some similar behaviors, have unique and unrepeatable resonances, and produce different effects in social life. There is, for example, very little evidence of a movement for "cinaedic" rights in Rome. Even neutral terms such as those referring to color in Latin are not directly translatable into English. Nor is there any reason to assume that the experience of canus, referring both to the gleaming white of the morning frost (Horace, Odes 1.4.4) and the gray of old age (Catullus 108.1), was identical with that of our white, gray, or something halfway in between (see Eco 1985). These terms were nonetheless very much a part of Roman reality. By the same token, the equations and mathematical formulae that constitute contemporary physics are themselves irreducible to the terms of conventional representation (e.g., a photon is both a wave and a particle), yet the effectiveness of these equations is not therefore called into question (Beer 1993).27 Indeed, the efficacy of modern physics is founded on its move toward a pure mathematicization, that is to say, on its retreat from both the ad hoc contingencies of the Real and the conventions of representation (Julien 1990: 146). Language, thus, can have effects in the Real, without necessarily reflecting it. This does not mean that all linguistic fictions are equally effective. In science, as Lacan notes, a concept is only retained if it functions, that is, if it traces its way in the Real (Lacan 1975: 183). But language's approach to the Real is always asymptotic at best, even when it is executed through such sophisticated apparatuses as modern science, and even when it has a revolutionary and enduring impact, such as Lacan's example of Newton's Laws (Jameson 1988: 104, 108).28

Language and other signifying systems construct a coherent network of representations, a reality, that allows us to function within the world, but does not present the world in its prelinguistic purity.29 All that exists outside the world of signs, and the categories of understanding they represent, can only be conceived of as a formless void of "inert presence" (Zizek 1989: 170), the literally unspeakable--what Jameson calls History, Lacan the Real, and Kant the Ding an sich30 (Ragland-Sullivan 1986: 190; Jameson 1988: 104; Zizek 1989: 132). The Real is not the system of narratives, conceptual structures, and affective investments we make in the world, but rather those systems' necessary limitations, the point at which they come up short before the world (Zizek 1989: 208-9; Jameson 1988: 107). The Real is the experience of suffering, the negation of our desire--which includes our desire for knowledge (Foucault 1977a: 27-28)--as we run up against a realm of necessity for which we cannot coherently account.31 The Real then is the beyond of the Symbolic, Lacan's term for the intersub-jective realm of language and knowledge (Julien 1990: 143; Jameson 1988: 106), and it is made manifest by the incoherencies and double binds created within the Symbolic (Zizek 1989: 21; Julien 1990: 173). It is simultaneously in and beyond the Symbolic, inscribed within it as the Symbolic's external limitation.32 As Joan Copjec writes, "the point where the real makes itself felt in the symbolic [is] the point at which the symbolic visibly fails to disambiguate itself" (1994: 176-77, emphasis in original). History, from the perspective of the Real, is not a narrative of events but a series of Symbolic systems, progressively collapsing before their own ineffable but changing beyond (Jameson 1981: 82; Zizek in Hanlon 2001: 15-16). As Ellie Ragland-Sullivan writes, "Symbolic systems build up into a state of near perfection and then decompose, as the Real and the Imaginary continually shift Symbolic order explanations and informations" (1986: 230-31). We return to the topic of the Imaginary and its role in the constitution of self-identity later in this chapter.

First, however, we must tarry a little longer with the Real, since it is the Real, as that which is unassimilable to the structures of explanation available to any given cultural formation, that makes the historical succession of cultural, linguistic, and political forms and institutions thinkable as something other than an inexplicable series of random transformations.33 For the Real marks the point at which the Symbolic meets its own systemic negation (Copjec 1994: 9, 121), its principle of finitude or limit. This moment of negation is necessary to any meaningful concept of historical change, because it is precisely this moment that figures the possibility of otherness within the reigning positive system.34 Without the negation of the Real, without this conception of the beyond of the Symbolic, the norms and systems of meaning that constitute the latter would absolutize themselves in a way that would allow no room to conceive of the radically other (Jameson 1981: 90-91; 1991: 405-6; Copjec 1994: 17, 23-24). As Derrida (1993: 68) has written, the concept of the future, of the yet-to-come (a-venir), depends on an opening within the Symbolic that cannot be reduced to the categories of either knowledge or ignorance, but must be seen as a systematic heterogeneity, an absolute otherness that exceeds the reigning positive system.

Here, however, we run into one of the cruxes of both postmodern historical thought and the new historicism, which trace their origins to the work of Michel Foucault: for this gesture of the reduction of the Symbolic to a pure positivity removed from the Real is recognizable in the latter's dictum that a discursive formation contains only one e ´piste ´me ` (Foucault 1966: 179; Laqueur 1990: 21; Macey 1993: 163; Flynn 1994: 33; Kittler 1999: 117).35 The problem here is simple. Such a system of historical investigation, as Foucault implicitly concedes,36 while potentially quite powerful in its descriptive mode, is inherently unable to explain how one discursive formation gives way to the next, since each moment is conceived of as a totalizing unity (Dreyfus and Rabinow 1982: 27; Macey 1993: 175). As Eve Bannet observes, Foucauldian "archeology seems to freeze history, describing discursive formations and seeking general rules which are valid for all points of time but neglecting the temporal series within these formations and reducing chronology to the point of rupture when one discursive formation is substituted for another" (1989: 108-9). In short, Foucault's analytic renders the conceptualization of historical change impossible.

This same methodological deficiency afflicts Foucault's followers as well. Thus Paul Veyne in what is in many ways a masterful book on elegy, completely avoids the question of how the genre arose or why it disappeared, in effect severing it from any meaningful relationship with Latin literary history or with the social world in which it was formed (Miller 1998a; Kennedy 1993: 91-100; Veyne 1988: 31-32).37 From Foucault's perspective, this isolation is the price we have to pay for preserving historical specificity, for not falling into a teleological vision of history that annihilates difference by sublimating it into an ideological master narrative of one sort or another (e.g., the decline and fall of the Roman Empire, the rise of the West, the birth of the individual). The Lacanian position, by founding the possibility of the Symbolic in its own inherent limitation, its severance and hence inability to account for the Real (Eagleton 1983: 168), makes the negation of the Symbolic a constitutive part of any given discursive formation. Thus where the Foucauldian position must homogenize a given period's e ´piste ´me ` in the name of historical difference (Foucault 1966: 77, 103, 171; Bannet 1989: 157; Sedgwick 1990: 46-47; Macey 1993: 163), and so cut that e ´piste ´me ` off from any narrative that would explain its position in the larger historical sequence (Kremer-Marietti 1985: 49), a Lacanian stance sees the Law of the Symbolic as always opposed to and constituted by its own principle of generation and succession, a principle that escapes the Symbolic itself (Copjec 1994: 122).38 A truly historical event, therefore, would be one that necessarily exposes the tension between the Symbolic and the Real and can never be wholly accounted for except in terms of that tension.

The collapse of the Roman Republic is a perfect illustration of this thesis: for it was not understandable in terms of the ideological categories of the traditional constitution, but only as a traumatic event that revealed the limits of those terms' applicability to the world as it existed. What could mos maiorum (the way of the ancestors), libertas (aristocratic free speech), and res publica (commonwealth) mean at a time of dictatorship and civil war that was not simultaneously the inverse of their normal signification? A good example of this ideological confusion can be seen in Cicero's letters to Atticus in the spring of 44 B.C.E. On the one hand, in the wake of Caesar's tyrannicide, Cicero feels obligated to reenter public life, but on the other hand he can no longer understand how he might do so. Thus on April 12, he writes:

I am able to find no way to enter politics. For nothing is so hors sens as when the tyrannicides are lauded to the skies and the deeds of the tyrant defended. But you see the consuls, you see the remaining magistriates, if magistrates they be, you see the lassitude of the noble orders.

On May 11, however, he does an about face and scolds Atticus, "Epicuri mentionem facis et audes dicere mhpoliteuesqai?" (Do you bring up Epicurus and dare to tell me not to enter politics; 14.20.5). The change in tone in striking. Nonetheless, in spite of his momentary optimism ("prorsus <melius> ibat res" [at the moment things have been going better], 14.20.4),39 Cicero has in fact still found no way to make the categories of the old world function in the new. His new attempts to save the republic meet with no greater success than did the old. Thus, we may indeed admire the courage of the later Philippics, but they hardly offer a way forward, a realistic political program, or an effective conceptualization of the systemic crisis gripping the Roman state.

Rather, it is precisely in Cicero's confusion, and that of others, that we see the index of real historical change, the sign of an event that exceeds the reach of the Symbolic categories available to account for it.40 Such aporias are a symptom of the eruption of the Real and therefore the ground of effective history. In fact this tension between the Symbolic and Real guarantees for us the possibility of a nonsubjective historiography and offers the point of purchase for a historical practice that is not merely a repetition of founding assumptions and ideological truisms. "The import of the event's divulgence of the relationship between the real and the symbolic . . . proceeds from the manner in which it expedites a nonsubjective historiography . . . Produced in the liminal space between them, the event qua history is therefore the effect of their differential relation" (Myers 2001: 42).

Difference, for Lacan, then, as opposed to Foucault, is not an external relation between regimes of power and knowledge but is internal to them (Copjec 1994: 9-10, 60). In this sense, the generative principle of any given discursive or ideological formation, its relation to that which exceeds it, also contains in nuce that formation's negation, the moment that escapes the dominant ideology's grasp and so makes change possible.41 Every form of hegemony is always haunted by the moment of repression through which it is constituted and is thus always already internally split (Derrida 1993: 69). This internal fissure is precisely the point at which we witness the emergence of the Real. It is the space of difference that this fissure implies, the space between the Symbolic and the Real, that prevents Symbolic totalization and thus makes history possible (Zizek 1989: 135).

The relation between Foucault and Lacan, however, is more complex than this brief sketch might lead us to believe. We cannot simply demonize Foucault and exonerate Lacan. On the one hand, while Lacan's concept of the Real, in its function as the negation of the Symbolic, makes historical explanation possible, he himself is often accused of being ahistorical (Butler 1990: 29, 55, 76; Irigaray 1977c: 97). This may be an oversimplification, but he is vulnerable to the charge in part because he never treats historical verities in any depth and tends to speak as if the topics he discusses exist in an eternal present. Hence, at times, he seems to treat the current formation of the Symbolic as if it were the Symbolic tout court. Even though he makes it clear that the Symbolic and the subject's insertion into it are open to historical modification,42 he never speaks in any detail about how that modification happens and the nature of its consequences. Thus his description of the Symbolic, particularly in relation to gender, can appear to be a prescription (Ragland-Sullivan 1986: 277). This is the gist of Luce Irigaray's feminist critique of her former teacher: not that his description of woman's exclusion from the Symbolic, as constituted by patriarchal reason, is inaccurate, but rather that his failure to posit an alternative to this situation implies that it is unalterable and perhaps even desirable (Irigaray 1977c: 92, 99; 1977d: 205; Weed 1994: 87, 99, 100-02).

Lacan, thus, may make historicization possible, but he does not himself historicize. Foucault, on the other hand, does historicize, but his suppression of the moment of the Real, of any effective negation of the Symbolic, makes the possibility of historicization inexplicable as anything more than the careful description of an arbitrary moment in a random series of events. What is needed is not an uncritical acceptance of either position, but a careful fusion of Foucault's practice of rigorous, nonteleological historicization, which emphasizes the radical difference between historically distinct modes of constituting the subject and power, with Lacan's ability to explain the succession of those modes. Such an amalgamation would result in the abandonment of any sort of purist Foucauldian or Lacanian position in favor of that ultimately more nuanced understanding that Jameson's concept of the "semantic conditions of possibility" demands.

Thus, the recognition that intelligible reality is only possible within a system of meaning or signification (Dowling 1984: 127) does not mean that language and fiction (i.e., the made) are all there is, or that all interpretive and linguistic positions are equally tenable (Lacan 1973: 278; Ragland-Sullivan 1986: 174). Rather, evidence of the inherent limitation of any given Symbolic regime can be seen in the fact that all literary formations are not equally possible at all times. Literary forms like people are finite. The possibility of specific fictions at specific times is historically limited (Frow 1986: 110, 114). History reveals itself precisely through this limitation, through the moment of negation inherent in any Symbolic form. Rimbaud would not only be foreign but unintelligible to a first-century Roman. The public context that made The Countess of Pembroke's Arcadia one of the most widely read and best-selling books in the English language throughout the seventeenth and eighteenth centuries has disappeared--few outside a trained minority find Sidney's great pastoral romance even readable. The world in which it could be produced and popularly consumed has vanished. The discursive formation that constituted its semantic conditions of possibility is no longer available, not because the structure of the sign has changed, but because the world--material, ideological, political, cultural--in which those utterances were immediately meaningful no longer exists (Eagleton 1976: 56; Jameson 1981: 148; Zizek 1991: 48; Moxey 1991: 987-88). None of this is to say that the value of the Arcadia was ever predicated on its ability to reflect the world that made it possible, anymore than that Vergil's Eclogues reflect the Augustan Real, but merely that the Arcadia and the Eclogues are not equally possible in all worlds. In addition, the fact that pastoral romance is not possible in our world, except as a curiosity, tells us that something crucial has changed in the period separating Sidney's time from our own. Likewise, the fact that love elegy was a viable genre in 30 B.C.E. but neither in 30 C.E. nor 60 B.C.E. tells us that the realm of possible meanings available to a Roman poet had changed in some fundamental fashion in the intervening period (Todorov 1984: 80; Cohen 1986a: 229-30; Morson and Emerson 1990: 277).

Literary history thus has a complex and overdetermined relation to the extraliterary. So complicated and ambivalent is this relationship that the binary opposition of literary and extraliterary that supports its conceptualization is untenable. This is true not only on the pragmatic level--elegy and the Arcadia are historically specific phenomena that neither necessarily reflect their worlds nor fail to transcend them as objects of consumption--but also on the more purely theoretical level of internal coherence. The opposition of literary and extraliterary, which subtends any such historical project, is based on a logic of thematics or reflection that rests ultimately on a fundamentally dualist ontology (inside-outside, literary-nonliterary), which in most accounts is never explicitly justified.43 At the same time, this very ontological opposition, while enabling literature to be conceived of as reflecting a reality separate from it, also insures that the moment of representation is infinitely deferred, since it is the fundamental and unbridgeable difference between the literary and the extraliterary that makes this logic of representation possible. Such self-deconstructing oppositions cannot be maintained. In point of fact, inasmuch as reality, as opposed to the unreflectable Real, is linguistically constructed, no difference in primal substance exists on which such a distinction can rest. Jameson's concept of the semantic conditions of possibility avoids the theoretical impasse of the traditional ontology of reflection. By determining what made elegy possible at a given moment in Roman history and when that possibility was closed off, the history of the subgenre and its multiform relation both to other cultural forms and to the profound transformations of the Roman Real that are taking place at this time can begin to be elucidated (Cohen 1986b: 213, 216). Those changes themselves are signaled by the progressive collapse of Roman republican ideology and the vast network of formal and informal institutions set up to maintain and be maintained by it, as well as through the subsequent rise and consolidation of the imperial state with its ideological state apparatuses. Neither of these ideological formations, the republican or the imperial, nor their various and often antagonistic subformations are capable of fully representing these changes. They can only produce narratives that respond to these shifts through the deployment of strategies of appropriation and containment (Jameson 1981: 52-53). Such responses are always necessarily more concerned with power and positioning than with pure representation. They are ideological and situated.

The challenge then is not only to provide a new, more accurate positive description of discursive practices but to trace within the heart of those practices their own systemic negations: to locate the emergence of the Real as a given form's simultaneous foundation and principle of finitude. Such moments of emergence will not be found within that form's structure of representation but rather in those moments of aporia and nonsense that are both constitutionally necessary and unrepresentable. Yet unlike for American deconstruction--in the hands of critics such as J. Hillis Miller (1981)--the task here is not simply to find and celebrate the undecideable but to map precisely how such moments are structured in different ways and at different places and times in the Symbolic system of a given society or societies. Different articulations of the Symbolic and its relation to the Real will produce different forms of aporia and different subject positions relative to those forms. The result of this investigation will be not the description of a unified positive system--which must by nature absolutize itself in an eternal present--but a detailed examination of the way such systems are constituted by that which they must exclude and of how this game of constitution and exclusion produces a shifting and unstable discursive field in which different Symbolic formations jockey for strategic advantage in relation both to one another and to an ever changing Real that by nature excedes their grasp.

What marks the Ciceronian age is a persistent tension between the old categories of political behavior and the onrush of events that appeared to outstrip them.
     --Gruen 1995: 49
Sed cum Lucius Domitius consulatus candidatus palam minaretur consulum se effecturum quod praetor nequisset adempturumque ei exercitus, Crassum Pompeiumque in urbem provinciae suae Lucam extractos conpulit [Caesar], ut detrudendi Domitii
causa consulatum alterum peterent, perfecitque per utrumque, ut in quinquennium sibi imperium prorogaretur. Qua fiducia ad legiones, quas a re publica acceperat, alias privato sumptu addidit, unam etiam ex Transalpinis conscriptam, vocabulo quoque Gallico--Alauda enim appellabatur--quam disciplina cultuque Romano institutam et ornatam postea universam civitate donavit.44
     --Suetonius, Divus Iulius 24

The ideological nature of the responses to the fundamental shifts in the nature of the Roman Real is evident throughout the period that concerns us. As Erich Gruen in his encyclopedic Last Generation of the Roman Republic has demonstrated, until the moment that Caesar crossed the Rubicon in 49 B.C.E., the political class in Rome and the populus as well thought not in terms of impending revolution but of the traditional republican politics of competitive elites. Likewise as the passages from Cicero's letters cited earlier(14.6.2, 14.20.4-5 ad Atticum) as well as Augustus's attempts to maintain republican governmental forms even after assuming effective power (Dio Cassius 53.12.1-3, 53.16.1) show, this tension between the categories of political experience and events on the ground remained throughout the civil wars and the rise of the principate. Caesar and Pompey were less self-conscious revolutionaries in the Leninist or Jacobin mode than faction leaders seeking dominance within a system they hoped to lead, not destroy (Gruen 1995: xix).

Nonetheless, Gruen's account, while invaluable and heroic in scope, is flawed. It accepts the statements of the political actors involved at face value, even as it acknowledges their being out of phase with events on the ground. It thus abstemiously refuses to draw any larger cultural or political consequences from Symbolic patterns manifest in both those statements and their specific repressions. Indeed, in his all but Humean distaste for concepts of causality, Gruen seems to claim that the collapse of the republican constitution and the outbreak of civil war were mere accidents without antecedent causes or necessary relations, one to the other. Thus, at one point, he argues: "Transformation of the state into a monarchical regime can be laid to the charge of a devastating civil war, rather than to the putative disintegration of institutions and morale in the previous decades" (1995: xx). A short time later, he asserts, "Civil war need not be read as a token of the Republic's collapse" (1995: 5). Such statements, if not directly contradictory, border on a reduction of history to mere chronology. This positivist refusal to integrate discrete facts into larger structures of intelligibility not only smacks of what Vernant (1990: 271) refers to as the "index card" method of classical scholarship but is also unable to draw significant connections between cultural and political phenomena, and thus between statements like the following by Cicero and the emergence of the elegists:

Quibus autem talis nulla sit causa, si despicere se dicant ea quae plerique mirentur, imperia et magistratus, iis non modo non laudi, verum etiam vitio dandum puto; quorum iudicium in eo quod gloriam contemnant et pro nihilo putent, difficile factu est non probare; sed videntur labores et molestias, tum offensionum et repulsarum quasi quamdam ignominiam timere et infamiam. Sunt enim qui in rebus contrariis parum sibi constent, voluptatem severissime contemnant, in dolore sint molliores, gloriam neglegant, frangantur infamia, atque ea quidem non satis constanter. (De Officiis 1.21.71)
Those, however, for whom there is no cause (such as ill health) to say they despise those things which many others admire, power and office, I think not only ought not to be praised but be given up to vice. It is not a difficult thing to approve the judgment of such people in this matter, since they look down on glory and think it worth nothing, but in fact they seem to fear hard work and difficulties, as well as the shame and dishonor of defeats and misfortunes. For there are those who are inconsistent in contrary affairs: they severely condemn pleasure and are rather effeminate in pain, they disregard glory but are crushed by shame, and even in this they are insufficiently consistent.

The picture Cicero paints here of a seemingly principled but nonetheless effeminate (mollis) withdrawal from public life is almost the exact image that the elegists project for themselves. The coincidence is striking, and not least because the characterization might just as easily be applied to Cicero himself. Thus, as we saw Cicero in 44 B.C.E. is writing letters to Atticus in which he says he can neither withdraw nor abstain from politics, even as he at that moment is composing a theoretical treatise that condemns this very inconstancy. In the same period, we also begin seeing the emergence of a class of poets as well as other members of the equestrian order who publicly renounce politics and question the traditional hegemony of the senate and its ruling ideology (cf. Cicero, Epistulae ad Atticum 1.17.8-9). Indeed, the passage from De Officiis is almost a paraphrase of Propertius's position in 1.6 and Tibullus's in 1.1. At the same time, we see the republic itself crumble and the emergence of a fundamentally new kind of state that will slowly elaborate its own ideological norms and expectations. If we accept Gruen's hard positivist view, this is a mere coincidence because at no point do the actors themselves self-consciously speak of revolution but always of traditional values, their transgression and potential restoration. Yet, it is precisely the inability of the existing Symbolic categories to comprehend the profound historical changes taking place that reveals the limits of those categories and the necessary and traumatic emergence of the Real.

Indeed, as the ten years immediately preceding the outbreak of open warfare clearly show, the collapse of the republic was anything but an accident. There was both a marked increase in the recorded cases of urban violence, electoral bribery, political warfare in the courts, and extraconstitutional political maneuvering by not only the triumvirs Pompey, Caesar, and Crassus, but also Clodius, Milo, and the reactionary Catonians.45 In each case, however, these events were presented, and presumably interpreted by the actors concerned, not as breeches of traditional republican legal principles and political norms but as actions designed to defend them (Gruen 1995: 119). Likewise, the response from the populus in the period was not rebellion but the continued return of the same small group of oligarchic families to office in the consular elections (Gruen 1995: ix, 158), while the senate continued to search for piecemeal and ad hoc methods of legislative reform (Gruen 1995: 506-7).

No one sought revolution, everyone restoration (Gruen 1995: xx, 153). Rather than the collapse of republican legal principles, what this period sees is their gradual hardening as the traditional governing norms of the mos maiorum became the object of increasingly bitter dispute. Indeed, the flurry of legislation witnessed by the last decade of the republic is not only a sign of the continuing viability of republican institutions but also a symptom of their crisis. What was once a flexible elite consensus now had to be defined contentiously and litigiously in exact legal terms (Gruen 1995: 258-59). The crisis reveals itself precisely as one of the impossibility, yet felt necessity, of consistent cultural meaning in a time of crisis. As Wallace-Hadrill argues:

A leitmotif of the late Roman Republic is on the one hand the crucial importance of following tradition, the mos maiorum, if Rome is to survive and succeed, and on the other an awareness that tradition is slipping away, and needs to be painstakingly reconstructed, both at a theoretical level by rediscovery of what lost tradition was, and at a political level by its re-establishment and reimposition. (1997: 13).

What Gruen demonstrates, then, is not so much the accidental nature of the collapse of the republic as the inability of republican ideology to conceptualize that collapse. It is precisely the crisis of Symbolic norms in relation to a changing Real that this period reveals, as Caesar, even in crossing the Rubicon, is not only able to present himself as the savior of the republic but also able to convince others to accept that claim (Bellum Civile 1.85, 3.57, 3.90-91; Cicero, Pro Ligario 18). His belief in, or cynicism about, this ideological position is not only ultimately unknowable but also less important than the fact that all sides deployed the same narrative structure. It is precisely the aporias of that structure, its progressive inability to account for the actions of those deploying it, as revealed by their mutually contradictory claims, that provides the truest window to the historical changes taking place. What is ultimately revealed is not so much "what actually happened" as the progressive foundering of the Romans' own ability to explain the changes that were happening.

The same pattern of misrecognition continues into the early imperial period. The Augustan program of moral revival and religious reconstruction was consistently presented in terms of a return to the virtues of the past, the mos maiorum, and the restoration of the republic (Augustus, Res Gestae 8, 19-21, 34), even as it laid the ideological groundwork for consolidating what was to be the most sweeping transformation of the Roman state since the expulsion of the Etruscan kings.46 At the same time, in spite of its self-proclaimed reactionary nature, the regime did not scruple to appropriate Egyptian concepts of sacred kingship and other ideas foreign to traditional republican political thought, where it deemed them useful (Koenen 1976: 128). Augustan imperial ideology was in fact less a coherent, theoretical edifice than an ad hoc construction designed to meet specific needs (Koenen 1977: 132; Kennedy 1993: 36-37).47 It was more concerned with power and stability than ideological purity or revolution (Ste. Croix 1981: 360).

The republican counterprogram, to the extent that such a thing existed (Tacitus, Annales 1.2), offered nothing more than the myth of a return to the golden age of senatorial rule, itself based on an oligarchical system that had ceased to be truly functional since the time of the Gracchi's proposed land reforms and subsequent murder one hundred years before.48 Despite the momentary stabilization brought about by the Sullan dictatorship and its reconstitution of the senatorial elite, the existing constitutional settlement had deteriorated to the point where civil war was a constant threat, elections had to be canceled in the face of riots, and the legions were used against the populace to quash uprisings for debt relief (Brunt 1971a: 127-32; Ste. Croix 1981: 354; Nippel 1995: 77-84).49 The professionalization of the army, which had begun with Marius, eroded the traditional ideology that bound the soldiers to a republic in which they as landholders had a definite stake. Instead, they increasingly professed loyalty to the individual commanders, the sole authorities who had an interest in insuring that they received their just rewards, generally an allotment of land supplied by despoiling their political enemies.50 Again Gruen offers a cautionary note, arguing that the Marian reforms merely acknowledged a state of affairs that had become increasingly the rule rather than the exception, and that many military leaders failed to establish strong bonds with their troops. This is certainly true, but it neither denies the nature of the change nor, in the cases of imperatores such as Caesar and Pompey, who were given commands of extraordinary length and power, that they were able as a consequence to establish power bases that were essentially beyond senatorial or popular control as demonstrated by Suetonius above (Gruen 1995: xvii, 112, 366, 377; see Shelton 1998: 244 n. 8).

In light of the changing military and political situation, the leadership of the senatorial oligarchy became beholden to the generals--Sulla, Marius, Pompey, Caesar, and eventually Octavian and Antony--to crush both their rivals and, if need be, popular resistance (Nippel 1995: 66-67, 83-84). The soldiers' alienation from the traditional instances of state power was in turn augmented by the practice of some of this senatorial elite of driving the families of these soldiers off their lands while they were away on campaign (Sallust, Bellum Jugurthinum 41.8; Brunt 1971b: 551-57; Ste. Croix 1981: 357). The ruling class thus became increasingly fragmented as it was forced to line up behind various warlords, like Caesar and Pompey, and the factions they represented, while constantly seeking to reassert its own dignitas and the values of a competitive elite in an increasingly fractious environment (Gruen 1995: 49, 66; Wallace-Hadrill 1997: 11). At the same time, certain members of the elite, such as Clodius Pulcher, who claimed to represent the interests of the plebs urbana, were engaging in new forms of political organizing that sought to create alternative organs of power through the mobilization of collegia, or guilds of tradesmen, vici, or neighborhood counsels, and the worshipers of nontraditional cults such as those of Isis and Bacchus.51 Thus, at this period in Roman history, power simultaneously became more diffuse as it broke away from its traditional institutional basis, even as it also became more concentrated in the hands of the generals whose troops became the final arbiters of Roman political conflict. It is no wonder that the traditional narratives that legitimated that power became increasingly suspect.

The republican constitution in effect no longer existed. The return to it was a dead letter. The fact was that no ideological alternative as yet existed. This was the breach into which the Augustan settlement would step. But where Augustus's program camouflaged the nature of its intervention in the reorganization of the state's political and cultural apparatuses and its centralization of the powers of military command, beneath the mask of a fictional, although not necessarily insincere, return to the status quo ante (Ste. Croix 1981: 391-92), the minimal republican opposition had no solutions to propose at all (Galinsky 1996: 8). It offered only an illusory nostalgia.

Neither of these narratives, the Augustan or the republican, can be said to provide an accurate reflection of the transformations occurring in the Roman Real, nor was this even their purpose, assuming that such a thing were possible. The picture that we have pieced together here is a synthetic product drawn forth by a variety of scholars from a disparate array of materials--historical, literary, epigraphical, and archaeological--no one of which is to be taken at face value. Indeed, our mosaic offers less a coherent narrative of meaningful events than points to a series of crises in meaning whose discursive functions are symptomatic rather than expository. What does it mean to be a senator, a landholder, a soldier, at the limit, a subject in Rome? Each of these questions points to a crisis that, as we shall see throughout this book, produced conflicting ideological valuations and even violence (Barton 1993: 85, 151-52, 189). These crises in turn are registered as moments of trauma and undecidabilty in the cultural productions of the period. They produce a profound ambiguity that is most satisfying to read and most horrible to live.

For the literary historian, however, it is not a question of taking sides or of determining which valuations to believe and which to condemn but rather of studying the conflicts themselves: when they arise, when they disappear, how they are linked to other moments of conflict and stabilization, and what is the logic of their transformation. Thus, what is of utmost importance is not the content per se of the narratives produced by those who participated in these events, but the moments of ideological contradiction and aporia found therein. These are the truest barometers of the architectonic shifts occurring in the Real, that is, the profound historical movements that are reshaping the world beyond the immediate consciousness and control of its actors and participants, as well as the cultural revolution accompanying them (Jameson 1981: 82-83, 96). These are the moments that reveal the limits of ideology and that mark those junctures where we must look for the emergence of history.

Tant la confusion était grande au cours de ces ann ées; tant il était difficile, au moment où on réévaluait "l'humanité," de déterminer la place que devait occuper la folie; tant il était difficile de la situer dans un espace social qui était en voie de restructuration.52
     --Foucault 1972: 445
The perception of the civil wars and the monarchy as permanently transgressing the discrimina ordinum brought into play a "physics" corresponding to René Girard's "Sacrificial Crisis: a collapse of the sensus communis, a shared system of categorizing and ordering."
     --Barton 1993: 146

In Le souci de soi, Michel Foucault (1984: 105-6, 117) observes that the period of the late republic and early empire witnesses a profound crisis in what it means to be a subject. This crisis manifested itself in part through an increasing anxiety over matters of sexual and personal conduct.53 The literature in and on the period is filled with anxious statements about the decline in the sexual morality of the upper classes, their increasing effeminacy, the emancipation of their women, and their progressive inability to replenish themselves (Riposati 1945: 30; Grimal 1986: 147).54 Thus, as Maria Wyke notes, starting in the middle of the second century B.C.E., when republican Rome was at the height of its power with the end of the Third Punic War, as vast new amounts of wealth flooded into the coffers of the aristocracy, creating greater social inequality and political disruption (Sallust, Bellum Catilinae 9-11), a series of female scare-figures begin to appear in the literature. These are powerful and sexually liberated women who are epitomized by Sallust's portrait of Sempronia (Bellum Catilinae 25) and Cicero's depiction of Clodia in the Pro Caelio. In all cases, as Wyke (1989: 37-40) observes, the portraits of these women are embedded in larger narratives of social decay and corruption and so cannot be seen as providing independent evidence regarding the lives of real Roman women, but rather as symptoms of a crisis in the way the desiring subject constituted itself in relation to traditional republican norms. Such women are not to be found in the Real (which is not to question their "reality") but precisely at the point where the Roman Symbolic reveals its inability to conceptualize such figures as anything other than impossible monsters.55 They are fictions of sexual and ideological aporia.

Indeed, it is precisely their status as textual constructs that allows these women to be seen as the forerunners of the elegists' own socially dubious mistresses: for such figures have become part of Rome's symbolic economy. Elegy's women are not so much revolutionary as symptomatic. The elegists refer to their beloveds as dominae, and hence as the mistresses of slaves, allowing elegy itself to become a tale of male subjection in which the elegiac ego assumes the position of the servus amoris (slave of love), in a world whose political norms have already been turned upside down. This position of dominance is doubly anomalous in that the elegists were all men of equestrian status and hence members of the Roman aristocracy. The incongruity of their position is presumably part of the fun but also serves as vivid testimony to a creeping awareness of the contingency of Symbolic norms in an environment of political disruption and social change. The dominae, however, are if anything more the products of textual construction than their "historical" avatars, for they are in fact lacking in the internal consistency one normally expects from literary characters, changing in manner, physical appearance, and social status from one poem to the next. There is, then, wide disagreement over who these dominae are meant to be: courtesans; married women practicing adultery; wealthy widows; or, as has been suggested in one case, the wife of the poet himself (see chapter 4).56 These women represent less simple identities than complex nexuses of conflicting Symbolic norms.

What they clearly do signify is a further development of the mounting anxiety over the moral and, hence, ideological fabric of the Roman state, in a period of increasing social unrest. Thus, as Wyke notes, the first century B.C.E. sees a "multiplication of discourses about . . . the female," in much the same way as Foucault has demonstrated the proliferation of discourses about sexuality in eighteenth-century Europe, also a period of widespread social reorganization and revolution (Wyke 1989: 41). Moreover, as Foucault notes in his discussion of the rise of "sexuality," that discourse--with its concerns about childhood masturbation, female hysteria, and the categorization of sexual deviants--cannot be seen as reflecting an actually existing state of affairs in Europe at the time. There is no reason, for example, to believe that children masturbated any more or less frequently at that time than they have before or since. Nor can the deployment of this discourse be seen as some sort of attempt to harness sexual power for the productive ends of the emerging bourgeoisie, a form of labor discipline; rather it represents a new articulation of the subject's relation to the body that is part of a complex response to other changes occurring simultaneously within the same material and cultural field. The rise of the discourse of sexuality is therefore a symptom of historical change, not its reflection (Foucault 1976: 137-73).

By the same token, the rise of figures like Clodia and Sempronia in the political and historical prose of the mid-first century, as well as of the elegiac dominae in the poetry of the next generation, is evidence of the rewriting of sexuality and gender's relation to the subject's innermost recesses and to its points of insertion in the networks of social and political power. This is the gist of Thomas Habinek's recent argument that

the elite sectors of Roman society in the late republic and early principate underwent a transition from a sex and gender system based on the principles of honor and shame associated with peasant societies to one more closely resembling that of contemporary urban centers. The lurid accounts of sexual outrages that proliferate in the era are at least in part attempts to identify appropriate limits on behavior in an era of radical change. (1998: 143-44)

In the move from a traditional society based on implicit concepts of honor and shame controlled by a self-regulating and self-perpetuating elite to an international empire in which the claims of traditional knowledge were constantly challenged by both social disruption and a Hellenistic model of universal knowledge, sexuality according to Habinek gradually became unmoored from its position as one regulated practice among many and assumed a new position as an autonomous realm of concern. This move is testified to not only by the emergence of figures like Clodia and Sempronia, not only by the increased concern over the nature of the mos maiorum, but also by the appearance of Latin love elegy itself (Habinek 1997; Kaster 1997: 17; Wallace-Hadril 1997: 9-10, 20-21). Indeed, where the literature of the second century B.C.E. is dominated by narratives either commissioned by or about the ruling class, and functioned as a means of subjection and normalization, the elegists occupied a far less stable place. Theirs was an interstitial space between masculine and feminine, active and passive, for which traditional Roman discourse had no terms (Habinek 1998: 45, 121; Fantham 1996: 11; Sharrock 1995: 166). They were in short the augurs of instability.

The sexual and moral issues, which were emblematized in the figures of the elegists and their dominae, were profoundly linked with anxieties about the nature of the Roman class structure (Walters 1997: 30). These anxieties only deepened as the role of the traditional senatorial oligarchy in the administration of the state diminished in the final years of the republic and the early principate (Edwards 1993: 182-86). These sexual issues, moreover, formed the basis of Augustus's program of moral and religious reform. In addition to its more strictly ideological goals of establishing the princeps as the guardian of traditional morality, this program was also designed to repopulate the upper echelons of Roman political society, which had been depleted by a hundred years of civil conflict. Such a restoration of the equestrian and senatorial orders would provide the princeps with a strong political base (Grimal 1986: 157; Edwards 1993: 37-42, 57) and at the same time work to restore the ideological hegemony of the state and its elites.57 The traditional categories of commitment to Roman political life and the public advancement of both self and household were no longer fully operative, but a complete retreat to the world of private pursuits, a realm whose existence traditional Roman ideology barely acknowledged, was not acceptable either (Nicolet 1966: 721; Veyne 1987: 95; Edwards 1993: 32; Fitzgerald 1995: 25). The categories of what constituted normative subjectivity in the Roman world are at this time in a state of flux as the verities of republican civic virtue find themselves more and more irrelevant in an increasingly administered society (Syme 1960: 508; Auerbach 1965: 247-48; Foucault 1984: 104-5; Fantham 1996: 56), whose primary claim to ideological legitimacy is that nothing has really changed. "The republic had been restored." "We have only returned to the way things really were all along." The principate's motto was "Back to the Future!":58 for Roman ideology was profoundly conservative and unable to think of the new (res novae) except as the threatening (Wilkinson 1974: 168).

[M]asculinity, as defined through political and social competition, was at the end of the first century B.C.E. an increasingly hollow form of theater, "a loathsome and bitter burlesque," as Carlin Barton has put it. Its recuperation in elegy is therefore parodic, but not simply funny; for its male authors, elegy's wounds are ambiguous metaphors for the transformation of elite masculinity into text.
     --Fredrick 1997: 172-73
[T]he imaginary is, in one sense, a thoroughly historicized concept.
     --Myers 2001: 37

Elegy is a symptom of this crisis in the Roman subject's self-conception. It does not so much reflect the lives and positions of a Tibullus or Propertius as it does a crisis in the categories of the Symbolic and the way the individual subject relates to them.59 This crisis, in turn, is the point where we witness of the emergence of the Real as a moment of non-sense beyond the existing categories of understanding, the point where History becomes visible as that which exceeds the artifices of consciousness. It is, as we shall see in the coming chapters, elegy's very aporetical character60 that is the truest marker of the historical change to which it testifies, and not the various and often self-contradictory meanings those experiencing these events attributed to those changes (Zizek 1989: 170-71; 1991: 30). That moment of emergence, I argue moreover, take place most radically within the erotic precisely because it is here that the conjugation of the private fantasmata of our Imaginary self-construction, our individually assumed self-image (Lacan 1966b: 90), and the publicly sanctioned realm of Symbolic norms, which recognize us as a subject, takes place with the greatest intensity and, hence, with the greatest possibility of conflict.61 Augustus's moral reform legislation recognized this fact but could not fully control it. Thus elegiac discourse offers a privileged vantage point for observing the production of this split in the Roman subject of the late republican and early imperial period.62

In a previous study, I have argued that Latin love elegy is a subgenre of lyric and that one part of what makes lyric subjectivity possible in first-century B.C.E. Rome is a phenomenon labeled "semiotic slippage" (1994: 45-47, 131-40). Semiotic slippage in this case refers to Catullus's, and presumably the other neoterics', use of traditional Roman value terms like fides, foedus, amicitia, otium, negotium, and officium to delineate internalized realms of feeling, rather than the more traditional reciprocal communal and political obligations to which these words commonly referred (Edwards 1993: 183; Conte 1994: 137, 323).63 Amicitia, for example, although often translated "friendship," is generally thought to refer to a set of conventionally accepted, mutual obligations agreed upon between two parties, a relationship sometimes closer to our notion of political alliance than to the affective bond we label friendship, although the latter is by no means excluded.64 Thus as Sallust famously notes, "sed haec inter bonos amicitia, inter malos factio est" (but what between good men is called friendship, between the bad is called faction; Bellum Jugurthinum 31.15). Amicitia is also used of formal treaty relations between countries where it is a synonym for foedus or "pact" (Caesar, Bellum Gallicum 4.6; Sallust, Bellum Jugurthinum 8.2, 104.4-5; Livy 7.43). Finally, while Cicero gives an idealized image of friendship in De Amicitia, in De Officiis (2.69) he recognizes that it was often a euphemism for patronage relations between members of the upper classes. Likewise, in a letter to Trebatius, he chides the latter for trading too openly on the beneficia Cicero's amicitia has secured him from Caesar (Ad Familiares 7.17.1; cf. 7.5.1). An amicitia thus was a relationship of mutual beneficia65 that could be formalized into a foedus, guaranteed by fides (Ad Familiares 23.16.2, 23.19.2),66 in which each party owed the other certain officia67 ranging from material and political support to simple friendship (Bowditch 2001: 17-24, 52; Helzle 1989: 22; Ste. Croix 1981: 343; Boucher 1980: 92). It could be broken off when one party or the other suffered iniuria and declared the other inimicus.

David Konstan rightly points out that the stress on reciprocal obligations in amicitia should not be thought automatically to exclude the affective dimension (1997: 122-48). W. Jeffrey Tatum is good on this when he notes, "Friendship and politics, while not identical, were inextricable in republican Rome" (1997: 484; see also 486-87). Tatum criticizes my stance in Lyric Texts and Lyric Consciousness (1994: 120-40), assimilating it to Wilamowitz's (1924: 305-10) position that Catullus's poetry is purely personal and in no way political. Unfortunately, my merit is not so great as the German philologist's, nor are our positions identical. My claim rather was that:

[T]he lyric genre requires a society with a group of educated people who possess sufficient standing in the community to lay claim to a voice in its collective discourse but who do not, at the same time, occupy so high a rank that their interests are perceived as being identical to that of the state moreover in order for these conditions to be such as would permit the emergence of the lyric genre, this attention must be focused on the poets as individuals and artists, not as representatives of their particular group, or of society's ruling class. (1994: 125)

Such a view hardly implies a complete abstraction from the power relations that constitute a society's various political fields broadly conceived (Tatum 1997: 482-83). Rather, as Leach (1999: 148 n.27) has recently pointed out, this position closely corresponds to the situation in which Cicero found himself during Caesar's dictatorship.

In Catullus's hands, amicitia and related terms came to designate those aspects of his relationship with Lesbia that transcended the more purely sexual realm denoted by the term amor (72.7-8, 76. 1-4, 109.6).68 Whether he was the first to make such use of these terms cannot be said, but we have no record of such usage before him. What these renovated definitions made possible was the creation of a vocabulary permitting the discussion of issues beyond those generally recognized as admissible to Roman public discourse, with its traditional concentration on questions of money, power, and politics (Lyne 1980: 68; Albrecht 1997: 752). When Augustus is presented as restoring the goddess Fides to her rightful place at Carmen Saeculare 57 and Aeneid 1.291, it is the normative use of the term that is being evoked, not the "faith" Catullus pledges to Lesbia (Boucher 1980: 101-2). Yet, as I contended in my previous study (1994), starting from the work of Carl Rubino (1975: 291-93), and as Platter (1995) and Kennedy (1993) have argued still more forcefully, these new semantic possibilities never simply broke free of, or replaced, the traditional meanings of such words (Conte 1994: 254). Instead, they existed in a constant and unstable dialectical tension with them. Precisely that tension made lyric subjectivity as a publicly recognized, private consciousness possible, with all the contradictions that such a formulation implies.

In the present work, I move beyond this earlier formulation and, with the help of concepts borrowed from Lacanian psychoanalysis, analyze more precisely what the nature of that semiotic slippage was, how it made the elegiac subject possible, and how that particular kind of slippage was ultimately stabilized and recuperated by the ideology and culture of the imperial Roman state. I argue that the nature of these particular slippages can best be understood as symptomatic of the opening of a new, more radical gap between the Imaginary and Symbolic registers of language and experience in Roman society, between the subject's sense of self and its recognition as a subject in the world of codified, signifying practices. The Imaginary represents the image with which we identify ourselves. It is how we picture ourselves to ourselves. The Symbolic on the other hand represents the world of codes and signifying and practices wherein we are recognized as subjects by the social world around us (Ragland-Sullivan 1986: 56; Zizek 1989: 109). In first-century B.C.E. Rome, certain changes in the Real--the expansion of the empire, the civil wars, and the general collapse of social order (Nippel 1995: 35, 77-84)--had produced a situation in which the Symbolic's capacities for interpreting the subject no longer meshed with that subject's own Imaginary self-identifications (Leach 1999: 148-49). This separation of the Imaginary and the Symbolic in turn permitted the emergence of new forms of self-representation and behavior. Indeed, as the gap between the Imaginary and the Symbolic, which structure all human subjectivity, became more visible, new modes of meaning and experience were possible. To take only one well-known example, Syme and Gruen have argued that Julius Caesar interpreted his political project through the paradigms of traditional patrician ideology (Syme 1960: 52-59, 68-70, 194-95; Gruen 1995: 491-92). Yet the only way he could achieve his goals in a way consistent with his own self-image or dignitas (Syme 1960: 13, 25, 48; Gruen 1995: 75, 496), and the demands of a state shattered by sixty years of internal conflict, was to destroy the structure that made that ideology possible, namely, the republic.69

Whereas gloria in Roman ideology represented the recognition the self receives from the instances of authority in society, and so established the subject's position in the Symbolic, dignitas marked for the Roman elite precisely that point where the Imaginary "provided the form of the subject's lived relation to society. Through this relation the subject was brought to accept as its own, to recognize itself in, the representations of social order" (Copjec 1994: 21).70 Dignitas, as the social manifestation of the Imaginary, in fact represented the point where the ego's self-construction came directly into play with the categories of self-conception, the socially approved subject positions, offered by the Symbolic (Julien 1990: 67; Zizek 1989: 110-11; Althusser 1996c: 26; Lacan 1966b: 90-91, 95). In more traditional psychoanalytic terms, it represented the hinge point between the Imaginary ego and the ego ideal (Julien 1990: 68-69; Gunderson 1997: 215-16). Thus, whereas pudor represented an emotion that was ideally completely internalized, and gloria a recognition that could only come from the Other, dignitas was precisely the Imaginary self-construction we present to the Other.71 To destroy that was to destroy Caesar. In the end, shifts on the level of the Real, such as the professionalization of the army, Sulla's proscriptions, and the Social Wars (Nippel 1995: 65-67), had produced a gap between the Imaginary and the Symbolic that could only be filled by destroying the social world that had given rise to it. Dignitas could only be preserved by a revolution that would shake to the roots the social categories that had made its construction and misrecognition both possible and necessary.

Elegiac subjectivity thus, I argue, is a product of this same gap between Imaginary identifications and Symbolic conceptions. It presents a schizoid personality, whose visibly split nature constitutes its ultimate, irrecuperable remainder (the excess that cannot be processed by any of the available Symbolicforms, be they republican or early Augustan). Indeed, the gap itself structures this mode of discourse and accounts for its ultimately unique, if not aberrational, nature in Roman culture. It also accounts for its bizarre status in traditional criticism: for, depending on whom one reads, elegy is either in league with the Augustan political regime (Kennedy 1993: 35-36; Newman 1997: 6) or implacably hostile to it and the traditional values it sought to promote;72 either political allegory (Edwards 1996: 24) or an apolitical, ludic discourse that gently mocks social custom (Veyne 1988: 31-32 104-8; Kennedy 1993: 95-96; Fantham 1996: 108); either exploitative of women (Kennedy 1993: 38, 56, 73) or bent on satirizing Roman misogyny (Greene 1994). In fact, almost everyone agrees that elegy was opposed to something, but nobody can agree on what that was (Santirocco 1995: 226-28). By focusing on the split itself rather than on the pros and cons of a given binary opposition, the endless debates that have dominated elegiac criticism can be surpassed and the structure that gave rise to those debates examined.

Moreover, this shift in the relation of Imaginary to the Symbolic, which made the schizoid discourse of the elegiac genre possible can itself only be accounted for by historicizing Lacan's third term, the Real, for no change in the relations maintained between the first two registers is possible except in relation to this dialectically necessary tertium quid,73 which, as Zizek (1989: 162) has shown, became the increasing focus of Lacan's later inquiries. The Real, as noted earlier, is that set of material circumstances beyond all Imaginary and Symbolic representations that makes the continued existence of both of those realms possible. The increasingly aporetical relations exemplified between the Imaginary and the Symbolic in Latin love elegy can also be seen as the momentary emergence of the Real into symptomatic visibility. Yet, that visibility exists in itself only as a moment of pure unintelligibility, a traumatic eruption that is beyond signification. As Sheri Benstock observes, "Knots or holes in language signal the Lacanian Real" (1991: 17). In modern literary practice, she notes, it is the poetry of Gertrude Stein's Tender Buttons that comes as close as possible to the moment in which language stumbles "up against the Lacanian Real" (1991: xxvii).

Elegy is another such stumble. It can never reflect the Real; it can only detect its seismic shifts through disruptions in the structures of signification. Latin love elegy is, on this view, made possible by certain subterranean movements of the Real, but these by nature can only be discussed through indirection. The symptoms of that movement, however, are clearly visible in the sudden emergence of this schizoid discourse during the period marking the final collapse of the Roman Republic, while the subsequent stabilization of those same architectonic forces is made visible by elegy's disappearance with the consolidation of the empire.

Return to Book Description

File created: 8/7/2007

Questions and comments to:
Princeton University Press

New Book E-mails
New In Print
PUP Blog
Princeton APPS
Sample Chapters
For Reviewers
Class Use
Recent Awards
Princeton Shorts
Freshman Reading
PUP Europe
About Us
Contact Us
PUP Home

Bookmark and Share