University of Southern California Dissertations and Theses

The vulnerable corpus of Propertius

(USC Thesis Other)

The vulnerable corpus of Propertius

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1 The Vulnerable Corpus of Propertius Robert Matera a dissertation presented to the Faculty of the Graduate School of the University of Southern California in partial fulfillment of the requirements for the degree of Doctor of Philosophy (Classics) for conferral of the degree on August 8, 2017 i For my mom, who first taught me to love words, and for Adam, with whom I learn daily what love is. ii Acknowledgements It is a great pleasure to express my gratitude to those who have helped and supported me along the way to finishing this dissertation. First I would like to thank the members of my committee. My supervisor, Tom Habinek, has seen this project through from beginning to end, and his guidance, always gracious, has corrected and enriched my work in matters broad and minute. For his vast expertise, for his generosity in sharing it, and for his belief in me and in my project, he has my deep admiration and gratitude. Claudia Moatti has similarly encouraged and guided me from the start of my graduate career and from when I first began to formulate this dissertation. I have been very fortunate to benefit from her mentorship and her marvelous knowledge of many fields. I thank her for kindly and patiently providing a κανών to set me right time and again. Alice Echols has given advice and encouragement at crucial moments during this project, and I am grateful to her for her valuable insights and for working with me on a project so far afield from her own research. Several other faculty members at USC have also helped me develop this dissertation. Tony Boyle and Ann Marie Yasin were keen readers of my prospectus and gave helpful feedback. Greg Thalmann, James Collins, and Lucas Herchenroeder listened to parts of the project and asked thoughtful questions. Sharon Hays helped me work through some early ideas in her seminar. My thanks to all of them. I wish I had space to acknowledge individually every friend and colleague at USC. I am especially grateful to Hanna Mason, Matt Chaldekas, Hannah Çulik-Baird, Scott Lepisto, Afroditi Manthati-Angelopoulou, Ambra Spinelli, and Christian Lehmann for years of friendship and for conversations both thoughtful and frivolous (they were all meaningful). Each of them has given a great deal to my dissertation and to me. Thanks also to Jess Wright and Jack Kelleher for their friendship and for talking with me about epistemologies, social justice, and jobs this past year. They and the rest of my friends and colleagues at USC have made our own little chicken-coop of the Muses collegial and productive. I am fortunate to have been welcomed into several communities in LA outside of USC, and I would like to thank in particular the Revs. Katie Cadigan and Nate Rugh, JK Hilbert and the St. A.’s Choir, and the congregation of St. Augustine by-the-Sea. The peace, joy, and love of the St. A.’s community have helped sustain me through the anxieties of dissertating and the final years of graduate school. My love and thanks also to my dear friends Steph, Rachel, Buzzy, Jeremy, Dash, and Byron for humor, support, long and wonderful conversations, quick check-ins, understanding when I have sometimes gone MIA in the forest of graduate work, and too-infrequent but always joyful reunions. For my family there can never be enough thanks. That I still refer to Boston as “home” is due in significant part to Robin, Kristin, Ashley, and Jack. Meg, Tim, El, Christine, and all the Grodins have been tremendously loving and supportive – I am especially grateful that they know about academia and “get it.” Kate has been a rock and a sounding-board. Mom and David have cheered me on all the way since forever, have heard far more about Latin poetry than they probably expected or wanted, and have reminded me to be a human being when my work has tempted me to expect too much or too little of myself. Their love has been extraordinary. My greatest thanks are for Adam, for his love and patience day in and day out, for listening to me go on about my work, for leaving my books where I put them down (all over the apartment), for joining conversations about classics with my colleagues with gusto, for dinners and laundry and all the quotidian things, and for dreaming with me. iii Contents Introduction: A Trope and Its Contexts 1 Chapter 1: Propertius’ Mutable Text 18 Chapter 2: The Vulnerable Text as a Literary Trope 52 Chapter 3: The Uses of Propertius’ Elegies 79 Section A: Early Interpolations 80 Section B: A Culture of Literary Recycling 95 Section C: Orality and Authorship 138 Appendix A 166 Appendix B 174 Chapter 4: The Propertian Corpus 179 Chapter 5: Vulnerabilites Compared 217 Conclusion: The Value of Being Worthless 239 Bibliography 244 1 Introduction: A Trope and its Contexts What worse fate could befall a love poet than for his ex-girlfriend’s ghost to visit him in his bed, twist his words, and replace them with her own? In Poem 4.7, the speaker of Propertius’ Elegies tells how this very thing happened to him, and over the course of the Elegies he imagines all sorts of people—a procuress, an accountant, other poets, and even the emperor’s best friend— changing and destroying his poetry. In fact, the Elegies begin with the susceptibility of a poet’s words to change. The first four verses of the first poem of Book 1 famously rewrite the beginning of an epigram by Meleager. 1 Five verses later the speaker reworks part of Eclogues 10, in which Vergil takes on the voice of Propertius’ friend and mentor Gallus and adapts Gallus’ elegiac poetics to his own pastoral poetic project. 2 The Propertian speaker shies away from acknowledging that his own words are similarly mutable in the first book of the Elegies, 3 but in the first poem of Book 2 he begins to explore how other people might change or destroy his words, and he presents several such scenarios in Books 2, 3, and 4. Greek and Roman authors had long considered possible receptions of their works by the time Propertius composed the Elegies—the concern is at least as old as the ideas of the ‘glories of men’ (κλέα ἀνδρῶν) and song’s power to immortalize—and a number of authors weighed in on the theme during Rome’s transition from Republic to Principate. Some of them reveal (or purport to reveal) how they took actual and possible receptions of their work into account in the processes of composing and revising. 1 Prop. 1.1.1-4 and Anth. Pal. 12.101.1-3. For recent discussions see Hollis (2006) 107-8 and Keith (2008) 45-7. The reworking of Meleager’s epigram enters my argument in the section of Chapter 3 on “Canonical Poets.” 2 See Chapter 1 and Ross (1975) 62ff.; Kennedy (1993) 65-6; Clausen (1994) 290-1; Cairns (2006b) 84; and Keith (2008) 66-8. 3 See Chapter 1. 2 For Cicero, revision was a collaborative, social practice. He took his lead from Isocrates, who, in Gurd’s analysis, had treated literary composition as an opportunity to train his students to discuss, debate, and offer and receive advice, so that these practices would become part of the students’ habitus and so that they would form a society of cooperative thinkers. 4 In Gurd’s reading, Cicero began from the end-point of Isocrates’ pedagogy, collegiality, and used single texts in the ways Isocrates used series of texts. 5 Gurd draws some of his evidence from Cicero’s letters. In a letter to his friend Atticus in 61 BCE, for example, Cicero wrote, “I shall include the topographical description of Misenum and Puteoli you requested in my speech. That ‘on December 3’ was a mistake I had already noticed. The things you praise in my speeches very much please me, believe me, though I didn’t dare to say so before. But now they seem much more Attic since they are approved by you. I added some things to that Metellinan oration. The book will be sent to you, since your love for me makes you such a rhetorophile.” τοποθεσίαν quam postulas Miseni et Puteolorum includam orationi meae. ‘a. d. III Non. Dec.’ mendose fuisse animadverteram. quae laudas ex orationibus, mihi crede, valde mihi placebant, sed non audebam antea dicere. nunc vero, quod a te probata sunt, multo mi Ἀττικώτερα videntur. in illam orationem Metellinam addidi quaedam. liber tibi mittetur, quoniam te amor nostri φιλορήτορα reddidit. (Cic. Att. 1.13.5 tr. Gurd) In the first sentence of this passage, Cicero reports that he has made a specific change to one of his texts at Atticus’ suggestion, and in the second sentence he tells Atticus that he has already made a suggested change. The following sentences speak more generally to Atticus’ influence on Cicero’s writings and vice versa. Here, Gurd reads Cicero playfully describing the effects of this mutual influence: “if Cicero’s writing made Atticus a lover of rhetoric, Atticus’s solicitous readership made Cicero’s works ‘more Attic’ (Ἀττικώτερα). Linking a stylistic virtue (Atticism) with a proper name (Atticus) portrays the revised text at partaking of both.” 6 The letter thus presents a model of shared authorship: “Atticus added his authority to Cicero’s in such a way that the results 4 Gurd (2012) 25-47, 50-2 5 Gurd (2012) 50-2 6 Gurd (2012) 53 3 could be imagined as shared property or ‘public things.’” 7 Gurd finds similar intellectual relationships on display in letters from Cicero to Tiro, in which Cicero says that his freedman is “the canon of my writings” (κανὼν…meorum scriptorum, Cic. Fam. 16.17.1) and shares the authorship of these writings with him (Cic. Fam. 16.10.2), 8 and in letters in which Cicero discusses the revisions of his de Republica (Cic. Q. Fr. 3.5, Att. 6.2). 9 We can perhaps glimpse in these letters some of Cicero’s thoughts and feelings about sharing authorship of works of which he was the original creator and which remained under his name. Gurd notes the affectionate and attentive tones of letters such as Att. 1.13 and Fam. 16.10 and 16.17. The first, third, and fourth sentences of Att. 1.13.5 (quoted above) suggest that Cicero is entirely receptive to and grateful for Atticus’ feedback. Amid these sentences, the second sentence, in which Cicero informs Atticus that he has already noticed and presumably corrected “on December 3” (a. d. III Non. Dec.), is conspicuously less gracious in implying that Atticus’ comment on the error was superfluous. To call this defensiveness on Cicero’s part may be too strong, but he wants Atticus to know that he has already caught and fixed the mistake. There is a hint of authorial vanity, here, which the following sentences pick up and make into a coy joke: “The things you praise in my speeches very much please me, believe me, though I didn’t dare to say so before. But now they seem much more Attic since they are approved by you” (quae laudas ex orationibus, mihi crede, valde mihi placebant, sed non audebam antea dicere. nunc vero, quod a te probata sunt, multo mi Ἀττικώτερα videntur, Cic. Fam. 1.13.5 tr. Gurd). Similarly, while Gurd rightly points out expressions of shared authorship in one of Cicero’s letters to Tiro, the sentences Gurd highlights also show ambivalence about shared authorship. 7 Gurd (2012) 53 8 Gurd (2012) 53 9 Gurd(2012) 53-6 4 When Cicero writes, “my texts—or rather ours are languishing out of longing for you” (Litterulae meae sive nostrae tui desiderio oblanguerunt, Cic. Fam. 16.10.2 tr. Gurd), he has kept the initial possessive adjective, “my” (meae) instead of removing it. This produces an effect of spontaneous correction, suggesting artless intimacy in a flirtatious sentence, but it also indicates that Cicero’s first impulse was to call the texts his rather than his and Tiro’s. Cicero also aligns the texts with himself when he says that the texts “are languishing out of longing for [Tiro],” which of course communicates that Cicero languishes and longs for Tiro. A little later, when he writes, “and when [Pomponius] wanted to hear some of our writing, I told him that everything of mine was silent without you” (ei cupienti audire nostra dixi sine te omnia mea muta esse, Cic. Fam. 16.10.2 tr. Gurd), he reinforces the ambiguity: First the writings are “ours” (nostra), but the first-person plural is often used in place of the first-person singular, “mine,” as in the letter to Atticus quoted above. Next the writings are distinctly “mine” (mea) but depend on Tiro to such a degree that they are “silent” (muta) without him. It is also worth keeping in mind that Tiro was Cicero’s slave and then freedman and that the writings were under Cicero’s name. Whether or not Cicero genuinely thought and felt the things I have just read in these letters—the letters are, after all, literary works by a masterful and politically motivated rhetorician 10 —he shows himself negotiating tensions over authority within a paradigm of collaborative authorship: Shared authorship does not necessarily mean equal authorship. 11 The dynamics of collaboration on display in the letters reveal, and in part produced, Cicero’s regard for his audiences’ receptions of his texts. We can, with Gurd, read Cicero’s collaborators as co-authors, but they are also audiences that receive his texts in various stages of 10 Butler (2011) 63-74, Gurd (2012) 49-76. 11 To be clear, Gurd does not say that it does. 5 formation. Furthermore, an author is the work’s first reader, as Butler reminds us, 12 so Cicero’s collaborators share with him the position of author-reader. When Cicero revised speeches and philosophical works on his friends’ advice, he made use of their reactions to the versions he had shared with them. In Gurd’s reading, Cicero’s care for his audiences’ receptions of his texts is both literary- scholarly and socio-political. The letters express Cicero’s interest in producing stylistically and intellectually excellent works and his interest in the social relationships that collaborative authorship entails. Gurd finds evidence for the politics of Cicero’s collaborative authorship in Cicero’s philosophical works and letters from the 40’s BCE. 13 Habinek has explored how Cicero’s positions on literary style promote the cultural authority of Rome’s intellectualist elites over and against that of its militarist elites. 14 Gurd argues that, for Cicero, processes of literary composition and the styles of the works they produce are bound up in the struggle between Caesar’s autocratic politics and his own republican politics. 15 Cicero advocated and practiced an Asianist style, brilliant but baroque and so prone to imperfections and always open to further revision, which provided opportunities for Cicero’s coterie to work together to refine their texts through discussion. 16 In the perfect Atticism of Caesar’s Commentarii, by contrast, Cicero saw the triumvir’s autocratic politics writ small: “For they are bare, direct, and beautiful, stripped of all ornamentation as though of a garment. But if he wanted others to have material that whoever wanted to write a history could take up, perhaps he did a favor for the incompetent, who would brand his texts with their hot irons; but prudent men he terrified away from writing, for there is nothing sweeter in history than pure and shining brevity.” nudi enim sunt, recti et venusti, omni ornatu orationis tamque veste detracta. sed dum voluit alios habere parata, unde sumerent qui vellent scribere historiam, ineptis gratum fortasse fecit, qui illa volent calamistris 12 Butler (2011) 13-27, 30, 36 13 Gurd (2012) 57-76 14 Habinek (1994) 15 Gurd (2012) 57ff. 16 Gurd (2012) 61ff. 6 inurere: sanos quidem homines a scribendo deterruit; nihil est enim in historia pura et inlustri brevitate dulcius. (Cic. de Rep. 262 tr. Gurd) Not only do Caesar’s writings obviate any need to re-work their material, they scare anyone prudent away from trying. 17 Whereas Cicero’s Asianism occasions dialogue that builds community, Caesar’s Atticism silences everyone else. Gurd thus argues that Cicero saw in his Asianism and collaborative authorship a hope for the continuation of the Republic and in Caesar’s Atticism and sole authorship the end of the Republic. That a text should be open to various receptions, then, was of paramount importance for Cicero’s republic of letters. Horace is less sanguine about collaboration and revision. In his Odes, the speaker presents his works as an immutable monument: “I have raised a monument more lasting than bronze…” (exegi monumentum aere perennius…, Hor. Carm. 3.30.1). 18 He also holds much of his audience at arms’ length and scorns their opinions: “I hate the vulgar crowd and keep it at a distance; hold your tongues” (odi profanum vulgus et arceo; / favete linguis, Hor. Carm. 3.1.1-2) and “Not deceitfully did Fate give me little fields and the slender breath of the Greek Muse and disdain for the malignant crowd” (mihi parva rura et / spiritum Graiae tenuem Camenae / Parca non mendax dedit et malignum / spernere vulgus, Hor. Carm. 2.16.38-41). In these verses, the speaker links his refined, Greco-Italian 19 poetics with his disdain for the crowd as gifts from the same goddess. Elsewhere, he boasts that his metrical dexterity elevates him above others (Hor. Carm. 1.1.29-36, 4.3, 4.6), and indeed the intricate lyric meters of the Odes and Horace’s mastery of them may have discouraged others from tinkering, as even Ovid says that Horace dazzled him: “And Horace the master of meters captivated our ears / while he played refined songs on his Ausonian lyre” (et 17 Gurd (2012) 58-9 18 See also Hor. Carm. 2.20, in which the speaker says that his poetry will make him immortal and, in its monumentality, will render a tomb a “superfluous honor” (supervacuous honores, 24). 19 On Camenae and the mixing of Greekness and Italianness in Latin poetry see, for example, Goldberg (1993) 22-8 and Dominik (1993) 38-56. 7 tenuit nostras numerosus Horatius aures, / dum ferit Ausonia carmina culta lyra, Ov. Trist. 4.10.49-50). No interest in collaboration here, then – these passages announce that the speaker wants only one reception from his audience (or at least, most of his audience): Shut up and listen. In the end, though, such a defensive reaction to the opinions of the crowd allows the crowd to influence the speaker’s poetics. Other passages in the Odes also betray ambivalence about receptions of the speaker’s texts. Verses 23-9 of Odes 1.28 instruct the addressee to bury the body of the ghost who is speaking so that the addressee may remain unaffected by literal and metapoetic storms: “however much Eurus will threaten the Italian waves, let the woods of Venusia be lashed while you are safe…” (quodcumque minabitur Eurus / fluctibus Hesperiis, Venusinae / plectantur silvae te sospite…, Hor. Carm. 1.28.25-7). Horace was from Venusia, and woods (silvae) are a common figure for poetic material in Greek and Latin poetry, 20 so the lashing of the woods of Venusia suggests savage criticism of Horace’s poetry. In verses 27-9, the ghost offers divine protection from such storms as an inducement to bury his corpse, presumably thinking that his addressee, the Horatian speaker, would desire this immunity. Elsewhere in the Odes, the speaker reports that a tree planted in his field fell on his head (Hor. Carm. 2.13, 2.17, 3.8). Forests consist of trees, which are also the material (materia) of literature in and of themselves, and in Odes 2.13, after his schetliastic opening, the speaker launches into a diatribe against carelessness that revisits material found in Homer, Simonides, Pindar, Lucilius, Catullus, and others. 21 He thus humorously enters a whole forest of trees that 20 On silva as a literary metaphor, translated from the Greek ὕλη, see Coleman (1988) xxii-xxiii; Hinds (1998) 11- 14; Heerink (2015) 8; OLD s.v. 5 and, for example, Cic. Orat. 12: “for indeed all fullness and, as it were, the silva of speaking has been drawn from them [i.e., philosophers’ debates]” (omnis enim ubertas et quasi silva dicendi ducta ab illis est). 21 Nisbet and Hubbard (1978) 201-22 8 might fall on him. In fact, the entire episode “is written within a literary tradition” in which “death [and injury] from falling objects” was a topos. 22 When the speaker brings up the tree again in Odes 2.17, he is telling Maecenas to stop “smothering [him] with complaints” (me querelis exanimas, Hor. Carm. 2.17.1) and just to be glad that he is still alive to compose poetry, considering what the tree almost did to him. He begins Odes 3.8 by entering another literary forest “with a parody of aetiology” 23 to explain to Maecenas that he is celebrating the anniversary of his survival, since he was “nearly killed by the blow from the tree” (prope funeratus / arboris ictu, Hor. Carm. 3.8.7- 8). The word for blow, ictus, is also used of stress in poetic meter, 24 so it draws another lexical connection between the tree and poetry. In Odes 2.13, 2.17, and 3.8, then, the speaker repeatedly links the falling tree that almost killed him with poetry. Several interpretations are possible – here is one: Material (the tree) in one of the speaker’s poems (his field), but which originated in another author’s work (planted by someone else), incurred criticism or got the speaker into trouble (fell on his head). In these readings of the lashing of the woods of Venusia and the falling tree, the speaker of the Odes acknowledges the power his audience has in its reception of his works. Horace considers reception more explicitly in his hexameter poetry, when he discusses authorial practices such as receiving criticism and revising. Gurd summarizes Horace’s position on revision in the hexameter poems thus: “Not collaboration but mortification was the chief characteristic of literary genesis in Horace; having to revise meant not getting it right the first time—but the good poet always had to revise, and therefore bore the weight of constantly falling short.” 25 Along with the emphasis on perfection in Alexandrian poetics, the Augustan poets had 22 Nisbet and Hubbard (1978) 202 23 Nisbet and Rudd (2004) 123 24 TLL 7.1.165.23-32. Horace unambiguously uses ictus with this meaning at Ars Poetica 253. 25 Gurd (2012) 77 9 adopted Hellenistic discomfort with the imperfection that revision implied. 26 In the world of Horace’s hexameter poetry, a poet who needed to revise could expect ridicule. Satires 2.3, for example, presents “the ruined art dealer turned Stoic diatribist Damasippus,” who evaluates the speaker not as “a poet who labors toward highly polished results” but “rather, an indolent and perpetually frutrated author whose constant failure to finish is the product of an interminable deferral.” 27 In the Ars Poetica, the speaker thus advises the Pisones to strictly limit who sees their texts while they are still in progress: “But once you have written something, let it enter Maecius’ ears for judgement and your father’s and mine, and let it be suppressed for nine years, the pages placed inside something. You can delete what you won’t submit for comments; a voice sent out doesn’t know how to turn back.” si quid tamen olim scripseris, in Maeci descendat iucidis auris et patris et nostras nonumque prematur in annum membranus intus positis. delere licebit, quod non edideris; nescit vox missa reverti. (Hor. Ars 386-90) In the same vein, he recommends in Satires 1.10, “nor let the crowd admire you and your labors / but be content with a few readers” (neque te ut miretur turba labores / contentus paucis lectoribus, Hor. Sat. 1.10.73-4). Seven verses later he produces a culled list of readers, though after naming Plotius, Varius, Maecenas, Vergil, Valgius, Octavius, Fuscus, the Visci, Pollio, Messalla, Messalla’s brother, Bibulus, Servius, and Furnius, he concedes that there are “many others, learned friends whom I discretely decline to name, whom I’d like these poems, such as they are, to please” (compluris alios, doctos ego quos et amicos / prudens praetereo, quibus haec, sint qualiacumque, / arridere velim, Hor. Sat. 1.10.87-9). These are more than a few readers, and in this perhaps the speaker is satirizing himself. In any case, Gurd notes that “this list has been trimmed down in a 26 Gurd (2012) 78-91 27 Gurd (2012) 94 10 kind of human or social revision, offering a catalogue that echoes or completes the circle opened by Satires 1.10.40-9... Horace is exploiting a set of similarities between revision, the selection of an elite group of readers, and the process of canon formation.” 28 In this reading, the Horatian speaker edits his group of readers much as he edits his texts and so exercises at least some control over the reception of the latter by the former. In contrast to Horace’s speaker, Ovid’s speaker proudly flaunts revisions he may or may not have actually made. To take one example, 29 the preface to the Amores has attracted much critical attention on this front: “We who just now had been Naso’s five books are three; the author preferred this work to that. Though it may still be no pleasure for you to have read us, at least, with two removed, your suffering will be lighter.” Qui modo Nasonis fueramus quinque libelli, tres sumus; hoc illi praetulit auctor opus. ut iam nulla tibi nos sit legisse voluptas, at levior demptis poena duobus erit. (Ov. Am. praef. 1-4) Whether Ovid first published a five-book edition of the Amores and later a three-book edition intended to replace it, or whether the preface is a wholly fictional exercise in metapoetics, 30 the preface positions the speaker as an author who revises out of consideration for the reader. Martelli observes that prefaces “make the point of entry for the reader a point of departure for the author, and initiate our reading experience by emphasizing the temporal disjunctions that separate our encounter with the text from its first creator’s.” 31 Ovid’s preface highlights the speaker’s control of the text, as its author, 32 even as it affects deference to the reader. With this maneuver, the 28 Gurd (2012) 97 29 See Martelli (2013) for an extended study of Ovid’s revisions as authorial self-making. 30 The large body of scholarship on this question includes Pohlenz (1913); Oliver (1945); Luck (1961) 166ff.; McKeown (1987) 75-89; Barchiesi (1988) and (2001) esp. 160-1; Holzberg (2002) 31ff.; and Martelli (2013) 35-67. 31 Martelli (2013) 13 and cf. Butler (2011) 13-27, 30, 36 on the author as a work’s first reader. 32 Martelli (2013) 14, Oliensis (2014) 209-10 11 preface makes the tension between the allegedly revised edition and “the idea of the original… a determining hermeneutic premise” for the reader. 33 It directs the reader, for example, to understand the speaker (aligned with Ovid) as “already an author” 34 who is “in the middle part of his career” 35 and whose earlier self is, in place of perhaps Propertius, “the most significant literary precursor for Ovid, author of [the revised edition].” 36 The preface thus exploits the real or imagined reception of a first edition of the Amores to present the allegedly revised edition and its author as the up-to-date standard in relation to the already outdated first edition and so to push the works of senior colleagues such as Propertius and Tibullus even farther back into literary archaicness. 37 Revising was but one literary practice in Rome that treated texts as fluid and changeable rather than fixed in a definitive version; another was the deployment of paratexts. As seen in the Ovidian epigram above, paratexts often muddled the boundaries of texts. O’Rourke finds paratexts within the ‘text proper’ of Book 4 of Propertius’ Elegies and argues, furthermore, that the paratexts bring other authors’ words and ideas and Propertius’ own words from earlier books into Book 4 to point out the text’s instability. 38 Peirano takes the literary sphragis, or seal, as a reflection of the physical seal, which is “not a direct referent for the literary sphragis” but rather “structures and informs the literary representation of the edges of the text” 39 under the sign of the author’s name. 40 33 Martelli (2013) 14 34 Oliensis (2014) 210 35 Farrell (2004) 46 36 Martelli (2013) 36 37 See Hinds (1998) 52-63 and 76-91 on newer authors constructing older authors as archaic. See Chapter 2 for discussion of this phenomenon in Propertius’ Elegies. 38 O’Rourke (2014) 39 Peirano (2014) 227 40 Peirano (2014) 12 Gibson and Rees explore how indices and intertitles can structure a reader’s experience of a text and can even attempt to deceive a reader about the layout of a work. 41 Sometimes people other than a text’s original author changed it. Interpolations could enter a text through a copyist’s error, though according to Reynolds and Wilson this was relatively rare in Roman texts before late antiquity. 42 More commonly, they argue, ancient textual critics corrupted the very texts they sought to correct. 43 Tarrant schematizes intentional changes to the original text (or what the critic thought was the original text) as “emendation, annotation, and collaboration,” 44 and in Chapter 3 I consider some possible early interpolations in Propertius’ Elegies in Tarrant’s terms. Eco writes that some texts are “open” to reworking in that they encourage their readers or audiences to participate in structuring their composition. 45 In cases of collaborative interpolation, in which it appears that the interpolator sought to improve or expand the original text, 46 the interpolator treats the original text as ‘open’ in this sense. This could happen in a performance of a text, as well. Donatus reports that a wag at one of Vergil’s recitations interrupted the poet to replace Vergil’s sententious ending of Georgics 1.299 with a comic one, and Martial accuses Fidentinus of changing his book so much that he becomes its new author just by reciting it badly (Mart. Epig. 138). 47 An author’s text could also change in the hands of another author: Roman authors frequently quoted or alluded to each other’s words and ideas and inevitably changed them when they did so. 48 In the works of some authors, including Propertius, the idea of 41 Gibson (2014) and Rees (2014) 42 Reynolds and Wilson (1968) 26-7 43 Reynolds and Wilson (1968) 27-30 44 Tarrant (1987) 126 45 Eco (1989) 15. See Chapter 2 for further discussion. 46 Tarrant (1987) 137 47 See Chapter 3 for further discussion. 48 Teasing out strands of quotation, citation, allusion, intertext, poetic memory, echo, and other models for relationships among ancient texts is an ongoing scholarly project. See Chapters 2 and 3 for discussion of some of this scholarship. 13 forests (silvae) of literary material (materia) links literary reworking with the writer’s wooden tablets. 49 This link was not just a metaphor – material factors in the processes of composition and publication also made texts susceptible to change, and some Roman writers incorporated their awareness of this into their discussions of their texts as texts. Butler has analyzed the politics of the erasability of words on wax tablets and papyrus in Cicero’s works: On the one hand, Cicero rocketed to fame by recognizing that Verres had exploited the ability to erase his name from the records of a tax collector, and on the other hand Cicero played with the erasability of his own words and name in his self-presentation as a friend, an author, and a politician. 50 Papyrus rolls and wax tablets were themselves vulnerable to misadventure. Farrell points out that the celebrated comparison of the physical and literary beauty of Catullus’ poetry book would have been written at the start of the papyrus roll. In this position, the papyrus and ink of a copy of Catullus 1 could quickly become less than pristine: “Because of the way in which they are handled, the outer part of the roll is especially liable to damage of every kind. If it is not actually torn away, it is very likely to become soiled through constant handling.” 51 Farrell then links this precarity with Catullus’ discussions of poetic permanence and impermanence, writing versus singing, and canonicity. 52 Roman finds Catullus, Propertius, Ovid, and Martial negotiating similar ideas when they lose, discard, or give away their tablets. 53 In contrast, a growing body of scholarship explores the meanings that inscriptions and inscriptional language embedded in ‘book poetry’ produce. 54 49 Roman (2006) 352 50 Butler (2011) 63-74 51 Farrell (2009) 167 52 Farrell (2009) 167-84 53 Roman (2006) 54 Fedeli (1989), Ramsby (2007), Videau (2010) 71-130, Dinter (2011) and (2013), Houghton (2013), Nelis-Clément and Nelis (2013), Bettenworth (2016) 14 Some of this scholarship considers how Roman elegy, in particular, uses inscriptions and inscriptional language both to claim permanence and to complicate that claim. 55 In this dissertation, I argue that the speaker of Propertius’ Elegies takes a radical view of textual fluidity when he anticipates what may happen to his texts when they leave his control. In Chapter 1, I analyze passages in Propertius’ Elegies in which the speaker explores the possibility that other people may change or destroy his texts. He begins in Book 1 by showing how other poets’ texts can be changed and coopted, but he does not yet acknowledge this likely reception of his own poetry. This acknowledgement begins in Book 2, and I consider possible reasons for the speaker’s new interest in the vulnerability of his text to change and destruction, including the wildfire success of Book 1 and the sudden fall from grace and subsequent suicide and possible censorship of his friend and mentor Cornelius Gallus. In Books 2, 3, and 4, the speaker presents scenarios in which other characters change his words, change the meanings of his words, or even erase his words. The particulars of these vignettes are as different as the particulars of real-life situations, and yet they all point to the ways in which users of Roman poetry sometimes changed or effaced the texts they were interacting with. Through repetition and variation, the vulnerability of the text thus becomes a trope which the speaker incorporates into his poetics. In the rest of my dissertation, I explore how this trope relates to broader literary and socio-political contexts. Chapter 2 situates the trope of the vulnerable text among ancient and modern literary- critical terms that describe ways in which texts interact with each other and ways in which authors reuse each other’s words and ideas. Of the many possible ancient terms and concepts, I focus especially on echo and anaklasis, because Propertius invokes echo as an acoustic phenomenon and a literary figure in Poems 1.18 and 1.20, and because he uses anaklasis frequently. Modern 55 Ramsby (2007), Videau (2010) 71-130, Dinter (2011), Houghton (2013), Bettenworth (2016) 15 scholarship has debated the concepts of allusion and intertextuality and their workings in Latin poetry for several decades. I assess how the vulnerable text of the Elegies relates to several scholars’ theories of allusion and intertextuality in Latin poetry: When the speaker of the Elegies shows other people quoting and changing his words and ideas, he inverts the process of allusion and provides an object lesson in the multiple directions of influence at play in moments of intertextuality, as scholars have theorized these two concepts. In some vignettes, he even shows the deformazione of his poetry in the hands and mouths of other people, such as the procuress Acanthis. I thus suggest that vulnerable textuality is a rhetorical maneuver with which the speaker of the Elegies anticipates how later texts will interact with his text and manages those anticipated interactions by integrating them into how he presents the Elegies as a text. I then consider how this rhetorical maneuver relates to Eco’s concept of an open text and how, with the combination of repetition and change that it anticipates, it fits within an understanding of Latin poetry as ritualized speech that can shape the world around it. In Chapter 3, I continue to address the question taken up in Chapter 2, to what features of Rome’s literary culture might the trope of the vulnerable text in Propertius’ Elegies respond, by looking at examples of the early reception of the Elegies. There are three sections to the chapter: In Section A, I examine evidence of very early interpolations in the text of the Elegies and analyze the changes these interpolations introduce into the text. In Section B, I explore how grammatical treatises, famous poems, and graffiti and other inscriptions change Propertius’ words and ideas while leaving them still recognizable as Propertian in origin. In Section C, I present evidence of an oral poetry culture alongside Rome’s written poetry culture and ask how certain features of oral poetry might affect ideas about authorship in written poetry and even efface words’ relationship with their original author. 16 Chapter 4 introduces another aspect of the Propertian speaker’s understanding of his text as a text, the connection between his words and his body. The speaker joins the Greek and Latin literary tradition of comparing a text with its author’s body, and this chapter surveys passages in which he does so. The associations between text and body are grouped thematically: association by physical presence and proximity, association through garland imagery, association or equation of the text with part of the speaker’s body, and association or equation of the text with the speaker’s whole body. I look briefly at how the Elegies adapt body metaphors from earlier authors, particularly Callimachus and Cicero. I then argue that presentations of the Elegies as performances and as books could incorporate them into the bodies of participants and audiences; theories of “new media” and embodied cognition help to understand how this might work (Malafouris’s discussion of the Mycenaean sword provides a useful comparandum). The different ways in which the speaker draws relationships between his text and his body explore how words could be part of a human body in Augustan Rome. In Chapter 5, I ask how the Elegies connect the literary culture discussed in Chapters 2 and 3 with the textual and human bodies (corpora) discussed in Chapter 4. Specifically, I consider what the speaker of the Elegies says or implies about these bodies when he imagines someone else changing or destroying his words. After briefly reviewing traditional Roman ideas about liberty (libertas) and elite manliness (virtus), I present several passages in the Elegies in which the speaker expresses anxieties about the integrity of his body and his speech. In the first few passages, the speaker is concerned with the loss of liberty, consisting in part of the vulnerability and violation of his body and curtailment of his speech, that his love affair with Cynthia inflicts. In the remaining passages, he links his erotic oppression with Octavian/Augustus. With these readings, I examine how the speaker of the Elegies thinks through several challenges to his identity as an elite, citizen 17 man by comparing the literary vulnerability of his text and his body with the perceived political precarity of his speech and his body. By looking at several modes of reception, such as allusion, shared authorship, and collaborative interpolation, from the perspective of an author anticipating what will happen to his text, this dissertation explores the relationship between a practical consideration of authorship and a poetic presentation of authorship. Hellegouarc’h writes that “the most general sense of esse auctor is ‘to act as guarantor of,’ and auctoritas is, in the most general and earliest sense of the word, ‘a guarantee.’” 56 In Gurd, Martelli, and Butler’s readings, Cicero, Horace, and Ovid may revise or claim to revise in response to their audiences’ receptions of their texts, but they remain the ultimate guarantors of their texts. Farrell and Roman argue that Roman poets concede their inability guarantee physical copies of texts only to reaffirm their ability to guarantee their texts as oral media. I argue that, in Propertius’ Elegies, the author acknowledges that he could guarantee nothing about the future of his text in general or in any individual copy or performance. This dissertation thus reevaluates how Propertius presents himself as an author and an elite Roman man in light of this break from more traditional Roman understandings of authorship. 56 Hellegouarc’h (1963) 296: “Le sens le plus général de esse auctor est « se porter garant de », et l’auctoritas c’est, au sens le plus général et le plus primitif du mot, « la garantie ».” 18 Chapter 1: Propertius’ Mutable Text Introduction Augustan poets could expect other people to reuse, change, and erase their words, as they themselves frequently quoted, alluded to, repurposed, and reworked each other’s and earlier poets’ words. 57 Nor was it just famous poets doing these things with each other’s words: In the previous generation, Varro had gathered snippets of poetry in his treatise On the Latin Language, and his successors, such as Verrius Flaccus and later Quintilian, Rutilius, and Caesius Bassus, realized Horace’s fear that his poetry would become a banal tool for teaching young boys (Hor. Epist. 20.17-8) by using verses from famous poets as examples and sometimes changing them in the process. Likewise, graffiti and other inscriptions quote, allude to, and mash-up canonical poets’ words. 58 Several poems in Book 1 of Propertius’ Elegies consider the susceptibility of other people’s poetry to change and criticism but ignore or deny their own susceptibility. Books 2, 3, and 4 present scenarios in which other people change, re-purpose, criticize, or destroy the speaker’s words. Ramsby has documented the Augustan elegists’ interest in the monumentality and permanence of their poetry, 59 and (as discussed in the Introduction) Farrell and Roman have teased out several authors’ ideas about permanence with regard to physical copies and oral performances of their poems. Janan, meanwhile, has analyzed how Book 4 of Propertius’ Elegies criticizes Propertian poetics. 60 Butler has highlighted some of Cicero’s thoughts on the erasibility of texts, Gurd has discussed Cicero’s collaborative revision and Horace’s apprehensions about being seen 57 Conte (1986); Barchiesi (1993); Hinds (1998); et al. 58 Lissberger (1934); Wachter (1998) 59 Ramsby (2007) 60 Janan (2001) 19 to revise, and Martelli has examined the processes and outcomes of Ovid’s revisions. 61 In this chapter, I explore how the speaker of Propertius’ Elegies incorporates the idea of his texts’ vulnerability into his poetics. Book 1 In Book 1 of Propertius’ Elegies, the speaker does not accept his text’s vulnerability to criticism, change, or destruction. He forbids criticism of his poetry, and he depicts other poets’ texts as changeable but not his own. The figure of Gallus, however, softens the speaker’s refusal of his text’s vulnerability. If the Gallus of Book 1 is the poet Cornelius Gallus who so influenced Propertius’ poetry, then the susceptibility of his text and his poetics to change provides a potential germ of the idea that the Propertian speaker’s text is also changeable. In Poems 1.4 and 1.7, the speaker acknowledges that other poets criticize his elegies but rejects their criticism and threatens punishment if they persist. The other poet in 1.4 is Bassus, probably the iambographer of Ovid’s Tristia 4.10.47, 62 who has been trying to get the speaker to leave Cynthia: “Why by praising so many girls, Bassus, do you try to make me change and leave my mistress?” (Quid mihi tam multas laudando, Basse, puellas / mutatum domina cogis abire mea? Prop. 1.4.1-2). Cynthia is both the speaker’s girlfriend and his poetry itself, 63 so breaking up with her would mean abandoning his poetic project. The speaker responds to Bassus’ attempts by comparing Cynthia’s beauty favorably with that of Bassus’ girls (Prop. 1.4.5-10) and elaborating additional reasons he loves her (Prop. 1.4.11-14). He then warns Bassus that Cynthia 61 Butler (2011) 63-74, Gurd (2012) 49-103, Martelli (2013) 62 Richardson (1977) 157 thinks the identification is possible, perhaps even likely (“…he may be the iambic poet Bassus linked by Ovid with the epic poet Ponticus…”), while Janan (2001) 36 treats the identification as certain and settled. 63 Book 2 explicitly assimilates Cynthia to the text of Book 1: “Cynthia is read all around the Forum” (...toto Cynthia lecta foro, Prop. 2.24.2). Wyke (2002) 46-77 gives an influential analysis of how the assimilation of Cynthia the fictional girlfriend to Cynthia the poetry plays out in Poems 2.10-2.13. 20 will take revenge for his meddling by badmouthing him to “all the other girls” (omnis alias…puellas, Prop. 1.4.21) so that none of them will have him anymore (Prop.1.4.22). Cynthia will thus inflict erotic failure on Bassus and perhaps also poetic failure, if girlfriends are to his poetry as Cynthia is to Propertius’. The speaker’s coup de grace is to dismiss Bassus as merely jealous of his love affair with Cynthia: “envious man, curb your annoying speech, and leave us to go the way we’re going, side by side” (invide, tu tandem voces compesce molestas / et sine nos cursu, quo sumus, ire pares! Prop. 5.1.1-2 64 ). In Poem 1.7, the speaker issues a similar warning to the epic poet Ponticus: 65 “You beware of scorning my poems in your arrogance: slow Love often comes with a high interest rate” (tu cave nostra tuo contemnas carmina fastu: / saepe venit magno faenore tardus Amor, Prop. 1.7.25-6). The interest (faenus) Ponticus will pay is the suffering and humiliation the speaker details in verses 15-22, which will culminate in Ponticus’ loss of poetic ability (Prop. 1.7.19-20) and forced acknowledgement of the speaker’s poetic superiority (Prop. 1.7.21-1). When the speaker then gloats about Ponticus’ eventual erotic and poetic failure in Poem 1.9, he is vindicating his refusal to allow other poets to interfere with his love affair and his poetry. While the speaker of Book 1 will not put up with other poets’ criticism, he himself sometimes rewrites his poetry. Poem 1.13, in which the speaker celebrates his friend Gallus’ new love affair, echoes and reworks ideas from Poem 1.10 (or vice versa), in which the speaker also celebrates Gallus’ new affair: “…when I saw you dying, Gallus, with your girl embracing you and drawing out your words with long delay! Though sleep was pressing my drooping eyes and the Moon was blushing between her horses in the sky, still I could not pull myself away from your sport: so great was the ardor in the words you two exchanged.” 64 I agree with Goold (1990) ad loc cit. and Heyworth (2007) 23 that Prop. 1.5.1-2 belong at the end of Poem 1.4 rather than the beginning of Prop. 1.5. 65 On the identification with Ovid’s Ponticus at Tristia 4.10.47 see Note 3 above. 21 …cum te complexa morientem, Galle, puella vidimus et longa ducere verba mora! quamvis labentes premeret mihi somnus ocellos et mediis caelo Luna ruberet equis, non tamen a vestro potui secedere lusu: tantus in alternis vocibus ardor erat. Prop.1.10.5-10 and “I saw you languish, clasped all around your neck, and weep, Gallus, with your hand thrown around her for a long time, and long to breath out your soul in wished-for words, and what happened then, my friend, my modesty conceals. I would not have been able to part your embraces, so great was the crazed passion between you two. vidi ego te toto vinctum languescere collo et flere iniectis, Galle, diu manibus, et cupere optatis animam deponere verbis, et quae deinde meus celat, amice, pudor. non ego complexus potui diducere vestros: tantus erat demens inter utrosque furor. Prop. 1.13.15-20 Some of the similarities between these two passages could more precisely be called variations. Variations suggest reworking, 66 though in this case the passages’ order of composition is elusive. Both passages are three couplets long. In the first couplets of both passages, the speaker addresses Gallus (Galle, Prop. 1.10.5 and 1.13.16) and says that he saw Gallus with his girlfriend (vidimus, Prop. 1.10.6 and vidi Prop 1.13.15). The word “dying” (morientem) in 1.10.5 corresponds to the words “languish” (languescere) and “weep” (flere) in 1.13.15-16. The couple embraces in both passages (“having embraced,” complexa, Prop. 1.10.5; “you entirely bound…around your neck,” te totum vinctum…collo, Prop. 1.13.15; and possibly “with hands thrown around her,” iniectis…manibus, 67 Prop. 1.13.16). The second couplets bring stronger variation: The motif of the lovers’ long and intense conversation enters Poem 1.13 in verse 17 (“and long to breath out 66 Cf. Martelli (2013) 8-11, 68-103 67 Richardson (1977) 182: “Enk would take this to mean that Gallus’ arms were embracing the girl; he compares Ovid, Am. 1.4.6 and Meta. 3.389. One might also take it to mean he was covering his face with his hands.” 22 your soul in wished-for words,” et cupere optatis animam deponere verbis 68 ), whereas it occurs in verse 6 of Poem 1.10 (“drawing out your words with long delay,” longa ducere verba mora). This leads the speaker to compress into one pentameter the reason he ought to have looked away (“and what happened then, my friend, my modesty conceals,” et quae deinde meus celat, amice, pudor, Prop. 1.13.18), which takes up the entire couplet in 1.10 (“Though sleep was pressing my drooping eyes and the Moon was blushing between her horses in the sky,” quamvis labentes premeret mihi somnus ocellos / et mediis caelo Luna ruberet equis, Prop. 1.10.7-8). In the third couplets, non tamen and non ego, potui secedere and potui diducere, and tantus and tantus occupy the same metrical positions on the one hand, while on the other hand the adjectives and nouns in the phrases vestro…lusu and complexus…vestros have switched places, and erat has moved from the end of the pentameter in 1.10.10 to straddle the first and second feet of 1.13.20. The sequence of ideas in 1.13 is more-or-less inverse to the sequence of ideas in 1.10. Poem 1.10 begins with the speaker expressing joy at having seen Gallus and his new girlfriend becoming intimate (vv. 1-4), whereas Poem 1.13 ends with this sentiment and with the speaker acting as a praeceptor amoris (vv. 21-36). Second in 1.10 comes the vignette of the speaker watching Gallus and his girlfriend (vv. 5-10), which is penultimate in 1.13 (vv. 15-20). After this vignette in 1.10, the speaker assures Gallus that he will not gossip about his new relationship (vv. 11-14). Immediately before the vignette in 1.13, the speaker tells Gallus that he did not learn about the relationship through gossip or augury but by seeing it himself (vv. 13-14). Finally in 1.10, the speaker proclaims himself a praeceptor amoris (vv. 15-30), but his boasts about his skill in love (vv.15-18) quickly shade into warnings about the unhappy lessons his affair with Cynthia has 68 While the mss read verbis, most critics—Shackleton Bailey (1956) 40, Fedeli (1984) ad loc. cit., Goold (1990) ad loc. cit., et al.—prefer Passerat’s suggestion of labris. They would win me over entirely as well, were it not for the parallels between this passage and Prop. 1.10.5-10, which offer some defense for verbis. 23 taught him (vv.19-30). Conversely, at the beginning of 1.13 the speaker is unhappy and warns that a girlfriend with whom one is head-over-heels in love can impose bitter and humbling lessons (vv. 1-12). The inversion is not perfect—the speaker acts as a praeceptor amoris at the ends of both poems—but it is close. Most importantly for our purposes, the vignette of the speaker watching Gallus and his girlfriend occupies the same place in the logics of both poems. By reworking ideas and phrases in these two passages, the speaker indicates that his poetry can be recycled so long as he remains in charge of it. Poem 1.16 returns to the idea of others appropriating a poet’s words for their own purposes. In this treatment of the exclusus amator trope, the door seems to be quoting the lover’s song verbatim in verses 17-44. Framed by the door’s complaint about the lover (Prop. 1.16.1-16, 45- 8), the song changes in purpose from persuading the door and demonstrating the lover’s grievance with it to persuading the reader and demonstrating the door’s grievance with the lover. The lover, as the door tells us, has pathetically run through the generic commonplaces of the paraclausithyron: He is apparently the most persistent (Prop. 1.16.13-16) of a number of would- be lovers who drunkenly attack the door (he claims in verse 37 that his words have never harmed the door, but the door disagrees in verses 14 and 47-8) and leave garlands and torches by it (Prop. 1.16.5-8); he never leaves his vigil at the doorstep (Prop. 1.16.15-16); he complains that the mistress and her door are cruel (Prop. 1.16.17-20, 29-30, 35ff.) and that it is cold out on the doorstep (Prop. 1.16.21-4); he jealously imagines her in another man’s arms (Prop. 1.16.33-4); and he recalls how many times he has sung to the door, given gifts to it, and kissed its steps (Prop. 1.16.36, 41-4). 69 While the lover’s song is stereotypical, the primary speaker of Book 1, here 69 Copley (1956) 123: “In sum, Propertius has written the definitive paraclausithyron, has made it say all that it could reasonably be expected to say, and has gathered together in one poem the essential facets and features of a literary and erotic tradition some seven hundred years long.” 24 aligned with the “external writer,” 70 innovatively and wittily makes the door the speaker of the poem, so that we see the lover and his song from its perspective. This innovation distinguishes the primary speaker of Book 1 from the lover in the poem. Furthermore, Richardson finds that “the lover does not much resemble the P. of the other poems; he is too gentle, too sweetly melancholy,” while Hubbard notes that “the lover in question has a taste for emotional diminutives that Propertius does not share.” 71 The door gives no indication of who the lover and his beloved are, but they are not the primary speaker and Cynthia. When the door changes the meaning of the lover’s song, then, it does not indicate that the poetry of the primary speaker of Book 1 is open to this sort of hostile appropriation. On the contrary, by distancing himself from the lover in both the style and the content of his poetry, the primary speaker suggests that his poetry should not be treated the way the door treats the lover’s. In Poem 1.18, the speaker imagines how a grove in which he is singing will reproduce his words. Certain features of the grove might efface his speech or use it in ways he does not want, but there are also images of his words lasting. The grove is “deserted” and “silent” (deserta...taciturna, Prop. 1.18.1). There is a breeze, but apparently it is so gentle that he does not worry that it will blow his words away (Prop, 1.18.2). Rather, he seems to think the emptiness and silence are suitable for him to grieve (querenti, Prop. 1.18.1). In verses 3 and 4, the change from an indicative (present general) apodosis to a subjunctive (future less vivid) protasis suggests that the speaker’s first impression of the grove as suitable for grieving shifts, after further inspection, to uncertainty about the choice of location: “Here I may tell my secret sorrows without 70 Hutchinson (2006) 99 describes the “external writer” of Book 4, and Janan (2001) Ch. 5, 6, and 7 also distinguishes between the external writer and the primary speaker of Book 4 (see below). As discussed below, several passages in Book 4 distance the external writer from the primary speaker by means of secondary speakers’ perspectives. My argument here is that in Poem 1.16 the external writer and the primary speaker of Book 1 are aligned and that the secondary speaker (the door) distances them from the tertiary speaker (the lover). 71 Richardson (1977) 189 and Hubbard (1974) 33 25 consequences, / if only the lonely rocks should keep faith” (hic licet occultos proferre impune dolores, / si modo sola queant saxa tenere fidem, Prop. 1.18.3-4). The rocks might break trust with the speaker by echoing his words, which could bring them to unfriendly ears and also could distort them. Verses 19 through 22 express more trust in the grove: “You will be witnesses, if trees have any loves, beech and pine, friend to the Arcadian god. Ah, how many times my words resound under your gentle shade, and ‘Cynthia’ is written in your bark!” vos eritis testes, si quos habet arbor amores, fagus et Arcadio pinus amica deo. a quotiens teneras resonant mea verba sub umbras, scribitur et vestris ‘Cynthia’ corticibus!” Prop. 1.18.19-22 Here the speaker is glad that the trees echo his words and preserve them carved in their bark, and he seems to think that his words are reproduced and preserved unchanged (resonant mea verba and scribitur...’Cynthia,’ Prop. 1.18.21-2). The final couplet of the poem confirms this: “But whatever sort you are, let the forests echo ‘Cynthia’ to me, and let the wasteland rocks not be empty of your name” (sed qualisqumque es, resonent mihi ‘Cynthia’ silvae, / nec deserta tuo nomine saxa vacent, Prop. 1.18.31-2). The worry expressed in verse 4 is forgotten, and speaker anticipates that the grove will repeat his words faithfully. As we have seen, in Book 1 the speaker explores several scenarios in which a poet’s text might be changed, but he only permits his own text to be changed when he himself is the one revising. In contrast, the programmatic first poem of Book 2 allows and even encourages other people to change the speaker’s text. One possible explanation for this development is the popularity of the Monobiblos. Poem 2.24 tells us that “Cynthia is read all around the Forum” (...toto Cynthia lecta foro, Prop. 2.24.2). The popularity of the Monobiblos authorized the speaker of the Elegies to talk about his poetry and himself as famous and potentially even canonical. The first poem of Book 2 also discusses fame and poetic immortality and introduces Maecenas as the 26 speaker’s patron (Prop. 2.1.17, 73). Fame and popularity did in fact create the conditions that speaker acknowledges in Books 2 through 4: Graffiti, tombstones, grammatical treatises, and other canonical poems recycle and change the Propertian speaker’s words for their own purposes. 72 Another possible explanation is the fate of Cornelius Gallus, Propertius’ poetic predecessor. Gallus was made prefect of Egypt in 30 BCE, 73 but in the following years he fell from Octavian/Augustus’ favor. The reasons for his fall are uncertain: Cairns lists as possibilities “financial misconduct involving peculation in Egypt, hubristic behavior there, and verbal, possibly drunken attacks on Augustus.” 74 A senatus consultum (Sue. Aug. 66.2; Dio 53.23) in 27 or 26 BCE condemned Gallus to exile and confiscation, and he committed suicide shortly afterwards. 75 Augustus may have ordered his works destroyed in a damnatio memoriae. 76 That Gallus influenced Propertius is certain. They seem to have been friends, and Gallus may have been Propertius’ patron before he fell and Propertius came under Maecenas’ patronage. 77 The Propertian speaker addresses Poems 1.5, 1.10, 1.13, and 1.20 to someone called Gallus. Some scholars argue that these poems address the poet Cornelius Gallus and that he is the speaker of Poem 1.21. 78 Several scholars have tried to reconstruct Gallan words, phrases, and metrics (what Cairns calls “Gallan verbal complexes” 79 ) as well as themes by comparing passages from surviving Augustan poets and, more recently, the few verses of Gallus from the Qaṣr Ibrîm papyrus. 80 72 See Chapter 3. 73 Stroh (2006) citing Sue. Aug. 66.1; Cairns (2006a) 70ff.; CIL iii Suppl. 14147 (5) 74 Cairns (2006a) 74 75 Sue. Aug. 66; Dio 53.23; Stroh (2006); Conte (1994) 324; Cairns (2006a) 74 76 Ross (1975) 39 believes this is the case. Miller (2004) 74 writes, “No formal book burning or proscription took place,” but he suggests that Gallus’ works may have been informally put away and allowed to disappear. 77 Cairns (2006a) 74-5 78 Cairns (1983) 88-91 and (2006) 70ff. citing and amplifying Skutsch (1901) and (1906); also Janan (2001) 36ff.; Miller (2004) 65ff.; and Keith (2008) 66-7. Against this identification see Syme (1978) 99-103. 79 Cairns (2006a) 82ff. 80 For example, Skutsch (1901) and (1906); Tränkle (1960); Ross (1975); Thomas (1979) 203-5; Janan (2001) 33- 52; Cairns (2006a); and Somerville (2009). 27 Tränkle, for example, proposes as a Gallanism the use of nota to mean “love bite” or “hickey” after habere at the end of a pentameter based on similarities between verses by Propertius, Tibullus, and Ovid. 81 With the benefit of the Qaṣr Ibrîm Papyrus, Cairns argues that the word nequitia, especially when it immediately follows the first foot of a hexameter or pentameter (where it occurs in the Qaṣr Ibrîm fragment), is a characteristic piece of Gallan vocabulary, a “shorthand for the elegiac life of love and social irresponsibility” which Propertius and Ovid continue to use. 82 Less controversial is the idea that Gallus plays an important thematic role in the poetry of Propertius. 83 Gallus’ forced suicide and the elimination, either formal or informal, of his poetry occurred in 27 or 26 BCE, between the publication of the Monobiblos in 29 or 28 BCE 84 and the publication of Book 2 of Propertius around 25 BCE 85 (certainly after Gallus’ death, which Prop. 2.34.91 mentions). The fate of Gallus and his poetry, especially in conjunction with the swift popularity of Book 1, may have helped Propertius develop the idea of his own texts’ vulnerability. Perhaps Propertius already associated the idea of a text’s vulnerability with Gallus, as Book 1 twice presents Gallus’ poetry as susceptible to change. 81 Tränkle (1960) 24 cites Prop. 1.18.8 (“now in your love I am forced to bear a mark,” nunc in amore tuo cogor habere notam); Ov. Am. 1.7.42 (“and for her neck to bear the mark of my fawning tooth,” et collum blandi dentis habere notam) and 3.14.34 (“and that her neck conspicuously bore the mark of a tooth,” collaque conspicio dentis habere notam); as well as Tib. 1.8.38 (“...kisses and set marks on his neck with her teeth,” ...oscula et in collo figere dente notas), which deviates slightly from the pattern. He comments, “Die inhaltlische Übereinstimmuing zwischen Ovid und Tibull, die formale zwischen Ovid und Properz führt wohl auf ein älteres Vorbild, in dem vielleicht von Liebesbissen die Rede war. Ist das richtig, so hat Properz den Vers inhaltlich stark abgewandelt. Bekräftigt wird diese Auffassung durch die Tatsache, daß wir in amore tuo an gleicher Stelle im Pentameter schon bei Cat. 87, 4 finden” (ibid.). 82 Cairns (2006a) 94-5. Cairns (2006a) 94 notes that “the term is absent from the vocabulary of both Catullus and Tibullus” and so believes that “Gallus must have pioneered its use as an elegiac Leitmotiv, and passed it on to the two successors who followed most closely in his wake.” On page 95 he includes a chart with numerous instances from Propertius and Ovid. 83 Janan (2001) 36ff.; Miller (2004) 65ff.; Cairns (2006a) passim; Keith (2008) passim; et al. 84 Neumeister (2006); Conte (1994) 331; Janan (2001) 40 85 Conte (1994) 331 28 In the programmatic Poem 1.1, the speaker presents the mutability of Gallus’ poetry, or at least his poetics, by way of Vergil’s Eclogue 10. The Milanion exemplum is a favorite passage of scholars seeking Gallus in Propertius: “Milanion by fleeing no labors, Tullus, broke the ferocity of Iasus’ harsh daughter. For now he was wandering, out of his mind, in Parthenian caves and he would often go and see the shaggy wild beasts; And he, stricken with a wound from Hylaeus’ club, injured, groaned on the Arcadian rocks. Thus he was able to tame the swift girl: so much are prayers and good deeds worth in love.” Milanion nullos fugiendo, Tulle, labores saevitiam durae contudit Iasidos. nam modo Partheniis amens errabat in antris, ibat et hirsutas saepe videre feras; ille etiam Hylaei percussus vulnere rami saucius Arcadiis rupibus ingemuit. ergo velocem potuit domuisse puellam: tantum in amore preces et bene facta valent. (Prop. 1.1.9-16) Several scholars have discussed verbal ties between the Milanion exemplum and Eclogue 10. 86 Ross, for example, notes that “Mt. Parthenius, in Arcadia, occurs for the first time, and for the only other time in Augustan poetry, in the Eclogues, specifically at that point in the Tenth Eclogue where Gallus himself speaks of hunting (non me ulla vetabunt | frigora Parthenios canibus circumdare saltus, 10.56-7).” 87 Virgil ventriloquizes Gallus in verses 31 through 69 of Eclogue 10: “Gallus now speaks in his own person, imitating himself, as it were; his self-dramatization, the shifting attitudes he adopts—Virgil’s Gallus is evidently imitating the ‘real’ Gallus, for similar passages or motifs occur later in the elegies of Propertius and Tibullus.” 88 As Clausen says, this is “Virgil’s Gallus:” Vergil puts a speech into the mouth of a persona he calls ‘Gallus,’ and the speech may even reflect elements of Gallus’ poetry, but it also suits Vergil’s poetic purposes. 86 For example, Ross (1975) 62ff.; Kennedy (1993) 65-6; Clausen (1994) 291; Keith (2008) 66-8 87 Ross (1975) 63 88 Clausen (1994) 290-1 29 Cairns suggests, for example, that “the pastoral setting in which Gallus is placed... may be no more than the means by which Virgil adapts the persona of Gallus to his own bucolic poetry.” 89 By constructing a ‘Gallan’ poetics that suits his own needs, Vergil creates open textuality for this Gallus-persona. When Propertius’ speaker invokes Vergil’s Gallus in the Milanion exemplum, he imports into Poem 1.1 the idea that other poets may change Gallus’ poetry. Support for this idea comes in Poem 1.20, which also suggests that Gallus’ text can be changed. In this poem, Hercules’ boyfriend Hylas has strayed away from the rest of the Argonauts at Pege, and water nymphs, smitten by his beauty, have dragged him into their pool. Presumably struggling, Hylas makes a noise, and “Hercules from far off calls back to him thrice, ‘Hylas!’: but the breeze carries back the name to him from distant hills” (cui procul Alcides ter ‘Hyla!’ respondet: at illi / nomen ab extremis montibus aura refert, Prop. 1.20.49-50 90 ). The breeze carries back the echo from the mountains, and it does not change the word (Hyla!), but it communicates something entirely different. Whereas Hercules’ call communicates fear, perhaps hope, and a request for information, the echo communicates that Hylas is gone and Hercules has called out in vain. Hercules stands in for Gallus in this exemplum of the dangers that can befall a beloved boyfriend when his lover does not watch over him carefully – Hercules and Gallus’ boyfriends are both endangered by nymphs, and they even have the same name. In verses 49-50, Hercules’ text 89 Cairns (2006b) 84 90 This is Goold’s (1992) text based on Fontein’s suggestion to replace iterat responsa; sed with ter ‘Hyla!’ respondet: at in verse 49. Heyworth also accepts this emendation, and I agree that nomen...refert in verse 50 suggests that Hercules should call out Hylas’ name in verse 49. Shackleton Bailey (1956) , Fedeli (1984), Goold (1992), et al. accept Heinsius’s emendation of fontibus to montibus in verse 50, but Richardson (1977: 206-7) argues that fontibus should stand and that Hylas is calling out his name in response to Hercules, and this is the nomen that the breeze carries back, for which Richardson adduces Theoc. 13.58-60 as a model (those favoring montibus adduce Verg. Ecl. 6.43-4 as a model of the landscape echoing Hercules’ call back to him). Richardson’s reading seems difficult as a narrative: Hylas has fallen and been dragged into and presumably under the water, since in verse 48 his body, not his voice, makes the noise to which Hercules responds in verse 49, so for Hylas to call his name back to Hercules in verse 50, we would have to imagine him resurfacing before ultimately being dragged under again. But if Richardson is right, and the breeze carries Hylas’ voice back to Hercules, the change of speaker still changes what the word communicates. 30 is mutable, and so by analogy is Gallus’. In addition to the figure of Gallus lurking behind Hercules, the high density of Gallanisms connects this poem with the Milanion exemplum. At 1.20.23-4, for example, the infinitive quaerere with a verb of motion (processerat) expresses purpose. Propertius only uses this archaism one other time in the Monobiblos: at 1.1.12, in the Milanion exemplum. 91 Ross and Cairns argue that this and other archaisms, imagery, metrics, and subject matter that link Poem 1.20 and the Milanion exemplum are Gallanisms. 92 Book 1 thus links Gallus with the idea of a mutable text in both its opening poem and the first poem of its three- part coda. 93 Perhaps as he was composing Book 2, inspired by Gallus’ death and the elimination of his works, Propertius gave further consideration to the questions of poetic authority that the deformazione of Gallan poetry in his own poetry 94 and Vergil’s raises. Book 2 In Poem 2.1, the programmatic opening of Book 2, the speaker presents himself as a love poet with Callimachean capabilities and aesthetics and rounds the poem off by considering his death, burial, and legacy. As he imagines his funeral, he invites his patron Maecenas to give a eulogy for him: “and, weeping, toss words such as these to the mute ashes: ‘A harsh girl was this wretch’s fate’ ” (taliaque illacrimans mutae iace verba favillae: / ‘Huic misero fatum dura puella fuit,’ Prop. 2.1.77-8). With the word “such” (talia ) 95 in verse 77, the speaker respectfully avoids dictating words to his patron but rather suggests a model which Maecenas might change or replace. 91 Ross (1975) 78. Ross also notes that Prop. 3.1.3-4 is the only occurrence of this construction in Books 2, 3, and 4 (ibid.) 92 Ross (1975) 74-81; Cairns (2006) 219ff. 93 For poems 20, 21, and 22 as a coda and other elaborate structures in the arrangement of poems in the Monobiblos see Skutsch 1963. 94 Cairns (2006) 106-8 95 As opposed, for example, to “these” (haec). haec atque or et haec or a number of other options could fill in metrically for taliaque. 31 In addition, the model the speaker gives is a pentameter, which would be anomalous as a stand- alone, especially since the tone of the pentameter suggests an epigram, so it might prompt Maecenas to provide his own hexameter to complete the distich. Thus the speaker invites Maecenas to alter or replace his words, which in this instance are the final verse of his poetic manifesto for the book. Readers other than Maecenas might also take up the challenge as a poetic game of the sort Catullus describes in Poem 50. Poems 2.11 and 2.13 present two more occasions for the audience to fill in a missing piece of the speaker’s text. In Poem 2.11, the speaker tries to forswear his love affair with Cynthia and any further poetry about her. He ends the epigram saying, “and the traveller will pass by, snubbing your bones, nor will he say, ‘This dust was a learned girl’ ” [et tua transibit contemnens ossa viator, / nec dicet: Cinis hic docta puella fuit, Prop. 2.11.5-6]. Richardson finds the poem as a whole not just “a perfect epigram” 96 but, “in effect, an epitaph, in form and brevity suitable for the inscription on a tombstone,” 97 while Ramsby takes the poem as an epigram but only the end as an epitaph. 98 Goold and Heyworth, on the other hand, argue that it is not focused enough to be a complete epigram rather than fragmentary or part of another poem. 99 Heyworth also notes the lack “of a vocative, or any other direct pointer to identify [the subject]” and the strangeness of a nominative form of alius in the first verse not beginning a priamel followed by a first person that introduces the speaker’s true idea (cf. Tib. 1.1.1-6 and Hor. Carm. 1.1-10). 100 I find Goold and Heyworth persuasive, and I agree with Ramsby in reading the end of the poem—“This dust was a learned girl,” Cinis hic docta puella fuit, Prop. 2.11.6—as an epitaph. Or rather, I read it as part 96 Richardson (1977) 244 97 Richardson (1977) 245 98 Ramsby (2007) 57 99 Goold (1990) 135; Heyworth (2007) 156-8. Günther (1997) 9-11 does not commit to either position. 100 Heyworth (2007) 157-8 32 of an epitaph: It is missing the hexameter and the first foot and a half of the pentameter, which might lead a reader to try to imagine what the rest of the couplet would have said. In Poem 2.13, the speaker has been imagining his death and funeral again, and when he comes to his burial, he gives instructions for his tombstone: “and let there be two verses: HE WHO NOW LIES AS GRIM DUST, / WAS ONCE A SLAVE OF A SINGLE LOVE” (et duo sint versus: QUI NVNC IACET HORRIDA PVLVIS, / UNIUS HIC QUONDAM SERVVS AMORIS ERAT, Prop. 2.13.35-6). The instructions ask for two verses, but the hexameter of the couplet the speaker provides is incomplete, because the words “and let there be two verses” (et duo sint versus) take up two and a half feet. In the speaker’s morbid fantasy, then, whoever erects his tombstone would need to fill in the missing two and a half feet to complete the two verses and fulfill his wishes – that is, whoever erects his tombstone would need to alter his text. The fact that the following poem includes a complete epigram with a similar introduction (“and under my name will be a poem such as this: / I, Propertius, set before your temples for you, Goddess, these / spoils, I who was received all night as a lover,” taleque sub nostro nomine carmen erit: / HAS PONO ANTE TUAS TIBI, DIVA, PROPERTIUS AEDIS / EXVVIAS, TOTA NOCTE RECEPTUS AMANS, Prop. 2.14.26-8) emphasizes the incompleteness of the epigram in 2.13 by contrast. Poems 2.5 and 2.11 hint more obliquely at the vulnerability of the text. In 2.5 the speaker remonstrates Cynthia for exposing herself to gossip and infamy. After realizing that he cannot leave her and will not beat her up, the speaker contents himself with threatening, “Therefore I shall write something you can’t erase in your lifetime: / ‘Cynthia, mighty beauty; Cynthia, fickle in words’ ” (scribam igitur, quod non umquam tua deleat aetas, / ‘Cynthia, forma potens: Cynthia verba levis, Prop. 2.5.26-7). The phrase “Cynthia, fickle in words” (Cynthia, verba levis) encourages us to follow Wyke in reading Cynthia both as the speaker’s poetry personified and as 33 the scripta puella who is composed of the speaker’s words. 101 Calling Cynthia “fickle in words” (verba levis), then, implies that the speaker’s poetry is changeable, perhaps by those other people with whom Cynthia has committed her indiscretions. Alternatively, Goold and Peiper read “slut” (verna) instead of “words” (verba). If this reading is correct and Cynthia is a fickle slut, then she, as the poetry personified and the scripta puella, is compromised by other men’s use. In either case, verses 1 through 4 accuse Cynthia of sleeping around and so can supply this same idea that Cynthia is compromised by other men. Similar ideas occur in Poem 2.11. The embittered speaker threatens to leave Cynthia, 102 saying, “Others may write about you or you may be unknown” (Scribant de te alii vel sis ignota licebit, Prop. 2.11.1). When the speaker invites others to write about the puella, who is his verbal creation, he invites others to appropriate his verbal creation. The word licebit (literally, “it will be permitted”) emphasizes this notion. The speaker also shows the vulnerability of his text through the motif of other people’s authority to command or judge his text. While he sometimes acknowledges the power of various divinities over his poetry, which is a poetic commonplace, less standard is his occasional concession of poetic authority to other humans. Even as powerful a patron as Maecenas cannot, for example, override the speaker’s recusal from composing martial epic in Poem 2.1. In Poem 2.13, though, the speaker says, “may it be my delight to read my writings in the lap of a learned girl / and to proof them by her faultless ear” (me iuvet in gremio doctae legisse puellae, / auribus et puris scripta probasse mea, Prop. 2.13.11-12). Verse 12 makes the girlfriend an editor, or at 101 Wyke (2002) 46-77 102 Poem 2.11 does not name Cynthia, but it does refer to its addressee as a docta puella (verse 6), which the speaker calls Cynthia at Prop. 2.13.11, and in Poem 2.5 he accuses Cynthia of spreading her affections all over Rome. 34 least an arbiter of tastes, and later the speaker says that she will be his preferred “judge” (iudice, Prop. 2.13.14). 103 A curious feature of Book 2 is that all the passages I have found that acknowledge the vulnerability of the text are in Poems 1 through 13. The longstanding debate over whether Book 2 is really a single book or whether it is really two books that were combined at some point in the manuscript tradition provides one possible explanation. A difficulty with this explanation is, of course, where to divide what is now Book 2. Among recent critics who support the division of Book 2, Richardson thinks that “a division before 2.24.1-10 would be hard to defend, and a point after 2.24.1-10 that will fulfill all requirements is hard to find;” 104 Goold thinks that Poem 2.10 is the end of one book, to which 2.11, 12, and 24A also belong, and 2.13 is the beginning of the next book; 105 Günther thinks that Poem 2.10 is the end of a book and is uncertain about the placement of 2.11 and 2.12; 106 and Heyworth thinks that Poem 2.10 is the end of a book, in which 2.11 also belongs. 107 None of these aligns perfectly with the idea of Poems 2.1-2.13 as a group in which the speaker acknowledges the vulnerability of the text. Furthermore, it is difficult to see why Propertius would drop this idea from a book made up of the remaining poems of what is now Book 2 only to bring it back in what are now Books 3 and 4. Another possible explanation is Butrica’s suggestion that a number of whole poems in Book 2 are interpolations, possibly written by Passennus Paullus. 108 While Butrica does not say 103 Horace, in Satires 1, recommends selecting a very few trusted editors and ignoring other people’s opinions. See below for comparison. 104 Richardson (1977) 8 105 Goold (1990) 19 106 Günther (1997) 10-12 107 Heyworth (2007) 153, 158 108 Butrica (2006) 35: “There also appear to be whole elegies interpolated in Book 2... Given that the verse is clearly classical in meter and style, it would appear that Book 2 has fortuitously incorporated substantial remains of another ancient elegiax poetry book—Silver threads among the gold should the author be Propertius’ descendant Passennus Paullus, praised by Pliny for writing elegies scarcely distinguishable from Propertius’ own (Ep. 9.22).” See Chapter 3 for further discussion. 35 outright that these poems are from the second half of the book, the four he lists as likely candidates (2.15, 2.21, 2.22, and 2.34) all are. If a later author, such as Passennus Paullus, mimicked Propertius’ style and worked with some but not all of his themes (while perhaps importing stylistic traits and themes of his or her own, such as using mythological figures’ names as pseudonyms for addressees 109 ), and if that other poet did in fact write some or all of Poems 2.15 through 2.34, that poet might not have included the idea of the vulnerability of the text in the poems that are now the second half of Book 2. Book 3 Book 3 develops the conflict between the speaker’s aspirations toward monumental poetry and his acceptance of the vulnerability of his text. The book begins with poems that look forward to poetic immortality, but the speaker’s confidence wavers as the book continues, with some poems imagining the occlusion or even the destruction of his texts. In the final poem, the speaker renounces Cynthia, and so also elegy, and prophesies that their beauty will fade with time. Poems 3.1 and 3.2 imagine the speaker’s fixed and canonized text bringing him everlasting fame. In 3.1, the speaker uses a reference to Catullus to indicate that his verse is polished and finished: “Let my verse go forth polished with a fine pumice stone” (exactus tenui pumice versus eat, Prop. 3.1.8). This verse alludes to Catullus’ dedication of his book: “To whom do I give my charming new little book, just polished with a dry pumice stone?” (cui dono lepidum novum libellum / arida modo pumice expolitum?, Cat. 1.1-2). The allusion to Catullus connects the idea of a polished, finished book of poetry with the idea of canonicity, and the following verses reaffirm this connection between fame and polished poetry, “by which [sc. “my verse”] lofty fame raises 109 Butrica (2006) 35 36 me from the ground (quo me Fama levat terra sublimis..., Prop. 3.1.9ff.). The speaker does admit that he will only achieve poetic immortality after he dies: “But what the envious crowd have stolen from me while I live, Honor will return with double interest after my death” (at mihi quod vivo detraxerit invida turba, / post obitum duplici faenore reddet Honos, Prop. 3.1.21-2). But this becomes a segue for him to compare his future fame to Homer’s fame, and not unfavorably (Prop. 3.1.25-38). Poem 3.2 returns the speaker’s thoughts to his world of elegiac love to contemplate poetic immortality in that setting. The speaker compares his power to collect a “crowd of girls” (turba puellarum, Prop. 3.2.10) with the power of three bards from mythology: Orpheus, Amphion, and Polyphemus (Prop. 3.10.3-10). Orpheus and Polyphemus are, respectively, exemplary tragic and comic poet-lovers who have achieved fama, while Amphion is both a bard and a founder of Thebes who built the city’s walls with his song. The series of these mythological references thus weaves together the themes of song, love, and monumentality. Later in the poem, the speaker recombines these three themes in his boast, “My songs will be so many monuments to your beauty” (carmina erunt formae tot monumenta tuae, Prop. 3.2.18). He then compares his poetry with the Pyramids, the Temple of Zeus at Olympia, and the Mausoleum (Prop. 3.2.19-22) before claiming poetic immortality outright: “But the name sought by genius will not perish because of time: the decoration for genius stands without death” (at non ingenio quaesitum nomen ab aevo / excidet: ingenio stat sine morte decus, Prop. 3.2.25-6). In Poem 3.16, on the other hand, the speaker wishes for impermanence and oblivion and imagines the vulnerability of his poetry to physical damage in connection with this wish. 110 He 110 Goold (1990) 281n47 comments, “This poem derives much charm from its blithe air of utter unreality: we are not of course to take it seriously.” On the one hand, I think Goold is not wrong; on the other hand, considering the general fictiveness and the frequent, many-layered humor and irony of the Elegies, as well as the vast scholarship documenting and analyzing these features, it seems unwise to take any of the poems as more or less “real” than the others. Another consideration: latrocinium (banditry) was a significant problem in the Roman world, even in Italy (Ebner 2006), perhaps on a major via publica such as the Via Flaminia (Plin. Ep. 6.25 mentions a traveler who disappeared just past Ocriculum, possibly a victim of bandits). Since latrocinium was a legitimate fear, and the 37 considers whether to risk the danger of a night journey to answer his mistress’s summons, and as he weighs the possibility of dying in an attack, his thoughts turn to his burial. He expects that his mistress will bring the trappings of elegiac love to his grave: “She will bring perfumes for me and will decorate my tomb with garlands, sitting as a guardian at my grave” (afferet haec unguenta mihi sertisque sepulcrum / ornabit custos ad mea busta sedens, Prop. 3.16.23-4). The garlands, again, suggest poetry – perhaps to be sung at his grave or perhaps inscribed on it. The speaker does not, however, want his grave to be easily visible: “let unfrequented ground cover me with trees’ leaves, or let me be buried and entrenched in dunes of unknown sand” (me tegat arborea devia terra coma, / aut humer ignotae cumulis vallatus harenae, Prop. 3.16-28-9). Nor is it clear whether he wants his name on his tomb at all: “it does not please me to have my name in the middle of the street” (non iuvat in media nomen habere via, Prop. 3.16.30). All of this amounts to erasure of his person, and he links his desire for oblivion to the idea that the tombs of elegiac lovers are at risk of desecration: “May the gods keep her from burying my bones in well-traveled ground, where the crowd makes its journey on a busy way! After death the tombs of lovers are thus defamed” (di faciant, mea ne terra locet ossa frequenti, / qua facit assiduo tramite vulgus iter! / post mortem tumuli sic infamantur amantum, Prop. 3.16.27). Richardson believes that the speaker is worried that travelers, having read the inscription and found out that the speaker had died of love, “would then try to avert any such fate from their own heads by apotropaic gestures and spitting as they passed, and the tomb might well become known in an unfortunate way.” 111 Fedeli cites Richardson but disagrees, arguing that an elegiac lover would take passers-by remembering his love as a point of pride. Rather, according to Fedeli, the speaker’s avoidance of the crowd as speaker’s reaction to the mistress’s letter and his self-involved fantasies about the Venus and Amor’s patronage are nothing out of the ordinary for a servus amoris, “reality” and “unreality” do not seem to me to be helpful categories here. 111 Richardson (1977) 386 38 professed in his epitaph in Poem 2.13, his isolation from friends as seen in Poem 1.1, and his retreat into nature as seen in Poem 1.18 indicate that the speaker wants his tomb to be secluded in order to avoid the vulgarity that accompanies popularity in his aesthetic imagination. 112 In either scenario, the poet imagines that passersby would somehow damage his tomb, including the poetry on it, and his sense of his aesthetic self. Erasure is again a concern in Poem 3.23, and in response to this danger the speaker produces a text that he must know will need to be changed. He has lost his writing tablets and assumes that his texts will be devalued and erased: “Well then, my so-learned tablets have been lost, on which equally so many good writings have been lost!” (Ergo tam doctae nobis periere tabellae, / scripta quibus pariter tot periere bona!, Prop. 3.23.1-2). After praising how well the tablets used to speak for him, despite their humble materials, and remembering or perhaps inventing messages girlfriends sent him on them, the speaker imagines their fate: “Wretched me, some miser writes his accounts on them and files them among his drudging ledgers!” (me miserum, his aliquis rationem scribit avarus / et ponit duras inter ephemeridas!, 3.23.19-20). The speaker’s writings may be erased and their materials repurposed, should they fall into the wrong hands. The speaker’s distaste for the “miser” (avarus) and his “drudging ledgers” (duras...ephemeridas) is the same fear that Fedeli reads in Poem 3.16, that contact with the vulgar will destroy his poetry and his aesthetic self. The speaker thus imagines the ‘death’ 113 of some of his texts in another of his texts and in this way incorporates into this present text the possibility of its own ‘death.’ At the end of the poem, he tells a slave, “Go, boy, and quickly post these words on some column, and write that your master lives on the Esquiline” (i puer, et citus haec aliqua propone columna, / et dominum Esquiliis scribe habitare tuum, Prop. 3.23.23-4). This last couplet creates the fiction that 112 Fedeli (1985) 508 113 The word periere, which I translate “have been lost” above, also means “have died.” 39 this text is also impermanent, since it is only a flier to be posted on some column. The last verse acknowledges that the text can be changed – in fact, it instructs the slave to add to the text. Because the speaker uses an indirect statement to tell the slave what to add, he relies on the slave to recompose the words in direct discourse, so they will no longer be his. Book 4 Book 4 explores the extended use of speakers other than the primary Propertian speaker. In some poems, the primary speaker reports and frames the others’ speech within his own speech; other poems appear as unmediated speeches or writings by other speakers, though they too are framed by the rest of the poetry book and so by the primary speaker’s poetics. These compositional techniques present new scenarios through which the primary speaker can explore the vulnerability of his text to change, criticism, and destruction. Poem 4.3 is a letter from a woman called Arethusa to her husband, Lycotas, who is away on campaign. After the salutation she warns him, “But if some part is missing for you to read, this blotting will have been made by my tears” (si qua tamen tibi lecturo pars oblita derit, / haec erit e lacrimis facta litura meis, Prop. 4.3.3-4). Arethusa knows that her text may be destroyed before it reaches her husband. The tears which may destroy it are hers, but she treats them, both grammatically and conceptually, as a source or cause of destruction over which she has no control. Arethusa writes directly and without immediate framing from the primary speaker of the Elegies. 114 In this way, she takes his place, and the possible destruction of her text can be 114 cf. Poem 4.4, in which Tarpeia speaks within the primary speaker’s narration of her story; Poem 4.5, in which Acanthis’ monologue, which takes up most of the poem, is not from Acanthis directly but instead reported by the primary speaker and framed by his introduction and conclusion; and Poem 4.7, in which the primary speaker tells how Cynthia’s ghost appeared to him and reports her speech, which, like Acanthis’ monologue in Poem 4.5, takes 40 understood as analogous to the possible destruction of his. On the other hand, her poem occurs within the larger context of the primary speaker’s book of poems. She is his fiction, so her poetics is internal to his, but since she is fictional, so is the threat her tears pose to the text. Hutchinson, however, notes that “the external reader is highly conscious of the external writer, whose creation of female discourse would have seemed analogous to prosopopoeia...” 115 In performance, 116 then, a speaker might actualize the danger to the text by producing real tears. In Poem 4.5, 117 the speaker complains that Acanthis, a procuress, has been undermining his overtures to the puella by turning his own words against him. After accusing Acanthis of wielding unnatural power through magic, the speaker then quotes her, or perhaps mimics a (stereo)typical speech of hers, in a prosopopoeia that makes up the bulk of the poem. Ventriloquized by the speaker, Acanthis urges the puella to press her lovers, including the speaker, for money and gifts at every opportunity (Prop. 4.5.20-8, 53-62) and coaches her in tricks and techniques for doing so (4.5.29-52). Some of her advice to the puella turns the speaker’s own poetry against him. When she advises the puella, “Always have fresh bite marks around your neck, so he’ll think they were given in back-and-forth quarrels” (semper habe morsus circa tua up most of the poem and is framed by the primary speaker’s introduction and conclusion. Poem 4.11, like Poem 4.3, presents a woman’s speech without framing by the primary speaker. 115 Hutchinson (2006) 99 116 See Chapters 3 and 4. 117 Janan (2001) 85-99 presents a reading of Poem 4.5 that is similar, in a number of ideas and sometimes even in wording, to my reading below. Whereas Janan grounds her reading in Lacanian psychoanalysis, I proceed from the materialism shown in James (2003) and the pragmatism shown in Habinek (1998). These approaches are not necessarily incompatible (Janan’s readings also involve materialism and pragmatism) and sometimes appear to express the same thing in different words—Janan is interested in how the Real pushes up against the Symbolic and cracks it, and I am interested in the material and ideological pragmatics of authorship—but Janan and I are aiming at different targets. Janan’s analysis leads to a beautiful argument that the incoherence of the speaker’s rhetoric and the power of Acanthis’ criticism of him in Poem 4.5 stem from “the lack of any signifier for Woman’s desire in a system implicitly centered on the signifier ‘Man,’ ” so that Acanthis’ voice (like those of Cynthia in 4.7 and Cornelia in 4.11) is necessary if Propertian elegy is to present “the whole story” of a woman (Janan 2001: 98-9). I am trying to understand the conditions under which the speaker says he creates his texts and under which they exist as texts, and I am trying to understand what these conditions mean for the speaker’s understanding of himself as an author. 41 colla recentis, / litibus alternis quos putet esse datos, Prop. 4.5.39-40), for example, she plays on the jealousy that the speaker has displayed in previous poems. 118 Similarly, when Acanthis says, “pretending to have a vir 119 also raises the price: use excuses! With a night’s delay, love returns greater” (et simulare virum pretium facit: utere causis! / maior dilata nocte recurret amor, Prop. 4.5.29-30), she seems to know that these tricks will work. Perhaps Acanthis knows this from other experiences, but perhaps she suggests these strategies because of what the speaker reveals in earlier poems. In Poems 2.23 and 3.14, he longs for love without these difficulties: “she doesn’t brook delay, if someone should want to approach; she won’t put you off, nor will she rant and demand what your stingy father often complains about giving, nor will she say: “I am afraid – hurry and get up now, I beg; you wretch, today my vir is coming from the countryside.” nec sinit esse moram, si quis adire velit; differet haec numquam, nec poscet garrula, quod te astrictus ploret saepe dedisse pater, nec dicet: ‘Timeo, propera iam surgere, quaeso; infelix, hodie vir mihi rure venit.’ (Prop. 2.23.16-20) and “nor fear heavy punishment from a severe vir. You yourself may speak of your affairs with no go-between: 118 e.g. Prop. 1.8, 1.11, 2.5, 2.6, 2.8, 2.9, 3.8 119 James (2003) 41-2 argues that the vir whom the puella sets up as a rival to the elegiac lover should not be understood as a husband but rather as a man who claims some other right to her exclusivity, such as concubinage or merces annua: “The primary organizing—not to say generative—principle in the matter of the rival is whether or not he separates lover-poet and puella by means of money, which is an issue of elegy rather than any pressing historical problem. Generically, marital status and rights are not very interesting to elegy, though they have been of great concern to scholars. Elegy focuses not on social or legal barriers but on persuasion, which by its nature requires independence of mind in its audience. That is, the independent will of the puella (combined with her constant financial needs), rather than anything as mundane as social or marital status, is the true problem for the elegiac lover-poet... the primary point about any Other Man in the docta puella’s life is that he operates as an obstacle, like a comic blocking character, between lover-poet and puella. There are two basic ways for him to perform this function: either by immediate payment or by some kind of rights, through marriage, concubinage, or perhaps a contract like the merces annua (yearly payment). Elegy differentiates these two types of rivals: the latter is usually designated a vir—which means primarily “man” but is commonly used of husbands... Though the lexicon of marriage infuses the passages dealing with this type of rival, in my view it is a red herring, applied not to describe a legitimate, socially recognized marital union but to characterize the lover-poet as standing outside an existing relationship.” 42 there is no discouragement from long delay” (nec gravis austeri poena cavenda viri. nullo praemisso de rebus tute loquaris ipse tuis: longae nulla repulsa morae (Prop. 3.14.24-6) Poem 3.14 is a fantasy about Spartan girls, while Poem 2.23 shows that the speaker is willing to pay for uncomplicated love, since the girl he describes is a prostitute in the Forum. These passages demonstrate that the tricks of the docta puella drive him to desperation–fantasy and unsophisticated prostitutes are poor substitutes for the docta puella—and Poem 2.14 reveals that relief from desperation might loosen his pocketbook. Here, the elegiac lover gushes that, after being scorned for some time (“I used to be called more worthless than a dried-up pond,” dicebar sicco vilior esse lacu, Prop 2.14.12), he is so excited by a night of long-awaited love that he is inclined to give great gifts to the Temple of Venus (“I shall affix great gifts to your columns, Cytherea,” magna ego dona tua figam, Cytherea columna, Prop. 2.14.25), which suggests he might be willing to spend money on the puella as well. Acanthis, then, advises the puella to take advantage of weaknesses in the elegiac lover that can be found in his earlier poetry. She turns his own poems against him to her advantage. The speaker also has Acanthis criticize his modus operandi as an elegiac lover. 120 In verses 55-6, 121 he has Acanthis use his own words from 1.2.1-2 to ask his mistress, “What good does it 120 Prop. 1.8.37-40 gives a clear statement of the elegiac lover’s m.o.: “Though he gave big gifts, though he was going to give [more] big gifts, still she did not greedily flee my embrace. I was able to sway her, not with gold, not with Indian pearls, but with the tribute of flattering poetry” (quamvis magna daret, quamvis magna daturus, / non tamen illa meos fugit avara sinus. / hanc ego non auro, non Indis flectere conchis, / sed potui blandi carminis obsequio). See James (2003) passim for an extended and detailed discussion of elegiac lovers’ techniques of persuasion. 121 Many critics mark verses 55 and 56 as interpolated (Knoche 1936: 16-28 and 52, Barber 1953: ad loc. cit., Fedeli 1965: 164, Richardson 1977: ad loc. cit. and 445, Heyworth 1986: 209, Goold 1990: ad loc. cit., Butrica 2006: 34, Hutchinson 2006: ad loc. cit. and 148, Tarrant 2006: 63; Heyworth 2007 does not mention the problem). Richardson (1977) 445 argues that “P. is not given to quoting himself, even when he has an obvious opportunity.” The self-quotation is indeed almost unique in Propertius’ Elegies (though see above on the correspondences between Prop. 1.10 and 1.13). But, as I have described, and as Shackleton Bailey (1953) 19 writes against this very argument in Knoche (1936) 16-28, the speaker is not quoting himself directly but rather having the lena quote him. On this point, Shackleton Bailey (1953) 19 also writes, “It is not surprising that a parallel phenomenon should be hard to 43 do you, my life, to go around with an ornate hairdo and to move the gossamer folds in a Coan dress?” (quid iuvat ornato procedere, vita, capillo / et tenues Coa veste movere sinus?, Prop. 4.5.55-6 and Prop. 1.2.1-2). In Poem 1.2, the speaker is trying to convince his beloved that she does not need fancy hairdos (1.2.1), Coan silks (1.2.2), Orontean perfume (1.2.3), and other “exotic gifts” (peregrinis...muneribus, Prop. 1.2.4). Instead, he would have her focus on her natural beauty (1.2.5-24) and intellectual cultivation (1.2.26-30)—which, conveniently, do not cost him money— and consider luxuries tedious (1.2.31-2). If the speaker is not entirely earnest in the sentiments he expresses here, the hope that his beloved will perceive him to be earnest is at least a conceit of the poem. In the context of Acanthis’ advice to the puella, the quoted couplet is a sarcastic, rhetorical question that mocks any would-be lover who would ask it. The preceding couplet—“look at the gold, not what sort of hand brings the gold! When you’ve heard verses, what will you get but words?” (aurum spectato, non quae manus afferat aurum! / versibus auditis quid nisi verba feres?, Prop. 4.5.53-4)—makes the quoted couplet a pointed reference to elegiac poet-lovers such as the speaker in Poem 1.2. To reinforce this point, Acanthis then opposes the speaker’s frequent, find, for the situation is peculiar,” and on the argument that vv. 55-6 do not follow the normal pattern of being marked by a word such as dixit, he writes, “...Reitzenstein has pointed out a Propertian exception (III, 6, 19) and I will add a Virgilian one: Aen. IV, 416” (ibid.). It is true that Coa veste (Prop. 4.5.56) is similar to ...Coae...vestis (Prop. 4.5.57) and so might be a sign that vv. 55-6 were at one time a marginal comment on v. 57. Shackleton Bailey (1953) 20 argues that there are many such repetitions in the Elegies, but he adduces no parallel example to this particular situation (repeated verse including words repeated in surrounding lines), perhaps because repeated verses are uncommon to begin with, let alone in this unusual poetic context. On the other hand, as Hutchinson (2006) 150 notes, in verse 67 the speaker quotes Acanthis’ phrase “I have seen” (vidi ego) from verse 61, throwing her words back in her face, which supports the use of this technique of sarcastic quotation in verses 55-6. The phrase vidi ego, furthermore, only occurs in one other place in Propertius, Prop. 1.13.14-15, where it begins two verses (but a pentameter and the following hexameter rather than two hexameters), as in Poem 4.5. These verses in Poem 1.13 introduce a passage that reworks or is reworked by a passage in Poem 1.10 (see above). When Acanthis picks up the phrase in Poem 4.5, she signals that she knows the speaker’s text can be reworked. By throwing it back in her face six verses later, the speaker tries to reclaim the phrase but in doing so acknowledges that Acanthis has left her mark on it. I do not know whether a definitive answer on the authenticity of vv. 55-6 is possible, and I must acknowledge that it is convenient for me if they are genuine. To my mind, the couplet is effective and even funny in its context, its sarcasm seems appropriate for Acanthis, and the arguments against it are as inconclusive as the arguments for it. I agree with Reitzenstein (1936) 94-108, Shackleton Bailey (1953) 17-20 and (1956) ad loc. cit., Luck (1955) 430-33, and Fedeli (1984) ad loc. cit., who believe the couplet genuine. Coutelle (2015) ad loc. cit. seems to follow this line of reasoning, as well, in writing, “La lena, en citant les vers mêmes du poète, inscrit ses paroles dans une polémique littéraire.” 44 cajoling suggestion that he has made the puella immortal through poetry with the reality that she will lose the bloom of youth all too soon (Prop. 4.5.59-62). The prosopopoeia of Acanthis allows the speaker, as a poet, to present a view of his elegiac ‘reality’ that he himself, as an elegiac lover, would rather not acknowledge. Hutchinson writes, “A.’s speech often implies [the speaker’s] errors in love. The dynamic central figure of the poem thus presents the world of elegy in a distorting mirror. The interaction between lover-poet and lena enables a vivacious and unexpected exploration of all P’s previous work.” 122 Perhaps, though, it is not Acanthis who presents a “distorted” vision of the world of elegy but rather the speaker himself. James, for instance, who reads elegy “from the viewpoint of the docta puella,” 123 argues that elegy as a genre is the fictional lover’s attempt to persuade a fictional courtesan to have sex with him for free 124 and that the lover’s persuasion entails such distortions of elegy’s fictional ‘reality’ as pleading poverty though in fact he is a member of the leisure class (probably an equestrian, which would require a census evaluation of 400,000 sesterces) as well as a client of Maecenas, 125 and yet he accuses the puella of unreasonable greed when in fact she is a professional offering her services for sale and needs the money to survive. 126 Acanthis’ critique impinges on the speaker’s fantasy, but since she, like everything else in the elegiac ‘reality,’ is the speaker’s creation, her critique is also the speaker’s confession of his own bad faith. 127 Acanthis interrogates the speaker’s poetry; she repurposes his ideas and even his exact words for her own persuasive ends and turns them against him. By elaborating a specific example of how his texts can be criticized and showing that, even when his words are quoted verbatim, their meanings can be 122 Hutchinson (2006) 138 123 James (2003) 35 124 James (2003) 35-41 and passim 125 James (2003) 36, 258 126 James (2003) 71 127 Janan (2001) 89 arrives at a similar reading and even uses the same phrase, “bad faith” – le mot juste, it seems. See Janan (2001) 106-9 on the reappearance of these ideas in Poem 4.7. 45 changed, the speaker incorporates the vulnerability of his text to criticism and change into the text itself. The speaker presents a similar scenario in Poem 4.7, when he recalls how Cynthia’s ghost appeared to him and berated him as he slept after her funeral. As with Acanthis’ speech in 4.5, so in 4.7 the speaker reports Cynthia’s speech so vividly as to suggest prosopopoeia and literalize the rhetorical figure mortuos ab inferis excitare. 128 Among her accusations and demands, Cynthia tells the speaker, “and whatever verses you made in my name, / burn them for me: stop praising me” (et quoscumque meo fecisti nomine versus, / ure mihi: laudes desine habere meas, Prop. 4.7.77-8). She then instructs him to clear her tomb of ivy (Prop. 4.7.79-82) and to inscribe on her grave an epitaph, which she dictates: Here write a poem worthy of me on my column, but a short one, so that traveler hurrying from the city may read it: HERE IN TIBURTINE GROUND LIES GOLDEN CYNTHIA: PRAISE HAS COME TO YOUR BANK, ANIENUS. hic carmen media dignum me scribe columna, sed breve, quod currens vector ab urbe legat: HIC TIBURTINA IACET AUREA CYNTHIA TERRA: ACCESSIT RIPAE LAUS, ANIENE, TUAE. (Prop. 4.7.83-6) Cynthia supplies the epitaph to be inscribed verbatim. In verse 83, Hutchinson (2006) prints “this poem” (hoc carmen) instead of “here...a poem” (hic carmen); 129 if he is correct, then hoc further emphasizes that the words should be exactly as Cynthia gives them. By calling the epitaph “a poem worthy of me” (carmen...dignum me) Cynthia implies that the speaker’s poems are unworthy of her, and so she demands that he destroy his words about her and replace them with hers. 128 See Lausberg (1960) 411-12 for the classical tradition and theories of this device. See Dufallo (2007) 13-35 for Cicero’s use of the device and 78-94 for a reading of Propertius’ adaptation of it in relation to Cicero’s use, including here in Poem 4.7. 129 He argues that the sense of the passage requires hoc, which occurred in the lost ms Π (Hutchinson 2006: 187) and is found in mss F L P D V (Fedeli 1984: ad loc. cit.). Barber (1953), Richardson (1977), Fedeli (1984), and Goold (1990) print hic. 46 Cynthia’s ghost also repurposes both the speaker’s fear from Poem 1.19 that she will forget him after he dies and neglect his grave and his promises to continue loving her even after he dies. Verbal parallels between their opening couplets signal a dialogue between this poem and Poem 1.19: “So ghosts do exist: death does not end all things, and a ghastly shade flees the vanquished pyre.” Sunt aliquid manes: letum non omnia finit, luridaque evictos effugit umbra rogos. (Prop. 4.7.1-2) and “Now I do not fear sad ghosts, my Cynthia, nor do I begrudge the fate owed to the final pyre. Non ego nunc tristes vereor, mea Cynthia, manes nec moror extremo debita fata rogo. (Prop. 1.19.1-2) Both first verses discuss “ghosts” (manes), and both second verses end in forms of the word “pyre” (rogus). Furthermore, the opening from Poem 4.7 reads like a response to or reevaluation of the thought at the opening of Poem 1.19, reached after a new and jarring experience, which the rest of Poem 4.7 describes. Later in Poem 1.19, the speaker frets, “but that my grave should lack your love, this fear is harder to bear than death itself” (sed ne forte tuo careat mihi funus amore, / hic timor est ipsis durior exsequiis, Prop. 1.19.3-4), and “how I fear that you should scorn my grave, Cynthia, and that unjust Love should drag you from my dust and force you against your will to dry your falling tears” (quam vereor ne te contempto, Cynthia, busto / abstrahat a nostro pulvere iniquus Amor, / cogat et invitam lacrimas siccare cadentes, Prop. 1.19.21-3). Cynthia’s ghost turns this worry back on the speaker as an accusation: “Who, then, saw you bowed over my grave, who saw your black toga grow hot with tears?” (denique quis nostro curvum te funere vidit, / atram quis lacrimis incaluisse togam?, Prop. 4.7.27-8). In Poem 1.19, the speaker fears that Cynthia will 47 not be present at his grave and that she will not weep for him; in Poem 4.7, Cynthia’s ghost charges the speaker with failing to do both of these things for her. 130 Near the end of the poem, Cynthia’s ghost turns the speaker’s fears and his promise from Poem 1.19 into an erotic threat. In the earlier poem, he worries that she will neglect his grave after he dies, and the words he uses for “death,” “grave,” “dust,” and “funeral” include fatum (2), funus (3), exsequiae (4), and pulvis (6). These words can also mean “corpse” or “remains,” 131 and Williams and Papanghelis argue that we should read them here with these meanings in addition to their more common meanings. 132 We can also understand the speaker, then, to be worrying lest his corpse, from which his love for Cynthia will never fade (Prop. 1.19.5-6), be without her love. In Poem 4.7, her ghost threatens to give him too much of what he has asked for. She leers at him, “Let other women have you now: soon I alone will hold you: you will be with me, and I will rub away your bones with my bones in intercourse 133 ” (nunc te possideant aliae: mox sola tenebo: / mecum eris, et mixtis ossibus ossa teram, Prop 4.7.93-4). Whereas he fears in Poem 1.19 that he will lack her love after he dies, she tells him in Poem 4.7 that when they are both dead she will make love to him with possessive, destructive intensity – she will literally love him to tiny little bits. 130 Cf. Janan (2001) 103 131 on fatum v. TLL 6.1.359.49ff.; on funus v. TLL 6.1.1605.36ff.; on exsequiae v. TLL 5.2.1848.71ff., though the examples are later than Propertius – TLL 5.2.1848.48ff. gives the usage of exsequiae here, at Prop. 1.19.4, as an example of the word meaning “death” by metonymy, so it is not a great stretch to understand it as “corpse” by metonymy, especially in this context; on pulvis used de reliquiis mortuorum fere cumbustis v. TLL 10.2.2631.72ff. 132 Williams (1968) 766ff. and Papanghelis (1987) 10-13 citing Williams 133 mixtis can mean simply “mingled,” but the context here suggests the erotic meaning, because Cynthia’s ghost is sexually jealous in the previous verse, and because teram also has sexual connotations (Adams 1982: 183-4, 219). For the erotic meaning of misceo, see Adams (1982) 180-1 and OLD “misceo” definition 4c. Propertius uses misceo to indicate sexual intercourse at 1.13.21 and has other bones mingling, possibly with sexual undertones since they are Antigone and Haemon’s bones, at 2.8.21-4. Cf. erotic uses of misceo at Cic. Div. 1.29.90, Verg. A. 7.661, and Ov. M. 13.866. For further examples see TLL 8.0.1080.5-8, 80.1081.47-55. 48 Cynthia’s ghost thus converts the speaker’s anxieties from an earlier poem into complaints, accusations, and a threat to use against him. Fabre-Serris reads Cynthia’s speech as a last will and testament, for both Cynthia the puella and Cynthia as Propertian elegy, with “une dimension méta- poétique.” 134 Cynthia’s ghost uses the speaker’s own words to point out the incoherence between those words and his actions, that is, to call him a self-involved hypocrite. She demands that he replace his poetry about her with her own, which is “worthy” (dignum, Prop. 4.7.83). The speaker frames her harangue with an introduction and a conclusion describing the arrival and departure of the ghost. Her ability to use the speaker’s own words against him thus becomes part of his poetics, just as Acanthis’ does in Poem 4.5. In Poem 4.8, the speaker invites the Appian Way to revise his account of a quarrel with Cynthia. The speaker’s version of what happened is a paragon of misogynist romantic comedy. Cynthia, he says, went to Lanuvium with a dandified playboy, so she was at fault in the first place (Prop. 4.8.15-26). The speaker’s response to her infidelity—throwing a party with Phyllis and Teia, whom he sketches as loose lushes who lacked Cynthia’s sophistication (Prop. 4.8.27-42) and ultimately failed to distract him (Prop. 4.8.45-8)—was admittedly not his finest moment, and he couldn’t even really bring himself to get Cynthia out of his head and enjoy the sexy ladies at hand (he’s such a sensitive guy). When Cynthia returned early, her entrance was thrilling (Prop. 4.8.49- 52), and the excitement stirred him from his spiraling lethargy (Prop. 4.8.53-4). Her response to his party was unreasonable, considering that she had cheated first, “but she’s sexy when she’s mad” (sed furibunda decens, Prop. 4.8.52). The speaker gleefully describes her catfight with Phyllis and Teia: Cynthia is magnificent, while Teia turns slapstick in her panic, chaos ensues, and finally the two lesser women flee pathetically (Prop. 4.8.51-62). The language then turns 134 Fabre-Serris (2009) 157 49 mock-epic, mock-legalistic, and mock-religious to tell how “Cynthia rejoiced in her spoils and, victorious, charged back” (Cynthia gaudet in exuviis victrixque recurrit, Prop. 4.8.63-4), overpowered and enslaved him (Prop. 4.8.70), imposed harsh terms of peace (Prop. 4.8.71-82), and cleansed the house and the speaker of Phyllis and Teia’s defilement (Prop. 4.8.83-6). 135 But all’s well that ends well, and Cynthia and the speaker end up in bed together (Prop. 4.8.88). Is this version of events honest? On the heels of Poem 4.7, in which Cynthia’s ghost accuses the speaker of perfidy and exposes his falsehoods, Poem 4.8 revisits the speaker’s same old tricks, 136 but now there is a third party to bear witness, the Appian Way. After describing the serpent rite of Lanuvium, the speaker starts in on the main point of his tale, beginning with Cynthia’s infidelity. He seems to call the Appian Way as a witness to his cause when he says, “Appian Way, tell, I beg, in how great a triumph she drove, with you as a witness, as her wheels shot across your stones” (Appia, dic, quaeso, quantum te teste triumphum / egerit effusis per tua saxa rotis, Prop. 4.8.17-18), and then he gives his own description of Cynthia’s ride (Prop. 4.8.21-4). 137 That the speaker all but tells the Appian Way what to say and then describes what happened himself, though, smacks of insecurity. Poem 4.7 has taught us to question the speaker’s truthfulness, and there is no reason to believe that in Poem 4.8 the Appian Way would confirm his account. Inviting the Appian Way to speak is in keeping with the compositional habit in Book 4 of introducing other speakers’ perspectives. But the invitation here is reminiscent of Poem 2.1, where the speaker says to Maecenas, “and, weeping, toss words such as these to the mute ashes” (taliaque illacrimans mutae 135 Cf. Habinek (1998) 128: “…she is situated as the upper half of an asymmetric master-slave arrangement, chiefly to be castigated as such.” 136 Komp (1988), Janan (2001) 114-15, and others have explored this and other dynamics in the the pairing of Poems 4.7 and 4.8. 137 Whether vv. 19-20 are entirely spurious (Hutchinson 2006: 194), are part of an early draft that should not have been kept (Shackleton Bailey 1956: 254-5), or actually belong after v. 2 (Richardson 1977 ad loc. cit. and Goold 1990 ad loc. cit.), they do not fit between v. 18 and v. 21. 50 iace verba favillae, Prop. 2.1.77). In Poem 2.1, the speaker tells Maecenas to say “such words” (talia...verba) as the ones he says in the following line. In Poem 4.8 he invites the Appian Way to describe a vignette, and he makes the invitation with an indirect question introduced by quantum. This construction, like talia...verba in Poem 2.1, invites the other speaker (here the Appian Way, there Maecenas) to insert its own words into his text or to replace his words entirely. Half of his brief description occurs within a preterition: “For I remain silent about her smooth-plucked playboy’s silken carriage and his dogs braceleted on their Molossian necks” (serica nam taceo vulsi carpenta nepotis / atque armillatos colla Molossa canes, Prop. 4.8.23-4). With the preterition, the speaker pretends that this part of the description does not happen. This trick aims to obfuscate what the speaker’s rhetoric is doing, but if we take him at his word (“I remain silent,” taceo), then we should ignore the description he supplies – another will need to be filled in. In this section of the poem (verses 15-26), the speaker tries to put Cynthia at fault for his misbehavior in the rest of the poem. Should the Appian Way contradict this version of events, it would undermine the speaker’s argument and his put-upon, hen-pecked, nice-guy persona. The extensive and sophisticated challenges to the speaker’s persona in Poems 4.5, 4.7, and 4.8 produce a distance between the speaker and what Hutchinson calls the “external writer.” 138 This distance allows us to read the challenges to the speaker’s persona as part of the external writer’s poetics. Conclusion In Book 1 it seems that the speaker of the Elegies is not ready to admit that what he sees happening to other people’s texts will happen to his own. When he begins in Book 2 to acknowledge his text’s susceptibility to change, criticism, repurposing, and even destruction, he 138 Hutchinson (2006) 99. See also Janan (2001) Ch. 5, 6, and 7. 51 frees himself to explore new (and, as we shall see, more realistic) ideas about his text’s relationship with the world in which it exists. When the speaker incorporates vulnerability into the way his poetry characterizes itself as poetry, he changes his relationship, as an author, to his poetry. Chapter 2 situates this vulnerable poetics among ancient and modern literary-critical concepts to understand what this new relationship entails and achieves. 52 Chapter 2: The Vulnerable Text as a Literary Trope Introduction Ancient literary criticism boasted a plethora of technical terms, including plenty to describe how authors reused each other’s words and ideas, but as far as I know it did not have a term for how authors anticipated such reuse. The speaker of Propertius’ Elegies does not acknowledge the vulnerability of his text to reuse in just one way, because other people might interact with his text in any number of ways. Instead he imagines scenarios in which others change and reuse his texts in different ways, for different reasons, and with different results, though he does repeat some scenarios, such as the incomplete epitaphs in Poems 2.1 and 2.13. Nevertheless, when viewed together, as in Chapter 1, these scenarios align with each other through the motif of anticipating receptions of the Elegies. The vulnerability of their text to more and less damaging receptions is thus a trope in the Elegies, and so we may ask how this trope relates to other tropes involving literary appropriation. This chapter, then, has two main goals: To investigate how the Propertian speaker’s vulnerable textuality relates to ancient ideas about relationships between texts, such as echo and anaklasis, 139 and to situate this understanding of the text as vulnerable to various receptions within modern scholarly discussions about intertextuality, allusion, deformazione, the openness of a text, and ritualization. Ancient Literary-Critical Terms: Echo and Anaklasis By the time Propertius was composing his Elegies, echo was a familiar literary figure. Hellenistic poets including Apollonius, Theocritus, and Nicander had written versions of the myth 139 I do not discuss μίμησις (mimesis) and imitatio as such. Echo and anaklasis are more specific sub-categories of them, and these more specific terms relate correspond more precisely to certain ways in which the Propertian speaker talks about his texts as vulnerable than the broader terms do. 53 of Hylas with its echo motif, for example, and by 29 BCE it had apparently become so common in Latin literature that Vergil could gripe, “Who hasn’t told of the boy Hylas?” (cui non dictus Hylas puer? Verg. Geo. 3.6). 140 Accius’ Phinidae may present the earliest surviving occurrence of the word echo in Latin literature: I. Hac ubi curuo litore latratu Unda sub undis labunda sonit. II. Simul et circum magna sonantibus Excita saxis suauisona echo Crepitu clangente cachinnat. (frs. I and II Ribbeck) “Here where on the arcing beach Waves under waves roll and roar … All around, resounding rocks Raise a smooth-toned echo. Crashing, clapping, cackling.” (tr. adapted from Boyle’s translation below) The repeated sounds in this passage—alliterations, assonances, and recurring words—simulate a chorus of mingling echoes. The text is uncertain, though, and the word “echo” is Ribbeck’s suggestion. 141 Even without the word itself, though, the passage treats the idea of an echo as a literary figure: The account of waves roaring and slapping against rocks on a shore describes a series of echoes, which the sound-effects of the words produce mimetically. We are on firmer ground with the verb “resound” (resonare). Verses by Ennius, Pacuvius, Accius, Catullus, and Macer use the verb resonare to describe echoes. 142 Among these verses, a fragment of a play by Pacuvius associates echo with a wedding song: “the age-mates sound the wedding song; the hall resounds with musical noise” (hymenaeum fremunt / Aequales, aula resonit 140 For recent surveys and readings of the myth of Hylas in Greek and Latin literature, see Mauerhofer (2004) and Heerink (2015). 141 Warmington reads Hac ubi curvo litore latratu / unda sub undis labunda sonit, / simul et circum stagna sonantibus / excita saxis saeva sonando / crepitu clangente cachinnant. Klotz reads simul et circum megna tonantibus / excita saxis saeva sonando in fr. 2 vv. 1-2. Boyle (2006) 115 reads, hac ubi curuo litore latratu / unda sub undis labunda sonit ... / simul et circum stagna tonantibus / excita saxis saeua sonando / crepitu clangente cachinnant, following Klotz’s text with a couple of changes, and translates, “Here where on the arcing beach / Waves under waves roll and roar … / All around, thundering rocks / Stir pools to savage sounds. / Crashing, clapping, cackling.” 142 Enn. Ann. VI. 363; Pac. trag. fr. I(4), V(2) Ribbeck; Acc. praet. fr. II(4), trag. fr. I(1), XI(8); Cat. Carm. 11.3; Mac. fr. 7.1 Blänsdorf 54 crepitu musico, frag. I[4] Ribbeck). In the poetry of Vergil, Propertius’ senior colleague, echoes present an allegory for the citational habit of Alexandrianizing poetry. At the start of Eclogue 1, for example, the goatherd Meliboeus speaks of Tityrus’ intentionally echoing song in five verses replete with echo-effects: Tityre, tu patulae recubans sub tegmine fagi silvestrem tenui musam meditaris avena: nos patriae fines et dulcia linquimus arva. nos patriam fugimus: tu, Tityre, lentus in umbra formosam resonare doces Amaryllida silvas. (Verg. Ecl. 1.1.1-5) “You, Tityrus, reclining under the cover of a beech tree, ponder the woodland muse on a slender reed: I am leaving the borders of my country and my pleasant fields. I am fleeing my country: you, Tityrus, relaxing in the shade, are teaching the woods to resound that Amaryllis is beautiful.” The name “Tityrus” itself nearly replicates its first syllable in its second and so produces the distorted repetition of sound sometimes associated with echoes. 143 Other sounds also repeat in verses 1-2: tu and patulae (which also echo the second syllable of Tityre), recubans and sub, tegmine and fagi, silvestrem and tenui, and silvestrem…musam meditaris. The start of verse 3, nos patria fines, recurs at the start of verse 4, nos patriam fugimus. The vowel u is sprinkled through the middle of verse 4, while a vowels surround it: nos patriam fugimus: tu, Tityre, lentus in umbra. Three instances of the u vowel in verse 4 are the syllable tu/ty, reprised from verse 1. The words tu, Tityre in verse 4 repeat the first two words of verse 1 but invert their order, and verses 4 and 5 repeat and invert the order of ideas from verses 1 through 3 – (a) you, Tityrus, are singing in the woods, (b) while I am leaving my home. The vowels o and a intertwine in verse 5 along with the liquid consonants, l and r, and the nasal consonants, m and n: formosam resonare doces Amaryllida silvas. The word silva echoes and subtly changes the last three syllables of 143 cf. Verg. Ecl. 6.43-4; Ov. Met. 3.499-501 55 Amaryllida. The beginning of the Eclogues thus associates song with echo as it introduces a book of poems that echo ideas and wording from earlier poetry. Vergil’s Georgics also use the verb resonare to talk about the echoing of wolves’ howls (1.485-6) and birdsong (3.338), and Eclogue 6 plays with an echo-effect in the meter of verse 44, a passage which Propertius 1.20 echoes. 144 Propertius’ speaker tells us that he has thought about echo as a figure and act of speech and writing. As discussed in Chapter 1, in Poem 1.18 the speaker concludes that the grove will preserve his words faithfully as an echo. The grove itself is a metaphor for literary material with a literary past: In verse 20, the naming of trees—“beech and pine, friend of the Arcadian god” (fagus et Arcadio pinus amica deo)—recalls the listing of trees in Ennius’ lumberjack scene (Ann. 175-9 Sk.), which itself looks back to the tree-felling at Iliad 23.114-20. 145 After naming the trees in the grove, the speaker describes how they have become the physical media on which he composes and publishes his poetry: “ah, how often my tender words echo under your shadows, and Cynthia is written in your bark!” (a quotiens teneras resonant mea verba sub umbras, / scribitur et vestris Cynthia corticibus! Prop. 1.18.21-2). Finally, just as the literary keyword, silva, 146 appears in the last verse of the Ennian passage, the Propertian speaker reserves it for his last couplet (Prop. 1.18.31-2). The speaker’s poetry echoes in the silva. It thus joins the rest of the literary material to be found there, but the speaker insists that the echoes will repeat his poetry without changing it. Bonadeo writes that “reproductivity and repetitiveness, however, are not sufficient components to fully characterize the acoustic reflection which, in its physical manifestations as in its literary 144 See below and Chapters 1 and 3. 145 Hinds (1998) 11-14 gives a famous intertextual reading of Hom. Il. 23.114-20, Enn. Ann. 175-9 Sk., and Verg. Aen. 6.179-82. 146 On silva as a literary metaphor, translated from the Greek ὕλη, see Coleman (1988) xxii-xxiii; Hinds (1998) 11- 14; Heerink (2015) 8; OLD s.v. 5 and, for example, Cic. Orat. 12: “for indeed all fullness and, as it were, the silva of speaking has been drawn from them [i.e., philosophers’ debates]” (omnis enim ubertas et quasi silva dicendi ducta ab illis est). 56 implications, is endowed with an intrinsic responsivity and dialogicity.” 147 She points out that sounds can become lost or garbled, especially in a polysyllabic echo, and that “even a formally unchanged repetition can overturn the sense of a statement, when situated in a different context or register than that to which it belongs.” 148 If the speaker in Poem 1.18 worries at first that the rocks may fail to “keep faith” (tenere fidem, Prop. 1.18.4) with him by distorting his words as they echo them, his anxiety matches Bonadeo’s assessment of echoes. But by the end of the poem he has rejected this assessment and decided instead that the echoing grove will reproduce his words verbatim. From the speaker’s perspective, there is no dialogue at this point. He pronounces his expectations with hortatory or perhaps wishing subjunctives (resonent and nec...vacent, Prop. 1.18.31-2), and he ignores any potential differences between what his words communicate when he says them and what they will communicate when the grove echoes them in his absence. Similarly, in Poem 1.1, the speaker says of whoever does not heed him now, “alas, with how much pain will he repeat my words” (heu referet quanto verba dolore mea, Prop. 1.1.38); the person who fails to take the speaker’s advice will echo his words and communicate the same thing with them that the speaker communicates. Propertius 1.20, on the other hand, dramatizes how repetition can change the context and register of words. Hercules responds to the sound of Hylas being dragged into the water— “Hercules from far off calls back to him thrice, ‘Hylas!’: but the breeze carries back the name to him from distant hills” (cui procul Alcides ter ‘Hyla!’ respondet: at illi / nomen ab extremis montibus aura refert, Prop. 1.20.49-50 149 )—and the natural world echoes what he says, apparently without change. But the pathos in Propertius’ telling arises from the difference between what 147 Bonadeo (2004) 245 (my translation) 148 Bondadeo (2004) 246 (my translation) 149 See Footnote 39 in Chapter 1 on the text of this couplet. 57 Hercules communicates when he calls out Hylas’ name and what the echo from the mountains communicates. Furthermore, Vergil’s version of the drowning of Hylas, an immediate predecessor (perhaps the immediate predecessor) of Propertius’ version, shows the distortion of sounds in echoes: “to these [tales] he adds by what spring the sailors called out to deserted Hylas, so that all the shore resounded, ‘Hylas, Hylas’” (his adiungit, Hylan nautae quo fonte relictum / clamassent, ut litus ‘Hyla, Hyla’ omne sonaret, Verg. Ecl. 6.43-4). In verse 44, the a in the second Hyla is shortened by correption, while the a in the first Hyla remains long. The sound-effect of the words Hy ̆ lā, Hy ̆ lă is thus the petering out of an echo, and both words are sonically diminished from the accusative Hylan in the previous verse. 150 Propertius’ version of the Hylas story follows Vergil’s and brings to mind the sound-effect that Vergil creates by repeating Hyla. Like Vergil’s before it, Propertius’ telling exploits Hylas’ name to signal allusions, but Propertius also ties in the setting of the episode. The name Hylas is linked etymologically and aetiologically to the word for grove (Ὕλας/ὕλη; Hylas/hylē), 151 and in Propertius’ version Hylas is abducted and drowned in a woodland spring (Prop. 1.20.35-46), whereas in Eclogue 6 there is no mention of woods. Propertius advertises the figura etymologica by ending consecutive verses with Hylae and silvae (Prop. 1.20.6-7) and ending the verse that precedes them with the phrase “not dissimilar in name” (non nomine dispar, Prop. 1.20.5). 152 150 See Wills (1996) 45ff. for discussion and classification of instances and types of gemination in Latin poetry, and see Hinds (1998) 5-6 for an intertextual reading of this passage along with Verg. Ecl. 3.78-9 (“Phyllis I love before all others, for she wept at my leaving and said, ‘a long farewell, farewell, beautiful Iollas,’” Phyllida amo ante alias; nam me discedere flevit / et ‘longum, formose, vale, vale,’ inquit, ‘Iolla,’) and Ov. Met. 3.499-501 (His last utterance as he looked into the familiar water was this: ‘Alas, boy beloved in vain!’ and the place sent back his very same words, and ‘farewell’ having been said, ‘farewell,’ said Echo too,” ultima vox solitam fuit haec spectantis in undam: / ‘heu frustra dilecte puer!’ totidemque remisit / verba locus, dictoque vale, ‘vale,’ inquit et Echo). 151 Hunter (1999) 262-3: “…an aetiology for a (real or believed) ritual practice of Mysia, cf. Strabo 12.4.3 ‘still to this day a festival is celebrated among the Prusians; it is a mountain festival (ὀρειβασία), in which they march in procession and call Hylas, as though making their expedition to the forests (ἐπὶ τὰς ὕλας) in quest of him’.” 152 Petrain (2000) 409-11 also notes the etymological play, here, and in 410n5 teases out implications of the phrase “not dissimilar in name” (non nomine dispar). 58 Citing intricate interactions with the Eclogues in Propertius 1.20, Petrain and Heerink argue that the Propertian speaker’s warnings to Gallus at the beginning and end of the poem to protect his loves (tuos…servabis amores, Prop. 1.20.51) are both erotic and metapoetic: Vergil has abducted and distorted Gallus’ poetic material in the Eclogues just as the nymphs abducted Hylas and drowned him, but the joke is that, in the course of warning Gallus not to let it happen again, Propertius has committed the very same theft from him and Vergil. 153 Whereas Poem 1.18 refuses to acknowledge that echoes in a grove might change the Propertian speaker’s words, Poem 1.20 shows how an echo can change what words communicate by changing their speaker and, by recalling Vergil’s version of the Hylas episode, how an echo can change the words themselves. As discussed in Chapter 1, the Propertian speaker does not acknowledge in Book 1 that others may change or destroy his text; accordingly, in Poem 1.18 he denies that his own words will be changed, while in Poem 1.20 he allows Hercules and Gallus’ words to be changed. Bonadeo argues that the Augustans and even some later Romans did not separate echo and poetic memory conceptually, 154 and the Elegies’ treatment of what is often called ‘quotation’ 155 does suggest similarities with echoes. Propertius may have known one type of quotation by its Greek name, anaklasis (ἀνάκλασις). As defined by Quintilian and Rutilius, anaklasis fits within the category Conte calls “reflective allusion.” Quintilian, a few generations after Propertius, says that “bordering on this [i.e., the opposite of paronomasia] is what is called anaklasis, a contrary 153 Petrain (2000); Heerink (2015) 83-112 154 Bonadeo (2004) 256 155 Conte (1986) 60 argues that the term quotation should be reserved for instances in which “the new verbal segment does not rework the old one dialectically; it simply inserts the only text statically within itself.” Instead, Conte would call the upcoming examples “reflective allusion[s]” (67). He defines this type of allusion thus: “...reflective allusion...involves intentional confrontation; the rhetorical figure that corresponds to it is the simile. Two items are juxtaposed and compared (here no ‘interpolating’ entity replaces the proper one on the model of a metaphor). The analytic character of comparison—especially if it is a simile—is recognizable here, by contrast with the synthetic character of a metaphor” (Conte 1986: 67). 59 signifying of the same word,” (cui confinis est quae ἀνάκλασις dicitur, eiusdem verbi contraria significatio,’ 156 Quint. Inst. 9.3.67). He gives as an example, “When Proculeius complained that his son was ‘waiting for’ his death, and the son said that he truly was not waiting for it, ‘no indeed,’ Proculeius answered, ‘I ask that you wait for it’” (cum Proculeius quereretur de filio, quod is mortem suam ‘expectaret’, et ille dixisset, ‘se non vero expectare’, ‘immo’, inquit, ‘rogo expectes,’ Quint. Inst. 9.3.67). In writing about figures of speech, Quintilian had recourse to P. Rutilius Lupus’ Latin abridgment of a treatise on this topic called Peri Schēmatōn by Gorgias of Athens (Quint. Inst. 9.2.102), who taught the younger Marcus Tullius Cicero in 44 BCE. 157 Rutilius writes, “Anaklasis: This figure generally occurs when that which was said by the other person is taken not with the intention which is understood but with another or an opposite” (Ἀνάκλασις. Hoc schema fieri solet, cum id quod ab altero dictum est, non in eam mentem quae intellegitur, sed in aliam aut contrariam accipitur, Rutil. 1.5.17). The testimony of Gorgias, Rutilius, and Quintilian indicates that anaklasis was used and discussed in Propertius’ time and that its technical definition as a rhetorical figure remained consistent. Sometimes the Propertian speaker uses anaklasis to criticize other people’s words, just as he uses echo to enter into a poetic contest with Gallus and Vergil and claim victory in Poem 1.20. In Poem 2.16, a praetor has returned to Rome from Illyria, and the speaker fears his competition for Cynthia’s attentions. The speaker compares himself unfavorably with men of action like the praetor, saying, “‘But you should be ashamed.’ Oh yes, I should be ashamed! Unless perhaps, as they say, shameful love generally has deaf ears” (‘At pudeat.’ certe, pudeat! nisi forte, quod aiunt, / turpis amor surdis auribus esse solet, Prop. 2.16.35-6). With the first pudeat in verse 35, the 156 This is the Teubner text of Buchheit’s 1971 revision of Radermacher. Bonadeo (2004) 252 cites this passage with ἀντανάκλασις instead of ἀνάκλασις but says they are interchangeable. 157 Weißeberger (2006) 60 speaker imagines someone chastising him. Wills gives this verse as an example when he writes that “full quotation is occasionally seen (especially in satire) when the narrator imagines the words of an interlocutor and then responds for his own part.” 158 The defiant invocation of the proverb makes it clear that the second pudeat signifies the opposite of the first and even mocks it. Sometimes, though, the speaker uses anaklasis or other kinds of repetition to acknowledge other people’s criticism of his own words. In Poem 4.5, Acanthis quotes the speaker’s words from Poem 1.2 to criticize him and his words: “What good does it do you, my life, to go around with an ornate hairdo and to move the gossamer folds in a Coan dress?” (quid iuvat ornato procedere, vita, capillo / et tenues Coa veste movere sinus?, Prop. 4.5.55-6 and Prop. 1.2.1-2). 159 More than two books of elegies and a decade between their respective dates of publication separate verses 1- 2 of Poem 1.2 and verses 55-6 of Poem 4.5, but the piquancy of the latter couplet derives from the reader’s recollection of the former. Through the reader’s memory, the verses from Poem 1.2 become part of Poem 4.5 but still as verses from Poem 1.2. Despite the physical and temporal separation of the couplets, which are greater than in Quintilian’s example, verses 55-6 of Poem 4.5 form an anaklasis in dialogue with verses 1-2 of Poem 1.2. This anaklasis is, in effect, an echo. The repetition in Poem 4.5 of the words from Poem 1.2 produces the same effect on them as the echo’s reflection of Hylas’ name in Poem 1.20. Whereas in the anaklasis in Poem 2.16 the speaker repeats the words of an imagined interlocutor and means something different by them, in Poem 4.5 the ventriloquized interlocutor repeats the words of the speaker and means something different by them. It is by moving in this second direction—from the interlocutor back to the speaker—that anaklasis in the Elegies can show other people criticizing the speaker’s words. 158 Wills (1996) 343 159 See Chapter 1 for discussion of open textuality in this couplet as well as questions of the couplet’s authenticity. 61 Another kind of echo occurs in Poem 4.7. As argued in Chapter 1, Cynthia’s ghost paraphrases the speaker’s fears and his promise from Poem 1.17 but uses them to make accusations and an erotic threat against him. This is not anaklasis as Rutilius and Quintilian define it, because Acanthis does not repeat the speaker’s words verbatim. She does repeat certain key words, though: “grave” (funere) at 4.7.27 repeats “grave” (funus) at 1.19.3 in the same metrical position, 160 and “who [saw] your black toga grow hot with tears” (atram quis lacrimis incaluisse togam) at 4.7.28 corresponds to “forced you against your will to dry your falling tears” (cogat et invitam lacrimas siccare cadentes) at 1.19.23. Furthermore, the speaker has already signaled a dialogue between these two poems through verbal parallels between their opening couplets. This signaling primes the reader to recognize when Cynthia repurposes key words later in Poem 4.7. By reprising these key words, Cynthia prepares the speaker and the reader to understand her threat, “Let other women have you now: soon I alone will hold you: you will be with me, and I will rub away your bones with my bones in intercourse” (nunc te possideant aliae: mox sola tenebo: / mecum eris, et mixtis ossibus ossa teram, Prop 4.7.93-4), as an answer to the speaker’s promise at 1.19.5-6 that “not so lightly has the boy cleaved to my eyes that my dust would be empty of our forgotten love” (non adeo leviter nostris puer haesit ocellis, / ut meus oblito pulvis amore vacet). In Poem 4.7, then, there is some repetition of sounds—funere/funus and lacrimis/lacrimas—from Poem 1.19. According to Bonadeo, this is a requisite component of the classical Roman understanding of echo and “the associative field of echo.” 161 In addition to sounds, Poem 4.7 repeats ideas from Poem 1.19. Through Cynthia’s ghost in Poem 4.7, the Propertian speaker explores how literary echoes might someday criticize his poems. 160 Or as close to the same position as possible, given their different lengths: funere is the fifth-foot dactyl of the hexameter in Prop. 4.7.27, and funus is the first two syllables of the fifth-foot dactyl of the hexameter in Prop. 1.19.3. 161 Bonadeo (2004) 256 62 Intertextuality and Allusion The proper understanding of intertextuality has been contested among classicists since the 1980’s. Kristeva coined the term in the late 1960’s, and the building-blocks of Kristevan intertextuality are the following ideas: 1) “The three dimensions of textual space…are writing subject, addressee, and exterior texts,” so “the word in the text belongs to both writing subject and addressee…and is oriented toward an anterior or synchronic literary corpus;” 2) “the addressee is included within a book’s discursive universe only as discourse itself” and so “fuses” with the anterior or synchronic literary corpus; thus 3) “each word (text) is an intersection of word (texts) where at least one other word (text) can be read,” so 4) “any text is constructed as a mosaic of quotations; any text is the absorption and transformation of another. The notion of intertextuality replaces that of intersubjectivity, and poetic language is to be read as at least double.” 162 Crucially for Kristeva, the intertextuality of every word makes readable within the word resonances from its literary context and the rest of its cultural/historical context: “The word as minimal textual unit thus turns out to occupy the status of mediator, linking structural models to cultural (historical) environment, as well as that of regulator, controlling mutations from diachrony to synchrony, i.e., to literary structure.” This means, in Edmunds’s lucid summations, that “in Kristeva’s sense, intertextuality marks the point at which the text enters history,” 163 and that a text is therefore, for Kristeva, nothing more than “the product of its historical circumstances.” 164 It is with this last idea that Kristeva’s intertextuality becomes less congenial to readers of Latin poetry, and so Edmunds seeks a model that can comprehend intertextuality “at the level of 162 Kristeva (1980) 65-6, which is an English translation of Kristeva (1969). Emphasis original. 163 Edmunds (2001) 10 164 Edmunds (2001) 14 63 diction and meter” as well as the level of “language (‘language as dynamic mark, a moving gram’).” 165 Kristeva’s brand of materialism, in other words, makes it difficult for her to conceptualize literary language as Wiedergebrauchsrede (“language that can be reused” as opposed to Verbrauchsrede (“language that is used up in the currency of everyday living”), 166 and it is difficult to understand in her terms what it means to call song “ritualized speech” 167 or “speech made special through meter, diction, accompanying bodily movement, or performance in ritual context.” 168 Edmunds understands intertextuality as produced entirely by the reader; the author has no place in this concept of intertextuality. It is the reader who connects parts of a text with the literary and non-literary past and present of the text. 169 Edmunds further argues that there is no semiotic marker of intertextuality. 170 Rather, intertextuality occurs when a reader sees a relationship between a text (T1) and an earlier text (T2), namely a quotation (Q1 – including “allusion, reference, echo”) in T1 of a part (Q2) of T2 which imports some of the context (C2) of Q2 into the context (C1) of Q1. 171 Edmunds’s understanding of intertextuality is not compatible with the Propertian speaker’s vulnerable textuality. Edmunds might say that the Propertian speaker has misidentified the source of his anxiety. Rather than worrying about other speakers, such as Acanthis and Cynthia’s ghost, changing his words, the Propertian speaker should worry about how readers of Acanthis and Cynthia’s ghost will change his words when they encounter them in the women’s texts. In Poem 4.5, the speaker is in fact worried about how the puella might change her behavior toward him if she listens to Acanthis, but he directs his spite at the madam not the courtesan. In Poem 4.7, the 165 Edmunds (2001) 12 166 Conte (1986) 40 167 Habinek (2005) 4 168 Habinek (2005) 1 169 Edmunds (2001) 163 170 Edumunds (2001) 153-7, 165-6 171 Edmunds (2001) 133, 137-8 64 speaker is the ghost’s only immediate reader. Repeating her words so that others may read them is a strategic error on his part – he would have no need to worry about losing control of his own words if he had not reproduced hers. In Edmunds’s terms, the Propertian speaker has misunderstood the referential relationships between texts in Poems 4.5 and 4.7. But, if we leave aside the vexed textual tradition of the Elegies for a moment, Propertius the historical poet did in fact compose Poems 4.5 and 4.7, and the Propertian speaker constructs himself as the author of the poems. This means that Propertius and his speaker have caused Acanthis and Cynthia’s ghost to quote Poems 1.2 and 1.19. Acanthis and the ghost quote the earlier poems to criticize them and the Propertian speaker – the new contexts of the quoted verses and ideas in Poems 4.5 and 4.7 defend this interpretation. When the Propertian speaker sets Acanthis and the ghost up as authors within the poems 172 and his girlfriend and himself up as their respective audiences, he guarantees that the audiences will recognize and understand the quotations and that the authors know they will. Edmunds’s intertextuality and Propertian vulnerable textuality are thus incompatible because their views of authorship are irreconcilable. 173 As we will see, Conte’s notion of cooperation between author and reader and Hinds’s analysis of the author function in intertext and allusion have more in common with the vulnerable textuality of the Elegies. Conte rarely uses the word ‘intertextuality’ in his work on poetic memory and allusion. He writes in a footnote, “The term ‘intertextuality’ is now widely accepted… Although these scholars [of the Tel Quel group] often extend the ideological import of the notion too far for the concrete needs of the philologist, who is a less 172 Acanthis “turns” and “versifies” (versat, Prop. 4.5.63) the mind of the speaker’s girlfriend, and Cynthia’s ghost dictates verses to the speaker (Prop. 4.7.85-6). 173 For a modern critique of intertextuality based on the conception of authorship it requires, see Calame (2004), especially p. 224: “C’est dire que l’étude de ses formes doit s’inspirer d’une analyse des discours sensible à la dimension culturelle et anthropologique d’une poétique de la citation.” 65 abstract analyst of texts, we should probably accept the term and seek to redefine it. I consider it equivalent to the less technical ‘poetic memory’—a strategic working equivalence suited to our needs.” 174 As Conte says in this note, he prefers the terms ‘poetic memory’ and ‘allusion,’ which he understands as “the activation of poetic memory.” 175 Despite his mild tone, Conte’s shift from intertextuality to poetic memory is polemical. It allows him to theorize allusion as a trope that relies on “complicity between poet and reader,” 176 in which “poetry, once it is sure of its survival and status, starts playing with its readers, offering them participation ‘in aenigmate’ (in mystery). But readers can attain to that source of pleasure only if they have acquired sufficient poetic ‘doctrine.’” 177 Conte differentiates between two types of allusion that such cooperations between author and reader can produce, “integrative allusion” and “reflective allusion.” 178 The vulnerable textuality of Propertius’ Elegies adds a third party to Conte’s idea of the “complicity between poet and reader.” 179 Conte argues that the author of an alluding text creates the conditions for this complicity: “...the author establishes the competence of the Model Reader, that is, the author constructs the addressee and motivates the text in order to do so. The text institutes strategic cooperation and regulates it.” 180 The two parties here are the author of the alluding text and the reader. By acknowledging the susceptibility of his text to allusions by later authors, the Propertian speaker tries to join the fun of Conte’s “complicity between [alluding] poet and reader [of the allusion]” as a third wheel, the author of the work alluded to. He makes himself complicit in the game the alluding author and the reader of the allusion will share by authorizing it ahead of time and incorporating it into his notion of his text as a text. The trouble for the 174 Conte (1986) 29n11 175 Conte (1986) 66 176 Conte (1986) 56 177 Conte (1986) 57 178 Conte (1986) 67 179 Conte (1986) 56 180 Conte (1986) 30 66 Propertian speaker is that the alluding author and the reader of the allusion do not need his authorization. From his perspective, he cannot enter the hypothetical scene as an equal participant with the alluding author and the reader. If we take Acanthis’ allusions to the speaker’s earlier poetry in Poem 4.5 as an example, we see that the speaker is an outsider with his nose pressed up against the window, the butt of the Acanthis’ (the alluding author’s) joke. Even in a scenario less hostile to the Propertian speaker, he is still at most a marginal participant in the relationship between the alluding author and the reader of the allusion. In Poem 2.11, he invites other poets to write about Cynthia and try to have sex with her: “Others may write about you, or you may be unknown: let him praise you, who sows his seeds in barren ground” (scribant de te alii vel sis ignota licebit: / laudet, qui sterili semina ponit humo, Prop. 2.11.1-2). In this poem, the speaker tries to renounce Cynthia and so sets himself up to have no involvement in an alluding poet’s love affair with Cynthia and the reader’s enjoyment of it beyond his status as Cynthia’s previous poet-lover. At the end of Poem 2.11, the speaker sneers, “and scorning your grave, a traveler will pass by and will not say, ‘This dust was a learned girl’” (et tua transibit contemnens ossa viator, / nec dicet: ‘Cinis hic docta puella fuit,’ Prop. 2.11.6). The speaker thus attempts to position the end of his poem as the future of any relationship another poet might hope to have with Cynthia. Barchiesi has explored some effects of precisely the temporal relationship that the speaker tries to create at the end of Poem 2.11: “But what happens when the older tradition enters a new text as a view of the future?” 181 But what the speaker envisions at the end of Poem 2.11 is still future and hypothetical for him, as well. Cynthia’s death is not yet written, so the speaker’s prediction does not constrain the fun another poet might have with Cynthia. It is not 181 Barchiesi (1993) 334 67 until Cynthia is actually dead in Poem 4.7—two books later—that a poet who alludes to her knows anything about her future from the time of Poem 2.11. Alternatively, another poet might write about Cynthia as she was before her relationship with the Propertian speaker. Such a poet would be writing in a genre other than elegy, because in elegy, as the Propertian speaker constructs the genre, to write about a girlfriend is to be involved in an erotic relationship with her in a static present. 182 Since, in elegy and elegiac love, “the beloved must remain, at least for the duration of their affair, a young woman, untouched, when her desirability is asserted, by the ravages of time,” 183 another elegist cannot have a relationship with Cynthia in which she has a future with Propertius, except on the occasions when her future is invoked to castigate and reject her. If another poet should write about Cynthia’s life before Propertius in a genre other than elegy, that poet could subject Cynthia and Propertius to some of the rich possibilities of irony that Barchiesi has explored in allusions in the Heroides. 184 In this case, again, the Propertian speaker’s acknowledgement of the vulnerability of his text cannot earn him any participation, beyond his part as the author of the text alluded to, in the game of signification that an alluding author and a reader of the allusion produce together. Different ways in which the speaker acknowledges the vulnerability of his text do, however, correspond to Conte’s two types of allusion, integrative and reflective. 185 In some passages in the Elegies, the speaker invites or instructs other characters to reformulate his words. At the end of Poem 2.1, for example, he asks Maecenas to say “words such as these” (talia…verba, Prop. 2.1.77) over his grave and provides and the pentameter of an elegiac couplet, which 182 Gardner (2013) 59-83 examines the Propertian speaker’s struggles with this static present and with its maintenance, while much of the rest of the book considers how Propertian speaker constructs Cynthia’s relationship with time and, by means of this, his own. 183 Gardner (2013) 14 184 Barchiesi (1993) 185 See above. 68 Maecenas may or may not use. Ultimately, the speaker’s goal in this passage is to have Maecenas memorialize him as an elegiac poet and lover. If Maecenas does so, whether or not he uses the speaker’s own words, then he will create “a condensation of two voices [his and the speaker’s] in a single image whose sense lies in an interdependence of meanings that become subjectively equivalent.” 186 In Poems 4.5 and 4.7, on the other hand, Acanthis and Cynthia’s ghost make reflective allusions to Poems 1.2 and 1.19. They create “intentional confrontation[s]” of meanings, in which the what the speaker meant in Poems 1.2 and 1.19 is “juxtaposed and compared” with what his words mean from Acanthis and the undead Cynthia’s perspectives three books later. 187 These alignments are tidy, in part, because the Propertian speaker and Conte largely agree about how authors allude to other authors. Charting a path between the positions of Edmunds and Conte (though his book appeared three years before Edmunds’s, and his path lies closer to Conte), Hinds presents yet another vision of the roles of authors in allusion and other kinds of intertext. Hinds readily allows that, with the term ‘allusion,’ we assume the intentions of the alluding poet, which in fact we cannot know, but he also urges, “let us not accept that as the end of the story.” 188 Instead, he advocates “embrac[ing] the fact (i.e. rather than occluding it) that one of the most persistent ways in which both Roman and modern readers construct the meaning of a poetic text is by attempting to construct from (and for) it an intention-bearing authorial voice.” 189 He therefore recommends that we retain allusion as a term and a model, while keeping in mind the fact that we ourselves construct the authorial intentions we read with it, so that we remain sensitive to “the possibilities of tendentiousness, 186 Conte (1986) 67 187 Conte (1986) 67 188 Hinds (1998) 47-8 189 Hinds (1998) 49 69 quirkiness or sheer surprise which add spice to the allusive practices of real authors.” 190 This modification of Conte’s understanding of allusion and authorship does not yield substantially different readings of the vulnerable textuality of Propertius’ Elegies than Conte’s version did above. 191 Where Hinds’s modification of Conte leads him on the subject of a literary tradition, however, helps to clarify some effects of Propertius’ vulnerable textuality. Catullus and Vergil treat Laevius and Ennius as old or archaic in order to present themselves as new, 192 and Propertius does the same thing to Gallus and Vergil in Poem 1.20. 193 “To generalize the lesson,” says Hinds, “proclamations of one poet’s newness are invariably proclamations of another poet’s oldness.” 194 An author is only ever old or archaic from the vantage point of another author who is positioning himself or herself in a literary tradition. By incorporating the anticipated reception of the Elegies by other authors, who are also readers, into his own poetics, the Propertian speaker lays claim to their eventual archaicness even as they are brand new and so forestalls it. The Elegies cannot be made truly archaic if anticipating and pre-claiming archaicness is part of their novelty, because treating them as archaic only reaffirms the terms of their freshness. 190 Hinds (1998) 50 191 I do not mean to say here that Conte believes that an author’s intention is knowable, as Conte (1986) 29 makes it clear that he does not. Nevertheless, he does involve the author’s intentions in his models of allusions in Conte (1986) 56-66. Hinds (1998) 50n64 argues that “Conte’s tendency to allow the author a little more presence in practice than he does in theory is a real strength” (emphasis original). For a very recent exploration of intertexuality and allusion that arrives at almost the same place as Hinds (1998) but from a position that more decidedly favors intertextuality over allusion, see Lewis (2016). 192 Hinds (1998) 52-63 and 76-91 193 Petrain (2000); Heerink (2015) 83-112 and see above. 194 Hinds (1998) 55 70 Deformazione Let us consider the Propertian speaker’s acknowledgement of the vulnerability of his text in relation to another critical term from recent discussions of Propertius and intertextuality: Cairns presents deformazione, the assimilation and/or misrepresentation of an author’s work by another author, as an obstacle to reconstructing Gallus’ poetics from their reception in later poets such as Vergil and Propertius. 195 In Poems 4.5 and 4.7, Acanthis and Cynthia’s ghost have subjected the primary speaker’s poetry to deformazione. In the poems from which the women have appropriated verses and ideas (Poems 1.2 and 1.19), the speaker professes his passion for Cynthia. Whether or not he is sincere in everything or anything he says, Cynthia must either believe he is sincere or find his insincerity charming or in some other way persuasive if his rhetorical strategies are to succeed at their goal, “to create unpurchased access to her bedroom.” 196 When Acanthis quotes the speaker at Prop. 4.5.55-6, she misrepresents his poetry, as he sees it, by reversing the persuasive aim of his words. Acanthis demonstrates to the puella that the speaker’s sincerity, even if believed, should be irrelevant to her decisions, and that his insincerity is not charming but inconsiderate. 197 In Poem 4.7, Cynthia’s ghost assimilates the speaker’s promise from Poem 1.19 to love her even after he dies in order to criticize him for being faithless to her. She then criticizes the promise itself by taking it to a perverse extreme in her erotic threat. 198 From the speaker’s point of view, this reductio ad absurdum distorts his promise for the sake of the ghost’s argument. 199 By 195 Cairns (2006) 104-9. Cairns (2006) 82n57 advocates leaving deformazione in Italian “because it can cover both misrepresentation and assimilation,” whereas “the closest English term, ‘deformation’, does not do so, and it has other, unwanted overtones.” He cites Perkell (1996) as an exemplary analysis of deformazione. 196 James (2003) 150 197 James (2003) passim esp. 71-107 discusses how the economic ‘realities’ of the elegiac puella affect her reception of the lover’s attempts at persuasion. 198 See Chapter 1. 199 Hutchinson (2006) 187 (ad loc. cit.): “The overtones of lis and perago suggest a legal attack (cf. OLD s.v. perago 12). Event and poem deny the narrator a reply; but he implies her unfairness. She concludes, rather than being forced to stop; she and nature refuse him an embrace (contrast 94). Complaint is so like her (1.3.18, 6.7-11, 3.6.18, 35).” 71 reporting these instances of deformazione and framing them within his own speech—that is, by showing how others have already subjected his poetry to deformazione—the speaker acknowledges that it may happen again. Vulnerability and Eco’s ‘Openness’ In some passages of the Elegies, however, the speaker predicts not the reuse of his texts but their total or partial destruction. In Poem 3.23, the speaker imagines his writing tablets in the hands of an accountant who erases the poems on them to use them as ledgers; in Poem 4.3, a woman called Arethusa explains to her husband that she is weeping for his prolonged absence as she writes to him and that her tears may blot out parts of her letter; in Poem 4.7, Cynthia’s ghost demands that the speaker destroy the poems he has written about her so that her own account of herself can replace them. Ramsby argues that the Augustan elegists display “an increasing awareness of the function of poetry as a means of establishing and preserving identity in an age of political and social upheaval.” 200 The Propertian speaker is anxious about the preservation of the self he is building in his verses, both when he acknowledges that his poetry may be erased or destroyed and when he acknowledges that it may be repurposed or changed. Eco has theorized two modes in which a text can be susceptible to reworking by people other than its original author: “However, the examples considered in the preceding section propose an ‘openness’ based on the theoretical, mental collaboration of the consumer, who must freely interpret an artistic datum, a product which has already been organized in its structural entirety (even if this structure allows for an indefinite plurality of interpretations). On the other hand, a composition like Scambi, by Pousseur, represents a fresh advance. Somebody listening to a work by Webern freely reorganizes and enjoys a series of interrelations inside the context of the sound system offered to him in that particular (already fully produced) composition. But in listening to Scambi the auditor is required to do some of this organizing and structuring of the musical discourse. He collaborates with the composer in making the composition.” 201 200 Ramsby (2007) 2 201 Eco (1989) 15, italics original. 72 When the Propertian speaker ventriloquizes Acanthis in Poem 4.5 and has her quote him verbatim to criticize his modus operandi as an elegiac lover, he produces openness that is closer to the first type that Eco describes than the second. Acanthis’ criticism is a particular interpretation of the speaker’s text, which is that Poem 1.2 is an unconvincing attempt at erotic persuasion by an inconsiderate egotist. Conversely, when the speaker dictates a flyer to his slave in indirect discourse in Poem 3.23 and expects the slave to recompose it in direct discourse, he dramatizes the second type of openness that Eco describes, because the slave will “collaborate[] with the [speaker] in making the composition.” Since the slave is an internal audience, the poem portrays Eco’s second type of openness but is not itself open in this way, unless perhaps an external audience takes up the slave’s task of recomposing the speaker’s words for the flyer. Likewise, the speaker presents the incomplete epitaphs in Poem 2.1 and 2.13 to internal audiences (Maecenas and Cynthia, respectively) to fill in or otherwise rework, though external audiences may also do so. Attempting to apply Eco’s second category of openness to Poem 4.7 helps to clarify the speaker’s vulnerable textuality because the second category of openness does not fit. When Cynthia’s ghost reuses the speaker’s words and ideas from an earlier elegy in Poem 4.7, she is not collaborating with the speaker to produce a poem based on a partially composed poem that the speaker has laid out for this purpose. She is instead appropriating part of a fully composed text for her own purposes. If this can be called ‘collaboration’ in any sense, the relationship between author and collaborator is different from the relationship Eco identifies in Pousseur’s Scambi. Poem 1.19 is not, on its own terms, open for collaboration the way Scambi is; instead, Cynthia’s ghost demonstrates how an audience might break open a poem to use part or all of it for their own ends. 73 The different ways of acknowledging the vulnerability of his texts to their audiences allow the Propertian speaker to manage this vulnerability to a degree. The speaker’s attempts to manage the vulnerability of his texts work toward a broader goal of managing their reception. The end of Poem 2.1 makes this double purpose clear. The speaker imagines Maecenas standing over his grave and says to his patron, “and, weeping, toss words such as these to the mute ashes: ‘A harsh girl was this wretch’s fate’ ” (taliaque illacrimans mutae iace verba favillae: / ‘Huic misero fatum dura puella fuit,’ Prop. 2.1.77-8). As discussed above, the speaker invites Maecenas to change his words here, but he also suggests how Maecenas should feel during the proceedings (“weeping,” illacrimans). However Maecenas may change the speaker’s words, it is important to the speaker that he receive them and the speaker (who is himself composed of words 202 ) with a certain emotion or at least with a certain display of emotion. Ritualized Speech and Mutability While the speaker of the Elegies presents his text as vulnerable in several ways, his attempts to manage this vulnerability boil down to the same strategy: He incorporates it into how he theorizes them as texts. But how does this incorporation of vulnerability work in a genre that is a type of ritualized speech? Roman primary sources and anthropological theories about ritual support the idea that Roman poetry is ritualized speech. Habinek argues that Roman poetry belongs to the category carmen, which he understands as “speech made special through meter, diction, accompanying bodily movement, or performance in ritual context” as opposed to “everyday speech.” 203 He also 202 See Chapter 4. 203 Habinek (2005) 1. Habinek (2005) 2 also notes that “song in Roman culture is not coextensive with performance and orality” and cites, among others, Cameron (1995) 87: there is “no conflict between song and book” and 102: “[i]t is a gross oversimplification to think of an age of reading succeeding an age of listening.” 74 argues that carmen makes people agents with authority formed within the ritualized context of carmen that extends to the broader culture outside the ritual contexts. 204 Habinek draws from Bell’s analyses of ritual and ritualization: “Ritualization... can be described as the strategic production of expedient schemes that structure an environment in such a way that the environment appears to be the source of the schemes and their values.” 205 Conte, in his study of poetic memory, borrows Lausberg’s two categories of speech, Verbrauchsrede (“language that is used up in the currency of everyday living”) and Wiedergebrauchsrede (“reusable language”). 206 Conte writes of the latter, “Reusable language exists precisely because society recognizes its value. Without this recognition it would not be freed from the obligation to communicate, or at least, to have communication as its sole and primary purpose,” and “Language that is reused is inevitably preserved in the poetic memory. In the case of poetry this process results in the shaping of a literary tradition... [which] does not exist in a vacuum but must work within the area of poetic discourse and must respect poetic norms.” 207 The ideas here of language that a society values for other capacities than just communication and the formation of a literary tradition correspond with Bell’s description of ritual in that the Wiedergebrauchsreden “structure” the literary tradition “in such a way that [the literary tradition] appears to be the source of the schemes and their values.” 208 Evidence for this last assertion comes in Conte’s discussions of how poetic memory functions in particular cases. He writes, for example, “In resisting the deterioration suffered by nonpoetic language at the hands of time, Catullus’s poetry projects toward us an autonomous system already endowed with its own 204 Habinek (2005) 3, and see his discussions of the Salian rites (8-33), convivia and sodales (34-57), and cinaedi (177-93) for illustrations of how this works. 205 Bell (1992) 140 quoted in Habinek (2005) 39, and see also Habinek (2005) 34-5. 206 Conte (1986) 40ff. 207 Conte 41-2 208 Bell (1992) 140 75 power to make its signs indelible.” 209 In other words, Catullus’ poetry constitutes the literary tradition that will both authorize it and provide its rules of operation. Some of the ways in which language makes itself “reusable” or “special” that Conte and Habinek point to also coincide with Bell and Rappaport’s descriptions of ritual. Conte writes, “[poetic language] boasts certain features that express and reflect this distance [from ordinary language]: rhythmical and metrical regularity, parallelism and effects of symmetry or proportion, unusual word order, repetition of expressive features and the exploitation of prosody (alliteration, assonance, onomatopoeia), to say nothing of the total effect produced by the figures of speech.” 210 Compare Habinek: “The Latin language differentiates between everyday speech and speech made special through meter, diction, accompanying bodily movement, or performance in ritual context” as opposed to “everyday speech.” 211 And compare Bell: “Ritualization, the production of ritualized acts, can be described, in part, as that way of acting that sets itself off from other ways of acting by virtue of the way in which it does what it does.” 212 The formality and repetition that Conte and Habinek describe correspond to some of the features of rituals that Rappaport lists. Rappaport argues that “ritual sequences are composed of conventional, even stereotyped elements, for instance stylized and often decorous gestures and postures and the arrangements of these elements in time and space are usually more or less fixed.” 213 His term for this is “formality (as decorum),” 214 and he sees as an inevitable consequence of this the quality of “invariance (more or less),” 215 by which he means that a ritual is performed in the same way every time, though “there is the possibility of, or even the necessity for, some 209 Conte (1986) 54 210 Conte (1986) 43 211 Habinek (2005) 1. Habinek (2005) 2 also notes that “song in Roman culture is not coextensive with performance and orality” and cites, among others, Cameron (1995) 87: there is “no conflict between song and book” and 102: “[i]t is a gross oversimplification to think of an age of reading succeeding an age of listening.” 212 Bell (1992) 140 213 Rappaport (1999) 33 214 Rappaport (1999) 33 215 Rappaport (1999) 36 76 choice to be exercised by performers even within the most invariant of liturgical orders.” 216 One way in which the language of Roman poetry is formalized is through the deployment of allusions, or “the activation of poetic memory.” 217 The material alluded to is conventional 218 and stylized. The requirements of fixity in time and space and invariance might at first seem at odds with allusion, which brings material from one text into another. But Conte writes that “[integrative allusion] produces a condensation of two voices in a single image whose sense lies in an interdependence of meanings that become subjectively equivalent.” 219 This subjective equivalence produces the effect of Rappaport’s “fixity in time and space.” What Conte calls “reflective allusions,” on the other hand, do not maintain the position of their material in time and space: In these allusions, “the literary process exploits the possibilities of comparison, not of substitution by an equivalent term. It considers the two items separately. Here the ‘propriety’ of the text is not violated.” 220 The text of a reflective allusion, then, causes the reader to revisit the text to which it alludes. In this way, the requirement of “fixity in time and space” is satisfied in reflective allusions, as well. Some literary devices with which the Propertian speaker anticipates other people reusing his text, such as echo and anaklasis, involve types of repetition and change that contribute to the ritualization of speech in Roman poetry. Bell and Rappaport both discuss the power of rituals to structure realities. Bell writes, “Ritualizing schemes invoke a series of privileged oppositions that, when acted in space and time through a series of movements, gestures, and sounds, effectively 216 Rappaport (1999) 36 217 Conte (1986) 66 218 Conte (1986) 57 219 Conte (1986) 67 220 Conte (1986) 67 77 structure and nuance an environment.” 221 Rappaport provides the helpful analogy of variables in a computer program: “The introduction to a textbook on circuit design tells us that ‘the successful operation of a real machine depends upon being able to separate the time intervals at which variables have their desired values from those in which they are changing’... As the values of variables in computers are contingent upon transformations in preceding intervals, so are social states in mundane periods in some degree outcomes of transformations in previous rituals, and while the states of affairs before and after ritual transformations can be distinguished by the digital logic of either/or (e.g., single/married, youth/man, war/peace), the logic of the interval, when transformation is actually effected, is that of neither/nor and analogical. ” 222 Habinek argues that “ritualization of language, that is, the transfiguration of speech into song, founds Roman culture and empowers agents within it.” 223 Propertius’ poems, for example, participate in and so help to “structure and nuance,” in Bell’s terms, Roman discourses on elite masculinities. 224 The speaker of the Elegies recognizes that types of repetition such as echo, anaklasis, and what modern scholars call allusion can change a poetic text that is itself participating in restructuring the world around it. The Propertian speaker thus builds mutability and impermanence into the structures (such as a non-traditional version of elite masculinity) of the fictive world of the Elegies, which will leak into the world outside of song and so participate in structuring that world. Conclusion This chapter has considered how certain ways in which the Propertian speaker acknowledges the vulnerability of his text relate to various concepts from ancient and modern literary and cultural criticism, particularly concepts that involve some sort of repetition. When the speaker predicts how other people may reuse his text, he shows readers how such reuse looks from 221 Bell (1992) 140 222 Rappaport (1992) 12-3 and cf. Rappaport (1999) 96-7 223 Habinek (2005) 4 224 See Chapters 3, 4, and 5 and the Conclusion. 78 the perspective of an author whose text will be reused rather than the perspective of the text that will do the reusing, which is the perspective generally taken by ancient discussions of echo and anaklasis and modern discussions of allusion and intertextuality. By anticipating how other people may reuse his text and making these possibilities part of the way he understands what it means for his text to be a text—i.e., that it is susceptible or vulnerable to reuse by other people—the speaker claims some control over the reception of his text and of the ideas he offers through his text to the world around him. Sometimes, though, the speaker anticipates the erasure or destruction of his text. These passages, as is perhaps self-evident, do not involve repetition. If we were to analyze them in Bell and Rappaport’s terms, they would be the removal of the ritual from both practice and record. As we shall see in Chapter 3, the speaker of Propertius’ Elegies had good reason to hypothesize that later texts would cite, imitate, and alter his words and that some users would destroy his texts or otherwise efface his authorship. 79 Chapter 3: The Uses of Propertius’ Elegies Introduction: To what features of Rome’s literary culture might the trope of the vulnerable text in Propertius’ Elegies—the speaker’s acknowledgement that others may change or destroy his text— respond? In Chapter 2 I discussed how the ways that Propertius’ Elegies theorize themselves as a set of texts open to change, reuse, and destruction relate to the array of concepts such as echo and allusion with which literary critics seek to understand connections between texts. In this chapter I turn to concrete examples of how poets and other creators and users of Latin literature reused and changed Propertius’ words. I do not aim to discuss or even list every quotation of or allusion to Propertius’ Elegies in Roman literature. Rather, I use examples from various genres and media to explore how later authors changed Propertius’ texts and how these changes affect the Propertian speaker’s self-presentation. The chapter is divided into three main sections: Section A examines evidence of very early interpolations in Propertius’ text and discusses what changes these interpolations, if they are correctly identified, introduce into the text. Section B explores how authors of several different types of texts, such as grammatical treatises and Pompeian graffiti, recycle Propertius’ words in ways that change them but leave them still recognizable as Propertian in origin. Section C considers evidence of an oral poetry culture alongside Rome’s written literary culture and discusses how certain elements of an oral poetry culture might change or efface the authorship of phrases from Propertius’ Elegies and how these features might more broadly affect Roman ideas about authorship. 80 Section A: Early Interpolations The textual tradition of Propertius’ Elegies is a notorious muddle. Even among other Latin and Greek texts with limited and difficult manuscripts, the Elegies stand out for the lateness of the surviving manuscripts, the depth of their known corruption, and the extent of their suspected corruption. Butrica writes, “All texts have suffered corruption to some degree in their transmission, but rarely has that corruption had so profound an impact upon reception,” 225 and “the text of Propertius as it emerged in the Middle Ages in the archetype of the [manuscript] tradition had unquestionably been deformed from its original character.” 226 Tarrant takes a more moderate position on the corruption of the Elegies but agrees that “the Propertian archetype was undoubtedly a wretched affair, defaced by scribal error and botched attempts at correction” 227 and that “critics of Propertius are faced with a highly fallible manuscript tradition.” 228 Even the conservative Fedeli writes at the beginning of his Teubner edition, “Persaepe in apparatu aliis locis allatis qui apti uidebantur, lectionem a me electam ueram ostendere conatus sum.” 229 Some of the many possible corruptions that critics and editors have identified may be ancient in origin. Butrica has pinpointed several that he believes are ancient, and he argues that a few of these entered the text of the Elegies very early – in one case perhaps even within a generation of the publication of the Elegies. This section examines these possible early interpolations and discusses how, if Butrica is correct, they changed Propertius’ text. There is, of course, a problem: If the text of the Elegies that has come down to us is so corrupt, how can I know what the original looked like before an early interpolation changed it? I cannot. The closest I know how to get to resolving this problem—carefully 225 Butrica (2006) 25 226 Butrica (2006) 26 227 Tarrant (2006) 50 228 Tarrant (2006) 52 229 Fedeli (1984) XXIV 81 considering each word on a case-by-case basis and in consultation with multiple critical editions and commentaries—is still imperfect. My analyses, then, might be taken as examples of the types of changes that interpolations could have introduced to Propertius’ texts soon after their publication, if not necessarily examples of actual changes. One verse that Butrica argues may have been corrupted in antiquity is Prop. 3.14.23. The consensus of modern editors is that in the archetype from which the surviving manuscripts descend (Ω) verses 23-4 read, “neither is there fear or any guarding of a girl kept shut in” (nec timor aut ulla est clausae tutela puellae / nec gravis austeri poena cavenda viri). Butrica writes, “To say absolutely that ‘there is not fear’ is too sweeping and too vague; and if timor is parallel to tutela, one is left with ‘fear of a secluded girl’—but there can be no such fear if there is also no such seclusion.” 230 In Butrica’s interpretation 231 of the text of the archetype, the girl’s guardian has shut her in because he fears that she will be seduced. This produces an awkward shift of perspective: In verses 21-2, the elegiac lover extols Spartan lovers’ freedom in public, and in verse 24 the elegiac lover fears punishment by a jealous husband, so it is difficult to accept that in verse 23 the speaker takes the perspective of a girl’s guardian. Fedeli wants instead to read the fear (timor) as the girl’s fear of a minder, 232 but this again produces an awkward shift of perspective. 233 To avoid these shifts, we would need to interpret the fear in verse 23 as the lover’s fear that his girl will be shut in—“neither is there fear or any guarding of a girl [being] shut in”—but this would normally 230 Butrica (1997) 184 231 Goold’s (1990) Loeb translation shares this interpretation. 232 Fedeli (1985) 462: “Il timor equivarrà, dunque, al custodium timor di 2,23,13-14 contra, reiecto quae libera uadit amictu, / custodum et nullo saepta timore, placet ; la sorveglianza, cioè, continuava anche nelle pubbliche passeggiate delle fanciulle, sotto forma di occhiuti custodi.” 233 In verse 22—“and it is permitted to be by her side in the crossroads” (et licet in triviis ad latus esse suae)—the feminine “her” (suae) indicates that verses 21-2 are specifically from the elegiac lover’s perspective, not the perspectives of both the lover and his girl. Likewise, the masculine “yourself” (ipse) in verse 26 indicates that verse 25 is from the elegiac lover’s perspective. So in Fedeli’s interpretation, the perspectives would shift as follows: verses 21-2, elegiac lover; verses 23-4, girl; verse 25, elegiac lover. 82 be expressed with a gerundive (timor… claudendae puellae) rather than the perfect passive participle. An argument that meter and a sort of zeugma 234 coordinating timor and tutela require the perfect passive participle is unconvincing when there is an elegant alternative. Butrica and Heyworth following him seek to revive Broekhuyzen’s proposal that verse 23 should read “neither is the guard of a shut-in girl a [source of] fear for anyone” (nec timor est ulli clausae tutela puellae), which “restores balance and contrast.” 235 Butrica offers as one possible source of this corruption “that ulli first became ulla through anticipation of tutela, then aut was added and the order of words changed to restore the metre.” 236 He argues that this change in word-order to restore the meter would have been ancient, “because medieval rules of scansion freely admitted the lengthening of short syllables at the principle caesura and no medieval reader would have balked at nec timor est ulla clausae tutela puellae.” 237 Butrica also argues that several entire poems in Book 2 of Propertius may have been interpolated into the book in antiquity. One of these is Poem 2.15. To begin with, “the juxtaposition of 2.14 and 15 is anomalous… since it yields a pair of consecutive poems with exactly the same subject matter in which the second seems unaware of the first,” whereas in Book 1 each part of the two-part elegies (8 and 11-12) and each of the “paired” elegies (7 and 9, 10 and 13) is in some way responsive to its partner. 238 Even more damning is the word Actiacus in verse 44. Butrica writes that Augustus “created Actiacus as a new cult name of Apollo in connection with the enlargement and rededication of his temple at Actium,” and he cites “a corrupt gloss of Servius on A. 8.704”: “The Thilo-Hagen edition prints the impossible quem postea Actium 234 It would also be unusual to understand this as a zeugma, because this kind of simple zeugma is usually formed by exploiting multiple meanings of a verb to link two distinct ideas (Quint. IO 9.3.62, and see Lausberg 1960: 348-9 for discussion and further citations and examples). 235 Butrica (1997) 184; Heyworth (2007) 362 236 Butrica (1997) 184 237 Butrica (1997) 184 238 Butrica (2006) 35 83 nominavit Augustus, but actiuum in mss LH surely points to Actiacum; cf. also Serv. A. 3.274 templum Actiaco Apollini constituit.” 239 If this is the case, it explains why the word Actius, not Actiacus, is used in all other Augustan poetry (including five occurrences in Propertius) until Ovid’s Metamorphoses and Fasti, which replace Actius with Actiacus, and why “Actiacus is all but universal thereafter.” 240 Furthermore, at 4.6.67, Propertius calls Apollo Actius when he talks about the victory monument at Actium. This poem is dated to 16 BCE by the reference to the Sygambri in verse 77, and “a coin datable to the very same year, no doubt minted for the same celebration of the ludi quinquennales for which Propertius probably wrote 4.6, carr[ies] an image reminiscent of the victory monument at Actium and bear[s] the legend Apollini Actio.” 241 As Butrica points out, “the moneyer is perhaps even less likely than the poet to have passed over Augustus’ own title for the god if it existed yet.” 242 If the title Actiacus did not exist when Propertius was composing Book 4, then it cannot have existed when he was composing Book 2. This fact and the strangeness of pairing Poems 2.14 and 2.15 suggest that Poem 2.15 is interpolated. If Poem 2.15 is interpolated, then the speaker of the elegies of Propertius we are accustomed to reading is more violent than the speaker of the elegies Propertius composed. One especially jarring introduction into Propertius’ poetry would be the only passage in the elegies in which the speaker openly and credibly threatens to beat up Cynthia or another puella: 243 239 Butrica (2006) 36 240 Butrica (2006) 35-6 241 Butrica (2006) 36 242 Butrica (2006) 36 243 At 2.1.13, the speaker discusses ripping off the puella’s dress, but the 3 rd -person plural euphemism, “then we truly compose long Iliads” (tum vero longas condimus Iliadas, Prop. 2.1.14), suggests that, in the speaker’s mind at least, this is a sex game that both participants enjoy. The speaker’s threat at 2.8.21-8 to stab the puella and then himself is an unrealistic fantasy: Papanghelis (1987) 125 calls it “Murder as one of the Fine Arts,” and James (2003) 192 writes, “Peculiarly, this poem does not appear on any commentator’s list of violence in elegy. It is certainly exaggerated in tone, and as Richardson (1977: 233) notes, it is deliberately incoherent in organization. Though it is exaggerated and not serious even on a superficial level, the death threat in this poem demonstrates the 84 “But if you willfully persist in going to be clothed, you will, with your dress ripped, feel my hands: Indeed, if anger carries me away even more, you will show your mother bruised arms.” quod si pertendens animo vestita cubaris, scissa veste meas experiere manus: quin etiam, si me ulterius provexerit ira, ostendes matri bracchia laesa tuae. (Prop. 2.15.17-20) This threat flies in the face of the speaker’s realization at 2.5.25-6 that such violence is the behavior of a “hick… whose head vines of ivy have not encircled” (rusticus… cuius non hederae circuiere caput). Fredrick has identified a pattern of “oscillation between Callimachean scopophilia and mock-epic voyeurism” in Poems 2.2 through 2.13. 244 Poems 2.14 and 2.15, he argues, are the “generically loaded climax” of this pattern. 245 Poem 2.14 provides mock epic voyeurism when it compares the speaker’s ecstasy at having successfully made love to the puella with Agamemnon’s victory in the Trojan War and other triumphs from epic (Prop. 2.14.1-10) and when the speaker says of his erotic conquest, “these will be my spoils, these my conquered kings, these my chariot” (haec spolia, haec reges, haec mihi currus erunt, Prop. 2.14.24). 246 After this, the epigram of thanks to Venus and the speaker’s wish that he might die if the puella should break off the ecstatic union could fittingly draw this sequence to a close. Poem 2.15 adds another ending to the sequence. In verses 1-20, the speaker describes the puella’s body and their lovemaking with physical details that allow the reader or audience to visualize the scene. In verses 17-20, the speaker threatens erotic violence if the puella continues to go to bed clothed and so by reverse implication suggests her nudity, which he has partially described in the preceding verses. Fredrick lover-poet’s rage at the puella, and it speaks particularly to his anger at not enjoying his usual position of mastery relative to women, particularly women of a lower class, a position ordinarily guaranteed him by his gender and status, an elite Roman male, but forbidden him by the rules of elegy. Since violence is forbidden, poetry is, as usual, his tool, this time for revenge.” 244 Fredrick (1998) 180 245 Fredrick (1998) 182 246 Fredrick (1998) 182 85 comments that “full disclosure of sexual difference in the bedroom overlaps with elegy’s obsession with discovering the puella’s guilt, the equivalent of breaking down the doors to find her with another man.” 247 He argues that in this way “scopophilia is transformed into voyeurism, and looking becomes a blow; the display of sexual difference is transposed into the bruised arms Cynthia shows to her mother.” 248 If Poem 2.15 is interpolated, then the sequence Fredrick has identified ends with Poem 2.14, and the development of “Callimachean scopophilia and mock-epic voyeurism” in Fredrick’s reading of Poem 2.15 is not Propertian. Perhaps Poem 2.15 is not such a climax, in any case, as its violence and mythological exempla are more suited to love poetry than to epic: The word rixa (Prop. 2.15.4) and its derivatives 249 do not appear in Ennius’ Annales, Vergil’s Aeneid, Lucan’s Pharsalia, Statius’ Thebaid, Valerius Flaccus’ Argonautica, Silius Italicus’ Punica, or even Ovid’s Metamorphoses; the word luctor (luctata, Prop. 2.15.5) occurs commonly in a wide range of genres, so it cannot be said to belong to any one of them; and the mythological exempla are the passions of Paris and Selene when they saw Helen and Endymion nude (Prop. 2.15.13-6). Butrica also suspects that Poems 2.21, 250 2.22, 251 and 2.34 were interpolated. These poems come under suspicion because they call some of the speaker’s acquaintances by mythological pseudonyms. In Poem 2.21, someone whom the speaker names “Panthus” (Prop. 2.21.1-2) has 247 Fredrick (1998) 182 248 Fredrick (1998) 182 249 rixo, rixor, rixator, rixatrix, rixatorius, rixosus 250 Günther (1997) 141 includes Poem 2.21 in an analysis of the structure of Book 2, so he seems not to think it was interpolated. Heyworth (2007) 198-200 also thinks the poem is genuine, albeit transmitted in shoddy condition, and wonders whether Panthus might have been a very young Ovid. 251 Günther (1997) 125 thinks that verses 11-12 are probably spurious and (1997) 19 that 22b (i.e., Poem 2.22 from verse 20 onward) does not belong in its transmitted position, as “xxii A, xxiii, and xxiv A form a cycle of three poems connected by a common theme and addressee,” and “the topic and sentiment of xxii B are as inappropriate between these poems as is the address to the domina.” Heyworth (2007) 200-1 thinks that, rather than interpolated, Poem 22 has been transmitted with verses out of order and maybe even with verses from more than one poem. Since, as he notes, Ovid “reworks the Demophoon section of [Prop.] 2.22” in Amores 2.4, perhaps this means that Prop. 2.22 is not interpolated. 86 written slanderous verses about the speaker and with them has seduced Cynthia, and now he spreads vicious gossip about her to other people. In Poem 2.22, the speaker complains to someone he calls “Demophoon” (Prop. 2.22.2, 13) about the trouble of loving multiple girlfriends, rebukes this “Demophoon” for asking why he subjects himself to this torment, and then claims that he is actually up to the challenge and that he is in fact better off loving two girlfriends than just one. In Poem 2.34, the speaker accuses someone he calls “Lynceus” (Prop. 2.34.9) of betraying him by trying to seduce his beloved. Panthus, Demophoon, and Lynceus are all mythological names, and Butrica notes that “there is no parallel in Propertius himself or elsewhere in Latin poetry for addresses with mythological pseudonyms either in an individual poem or throughout a series of poems—and certainly not for their use in only one part of a book.” 252 He suspects that these mythological pseudonyms indicate an author other than Propertius, and the fact that the pseudonyms occur in the second half of Book 2, where there are many problems with divisions between poems, compounds his suspicion, as does the unusual length of Book 2 (1362 verses as transmitted). 253 To condemn these three poems as interpolated is aggressive and perhaps even overreaching, and Butrica acknowledges that the evidence is only enough for suspicion. 254 Poem 2.34 is the famous concluding poem of the book, 255 and so it is difficult to do away with it. If Poem 2.34 stands, then its use of a mythological pseudonym defends those in Poems 2.21 and 2.22. Nevertheless, Butrica has enough evidence for suspicion, and other critics have questioned various parts of these poems, 256 though not the entire poems. 252 Butrica (2006) 35. Strictly speaking, Poem 2.21 does not address Panthus but rather speaks about him in the third person to the puella, whom it addresses. 253 Butrica (2006) 35 254 Butrica (2006) 35 255 See, for example, Heyworth (2007) 264: “The whole discursive piece works well as a closing poem; the previous two introductory poems II i and II xiii move from reflections on genre to more precisely erotic material, and this poem reverses the pattern.” 256 Günther (1997) 83-5, for example, argues that verse 39 of Poem 2.34 is an interpolation in a lacuna. 87 In an earlier essay, Butrica presents more detailed evidence to suggest that the praise of Vergil’s forthcoming Aeneid at Prop. 2.34.65-84 is interpolated. 257 He first argues that verses 65- 6 are Propertius’ own words and were interpolated from Donatus’ Life of Vergil in antiquity or the middle ages. 258 Donatus writes, “So much hype arose about the Aeneid, though it was scarcely begun, that Sextus Propertius did not hesitate to proclaim, ‘Out of the way…’,” (Aeneidos vixdum coeptae tanta exstitit fama ut Sextus Propertius non dubitaverit sic praedicare, Cedite… Don. Verg.). Butrica notes that Donatus’ verb “to proclaim” (praedicare) suggests that verses 65-6 were “a spontaneous utterance, no doubt in the context of some banquet or recitation,” and he reads them as “a humorous, perhaps even ironic or sarcastic, reaction to the noise—emanating from Maecenas’ circle in the first instance and perhaps also from the general direction of the Palatine— that greeted the announcement of Virgil’s first work on the Aeneid.” 259 In contrast, Butrica does not think that verses 67-84 are Propertian at all. He points to the “corrupt and therefore incomprehensible final couplet” as the most obvious indication of a problem. 260 He then demonstrates that the text of Poem 2.34 reads smoothly and makes good sense without verses 65- 84, and he notes several problems of sense in these verses: the confusing inconsistency and lack of transition between the discussion of Vergil in the third person in verses 61-4 and the second- person pronoun tu addressing Vergil in verse 67, for example, and the problem of what the word haec refers to in verses 85, 87, and 89. 261 He also points out a metrical feature of verses 67-84 inconsistent with the rest of Book 2, namely that the “high concentration of pentameters with polysyllabic endings… make it look as though Propertius at the very end of Book 2 suddenly 257 Butrica (1997) 201 258 Butrica (1997) 201 259 Butrica (1997) 201 260 Butrica (1997) 202-3 261 Butrica (1997) 202-4 88 reverted to his practices in Book 1… a metrical practice he had already in large part abandoned.” 262 Butrica further argues that this interpolation must be ancient and perhaps even early Augustan in origin, because medieval scholarship and poetry show no interest in polysyllabic pentameter endings. 263 Butrica also suggests a source for many of these interpolations: “Given that the verse is clearly classical in meter and style, it would appear that Book 2 has fortuitously incorporated substantial remains of another ancient elegiac poetry book.” 264 He proposes as a candidate for authorship of these interpolations Passennus Paulus, about whom Pliny the Younger writes, 265 “Besides, in literary composition he rivals, imitates, and restores the old [poets], first and foremost Propertius, from whom he descends, true offspring and very similar to him [i.e., Propertius] in that in which [Propertius] was distinguished. If you pick up his elegies, you will read a polished, tender, delightful work, clearly written in the household of Propertius.” praeterea in litteris veteres aemulatur, exprimit, reddit, Propertium in primis, a quo genus ducit, vera suboles eoque simillima illi, in quo ille praecipuus. si elegos eius in manus sumpseris, leges opus tersum, molle, iucundum et plane in Properti domo scriptum. (Plin. Ep. 9.22) Pliny recommends Passennus Paulus’ elegies as very similar to Propertius’ – the phrases “he… restores the old [poets]” (veteres… reddit) and “clearly written in the household of Propertius” (plane in Properti domo scriptum) suggests that a fan of Propertius’ elegies might enjoy Passennus Paulus’ as an extension or a continuation of them. The relationship that the word genus in the phrase “from whom he descends” (a quo genus ducit) indicates, furthermore, may be one of literary genre 266 as well as familial descent. Even if Passennus Paulus is not the author of Poems 2.21, 262 Butrica (1997) 204: “While the 18 lines preceding the interpolation offer three scattered examples (48 laqueis, 58 ingenio and 64 litoribus), the 18 lines of the ‘Virgilian’ passage contain a cluster of five examples (68 harundinibus, 70 uberibus, 74 delicias, 76 Hamadryadas and 80 articulis); the eight examples found within these 36 lines actually surpass the frequency found in 1.1, which had seven examples within 38 lines.” 263 Butrica (1997) 204, (2006) 34 264 Butrica (2006) 35 265 Butrica (2006) 35 266 TLL 6.2.1898.34ff. gives examples of the word genus used of literary type. Evidence of genre-awareness in Latin literature abounds. See, for example, Cic. Orat. 61-8; Verg. Ecl. 6.1-12, Geo. 2.176; Hor. Carm. 1.6, Ars 73-85; Prop. 1.7, 1.9, 2.1, 3.1, 4.1 inter alia; Quint. Inst. 10.1.85-131. For discussion see Cairns (1972) and Conte (1994b). 89 2.22, and 2.34, then, Pliny’s letter indicates that the work of a less famous poet in the style of a more famous poet could come to be seen as virtually a part of the more famous poet’s corpus. If certain parts or all of these three poems are interpolated, how have they changed Propertius’ text? If the mythological pseudonyms are not Propertius’ work, then they present as Propertian a poetic practice that is not. Propertius’ speaker does make frequent and sometimes obscure references to mythology, so the introduction of further mythological content does not much change his poetics. He generally addresses or refers to his friends, relatives, rivals, and other acquaintances by their own names, though, or at least by Roman names. These real or at least Roman names help to anchor the poems in the speaker’s present and so to make his vignettes feel true and realistic. Mythological pseudonyms, on the other hand, flaunt the speaker’s artfulness. 267 If large parts of Poem 2.34 are interpolated, the conclusion of Book 2 may originally have been substantially different from the conclusion as we now know it. The effects of whole poems or large sections or even couplets and verses being interpolated can probably be seen in many of the difficulties that exercise textual critics in Book 2. As Butrica says, the book is too long, so heavy interpolation seems likely. Tarrant has developed a typology of interpolations with three basic categories, which he calls “emendation, annotation, and collaboration.” 268 If Butrica’s scenario for the corruption of Prop. 3.14.23 is correct, 269 then the interpolations in this verse are of the emending variety: One scribe misunderstood the syntax of the verse and changed ulli to ulla to agree with tutela, and later another scribe added aut and changed the word order to fix the meter. Likwise, verse 39 of Poem 2.34 is an emending interpolation if Günther is correct that it is an interpolation in a lacuna. 270 In 267 Veyne (1988) and Kennedy (1993) explore reality, realism, art, and artfulness in detail. 268 Tarrant (1987) 126 269 See above. 270 Günther (1997) 83-5 90 contrast, Tarrant’s term “collaborative” suits the other interpolations discussed above. In a “collaborative” interpolation, “the reader seems to take on the role of a co-author who revises, expands, or varies the text, not because it appears defective or obscure but simply because it allows for further elaboration, because it has not yet exhausted the possibilities of the material.” 271 If Poem 2.15 is interpolated, for example, its author may have composed it with the idea of further developing themes from Poem 2.14 and earlier poems in Book 2. This means that there may be multiple readers’ voices alongside Propertius’ in the texts we understand as authored and authorized by Propertius. There may also be additional acts of authorship at work here. The author of Poem 2.15 may never have meant for it to be inserted into Propertius’ book – the names Cynthia and Propertius do not appear in this poem, nor does any other detail that would securely identify the poem as intended by its author for its current position. In this case, a third party may have added it to Book 2 and so acted as a coauthor both to Poem 2.15 and to Book 2. As he describes collaborative interpolations, Tarrant further distinguishes three subtypes: “those which smooth a transition or fill an apparent ellipse in the argument or narrative; those which extend, amplify, or heighten a point; and those which add emphasis or weight to a conclusion.” 272 Poem 2.15 extends the speaker’s proclamation and consideration of his joy at having had sex with the puella. It also reformulates these ideas so that the speaker’s wish at the end of Poem 2.14 to die if the puella should leave him becomes a meditation on a joint Liebestod: “But just as the petals have left the wilting garlands, and you see them strewn everywhere, swimming in the cups, so for us, who now breath deep, being lovers, perhaps the morrow day will close our fates.” ac veluti folia arentis liquere corollas, quae passim calathis strata natare vides, sic nobis, qui nunc magnum spiramus amantes, forsitan includet crastina fata dies. (Prop. 2.15.51-4) 271 Tarrant (1987) 137 272 Tarrant (1987) 137 91 This revised fantasy of death, with its flower simile and sympotic imagery, is a dramatic and satisfying conclusion to an elegiac discussion of ecstatic erotic union. In contrast, the last four verses (29-32) of Poem 2.14 read like an epilogue after the climactic epigram in verses 27-8, “these spoils I, Propertius, place before your temple, Goddess, I who was received for a whole night as a lover” (HAS PONO ANTE TUAM, TIBI, DIVA, PROPERTIUS AEDEM / EXUVIAS, TOTA NOCTE RECEPTUS AMANS). 273 This is not to say that the last four verses of Poem 2.14 are unsuitable as an ending—an ending in which the Propertian speaker pathetically undercuts his own grandeur suits him well—but rather that, in a scenario in which Poem 2.15 is interpolated, its author has chosen an ending that seals the speaker’s erotic triumph rather than doubting its permanence, and some readers might find this ending more “empha[tic]” or “weight[y],” to use Tarrant’s terms, than the epilogic ending of Poem 2.14. This example of the concluding type of collaborative interpolation may not seem entirely distinct from the amplifying type – Tarrant writes, “[the concluding] type may be considered a subdivision of the amplifying/heightening interpolation placed at the end of a paragraph, speech, or episode.” 274 Tarrant gives examples of collaborative interpolations in texts by other authors, and some of them may be ancient in origin. Verse 197a of Book 9 of Ovid’s Metamorphoses, for example, reads, his Cacus horrendum Tiberino litore monstrum. As Tarrant notes, the author of the verse “scanned Cacus with a short first syllable, a solecism suggesting either amateur standing or a post- 273 The text of the archetype (Ω) read tuas…aedes. Goold’s Loeb text and Heyworth (2007) 172 accept Scaliger’s correction to tuam…aedem, and I find Heyworth persuasive here: “the appropriate sense here is ‘temple’, not ‘house’… [Scaliger’s correction] avoids the awkward interveaving of feminine accusative plurals in has…tuas…aedes | exuvias: the cause of corruption lies in has…exuvias as well.” 274 Tarrant (1987) 150 92 classical date.” 275 Likewise non-Ovidian are five verses that some manuscripts include in the battle of the Lapiths and Centaurs in Book 12 of the Metamorphoses: “the broad dome of his head 276 was smashed, and through his mouth and through the hollows of his nose and eyes and ears his soft brain flowed, like cottage cheese from an oak-twig basket or like a thick liquid flows under the weight of a fine sieve and is pressed out through the tight openings.” fracta volubilitas capitis latissima, perque os perque cavas nares oculosque auresque cerebrum molle fluit, veluti concretum vimine querno lac solet utve liquor rari sub pondere cribri manat et exprimitur per densa foramina spissus. (Ov. Met. 12.434-8) None of the older manuscripts of the Metamorphoses include these verses, and while, on the one hand “the lines hardly lack imagination and the writing displays a certain demented vigor,” on the other hand, “the Latinity is plainly not Ovidian and perhaps not even classical.” 277 Verses 125 and 130 from Juvenal’s description in Satire 6 of Messalina moonlighting in a brothel, if they are in fact interpolations, “are almost certainly ancient in origin since they employ (albeit with much less skill) the same rhetorical and stylistic procedures favored by Juvenal himself.” 278 These and other examples suggest a literary culture in which collaborative interpolation was a common way of interacting with poetry (and perhaps also prose). Authors composing in such a culture were likely aware of this tendency in their readers and audiences: Donatus’ Life of Vergil reports that a member of the audience at a recitatio interrupted Vergil to supply his own ending to verse 299 of Georgics 1. Vergil’s line reads, “[in the summer] plow nude, sow nude; winter is the lazy time for a farmer (nudus ara, sere nudus; hiems ignava colono, Verg. Geor. 1.299). The wit in the audience chimed in after the second nudus and replaced “winter is the lazy 275 Tarrant (1987) 140 276 Tarrant (1987) 146: “volubilitas captitis latissima” for (I suppose) ‘the broad dome of his head’” 277 Tarrant (1987) 146 278 Tarrant (1987) 150 93 time for a farmer” (hiems ignava colono) with “you’ll catch a fever from the cold” (habebis frigore febrem). 279 As Tarrant notes, this story may be apocryphal, but “it presupposes” an audience and a literary culture in which the story of such an interaction would have been credible. 280 Dating these interpolations precisely is probably not possible – it is often difficult even to be certain of their antiquity. Tarrant suggests that the periods from the death of Ovid in 17 CE to the death of Juvenal in the early 2 nd c. CE and from about 350-420CE are “especially likely” time- frames for the bulk of the collaborative interpolations in Latin literature. 281 In support of the first of these periods, he points out the relationship between the “quintessentially ‘rhetorical’” “character” of several texts in this century and the practice of declamation in rhetorical education. 282 He notes the many variations on themes such as the Judgement of Paris in the Latin Anthology and the common rhetorical exercises in paraphrase and prosopopoeia, 283 and he argues that prosopopoeia “probably accounts for the bulk of pseudonymous literature, both prose and verse, of the late Republic and early Empire… Here the analogy with interpolation is especially close…” 284 The death of Ovid is a later beginning than necessary for Tarrant’s first period of heavy interpolation. As discussed in Chapter 1, the incomplete epitaphs in Poems 2.1 and 2.13 invite collaborative interpolations by the addressees and audience to “fill an apparent ellipse in the argument or narrative” 285 or perhaps interpolations to emend their texts. 286 Similarly, in Poem 4.8 (also discussed in Chapter 1) the speaker asks the Appian Way to describe Cynthia’s triumphal 279 See Chapter 3 Section C for discussion of Vergil’s own extemporizing in Donatus’ Vita. 280 Tarrant (1987) 161 281 Tarrant (1987) 156 282 Tarrant (1987) 158, 159-62 283 Tarrant (1987) 159 284 Tarrant (1987) 160 285 Tarrant (1987) 143 286 Tarrant (1987) 126-30 94 ride to Lanuvium in the carriage of a young fop: “Appian Way, tell, I beg, in how great a triumph she drove, with you as a witness, as her wheels shot across your stones” (Appia, dic, quaeso, quantum te teste triumphum / egerit effusis per tua saxa rotis, Prop. 4.8.17-18). The indirect question introduced by quantum invites the Appian way to insert its version of the story into Propertius’ account, and this collaborative interpolation would “extend, amplify, or heighten a point.” 287 Martial’s epigram against Fidentinus, quoted as the epigraph to Section C, indicates that other people might, accidentally or on purpose, make an author’s poetry their own when they recited it, even while the original author was still alive and perhaps even with the original author in the audience. 288 Martial accuses Fidentinus of performing deformazione on his poetry just by reciting it poorly. In Eclogues 10, Vergil composes deformazione of Gallus’ poetry, and the Propertian speaker warns Gallus to beware of deformazione even as he does it to both Gallus and Vergil’s poetry in Poem 1.20. 289 Modern textual criticism does not think of such examples of deformazione as interpolations, but in Augustan Rome, where poetry was performed in recitationes, where people might behave as Vergil’s audience member mentioned above did, and where poetry was played with 290 and read in private, deformazione and interpolation may not have seemed so different. 291 287 Tarrant (1987) 143 288 “The book you’re reciting, Fidentinus, is mine, but when you recite it badly, it begins to be yours!” (quem recitas meus est, o Fidentine, libellus; / sed male cum recitas, incipit esse tuus! Mart. Epig. 1.38) 289 See Chapter 2. 290 See discussion of Catullus 50 in Section C. 291 Greek and Roman literature were concerned about interpolation as early as the poetry contest in Aristophanes’ Frogs, in which Aeschylus metaphorically fucks Euripides’ verses by interpolating the phrase “he lost his little lekythos” (ληκύθιον ἀπώλεσεν, Ar. Ran. 1197-1250) into them (a lekythos is a flask, and Dover 1997: 204 reviews the philological and physical reasons for reading Aeschylus’ little lekythos as phallic, including the fact that “one common type looks remarkably like a penis”). Cicero accuses Verres of cooking the books, in part by interpolating, to conceal his abuses as Governor of Sicily: “In this way that man learned to look out for himself and his own wellbeing by entering things into both his private records and the public records that had not happened, by taking away things that had happened, and continually removing something, changing something, interpolating something” (hoc modo iste sibi et saluti suae prospicere didicit referendo in tabulas et privatas et publicas quod gestum non esset, tollendo quod esset, et semper aliquid demendo, mutando, interpolando, Cic. Verr. 2.1.158). 95 Tarrant’s insights about rhetoric and declamation can also support extending his first period of heavy interpolation back at least as far as Propertius’ floruit in the 20’s and 10’s BCE. Rhetorical training began in Rome long before Propertius’ time, and even the “new world” 292 of declamation, to which Tarrant refers, 293 was underway when Propertius was composing the Elegies. Some of the declaimers in the Controversiae and Suasoriae of Seneca the Elder, for example, must have been active in the 20’s BCE and perhaps even the 30’s, since Seneca says that he heard some of them in his youth (Sen. Controv. 1.pr.3-5 294 ). Keith has demonstrated the debt of Propertius’ Elegies to Ciceronian oratory 295 and documented the evidence of Propertius’ own training in schools of rhetoric, which he would have entered in the years following the Perusine War. 296 Prosopopoeia, which Tarrant mentions as a rhetorical technique that may have caused pseudonymous literature and interpolation, 297 is one of many rhetorical techniques that Propertius’ Elegies employ (Prop. 1.16, 3.6.19-34, 3.7.57-64, 4.1B, 4.2, 4.3, 4.5, 4.7, and 4.11). Internal and external evidence both suggest that one way in which other people in Propertius’ time and soon after likely interacted with the Elegies, perhaps even with Propertius present, was to change and add on to them. Section B: A Culture of Literary Recycling Interpolations were not the only changes that commonly befell texts of poetry in the late Republic and early Empire. Authors of literature in various genres alluded to and quoted each other’s words, and in doing so they changed each other’s texts in a number of ways. Sometimes 292 Clarke (1953) 85 293 Tarrant (1987) 158, 159-62 294 See Chapter 5 for further citations and discussion. 295 Keith (1999) 52-7 296 Keith (2008) 19-44 297 Tarrant (1987) 160 96 they changed the actual words of the source text while leaving enough of the phrase intact to be recognizable or attributing the phrase to its original author. Even when it did not change the words of the source text, an alluding or quoting text set the quoted phrase or verse into a new context, sometimes one drastically different from the original context, and provided an interpretation that might preempt a reader from thinking of other interpretations. In this section, I explore how technical and philosophical treatises, canonical poetry, and epigraphic poetry reuse and change Propertius’ text. Grammarians and other Writers of Treatises In the late Roman Republic and the Roman Empire, grammarians (grammatici) were teachers who received the children of the elite after their elementary education and prepared them for advanced training in rhetoric. Some of them were also poets and scholars in their own right. There are twelve known quotations of Propertius in treatises by grammarians and one quotation of a passage ascribed to Propertius. 298 Two of these quotations can be traced to the first century CE and so reflect literary practices shortly after Propertius was composing and publishing. These two quotations remove Propertius’ words from their original, poetic contexts and use them to communicate information. In Conte’s terms discussed in Chapter 2, they change Propertius’ words from Wiedergebrauchsrede (“language that can be reused”) into Verbrauchsrede (“language that is used up in the currency of everyday living”). 299 The earlier of these survives in the Ars grammatica of the fourth-century grammarian Charisius. Charisius gives several exempla from Propertius, and one of these quotations can be traced back to the now lost work of the influential grammarian Remmius Palaemon, who taught Persius the satirist and Quintilian and lived during the reigns of Tiberius and Claudius. The 298 Menes (1983) 137-9 299 Conte (1986) 40 97 passage from Charisius’ Ars Grammatica reads, “for the ancients sometimes ended masculine nouns with the -es and sometimes with -a in the nominative case, such as Anchises Anchisa, Chryses Chrysa, Attes Atta; likewise Atrides Atrida, as Propertius says, ‘Not thus did Atrida [i.e., the son of Atreus] rejoice in his Dardanian victory” (nam masculina modo es modo a nominativo casu veteres terminaverunt, velut Anchises Anchisa, Chryses Chrysa, Attes Atta; similiter Atrides Atrida, ut Propertius ‘non ita Dardanio gavisus Atrida triumpho,’ Char. Ars. 85.6-10B). Barwick has developed a list of verbal tics that occur in the Ars when Charisius borrows material from Remmius Palaemon. One of these, which Bonnet endorses enthusiastically, 300 is the use of velut to introduce a list of “individual words or self-composed sentences as examples.” 301 In the passage of Charisius above, the word velut introduces the list of examples of the formation of masculine nominatives in -a as an alternative to -es. At the end of this list, Charisius quotes Propertius 2.14.1, “Not thus did the son of Atreus rejoice in his Dardanian victory” (non ita Dardanio gavisus Atrida triumpho est), but he omits est at the end of the verse. 302 An element of Charisius’ list might seem to suggest that the material from Palaemon ends at Atta: The word “likewise” (similiter) separates Atrides Atrida and the quotation of Propertius from the rest of the list and so might indicate an addition to the original thought. But, as Barwick points out, Ars 168.34-175.27, a passage similar in structure to Ars 85.6-10B, rightly belongs in his sampling of passages in which velut introduces examples which he considers drawn from Palaemon. velut occurs once near the beginning of this lengthy passage (169.2), and the examples that follow it 300 Bonnet (2009) 25 301 Barwick (1922) 112: “einzelne Worte oder selber gebildete Sätze als Beispiele” 302 This est at the end of Prop. 2.14.1 is, admittedly, controversial. While it is the unanimous reading of the older codices, a group of codices deteriores usually designated as ζ omits it as Charisius does, and Volscus’s Venetian edition of 1488 CE conjectures es instead of est and reads Atrida as a vocative (Bejarano Sánchez 1990: 28-9). Modern editors and critics are also divided. Fedeli’s Teubner, for instance, prints …triumpho est, while Goold’s Loeb prints …triumpho’s, and Bejarano Sánchez (1990) 29 believes that Charisius was accurate and the verse should end with …triumpho. 98 include individual words, quotations of poets, analyses, and transitions such as “Pomponius also” (Pomponius quoque, 170.2) and “likewise” (item, 170.7). Barwick is especially certain that this material is drawn from Palaemon: “That these last two sections [168.34-175.27 and 243-264] belong to Palaemon is supported by the notice of Consentius 375.6,” 303 which concerns Palaemon’s distinctive understanding of the relationship between the subjunctive and the optative. The other citation of Propertius by a grammarian in the first century CE is in the De Metris of Caesius Bassus. 304 The De Metris survives in fragments under the name Atilius Fortunatianus; Keil and Leonhardt identify this text with the treatise on metrics that Rufinus dates to the time of Nero, 305 and they identify the author of this text as the poet Caesius Bassus to whom Persius dedicated his sixth satire. 306 Caesius quotes Propertius 2.1.2 accurately and then quotes it again in altered form to demonstrate a metrical technique: “with respect to a heroic pentameter that begins with two dactyls, like this one ‘from what source my soft book comes into the mouth(s),’ with two long syllables having been added to the end you will make a choriambic from the heroic pentameter thus: ‘from what source now my soft book comes into these mouths’ ” ad summam pentametrum heroum, qui habet dactylos primos duos, velut hunc ‘unde meus veniat mollis in ora liber’, adiectis duabus syllabis longis facies choriambicum ex heroo pentametro sic: 'unde meus nunc veniat mollis in haec ora liber' (Mazzarino 141.250-54) This passage demonstrates a willingness to adapt poetic texts and suggests a song culture in which this kind of playing is common. In Propertius’ verse it is difficult to know whether to read ora as a poetic plural (standing in for singular) or as “lips” (Goold’s translation)—both of which might lead us to understand the verse to refer to the speaker’s mouth (“from what source my soft book 303 Barwick (1922) 113: “Die Zugehörigkeit der beiden letzten Abschnitte zu Palaemon erhält noch durch die Notiz des Consentius 375.6” 304 Leonhardt (2006) 305 When Butrica (1981) 329 writes that this is “the earliest citation of Propertius, accurate or otherwise… if [the author] is the same as the first-century poet Caesius Bassus,” he may not agree that the material in passage from Charisius discussed above can be traced to Remmius Palaemon, or he may not be considering it. 306 Keil (1874) 245-54; Leonhardt (2006) 99 comes onto my lips” or “…into my mouth”)—or whether to read ora as actually plural and so referring to multiple mouths—the speaker’s and other readers’. Richardson and Heyworth favor the latter reading, 307 and Goold does not make his choice clear, 308 but to my mind the preceding and following verses suggest that the speaker is talking about his process of composition and performance rather about than his popularity, and so the former reading is preferable: “You ask from what source my loves are written so often,… Not Calliope, not Apollo sings these things to me” (Quaeritis unde mihi totiens scribantur amores,… non haec Calliope, non haec mihi cantat Apollo, Prop. 2.1.1, 3). Caesius’ addition of the word “now” (nunc) into verse 2 specifies the scenario that Poem 2.1 creates for itself. If the addressees ask the speaker from what source his book comes into his mouth right now, then the speaker is reciting to the addressees and answering their question of the moment. In Propertius’ verse, this may be the scenario, but it may as easily be that the speaker is responding more generally to questions that people have been asking him recently. Caesius’ addition of the word “these” (haec), meanwhile, specifies that it is into the speaker’s mouth or the mouths of his immediate audience that the book comes but not the mouths of the general public. Furthermore, the word haec occurs twice in the following verse (verse 3, above), which may have inspired Caesius’ exemplum. If this is the case, Caesius is also repurposing parts of verse 3. Caesius’ change in meter is more drastic – the shift from pentameter to choriambic pulls the verse out of the genre of elegy, so the remade verse no longer participates in Propertius’ elegiac 307 Richardson (1977) 211; Heyworth (2007) 104 308 Goold’s translation (ad loc.) reads, “…how it is that my book sounds so soft upon the lips.” At the bottom of the page, Goold quotes Thomas Gray’s translation from 1742: “You ask, why thus my Loves I still rehearse? / Whence the soft Strain and ever-melting Verse.” This may be given as a comparandum or a point of personal interest, but it may also suggest that Goold prefers the same reading I do. 100 poetics, including his poetics of speech and masculinity, yet it remains recognizably Propertian and so retains its reference to the speaker of Propertius’ elegies. Both Palaemon and Caesius’ quotations of Propertius remove the poet’s words from their original contexts and set them into radically different contexts. Some ancient readers could probably remember the Propertian contexts, but some may not have been able to. The verse that Caesius quotes indicates at least something about its original context, because it is clearly talking about the speaker’s poetry, and because the adverb “from what source” (unde) and the subjunctive veniat together suggest that the verse is an indirect question. This is a far cry from the dense web of meanings in which this verse participates in its original context, and there is even less to work with in the verse that Palaemon quotes, “Not thus did the son of Atreus rejoice in his Dardanian victory” (non ita Dardanio gavisus Atrida triumpho, Prop. 2.14.1), from which a reader can only deduce that there was a happy occasion of some sort, perhaps an achievement. In their new contexts, the two verses are used not to convey any kind of elegiac poetics but rather as a grammatical example and a metrical example. The meanings of the verses do not matter to Palaemon and Caesius’ points. While Palaemon and Caesius do not actually destroy the texts, they use the verses in ways that all but void them of their original meanings. Propertius’ poetics is mostly erased from the verses, except to the extent that the reader can remember the rest of each poem, and in this way Palaemon and Caesius are similar to the accountant in Prop. 3.23, who erases the speaker’s tablets to write his ledgers on them, and to Cynthia’s ghost in Prop. 4.7, who demands that the speaker destroy what he has written about her. 101 Canonical Poets Rome’s celebrity poets made quotation, allusion, and other types of literary borrowing one of Latin poetry’s most prominent and widespread features. Propertius himself participates in these practices right from his opening verses: The first two couplets of Propertius 1.1 adapt verses 1-3 of an epigram by Meleager (AP 12.101) and, among other changes, replace the beloved boy Myiskos with Cynthia. Six verses later, in the Milanion exemplum, Propertius echoes Vergil’s 10 th Eclogue, which, in turn, is thought to reproduce words and themes from Gallus’ now lost poetry. 309 It is no surprise, then, that other poets found phrases and ideas to borrow and adapt in Propertius’ Elegies. Here I look at examples from Ovid, one of the most frequent and renowned borrowers from Propertius, and from the Consolatio ad Liviam, another early adapter of Propertian words and ideas. The first example from Ovid is one of the most written-about examples of allusion in modern scholarship on Latin poetry and occurs at the start of the Amores. It represents an especially conspicuous literary theft. The remaining examples also typify literary recycling, and the ways that borrowing happens in the famous example and the less famous examples are not very different. The main differences between them are the prominence of their positions in their respective texts and the boldness and poetic importance of the changes they make. Ovid famously borrows and repurposes words, phrases, and ideas from his friend and senior colleague, Propertius. At the beginning of Ovid’s Amores, for example, the speaker complains about Cupid’s inteference with his poetry: “the lower verse was equal; Cupid is said to have laughed and stolen a foot” (par erat inferior versus; risisse Cupido / dicitur atque unum surripuisse pedem, Ov. Am. 1.1.3-4). Keith argues that the words “is said” (dicitur) and “to have stolen” (surripuisse) should put readers or audience members on the lookout for allusions, or 309 Ross (1975) 63; Clausen (1994) 290-1; Cairns (2006b) 84 102 literary ‘thefts,’ and Hardie joins her in arguing that the “foot” (pedem) that Cupid steals alludes to the “feet” (pedibus, Prop. 1.1.4) with which Amor pushes down the Propertian speaker’s head. 310 Ovid’s pedem and Propertius’ pedibus are both metrical puns on feet, and both are the last words of the fourth verses of the first poems in the poets’ respective first books. The idea that this is a literary theft (surripuisse) suggests that Ovid’s speaker is taking possession of his Propertian model and may use it as he pleases and to his own ends rather than faithfully re-rendering it. In these verses, Ovid’s speaker performs a style of masculinity both alike to and different from the Propertian speaker’s. The Propertian speaker tells us, “Cynthia first captured wretched me with her eyes, me, before then infected with no desires. Then Love cast down my gaze of constant arrogance and pressed down my head with his feet set upon it until he taught me to hate chaste girls, the jerk, and to live without prudence.” Cynthia prima suis miserum me cepit ocellis contactum nullis ante cupidinibus. tum mihi constantis deiecit lumina fastus et caput impositis pressit Amor pedibus donec me docuit castas odisse puellas improbus, et nullo vivere consilio. (Prop. 1.1.1-6) The speaker here is abject and dominated both physically and mentally by Love. The Ovidian speaker, on the other hand, says this: “Arms in a serious meter and wars I was preparing to tell, the material suiting the measure. The lower verse was equal; Cupid is said to have laughed and stolen a foot. ‘Who, savage boy, gave you this authority over songs? We poets are the Muses’, not your crowd...’” Arma gravi numero violentaque bella parabam edere, materia conveniente modis. par erat inferior versus; risisse Cupido dicitur atque unum surripuisse pedem. ‘quis tibi, saeve puer, dedit hoc in carmina iuris? Pieridum vates, non tua turba, sumus...’ (Ov. Am. 1.1.1-6) 310 Keith (1992) 340-4; Hardie (2002) 36-7 103 Ovid’s speaker adopts a posture similar in some ways to that of Propertius’ speaker. Cupid has interfered with his meter in the same bodily pun as in Propertius, and he has claimed authority over the speaker’s poetry. But in Ovid the tone is playful and light rather than (at face value, at least) woeful and intense. The speaker is not dejected; he even chides Cupid in legalistic language (iuris). When he submits to Cupid’s force later in the poem (Ov. Am. 1.1.21-30) and echoes Propertius’ language of abjectness saying, “wretched me” (me miserum, Ov. Am. 1.1.25, cf. miserum me, Prop. 1.1.1), the Ovidian speaker’s tone remains playful. He says, “I am burned, and Love reigns in my empty chest” (uror, et in vacuo pectore regnat Amor, Ov. Am. 1.1.26), but the next verses, the last four in the poem, focus on the effects on his poetry, in meter and so in subject, of Love’s reign. Verse 4 presages these last verses with the word dicitur, an “Alexandrian footnote” 311 which “transports the reality of Cupid’s epiphany to a realm of tradition and textual allusion.” 312 Also, the speaker seems readier to celebrate than to mourn, as he says, “bind your temples with yellow myrtle from the shore, Muse, and sing on in eleven feet” (cingere litorea flaventia tempora myrto, / Musa per undenos emodulanda pedes, Ov. Am. 1.1.29-30). In contrast, the Propertian speaker specifies how Love has taken over his entire life: “he taught me to hate chaste girls, the jerk, and to live without prudence” (...me docuit castas odisse puellas / improbus, et nullo vivere consilio, Prop. 1.1.5-6), and “our Venus works bitter nights in me, and at no time does empty Love wane” (in me nostra Venus noctes exercet amaras, / et nullo vacuus tempore defit Amor, Prop. 1.1.33-4). Then the Propertian speaker ends the poem with a warning that culminates in the words “my pain” (dolore mea, Prop. 1.1.38). He is of course playing with literary allusion and changes, too: his 311 Hinds (1998) 1-2 cited in Hardie (2002) 36 312 Hardie (2002) 36 104 first two couplets adapt part of a poem by Meleager (Anth. Pal. 12.101). But the Propertian speaker maintains a consistently dejected attitude as a layer over the allusive play in his first poem, whereas the Ovidian speaker borrows his wording to gleefully and conspicuously point out his own cleverness. In doing so, the Ovidian speaker claims primacy in the genre of erotic elegy and refigures the history of the genre, diminishing Propertius’ place in the genre just as Propertius had done to Gallus in Book 1 of the Elegies. 313 Some cases of Ovidian borrowing might more accurately be described as participation in literary conversations (or even struggles) about words and phrases that carry heavy ideological loads. About a third of the way through Book 3 of the Ars Amatoria, for example, Ovid suggests ways that women who do not meet his standards of beauty can hide their physical imperfections. One recommendation is, “let a bandage roll go around a scrawy chest” (angustum circa fascia pectus eat, Ov. Ars 3.274). The words “scrawny chest” (angustum pectus) here echo the “slender chest” (angusto pectore) with which Callimachus and the Propertian speaker, who identifies with him, sing in Poem 2.1: “But neither the Phlegraean din of Jove and Enceladus would Callimachus with his slender chest thunder forth nor is my diaphragm suited, with tough verse to establish Caesar’s name among his Phrygian forefathers.” Sed neque Phlegraeos Iovis Enceladique tumultus intonet angusto pectore Callimachus, nec mea conveniunt duro praecordia versu Caesaris in Phrygios condere nomen avos. (Prop. 2.1.39-41) Propertius himself seems to have drawn the phrase from Cicero’s discussions of rhetorical training and technique, as Keith argues. 314 In the In Pisonem, Cicero mocks Piso for his “skinniness of 313 See Chapter 2. 314 Keith (1999) 52ff., and see “Body Terms Borrowed from Oratory” in Chapter 5. 105 chest” (angustiae pectoris, Cic. Pis. 24), which he links to weakness of spirit and intellect. 315 In the Brutus, Cicero characterizes Atticist oratory as “scrawny and feeble” (anguste et exiliter dicere, Cic. Brut. 289), and in the Orator he advises that an Atticist speaker should not be trained or expected to attempt dramatic rhetorical figures and long periods, lest his flanks—that is, his breath support apparatus—be overwhelmed. 316 The slender chest of the Propertian speaker also recalls Vergil’s diminutive but stout bees in Georgics 4. Vergil praises the king bees’ courage in battle, which belies their tiny chests (ingentis animos angusto in pectore versant, Verg. Geo. 4.83); Vergil thus valorizes the angustum pectus in the context of Callimachean poetics. 317 Whereas Vergil’s bees have great courage in battle despite their tiny chests, in Propertius’ elegiac lover in Poem 2.1 claims the angustum pectus for himself while recusing himself from military service or even poetry about war. Instead, Propertius’s speaker compares himself to Callimachus in order to valorize his angustum pectus and so the slender and unwarlike elegiac poetics and masculinity of which his angustum pectus is an emblem. Ovid turns this vindication on its head when he advises the puella to bulk up her angustum pectus with a bandage roll to make herself more sexually attractive. 315 Cic. Pis. 24: “Great is the name, great is the semblance, great is the dignity, great is the majesty of a consul; the slightness of your chest cannot comprehend it, your lightness cannot take it in, nor your poverty of mind; not the weakness of your intellect, not your inexperience of favorable circumstances can support so great, so weighty, so serious a character” (magnum nomen est, magna species, magna dignitas, magna maiestas consulis; non capiunt angustiae pectoris tui, non recipit levitas ista, non egestas animi; non infirmitas ingenii sustinet, non insolentia rerum secundarum tantam personam, tam gravem, tam severam). 316 Cic: Or. 85 “He will not make the Republic speak nor raise the dead from the lower regions nor tie together many things in a heap in one period, crowding them in. These are for stronger flanks and we must not hope for them or require them from this man when we educate him…” (Non faciet rem publicam loquentem nec ab inferis mortuos excitabit nec acervatim multa frequentans una complexione devinciet. valentiorum haec laterum sunt nec ab hoc quem informamus aut exspectanda aut postulanda...) 317 The Fourth Georgic announces its Callimachean poetics in its proem: admiranda tibi levium spectacula rerum (Verg. Geo. 4.3) and (in tenui labor; at tenuis non gloria…, Verg. Geo. 4.6). 106 Another early adapter of Propertian phrases is the Consolatio ad Liviam. 318 Shackleton Bailey has identified three “possible imitations or reminiscences of Propertius” in the Consolatio, 319 and I will discuss two of them here. The clearest borrowing, in verse 313, reworks Prop. 4.1.69: “rivers and mountains and grand names of places” fluminaque et montes et nomina magna locorum (Cons. Liv. 313) and “I shall sing rites and gods and ancient names of places” sacra deosque canam et cognomina prisca locorum (Prop. 4.1.69) 320 Both verses begin with nouns that are the objects of verbs of narration followed by -que, and the differences between the last four words of each verse are only the subtraction of a prefix (co[g]-) and the change of an adjective (prisca to magna). The Consolatio sympathizes with Livia here for the loss of her son and for how he will not be able to tell her about all that he saw in his military adventures. This is a change from the Propertian verse, in which the speaker announces the planned subjects of his forthcoming poetry. Furthermore, the theme of the Propertian narrator’s hypothetical narration is traditions, but the Consolatio changes the theme to geography. The difference between the Propertian phrase “ancient names of places” (cognomina prisca locorum) 318 The date of this poem is contested, but most put it in the early or middle first century CE: Richmond (2006) writes, “A reference to the temple of Castor and Pollux in Rome (283-288) suggests a composition date after 6 AD. The probable imitation of Ovid’s poems of exile puts the terminus post quem at AD 12. The restrained language of this insipid poem, its dignified style and metric quality argue against its being post-Augustan;” Schrijvers (1988) 382-4 suggests that the poem dates to 20 CE as a response to the death of Germanicus; Schoonhoven (1992) 22ff. argues that the poem was written to support Britannicus’ claim to the throne, so under Nero’s reign but while Britannicus was still alive, i.e. 54 CE, but Butrica (1993) 265 finds that “nothing adduced in proof is incompatible with an Augustan date,” while Pinotti (1996) 501 thinks that none of Schoonhoven’s evidence contradicts Schrijvers’s suggestion. 319 Shackleton Bailey (1952) 307-8 320 Fedeli’s Teubner text reads sacra diesque…, but I am persuaded by Heyworth’s (2007: 423-4) argument for sacra deosque… Sullivan (1976) 141 and Goold’s 1992 Loeb agree. Wellesley (1969) 96 seems to have presented the idea first. 107 and the phrase “grand names of places” (nomina magna locorum) in the Consolatio highlights this change. In Propertius the adjective “ancient” (prisca) belongs both in form and sense with the noun it modifies, “names” (cognomina). In the Consolatio the adjective “grand” (magna) is a transferred epithet, belonging in form with the noun it modifies, “names” (nomina), but in sense with the noun “places” (locorum). The Consolatio also recycles words from Prop. 4.3.10 and changes their focus in verse 386: “and darkened Indians, burned by Eastern water” ustus et Eoa decolor Indus aqua (Prop. 4.3.10) 321 and “darkened Isargus, a witness with stained water” decolor infecta testis Isargus aqua (Cons. Liv. 386) Both verses contain the same elements: 1. a proper noun that begins with I (Indus and Isargus) 2. modified by decolor and a second, disyllabic modifier (ustus and testis), and 3. aqua in the ablative separated from its adjective (Eoa and infecta). The word order has changed only by switching the positions of decolor and the other modifier of the river (ustus…decolor and decolor…testis). 322 In the Propertian verse, a woman who gives herself the pseudonym Arethusa complains that various foreign peoples have gotten to see her husband on his military campaigns (te modo viderunt, Prop. 4.3.7), while she has not. Her list of foreign peoples expresses frustration, grief, and loneliness. The Consolatio, on the other hand, lists faraway lands to comfort Livia by reminding her of the accomplishments of her family. The Consolatio shifts the words from a human focus to a 321 This verse has probably come down to us with at least one corruption. N F L P D V Vo have discolor, but ς (the codices deteriores) has decolor, and modern editors unanimously (or virtually unanimously) prefer this latter reading. More controversial is the first word of the verse: Ω probably had ustus, but this has dissatisfied many editors, and conjectures have included tunsus or tusus (Housmann, followed by Shackleton Bailey and Goold), tinctus (Postgate), and mixtus (Cameron). Hutchinson (2006) 103 suggests, “perhaps some different kind of word, like say vidit, has been corrupted.” 322 These similarities suggest that the verse in the Consolatio is patterned after Prop. 4.3.10 rather than Ov. Ars 3.130 (quos legit in viridi decolor Indus aqua) or Ov. Trist. 5.3.24 (et quascumque bibit decolor Indus aquas). 108 geographical one here, as well. The word Indus in Propertius 4.3 can mean an Indian person or the Indus River. Verses 8-9 and possibly 7 focus on peoples rather than geographical features, 323 so it fits the sequence to read Indus in verse 10 as the people rather than the river. The focus on peoples also makes sense in the context of Arethusa’s complaint: while geographical features could, through personification, have “seen” Arethusa’s husband, the expression is more direct and vivid if Arethusa is complaining that other people have seen her husband. The Consolatio uses Propertius’ words, but, though it is listing “beaten enem[ies]” (pulsus…hostis, Cons. Liv. 384), it shifts the focus to rivers, valleys, and other places in verses 385-91. The word in verse 386 that replaces Indus is Isargus, which is unambiguously a river. Someone who remembers Propertius’ words while reading the Consolatio could thus easily take Indus in Prop. 4.3.10 as the Indus River. This reader would then remember Arethusa’s complaint as less direct and more figurative. Poetry in Graffiti and Other Inscriptions About 100 versified graffiti and other inscriptions reuse words and phrases from Propertius, and about 85 share what look like oral formulas with Propertius’ Elegies (there is some overlap – i.e., some graffiti and inscriptions contain both intertexts and formulas). 324 I discuss formulas in 323 This is true of verses 8-9, whichever text of 8 one prefers (Britannia in verse 9 is a place, but the detail with which the speaker characterizes it, the “painted chariot” [picto…curru] is anthropological). Verse 7, which begins the sequence, however, may or may not fit this pattern: Ω read “Bactra has recently seen you in its well-traveled East” (te modo viderunt iteratos Bactra per ortus), which Barber and Fedeli accept in their OCT and Teubner texts and Shackleton Bailey (1953) 229-30 defends. Richardson and Hutchinson print it but do not accept it, and neither does Günther (1997) 71-2. Housman (1972) 277f. suggests Ituraeos viderunt Bactra per arcus, which Morgan (1986) 189f. changes to te modo viderunt intentos Bactra per arcus, and Goold prints this text. I find the Ω text suspect, expecially because N does not have Bactra per ortus, and Hutchinson (2006) 104 notes that “words, phrases, and lines that do not appear in N show signs of spuriousness with remarkable frequency.” I can see good reasons to join Richardson, Hutchinson, and Günther in throwing up my hands at this verse, but I also find Morgan and Goold’s text plausible and attractive. If Morgan and Goold are correct, then verses 7-9 give a sequence of foreign peoples who have seen Lycotas; if they are not correct, then verses 8-9 give a sequence of foreign peoples who have seen Lycotas. Either way there is a list of peoples preceding verse 10, which suggests that Indus in verse 10 should be taken to mean “Indians” (literally “an Indian”) rather than the Indus River. 324 See Appendix A for a full list. I cite approximate numbers here to acknowledge the uncertainty of some of these intertexts. There is not yet a centralized, searchable database of Latin inscriptions, so I have gathered my collection 109 Section C. Here I discuss five examples that resemble allusions in the works of famous poets, such as those discussed above. Three of the examples are erotic graffiti from Pompeii, which have several advantages for my study: They were produced close in time (before 79 CE) and space to Propertius’ elegies, two of the three survive mostly intact, and much is known or can be deduced about their original physical contexts. Epitaphs are the other most common source of epigraphic poetry. I focus here on two examples from Rome itself, CIL VI 6592 and CIL VI 29869, because the former is close in time and place to Propertius, and the latter is especially striking with its complex allusions and its play with the ideologically loaded word doctus. The graffitists of Pompeii and Herculaneum produced a wealth of Latin poetry which the ashes of Vesuvius have preserved, including their borrowings from Propertius and other poets. One such graffito was found etched into the plaster of a wall in the atrium of House of the Scientists (6. 14. 43), 325 which Milnor describes as “one of the large and well-appointed atrium houses in the wealthy north-west quadrant of the city.” 326 Here is a transcription based on the line drawing in the CIL: 327 CANDIDAMEDOCVIT NIGRAS ODISSEPVELLAS · ODERO·SEPOTERO·S[ED]NONINVIT[U]S AMABO· SCRIPSIT VENVS· FISICA·: POMPEIANA The CIL gives the edited text of the couplet as follows with my translation: Candida me docuit nigras odisse puellas: Odera s[i] potero, sed non invitus amabo (CIL IV 1520) “Fair in complexion, she taught me to hate black girls: I’ll hate them if I can, but not unwillingly shall I love them.” from various sources, especially the Carmina Latina Epigraphica, the EAGLE Epigraphic Database, and Lissberger’s study, Das Fortleben der römischen Elegiker in den Carmina Latina Epigraphica. 325 CIL IV 1520; Varone (2002) 56n71; Milnor (2014) 92, who lists the House of the Scientists as 5. 14. 43 (perhaps a slip of the keyboard, as her map on p. 16 shows Block 14 in Region 6, where Varone places the House of the Scientists). 326 Milnor (2014) 92 327 Appendix B Figure 4 110 The CIL does not give an edited text of the signature, “Venus fisica of Pompeii wrote this” (scripsit Venus fisica Pompeiana). The graffito is scratched in a stylish script with many letters larger than others, often by virtue of long, curving tails. The first line is slightly indented, and the third line is more substantially indented. The A at the start of amabo has a prominent right angle (one of the few right angles in the inscription, and the others are much smaller), which forms a box larger than any other letter, though the space inside the curve of the C in candida would be about as large if the A following did not interrupt it, and the longest stroke in the graffito is the C in fisica (but the A in fisica and the P in Pompeiana intersect it a total of 3 times and break up its visual impact). 328 There are interpuncts in the spaces between puellas and odero, odero and sepotero, and sepotero and sed, and also following amabo, Venus, and Fisica. 329 The whole inscription is bounded by two parallel, vertical lines in which the plaster is, or at least once was, painted red. 330 The text of the inscription forms two hexameters broken up across the first three lines and followed in the fourth line by the non-metrical signature. The speaker of the graffito expresses a preference for fair-complexioned girls over black or dark-complexioned girls but not a very serious preference. After all, he says in the second verse that he will not turn down a black girl if she is all he can get. He is a connoisseur but not to the point of impracticality. The hexameter of the graffito recalls verse 5 from Propertius 1.1: “until he taught me to hate chaste girls” (donec me docuit castas odisse puellas, Prop. 1.1.5) by placing me docuit and odisse puellas in the same metrical positions, but the erotic preferences are not the same. Whereas the speaker of the graffito expresses a preference for fair girls over dark girls, Propertius’ speaker in 1.1 expresses hatred for chaste girls. Furthermore, the graffito is cheerful 328 Milnor (2014) 92 includes a line drawing of the inscription from CIL IV. 329 CIL IV 1520 330 CIL IV 1520 111 in tone, while the tone Propertius 1.1 is of suffering and pleasure mixed. Veyne argues that the fictiveness of Propertius’ stereotyped Liebesschmerz makes Propertius’ self-presentation into a joke, 331 and if this is the case then the tone of Propertius 1.1 is operating in two layers, one of Liebesschmerz and one of humor. The graffito replaces the lover’s agony with erotic confidence. In the second verse, the speaker reveals that his hate for black girls is only casual and situational, and the prominence of the word “I shall love” (amabo)—alone on its line and beginning with a large and geometrically striking A—emphasizes the message that, one way or another, the speaker will make love. The graffito also alludes to two poems from Ovid’s Amores: A fair girl will capture me, capture me a blonde girl will, but Venus is also welcome in a dark color. If dark locks hang on her snowy neck, Leda was especially admired with her black hair; if they are blonde, Aurora was pleasing with saffron locks. Candida me capiet, capiet me flava puella est etiam in fusco grata colore Venus. Seu pendent nivea pulli cervice capilli, Leda fuit nigra conspicienda coma; seu flavent, placuit croceis Aurora capillis. (Ov. Am. 2.4.39-43) and They fight and stretch my fickle chest in opposite directions; here is love, here is hate, but, I think, love is winning. I’ll hate her if I can; if not, I’ll love her unwillingly. Luctantur pectusque leve in contraria tendunt hac amor hac odium, sed, puto, vincit amor. Odero si potero, si non, invitus amabo. (Ov. Am. 3.11.33-5) From Amores 2.4.39 the graffito borrows the words candida me 332 to begin its hexameter and puella(s) to end it, but while in Ovid’s poem the speaker is the object of capture by the fair girl, in 331 Veyne (1988) 26, 29-37 332 Putting me in the second position is standard Latin word-order, so whileit is notable that the graffito repeats candida at the start of a hexameter is notable, me in the graffito is notable in that it is borrowed along with candida, 112 the graffito he is the recipient of her teaching—she shapes his preference and agency, but he still prefers and acts, in contrast to the speaker of Ovid’s poem, who postures himself as having lost his agency for a moment, or at least having let it slip out of the reader or audience’s focus. Ovid’s following line introduces the possibility of loving a dark girl, but unlike the speaker of the graffito, the speaker in Ovid does not express any particular distaste for dark girls. The ideas are similar, but Ovid’s speaker is more gracious, while the speaker of the graffito is more aggressive. The pentameter of the graffito reworks a verse from Amores 3.4. Ovid’s verse, “I’ll hate her if I can, but if not I’ll love her unwillingly” (Odero si potero, si non, invitus amabo, Ov. Am. 3.4.35), shows an erotic ambivalence reminiscent of Propertius’. The pentameter of the graffito, as noted above, is more cheerful and sexually confident, a change it effects by changing Ovid’s “if not” (si non) to “but not” (sed non). Furthermore, by changing Ovid’s first si to se, the graffito replaces the formally classical spelling with a colloquial variant. The graffito’s speaker, then, mashes up (to borrow a term from contemporary music) Propertius and Ovid’s words but yet differs from Propertius and Ovid’s speakers: He takes pride in his taste in girls but will let lust override taste if need be; he is more cheerful and sexually confident than Propertius and Ovid; and he is more aggressive and less formal or proper than Ovid (and also than Propertius, who does not use se as a variant for si). The graffito follows the two hexameters with a signature that complicates its speaker’s performance of masculinity: “Venus fisica of Pompeii wrote this” (scripsit Venus fisica Pompeiana). The CIL does not know what to make of this, 333 and Varone notes its presence but offers no further comment. 334 The word for “wrote,” scripsit, may indicate that this Venus so that the graffito begins with the same idea as Amores 2.4.39, a relationship between a fair girl and the speaker in which she is subject and he is the object. 333 CIL IV 1520 334 Varone (2002) 56-7n71 113 composed the text of the graffito, composed the text and scratched it into the wall, or scratched the text, which someone else had composed, into the wall. 335 The exact meaning of fisica, meanwhile, is totally uncertain. 336 Milnor offers an appealing but unlikely solution: “CIL 4. 1520 is a couplet with an erotic theme signed by ‘Venus Fisica Pompeiana’, so that Pompeii’s own goddess of bodily love is made to claim authorship of a poem about bodily love.” 337 This is not what fisicus usually means, though. Rather, it refers to ideas about nature and what is natural, and the only surviving literary passage from before late antiquity that connects it to bodies needs the phrase “nature and natural workings” (naturam et naturales operationes) to clarify the connection. 338 Adding to the confusion, Venus and fisica are feminine, while the gender of invitus in the second hexameter indicates that the speaker is masculine. Perhaps this is a woman writing in a masculine poetic voice, or perhaps it is a man writing as a woman writing in a masculine poetic voice. Wyke 339 and Miller 340 have written about the speakers of the famous elegists, especially Propertius, doing what the sociologist Barrie Thorne calls “crossing,” which is “the process by which a girl or boy may seek access to groups and activities of the other gender,” 341 but it is difficult to know which of Wyke and Miller’s ideas might apply here. Veyne’s argument about 335 OLD scribo 1-2, 12 336 The Thesaurus Linguae Latinae writes “concerning gods (namely, for a nickname; we list passages here uncertainly…)” [de deis (sc. pro cognomine; locos huc dubitanter afferimus...) and gives CIL IV 1520 as an example (TLL 10.1.2063.21ff.). I have found the adjective fisicus in three other inscriptions (CIL IV 6865, CIL X 203, and CIL X 928), none of which provides helpful context for this word. 337 Milnor (2014) 150 338 “the fisica part of medicine is that which gets to know the nature and natural workings of the body” (fisica pars medicinae est, quae corporis naturam et naturales operationes cognoscit, ps.-Soranus Quaest. med. 25). Lucilius and Cicero use the Greek spelling to discuss natural philosophers’ studies of bodies, but these two passages cannot support reading fisica as “bodily”: “In the first place, all natural philosophers say that a person consists of the soul and the body” (principio physici omnes constare hominem ex anima et corpore dicunt, Lucil. 635); “as not only the natural philosophers teach, but truly also the doctors, who have seen [bodies] opened up and revealed” (neque est enim ullus sensus in corpore, sed, ut non physici solum docent, verum etiam medici, qui ista aperta et patefacta viderunt, Cic. Tusc. 1.46). Cf. TLL 10.1.2064.9-10, 16-7, and 45-7. 339 Wyke (2002) 155-91 340 Miller (2004) 130-59 341 Thorne 1993: 122 114 fictiveness seems more obviously useful: If we interpret the sphragis of CIL IV 1520 in either of the two ways I have suggested—that is, as indicating that the speaker is a woman writing in a masculine voice or a man writing as a woman writing in a masculine voice—then the sphragis highlights the fictiveness of the gender performance. The graffito suggests that the performances of masculinities by Propertius and Ovid’s speakers have some usable elements, but for some reason they do not suit the graffito’s speaker exactly, and so he or she modifies them. If Veyne is right that “the Roman elegists liked to describe general ways of doing things through the fiction of the first person singular” 342 and that elegy is didactic and normative, 343 then the speaker of the graffito is criticizing Propertius and Ovid’s performances of masculinity by changing them. Then the speaker of the graffito destabilizes his or her own gender performance with the sphragis, perhaps. An almost identical graffito was found to the left of a doorway leading from the garden and vineyard into a room with a counter in the Caupona of Euxinus (1.11.10-11): 344 candida me docuit nigras o[d]isse puellas | odero si potero si non invitus amabo (CIL IV 9847) 345 This graffito forms two hexameters, the first of which is the same as the first hexameter of CIL IV 1520, while the second differs slightly but significantly from the second hexameter of CIL IV 1520: Fair in complexion she taught me to hate black girls. I’ll hate them if I can; if not, I’ll love them unwillingly. candida me docuit nigras odisse puellas odero si potero si non invitus amabo 342 Veyne (1988) 39 343 Veyne (1988) 54 344 Jashemski (1967) 43-4; Varone (2002) 56n71; Milnor (2014) 87ff. 345 This graffito and the graffiti and images around it were done in paint, and unfortunately they were left unprotected after their excavation in 1955. Air and rain had washed them away by 1964, when Jashemski excavated, and no photographs had been taken (Milnor 2014: 88-9). Milnor (2014) 89 records slightly different restitutions: candida me docuit nigras o[di]sse / puellas | odero si potero [s]i non / invitus amabo. 115 The second hexameter in this graffito quotes Ovid’s Amores 3.11.35 346 verbatim instead of changing it, as the second hexameter in CIL IV 1520 does. In CIL IV 9847, the speaker maintains the Ovidian erotic ambivalence, and the visual arrangement of the graffito heightens this effect. The two hexameters are spread across three lines with the last two words alone on the third line. This placement “isolate[s] and underscore[s] the paradoxical invitus amabo, which appears in the transcription provided by the Notizie degli scavi neatly centred below the other two lines and in letters perhaps slightly larger than the rest.” 347 This graffito and CIL IV 1520, then, use the same visual techniques (placement and size) to underscore meanings that their second hexameters produce, and these meanings are nearly opposite. Though it expresses different meanings than CIL IV 1520, CIL IV 9847 still changes Propertius’ text. Propertius’ words, of course, are changed as in CIL IV 1520 so that the speaker of the graffito says that he hates black girls instead of chaste girls, and he tells us that his beloved is fair, whereas the Propertian speaker tells us nothing about Cynthia’s appearance in Poem 1.1. In tone, CIL IV 9847 is closer than CIL IV 1520 to Propertius 1.1: The lover is ambivalent, perhaps even tormented. CIL IV 9847 also interacted with its physical surroundings to produce meanings that further differ from Propertius’ poetics in Book 1. 348 Under it was painted, 349 Here two rivals sing. One girl holds the fasces Tr ---------------s for whom it is right Ar------------------ae Hic duo rivales ca<n>ont Una puella tenet fasces Tr [---------------]s cui fas 346 See above. 347 Milnor (2014) 89-90 348 See Appendix B Figures 5A and 5B for a plan and photo of the findspot. 349 Milnor (2014) 89 116 Ar[------------------]ae (CIL IV 9848) To the right of the door was painted a hexameter: “We have come here desirous; we desire to leave much more” ([venimus h]oc cupidi, multo magis ire cupimus, CIL IV 9849). 350 This verse has been found, whole and in fragments, 9 times in Pompeii and once in Herculaneum, “in material contexts which range from the peristyles of wealthy houses to the walls of cookshops,” twice with (different) pentameters added to the hexameter. 351 It is difficult to say in what order these three graffiti were painted and whether they were painted together or separately. CIL IV 9847 and 9848 were both written in red paint, 352 and their placement, 9848 under 9847 and both under a picture of Priapus, suggests that they may have been written together and that 9848 may have followed 9847. If this is correct, then we can read 9848 as a comment that advises the reader on how to engage with 9847. 353 Milnor has suggested that 9848 “evoke[s] the bucolic atmosphere associated with [singing contests],” 354 especially since these graffiti are on a wall of a vineyard with two large trees, an arbor, and two altars, and 9847 and 9848 were painted under a picture of Priapus, while 9849 was painted under a picture of Bacchus. 355 Jashemski and Milnor also discuss the similarities of this setting to the depiction of an urban-bucolic picnic in the garden of a taberna in the pseudo-Vergilian Copa and suggest that bringing a little bit of the country into the city may have been a popular way of playing with 350 Jashemski (1967) 43; Milnor (2014) 89-90. 351 Milnor (2014) 182-4. Milnor discusses the variants of the poem and the theory that the hexameter is a quotation from a source that no longer survives. She gives the CIL numbers IV 1227, 2995, 8114, 8231, 8891, 9849, 10065a, 10640 and, for the instance of the hexameter with a pentameter in the House of Maius Castricius, cites Solin (1975) 252. 352 Milnor (2014) 89 353 For explorations of how ancient Roman paratexts can direct a reader’s engagement with the respective ‘primary’ texts and other relationships between paratexts and their ‘primary’ texts, see Gibson (2014); O’Rourke (2014), which considers this phenomenon in Propertius’ elegies; and other essays in The Roman Paratext, Ed. L. Jansen. 354 Milnor (2014) 90 355 Jashemski (1967) 40-44; Milnor (2014) 88-90, 93 117 conventions of space, behavior, and genre. 356 The meter of the couplet—both verses are hexameters—may also recall Vergil’s Eclogues, given the couplet’s erotic themes and countryside setting. Placing an adaptation of Propertius’ words from Poem 1.1 in the setting of a bucolic song contest further changes Propertius’ words by changing their context to one alien to Propertius’ poetics in Book 1. While Book 1 sometimes sets the elegiac lover in wilderness locations, Propertius does not explore the elegiac possibilities of the more domesticated bucolic countryside until Book 2, following Tibullus’ introduction of the rusticating elegiac lover in his first book of elegies a year after Book 1 of Propertius. 357 Also, the image of Priapus may call into question the sincerity of the speaker’s erotic ambivalence in CIL IV 9847, given the ithyphallic god’s notorious sexual aggression. This would further undermine Propertian poetics by openly displaying the insincerity of the lover’s posturing 358 rather than leaving it as a joke that lurks behind and between multiple layers of meaning. The visual context of the graffito creates “paratextual ‘special effects’” that limit the play of signification of the words by themselves 359 but replace it with play between the words and their visual surroundings. The text of CIL IV 9847 changes the meaning of Propertius’ words, and their verbal and visual surroundings in the Caupona of Euxinus bring further influences to bear. This, of course, is a text-centric analysis. 360 Bergmann rightly asks, “where… does the balance lie between art and text?” and answers that, “since we will never know the intentions of 356 Jashemski (1964) 332-49; Milnor (2014) 90-91. See also Cutolo (1990) 115-19 on the Copa. 357 Keith (2008) 68-9 358 See discussion in Chapter 2 of insincerity in Poems 1.2 and 1.19 held against the speaker in Poems 4.5 and 4.7. 359 O’Rourke (2014) 174, and cf. Habinek (2009) 136, part of which O’Rourke quotes: “…in separating writing from its connection with speech, using it to defamiliarize processes of visual and auditory perception, at least some Romans expose the materiality of the word, its groundedness in the realm of the phenomenal. Roman writing, to be sure, is an aid to signification. But more often than we are accustomed to acknowledge, it denies the freedom of the signifier and limits production of meaning to direct encounters with the very system of inscription recognized as such.” 360 See Squire (2009) for a thought-provoking analysis of modes of viewing and reading images in relation to words. 118 the ancient patron or painter, answers to these questions must depend upon who was, or is, looking.” 361 Bergmann suggests elsewhere that viewers of a painting of Polyphemus in room z of the Casa del Marinaio in Pompeii may have sung the cyclops’s serenade, inspired by recollections of Theocritus and Ovid’s versions of his love for the nymph Galatea. 362 Viewers of CIL IV 9847 may similarly have recited Propertius and Ovid and tried out their own variations on the canonical texts, perhaps even entering into a ‘bucolic’ song contest. These people’s responses to the graffiti, the paintings, the garden, the wine, and any number of other factors may have interacted with Propertius’ poetry to produce ephemeral variations far beyond his ability to anticipate, let alone account for. Whether we center the text, the paintings, or the vineyard or analyze the scene in a more egalitarian way, the overall effect on Propertius’ words in CIL IV 9847 is the same: Authority over the production of meaning is transferred from the original author of the words, Propertius, to their adapter(s) in the graffito 363 to the creator(s) and visitors of the garden with all its plants, pictures, and poetry set up as entertainments in a tavern. Another graffito employs a phrase that probably originated as a Propertianism but reworks and criticizes Propertian (and Ovidian) elegiac poetics: 364 361 Bergmann (2007) 60 362 Bergmann (1999) 93 and see Squire (2009) 319-21 for further discussion. 363 The person who inscribed the CIL IV 9847 may not be the person who composed the couplet. Given that a very similar couplet, CIL IV 1520 was inscribed in another part of the city entirely (the atrium of a wealthy house in Region 6 versus the garden of a tavern in Region 1), it is possible that at least one of them is not “the original.” The differences between CIL IV 9847 and 1520 indicate that, if one or both were based on a model text, then one or both do not follow the model exactly, and this would add yet another layer of authorship. As Milnor (2014) 204 notes, “there is significant evidence that many of the poetic fragments found on Pompeian walls originated in other media, either written texts or ‘popular’ oral performances. Thus, those who first composed the verses may very well not have been those who inscribed them. It would be a mistake, however, to underestimate the role of the inscriber in the creation of the text as we have it: in the same way that, for instance, artists who painted the frescoes on the walls of elite houses were often working from models and yet were able to imprint their own style on individual images, so too those who wrote poetic graffiti were not mere copyists.” 364 See Appendix B Figure 6 for a line drawing. 119 OVTINAM·LICEAT· COLLO· COMPLEXA TENERE · BRACIOLA·ET·TENERIS OSCVLA·FERRE·LABELIS·INVNC·VENTIS·TVA·GAVDIA PVPVLA· CREDE CREDE·MIHI·LEVIS· EST NATVRA·VIRORVM· SAEPE·E GO CV MEDIA VIGILARE · PERDITA · NOCTE·HAEC· ME CVM· METIDAS MVLTOS FORTVNA·QVOS·SVPSTVLIT· ALTE · HAS MODO · PROIE CTOS SVBITO· PRAECIPITESQVE PREMVT· SIC· VENVS·VT· SVBITO· COIVNXIT CORPORA·AMANTIVM·DIVIDIT· LUXET ·SE [The last line of the inscription has 3 words. The middle word is QVID, followed by an interpunct. The other two words include letters and symbols that are distinguishable, but which I cannot interpret.] While the meter of this graffito is irregular, verses are still readily recognizable: Oh, if only it were permitted to grasp with my neck your little arms, when they’ve wrapped around it, and to give kisses to your little lips. Go now, darling girl, trust your joys to the winds. Trust me, the nature of men is fickle. Often while I have kept my vigil, lovelorn, in the middle of the night, you ponder these things with me: many whom Fortune raised on high, these now, cast down suddenly and head-over-heels, she oppresses. Thus as Venus on a sudden joined lovers’ bodies, the light divides them and se… O utinam liceat collo complexa tenere brac<ch>iola et teneris oscula ferre label<l>is. I nunc, ventis tua gaudia, pupula, crede. Crede mihi, levis est natura virorum. Saepe ego cu<m> media vigilare<m> perdita nocte, haec mecum medita<ri>s: multos Fortuna quos supstulit alte hos modo proiectos subito praecipitesque premit. Sic Venus ut subito coiunxit corpora amantum, dividit lux et se... 365 (CIL IV 5296) The inscription begins with a phrase, o utinam, which Propertius was the first in the surviving literature to use at the start of a hexameter (the only surviving occurrence of the phrase from before Propertius is in Turpilius 125, a fragment of a comedy). Propertius begins five hexameters with the phrase, 366 and then Ovid begins 13 with it. 367 After Ovid, o utinam begins hexameters by 365 I follow Milnor (2014) 197-8 in using the text given at Sogliano (1888) 519, the original excavation report of this graffito in Notizie degli scavi. Sogliano emends meditas to meditaris, which produces a sentence with a main clause, whereas Mau’s (1889) 122-3 suggestion of meditans yields a sentence without a main clause. The other sentences in the graffito are complete. Milnor notes that Pompeian graffiti often use non-deponent forms of verbs that are deponent in the classical dialect. 366 Prop. 1.3.39, 1.8.9, 1.16.27, 4.4.33, 4.4.51 367 Ov. Am. 2.5.7, 2.11.5, 2.15.9, 3.6.67, Fasti 3.477, Her. 1.5, 4.125, 11.21, 19.115, Met. 1.363, 3.467, 8.501, Trist. 4.4b.33 120 Lucan, Valerius Flaccus, and Statius 368 as well as in seven inscriptions including CIL IV 5296. 369 It also occurs eight times in verses other than hexameters from Statius and earlier. 370 Given the amount of Roman literature that has not survived, it is impossible to know for certain whether Propertius was the first to use o utinam at the start of a hexameter. But the facts that in the surviving literature only Propertius uses it before Ovid, who uses it 13 times, and that various other poets then begin to use it, suggest that it was originally a Propertianism. Based on this phrase alone, I cannot say whether the author of CIL IV 5296 had Propertius, Ovid, another poet, or no poet in particular in mind. 371 The rest of the graffito, though, plays with elegiac themes and images—intertwining arms, lips and kisses, Venus, sudden reversals of Fortune, sleep and wakefulness, the exclusus amator, and the fickleness of women—which, in a poem of mixed hexameters and pentameters, 372 suggests Propertius and/or Ovid. Here, for example, is part of Propertius 2.15: “She opened my eyes fallen shut with sleep and said with her mouth, ‘Are you going to lie there like that, lazybones? In what varied embraces we mixed our arms! How often our kisses lingered on your lips! It’s not nice to ruin Venus in blind movement: if you don’t know, eyes are the leaders in love.” illa meos somno lapsos patefecit ocellos ore suo et dixit: ‘Sicine, lente, iaces?’ quam vario amplexu mutamus bracchia! quantum oscula sunt labris nostra morata tuis! non iuvat in caeco Venerem corrumpere motu: si nescis, oculi sunt in amore duces. (Prop. 2.15.7-12) This passage weaves together several of the themes and images listed above. One might also compare Propertius 1.3, where Propertius first uses o utinam (39), and where embraces (15), kisses 368 Luc. Bell. 2.306, 4.509, 8.86, 8.306; Val. Flacc. 1.111, 3.615, 7.133, 7.534, 8.439; Stat. Silv. 5.1.176 369 CIL IV 5296, VI 23472, 26489, 36658 (if Buecheler’s restitution of verse 6 is correct), XII 825, an inscription published in L’Année Épigraphique (2008) 1099, and an inscription published in Ben Abdallah et al. (2005) 103 and L’Année Épigraphique (2005) 1671, on which see below. 370 Turp. 125; Tib. 1.3.2; Hor. Carm. 1.35.38, 4.5.37; Calp. Sic. 0.79, 0.86; Stat. Theb. 5.472, 7.358 371 See Section C for discussion of o utinam at the start of a hexameter as a formula. 372 On the meter of this inscription, see Milnor (2014) 206-13 and below. 121 (16), lovers’ faithlessness (35-8), sudden reversals (31ff), sleep (passim), and lovers’ nightime vigils (39-44) all appear. While CIL IV 5296 uses themes and images that are common in elite elegy as well as a likely Propertianism, it does something different with them than Propertius does. Scholars have debated the gender of the speaker in CIL IV 5296, the number speakers the poem presents, and whether the poem is even a unified whole. 373 Of particular concern has been the fact that a speaker identified as feminine by the adjective perdita in verse 5 addresses a beloved who also seems to be feminine from the word pupula in verse 3 and the criticism of men in verse 4. Thus, “in the past, scholars have wanted to attribute the change of tone after line 3 to a change of speaker in the poem, and argument which (to their minds) has the merit of resolving the apparent lesbianism of the poem.” 374 More recent scholarship, on the other hand, is reversing this tendency. Varone takes the genders of perdita and pupula at face-value and without comment reads the poem as a unified composition with a single speaker. 375 He also adduces CIL IV 8321a as evidence that love between women was not alien to Pompeian graffiti. 376 Milnor carefully exposes the weaknesses of arguments that deny that CIL IV 5296 presents love between women but takes a more cautious stance than Varone: “by seriously engaging with the possibility that CIL 4. 5296 represents an erotic communication between women, we can actually understand the logic of the poem in different terms, and hear in it a voice not elsewhere represented in Latin poetry.” 377 373 See Milnor (2014) 207-12 for summary and discussion. 374 Milnor (2014) 210 375 Varone (2002) 100-1 376 Varone (2002) 102: “Chloe greets Eutychia: Eutychia, you do not care about me. With your hope steadfast you love Ruf(us?)” (Chloe Eutychiae s(alutem): | Non me curas, Euty | chia. Spe firma | tua Ruf(um?) amas, CIL IV 8321a, my translation) 377 Milnor (2014) 217-18. Milnor also addresses here the question of the actual sex and gender of the person who wrote CIL IV 5296 and notes that “as often as male poets may adopt a female persona in which to speak, they rarely do so in order to articulate female homoerotic desire, and when they do, they tend to represent women who love women as freakishly masculine.” 122 Reading CIL IV 5296 as a poem about love between women yields meanings that stray away from Propertius’ poetics. Propertius uses the phrase o utinam, for example, in five passages about fickleness and faithlessness. In Poem 1.3, Cynthia has awakened and complains that she has been anxiously awake all night waiting for the speaker to come to her, while he has been out partying. She wishes, “oh, if only you might pass such nights, you jerk, as these that you are always forcing poor me to have” (o utinam talis perducas, improbe, noctes, / me miseram qualis semper habere iubes, 39-40). In Poem 1.8, Cynthia is about to sail to Illyria, and though the speaker calls her “vow-breaker” (periura, 17), he wishes, “oh, if only the time of wintry storms might be doubled and the sailor be stuck not moving with a late rising of the Pleiades” (o utinam hibernae duplicentur tempora brumae / et sit iners tardis navita Vergiliis, 9-10). In Poem 1.16, a door recites a typical complaint of a locked-out lover who protests his devotion to his mistress despite her fickleness: “oh, if only my little voice, carried through a hollow crack, might make a turn into my mistress’s little ears and strike them… now she lies snuggled by another man’s happy arm” (o utinam traiecta cava mea vocula rima / percussas dominae vertat in auriculas… nunc iacet alterius felici nixa lacerto, 27-8, 33). In Poem 4.4, Tarpeia is preparing to betray Rome out of her infatuation with Tatius, and she wishes, “oh, if only I might sit captive before your household gods, and while captive catch sight of my Tatius’ face!” (o utinam ad vestros sedeam captiva Penatis, / dum captiva mei conspicer ora Tati!, 33-4 378 ), and, “oh, if only I knew the spells of the magical Muse! This tongue, too, would bring aid to a handsome man” (o utinam magicae nossem cantamina Musae! / haec quoque formoso lingua tulisset opem, 51-2). Poems 1.8, 1.16, and 4.4, men discuss or complain about women’s fickleness. In Poem 1.3, a woman complains about a 378 I follow Günther (1997) 106, Goold’s (1999) Loeb, and Hutchinson (2006) 125 in accepting Gronovius’ emendation of esse to ora in verse 34, because conspicer is deponent, and so esse makes little sense. The arguments for reading conspicer as deponent with passive meaning, while not impossible or unprecedented in Propertius, seem weaker to me in this case than the arguments for reading it with active meaning (i.e., regularly). 123 man’s faithlessness, but this is in service of a larger point in the poem: how Cynthia’s volatile temper ruins the speaker’s erotic fantasy so that she changes suddenly from the dream-girl he adores to the nagging harpy he is stuck in love with. As Milnor notes, 379 women’s fickleness is a topos of Latin love poetry. Propertius makes an aphorism of it in Poem 2.28: “Whatever they have sworn wind and wave snatch away” (quidquid iurarunt ventus et unda rapit, 8). 380 The speaker of CIL IV 5296, on the other hand, criticizes the faithlessness of men in attempt to persuade the girl she loves to return to her. Such valorizing of love between women at the expense of men and of different-sex erotic pairings is entirely foreign to Propertius. Spellings of certain words and metrical irregularities also distinguish CIL IV 5296 from Propertius’ classical dialect and his command of the elegiac couplet. A particularly telling word is labelis in verse 2 of the inscription. Propertius 2.15.10 accommodates a different but equivalent metrical position with the word labris for “lips.” In the inscription, labelis produces an unmetrical pentameter where labris would have avoided the problem. Furthermore, labelis is spelled labellis in Propertius’ classical, high-literary dialect at the end of the hexameter at Prop. 2.13.29, but labellis still would not fix the meter of the second line of the inscription. Other non-classical spellings are braciola instead of brachiola, cu instead of cum, vigilare instead of vigilarem, meditas instead of meditaris, supstulit instead of sustulit (or even substulit in Plautus’ archaic dialect), and coiunxit instead of coniunxit. 381 In addition to verse 2, most of the other verses are also metrically incorrect. Verse 3 is one syllable (such as et placed after nunc) shy of a correct hexameter; verse 4 is dactylic and has five feet but is not a dactylic pentameter; verse 6 could be a 379 Milnor (2014) 215 380 cf. also Cat. Carm. 70.3-4: “What a woman says to an eager lover should be written on wind and swift water” (mulier cupido duo dicit amanti / in vento et rapida scribere oportet aqua). 381 Whether or not the nasal consonants at the ends of words (or prefixes) in cum, vigilarem, and coniunxit were pronounced in the classical dialect, inscriptions such as the Res Gestae Divi Augusti (CIL III 774) and the epitaph for Margarita (CIL VI 29896, below) suggest that they were written, and meditor, as noted above, is deponent in the classical dialect. 124 hexameter with three extra syllables (-tuna quos) or, as Milnor analyzes it, “a half-pentameter, which falls apart after meditas… [but] ends with the dactyl and spondee combination (supstulit alte) which would mark a hexameter” 382 ; and verse 7 should be a hexameter but is not and instead would be a correct pentameter without subito. Verses 1, 5, and 8 are correct hexameters, though 8 should be a pentameter if the meter is elegiac couplets as the incorrect pentameters in 2, 4, and 6 suggest. 383 Thus, while the use of o utinam and the themes and images of the inscription create a partially Propertian effect, the non-classical spellings and metrical mistakes differentiate the inscription from elite literary culture. One way to interpret this difference is that the inscription’s author has tried to emulate Propertius’ meter and language while adapting his ideas but lacked the education and perhaps the talent to do the former successfully and so instead has produced a slightly déclassé version. On the other hand, given that the inscription has three successful hexameters, the fact that some of the metrical errors are, to borrow a term from tennis, unforced—labelis instead of labris in line 2, asyndeton in verse 3 where the addition of et would have maintained the sense and corrected the meter, and especially the unnecessary subito in verse 7—might suggest that the incompetence itself is artfully composed. Such artful incompetence does not appear anywhere in Propertius’ Elegies; it would be inconsistent with their Callimacheanism. I think, though, that an interpretation somewhere between these two is best. CIL IV 5296 is in many ways a sophisticated piece of poetry. As Milnor notes, “the language of the first lines actually enacts the embraces it describes.” 384 The choice to make collo ablative rather than the accusative object of amplexa produces “an appropriately convoluted image” and grammatical 382 Milnor (2014) 207 383 Milnor (2014) 206-7 gives a similar analysis of the metrical irregularities in this poem. 384 Milnor (2014) 214 125 structure: The speaker wishes to have the beloved’s arms, which have embraced her, held by her (i.e., the speaker’s) neck. 385 Furthermore, nouns embrace verbs in a chiasmus in the phrase collo complexa tenere braciola, while at the same time the words for the speaker’s body and her action (collo and tenere) are “intertwined” 386 in a synchisis with the words for the beloved’s action and body (complexa and braciola). Similarly, in verse two the accusative nouns braciola and oscula interlock with the dative teneris…labelis. The overall effect is of crisscrosses and movement like bodies embracing and kissing. The poem also shows originality and artistry in transforming the misogynistic topos of women’s fickleness into a gnomic statement on men’s fickleness and using this as part of a lover’s persuasion in a relationship between women. The words and meter of CIL IV 5296 suggest some schooling but not the extreme refinement of Propertius. The dialect as reflected in the spellings, however, seems to have been, to some degree, a choice. While the graffitist wrote cu instead of cum, they also wrote mecum instead of mecu, and while vigilare in the graffito lacks an m at the end, utinam, virorum, and amantum have it. Furthermore, as noted above, some (but not all) of the metrical errors are so easily fixed that they could be taken as deliberate. The inscription’s physical setting also provides clues about who wrote it. It was found in the entryway of a modest but not poor house in Pompeii (9.9.6): The house has “two tiny rooms, a balcony, and a small open court.” In this space there were “a tiny but workable hearth, a cistern for storing water, a small built-in cabinet, and a number of cooking pots.” The entryway “sports a simple but attractive marble-chip mosaic; the length of its north wall was, at one time, decorated with painted plaster;” and “a small decorative statue in bronze of Zeus brandishing a thunderbolt” 385 Milnor (2014) 214 386 Milnor (2014) 214 126 was found in the house. 387 Milnor compares the speaker of CIL IV 5296 to the speaker of Sulpicia’s elegies because of their shared “interest in engaging with and subverting the cultural prejudices which would restrict a women’s right to speak.” 388 Elite Roman men constructed their literary culture, in part, to limit women’s participation, 389 and in this graffito insight, learning, and artistry appear to struggle against the intersecting confines of womanhood and non-elite social class in Roman culture. Though the speaker of Propertius’ elegies presents himself as limited by his less-than-alpha masculinity, he is still an elite Roman man and a doctus poeta with almost 390 “all the perquisites” 391 of speech associated with this status, and self-consciously so. Erotic graffiti are not the only type of inscription to recycle words and ideas from Propertius’ elegies: We find similar practices of reuse and adaptation in funerary inscriptions from many parts of the Roman world. CIL VI 6592 is one of the more than 400 inscriptions 392 discovered in the Columbarium of the Statilii, a tomb on the Esquiline found along with two others along the north side of the Via Praenestina-Labicana and just inside the Porta Praenestina (now the Porta Maggiore). 393 The Columbarium of the Statilii was for the Statilii and their slaves and freedmen and may originally have been part of a large garden complex. 394 This columbarium “originally… provided equal semicircular niches to all its occupants,” but “an extension was added 387 Milnor (2014) 225-6 388 Milnor (2014) 217 389 Habinek (1998) 122-36 390 For discussion of the limitations under which the Propertian speaker perceives himself to labor, see Chapter 5. 391 Habinek (1998) 127 392 CIL VI 6213-6640 393 Holliday (2005) 90-1 394 Holliday (2005) 93: “The columbaria originally stood in the horti Tauriani, which belonged to the influential Statilii family, outside the original limits of the Gardens of Maecenas (although part of the horti Tauriani was incorporated into them in A.D. 35). The horti Tauriani came under imperial control in A.D. 53 when Agrippa, who coveted them, accused their owner, Titus Statilius Taurus (cos. A.D. 44), of extortion and practicing magic; he committed suicide (Tac. Ann. 12.59). The gardens were subsequently divided and given to two powerful imperial freedmen, Pallantes and Epafroditus, and thus became known as the horti Pallantiani (ed Epafroditiani). After their downfall, the horti were again confiscated and passed into imperial possession (see. Suet. Dom. 14).” Borbonus (2014) 97 and 240n54 argues that the inclusion of the columbarium in the gardens in uncertain and reviews scholarship on the problem. 127 that not only allowed personalization of individual burials but required it” 395 after Statilia Messalina married Nero in 66 CE 396 There was a painted frieze of scenes from the myths of Rome’s founding separated by “representations of pilasters.” 397 Holliday reconstructs the scenes as follows: On the west wall, the Trojans and Latins under Iulus fight the Rutuli, and Iulus builds Alba Longa; on the south wall, Aeneas battles the Rutuli, Aeneas and the Rutuli strike a truce, Aeneas marries Lavinia, and the walls of Lavinium are built; Rhea Silvia is with the other Vestal Virgins, Mars seduces her, and her pregnancy is discovered; on part of the north wall, the infant Romulus and Remus are left out to die, the she-wolf suckles them, and they spend their youth as shepherds – the rest of the frieze on the north wall is unknown. 398 The frieze seems to be a “condensed” or “excerpt[ed]” version of the frieze of the Basilica Aemilia. 399 One epitaph found in the main, original section of the columbarium (n in Brizio’s drawing) is CIL VI 5296. 400 The inscription reads, SI·QUA·MANENT·OBITIS·UI[… PRAEMIA·SUB·TERRIS·[… IÚDICAFID·CONIÚNX·ÉREPTÁ·QUÁ·SIBI MAERENS DEUOUET · INUISI · NOXSIA · REGNA · DEI (CIL VI 6592) 401 395 Borbonus (2014) 93. Borbonus (2014) 97 and 240n55 explains the difficulties with securely identifying the extension as actually belonging to the Columbarium of the Statilii. 396 Borbonus (2014) 95 397 Holliday (2005) 94 398 Holliday (2005) 94 399 Holliday (2005) 108 400 Borbonus (2014) 196 401 See Appendix B Figures 1 and 2 for images. This is the text given in the CIL. As I look at the photos of the inscription in Epigraphic Database Roma, I do not see the interpunct that the CIL records between SI and QUA in the first line, and I am uncertain of the interpunct between IUDICAFID and CONIUNX, whereas the other interpuncts are all clearly visible. The U’s here are written as V’s in the CIL, as they are in the inscription, but I cannot find a way to type a V with an acute accent. 128 The inscription is now in 3 fragments, one with the first couplet and the other two with the second. 402 As epigraphers have restored the second halves of the first two verses and emended the Latin, the poem reads, “If any rewards of a life led without a misdeed remain for the dead under the earth, Amaryllis has these. Her husband judges it so and, grieving for himself that she was snatched away, execrates the noxious realms of the hateful god.” Si qua manent obitis vi[tae sine fraude peractae] praemia sub terris [haec Amaryllis habet] iudica(t) i(d) coniunx, erepta qua sibi maerens devovet invisi noxia regna dei. 403 While the text of the first two lines is a hypothesis (however well informed), and there is some debate over emendations to the third line, the text of the fourth line is clear and, as far as I know, undisputed. This fourth line is where the epitaph echoes Propertius. In a poem complaining about greedy girlfriends, Propertius’ speaker recalls Brennus, a Gallic chieftain who invaded Greece in 279 BCE and was defeated in 278 BCE when he attempted to sack and pillage the Sanctuary of Apollo at Delphi: 404 “Charred doorways testify that Brennus was sacrilegious when he attacked the Pythian realm of the unshorn god.” Torrida sacrilegum testantur limina Brennum, dum petit intonsi Pythia regna dei. (Prop. 3.13.51-2) Verse 4 of CIL VI 6592 appears to be closely patterned after verse 52 of Propertius 3.13: 405 402 Caldelli and Ricci (1999) 88; EAGLE: Electronic Archive of Greek and Latin Epigraphy 403 CIL VI 6592 emends IUDICAFID to iudicavit; CLE 1031 emends it to iudicat id. 404 Richardson (1977) 303; Spickermann (2006) 405 There is another possible model for CIL VI 6592.4: Tib. [Lygd.] 3.5.22 reads duraque sortiti tertia regna dei. The dates of the Lygdamus poems are difficult to establish, though. Büchner (1965) and Conte (1994a) believe they are contemporary with Ovid, but others such as Axelson (1960) Tränkle (1990) favor the Flavian period. If the latter are correct, then the Lygdamus poem cannot have influenced CIL VI 6592, and there is even a slight possibility that the epitaph influenced the Lygdamus poem. The fact that both the epitaph and the Lygdamus poem change the god from Apollo to Pluto is perhaps suggestive. The only other canonical parallel is very weak: vixque bene audito nomine regna dedi (Ov. Her. 7.90) 129 devovet invisi noxsia regna dei. (CIL VI 6592.4) The most obvious similarity is the ending of both verses, regna dei. Preceding this ending in each verse is a dactyllic adjective modifying regna (Pythia and noxsia). Preceding these adjectives, in turn, are trisyllabic adjectives that begin with in-, 406 have all long syllables, and modify dei. The inscription thus reproduces the chiasmus from Propertius, so that in both verses the gods and their adjectives ‘hold’ the realms and their adjectives. Finally, each verse begins with with a dactyl that starts with d and includes or is the 3 rd -person singular, present, active, indicative verb of which regna is the direct object. The setting of CIL VI 6592 in the Columbarium of the Statilii also suggests commonalities between its speaker and the speaker of Propertius’ Elegies. 407 Holliday argues that the slaves and especially the freedmen entombed in the Columbarium of the Statilii were outsiders looking for new and better places in the new order of Rome and through ideas of Romanness ushered in under Augustus. By depicting myths of Rome’s founding, Holliday says, the frieze in the Columbarium of the Statilii functions as the buried slaves’ and freedmens’ “plea for inclusion in the energizing vision of a unified Rome.” 408 The speaker of Propertius’ elegies also feels insecure about his place in the new order. He may be an equestrian, but he is also Umbrian, as he points out prominently in the last poem of Book 1 and the first poem of Book 4. He is anxious, as well, about his social class and his gender. He perceives that he does not enjoy certain prerogatives of public speech and bodily inviolability that equestrian viri were traditionally supposed to have (Chapter 5 explores how he tries to negotiate what it means to be an equestrian vir in light of these deprivations). It is 406 It is not the same in- in both adjectives—the in- that begins intonsi means “not,” while the in- that begins invisi is the preposition of motion, here meaning “against”—but the sound and spelling matter more here than the meanings of the prefixes. 407 See Appendix B Figure 3 for a plan of the Coumbarium of the Statilii. 408 Holliday (2005) 124 130 not an accident that one of his primary ways to describe his condition is with the metaphor of “slavery to love” (servitium amoris). Furthermore, just as the freedmen depended for their social position on their influential patrons, the gens Statilia, the Propertian speaker depended for his social station and his income on his patron, Maecenas. While these relationships with patrons wove the freedmen and the Propertian speaker into Roman society, a patron’s favor was not guaranteed to continue. The freedmen of the Statilii ‘speak’ with words and images that seek to define them as Romans, and the speaker of Propertius’ elegies speaks as a relatively newly “made” 409 Roman seeking to define his identity in his society. But while verse 52 of Propertius 3.13 and verse 4 of CIL VI 6592 share much in form and context, they differ in meaning. The god in Propertius’ verse is Apollo, who is described flatteringly as “unshorn” (intonsi), that is, eternally youthful and beautiful. In CIL VI 6592, the god is Pluto, who is “hateful” (invisi). The poems also describe the realms of the gods in opposite ways. In Propertius 3.13, Apollo’s realm deserves reverence, which is why, in the following couplet—“but soon Parnassus, struck on its laurel-bearing peak, cast terrible snows onto Gallic arms” (at mox laurigero concussus vertice diras / Gallica Parnasus sparsit in arma nives, Prop. 3.13.53-4)—Apollo and Jupiter or Minerva, indicated by the laurels (laurigero) and the lightning bolt (concussus) respectively, vindicate Delphi and Apollo’s divinity by casting ruin upon the invading and sacrilegious Gauls. Furthermore, with the epithet “Pythian” (Pythia), the speaker evokes Apollo’s slaying of the monstrous serpent, a heroic victory in which Apollo earned Delphi as his realm. In contrast, Pluto’s realm in CIL VI 6592 is “noxious” (noxsia), and Amaryllis’ husband “execrates” (devovet) it for “snatch[ing] away” (eripuit) his wife. Both poems are angry complaints, but the Propertian speaker is angry with girlfriends and laments his empty wallet, 409 Feldherr (1998) 113-14 (“Romans are made, not born”) qtd. in Holliday (2005) 123 131 whereas Amaryllis’ husband mourns his dead wife and is angry with Pluto. Even as they use similar words and emerge from comparable social pressures, the poems differ in their emotional registers. While the setting of CIL VI 6592 in the Columbarium of the Statilii suggests similarities between the speakers of the epitaph and Propertius 3.13, it also complicates those similarities and creates differences alongside them. The simple fact of having a fixed geographical location and relatively fixed surroundings, as opposed to a poem that might be read or recited in a wide variety of locations, invests the text of CIL VI 6592 with a closer relationship to its setting than Propertius 3.13 has to its various possible settings. A columbarium, argues Borbonus, fostered a sense of collective identity through architecture, ritual, written texts, and social participation 410 in the face of a society that “placed freedmen in contradictory positions between success and marginalization.” 411 The Propertian speaker, in contrast, proclaims his individuality and alienation at Prop. 3.13.59-66, and elsewhere he distinguishes himself from other poets 412 and declares himself unable to participate in the social obligations of his peer group, such as attending his friends’ sickbeds. 413 CIL VI 6592 thus coopts part of a poem that constructs an individualistic, alienated identity of a free-born equestrian who refuses to follow social conventions and who has himself coopted the vocabulary of enslavement to describe his status. It uses those words to participate in building a collective identity, probably of freedmen, that contrasts itself with the social and political identities of traditional Roman elites. 410 Borbonus (2014) passim 411 Borbonus (2014) 159 412 Prop. 1.7 and 2.34.59-94, e.g. 413 “I will not be able to bring comfort to you when you ask, since there is no medicine for my own sickness” (non ego tum potero solacia ferre roganti, / cum mihi nulla mei sit medicina mali, Prop. 1.5.27-8). Varhelyi (2010) 78-90 examines the importance of attendance at peers’ sickbeds for communal identity building among elite men in the early Empire. See further discussion in the Conclusion. 132 On the other hand, CIL VI 6592 may assert more individuality than many epitaphs in columbaria do, in part through its reuse of Prop. 3.13.52. Typically, “columbarium inscriptions are cursory and formulaic and articulate only a limited amount of information about the person they commemorate.” 414 Verses 1-3 of CIL VI 6592 contain several formulaic phrases and ideas, such as si qua at the start of the first hexameter, 415 and they do not communicate many particulars about Amaryllis, the deceased. Verse 4 also says little about Amaryllis, but it says rather more about her husband. Whether or not her husband composed the epitaph, it is in his voice, so he is the one we read reusing Prop. 3.13.52. This verse is buried in the middle of a poem in the middle of a poetry book – it is not what one would find at the beginning of a papyrus roll. Verse 4 of CIL VI 6592, then, communicates that Amaryllis’s husband had more than a passing familiarity with Propertius’ Elegies 416 to anyone else who is in the know. To readers less familiar with Propertius, verse 4 may not have stood out. Propertius’ transmuted text becomes a vehicle for Amaryllis’ husband to create two circles of collegiality with the other users of the columbarium, a well-read circle and a less well-read one. CIL VI 29896, the epitaph for Pearl (Margarita) the dog, also plays complex literary and social games. Pearl’s epitaph was found in front of the Porta Pinciana in Rome. 417 The marble is inscribed with small, fairly neat capital letters, with the text aligned to the left margin and the pentameters indented. 418 The name Margarita is written on the left underneath the epitaph and a palm leaf is drawn on the right opposite the name. 419 There is a frame of round, convex molding. 414 Borbonus (2014) 106. Borbonus (2014) 116-34 discusses the most common types of words used in epitaphs in columbaria. 415 See Section C on formulas in epigraphic and canonical poetry. 416 It is possible that whoever composed the epitaph used a manual of formulas and quotable verses appropriate to particular types of inscriptions, on which see Cagnat (1889), Susini (1973) 47-8, and Cooley (2012) 291-2. The effect on the reader nevertheless remains the same, that the speaker is well-versed in Propertius. 417 CIL VI 29896 418 See Appendix B Figure 7 for a photo of Pearl’s epitaph. 419 Buecheler (1895) 1175 133 It is dated to the 2 nd century CE. 420 The text of the inscription quotes Propertius and several other famous poets: Gallia gave birth to me, the name of the ocean’s wealth a shell gave me; the honor of the name befits my beauty. Learned in running bravely through dim forests and in flushing out furry beasts in the hills, I was not accustomed ever to being held in heavy chains, 5 nor to suffering savage beatings on my snowy body. For I would lie on the soft lap of my master and mistress, and I knew how to sleep on a made bed when I was tired, and I spoke more than was given to a dog’s unspeaking mouth: nobody was very afraid of my barks. 10 But now I have died, cast down by giving ill-favored birth, and now the earth covers me under a small marble. Gallia me genuit, nomen mihi divitis undae concha dedit, formae nominis aptus honos. docta per incertas audax discurrere silvas collibus hirsutas atque agitare feras, non gravibus vinclis unquam consueta teneri 5 verbera nec niveo corpore saeva pati. molli namque sinu domini dominaeque iacebam et noram in strato lassa cubare toro et plus quam licuit muto canis ore loquebar: nulli latratus pertimuere meos. 10 sed iam fata subii partu iactata sinistro, quam nunc sub parvo marmore terra tegit. In verse 4, Pearl uses the words “furry beasts” (hirsutas... feras) in the same metrical positions where they occur in verse 12 of Propertius 1.1 (ibat et hirsutas saepe videre feras, Prop. 1.1.12), in the Milanion exemplum, and the subjects in both verses are envisioned moving through the wilderness (forests in the epitaph, caves in Propertius). Likewise, verse 8 of Pearl’s epitaph, “For I would lie on the soft lap of my master and mistress” (molli namque sinu domini dominaeque iacebam, 8), recalls Propertius 2.13, “let me have the pleasure of reading in the lap of a docta puella,” (me iuvet in gremio doctae legisse puellae, Prop. 2.13.11). Pearl thus appropriates elements of the Propertian speaker’s performances of elegiac masculinity for her own performance of feminine and mourned-for caninity. 420 Cecere (1994) 417-8 134 But these are not Pearl’s only allusions. Her opening words, “Gallia gave birth to me” (Gallia me genuit, 1) recycle the beginning of the famous epitaph for Vergil, “Mantua gave birth to me” (Mantua me genuit). In verse 6 of the epitaph, the words “snowy body” (niveo corpore, 6) occur in the same metrical position as in verse 30 of Tibullus 3.4, one of the Lygdamus poems. This is also where these words occur in verse 42 of Ovid’s Amores 3.2, which, like Propertius 1.1 and Pearl’s epitaph, uses Atalanta and Milanion as an erotic exemplum (Ov. Am. 3.2.29-30). 421 The allusions to Vergil and Propertius compare Pearl with the speakers, the personae of Vergil and Propertius presented in these poems, who are docti poetae; the allusion to Tibullus/Lygdamus and Ovid compares her with erotic objects in these poems, a beautiful boy and a puella. In addition to Propertius 2.13.11 (as noted above), verse 8 of Pearl’s epitaph, “For I would lie on the soft lap of my master and mistress” (molli namque sinu domini dominaeque iacebam, 8), recalls Catullus’ sparrow poems (Cat. Carm. 2, 3). The fact that Pearl’s epitaph is an epigram about a dog and Pearl’s discussion of her sleeping habits (cubare, 8) and her speech (loquebar, 9) recall Martial’s epigram on the dog Issa (Mart. Epig. 1.109), which uses the same verbs to dicuss Issa’s sleeping habits (cubat, 8) and speech (loqui, 6), and which itself refers directly to the sparrow poems (“Issa is naughtier than Catullus’ sparrow,” Issa est passere nequior Catulli, 1). By alluding to the sparrow poems and the epigram on Issa, Pearl compares herself to objects of affection spoken of in eroticizing terms, and by alluding to the Propertius 2.13, she compares herself to an elegiac lover, a doctus poeta who is performing his doctitude for a docta puella who in the following verse, by her discerning evaluation of the Propertian speaker’s poetry, demonstrates her own doctitude. 421 See also nitido corpore (Tib. 3.4.36) in the same metrical position, but following spondee + long rather than dactyl + long. 135 “Doctitude” is the quality of being doctus. Habinek parses out the elements of doctitude as “learning, discernment, and [public or semi-public] performance” 422 for men; he argues that women’s doctitude also involved learning and discernment but not usually performance, at least in the highest echelons of society, since the men who were literary producers in these milieux often stifled or coopted literary output by the women in their lives. 423 As several scholars have documented, doctitude was a quality valued in speakers, listeners, readers, and other users of Latin literature from at least the “Good Friend” in Ennius through Aulus Gellius. 424 Pearl shows off her doctitude with clever poetic composition in addition to her literary allusions. She uses prosody expressively, for example. The first five hexameters, in which she describes her life, scan normally. In verse 11, the last hexameter, in which she describes her death, Pearl uses synizesis to combine the two i’s in subii and so resolve the problem that the first i in subii is short and the a in partu is long. She also presents the reader with a riddle in the first couplet: “the name of the ocean’s wealth a shell gave me; the honor of the name befits my beauty” (nomen mihi divitis undae / concha dedit, formae nominis aptus honos, 1-2). Pearl tells us that she was learned in doggy skills, as well. She says that she was docta in “running bravely through dim forests and in flushing out furry beasts in the hills” (per incertas audax discurrere silvas / collibus hirsutas atque agitare feras, 3-4); that she “...would lie on the soft lap of my master and mistress” (molli namque sinu domini dominaeque iacebam, 7); and that she “knew how to sleep on a made bed when [she] was tired” (et noram in strato lassa cubare toro, 8). In these phrases, Pearl describes her canine knowledge, and the words docta and noram 425 422 Habinek (1998) 131 423 Habinek (1998) 130ff. Since Pearl was a female dog, her admission in verse 9 that she spoke excessively may play on this social dynamic. 424 inter alios, Habinek (1998) 122-36; James (2003) passim; and Rust (2009) vi, 100, 172 425 Lewis and Short ad loc cit. 136 suggest that it is knowledge gained though learning rather than innate knowledge. She also confesses to a little misbehavior behavior—“and I spoke more than it was given to a dog’s unspeaking mouth to speak” (et plus quam licuit muto canis ore loquebar, 9). On the one hand, this confession suggests a lack of decorum, an important element of doctitude. 426 On the other hand, the confession is a literary pose. It invites the reader to indulge and even to be charmed by Pearl’s naughtiness – after all, she says, “nobody was very afraid of my barks” (nulli latratus pertimuere meos, 10). Pearl, then, is telling us about her dogtitude: She knew how to behave well as a dog, and she also knew and still knows how to charm people with a little misbehavior. But to return to the repurposing of Propertius’ poetry: Not only does Pearl pull the Propertian speaker’s words and ideas from elegiac masculinity into feminine caninity, but she also plays with the idea of doctitude, which is a key element of Propertius’ Callimacheanism. Pearl is a humorous, miniaturized docta puella with her dogtitude and her allusion to Martial’s epigram on Issa. At the same time, she presents herself as a proper docta puella with her allusions to Tibullus 3.4, and Ovid’s Amores 3.2. She is also a docta poeta with her allusions to Propertius 1.1 and 2.13 and Vergil’s epitaph and with her skillful composition and sophisticated combinations of allusions. When Pearl says that she is “learned in running bravely through dim forests” (docta per incertas audax discurrere silvas, CIL VI 29896), the word “forests” (silvas) refers both to these combinations of allusions, as a metapoetic buzzword, 427 and to actual woods that a dog might run through. Doctitude and dogtitude thus converge in Pearl’s voice and especially in the word docta, which simultaneously sets up and undercuts the miniaturization of doctitude. Pearl is both a humorist and a lost object of affection, and doctitude is both parodied and cherished. A reader of Pearl’s epitaph finds Propertius’ words and ideas mischievously reimagined. 426 Habinek (1998) 124-32 427 See Chapter 2. 137 While Pearl’s persona is engaging, this epitaph and tombstone were, of course, created by a human or humans. On the one hand, Pearl is the speaker of her own epitaph; on the other hand, she is dead so she cannot speak, and she is a dog so she cannot speak. The gravestone itself, with its border and palm leaf and Margarita’s name, physically frames her speech just as, for example, the speaker in Propertius 4.5 frames the dead Acanthis’ speech, the speaker in Propertius 4.7 frames the dead Cynthia’s speech, and, in Propertius 4.11, the poetry book frames the dead Cornelia’s speech. 428 Like Acanthis, Cynthia, and Cornelia, Pearl is ventriloquized. Her speech sallies forth into the game of allusions and variations in Latin poetry that scholars such as Conte and Hinds have elaborated 429 and into the broader games of learned performance and literary production that Habinek, Rust, Gurd, and others have described. 430 Whoever put a composition so laced with doctitude into the voice of Pearl the dog was an especially skilled player. Propertius imagined multiple scenarios in which other people would reuse and change his poetry, and his readers and listeners exceeded what he anticipated in the Elegies. It is not surprising that members of Rome’s intellectual elite, such as Ovid and Remmius Palaemon, appropriated words and ideas from the Elegies as they continued the production of Rome’s elite culture. Perhaps less expected is the evidence that people from less privileged social strata also reused Propertius’ text as they participated in Rome’s literary culture. While some wealth was involved in producing the epitaph for Pearl the dog (CIL VI 29896), this does not necessarily signal that Pearl’s owners were equestrians like Ovid. The social classes of graffitists are notoriously difficult to establish – to assume, for example, that an equestrian or even a senator would not have visited the Caupona of Euxinus in Pompeii would not be sound. On the other hand, the love poem 428 cf. Habinek (1998) 130 429 Conte (1986) passim; Barchiesi (1993) passim; Hinds (1998) 123-34 and passim 430 Habinek (1998) passim, (2005) passim, (2009) 121-24; Johnson (2009) 320-30; Rust (2009) passim; Gurd (2011) 77-104 138 from one woman to another (CIL IV 5296) was found inside a modest house, “somewhere between working class and middle class.” 431 Because the graffito was written inside a private dwelling, it seems safer in this case to suppose that the graffitist lived there or at least associated closely with those who did. Because the epitaph for Amaryllis (CIL VI 6592) is from the Columbarium of the Statilii, we can say fairly securely that Amaryllis and her husband were social and perhaps legal dependents of the Statilii and probably freedmen. The Propertian speaker all but invites other people to change and reuse his text when he presents it as vulnerable, and in Poem 2.1 he actually issues the invitation to Maecenas and implicitly to his readers and listeners. People from various walks of life took him up on it. Some people, however, may not have been aware that they were reusing Propertius’ words. Section C will explore how Rome’s oral literary culture interacted with its written literary culture in ways that could obscure the link between a phrase and its original author. Section C: Orality and Authorship “The book you’re reciting, Fidentinus, is mine, but when you recite it badly, it begins to be yours!” quem recitas meus est, o Fidentine, libellus; sed male cum recitas, incipit esse tuus! (Mart. Epig. 1.38) Oral composition is familiar from Homeric poetry, but in the last few decades scholars have described a distinctively Roman oral literary culture that existed alongside and in a mutually formative relationship with the Romans’ written literature. Oral composition was part of the creative processes of even some of Rome’s most Alexandrianizing poets as well as of people 431 Milnor (2014) 226 139 producing poetry in less elite literary settings. My concern here is with the effects of this orality on authorship: While the written word is changeable, the orally transmitted word is even more so, and the use of formulas characteristic of oral composition could disconnect a phrase from its original author. These and other aspects of orality in Roman literature challenge notions of a poet’s authority over his or her texts, such as those we saw Horace clinging to in Chapter 2. I begin this section with a picture, by necessity in rather broad strokes, of orality in Rome at the end of the Republic and the beginning of the Empire. I review evidence for and scholarship about oral performance and oral composition among Rome’s leading poets, including Propertius. Moving from elite literature to inscriptions, I explore epigraphic evidence of oral composition by less famous poets. This evidence indicates that some of Roman oral poetry culture was shared by poets ranging from whoever scratched graffiti into the walls of Pompeii all the way up to superstars such as Vergil, Propertius, and Ovid and that it reached from Rome and Italy to Gaul, Africa, Spain, and other parts of the empire. I also consider the implications for authorship when a famous poet’s words could be taken from their original context and eventually grow so far removed from it as to become a formula. Oral performance of special words—made special by planned composition, meter or rhythm, poetic devices, formulas, or other means of ritualization 432 —happened in many quarters of Roman life. Romans memorized and chanted literature in school 433 and gave recitals of poetry and prose. They attended theatrical productions. They read statutes aloud to publicize them. They legislated against certain kinds of songs. 434 Horsfall describes the participation of the Roman plebeians in oral literary culture, from “popular song and chant” 435 to poets seeking audiences in 432 Conte (1986) 40-56; Habinek (2005) 1-4; and cf. Nagy (1990a) 31 433 Quint. I.O. 11.1.56, 11.2.33, and 11.3.57. Gamel (1998) 81; Markus (2000) 142; Horsfall (2003) 11-2 434 occentatio, for example, was punishable by death (Lex XII tab. 8.1 b; Cic. De Rep. 4.11-2) 435 Horsfall (2003) 36 140 the streets and baths and even latrines 436 to the crowds’ adoption of songs heard at the theater 437 to Toronius Gallus’ musical group, which often played at Augustus’ dinners. 438 These last two examples indicate that the boundaries between elite and non-elite literary culture were porous. Another of Horsfall’s examples captures the interdependence of written and oral literary cultures among the Roman plebs: “Caesar had complained of the versiculis de Mamurra perpetua stigmata imposita, the indelible brands laid upon him by the epigrams upon Mamurra (Suet. Caes. 73.1): scribbled up… shouted out over a drink in the schola (club-house) of a collegium, chanted in the street, even sung; we are left with a positive excess of possible methods of embarrassing diffusion…” 439 Some of the Pompeiian graffiti and other inscriptions discussed below also suggest influence moving in both directions between written and oral and between elite and non-elite literature. Elite Roman literature reports multiple types of oral performance. Some performances were fully improvised. Cicero mentions two poets who composed orally, Antipater of Sidon, who was active a generation or so before Cicero, and Archias, Cicero’s contemporary: “…that Antipater of Sidon, whom you remember well, Catulus, used to pour forth hexameter verses and others in various modes and meters extemporaneously…” …Antipater ille Sidonius, quem tu probe, Catule, meministi, solitus est versus hexametros aliosque variis modis atque numeris fundere ex tempore… (Cic. de Orat. 3.50.194) “How often have I seen this Archias, judges… how often have I seen him, though he had written not one letter, speak extemporaneously a great number of very fine verses on the very matters then under discussion! How often have I seen him, called back, declaim on the same subject with entirely changed words and expressions!” Quotiens ego hunc Archiam vidi, iudices...quotiens hunc vidi, cum litteram scripsisset nullam, magnum numerum optimorum versuum de eis ipsis rebus quae tum agerentur dicere ex tempore! Quotiens revocatum eandem rem dicere, commutatis verbis atque sententiis! (Cic. Arch. 18) Cicero uses the same phrase, “extemporaneously” (ex tempore), to describe how both Antipater and Archias composed, and he even specifies that Archias did not write anything down before 436 Horsfall (2003) 55 437 Horsall (2003) 13 438 Horsfall (2003) 34 439 Horsfall (2003) 42 141 performing. Antipater and Archias, in Cicero’s reports, composed entirely orally in performance, though Archias also used writing in his studies (Cic. Arch. 4) and wrote texts for others to read (Cic. Arch. 18). Both Antipater and Archias were Greek, and the surviving poetry attributed to them is in Greek not Latin. Even if their extemporaneous compositions were also in Greek, though, they were still instances of orally composed poetry performed at Rome. Furthermore, fluent command of Greek was de rigueur among the Roman elite, so Antipater and Archias would not have been linguistically exotic to them. Some, such as Cicero, even mixed Greek and Latin in personal communications. Cicero also suggests that the judges should think of Archias within the Roman poetic tradition: “And we have accepted it from the best and most learned men, that the pursuit of other things depends on learning and precepts and skill, but that a poet takes his strength from nature itself, and that he is awakened by the power if the mind, and that he is filled, as it were, by a certain divine spirit. Wherefore in his own right our Ennius calls poets sacred, because they seem to have been entrusted to us as if by some gift or good offices of the gods.” Atque sic a summis hominibus eruditissimisque accepimus, ceterarum rerum studia et doctrina et praeceptis et arte constare: poetam natura ipsa valere, et mentis viribus excitari, et quasi divino quodam spiritu inflari. Qua re suo iure noster ille Ennius sanctos appellat poetas, quod quasi deorum aliquo dono atque munere commendati nobis videantur. (Cic. Arch. 18) In the first sentence Cicero speaks universally about poets, setting Archias’ Greekness to the side. Even while he does this, though, he hints at Romanness – the phrase “we have accepted it from the best and most learned men” (sic a summis hominibus eruditissimisque accepimus) sounds similar to the customary Roman appeal to the authority of their ancestors. In the second sentence, Cicero brings Archias under the ideological protection of Ennius, whom he presents as an archetypal Roman poet (noster ille Ennius). Cicero is perhaps stretching ideas of Romanness to cover Archias here; this speech is, after all, a legal defense against a prosecution alleging false claims of citizenship. But Cicero is unlikely to have made this argument if it lacked all credibility, 142 and he is similarly unlikely to have defended Archias’ citizenship by emphasizing his skill at oral composition if such a practice had seemed entirely out of the ordinary at Rome. 440 Other sources report oral composition by leading Roman poets. Donatus, perhaps drawing his material from a now lost biography of Vergil by Suetonius, tells us, “They say that Eros, [Vergil’s] secretary and freedman, used to tell in his old age of how one time when reciting [Vergil] completed two half-verses extemporaneously. For where, up to that point, he had ‘Misenus son of Aeolus’ he added ‘than whom there was not another more outstanding,’ and likewise to this “at stirring up men with a bronze [trumpet],” similarly fired up, he added ‘and at setting Mars ablaze with the song’ and immediately instructed [Eros] to write them both down in his book.” Erotem librarium et libertum eius exactae iam senectutis tradunt referre solitum, quondam eum in recitando duos dimidiatos versus complesse ex tempore. Nam cum hactenus haberet: “Misenum Aeoliden,” adiecisse: “quo non praestantior alter,” item huic “Aere ciere viros,” simili calore iactatum subiunxisse: “Martemque accendere cantu,” statimque sibi imperasse ut utrumque volumini adscriberet. (Brugnoli and Stok, Vitae Vergilianae Antiquae, 32.8-33.5) According this anecdote, then, Vergil recited his poetry before it was entirely finished, and sometimes he extemporized. Improvised oral composition was part of the creative process of even so painstaking and polished a poet and even when the words would ultimately be written down. In fact, poetry was often dictated to secretaries, and this is oral composition 441 albeit distinguished from the oral composition that Parry and Lord describe by the poet’s ability to pause and look over what has been written down. Some of Rome’s most famous poets also speak of oral performances. Ovid says that several poets he admired when he was young performed their poetry aloud: “Often Macer, who was greater in age, read to me about his birds, and about which snakes harm you and which plants help you. Often Propertius used to recite his flames – he had been joined to me by right of fellowship. Ponticus with heroic verse and Bassus, famous for iambics, were dear members of my circle. And Horace the master of meters captivated our ears while he played refined songs on his Ausonian lyre.” saepe suas volucres legit mihi grandior aevo, quaeque nocet serpens, quae iuvat herba, Macer. saepe suos solitus recitare Propertius ignes, 440 See Ziolkowski (2005) 132-3 for further discussion of Antipater and Archias. 441 Gamel (1998) 81 citing Hor. Sat 1.10.92, Quint. I.O. 10.7.29, and Pliny Ep. 9.36 143 iure sodalicii qui mihi iunctus erat. Ponticus heroo, Bassus quoque clarus iambis dulcia convictus membra fuere mei. et tenuit nostras numerosus Horatius aures, dum ferit Ausonia carmina culta lyra. (Ov. Trist. 4.10.43-50) Ovid describes a few different types of orality in this reminiscence. Macer read aloud to him in a private setting, whereas Propertius gave recitals, and Horace sang or recited to music. If Horace was playing the lyre, he could not have held a text in his hands, and a scroll would have been unwieldy on a music stand. Horace was probably singing or reciting from memory, then, rather than reading. The setting of Horace’s performances is less clear here than those of Macer and Propertius’ performances. Horace himself says he does not like performing at a recitatio: “I’m ashamed to recite unworthy writings in crowded auditoriums 442 and add weight to the nonsense” (…spissis indigna theatris / scripta pudet recitare et nugis addere pondus, Hor. Epist. 1.19.40-1). He says he only recites for friends and only grudgingly: “I don’t recite for anyone except friends and I only do it when forced, and not just anywhere or in front of just anybody” (nec recito cuiquam nisi amicis idque coactus, / non ubivis coramve quibuslibet, Hor. Sat. 1.4.73-4). This passage suggests that, while the recitatio was a formalized social institution in Augustan Rome, some recitals could be less formal 443 – if the Horatian speaker only recites when pressed by friends, these are not necessarily planned performances and certainly not in a hired auditorium, which he scorns in Epistle 1.19 above. 444 Lefèvre notes that Horace’s reluctance here is in part because he does not 442 Mayer (1994) 267 argues for reading theatris as “auditoria” rather than “theatres” here and cites Nisbet and Hubbard (1978) 16-17, who argue that desit theatris at Hor. Carm. 2.1.10 refers to recitationes, given the association with Asinius Pollio. Nisbet and Hubbard also suggest a wider semantic range for theatrum than just ‘theater,’ and they cite the present passage (Hor. Epist. 1.19.40-1) along with Petron. 90.5, Tac. Dial. 10.5 “with Gudeman’s note,” Gell. NA 18.5.2, and Apul. Flor. 16 as examples of the word theatrum used in the context of a recitatio. 443 Markus (2000) 139ff 444 Dupont (1997) explores and describes various aspects of recitatio as a social institution, and Habinek (2005) 11- 19 discusses Pliny’s portrayal of his own participation in recitationes. Dupont insists (45) that a recitatio was a reading of an already written text. Eros’ anecdote about Vergil in the Vita by Donatus (above) indicates that some 144 trust most audiences to be refined enough to receive his poetry appropriately, 445 the same mistrust that Gurd discusses. 446 Wherever Horace deigned to recite, and whether he read from a written copy or spoke from memory, his audience encountered his poetry in spoken rather than, or possibly in addition to, written form on these occasions. Sometimes people other than the poets recited the poets’ works. In Catullus 50, the speaker and his friend “played around with [the speaker’s] tablets” (multum lusimus in meis tabellis, Cat. Carm. 50.2). The setting was casual and intimate, and the two men were composing off the cuff and riffing on each other’s verses: “we were relaxing… writing little verses, each of us was playing now in this meter and now in that, responding each to the other over laughter and wine” (otiosi…scribens versiculos uterque nostrum / ludebat numero modo hoc modo illoc, / reddens mutua per iocum atque vinum, Cat. Carm. 50.1, 4-6). Gamel argues that we can understand Catullus 50 best as a script for performance, both because of its sound-play and because the performer’s interpretive choices resolve for the audience issues that a reader must encounter as unresolved and possibly unresolvable. 447 In this understanding, the text of the poem recommends its own oral performance. Gamel also writes that “every elegist composed with the expectation that his poems would be performed in dramatic readings.” 448 The puella of Propertius 2.26 recites the speaker’s poems frequently: “for when she recites my poems, she says she hates rich men: / no girl worships poetry so religiously” (nam mea cum recitat, dicit se odisse beatos: / carmina tam improvisation could take place at a recitatio, but it is unclear from the anecdote whether Vergil was reading his text or delivering it from memory. Ovid does not describe Horace’s performances with the verb recito or any related words, but this may have more to do with varying his wording than drawing a distinction between Horace’s performances and Propertius’. Horace uses recito of his own performances. 445 Lefèvre (1990) 11 446 Gurd (2011) 95 447 Gamel (1998) 83-5. Gamel (1998) 84 writes, “Daniel Selden descriges how certain poems of Catullus generate readings that are ‘not simply divergent, but diametrically opposed’ (465), putting the reader into ‘a state of empirical suspense between two irreconcilable positions whose resolution the soliloquy per se will not allow’ (471).” 448 Gamel (1998) 79 145 sancte nulla puella colit, Prop. 2.26.25-6). The epigram by Martial at the head of this section reveals a danger to the poet of other people reciting his or her poetry: “The book you’re reciting, Fidentinus, is mine, / but when you recite it badly, it begins to be yours!” (quem recitas meus est, o Fidentine, libellus; / sed male cum recitas, incipit esse tuus!, Mart. Epig. 1.38). Fidentinus’ poor performance may reflect poorly on his (Martial’s) poetry or may even mangle it beyond recognition. Either way Martial loses. Rust argues that Gellius’ Noctes Atticae was a sort of workbook for elite men to practice performing doctitude, which social competition demanded. 449 Rust’s idea of the pressure on elite men to be able to perform doctitude by quoting apropos sententiae on the spot implies some measure of orality, precisely in the act of recalling an appropriate quotation and saying it out loud. While this ability is not exactly the same as an oral poet’s ability to fit the right formula to the right context (content and sedes), both involve the quick selection of a set-piece from an established stock of set-pieces and sometimes a slight adjustment of the set-piece to fit the needs of the moment. So how to characterize Roman orality? By the first century BCE, orality in Roman culture shared some characteristics with Homeric orality as Parry and Lord and the proponents of Oral Theory who have followed them have characterized it, but the Roman evidence also indicates significant differences. Homeric specialists disagree about how early and to what degree writing became a factor in the composition and transmission of Homeric poetry. 450 In contrast, it is clear that the presence of writing heavily influenced Roman oral literary culture. Some Romanists have 449 Rust (2009) vi, 100, 172. While Gellius was writing in the late 2 nd c. CE, 200 years after Propertius, negotiation of social status through literary performance was not new to his time. Habinek (2005) 133-4 reads Catullus 50 as a “challenge match in which Catullus and Calvus made each other compose verses in various meters” and through which they explored their relationship and after which Catullus remains unsatisfied until he has consituted them retrospectively as amici with this poem. In Catullus 25, Thallus makes a performance of his possession of Catullus’ writing tablets, apparently with the idea that his physical possession of Catullus’ poetry enhanced his social status, much as Gellius’ contemporaries’ mental mastery of famous poetry could enhance theirs. See also Markus (2000) 147. 450 Nagy (1990a), (1990b) (2004); West (2011) esp. 1-77; Montanari (2012) 146 adopted Ong’s term “secondary orality” to describe the relationship between orality and writing in Roman culture. 451 Ong defines “primary orality” and “secondary orality” in contrast with each other: “I style the orality of a culture totally untouched by any knowledge of writing or print, ‘primary orality’. It is ‘primary’ by contrast with the ‘secondary orality’ of present-day high-technology culture, in which a new orality is sustained by telephone, radio, television and other electronic devices that depend for their existence and functioning on writing and print.” 452 It is useful to differentiate between Ong’s “primary orality” and the Roman situation, but “secondary orality,” as Ong defines it, depends largely on mass-diffusion of audio recordings. Writing is involved in that it makes possible the technology that produces and distributes the audio recordings. These recordings can be listened to by millions of people in different times and places with no variations in the performances, and a listener who enjoys a particular rendition can return to it over and over. 453 Writing has made this possible in the 20 th and 21 st centuries but of course not in ancient Rome, where every performance was ephemeral, and where writing recorded what was said or sung but not the speaking or singing itself. Ziolkowski instead contrasts “primary orality” with “academic orality,” citing Chapter 3 of Ong. 454 Ong does not actually use the phrase “academic orality” but rather discusses the effect of “academic rhetoric” on orality in Europe from Classical Athens through the Romantic Period. 455 We can, with Ziolkowski, try to extrapolate an idea of “academic orality” from what Ong has written, with such key concepts as “expression as basically agonistic and formulaic” with loci communes 456 and the self-consciously learned use of 451 Gamel (1998) 81 452 Ong (1982) 11 453 The listeners’ surroundings and audio-play devices may affect how they listen and what they hear, but Ella Fitzgerald’s scatting will never change, Madonna will always sigh exactly the same way at exactly the same moments in “Like a Virgin,” and, regrettably, certain Queens of the Night will always miss their top F’s. 454 Ziolkowski (2005) 129 455 Ong (1982) 108-12 456 Ong (1982) 109 147 these loci communes by speakers who were formally trained in schools. 457 This idea of “academic orality” seems to fit elite Roman literature and its producers, but it is less appropriate to the non- elite oralities that Horsfall describes (above). Vogt-Spira argues that, “for [ancient] practices of reading, the systematic distinction between phonic and graphic is irrelevant.” 458 His conclusion that “orality and writing are not rival media of the same order” 459 is in the context of a discussion of reading aloud, so it may tell part of the story but not the whole story of Roman orality. 460 Further evidence for the relationship between writing and oral composition in the Roman world, in particular as regards notions of authorship and authority over a text, can be found in versified graffiti and inscriptions. A number of graffiti from Pompeii and other inscriptions from Rome and other parts of the empire show some degree of oral composition, both through errors in spelling and grammar and through the use of formulaic phrases characteristic of oral poetry. So far, this phenomenon has been studied only in the graffiti of Pompeii, perhaps because they are a relatively densely clustered data set compared to the appearances of these types of phrases and motifs in Latin inscriptions from other places. But even the Pompeiian graffiti have, to my knowledge, only received one article focused on the evidence for orality. 461 Some inscriptions have errors in their citations of famous poetry that suggest oral composition. 462 Wachter lists seven examples, including these two: tu dea tu prese nostro succurre labore (CIL IV 2310k, add. p. 216) quotes Aeneid 9.404 but gets Vergil’s words praesens 457 Ong (1982) 111-12 458 Vogt-Spira (1991) 327, translated 459 Vogt-Spira (1991) 327, translated 460 Cf. Nagy (1990b) 9: “If something is 'oral', we tend to assume a conflict with the notion of 'written'. From the standpoint of cultural anthropology, however, it is 'written' that has to be defined in terms of 'oral'. 'Written' is not something that is not 'oral', rather, it is something in addition to being oral, and that additional something will vary from society to society.” 461 Wachter (1998) 462 Wachter (1998) 74 148 and labori wrong, and Idai cernu nemura (CIL IV 6698) changes grammar and spelling in verse 730 of Seneca’s Agamemnon, which reads Idaea cerno nermora. 463 Similarly, variants on a common couplet in Pompeian graffiti, 464 quis(quis) amat valeat, pereat qui nescit amare, / bis tanto pereat, quisquis amare vetat (CIL IV 4091), include quisquis ama(t) valia(t), peria(t) qui no(n)sci(t) amare, / bis [t]anti peria[t], quisquis amare vota(t) (CIL IV 1173) and cuscus amat valeat, pereat qui no(n)scit amare (CIL IV 3199). Wachter argues that the inscribers of these graffiti probably did not have written copies of the texts in front of them but rather were likely working from memory. Milnor reaches a similar conclusion in her analysis of CIL IV 5296, the love poem from one woman to another discussed in Section B, but she but also considers the material evidence of the graffito to suggest a possible scenario for its production: “CIL 4. 5296 contains certain linguistic indicators—the loss of the final ‘m’ from cum and vigilarem most notably 465 — which suggest some relationship with Latin as it was spoken, but, on the other hand as I have noted, the use of interpuncts is an affectation drawn strictly from the world of writing. In addition, even a cursory examination shows that the graffito is surprisingly neat, carefully inserted into the corner of the painted wall decoration where the lines turned to run parallel to the ceiling; there is an even space between the first letters of the lines and the decorative border which originally framed the white panel into which it was written. The text may well have been composed orally elsewhere, but there was real care and interest given to the production of CIL 4. 5296’s written form.” 466 Milnor argues that the first three verses of CIL IV 5296 may have come from one source text, while the rest of the verses may have come from one or more other source texts, and this is why she suggests that oral composition would have happened “elsewhere.” In Milnor’s scenario, one or more of the source texts may have been composed mentally or even aloud rather than on, for example, wax tablet. The graffitist, in turn, may have combined source texts in her head or, even 463 Wachter (1998) 74 464 Wachter (1998) 76n15: “Bü. 945, 946, 946 Anm. und Lo. 2063 (mit CIL 6782 quisquis amat valeat / rumna / quis, und dem späteren CIL 9130 [= Not. 1913 p. 147 bei Lo.] quisquis amat valeat pereat), ferner CIL 5272 (quisquis amat v[...]) und So. 66 (Schluß) quisquis amat valeat. - Unklar ist mir CIL 9202 (falsche Metrik oder falsche Lesung?).” 465 CIL IV 5296.5: saepe ego cu(m) media vigilare(m) perdita nocte 466 Milnor (2014) 219 149 if she 467 drafted the composition on a tablet, she may have reproduced the verses from memory rather than from written copies. Wachter also identifies several phrases that, both in form and employment, look like the formulas familiar from Homeric poetry. Parry writes, “The formula in the Homeric poems may be defined as a group of words which is regularly employed under the same metrical conditions to express a given essential idea.” 468 The phrases discussed below fit this description. The phrase quisquis amat valeat, for example, begins the hexameter of the famous couplet discussed above: quis(quis) amat valeat, pereat qui nescit amare, bis tanto pereat, quisquis amare vetat (CIL IV 4091 = Bü. 945, see above). But some graffiti change the phrase to mean the opposite by replacing valeat with the metrically equivalent pereat 469 : Quisquis amat pereat..[], (CIL IV 4659) Quisquis amat perea[], (CIL IV 4463) Quisquis amat per[]. (CIL IV 5186) This example of change fits Lord’s description of an oral formula as “a living phenomenon of metrical language” that has “a dynamic life.” 470 Other phrases that Wachter identifies as formulaic include si quis and si quis amat at the beginning of a hexameter, which are metrically equivalent to quisquis and quisquis amat and similar enough in meaning to be sometimes interchangeable (“if someone/anyone…” and “whoever…,” respectively) but also different enough to convey distinct 467 Milnor (2014) 196-218 argues that the speaker of the poem and the beloved are both women and that, since the graffito takes “an ironic approach…. to gendered language and stereotypes” (218) and portrays homoerotic desire between women sympathetically, whereas Roman men tended to portray it as monstrous and disgusting, the common default assumption that the graffitist was a man is problematic at best and runs afoul of evidence we can glean from the graffito. 468 Parry (1930) 80 469 Wachter (1998) 76 470 Lord (2000) 30 and cf. Parry (1930) 84-5 on changes in formulas 150 meanings; 471 quisquis am… and si quis am… after the caesura of a pentameter; 472 …amare potest at the end of a pentameter; 473 and non amplius + disyllable at the end of a hexameter. 474 Milnor treats this material briefly, and her stance, while more cautious, still seems to support Wachter’s thesis: “Certain stock phrases—such as quisquis amat valeat or venimus hoc cupidi—which show up in multiple graffiti, in various permutations, would seem likely to have been the building blocks of popular songs.” 475 Similarly formulaic phrases (and, in fact, some of the same ones) appear in inscriptions from other parts of the Roman world as well. Many inscriptions are also only approximately dated or have not yet been dated as far as I have seen. I have found dates, both specific and approximate, for some, though, and have included inscriptions dating from the 1 st century BCE to late antiquity. I have indicated the earlier inscriptions, in order to show that evidence of oral composition appears in inscriptions from before and around the time of Propertius as well as after. Original locations are often easier to establish, and I have given them where possible. The formula si quis at the start of a hexameter occurs in inscriptions from Rome 476 and Algeria 477 in addition to Pompeiian graffiti. Several other formulaic phrases also occur in inscriptions found at Rome. One example is causa doloris, sometimes preceded by a dative pronoun, at the end of a hexameter: 471 Wachter (1998) 76 472 Wachter (1998) 76 473 Wachter (1998) 77 474 Wachter (1998) 83-5 475 Milnor (2014) 219 476 seiquis havet nostro conferre dolore(m) (CIL I.2 1222.1). Archaic spellings suggest a relatively early date for this inscription. The CIL lists it in volume I.2, part 2 fasc. 1, called Inscriptiones Latinae Antiquissimae. Warmington’s Loeb lists it among others from the 1 st c. BCE (p. 28). The inscription also has another formula: ferre, usually with a prefix, followed by a trisyllabic form of dolor at the end of a hexameter occurs in Lucr. DRN 3.990; Verg. Aen. 6.464, 9.426; Prop. 2.15.35; Tib. (Lyg.) 2.3.2; and Val. Flac. Arg. 1.766. The inscribed verse is not a full hexameter, but it stands in place of one: While it has only five feet, it is the first verse in an inscription made up of elegiac couplets, and it is dactyllic, but it is a dactyllic hexameter missing a foot before the fifth-foot dactyl rather than a dactyllic pentameter. 477 si quis forte v[elit curiose scire viator (CLE 1952.1) 151 ne tibi sim penitus quaerenti causa doloris (CIL VI 10969.7 – Rome, 117-138 CE 478 ) heu memorande pater, longi mihi causa doloris (CIL VI 31937.1 – Rome 479 ) nec mea plus i]usto sit mors tibi [causa doloris 480 (CLE 389.6) Another formula found at Rome is a reference to death (sometimes a metaphor) followed by a disyllabic verb meaning “shut” followed by ocellos, which occurs at the end of a hexameter: Vis] nulla est animi, non somnus claudit ocellos, (CLE 943.1) . . . . . somnus claudit ocellos (CLE 1811) 481 . . m]eos morientes texit ocellos (CLE 452.2) Note that the verbs in these three inscriptions, claudit and texit, are both disyllabic; when we revisit this formula below, the examples will use other verbs, some of which will be disyllabic, and some of which will not. Yet another formula in inscriptions found at Rome is corpus inane followed by a disyllabic word in the second half of a pentameter: desine mirari: corpus inane iacet. (CIL VI 17342.2- Rome) 482 corpora sémanimés corpus inane colunt. (CIL VI 12307.2 – Rome) Some formulas are found in inscriptions from both Rome and elsewhere. One of these is ossa sepulcro or membra sepulcro at the end of a hexameter, sometimes preceded by requiescunt: hic mea ferali requiescunt ossa sepulcro (CIL VI 11407.3 – Rome) 478 EAGLE Epigraphic Database number EDR133273 479 This is perhaps the latest inscription in this section, dated to 534 CE (CLE 707). 480 The identification of the formula in this inscription depends, of course, on accepting the editors’ restitution of the line. The inscription was edited and published by Bormann in Inscr. Lat. Nov., p. 8, and Rossi in Act. Inst. Arch. 1877, p. 56 481 It is possible that this is the same inscription as CLE 943. Bücheler ad loc.: “inspice 943, aut bis idem scriptum murus obtulit aut facinus factum est indignum.” 482 Late 2 nd c. CE (CIL ad loc. cit.) 152 condidit et ca]rus mihi con[iux ossa sepulcro (CLE 1296.5 – Rome) 483 Hoc rudis aurigae requiescunt ossa sepulchro (CIL II 4314.1 – Hispania) Non clausa in tumulo requiescunt ossa sepulcro (CIL III 1871.1 – Dalmatia) huc veniens placido posuit pia membra sephulcro (CLE 612.2 - Rome) In these examples, ossa sepulcro or membra sepulcro fills the fifth and sixth feet of the hexameter, and when requiescunt is there, the formula requiescunt ossa sepulcro fills out the verse from the strong penthemimeral caesura. The importance of caesuras, especially the penthemimeral caesura, in the formulation of Greek oral hexameters is well documented. 484 The group of formulas requiescunt ossa sepulcro and ossa sepulcro 485 is comparable with Homeric formula groups that include variants which go from the strong penthemimeral caesura to the end of the verse, such as πάθεν ἄλγεα ὅν κατὰ θυμόν and χαλεπὸν δέ τοι ἔσσεται ἄλγος, and other variants which fit into just the fifth and sixth feet (i.e., after the bucolic diaeresis 486 ), such as ἄλγεα πάσχει, ἄλγε’ ἔχουσιν, πήματα πάσχει, and πῆμα πάθηισιν. 487 In each formula group, the members express the same basic idea, 488 and both formula groups have members to fit into these two important metrical positions (from the strong penthemimeral caesura to the end of the verse and from the bucolic diaeresis to the end of the verse). Another formula found in inscriptions from both Rome and one of the provinces is quandocumque at start of hexameter about being dead, which occurs in inscriptions from Rome and Africa proconsularis: Quandocumque levis tellus mea conteget ossa incisum et duro nomen erit lapide, 483 The identification of the formula in this inscription also depends on accepting the editors’ restitution of the line. 484 Parry (1930) 86, 127, 131; Nagy (1990a) 416-17 485 See below for more members of this formula group, as well. 486 Note that these four inscriptions break with the preference for a bucolic diaeresis following a trisyllabic fourth foot and against a bucolic diaeresis following a disyllabic fourth foot. 487 Parry (1930) 130-1 488 Parry (1930) 130 153 quod si forte tibi fuerit fatorum cura meorum ne grave sit tumulum visere saepe meum. (CLE 965 – Rome) 489 Quandocumq. tellus mea conteget ossa, incisum in duro nomen erit titulo, tum tibi si qua mei fatorum cura manebit, ne grave sit tumulum visere saepe meum. (CIL VIII suppl. 15716 – Afr. procos.) The use of quandocumque in these inscriptions may not at first seem oral-formulaic. To begin with, the inscriptions’ first couplets are nearly identical, and their fourth verses match verbatim. But orality does not necessarily imply a poetics that is, in principle, open to varying a text, though variation may occur in practice whether or not it is intended. 490 Also, the first line of CIL VIII suppl. 15716 is missing the word levis, so the hexameter is incomplete. This error suggests that the similarities are more likely the result of memorization than of written copying, as does the change from et to in in the respective second verses. Whether the composer of CIL VIII suppl. 15716 borrowed from CLE 965 or whether they both borrowed from another source is hard to tell. The error in the first verse of CIL VIII suppl. 15716 also suggests oral or mental composition rather than written composition, where metrical errors are easier to avoid. Another feature of these inscriptions that may suggest oral composition is their lack of enjambment. Lord writes, “This absence of necessary enjambement is a characteristic of oral composition and is one of the easiest touchstones to apply in testing the orality of a poem. Milman Parry has called it an ‘adding style’; the term is apt.” 491 A factor that may shed doubt on whether this use of quandocumque is oral-formulaic is that fact that it is only one word, while Parry’s definition of a formula is “a group of words…” 492 489 Buecheler ad loc. dates this inscription (CLE 965) to 10 CE based on the consuls listed in the prescript: Rusticellia M. I. Cytheris debitum reddidit X k. Sept. Maluginense et Blaseo cos. 490 Nagy (1990a) 54-5 491 Lord (2000) 54, though cf. Parry (1930) 127 on a method of enjambment common in Homer 492 Parry (1930) 80 154 At four syllables spanning a spondee and most of a dactyl (unless elided), 493 though, quandocumque is as long and metrically particular as some Homeric formulas. 494 Further support for the idea that this use of quandocumque is oral-formulaic comes from its appearances in the same context (verses about being dead) and metrical position in famous poems: 495 quandocumque igitur vitam mea fata reposcent et breve in exiguo marmore nomen ero (Prop. 2.1.71-2) Quandocumque igitur nostros mors claudet ocellos (Prop. 2.13.17) quandocumque hominem me longa receperit aetas (Lydg.[Tib.] 3.7.210) In fact, occurrences of several phrases in particular metrical positions in verses by famous authors help to identify these phrases as formulas when they occur in inscriptions. An example is the phrase quod si at the start of a verse, especially a hexameter: quod si fata mihi dedissent luce videre (CIL VI 3608 – Rome) From the inscription by itself, the phrase looks generic rather than formulaic. But it appears so many times in the verses of famous poets 496 that its sheer frequency and variety of contexts suggest formulaic use. Furthermore, several variants are also attested. One of the inscriptions with quandocumque (above) also has the phrase quod si forte at the beginning of a hexameter: 493 Or 7 units of time, if each short syllable is a unit. 494 πάθεν ἄλγεα, Parry (1930) 130; νήπιοι οἱ and δυσμόρωι ὅς, Parry (1930) 127 – these two are particularly apt for comparison with quandocumque because they, like it, are connecting words rather than words carrying the main idea of the sentence: “Very common in Homer is the device of using and adjective followed by a relative clause to continue a sentence which might have come to an end with the preceding verse” (Parry 1930: 127). 495 quandocumque also begins hexameters in contexts unrelated to death at Hor. Ep. 1.14.17, 1.16.58. In Ovid, quandocumque begins hexameters that can be related to metaphorical deaths at Met. 6.544 and Trist. 3.1.57. 496 Accius frag. 312; Paut. Curc. 69, Pers. 402, Trinum. 217; Ter. Andria 244, Eun. 35, Eun. 924, Heaut. 157, Hec. 24, 519, Phorm. 9, 171, 201, 586, 738; Lucil. Sat. frag. 174, 694, 696; Rutilius Rufus frag. 14.4; Cat. Carm 14.8, 15.14, 42.16; Lucr. DRN 1.180, 1.213, 1.234, 1.335, 1.410, 1.904, 2.47, 2.481, 2.772, 2.985, 3.612, 3.714, 3.748, 3.766, 3.790, 5.134, 5.195, 6.387; Verg. Aen. 6.133, 7.310, 11.166, 11.357, 11.434; Prop. 1.1.37, 1.5.25, 2.7.15, 2.10.5, 2.15.17, 2.20.19, 2.24a.5, 2.28.25, 2.30.31, 2.32.61, 3.7.43, 3.14.35, 3.16.21, 3.17.13; Tib. 1.3.53, 2.6.7; et al. 155 quod si forte tibi fuerit fatorum cura meorum (CLE 965.3 – Rome) 10 CE 497 This variant, sometimes adjusted to suit different grammatical contexts, occurs in a number of famous poets’ verses, as well. 498 I have found other variants, quod mihi si and quod tibi si, at the beginnings of hexameters in famous poets’ works 499 but not in inscriptions. Taken together, these phrases resemble Parry’s Homeric formula groups discussed above with ossa sepulcro and requiescunt ossa sepulcro. 500 Other phrases from inscriptions whose identification as formulas is supported by evidence from famous poets include causa doloris at the end of a hexameter 501 and the group of a 4-syllable verb ending in a vowel (elided) + amantes/amantis at the end of a hexameter 502 or the metrically equivalent 3-syllable verb ending in a consonant + amantes/amantis at the end of a hexameter. 503 A closer look at one of these formulas will lead to some observations about formulas and oral composition in Roman poetry. The phrase ossa sepulcro and its variants occur not only in the inscriptions above but also at the ends of the following hexameters by famous poets: heu quam crudeli condebat membra sepulcro (Enn. Ann. 2.139) de superis queror; haud parvo data membra sepulcro (Sil. Pun. 13.691) exitus instanti praemisit membra sepulcro (Stat. Silv. 5.3.259) 497 Buecheler ad loc. dates this inscription (CLE 965) to 10 CE based on the consuls listed in the prescript: Rusticellia M. I. Cytheris debitum reddidit X k. Sept. Maluginense et Blaseo cos. 498 Lucr. DRN 1.391, 1.665, 2.225, 2.924, 2.931, 3.533, 3.698, 3.722, 3.819, 5.338; Prop. 1.5.9, 2.13.15 (quae si forte), 2.14.31, 2.22.11 (quae si forte), 2.26.13; Ov. Fast. 3.87, Epist. 2.5.33 (qui si forte); App. Verg. Aetna 307, 330, 402 (quem si forte), 492 499 quod mihi si: Prop. 2.1.17, 2.15.37, 2.26.57, 3.6.41; Ov. Pont. 3.4.14. quod tibi si: Cat. Carm. 64.228; Tib. 3.7.201; Ov. Met. 13.113 500 Parry (1930) 130-1 501 CIL VI 10969.7, CLE 707.1, and CLE 389.6 above and Prop. 1.16.35; Verg. Aen. 9.216; Ov. Her. 15.117, Met. 1.508, 1.734, Trist. 4.3.33; Sil. Pun. 5.599, Mart. Epigr. 10.41.3 502 CIL VIII 27380.5 (crudelis, quae sola potest disiungere amantes) and Prop. 1.10.15, 1.13.25, 2.7.3, 2.14.19, 2.26.33, and 3.14.21 503 Verg. Ecl. 8.26; Prop. 1.16.45, 1.19.25, 2.5.13, 3.17.5; Tib. 1.4.77, 1.8.71, 2.4.39, Tib.[Sulp.] 4.4.19 [non laedit amantes], Tib.[Sulp.] 4.6.7; Ov. Am. 1.9.25, Ars 2.515 [quod laedat amantes cf. Tib. [Sulp.] 4.4.19], Met. 4.68, 4.679, 7.719, Rem. 611, 755; Stat. Sil. 1.5.18, 3.5.46 156 hoc ultor Capaneus operit tua membra sepulcro (Stat. Theb. 9.565) ne mea contempto lapis indicet ossa sepulcro (Prop. 3.1.37) illa mihi sancta est, illius dona sepulcro (Tib. 2.6.31) per tamen ossa viri subito male tecta sepulcro (Ov. Her. 3.103) Membra pater Libyco posuit male tecta sepulcro (ps.-Sen. Epigr. 402.3) strage virum mersus Trebia est, atque ora sepulcro (Sil. Pun. 9.189) The switching between membra and ossa looks like what Parry and Lord call a “[formula] system” 504 or a “substitution system.” 505 In this system, membra sepulcro is the earlier variant, and it survives into the later examples as well. It also appears in CLE 612.2, which is probably from no earlier than the 3 rd c. CE. 506 The more common variant, by virtue of several appearances in inscriptions (above: CIL VI 11407.3, CLE 1296.5, CIL II 4314.1, and CIL III 1871.1), is ossa sepulcro, which strictly speaking appears only in Propertius among the famous poets, though Heroides 3.103 makes reference to it, and Silius Italicus’ ora sepulcro may also play on it. Heroides 3.103, in playful Ovidian fashion, varies the formula and appears to self- consciously highlight the variation by putting ossa earlier in the verse. The ps.-Senecan Epigram 402 seems to have picked up on this trick: Like Heroides 3.103, verse 3 of the epigram ends in the phrase male tecta sepulcro, but where the Ovidian verse places ossa in the second foot, the epigram has membra in the first foot, which varies Ovid’s conspicuous display of his own 504 Lord (1960) 35: “…the many words that can be substituted for the key word in such formulas. For example, in the Prilip formulas above, any name of a city with a dative of three syllables can be used instead of Prilip: u Stambolu. u Travniku, u Kladuši. Instead of a u kuli, ‘in the tower,’ one can say a u dvoru, ‘in the castle,’ or a u kući, ‘in the house.’ These formulas can be grouped together in what Parry, when studying the traditional epithets in Homer, termed ‘systems.’” Parry (1930) 132 505 Lord (1960) 36 506 Marini (1785) 31 found this inscription in the Cemetery of Ciriaca, now the Cimitero del Verano, where the earliest layer of construction it is likely to have been set in is from the 3 rd c. CE. 157 variation. In these verses, ossa and membra are ‘Alexandrian footnotes’ 507 that link the variant back to the original formula. But to which version does the footnote in each of these verses point? The earlier verse, from the Heroides, appears to point back to Propertius 3.1.37 with ossa (though of course an earlier instance of ossa sepulcro may simply not have survived), while the later, ps.- Senecan verse, which quotes the words male tecta sepulcro from the Ovidian verse, seems to draw the word membra from Ennius’ Annales 2.139, which predates Propertius’ elegies by a century and a half (though, again, other examples from between Ennius and the ps.-Senecan author may have been lost). Does the Propertian ossa sepulcro point back to the Ennian membra sepulcro or to another, now lost predecessor? Or was membra/ossa sepulcro a formula by Propertius’ time? If so, is Propertius’ use of it formulaic, in which case it does not point back to any predecessor in particular? Or is Propertius’ use ‘Alexandrian,’ in which case it does point back to a predecessor? Or did Propertius coin the phrase ossa sepulcro and place it at the end of a hexameter independently? We cannot answer all of these questions, but we can nevertheless draw some conclusions from amid the ambiguities. Phrases such as membra/ossa sepulcro may not have been formulaic in their first occurrences but may rather have become formulaic over time. Our first attested use of membra sepulcro at the end of a hexameter, in Ennius’ Annales 2.139, may have been the original occurrence of what would become a formula. Ennius uses it once in his surviving verses, so we cannot say whether it was a formula for him, though the fact that the phrase and its variants do not appear in the same metrical position again until Propertius supports the guess that Annales 2.139 was the only place Ennius used it. Subsequent poets pick up the phrase, and some of them— Statius, Silius Italicus, and especially Ovid—use it multiple times, and sometimes they vary it, all 507 Hinds (1998) 1-2 158 of which chimes with how Lord describes individual poets’ development and use of oral formulas. 508 Poets such as Propertius, Ovid, and Statius were often specific and precise with their allusions, but that does not mean that all of their audiences or readers caught every allusion or correctly identified its source. When a listener or reader might encounter a phrase and its variants in the works of famous poets and (probably) not-so-famous poets, as with membra/ossa sepulcro in the famous verses and inscriptions listed above, the original meaning of what was becoming a formula, and its attribution to its originator, could be lost: “It is certainly possible that a formula that entered the poetry because its acoustical patterns emphasized by repetition a potent word or idea was kept after the particular potency which it symbolized and which one might say it even was intended to make effective was lost – kept because the fragrance of its past importance still clung vaguely to it and kept also because it was now useful in composition. It is then that the repeated phrases, hitherto a driving force in the direction of accomplishment of those blessings to be conferred by the story in song, began to lose their precision through frequent use. Meaning in them became vestigial, connotative rather than denotative.” 509 The process of becoming a formula, then, can efface the authorship of a phrase, in terms both of who the author was and of any notion of authority over how the phrase is used and what it means. This seems to have happened to at least one usage that Propertius coined: o utinam (“oh, if only”) at the start of a verse. The phrase o utinam is extant only once in a famous author’s work before Propertius, 510 in a fragment of a comedy by Turpilius from the 2 nd century B.C.E., and there the o is reconstructed. Propertius begins five hexameters with this phrase, 511 and after that there are more than two hundred instances of the phrase in other authors from Antiquity and the Middle Ages. It also occurs in inscriptions: 508 Lord (1960) 34 509 Lord (1960) 65 510 Tibullus uses the phrase once, at 1.3.2. Conte (1994a) 326 writes that Tibullus’ first book was “begun after 32 and published in 26 or 25,” so it was composed at the same time as Propertius’ first book and published after it. I have been unable to find a date for CIL IV suppl. 5296. The earliest securely dated Pompeiian graffito is CIL IV 1842, from 78 BCE (Milnor 2014: 17), and no Pompeiian graffito could have been made after Vesuvius’ eruption in79 CE. 511 Prop. 1.3.39, 1.8.9, 1.16.27, 4.4.33, and 4.4.51 159 o utinam t[- - -] (Ben Abdallah et al. 2005: 103 – Ammaedara) 512 o utinam Italiae potius mea fata dedissent (AE 2008: 1099.14-15 – Carnuntum 31-70CE) o utinam vivo potuissem praemia morum (CIL VI 23472 – Rome) o utinam iunctos licuisset frui... (CIL VI 26489.7 – Rome) 513 [o utinam d]igitis operirent lumen inane (CIL VI 36658 – Rome) 514 o utinam possit reparari spiritus ille (CIL XII 825 – Arelate) O utinam liceat collo complexa tenere (CIL IV suppl. 5296.1 – Pompeii) Poets, such as Tibullus and Ovid, who used the phrase soon after Propertius, may have thought of it as a Propertianism. But later poets, including the poets of the inscription from Ammaedara and the graffito from Pompeii above, may just as easily have drawn the phrase from Ovid or used it without a particular reference in mind. On the other hand, some inscriptions combine formulas in ways that suggest specific allusions to famous poets: Quandocumque levis tellus mea conteget ossa incisum et duro nomen erit lapide, quod si forte tibi fuerit fatorum cura meorum ne grave sit tumulum visere saepe meum. (CLE 965 – Rome) 10 CE 515 As discussed above, this inscription uses the formulas quandocumque at the start of a hexameter and quod si forte at the start of a hexameter. The inscription deploys these formulas in contexts similar to those in which they appear in Propertius: quandocumque igitur vitam mea fata reposcent et breve in exiguo marmore nomen ero 512 Ben Abdallah et al. (2005) 103 tentatively date the Ammaedaran inscription from which this fragment of a verse comes to the late 2 nd c. CE and argue that the verses in this inscription were probably hexameters. 513 The CIL entry identifies this verse as the hexameter of an elegiac couplet. 514 The restitution is by Buecheler. 515 Buecheler (ad loc. cit.) dates this inscription (CLE 965) to 10 CE based on the consuls listed in the prescript: Rusticellia M. I. Cytheris debitum reddidit X k. Sept. Maluginense et Blaseo cos. 160 ... ... si te forte meo ducet via proxima busto (Prop. 2.1.71-5) Quandocumque igitur nostros mors claudet ocellos, accipe quae serves funeris acta mei. (Prop. 2.13.17-18 516 ) quod si forte tibi properarint fata quietem, illa sepulturae fata beata tuae (Prop. 2.28.25-6) Like CLE 965, these verses are about death and burial. While the positions of words are changed, there are similarities of phrasing and ideas in addition to the formulas. In CLE 965.2, duro…lapide and nomen erit echo exiguo marmore and nomen ero in Prop. 2.1.72. Similarly, both CLE 965.3 and Prop. 2.28.25 extend the formula quod si forte to quod si forte tibi, and both follow it with future perfect verbs, a not uncommon variation on the normal form of a future more vivid condition, which Latin grammars are at pains to explain variously, but which they generally agree indicates some extra emphasis or temporal specificity. 517 Also, Prop. 2.1.75, Prop. 2.13.18, and the second couplet of the inscription display uncertainties about the addressees’ reactions to the speakers’ deaths. Given these similarities, CLE 965.3 looks as though it may have borrowed phrasing from Prop. 2.28.25 and ideas from Prop. 2.13.18 and Prop. 2.1.75, especially as si te forte and quod si forte tibi are not very different. CLE 965 also shares phrasing and ideas with verses 198-210 of the Panegyricus Messallae: …mea conteget ossa (CLE 965.1) resembles mea…texerit ossa (Panegyricus 204), and quandocumque… begins hexameters about dying in both poems (CLE 965.1; Panegyricus 210). 518 Verse 4 of CIL VI 3608 (quod si fata mihi dedissent luce videre) 516 Fedeli (1984) splits Poem 2.13 in two after verse 16, so in his text these verses are 2.13b.1-2. I find Richardson’s argument for the unity of Poem 2.13 convincing (1977: 247), especially as Shackleton Bailey (1956: 89) and Goold (1999) leave 2.13 as a single poem, and Heyworth (2007: 163) also argues for unity. 517 Allen and Greenough’s New Latin Grammar for Schools and Colleges §516c; Bradley’s Arnold Latin Prose Composition §456; Moreland and Fleischer, Latin: An Intensive Course p. 38; Wheelock’s Latin p. 274. 518 Whether the Panegyricus Messallae was written before or after Book 2 of Propertius is difficult to say. Conte believes that the Panegyricus was “probably written not long after 31, the year of [Messalla’s] consulship” (1994a: 331). On the other hand, Glock (2006) writes that it “is generally regarded as impossible” that the Panegyricus was 161 similarly combines the formulas quod mihi si at the start of a hexameter and si fata dedissent. This combination of formulas may point to Prop. 2.1.17 (Quod mihi si tantum, Maecenas, fata dedissent), but more likely it points to Lucan’s Pharsalia 1.114-15 (…quod si tibi fata dedissent / maiores in luce moras…) – the word luce is telling, as is the synecdoche of seeing light or spending time in the light for living longer. While CLE 965 and CIL VI 3608 use formulas, then, the particular combinations of formulas and the words among which the inscriptions set the formulas reassert specific allusions against the general tendency of formulas to lose that specificity. Inscriptions also use motifs in ways characteristic of oral poetry and other kinds of orality. An example is the idea of placing flowers or other plants on a grave: sit tibi terra levis, cineres quoque flore tegantur (CLE 451.3 – Rome) hic in flore cubat longum securus in aevom (CIL XIV 3826.4 – Rome) et cingant suaves | ossa sepulta rosae (CIL VI 20466.4 – Rome) floribus ut saltem re[qu]ie[s]cant membra iucundis (CIL III 754.1 – Moesia inferior) These verses all come from funerary inscriptions. Propertius uses the same motif several times 519 in different rhetorical contexts than the funerary inscriptions. In Propertius 1.17, 2.13, and 3.16 the speaker is not actually dead but rather fantasizing about his own death and using the flowers as a symbol of the loyalty he wishes the puella would show. In Propertius 4.7, the speaker ventriloquizes Cynthia’s ghost and has her accuse him of neglecting her grave and demand that he tend it properly, which for her includes planting ivy over it. The funerary inscriptions, in contrast, assume that the flowers are already on the grave and wish the occupants of the graves a peaceful rest. The relationships between the dead and those who place flowers over their graves are secure, written before “linguistically comparable passages by Augustan poets” and cites Prop. 2.10.5f., Hor. Carm. 2.20.1-3, and Verg. Aen. 1.126f. She concludes that “we can only assume that it was written after 20 BC.” 519 Prop. 1.17.22, 2.13.33-4 (a laurel rather than flowers), 3.16.23-4, and 4.7.79-80 (ivy rather than flowers) 162 at least in that respect. Other motifs shared between inscriptions and elite poets include life as a road with death as a goal at the end, 520 living honestly and with good character, 521 enjoying “the fruit of one’s hard work,” 522 and more. 523 Lord calls such phenomena “themes” in oral poetry: “The theme can be defined as a recurrent element of narration or description in traditional oral poetry. It is not restricted, as is the formula, by metrical consideration; hence, it should not be limited to exact word-for-word repetition.” 524 Lord is talking about what Ong calls “primary orality,” 525 though. With writing more firmly a part of the Roman scene than the Homeric, identifying oral composition with themes as opposed to written composition with themes becomes more difficult. Ong points out that the oral use of themes can also occur in highly literate contexts: “Developing a subject was thought of as a process of ‘invention’, that is, of finding in the store of arguments that others had already exploited those arguments which were applicable to your case. These arguments… were often called loci communes or commonplaces when they were thought of as providing arguments common to any and all subject matter.” 526 For Ong, these loci communes are an element of oral composition in the practice of “academic rhetoric.” So to take an example, when Vergil, Propertius, and Ovid played on the theme of the drowning of Hylas, 527 to what degree was oral composition involved? In Vergil’s version, sound- effects seem to beg for oral performance: 528 The words Hy ̆ lā, Hy ̆ lă reproduce the petering out of 520 CLE 998.1-2, Prop. 3.18.22, and see Lissberger (1934) 33 521 CLE 991.1, CLE 992.1, Prop. 2.32.34, and see Lissberger (1934) 101 522 Lissberger (1934) 111, translated. CLE 1988.13, Prop. 3.13.17, and see Lissberger (1934) 111. 523 See also Milnor (2014) 192-3. 524 Lord (1951) 73 525 See above. 526 Ong (1982) 110 527 Verg. Ecl. 6 and Prop. 1.20. See Chapter 2 and Petrain (2000) and Heerink (2015) 83-112 for discussion of Hylas, echo, and intertextuality. 528 Cf. Gamel (1998) 83-4 on sound effects in Catullus 50. 163 an echo with the second “a” shortened by correption (Verg. Ecl. 6.43-4). Vergil could conceivably have produced this effect through either written or oral composition, though he composed orally and extemporaneously and had his secretary write the words down on at least one occasion. 529 Propertius’ version is also conscious of sound, in the way it suggests opposite subtexts for Hylas’ name when Hercules calls it out and the mountains echo it back (Prop. 1.20.49-50), though his text itself does not produce any sound effects. Ovid appropriates the echo effect of correption in his telling of Echo and Narcissus (Ov. Met. 3.499-501). Hylas and Narcissus are both beautiful youths who die by water, and echoes are involved in both of their stories. Ovid borrows the phrase vale, vale with correption of the second “e” from another Vergilian passage about lovers parted (Verg. Ecl. 3.78-9) but adds the insight from Eclogue 6 that repetition with correption can mimic an echo. He thus uses a Vergilian formula—repeated disyllable with correption of the final vowel 530 —to conflate two similar themes. Lord explains that “once a singer has a command of the common themes of the tradition, he has merely to hear a song which is new to him only once to be able to perform it himself” and that this is how themes facilitate oral composition. 531 Ovid probably encountered Vergil and Propertius’ poems more than once, but the way he combines elements from Eclogue 3 and Eclogue 6 makes it difficult to distinguish between brilliant allusivity and brilliant use of formulas and themes to facilitate composition. Perhaps they do not always need to be distinguished. Conclusion As we saw in Chapter 1, Propertius’ speaker accepts that other users of poetry could change his text in some of the ways discussed in this chapter. While altering the words themselves is the 529 See above. 530 Cf. Parry (1930) 132 and Lord (1960) 35 on substitution of key words in formulas, discussed above. 531 Lord (1951) 73-4 164 most obvious way that other people changed Propertius’ text, displacing Propertius’ words from their original contexts and placing them into new contexts could also radically change what the words communicate. Some of these new contexts are literary, as when Caesius Bassus or Ovid appropriates words from Propertius for his own ends. From the graffiti and epitaphs, though, we have seen how physical contexts can interact with the meanings of poetry set into them. We do not know exactly where Propertius composed or first recited his poems, but once he published them their physical setting changed with each occasion of reading or recitation. Rome’s oral poetry culture made his words especially portable and susceptible all manner of changes of context. On the other hand, when an inscription reused Propertius’ words, it tied them to particular physical location. Depending on the surroundings, this could both limit and add to the meanings that the words communicated. A poet such as Propertius’ speaker who imagined that his poetry would be recycled in inscriptions may further have understood that architecture, images, and other ways in which physical spaces communicate would affect the way people sometimes experienced his words. Creation and performance of texts were important elements of the self-fashioning of elite Roman men as elite Roman men, both individually and as a group, in the Augustan period. Dupont argues that in recitationes elite Roman men could “accomplish the rare feat of reconstituting, under the Empire and by means of writing, a space for libertas.” 532 Gamel and Markus discuss how poetic performances could affect people’s perceptions of a man’s gender presentation. 533 When others performed or even recomposed a poet’s text, then, they could refashion or even erase part of his social identity. Nagy writes, “The key to loss of identity as a composer is loss of control 532 Dupont (1997) 59 533 Gamel (1998); Markus (2000), e.g. 140: “In fact, behind criticisms of the epic recital often lie issues about the performance of gender and social status. In that regard, epic’s position is parallel to that of rhetoric.” 165 over performance. Once the factor of performance slips out of the poet’s control—even if the performers of the poet’s poetry have traditional comments about the poet as a composer—the poet becomes a myth; more accurately the poet becomes part of a myth, and the myth-making structure appropriates his or her identity.” 534 Nagy is speaking about oral poetics, but written poetry can also slip from its original author’s control, as this chapter has shown. And, as we have seen, written and oral poetics can be difficult to tell apart in Augustan Rome. The evidence examined in this chapter, then, indicates that Propertius’ speaker was not paranoid: The loss of control he anticipates was not so much a risk as a near-certain hazard of composing poetry for publication in Augustan Rome. 534 Nagy (1990a) 80 166 Chapter 3 Appendix A Propertius’ Elegies and Epigraphic Poetry: Allusions, Formulas, and Motifs n.b. An equals sign (=) between the citation from Propertius and the CLE number indicates shared or similar wording and/or ideas, not identical wording. An equals sign (=) between the CLE number and the CIL number indicates that they are the same inscription. Words and Ideas from Propertius’ Elegies Repeated in Epigraphic Poetry: Prop. 1.1.5 = CLE 354.1 = CIL IV 1520.1 = CLE 1939.1 = CIL IV 9847.1 Prop. 1.1.5 = CLE 1944.1 = CIL XI 7476.1 (see Prop. 1.16.33; Prop. 1.22.1, 6; Prop. 2.15.23; Prop. 4.1.121) Prop 1.1.12 = CLE 1175.4 = CIL VI 29896.4 Prop. 1.2.24 = CLE 1123.4 = CIL II 1699.4 = CLE 2113.2 Prop. 1.3.39 = CIL IV 5296.1 = inscription from Ammaedara in Ben Abdallah et al. (2005) 103 Prop. 1.9.2 = CLE 2131.4 = CIL VIII 27916.4 Prop. 1.9.16 = CLE 1745.2 (no CIL given; Bulletin d. travaux hist. arch. 1894 p. 234 n. 10 descr. Gauckler) Prop. 1.10.1, 3 = CLE 963.1 = CIL VI 17130.1 Prop. 1.10.15 = CLE 1971.5-6 = CIL VIII 27380.5-6 Prop. 1.10.24 = CLE 952 = CIL IV 1118 (see Prop. 3.23.6) Prop. 1.11.23 = CLE 542.6 = CIL II 4427.6 Prop 1.11.29 = CLE 249.11 = CIL XIV 2852.11 Prop. 1.13.20 = CLE 939.2 Prop. 1.15.21 = CLE 389.3 Prop.1.16.13 = CLE 750.6 = CIL III 9610.6 167 Prop. 1.16.28 = CLE 2121.4 Prop. 1.16.33 = CLE 1944.4 = CIL XI 7476.4 (see Prop. 1.1.5; Prop. 1.22.1, 6; Prop. 2.15.23; Prop. 4.1.121) Prop. 1.18.32 = CLE 371.7 = CIL VI 10237.7 Prop. 1.20.24 = CLE 994.2 = CIL IV 1649.2 Prop. 1.21.1, 6 = CLE 1944.4 = CIL XI 7476.4 (see Prop. 1.1.5; Prop. 1.16.33; Prop. 2.15.23; Prop. 4.1.121) Prop. 1.22.8 = CLE 1021.2 = CIL XII 5271.2 = CLE 1252.6 Prop. 2.1.48 = CLE 1134.5 = CIL VI 26489.5 Prop. 2.1.57-8 = CLE 902.5-6 (see Prop. 2.18.12) Prop. 2.1.71ff = CLE 965.1ff = CLE 966 = CIL VIII 15716 Prop. 2.3.16 = CLE 1405.10 Prop. 2.3.23 = CLE 1021.4 = CIL XII 5271.4 Prop. 2.3.25 = CLE 1373.5 = CLE 2036.1 Prop. 2.5.9f = CLE 1785 = CIL IV 4491 Prop. 2.5.14 = CLE 1369.4 = CIL XII 338.4 = CLE 1744.4 = CIL XII 975.4 Prop. 2.6.35 = CLE 2107A.3 Prop. 2.7.3 = CLE 1971.5 = CIL VIII 27380.5 Prop. 2.8.23 = CLE 1136.2 = CIL VI 9693.2 Prop. 2.9.13 = CLE 1233.3 = CIL III 686.3 Prop. 2.12.18 = CLE 2052.1 = CIL IV 6626.1 Prop. 2.13.24 = CLE 531.1 = CIL VIII 15220.1 168 = CLE 1001.2 = CIL VI 6502.2 Prop. 2.13.47 = CLE 777.9 = CIL V 6817.9 Prop. 2.14.10 = CLE 2099.7 = CIL XIII 128.7 Prop. 2.15.9f = CLE 950.1f = CIL IV 5296.1-2 Prop. 2.15.19 = CLE 2107A.5 Prop. 2.15.23 = CLE 1944.3 = CIL XI 7476.3 (see Prop. 1.1.5; Prop. 1.16.33; Prop. 1.22.1, 6; Prop. 4.5.121) Prop. 2.15.41 = CLE 610.2 = CIL V 6693.2 Prop. 2.15.51f = CLE 1040.1ff = CIL VI 22377.1ff (see Prop. 2.18.29) Prop. 2.16.41 = CLE 1905.1 (cf. Verg. Aen. 11.444) Prop. 2.16.50 = CLE 1340.4 = CIL IX 952.4 Prop. 2.18.8 = CLE 1180.4 = CIL V 7404.4 Prop. 2.18.12 = CLE 902.2 (see Prop. 2.1.57-8) Prop. 2.18.14 = CLE 1025.2 = CIL XIV 3838.2 Prop. 2.18.29 = CLE 1040.3 = CIL VI 22377.3 (see Prop. 2.15.51f) Prop. 2.20.27 = CLE 652.4 Prop. 2.24.24 = CLE 1287.2 = CIL VIII 8123.2 Prop. 2.28.25 = CLE 965.3 (allusion and formula) Prop. 2.30.19 = CLE 2107A.3 Prop 3.1.16 = CLE 1141.16 = CIL III 2964.16 Prop. 3.1.31 = CLE 1988.11 169 Prop. 3.5.13-14 = CLE 434.11 Prop. 3.7.56 = CLE 2040.2 (see Prop. 4.6.64) Prop. 3.7.59 = CLE 587.5 = CIL III 8385.5 = CLE 1547.1 = CIL V 116.1 Prop. 3.10.10 = CLE 1109.26 = CIL VI 21521.6 (cf. Ov. Trist. 2.390) Prop. 3.11.8 = CLE 1287.2 = CIL VIII 8123.2 Prop. 3.11.63 = CLE 899.1 = CIL VI 1199b.1 Prop. 3.12.37 = CLE 2107A.8 Prop. 3.13.52 = CLE 1031.4 = CIL VI 6592.4 Prop. 3.14.28 = CLE 1095.5 = CIL V 3415.5 Prop. 3.16.13-14 = CLE 1785 = CIL IV 1950 Prop. 3.18.6 = CLE 513.7 = CIL XI 627.7 Prop. 3.18.30 = CLE 2103.8 Prop. 3.19.27 = CLE 2107° Prop. 3.20.7 = CLE 1109.33 = CIL VI 21521.33 (cf. Verg. Aen. 2.15) Prop. 3.20.9 = CLE 1536.1 = CIL III 2722.1, suppl. 9729.1 Prop. 3.20.12 = CLE 254.15 = CIL VIII 4635.15 Prop. 3.20.24 = CLE 378.1 = CIL X 5665.1 Prop. 3.21.34 = CLE 1155.4 = CIL VI 30110.4 Prop. 3.23.6 = CLE 952 = CIL IV 1118 add. p. 203 Prop. 4.1.121= CLE 1944.1 = CIL XI 7476.1 (see Prop. 1.1.5; Prop. 1.16.33; Prop. 1.22.1, 6; Prop. 2.15.23) Prop. 4.4.25 = CLE 467.6 = CIL VI 9118.6 170 Prop 4.4.94 = CLE 1975.1 (cf. Cat. Carm. 64.157) Prop. 4.5.47f = CLE 1785 = CIL IV 1894 Prop. 4.6.42 = CLE 2046.9 Prop. 4.6.64 = CLE 2040.2 (see Prop. 3.7.56) Prop. 4.7.1 = CLE 1269.3 = CIL VI 22513.3 = CLE 1329.5 = CIL VIII 403.5 = CLE 1497.3 = CIL VIII 1165.3 Prop. 4.7.94 = CLE 1136.2 = CIL VI 9693.2 = CLE 1571.2 = CIL VI 19008.2 Prop. 4.9.53 = CLE 443.3 = CIL VI 10969.3 Prop. 4.11.2 = CLE 2075.2 Prop. 4.11.17 = CLE 492.7f Prop. 4.11.46 = CLE 2080.5 Prop. 4.11.68 = CLE 1287 = CIL VIII 8123 Prop. 4.11.73 = CLE 682.1 Formulas and Motifs Occurring in Propertius’ Elegies and Epigraphic Poetry: Prop. 1.2.14 = CLE 468.3 = CIL VIII 7854.3 Prop. 1.2.24 = CLE 1123.4 = CIL II 1699.4 Prop. 1.2.31 = CLE 466.3 = CIL XII 218.3 Prop. 1.10.7 = CLE 943 Prop. 1.10.12 = CLE 1926 = CIL XIII 10018 Prop. 1.15.3 = CLE 569.1 = CIL VIII 8567.1 Prop. 1.15.22 = CLE 1123.4 = CIL II 1699.4 171 Prop. 1.16.35 = CLE 389.6 = CLE 443.7 = CIL VI 10969.7 = CLE 707.1 Prop. 1.17.22 = CLE 451.3 = CLE 488.4 = CIL XIV 3826.4 = CLE 492.1 = CIL III 754.1 = CLE 1064.4 = CIL VI 20466.4 Prop 1.17.24 = CLE 429.10 Prop. 1.19.11 = CLE 1233.21 = CIL III 686.21 Prop. 1.20.45 = CLE 1997.4 = CLE 2151.12 = CIL VIII 27764.12 Prop. 1.21.1 = CLE 501.7-8 Prop. 1.21.5 = CLE 369.4 = CIL V 2435.4 = CLE 369.2 = CIL V 2435.2 = CLE 446.3 = CIL XIII 8404.3 Prop. 2.1.17 = CLE 475.4 = CIL VI 3608.4 Prop. 2.4.2 (Fedeli, Prop. 2.3b.2; Barber, Prop. 2.3.46; Goold 2.3.46, listed under 2.4) = CLE 2103.8 Prop. 2.9.14 = CLE 1033.2 = CIL VI 6976.2 = CLE 1201.2 = CIL XII 879.2 Prop. 2.13.17 = CLE 452.2 = CLE 943.1 = CLE 1811 Prop. 2.13.35 = CLE 1944.4 = CIL XI 7476.4 Prop. 2.15.23 = CLE 2075.1 Prop. 2.23.16 = CLE 1952.1 = CLE 953.1 = CIL IV 1645.1 Prop. 2.24.4 = CLE 2051.2 = CIL IV 7065.2 Prop. 2.24.20 = CLE 2138.2 172 Prop. 2.26.57 = CLE 704.1 = CIL V 6723 Prop. 2.28.15 = CLE 514.3 = CIL VIII 16463.3 Prop. 2.28.22 = CLE 2036.3 = CIL VIII 26492.3 Prop. 2.28.25 = CLE 965.3 (allusion and formula) Prop. 2.28.42 = CLE 1439.9f Prop. 2.28.58 = CLE 1877.5 = CIL XI 4565.5 Prop. 2.32.34 = CLE 991 = CIL VI 2489 = CLE 992 = CIL III 2835 Prop. 3.1.37 = CLE 609.1 = CIL III 1871.1 = CLE 1222.3 = CIL VI 11407.3 = CLE 1279.1 = CIL II 4314.1 = CLE 1296.5 Prop. 3.10.11 = CLE 1996.2 Prop. 3.11.70 = CLE 1159.2 = CIL VI 29436.2 = CLE 2106.2 Prop. 3.13.17 = CLE 1988.13 Prop. 3.16.27 = CLE 2103.3 Prop. 3.17.5 = CLE 1971.5-6 = CIL VIII 27380.5-6 (cf. Prop. 1.10.15) Prop 3.18.9 = CLE 523.4 = CIL VIII 696.4 = CLE 1223.7 = CIL VI 25128.7 Prop. 3.18.22 = CLE 998.1f = CIL V 2411 Prop. 3.18.28 = CLE 970.14 = CIL VI 23551, X 6620 = CLE 971.15 = CIL VI 7872 [this inscription shares verses verbatim with CIL VI 23551] = CLE 1068.4 = CIL VI 5953 Prop. 3.18.32 = CLE 1049.2 = CIL VI 17342.2 = CLE 1050.2 = CIL VI 12307.2 173 Prop. 4.1.97 = CLE 1032.1 = CIL XIV 2553.1 Prop. 4.5.1 = CLE 1181.6 = CIL XI 911.6 Prop. 4.5.3 = CLE 197.2 Prop. 4.5.77 = CLE 945 = CLE 446al = CIL IV 1173 = CLE 1288.1 = CIL VIII 7427 = CLE 2056 = CLE 2063 Prop. 4.6.9 = CLE 443.2 Prop. 4.9.65 = CLE 422.8 = CIL VI 7578.8 = CLE 1005.10 Prop. 4.9.71 = CLE 1504.1, 12, 17, 22, 32, 52 = CIL XIV 3565 = CLE 1956.1 Prop. 4.11.1 = CLE 679.3 Prop. 4.11.19 = CLE 1109.23 = CIL VI 21521.23 Prop. 4.11.36 = CLE 1038.6 = CIL VI 14404.6 Prop. 4.11.63 = CLE 518.2al. Prop. 4.11.75 = CLE 710.7 = CIL XI 312 = CLE 2103.5 Prop. 4.11.82 = CLE 135.2 = CIL VIII 13265.2 Prop. 4.11.95 = CLE 995.25 = CIL VI 12652.25 Prop. 4.11.98 = CLE 213.4 = CIL X 1275.4 = CLE 422.13 = CIL VI 7578.13 174 Chapter 3 Appendix B Figure 1: CIL VI 6592 part a. Image PH0005515 in Corpus Inscriptionum Latinarum Online Database 175 Figure 2: CIL VI 6592 part b. Image PH0005527 in Corpus Inscriptionum Latinarum Online Database Figure 3: Columbarium of the Statilii: plan. Brizio (1876) pl. 1 in Borbonus (2014) 94 176 Figure 4: CIL IV 1520, line drawing from CIL IV Table XIV 177 Figures 5A and 5B: A) Plan of the Caupona of Euxinus from Jashemski (1967) 39. The arrow marks the findspot of CIL IV 9847-9 following Milnor (2014) 88. B) Photo of room d showing the doorway by which CIL IV 9847-9 were found, photo from Jashemski (1967) 44. A B Figure 6: CIL VI 5296, line drawing and text from CIL IV supp. 2. 178 Figure 7: CIL VI 29896, the epitaph for Pearl (Margarita) the dog. Image PH0005774 in Corpus Inscriptionum Latinarum online database. 179 Chapter 4: The Propertian Corpus Propertius’ Elegies rarely play directly on the double meaning of corpus as the body of a person and a book or an author’s body of work, and yet the idea pervades them. The double meaning of corpus has long standing in Latin literature before, contemporary with, and after Propertius. Cicero refers to works of literature as corpora and a corpus in his letters (Cic. ad Q. fr. 2.12[11].4; Fam. 5.12.4), as do Ovid (Trist. 2.535) and Vitruvius (2.1.8, 6 praef. 7, 9.8.15), and Livy writes of “the corpus of all Roman law” (corpus omnis Romani iuris, Liv. AUC 3.34.7). Ovid famously plays with the double meaning at the start of the Metamorphoses: “My mind brings me to speak forms changed into new corpora” (In nova fert animus mutatas dicere formas / corpora, Ov. Met. 1.1-2). The Greek and Latin literary traditions, furthermore, had long compared the texts of speeches and poems with human bodies and especially with the bodies of their authors by Propertius’ time. The speaker of Propertius’ Elegies explores several different relationships between his text and his body with a variety of poetic techniques. Physical Location The speaker of the Elegies uses the idea of physical presence in certain places to represent composing poetry. Sometimes he relies on associations from myth. In Poem 2.10, the speaker considers a change of genre, and dancing around Mt. Helicon stands in for composing poetry: “But it’s time to circle Helicon in other dances and time to give the field to the Haemonian 535 horse” (Sed tempus lustrare aliis Helicona choreis / et campum Haemonio iam dare tempus equo, Prop. 2.10.1-2). Similarly, the speaker describes his dream in Poem 3.3, saying, “I had seemed to 535 The Haemonian horse represents epic poetry, here. Haemonian means Thessalian, and Achilles’ famous war horses were Thessalian – in fact, Prop. 2.8.38 (“he dragged that brave Hector with Haemonian horses” fortem illum Haemoniis Hectora traxit equis) refers to Achilles’ horses with the same adjective as Prop. 2.10.2. Richardson (1977) 241 also adduces Verg. Geor. 2.541-2 and Lucan 6.397 as references to the fame of Thessalian war horses. 180 be reclining in the gentle shade of Helicon, where the spring of Bellerophon’s horse 536 flows” (Visus eram molli recubans Heliconis in umbra, / Bellerophonti qua fluit umor equi, Prop. 3.3.1- 2). Scholars have noted the associations of Helicon, Hippocrene, and Pegasus with epic poetry in general and in Propertius. 537 Being near Mt. Helicon, then, stands in specifically for composing epic poetry here, 538 and in verses 3-12 the speaker is on the verge of singing Ennian historical epic. In both Poems 2.10 and 3.3, though, the speaker is unable to follow through with the composing in genres other than elegy: “Just as, when one can’t touch the head of great statues, the crown is placed here down at their feet, thus we, unable to mount the peak of praise, give lowly incense in paupers’ rites. Not yet do my verses know the Ascraean 539 fonts, but only in the stream of Permessus has love washed them.” At caput in magnis ubi non est tangere signis, ponitur hac 540 imos ante corona pedes; sic nos nunc, inopes laudis conscendere culmen 541 , pauperibus sacris vilia tura damus. nondum etiam Ascraeos norunt mea carmina fontis, sed modo Permessi flumine lavit Amor. (Prop 2.10.21-6) and “ ‘What is there for you in such a stream, madman? Who has bidden you to touch the work of heroic song? 536 Bellerophon’s horse is Pegasus, and Hippocrene is the spring on Mt. Helicon said to have been created when Pegasus’ hoof struck the rock (Richardson 1977: 325; Pausanias 9.31). 537 Nethercut (1961) 392; Fedeli (1985) 110; Heyworth (2007) 291 538 See also Fedeli (1985) 116 on the adjective mollis as a signal of Callimacheanism that foreshadows the recusal from composing epic later in the poem. 539 The Ascraean fonts are Hippocrene and Aganippe (Richardson 1977: 244). 540 hac has long aroused critical suspicion. Shackleton Bailey (1956) 83-4, Richardson (1977) 243, Fedeli (1984) 72, Goold (1990) 132, Günther (1997) 2, and Heyworth (2007) 154 all doubt it and, with varying degrees of assuredness, offer other possibilities. Even some of the more aggressive critics, such as Heyworth, are diffident, so I agree with Fedeli’s assessment: sed nulla emendatio certa mihi videtur. 541 ω has carmen, and Shackleton Bailey (1956) 84 and Richardson (1977) 244 defend this reading. Goold (1990) 134 prefers currum. Fedeli (1984) 72 adopts culmen, and Heyworth (2007) 154 writes persuasively, “In verse 23 conscendere introduces the notion of ascent… the phrase conscendere carmen is unparalleled, laudis is sufficient to explain the metaphor, and with carmina occurring also in 25 there is a strong case for conjecture. Markland’s currum introduces an object Propertius has elsewhere with conscendere (II xviii 13), and an image that recurs in programmatic contexts… But it does not fit with inopes: it is not mounting a chariot that is difficult, but driving it. Unless there has been some more radical corruption, such as postulated by Jeverus’s lauto contendere cultu, one of the conjectures confirming the image of ascent is almost certainly right; and the easiest paleopgraphically is culmen, which I print. Support is regularly found in Hannibal’s address to his troops before the assault on the Alps at Sil. 3.509-10.” 181 You must not hope for any fame here, Propertius…’ ” ‘Quid tibi cum tali, demens, est flumine? quis te carminis heroi tangere iussit opus? non hic ulla tibi speranda est fama, Properti…’ (Prop. 3.3.15-17) In these final verses of Poem 2.10, the speaker admits that his poetry has no place at the springs of Hippocrene and Aganippe on Helicon but belongs instead at the humbler stream of Permessus. Poem 3.3 takes an even stronger tack: Apollo calls the speaker a “madman” (demens) for resting by Helicon and drinking from Hippocrene and sends him away. Conversely, being near or on Helicon represents composing elegy in Poems 2.13 and 3.5. Love has dictated where the speaker, as an elegiac poet, should live in Poem 2.13: “He has forbidden me to scorn such plain Muses and has ordered me to dwell thus in the Ascraean grove, not so that the Pierian oaks may follow my words, or so that I can lead along wild beasts in the Ismarian vale, but rather so that Cynthia might be dumbfounded by my verse.” hic me tam gracilis vetuit contemnere Musas, iussit et Ascraeum sic habitare nemus, non ut Pieriae quercus mea verba sequantur, aut possim Ismaria ducere valle feras, sed magis ut nostro stupefiat Cynthia versu (Prop. 2.13.3-7) Love has confined the speaker to the Ascraean grove, on Helicon. The speaker is not to go to Pieria or Ismarus to draw the trees and wild beasts to him with his song like Orpheus. Verse 7 reveals that Love has commanded the speaker to live in the Ascraean grove so that the speaker may compose a particular kind of poetry, namely poetry to astonish Cynthia. The speaker also says that he has lived on Helicon in Poem 3.5: “May I enjoy having dwelt on Helicon in my early youth” (me iuvet in prima coluisse Helicona iuventa, Prop. 3.5.19). Gardner notes that verses 23- 6 of Poem 3.5 associate maturity with poetry concerning philosophy and natural science. 542 Verses 542 Gardner (2013) 233-5 182 19-22 describe delights of youth and associate them with elegy both by contrast with verses 23-6 and with the motif of the garland. 543 The mythological locations in these passages are associated with poetry by metonymy. Ascra is the birthplace of Hesiod, Hesiod’s Muses are Heliconian at the start of the Theogony and Pierian at the start of the Works and Days, and Ismarus is in Thrace, the birthplace of Orpheus. The Callimachean fondness for learned references would have primed at least some of Propertius’ audience to appreciate these associations. But just in case, the phrase “more famous for my art than Inachian Linus” (Inachio notior arte Lino, Prop. 2.13.8) forces the issue of geographical metonymy. Inachus was the first king of Argos; Linus, in Callimachus’ telling (Aetia fr. 26-8), was the son of Apollo and Psamathe, an Argive princess, and in a yearly festival Argive women mourned the killing of Linus by the dogs of Psamathe’s father, Crotopus. Geography thus provides the simplest association between Linus and Inachus. 544 Thebes, however, also claimed Linus as a culture hero, music teacher to Hercules. 545 As a celebrated singer of songs, the Theban Linus fits Propertius’ comparison better than the Callimachean-Argive Linus, who is only a subject of songs. An interpreter is thus left to manage the disagreement between the stated geographical metonymy and the suggested association, and so the verse calls attention to metonymic geographical association as a poetic device. By invoking the well-known metonymic associations of Ascra, Helicon, Pieria, and Ismarus with famous poets, the Propertian speaker can use being in these places as metaphors for composing poetry. The speaker also uses being in certain generic places—i.e., certain types of places associated with elegy as a genre rather than specific places with mythological significance—as a 543 See section on “Garlands and the Speaker’s Body” below. 544 An audience can associate Linus with Inachus by genealogy alone, but this requires five steps: Linus is the son of Psamathe (1), daughter of Crotopus (2), son of Agenor (3), son of Phoroneus (4), son of Inachus (5). 545 Bremmer citing Alexis, fr. 140 K.-A.; Anaxandrines, fr. 16 K.-A.; and Achaeus TGF 20 F 26. 183 metaphor composing elegy. In Poem 3.3, Apollo, having sent the speaker down from Helicon and Hippocrene, instructs him that, instead of high peaks, “soft meadows should have tracks made in them by [his] little wheels” (mollia sunt parvis prata terenda rotis, Prop. 3.3.18). This turning of little wheels is the making of a “little book” (libellus, Prop. 3.3.19), which implies short, refined poems. 546 Similar associations between being in certain places and composing certain types of poetry begin the programmatic first poem of Book 3: “Callimachus’ ghost and holies of Coan Philitas, I beseech you, let me enter your grove! I enter as the first priest to proffer, from a pure wellspring, Italian mysteries through Greeks dances. Tell me, in what grotto did you likewise spin out your song? On which foot did you enter? Which water did you drink? Bye-bye to whoever detains Phoebus amid weapons! Let verse go forth finished with a fine-grained pumice” Callimachi manes et Coi sacra Philitae, in vestrum, quaeso, me sinite ire nemus! primus ego ingredior puro de fonte sacerdos Itala per Graios orgia ferre choros. Dicite, quo pariter carmen tenuastis in antro? quove pede ingressi? quamve bibistis aquam? a valeat, Phoebum quicumque moratur in armis! exactus tenui pumice versus eat (Prop. 3.1.1-8) In the first couplet, the speaker wishes to enter the grove of Callimachus and Philitas, both of whom often wrote in elegiac couplets and whose refined poetry the speaker admires. For the speaker, entering the grove means fulfilling his ambition “to proffer Italian mysteries through Greek dances” (Itala per Graios orgia ferre choros, Prop. 3.1.4). Next the speaker asks Callimachus and Philitas where they composed their poetry, and the next two questions—“On which foot did you enter?” (quove pede ingressi?) and “Which water did you drink?” (quamve bibistis aquam?)—emphasize bodily presence in the association between being in a particular place and composing poetry. 546 See below on the allusion to Poem 1 of Catullus. There is also an allusion to the wagon tracks in Callim. Aet. 1.25-8 (on this poem, see the section on “Body Terms Borrowed from Callimachus and Oratory” below). 184 Propertius’ speaker associates composing poetry with being near Cynthia when he considers leaving her. His elegiac life and love affair have made him sick and wretched in Poem 1.1, and he says to his friends, “Carry me among faraway peoples and through the waves, where no woman may know my route” (ferte per extremas gentes et ferte per undas, / qua non ulla meum femina norit iter, Prop. 1.1.29-30). It is physical proximity to Cynthia, here, that he seeks to escape in order to recover from his elegiac malady. The speaker only implicitly connects being near Cynthia with composing elegy, in that composing elegy is a sine qua non of his elegiac life and love affair. The couplet immediately preceding, though, puts the reader or listener in mind of words and poetry: “I will bravely suffer iron and savage fires; let there only be freedom to speak the things my anger wants” (fortiter et ferrum saevos patiemur et ignes, / sit modo libertas quae velit ira loqui, Prop. 1.1.27-8). Being near and with Cynthia has limited his speech to that of an elegiac lover, i.e. to elegy: To be near Cynthia is, for the speaker, to compose elegy. The speaker draws this connection explicitly in Poem 3.21, when he again fantasizes about leaving Cynthia. In the first couplet, he says that he is forced (cogor, Prop. 3.21.1) to travel to Athens “so that the long way may free [him] from oppressive love” (ut me longa gravi solvat amore via, Prop. 3.21.2). Several verses later he underscores his point: “There will be one remedy: as far from my eyes as Cynthia will be when I’m abroad, love will be that far from my mind (unum erit auxilium: mutatis Cynthia terris / quantum oculis, animo tam procul ibit amor, Prop. 3.21.9-10). In Athens he will write in other genres such as philosophy (Prop. 3.21.25-6), oratory (Prop. 3.21.27), or comedy (Prop. 3.21.28), or perhaps he will even switch to another art form entirely, such as painting, carving ivory, or sculpting in bronze (Prop. 3.21.29-30). Just as being in Athens means working in these other genres and art forms, being near Cynthia means composing elegy, and when the speaker actually leaves Cynthia in Poem 3.24 he renounces love elegy as a genre. Because Cynthia 185 does not exist outside of the speaker’s poetry, and so the poetry is the cause of Cynthia, while at the same time the poetry posits Cynthia as its own cause, the relationship of being near Cynthia to composing elegy cannot be entirely causal but must also be metaphorical. In these passages, it is not the places themselves but rather the speaker’s physical presence in the places that constitutes the metaphors for composing poetry, and so his body is part of the metaphors. The terms of the metaphors produce an association, albeit a loose one, between the speaker’s body and his poetry. Garlands and the Speaker’s Body By the Augustan Period, there was a longstanding association in Greek and Latin poetry between garlands and poetry itself. The association occurs, in part, by metonymy: Garlands and poetry were both common features of symposia. Garlands are also sometimes used as a metaphor for poetry, and weaving as a metaphor for poetry reaches back to the Iliad and the Odyssey and even farther back into the Indo-European tradition. 547 More directly influential on Propertius’ elegies was Meleager’s Garland, part of which Propertius adapts in the first four verses of Book 1. 548 Propertius’ elegies use the metonymy and the metaphor of garlands for poetry, and some passages also associate garlands with the speaker’s body. Some of the Propertian speaker’s descriptions link garlands to his body through close physical proximity and entwining or even intermingling. As he imagines his death and burial in one poem, the speaker hopes that his mistress “will bring perfumes for [him] and will decorate [his] grave with garlands, sitting as a guardian by [his] mound” (afferet haec unguenta mihi sertisque sepulcrum / ornabit custos ad mea busta sedens, Prop. 3.16.23-4). The speaker wants 547 Schmitt (1967) 298-301 548 See section on “Canonical Poets” in Chapter 3. 186 garlands placed on his grave, which is both a sign and a receptacle for the remains of his body, and which an inscription bearing some his words may mark. In another poem, he revels in his life of love, parties, poetry, and nequitia: “let it please me to languish, lying amid yesterday’s garlands” (me iuvet hesternis positum languere corollis, Prop. 2.34.59). The speaker envisions his body entwined with garlands, which, being yesterday’s, would be wilting as he languishes. There is physical proximity, which is the basis of metonymy, between the garlands and the speaker’s body, while the word “languish” (languere) also implies an analogy between them in that both the garlands and the speaker’s body are the worse for the wear the morning after the party. He celebrates this lifestyle in another poem as well, saying, “It delights me both to bind my mind with lots of wine and always to have my head in spring roses” (me iuvat 549 et multo mentem vincire Lyaeo / et caput in verna semper habere rosa, Prop. 3.5.21-2). In verse 21, the speaker’s mind is bound with wine. This is a physical metaphor for an effect on something non-physical, the mind (mentem), and this effect is brought about by the bodily effects of wine, so the body and the mind blur together. The verb “to bind” (vincire) compares the wine to a garland, and the discussion of 549 There has been much debate over whether verses 19 and 21 of Prop. 3.5 should have iuvat or iuvet. Verses 19-22 read, me iuvet in prima coluisse Helicona iuventa / Musarumque choris implicuisse manus: / me iuvat et multo mentem vincire Lyaeo / et caput in verna semper habere rosa. Some (e.g., Fedeli 1984: 154) who prefer iuvet argue that the verses allude to Prop. 2.13.11 (me iuvet in gremio doctae legisse puellae), while some (e.g., Heyworth 2007: 301 citing Conte 2000: 307-10) in favor of iuvat suggest an allusion to Lucr. DRN 4.1ff (Avia Pieridum peragro loca nullius ante / trita solo. iuvat integros accedere fontis / atque haurire, iuvatque novos decerpere flores / insignemque meo capiti petere inde coronam, / unde prius nulli velarint tempora musae…). Both sides present strong arguments, and neither convinces me that the other is less plausible. I also think that we may still read the allusion to Prop. 2.13.11 with iuvat or the allusion to Lucr. DRN 4.1ff with iuvet. I disagree, however, with Heyworth’s (2007) 301 statement that “there is no reason for differentiation here; each line should have the same mood,” because the tenses of the infinitives differentiate the two couplets and may provide a clue as to the mood(s) of iuvat/iuvet in verses 19 and 21. The rest of the poem indicates that the speaker is still an elegiac lover (see especially verses 1-6) and so still in in his prima iuventa (see Gardner 2013: 233-5 on the youth of the lover). The perfect infinitives coluisse (19) and implicuisse (20) indicate actions that have come to a close in the present; the youthful elegiac lover is not yet finished doing these things, but rather he is imagining himself being finished and looking back on his actions, and the subjunctive iuvet would express this hope for future enjoyment of reminiscing about his present actions. The infinitives vincire and habere, on the other hand, are present and so express what currently delights the speaker, suggesting the indicative iuvat. If this is correct, verses 21-2 do not express a second wish but rather elaborate and explain the speaker’s wish in verses 19-20. The sequence of thought, then, would be this: “I hope that in the future I look back fondly on my youth spent as an elegiac lover-poet; what I like now is getting drunk on wine and wearing garlands of roses.” 187 actual garlands in the next verse reinforces the association. In verse 22, the speaker’s head (caput) is grammatically parallel to his mind (mentem) in verse 21; this parallelism suggests the analogy that the speaker’s head is to the garland of roses as the speaker’s mind is to wine. Through this comparison, the garland is incorporated into the speaker’s body and mind like the wine. Other passages draw likenesses between garlands and the speaker’s body through simile and analogy. In Poem 2.15, the speaker has been celebrating a night of lovemaking, and he hopes that the puella will grant him many more such nights. He thus concludes with a carpe diem exhortation: “Only, while there is light, don’t you give up the enjoyment of life! If you will give me all your kisses, you will still give too few. But just as the petals have forsaken the withering garlands, and now you see them strewn all over, floating in the wine cups, so for us, who now breathe deep as lovers, perhaps tomorrow’s day will close in our fates.” tu modo, dum lucet, fructum ne desere vitae! omnia si dederis oscula, pauca dabis. ac veluti folia arentis liquere corollas, quae passim calathis strata natare vides, sic nobis, qui nunc magnum spiramus amantes, forsitan includet crastina fata dies. (Prop. 2.15.49-54) The comparison between the garlands and the speaker and his girlfriend’s bodies is simple and direct: Just as the garlands have withered, so will they. Poem 1.3 presents a subtler comparison of a garland to the speaker’s body: “Her, not yet having mislaid all my senses, I gently try to approach, making the bed dip; and though they bade me, while I was seized with two-fold passion, here Amor and there Liber, each a harsh god, to try to slip my arm under her as she lay there and advancing my hand to take kisses and arms, even so I dared not disturb my lady’s rest fearing strife with harshness I’d experienced before; thus, rather, I stayed fast, transfixed with eyes intent, like Argus watching Io’s unfamiliar horns. And I was just untying the garland from my forehead and placing it, Cynthia, around your temples.” hanc ego, nondum etiam sensus deperditus omnis, molliter impresso conor adire toro; 188 et quamvis duplici correptum ardore iuberent hac Amor hac Liber, durus uterque deus, subiecto leviter positam temptare lacerto osculaque admota sumere et arma manu, non tamen ausus eram dominae turbare quietem, expertae metuens iurgia saevitiae; sed sic intentis harebam fixus ocellis, Argus ut ignotis cornibus Inachidos. et modo solvebam nostra de fronte corollas ponebamque tuis, Cynthia, temporibus (Prop. 1.3.11-22) In verses 11-12, the speaker attempted to get into Cynthia’s bed without waking her up but could not manage it, as the verb “try” (conor) suggests and verses 19-20 confirm. Amor and Liber tempted him to try to wrap his arms around Cynthia, but “even so” (tamen) he did not dare (non ausus eram). Instead, he took the garland off his own head and put it around hers. The garland did what he wished his arms could do, and so it took the place of his body. Membra: Text and Body Parts In some passages of the elegies, the speaker associates his text or parts of his text with parts of his body through comparisons and physical relationships. In Poem 2.7, he rejoices at the repeal of a law that might have forced him to marry and so to abandon his love affair with Cynthia and his life as an elegiac lover-poet. The potential conflict between his love and the law allayed, he explains his relief saying, “For I would sooner suffer my head to part from my body than I could put out the torch of love at a bride’s whim” (nam citius paterer caput hoc discedere collo, / quam possem nuptae perdere more faces, Prop. 2.7.7-8). At the beginning of Book 2, the speaker conspicuously calls his poetry Amores, so in this couplet he compares the loss not only of his love affair but also of elegy itself with decapitation. The speaker also employs a number of puns and double meanings that link his body and his text. Most of these are puns on feet as both body parts and units of meter. He describes the plight of an elegiac lover-poet who has fallen out of his mistress’s favor in Poem 2.4: 189 “You must first lament many wrongs by your mistress, often ask for something, often leave rejected, and often wreck your undeserving nails with your teeth, and anger must stir you to drum your wavering foot! In vain I poured fragrant oil on my hair, and my slowed foot moved with a measured step.” Multa prius dominae delicta queraris oportet, saepe roges aliquid, saepe repulsus eas, et saepe immeritos corrumpas dentibus unguis et crepitum dubio suscitet ira pede! Nequiquam perfusa meis unguenta capillis, ibat et expenso planta morata gradu. (Prop. 2.4.1-6) The speaker may as well be telling his own story in verses 1-4, and indeed he slips into the first person and his own story in verse 5. In verse 4, he prescribes that an elegiac lover-poet, mistreated by his mistress, should express his anger with the rhythm of his foot. He should also transfer his confusion (dubio) at his mistreatment to his foot. This refers both to the bodily sensations that can accompany feelings of frustration and bewilderment and to the trope of the ‘deficiency’ of the elegiac couplet, in which the pentameter is ‘missing a foot’ in comparison with hexametric poetry and does not even form five feet properly but rather is two hemiepes separated by a diaeresis. Next the speaker expresses regret for the care he has wasted with his hair and with the way he walks. The words for the sole of his foot (planta) and his step (gradu) in verse 6 pick up on the more common metrical pun, pes, in verse 4, so his hesitating foot and his measured step also describe the carefully wrought poetry he composes to try to win over his mistress. He overlays a chiasmus of expenso and gradu around planta and morata with a synchisis of participles (expenso and morata) and nouns (planta and gradu): This demonstration of his verbal art involves the words for “foot” and “step” so that the form of the verse reinforces the association between his text and his body. Sometimes the speaker refers to feet less directly. At the start of Poem 2.10, he talks about dancing around Helicon: “But it’s time to circle Helicon in other dances and time to give the field 190 to the Haemonian 550 horse” (Sed tempus lustrare aliis Helicona choreis / et campum Haemonio iam dare tempus equo, Prop. 2.10.1-2). As discussed above, dancing around Helicon means composing poetry, and doing so “in other dances” (aliis…choreis) means composing in meters other than his usual elegiac couplets, in this case hexameters, as verse 2 indicates. Since dancing involves the rhythmic movement of feet, and hexameters are a rhythmic arrangement of feet, verse 1 implies the pun on feet through the metonymy of dance for feet and the metaphor and metonymy of dance for poetic composition. Similarly, the speaker promises in a prayer to Bacchus, “I shall call to mind these things to be recounted with a buskin not unassuming” (haec ego non humili referam memoranda cothurno, Prop. 3.17.39). The buskin (cothurno) is the high shoe worn by a tragic actor and is associated with poetry in an elevated registers. A shoe also brings to mind a foot, and so the buskin engenders in verse 39 the double meaning of foot as body part and foot as unit of meter. In Poem 2.34, the speaker uses the same word when he advises Lynceus, “also stop composing your words with an Aeschylean boot – stop, and relax your limbs (clauses) into soft dances” (desine et Aeschyleo componere verba coturno, / desine, et ad molles membra resolve choros, Prop. 2.34.41-2). The ablative or dative Aeschyleo...coturno may be read “with verba... ‘verse shod with the buskin of Aeschylus,’ as a descriptive ablative modifying the subject of desine, or as a Propertian dative with componere.” 551 The simultaneous possibilities of Lynceus wearing the coturnus and his words wearing it equate Lynceus’ body with his text. Another meaning of the verb compono, which is “to fit words and music, or words and rhythm together,” 552 furthers the play on words here. 550 See Footnote 482 above. 551 Richardson (1976) 314 552 Habinek (2005) 15, 264: “See Ov. Ars Am. 3.345, where compono seems to mean “apply a melody” to a preexistent text; on the fitting together of voice and rhythm, see Marius Victorinus, GLK 6.184.” 191 The speaker associates his text with limbs and bones, as well as feet. In verse 42 of Poem 2.34 above, the speaker advises his friend Lynceus to ease and loosen his membra. The word membra means both “limbs” and “clauses,” 553 both of which may relax to form “soft dances” (molles choros), and so the speaker equates the gentle play and rhythm of the body with those of the text of elegy, which he encourages Lynceus to compose. In Poem 2.24, the speaker complains that the puella is unfaithful, and he fantasizes about his death and her ministrations to his burial: “You will arrange me and say, ‘these are your bones, Propertius: alas, you were steadfast to me” (tu me compones et dices ‘ossa, Properti, / haec tua sunt: eheu tu mihi certus eras, Prop. 2.24.35- 6). In a funeral context, the verb compono is occasionally used of laying out the corpse on a bed 554 but more frequently refers to gathering the ashes and shards of bone after cremation. 555 As noted above, compono can also be used of poetic composition. With the phrase “you will arrange me” (tu me compones), then, the speaker of Propertius 2.24 imagines both that the puella will lay out his body and/or collect his remains for burial after cremation and that she will speak the words he provides to sum him(self) up. The text of this couplet, however, is disputed. 556 Whether the first word of verse 35 is tu or tum makes no difference to my analysis. If the second word of verse 35 is mea instead of me, 553 See Keith (1999) 57 for discussion. 554 TLL 3.0.2116.36-7 lists Ov. Met. 9.504 as the only Augustan example. 555 TLL 3.0.2116.39-50. For cremation as the normal funereal practice at Rome in the 1 st century CE, see Morris (1992) 42-3 and Noy (2005). See Habinek (2016) on cremation and cremated remains in Latin literature. 556 Fedeli (1984) 105 prints tu me[a] and argues, “ego non equidem ignoro vocem ossa e sequentibus suppleri posse; quod tamen durius mihi videtur, cum de Cynthiae verbis agatur.” ω reads tu mea instead of tu me in verse 35. Shackleton Bailey (1956) 113 prefers this reading, but apparently with reservations, since he writes, “For the agreement of mea with ossa in direct speech, 2. 26. 25 is no parallel; see ad loc. Stat. Theb. 9. 900 ‘hunc tamen, orba parens, crinem’, dextraque secandum / praebuit, ‘hunc toto capies pro corpore crinem’ (probably from Virg. Aen. 4. 702ff.) is the nearest I can find.” Richardson (1977) 78 and Heyworth (2007) 215 follow him, and the latter adds, “there is considerable awkwardness in the emphatic positioning of ossa if it is not doing double duty.” ς reads tum instead of tu, and Porson emended mea to me, producing tum me, which Goold (1990) 172 prints. I find both Fedeli and Heyworth’s arguments compelling. I do not, however, see why Shackleton Bailey (1956) 113 writes, “A question-mark after sunt seems preferable to the usual colon,” and why Richardson (1977) 78, Fedeli (1984) 105, and Heyworth (2007) 215 follow him in this. I prefer the colon, following Goold (1990) 172. Günther (1997) says nothing on this verse. 192 though, then my analysis changes in the details but reaches essentially the same conclusion: Before the reader or listener reaches the word “bones” (ossa), the phrase “you will arrange my…” (tu mea compones) can be understood as “you will arrange my affairs,” and the verb “you will say” (dices) suggests that compono means “to fit words and music, or words and rhythm together.” 557 A reader or listener might even momentarily infer “words” (verba) from tu mea compones and dices. Then the reader or listener comes to the word “bones” (ossa) and the image of the puella speaking as she organizes his remains. In this reading, the idea of poetic composition is suggested and then displaced by the image of laying out the speaker’s body or gathering his ashes, so the body takes the place of the text. Furthermore, without the aid of quotation marks a reader or listener may at first understand the phrase dices ossa Properti as “you will speak the bones of Propertius.” The verb dico refers to an authoritative utterance sometimes associated with song. 558 To speak the bones of Propertius authoritatively and perhaps as a song, then, would mean to speak parts of his poetry. Only after the turn of the verse do the words that follow reveal that this phrase is instead part of the hypothetical direct speech of the puella. Whichever reading one prefers, this couplet links the speaker’s body and his words, even when someone else speaks them, through two meanings of the verb compono. 559 The link between the speaker’s bones and his text reappears later in the poem, when the speaker maintains that the puella should ignore her wealthy would-be lovers, because “hardly even one comes to gather your bones when you die” (vix venit, extremo qui legat ossa die, Prop. 2.24.50). The Latin word for “to gather,” lego, has “to read” as a secondary meaning, and verses 557 Habinek (2005) 15, 264: “See Ov. Ars Am. 3.345, where compono seems to mean “apply a melody” to a preexistent text; on the fitting together of voice and rhythm, see Marius Victorinus, GLK 6.184.” 558 Habinek (2005) 63, esp. “any carmen can be construed as a dictum, although not every dictum need be a carmen.” 559 See above on compono. 193 35-6 have already established a conceptual link between bodies—and especially bones—and texts in this poem. While the phrase qui legat ossa means “to gather your bones” in verse 50, then, the placement of the words legat and ossa next to each other evokes this conceptual link. The idea of reading the bones of the scripta puella, as she is called at Prop. 2.10.8, is not entirely nonsensical, either: What could her bones be but the poetry itself? And just as hardly any of the rich would- be lovers will come to gather the bones of the puella, hardly any of them can truly read and appreciate the poetry that is the scripta puella. The next couplet corroborates this reading: “I will be this for you: but I pray that instead you mourn me, bare-breasted and with your hair let down” (hic tibi nos erimus: sed tu potius precor ut me / demissis plangas pectora nuda comis, Prop. 2.24.51-2). The speaker offers here to be the person who will gather and read (lego) the bones of the puella but says that he would prefer to be the one who dies first and is mourned, reversing the roles he has just been talking about. In this revised wish, it will be the puella who gathers and reads the speaker’s bones. The fantasy of verses 35-6 thus returns in verses 50-2, putting the two words for gathering and arranging the speaker’s body and bones, compones (35) and legat (50), in dialogue with each other. The juxtaposition of these two verbs strengthens the double meaning of each of them and so the association of the speaker’s body with his text. One of the body parts that the speaker links to his text most frequently is his hands. In verse 19 of Poem 3.5, he hopes that he will enjoy having lived on Helicon in his youth, as discussed above, 560 and in verse 20 he adds the second part of the wish: “...and [may it please me] to have entwined my hands in the dances of the Muses” (Musarumque choris implicuisse manus). Like living on Helicon in verse 19, entwining his hands in dances in verse 20 is a metaphor for composing elegy. In this metaphor, dances stand in for elegy, and the speaker’s hands are 560 See the section on “Physical Location.” 194 intimately involved, as it is they that are doing the dancing. Furthermore, the idea of entwining anticipates the image of the garland, also associated with both poetry and the speaker’s body, 561 that follows in the next couplet (Prop. 3.5.21-2 562 ). There is a similar relationship of close involvement in Prop. 3.23.3, where frequent use and contact, a relationship of physical proximity such as metonymy expresses, has left the writing tablets on which the speaker used to compose his elegies (before he lost the tablets) marked by wear from his hands (has quondam nostris manibus detriverat usus). At the end of Book 3, when the speaker foreswears the elegiac life, the images and words in one of his complaints suggest that the elegiac life brought his hands into close contact with garlands and even versified his hands: “Seized, I was being scorched in Venus’ fierce cauldron; I had been bound with my hands twisted onto my back. But look, the crowned keel has reached the port, having been led across the sandbars, and my anchor has been cast. Now, worn out, I recover at length from the immense roiling, and my wounds now move together toward health.” correptus saevo Veneris torrebar aeno; vinctus eram versas in mea terga manus. ecce coronatae portum tetigere carinae, traiectae Syrtes, ancora iacta mihi est. nunc demum vasto fessi resipiscimus aestu, vulneraque ad sanum nunc coiere mea. (Prop. 3.24.13-18) In verse 14, the speaker says that his hands had been twisted and bound behind his back. Venus, standing in metonymically for his love affair and so also his love poetry, seems to have done this to him. In the sound and etymology of the word “twisted” (versas), then, can be heard the word for verse (versus). The word “bound” (vinctus) supports this figura etymologica, as Propertius 561 See the section on “Garlands and the Speaker’s Body” above. 562 See above. 195 sometimes uses it with garlands, 563 and the word coronatae 564 (15) in the following verse reinforces the association. Through this wordplay the speaker says that his hands were twisted and versified and then bound, perhaps even with a garland, when he was an elegiac lover. The speaker’s hands, then, are in close physical association with garlands and are “turned” into verse. Corpus: Text and Body In addition to associating his text with parts of his body through various images and devices, the speaker of the Elegies sometimes associates his text with his body more generally or even with his body as a whole. Just as he deploys the double meaning of “foot” (pes) in several poems, in Poem 4.2 the speaker exploits the double meaning of corpus for metaliterary commentary. The poem begins, “Why do you marvel that I have so many forms in one corpus?” (Quid mirare meas tot in uno corpore formas? Prop. 4.2.1). On the one hand, this is “an inscriptional start,” 565 which might suggest the text on a statue or a tombstone. On the other hand, this verse follows a programmatic first poem that expresses considerable anxiety about genre, and indeed Book 4 includes love elegy, aetiological poetry, epistolary poetry, martial epic, invective, and funeral laudation – many forms in one corpus. O’Rourke compares this verse with the beginning of Book 2, “You ask from what source my loves are written so often” (Quaeritis unde mihi totiens scribantur amores, Prop. 2.1.1 566 ), to argue that “on first reading, 4.2.1 might likewise 563 e.g., Prop. 3.5.21-2 (see section on Garlands above). 564 Richardson (1977) 411 notes “the custom of garlanding a ship on safe arrival in port” and cites Verg. Geo. 1.303- 4. In the Propertian speaker’s idiom, the word “crowned” (coronatae, 15) has an ambiguous relationship with garlands: In the programmatic first poem of the book, he predicts that refined elegiac poetry will lead his Muse to ride in triumph coronatis...equis (Prop. 3.1.10). Richardson (1977) 319 writes, “The horses are garlanded especially because of the association of garlands with the Muses,” and Goold (1990) 221 translates coronatis as “garlanded.” Nine verses later, though, the speaker opposes garlands (serta), which he associates with elegy, to a crown (corona), which he associates with epic: “Give soft garlands, Muses, to your poet: / a hard crown will not do for my head” (mollia, Pegasides, date vestro serta poetae: / non faciet capiti dura corona meo, Prop. 3.1.19-20). 565 Hutchinson (2006) 89 566 Translation mine; underlining in Latin O’Rourke’s. 196 construct a reader who has posed a question (quid mirare?) pertaining to the plenitude of the plenitude of the poet’s output (meas tot).” 567 Verse 2 of Poem 4.2 supports the double meaning in the first verse: “Receive the ancestral signs of the god Vertumnus” (accipe Vertumni signa paterna dei, Prop. 4.2.2). A reader or listener can take signa as “statue” 568 —i.e., the simulacrum of Vertumnus’ body, possibly to be worshipped as his actual body—but also as “tokens,” 569 and even a meaning close to our modern semiotic sense of “signs” was available. 570 As O’Rourke notes, an audience could easily take the poem to be in the Propertian speaker’s own voice until verse 2, which reveals that the speaker is Vertumnus. 571 Vertumnus, though, is ventriloquized, and the potential misdirection in verse 1 reminds the audience that the Propertian speaker is behind Vertumnus the whole while. The word corpus in verse 1 can thus be taken as both Vertumnus’ and the Propertian speaker’s. The speaker also associates his text with his whole body in less direct ways. In the verses from Poem 2.13 discussed above, the speaker’s physical presence on Helicon represents composing elegy, and in verse 5 the speaker’s text takes the place of his body. Verse 4 concerns the speaker’s physical location. In verse 5—“not so that the Pierian oaks may follow my words” (non ut Pieriae quercus mea verba sequantur)—the focus shifts from the speaker himself to his words, which the Pierian oaks might hypothetically follow as they followed Orpheus’ song. The speaker and Apollo, whose command he is following, treat the speaker’s words as a potential index for his physical person: Whether the trees might try to follow his words to him or might just trail 567 O’Rourke (2014) 163, and see the pages following in O’Rourke’s essay for fuller discussion of metatextuality, intertextuality, and paratextuality in Prop. 4.2. 568 Hutchinson (2006) 89; O’Rourke (2014) 163; and Richardson (1977) 424: “the meaning ‘the statue types in which I am worshipped as pater’ may be implicit.” Against this reading see Heyworth (2007) 435. 569 Goold (1990) 321 570 “A sign is that which falls under some sense and means something, which seems to have come forth from it itself” (Signum est, quod sub sensum aliquem cadit et quiddam significat, quod ex ipso profectum videtur, Cic. Inv. 1.30.47). 571 O’Rourke (2014) 164 197 along after him, following his words as he sings, his words being audible would indicate his physical proximity. Verse 6—“or so that I can lead along wild beasts in the Ismarian vale” (aut possim Ismaria ducere valle feras)—is grammatically parallel to verse 5, and it presents the speaker himself leading wild beasts along as his words might lead the oaks. The parallelism of the grammar and ideas of verses 5 and 6 makes the speaker’s text and body interchangeable. His body could replace his words in verse 5, and his words could replace his body in verse 6 without harming the sense of either verse, the couplet, or the poem. Whereas Poem 2.13 equates the speaker’s poetry as sound with his body, in Poem 3.23 the speaker’s writing tablets, a visual manifestation of his poetry, take the place of his body. He has lost the tablets and, as discussed above, 572 describes how close and frequent contact with his hands had left the tablets worn and marked. He then reflects on how “by now they knew how to please girls without me and without me to say some choice words” (illae iam sine me norant placare puellas / et quaedam sine me verba diserta loqui. Prop. 3.23.5-6). The tablets, then, knew how to do what the speaker, as an elegiac lover, fancies himself able to do—they could charm girls with eloquent and refined words—and what he perhaps wishes he could manage: “they always earned good results” (semper et effectus promeruere bonos, Prop. 3.23.10). In this conceit, the letters spoke for the speaker when he sent messages on them. They took his place, being physically present with the girls when he was not there and speaking to them. Though of course it was the speaker’s words that the tablets relayed to the girls, the conceit that the tablets were learned (doctae, Prop. 3.23.1) and spoke of their own accord reinforces the idea that they stood in for the speaker. In the tablets, furthermore, the speaker saw an idealized, always-successful version of himself. The tablets did not always take the speaker’s place, though. Verses 11-18 recall that 572 See the section “Membra: Text and Body Parts” above. 198 sometimes the speaker’s girlfriends wrote messages back to him on the tablets. It was not the tablets themselves, the medium on which he wrote, then, that took the speaker’s place; it was the tablets as written on by the speaker, a version of a portion of his book of poetry. The speaker’s text also replaces his body in a mythological exemplum. In Poem 2.18, he plays with the story of Aurora and Tithonus to reproach the puella for spurning him: “But Aurora did not scorn Tithonus’ old age and let him lie abandoned in the House of Dawn ... It did not shame such a girl to sleep with an old man, and so often to kiss his white hair. But you hate me though I’m a young man, faithless girl, though you yourself will be a hunchbacked crone someday not far off.” at non Tithoni spernens Aurora senectam desertum Eoa passa iacere domo est ... cum sene non puduit talem dormire puellam et canae totiens oscula ferre comae. at tu etiam iuvenem odisti me, perfida, cum sis ipsa anus haud longa curva futura die. (Prop. 2.18.7-8, 17-20) As told in the Homeric Hymn to Aphrodite, the goddess Aurora (Eos in Greek) fell in love with the mortal man Tithonus. When she convinced Zeus to make him immortal she neglected to ask for eternal youth for him as well, and so he grew old. Though she continued to cherish him, she no longer slept with him. Aphrodite uses the story of Aurora and Tithonus to convince Anchises not to strive for immortality. 573 The Propertian speaker adapts the story to his own rhetorical goals: His version of Aurora never abandons Tithonus’ bed. Gardner “read[s] Tithonus and Aurora as foils for the amator and puella within the larger corpus of Latin love elegy,” and within this analytic framework she writes, “The amator, while expressing anxieties about his future aging, challenges mortality through the permanence of his art: his alignment with Tithonus—who was by 573 Gardner (2013) 4 summarizes this part of the Hymn to Aphrodite to introduce her analysis of Prop. 2.18a.17-20. 199 some accounts released from his suffering through transformation into a tuneful cicada— ultimately confirms the power of song to outlive the physical form.” 574 In this reading, the speaker hopes to achieve immortality through his poetry, which will replace his body. Poem 2.24 similarly compares speaker’s body with his poetry book, specifically the Monobiblos. In the first couplet, we find out that the Monobiblos, here called Cynthia after its first word, has become a best-seller and has made the speaker an instant celebrity: “Are you speaking, now that you’re the talk of the town for your famous book, and your Cynthia is read all over the forum?” (‘tu loqueris, cum sis iam noto fabula libro / et tua sit toto Cynthia lecta foro?’ Prop. 2.24.1-2). A few verses later, he complains that, if Cynthia were nicer to him, he “would not be dragged thus in infamy all over the city” (nec sic per totam infamis traducerer urbem, Prop. 2.24.7). Just as his book has been spread around publicly, so he himself has been paraded around as a public spectacle. The similar structures of the verse describing his book’s publication and the verse describing his public humiliation strengthen the comparison. Both verses begin with coordinating conjunctions: et (2) and nec (7). The word tua in verse 2 has no echo in verse 7, and sit (2) and sic (7) mean different things, but they look and sound similar. Following these are the words (toto) and (per totam), both in cases indicating location. Next come nominatives – Cynthia, which is the subject of verse 2, and infamis, best read as a predicate nominative modifying the subject expressed in the ending of the verb in verse 7. After the subjects come their passive actions, “read” (lecta) in verse 2 and “I would be dragged” (traducerer) in verse 7. Finally, the words foro (2) and urbem (7), modified by toto and totam respectively, give the actions their settings, which are almost interchangeable, the forum being one of the primary public spaces in the city. The 574 Gardner (2013) 5 200 speaker’s body and the Monobiblos thus suffer similar treatments in verses closely patterned after each other. Sometimes the speaker alleges that abuse from Cynthia turns his body into a legible text. In Poem 3.11, he asks an interlocutor, perhaps the reader or audience, “Why are you amazed that a woman turns my life and drags me, a man, bound under her rule, why do you devise shameful charges of idleness against me, because I cannot burst my bonds and shiver my yoke?” Quid mirare meam si versat femina vitam et trahit addictum sub sua iura virum, criminaque ignavi capitis mihi turpia fingis, quod nequeam fracto rumpere vincla iugo? (Prop. 3.11.1-4) The speaker says here that a woman controls his life, and the verb “turns” (versat, 1) describes this control with a physical metaphor of directing or steering. In verses 2 and 4, more specific metaphors describe her control over his life as control over his body: She “drags” (trahit, 2) him, and she hold him in “bonds” (vincla, 4) and a “yoke” (iugo, 4). The sound and etymology of the verb versat in line 1 also bring to mind the word for verse (versus), as with the word versas in Poem 3.24 above. 575 The verb verso, furthermore, can mean “turn over in one’s mind” and “ponder” as well as “turn” in a physical sense. Horace even uses it of reading: “Turn the Greek exemplars with your night-time hand, turn them with your day-time hand (vos exemplaria Graeca / nocturna versate manu versate diurna (Hor. A.P. 268-9). In the word versate, Horace may be playing with the meaning “ponder,” but his emphasis on the hand indicates physical turning, perhaps of the scroll. To read, here, is both to turn over a text in one’s mind and to physically turn the text to move through the columns of the scroll. The elegiac lover’s life is both poetry and 575 See section on “Membra: Text and Body Parts.” The word vincla in verse 4 of Poem 3.11 shares its root with the word vinctus in Poem 3.24. As discussed above, Propertius sometimes uses the verb vincio of garlands, so vinctus supports the play between versas and versus in the same verse. In Propertian elegy it is perhaps poetic and perhaps a stretch to connect vincla to garlands. 201 enslavement to love and to his mistress, and so, just as Venus versified the speaker in Poem 3.24, the controlling woman versifies him in Poem 3.11. Frederick (1997) explores the use of bodies as texts when elegiac speakers are violent toward their girlfriends and boyfriends; a similar treatment of bodies as texts occurs in the inverse scenario in Propertius. In Poem 3.8, the Propertian speaker encourages his girlfriend to get rough with him again: “I enjoyed our brawl by the lanterns last night, and so many abuses from your crazy voice. I mean it – dare now to attack my hair and mark my face with your gorgeous nails, threaten to burn out my eyes with a flame brought to bear, rip back my tunic to bare my chest! When, maddened with wine, you overturn the table and throw your full goblet at me with a crazy hand, without a doubt signs of true ardor are given to me: for without serious love no woman hurts.” 1 Dulcis ad hesternas fuerat mihi rixa lucernas vocis et insanae tot maledicta tuae. 5 tu vero nostros audax invade capillos et mea formosis unguibus ora nota, tu minitare oculos subiecta exurere flamma, fac mea rescisso pectora nuda sinu! 3 cum furibunda mero mensam propellis et in me 4 proicis insana cymbia plena manu, 9 nimirum veri dantur mihi signa caloris: nam sine amore gravi femina nulla dolet. (Prop. 3.8.1-10) 576 The speaker encourages the puella, here, to switch from words (1-2) to a physical attack (5-10) that will leave marks. In verse 5, she is to pull his hair, perhaps hard enough to rip some of it out; in verse 6 she is to scratch his face; in verse 7 she is to burn his eyes out; in verse 8 she is to tear off his clothing; in verses 3 and 4 the upended table and hurled cup could bruise him, and the wine will stain him and his clothing. The words “mark” (nota, 6) and “signs” (signa, 9) specifically 576 I follow Goold (1990) 246 and Günther (1997) 50 in accepting Heyworth’s (1986) 204 correction of the received order of verses in this passage (verses 3 and 4 moved to between verses 8 and 9). ω has cur at the start of verse 3; Beroaldus corrected it to cum, and most if not all modern editors accept this correction. 202 indicate that the speaker intends the marks the puella will leave to signify and so to make his body a legible text. This violence of the puella toward the speaker is the mirror image of the elegiac speakers’ violence toward their girlfriends and boyfriends that Fredrick discusses. 577 The Propertian speaker even uses the same images and language to describe the puella’s violence as he and Tibullus’ speaker use earlier for their own violence. Looking forward to attacking his wife’s hair, the Tibullan farmer-elegist says, “let it be enough to pull apart her coiffure” (sit satis ornatus dissoluisse comae, Tib.1.10.62), and the Propertian speaker threatens to bruise Cynthia, saying, “you’ll show bruised arms to your mother” (ostendes matri bracchia laesa tuae, Prop.2.15.20). These bruises will be marks for Cynthia to show her mother – she and her mother will read in them the speaker’s passion and his violent, domineering masculinity. In another poem, he says he will not pull Cynthia’s hair or bruise her: “neither would I dare to angrily pull apart your plaited hair nor to to wound you with harsh thumbs” (nec tibi conexos iratus carpere crinis / nec duris ausim laedere pollicibus, Prop. 2.5.23-4). When the speaker eggs the puella on to violence in Poem 3.8 above, he encourages her to pull his hair (5), to scratch his face (6), possibly to bruise him (3-4), and to leave marks (9-10). Another verse in Poem 3.8, “rip back my tunic to bare my chest” (fac mea rescisso pectora nuda sinu, Prop. 3.8.8), echoes Prop. 2.5.21, “I will not rip the dress from your deceitful body” (nec tibi periuro scindam de corpore vestis), as well as Prop. 2.15.18, “your dress torn away, you will come to know my hands” (scissa veste meas experiere manus), and Tib. 1.10.61, “let it be enough to tear back the thin dress from her limbs” (sit satis e membris tenuem rescindere vestem). All four of these verses use forms of the same verb for tearing, as well: rescisso, scindam, scissa, and rescindere. The Propertian speaker, then, wants the puella to 577 Fredrick (1997) 203 commit the same violence against him that elegiac lovers commit against their beloveds and to mark his body in ways that signify just as elegiac lovers mark the bodies of their beloveds. Fredrick argues that Sulpicia’s elegies, by inverting the usual gender order of Augustan elegy, expose “not only an instability of gender intrinsic to elegy, but a confusion of the inscriptive metaphor itself...” 578 To this we may add confusion between and possible conflation of the speakers’ bodies and their beloveds’ bodies as texts and specifically the text of elegy itself. This confusion chimes with Flaschenriem’s argument, followed and expanded upon by Wyke, 579 that “we could read the spectral Cynthia of 4.7 as a metaphor for a feminine perspective, or a subjectivity, which the poet’s earlier erotic fictions acknowledge intermittently, but generally appropriate as a part of the male narrator’s literary repertoire.” 580 As Wyke notes, 581 this dynamic is similar to what Zeitlin famously describes in Athenian drama: “Even when female characters struggle with the conflicts generated by the peculiarities of their subordinate social position, their demands for identity and self-esteem are still designed primarily for exploring the male project of selfhood in the larger world.” 582 Propertian elegy presents the scripta puella (Prop. 2.10.8) as part of the self-making of the scriptus poeta. 583 Body Terms Borrowed from Callimachus and Oratory The metaphor of text as body or body part is far older than Propertius. The Athenians of the 5 th century BCE applied vocabulary for describing bodies to speeches, and Propertius’ hero 578 Fredrick (1997) 188 579 Wyke (2002) 185-91 580 Flaschenriem (1998) 63 581 Wyke (2002) 183-4 582 Zeitlin (1996) 347 583 cf. McCarthy (1998) 175: “neither the suffering lover nor the triumphant poet alone reflects the position of the real author; it is the author’s stake in both of these identities that promotes elegy’s ‘male project of selfhood.’ This poetry reflects the elegist’s attempt to stake out a place for himself in the complex hierarchies that shape Roman life...” 204 and model Callimachus took the metaphor for granted to such an extent that he deployed it through puns in the programmatic opening poem of his Aetia: 584 “…to thunder is not mine, but Zeus’.’ For when I first set a tablet on my knees, Lycian Apollo said to me, ‘………. poet, feed the victim to be as fat as possible, but keep the Muse slender, my good man; and I bid you this: tread a path wagons don’t trample, along the same tracks as others don’t drive your chariot, and not along a wide road, but along paths unworn, even if you will drive a tighter course. For we sing among those who love the shrill voice of cicadas and don’t love the din of asses.’ Let someone else bray like the long-eared beast, but let me be the exquisite one, the winged one…” ...βροντᾶν οὐκ ἐμόν, ἀλλὰ Διός.” καὶ γὰρ ὅτε πρώτιστον ἐμοῖς ἐπὶ δέλτον ἔθηκα γούνασιν, Ἀπόλλων εἶπεν ὅ μοι Λύκιος· “……….] ἀοιδέ, τὸ μὲν θύος ὅττι πάχιστον θρέψαι, τὴ]ν Μοῦσαν δ’ ὠγαθὲ λεπταλέην· πρὸς δέ σε] καὶ τόδ’ ἄνωγα, τὰ μὴ πατέουσιν ἅμαξαι τὰ στείβειν, ἑτέρων δ’ ἴχνια μὴ καθ’ ὁμά δίφρον ἐλ]ᾶν μηδ’ οἶμον ἀνὰ πλατύν, ἀλλὰ κελεύθους ἀτρίπτο]υς, εἰ καὶ στεινοτέρην ἐλάσεις. τεττίγω]ν ἐνὶ τοῖς γὰρ ἀείδομεν οἳ λιγὺν ἦχον θ]όρυβον δ’ οὐκ ἐφίλησαν ὄνων.” θηρὶ μὲν οὐατόεντι πανείκελον ὀγκήσαιτο ἄλλος, ἐγ]ὼ δ’ εἴην οὑλαχύς, ὁ πτερόεις… (Callim. Aet. 1.20-32) Apollo demands that the Muse, standing in metonymically for Callimachus’ poetry, be “slender” (λεπταλέην, 24). The speaker loves and seeks to emulate the “shrill voice of cicadas” (τεττίγων...λιγὺν ἦχον, 29) and wishes to be “exquisite” (οὑλαχύς, 32) and “winged” (πτερόεις, 32). The Propertian speaker invokes these verses and their rhetoric about bodies and texts in Poem 2.1, one of his own programmatic poems: “But neither the Phlegraean din of Jove and Enceladus would Callimachus with his slender chest thunder forth nor is my diaphragm suited, with tough verse to establish Caesar’s name among his Phrygian forefathers.” 584 Keith (1999) 41 205 sed neque Phlegraeos Iovis Enceladi tumultus intonet angusto pectore Callimachus, nec mea conveniunt duro praecordia versu Caesaris in Phrygios condere nomen avos (Prop. 2.1.39-41) In addition, the Callimachean speaker says that his lap is his place of composition (Callim. Aet. 1.21-2), again tying his writing to his body, and the Propertian speaker borrows the body as site of composition but changes the specific location to his “mouth” (ora, Prop. 2.1.2). Wimmel has catalogued and explored Propertius and other Augustan poets’ reuse of Callimachus’ ideas and rhetoric at length. 585 Horace, Cicero, and other Roman practitioners and scholars of poetry and rhetoric appropriated the Greek terminology associating the text with the author’s body for their own context and purposes, and Propertius drew from their adaptations of the Greek tradition as well as from the Greek tradition itself. 586 Horace, for example, says that Lucilius is not a real poet because if you rearrange his words “you will not, as you would if you should take apart ‘after loathsome Discord shattered the iron posts and gates of War,’ find the limbs of even a dismembered poet” (non, ut si solvas ‘postquam Discordia taetra / Belli feratos postis portasque refregit,’ invenias etiam disiecti membra poetae, Hor. Sat. 1.4.60-2). For Horace, Lucilius is not a real poet because his text is not made of his body. 587 On Cicero’s deliberate conflation of an author’s text and body, Keith writes, “In the late Republic the metaphor that shapes a literary text in the image of its author’s body is especially concentrated in Cicero’s discussions of the ‘plain’ oratorical style (genus humile or tenue) which he associates with the Roman ‘Atticists.’” 588 Cicero says that, while Lysias’ speeches have “muscle” (lacerti, Cic. Brut. 64), their style is still “rather shrivelled” 585 Wimmel (1960). See also Keith (2008) 45-85 586 Keith (1999) 587 Cf. Keith (1999) 41. It is also possible to read this passage as saying that Lucilius’ poetry may be made of his body, but his body is not a poet’s body, so Lucilius is not a poet. I am not sure whether to impute such circular logic to Horace, but the point stands that, for Horace, poetry is made of the poet’s body. 588 Keith (1999) 42-3 206 (strigosior, Cic. Brut. 64) and that the Roman Atticists whole follow Lysias’ have scrawny bodies (Cic. Brut. 64) and “speak concisely and thinly” (anguste et exiliter dicere, Cic. Brut. 289) in orations characterized by “meagerness, dryness, and impotence” (ieiunitatem et siccitatem et inopiam, Cic. Brut. 285). 589 Keith demonstrates that the Augustan elegists in general 590 and Propertius in particular 591 received extensive education in oratory and law that would have exposed them to Cicero’s rhetoric about bodies and texts. She then argues that the Augustan elegists define their genre, in part, by deploying adjectives such as mollis, tenuus, and angustus and other body metaphors acquired from Cicero to discuss their own texts. 592 The Propertian speaker adapts the Greek and Roman traditions of associating a text with its author’s body to his own ends. In doing so, he involves quotation and allusion in his association of his text with his body, as his recommendation to Lynceus indicates: “Instead, imitate the lighter Muse of Philitas and the dreams of not overblown Callimachus” (tu satius Musam leviorem imitere Philitae / et non inflati somnia Callimachi, Prop. 2.34.31-2). 593 In these verses, a slender style is achieved by copying the right models, whom the speaker describes using terms drawn from rhetorical discussions of text as body, “light” (leviorem 594 ) and “not overblown” (non inflati 595 ). 589 Keith (1999) 42-5, and see discussion of Propertius, Cicero, and Vergil’s king bees in Chapter 3. 590 Keith (1999) 44-6 591 Keith (2008) 19-44 592 Keith (1999) 46ff. 593 Verse 31 is highly contested. The traditions stemming from the Π and Λ read tu satius musis memorem…, which is unmetrical. N reads tu satius memorem musis…, of which many editors are suspicious. Fedeli’s Teubner prints the N text, which Heyworth also prefers, while Goold (1990) 210 prints tu satius Musam leviorem…, and Günther (1997) 11 suggests tu satius tenues Musas imitere Philitae. Nobody seems especially confident. I have preferred Goold’s text here for the parallel grammar it creates between Musam…Philitae and somnia…Callimachi. This reading adds the word leviorem, which supports my argument, but the presence of the phrase non inflati in the next verse allows the argument to work without leviorem. 594 E.g., Cic. Pis. 24: “Great is the name, great is the semblance, great is the dignity, great is the majesty of a consul; the slightness of your chest cannot comprehend it, your lightness cannot take it in, nor your poverty of mind; not the weakness of your intellect, not your inexperience of favorable circumstances can support so great, so weighty, so serious a character” (magnum nomen est, magna species, magna dignitas, magna maiestas consulis; non capiunt angustiae pectoris tui, non recipit levitas ista, non egestas animi; non infirmitas ingenii sustinet, non insolentia rerum secundarum tantam personam, tam gravem, tam severam). 595 Keith (1999) 44 cites Tac. Dial. 18.4 on what Cicero’s contemporary critics said about his texts: “the sort to whom he seemed overblown and swollen” (quibus inflatus et tumens…videretur). 207 A few generations later, Quintilian explicitly links quotable quotes (sententiae) with the idea of text as body in his Institutio Oratoria: “In truth, I consider these highlights of oratory to be, as it were, some eyes of eloquence. But I would not want eyes to be all over the body, lest other body parts lose their function” (ego vero haec lumina orationis velut oculos quosdam esse eloquentia credo. sed neque oculos esse toto corpore velim, ne cetera membra officium suum perdant Quint. Inst. 8.5.34). In this passage, Quintilian compares sententiae to beautiful, bright eyes and the full text of a speech to the whole body. Similarly, in the Dialogus de Oratoribus, attributed to Tacitus, the character Messalla 596 laments that orators of his day “reduce eloquence to very few senses and restricted sententiae as if she were driven out from her own kingdom, so that she who once as mistress of all the arts filled people’s chests with her very beautiful retinue but now, clipped on all sides and amputated, without her trappings, without mark of honor, I would almost have said without freeborn status, is learned as if one from among the most sordid crafts.” in paucissimos sensus et angustas sententias detrudunt eloquentiam velut expulsam regno suo, ut quae olim omnium artium domina pulcherrimo comitatu pectora implebat, nunc circumcisa et amputata, sine apparatu, sine honore, paene dixerim sine ingenuitate, quasi una ex sordidissimis artificiis discatur. (Tac. Dial. 32.4) Here, Tacitus’ Messala compares quotable quotes with body parts and personal effects associated so closely with the body that they become part of a person’s body schema. 597 O’Gorman has pointed out that the treatment of sententiae as body parts in these two passages follows logically from “the traditional parallel between the organic body and the well-structured speech,” 598 which was familiar to Propertius. 599 And, as discussed above, the Propertian speaker often compares his text or parts of his text with parts of his body. Performance, Books, and the Incorporation of Text into Body 596 Lucius Vipstanus Messalla not Marcus Valerius Messalla Corvinus, patron of Tibullus and contemporary of Propertius 597 On body schema see the section on “New Media and the Incorporation of Text into Body” below. 598 O’Gorman (2016). This was a paper delivered at a conference, and I thank O’Gorman for her friendly permission to quote her and to work with ideas she is developing. 599 Keith (1999) 44-6; (2008) 19-44, and see above. 208 In addition, the speaker incorporated the text of the elegies into his body when he performed them. Romans understood performances to open the body and gender of the performer to scrutiny and interpretation. Gamel cites, for example, the anxieties surrounding gladiators and orators and their bodies, movements, and voices in their respective performances 600 and notes “the physical, emotional, and ethical investment involved in performing,” 601 especially considering “the way the activity and significance of performance was conceived in Rome.” 602 Narrowing her focus to Propertius, Gamel elaborates on the nature of these investments: “As a man performed the roles of both amator and puella, putting his own voice into the pronouns and his own body into the gestures, the reader/performer of this poetry performed his own involvement in the system of gender construction more consciously and more actively than did silent readers or audience members watching comedies.” 603 The text of the poem he was performing encouraged the speaker to use his body in certain ways. As Gamel suggests, he may have differentiated among the amator, the puella, and any other personas with his gestures and his overall physical carriage. He may also have used his voice to differentiate and so have involved his diaphragm, back and intercostal muscles, throat muscles, vocal chords and larynx, soft and hard palates, pharynx, jaw, tongue, lips, and other body parts 604 in the production of lower or high, softer or louder, smoother or harsher tones. Similarly, if he changed accents for different characters, he may have changed the shapes his mouth formed and the placement of his voice in his mouth. In these and/or other ways of conveying character, the speaker worked the words as well as the ideas and implications of the text into his body. It is unlikely that a performer of the Elegies, especially Propertius himself, would have spoken them flatly and with minimal expression, as “authors were known not only by the nature of their 600 Gamel (1998) 91 601 Gamel (1998) 92 602 Gamel (1998) 92 603 Gamel (1998) 92 604 Vocal production can sometimes recruit parts of the body with no obvious bearing on the voice. Many singers have, for example, raised their eyebrows in search of high notes. 209 compositions but by the effectiveness with which they performed them,” 605 but even if he did, his vocal apparatus would still have produced the words, and his body would still have been subjected to the gaze of the audience. The effects on the speaker’s body continued beyond the temporal and spatial confines of a performance of any one poem, as well. In performing an elegy, the speaker took on a persona or personas with particular and sometimes unconventional ideas about bodies and gender to sing a text that exposed the constructedness of bodies and the performativity of gender. Performing an elegy changed how the speaker interacted with the world, whether it challenged the norms in his mind and opened them to further query and play or whether playing across a boundary briefly exposed categories and cultural constructs to query and play only to leave them reinforced when the speaker returned to his normal roles. 606 The text and its performance, furthermore, drew from and contributed to Roman discourses about bodies and gender; through participation in these discourses, the text and its performance inevitably affected them. 607 This is part Gamel’s insight in her phrase “[the reader/performer’s] own involvement in the system of gender construction.” Latin poetry, including Augustan elegy, is ritualized language, 608 and theorists of rituals argue that part of the effectiveness of a ritual is that some of the changes to reality that occur within the special rules of operation that the ritual temporarily instantiates leak out of the ritual time and place and become part of ordinary reality. 609 Habinek thus refers to “song’s role in the foundation and 605 Gamel (1998) 82 cites Juvenal 7.83-7 and notes that “Vergil and Statius were especially gifted.” 606 See Thorne (1993) 122-123 on the concept of “crossing.” 607 Persius provides an example, albeit hyperbolic and polemical, of some of the effects on a Roman audience of a recitation of certain kinds of poetry: “Then with neither moral decency nor a calm voice you will see huge Tituses tremble, when your songs enter their loins and they are titillated in their deepest parts by your pulsing verse” (tunc neque more probo videas nec voce serena / ingentis trepidare Titos, cum carmina lumbum intrant et tremulo scalpuntur ubi intima versu, Pers. Sat. 1.19-21). 608 Habinek (2005) 4-6 609 Geertz (1973) 112; Rappaport (1992) 12-3, (1999) 96-7; Bell (1993) 140-1; Butler (1993) 10, 95; Hollywood (2006) 252-6, 267-9 210 maintenance of social order.” 610 A performance of an elegy by Propertius, then, changed the contemporary discourses on bodies and gender and the speaker’s relationships with those discourses. It became part of the speaker’s experience of his body and of other people’s experience of his body. The written forms and performances of Propertius’ elegies together formed the speaker’s body. Fitzgerald writes, “The ambiguous status of the Catullan poem enhances its elusive quality, the fact that it seems both casually occasional and yet fixed in its polished rightness of expression.” 611 The same can be said of Propertius’ elegies, at least of the passages not known or suspected to be so very corrupt. They present themselves with immediacy as windows onto the many agonies and trials and the occasional triumphs and joys of the elegiac lover-poet, and yet they are at the same time self-professedly Callimachean. Fitzgerald suggests a possible reason for this ambiguity: “Perhaps this is rather a function of the ambiguous status of [the poem’s] receivers, who have different kinds of access to the poem depending on whether they situate themselves as audience or as readership. The poem may be performed, or else read as a text; either way, the consumption of the poem is incomplete. Because the written poem is the record of a performance we have missed, and the spoken poem can never realize the possibilities of the written text, the poem is always either more immediate or more enduring than what we experience.” 612 At the same time, oral performance by the author largely maintained the text as embedded in a particular context determined by the author (though a listener might remember some or all of the words of a text but forget or ‘improve’ the gestures that accompanied it in performance) but limited 610 Habinek (2005) 9 611 Fitzgerald (1995) 6 612 Fitzgerald (1995) 6. When Fitzgerald says that “the written poem is the record of a performance we have missed,” he seems to take a strong stance on the roles of performance in the composition and publication of Catullus’ poems. A critic wanting to take a milder stance with regard to the poems of Propertius can still argue that the poems present themselves as the written records of performances (the speaker performs his status as an elegiac lover-poet in all of his verbal actions) and that Propertius did in fact perform his poems, if Ovid is to be believed (Ov. Trist. 4.10.45-6). This may actually be what Fitzgerald is saying about Catullus’ poems, given the discussion that precedes and follows the comment I have quoted. 211 the exposure of the text, whereas written copies expanded publication but also allowed readers to more easily disembed the text from authorially controlled contexts. 613 Written copies and oral performances of Propertius’ elegies presented the text in complementary ways, and so in complementary ways they incorporated the text into the speaker’s body. New Media and the Incorporation of Text into Body New media for written and oral publication had recently become popular when Propertius composed his elegies: Catullus’ poetry-book is “the first extant example of a true lyric collection... A qualitative break separates his work from that of the earlier Greek monodists,” 614 and poetry books “became fashionable” in the 30’s and 20’s BCE. 615 At the same time, the institution of the recitatio came into vogue with Asinius Pollio leading the way: 616 “Asinius Pollio never declaimed for a crowd that had been granted entry, but neither was ambition in his pursuits lacking in him: For he was the first of all Romans to recite his own writings for invited people” (Pollio Asinius numquam admissa multitudine declamavit; nec illi ambitio in studiis defuit: primus enim omnium Romanorum advocatis hominibus sua scripta recitavit, Sen. Contr. 4 praef. 2). 617 Via these new 613 Habinek (1998) 106-8 614 Miller (1994) 52 615 Wyke (2002) 181 616 Dupont (1997) 45; Habinek (1998) 107; Wyke (2002) 181 617 Against the idea that Pollio was the first to perform his own work in a recitatio, Dalzell (1955) 26 writes, “The phrase ‘advocatis hominibus’ is ambiguous. It could mean that Pollio issued invitations to chosen individuals or it could imply a general summons as in Cicero, De Domo Sua, 124, ‘contione advocata’ and Livy I, 59, 7, ‘populum advocavit’. That the second is the meaning here seems certain from ‘admissa multitudine’ in the previous sentence, to which ‘advocatis hominibus’ is clearly parallel. Unless these words indicate that Pollio recited before a large audience, the sentence would not be proof of his ambitio, which is the reason for its introduction. If this argument is granted, then we may conclude that Pollio did not limit his invitations to a select few, but was prepared to read his works to all who were interested or could afford the time to listen.” This argument misses the contrast expressed in the word nec between never (numquam) declaiming for a crowd that had been granted audience (i.e., had arrived of its own accord and been permitted to listen) and being the first Roman to recite for invited guests (i.e., people whose presence had been requested). Dalzell (1955) 26-8 does, however, provide suggestive evidence about the establishment of libraries as places of performance in Rome to argue that Pollio formalized the institution of the recitatio in libraries by convening recitationes in the Atrium Libertatis. 212 media of diffusion arrived Augustan elegy as a new genre. Gallus had begun it or something very close to it in the 40’s, but his work may not have survived as widespread past the early 20’s BCE. 618 Right around this time, Propertius published his first book of elegies and became an overnight sensation. According to McLuhan, “any invention or technology is an extension or self- amputation of our physical bodies, and such extension also demands new ratios and new equilibriums among the other organs and extensions of the body.” 619 If McLuhan is right, 620 Propertius’ elegies became part of the bodies of all of their participants, including the speaker and author. 621 The idea of embodied cognition, developed in cognitive psychology, supports this part of McLuhan’s thesis, and its application to Mycenaean archaeology provides a comparandum for the Propertian text as body part. Malafouris writes, 618 Gallus’ works did not survive antiquity except in fragments quoted in other authors and in a recently discovered papyrus. There is reason to believe that Augustus may have put a stop to their publication and continued diffusion in the early 20s BCE. A senatus consultum (Sue. Aug. 66.2) in 27 or 26 BCE condemned Gallus to exile and confiscation, and he committed suicide shortly afterwards. Ross (1975) 39 believes that Augustus ordered Gallus’ works destroyed in a damnatio memoriae, whereas Miller (2004) 74 writes that “No formal book burning or proscription took place” but that Gallus’ works may have been informally put away and allowed to disappear. 619 McLuhan (1964) 45 620 Of the many critiques of McLuhan’s work, some of the most famous are Eco’s (1986) 231-4 accusations that “he offers us the co-presence of arguments as if it were a logical suggession” and that he conflates the medium with its content, and that “To say that [the medium] represents an extension of our bodies still means little. The wheel extends the capacity of the foot and the lever that of the arm, but the alphabet reduces, according to criteria of a particular economy, the possibilities of the sound-making organs in order to allow a certain codification of experience. The sense in which the press is a medium is not the same as that in which language is a medium. The press does not change the coding of experience, with respect to written language, but fosters its diffusion and increments certain developments in the direction of precision, standardization, and so on.” On this last point, Eco goes too far. He cites increased “precision, standardization, and so on” to prove that “the press does not change the coding of experience,” but these very developments in fact contradict his argument: Increased standardization, for example, is a change in the code itself. Debray (1996) 71, 73-4 criticizes both McLuhan and Eco and takes a compromise position between theirs: “Clearly [McLuhan] was unfamiliar with Saussure (and no less with Seignobos). Though it may also be that Thomas Aquinas’ close reader Eco pays no heed to the complex materialities of the medium. Whereupon arises this unfortunate symmetry: against the medium minus code and message of the McLuhanites, the semiologists set up codes minus medium and milieu,” and “a choice must be made to emphasize code or codex, the language system or its material embodiment, perhaps one will say. But that’s exactly what one had best avoid doing by carefully examining watersheds and precise turning points. What good is it to hock one fetishism for another—structure for material implementation?” 621 cf. Wyke (2002) 189: “the elegiac genre as itself a technology of gender” and “Propertian elegy was part of an institutionalized system of representation—a social technology—through which gender was performed and, therefore, constructed at Rome.” 213 “The general idea behind embodied cognition is quite simple: the body is not as conventionally held, a passive external container of the human mind that has little to do with cognition per se but a constitutive and integral component of the way we think. In other words, the mind does not inhabit the body, it is rather the body that inhabits the mind. The task is not to understand how the body contains the mind, but how the body shapes the mind.” 622 Malafouris uses this idea to explore the relationships between the Mycenaean sword, a new technology that spread quickly and brought with it changes in military practice and in “personal and cultural Mycenaean identity,” 623 and the bodies and cognition of the Mycenaean warriors who used it: “The early Mycenaean warrior is not simply using a new weapon but is extending and transforming his very self. He is not the same warrior in possession of a better weapon but a substantively different human/non- human hybrid. The sword does not merely represent a new aspect of the emerging Mycenaean world, but constitutes a novel concrete situational perspective of being-in-the-Mycenaean world. The intentional stance of the Mycenaean person is partially determined by the skilled embodied engagements made possible by the use of the sword. Representational content and ‘aboutness’ are not to be found inside the cabinet of the Mycenaean head they are instead negotiated between the hand and the sword.” 624 The sword becomes part of how the Mycenaean warrior encounters, understands, and interacts with the world around him, and it influences those encounters, understandings, and interactions. As discussed above, the text of Propertius’ elegies, both in performance and in written copies, becomes part of how the speaker encounters, understands, and interacts with the world around him. The text occasions certain actions and experiences, such as recitation or private reading, and takes part in determining their features (who else is present, how they interact with the speaker, et al.). Even in situations not initially brought about by the text—a casual conversation in the Forum or a trip to Baiae, for example—the user of the text, who has become the speaker if they have read it aloud, may remember a reference in the text to a building or place that colors their experience of that building or place, or the user may perform their gender in ways that the text has influenced. 622 Malafouris (2008) 116 623 Malafouris (2008) 117-19 624 Malafouris (2008) 122 214 The Mycenaean sword as body part is not a metaphor. 625 To explain how a sword can actually become part of a human body, Malafouris introduces the concept of the body schema: “The notion of the ‘body schema’ was first introduced by Head and Holmes (1911–1912) (Oldfield and Zangwill, 1942) and currently denotes in cognitive neuroscience the complicated neuronal network responsible for continually tracking the position of our body in space, the dynamic configurations of our limb segments and the shape of our body surface. In other words it can be understood as an unconscious body map responsible for the constant monitoring of the execution of actions with the different body parts. According to Melzack (1990) the body schema although largely prewired by genetics it is open to continuous shaping influences of experience, and what is important to note in this context is the effect that external objects and prostheses appear to have in the cognitive topography of this space. More specifically, not only behavioural and imaging studies of visuotactile interactions have shown that tool-use extends the ‘peripersonal space’ – i.e. the behavioural space that immediately surrounds the body – but more important, recent neuroscientfic findings suggest that the systematic association between the body and inanimate objects (like clothes, jewelry, tools, etc.) can result into a temporary or permanent incorporation of the latter into the body schema (Berti and Frassinetti 2000; Farne and Ladavas, 2000; Farne et al. 2005; Flugel 1930; Graziano et al. 2002; Holmes and Spence 2006; Holmes et al. 2005: 62, 2004; Iriki et al. 1996; Maravita and Iriki 2004; Maravita et al. 2002, 2003). An observation which essentially means that objects and tools attached to the body can become a part of the body as the physical body itself. Head and Holmes referred to this phenomenon with their famous comment that ‘a woman’s power of localization may extend to the feather in her hat’ (1911–12: 188).” 626 The Mycenaean sword, argues Malafouris, was incorporated into the Mycenaean warrior’s body schema. The sword caused the warrior’s neural pathways to re-order themselves in response to changes in the warrior’s experience of his body and the world around him. 627 If the Propertian speaker’s text was part of his body schema, then it caused analogous changes in his cerebral cortex. This last argument faces the challenge that the speaker of the elegies is a fictional persona, not a real person with a real neural network. But in the various circumstances of composition, writing, performance, and reading and the various forms the text took in those circumstances, there was one or more real, historical person behind the fictional persona. In performance, the performer was a real person who took on the persona of Propertius the elegiac lover-poet. People reading to themselves may or may not have identified with the persona of Propertius. In any case, Propertius 625 Malafouris (2008) 122 626 Malafouris (2008) 123, and see Malafouris (2008) 123-5 on evidence that studies of neuropathology provide in support of this thesis. 627 Malafouris (2008) 125 215 the fictional persona is a product of Propertius the historical person. He is formed from the experiences of Propertius the historical person, whether or not the historical person experienced all or most or even any of what the fictional persona experiences. It does seem, though, that the historical person fashioned the fictional persona with some of his own experiences: While Propertian scholarship has rightfully repudiated the quest to understand the elegies as a largely nonfictional account of a real love affair between Propertius the historical person and a real, historical woman named Hostia (Apul. Apol. 10), scholars still routinely, albeit cautiously, draw biographical details about Propertius the historical person from the account of Propertius the fictional persona, especially from Poems 1.22 and 4.1. Propertius the author, the historical person, thus remained and remains a real person behind the fictionalized persona of Propertius the elegiac lover-poet. The Mycenaean sword as body part is not a metaphor, but sometimes the Propertian speaker equates his text to his body or to body parts through metaphors. This is perhaps an attempt to articulate the intuition or realization that his text seemed to be part of his body with the conceptual apparatus that he could access or invent. Propertius, of course, did not have access to 20th- and 21st-century theories of ritualization, distributed cognition, and cognitive body schemas. But neither did or do many of the people whose experiences have informed the scholarship that has developed these theories. The Mycenaean warrior did not, in all likelihood, know how to say that his sword was part of his body schema. The theorists describe some of the processes at work in the experiences of the people they study, and the passages from Propertius’ elegies above, which compare the text to part or all of the speaker’s body or treat it as part or all of his body, explore similar workings in light of social, political, and technological changes. Some ideas about text as body or body part, such as Cicero’s and perhaps Quintilian’s, were current in Propertius’ day. The 216 different ways in which the speaker draws relationships between his text and his body pose questions—how can words be part of a body, and what does that mean?—and try out possible answers to them. 217 Chapter 5: Vulnerabilities Compared Introduction In this chapter, I draw together the threads of the previous chapters. Chapters 2 and 3 explored the literary culture in which Propertius composed the Elegies and created their vulnerable textuality. Those chapters paid particular attention to that literary culture’s practices of reuse and erasure, with which the vulnerable textuality of the Elegies engages. Chapter 4 then examined passages in which the speaker equates his text with his body or part of his body and argued that the texts of the Elegies were part or all of the fictive Propertian speaker’s body as well as part of the bodies of anyone who performed them and of Propertius the historical author. This chapter will investigate what the speaker of the Elegies says about the socio-political circumstances of his body and speech in order to interpret how his negotiation of his literary milieu, as discussed in Chapters 1, 2, and 3, relates to the corpora of Chapter 4. If the text is the body or part of the bodies of the fictive Propertian speaker, a real person who performs Propertius’ poetry, and the historical Propertius himself, what do the Elegies say or imply that it means for their bodies when someone other than the speaker changes or destroys part of the text? I seek the beginning of an answer to this question (more of the answer will come in the Conclusion) in the Propertian speaker’s ambivalence toward Octavian/Augustus. Scholars disagree about the overall tone of Octavian/Augustus’ relationships with the senatorial and equestrian elites and about how to interpret particular relationships, policies, remarks, and incidents. It is not my aim here to resolve these problems but rather to examine passages in which the speaker of Propertius’ Elegies talks about the physical oppression he suffers and the constraints 218 on his speech as a loss of liberty and passages in which he connects his erotic oppression with Octavian/Augustus. Oriented thus, this chapter risks portraying the Elegies as only critical and never appreciative of Octavian/Augustus. This would, of course, misrepresent what the Elegies say outright about the princeps and would hazardously ignore ironies and problems of art and historicity. 628 Following many other interpreters, I read in the Elegies mixed feelings about Octavian/Augustus. 629 Many passages in the Elegies can be read as praising Octavian/Augustus, including some of the passages in which I point out anger and mistrust toward him below. With this clarified, the following analyses focus on the Propertian speaker’s erotic anxieties about his body and speech and then on his anxieties about them in relation to Octavian/Augustus. First, though, it will be helpful to lay out the traditional Roman ideas about libertas and virtus in relation to bodies and speech, with which the Propertian speaker’s ideas are in dialogue. Libertas and Virtus An important element of libertas was the idea that the body of an adult, citizen man was legally inviolable. In the De Republica, Cicero narrates how, in the first few years of the Republic, P. Valerius Publicola instituted a law forbidding any magistrate to beat or execute a citizen who exercised the right of appeal to the people (provocatio). 630 In a speech defending Rabirius, he reminds the Senate that it is illegal to beat a citizen, and he connects this protection directly with libertas: “The Porcian Law cast off the rods from the body of all Roman citizens...The Porcian 628 See Veyne (1988) and Kennedy (1993) for discussion. 629 For example, Stahl (1985), White (1993) and (2005), DeBrohun (2003) esp. 33-117 and 210-35, Miller (2004), Keith (2008) 139-65, and Breed (2010) 630 Cic. Rep. 2.53. Nicolet (1976) 430 notes that this traditional narrative is “without doubt” anachronistic and sets the date for the establishment of the law too early. 219 Law set the libertas of the citizens free from the lictor” (Porcia lex virgas ab omnium civium Romanorum corpore amovit...Porcia lex libertatem civium lictori eripuit, Cic. Rab. Perd. 4.12). These laws protected the bodies of citizens from violence by the state, which fell under the category of coercion by the state (coercitio). 631 Nicolet writes that Roman primary sources treat this protection as “the most precious privilege of Roman liberty.” 632 Thus, under these laws and social expectations, “Cicero (In Verrem 2.5.160-73) alienated support from the corrupt governor Verres by stressing his breach of libertas in the public beating of the Roman eques, Gavisa, in Sicily.” 633 Legal protection from violence by the state, however, was not the only meaning of the bodily inviolability of a vir. There were also protections against violence by individuals acting privately. 634 In addition, the body of a vir was not supposed to be penetrated. This put citizen soldiers in a difficult ideological position that changed over time; by the late Republic “the increasing emphasis on libertas encouraged a growing disjunction between military service, especially among ‘the herd’, and the status of vir.” 635 Sexual penetration was mostly a clearer case. The Romans saw sexual penetration as a form of domination, 636 so for a vir to accept or, worse yet, desire penetration was for him to accept or desire domination, or in other terms, to allow himself to be treated or to desire to be treated like a woman, a boy (and not even a respectable, 631 Nicolet (1976) 430 632 Nicolet (1976) 430 633 Alston (1998) 207 634 Lintott (1968) 107-31 provides a useful discussion of these laws and remedies. I do not review this issue here, because my dissertation will concern violence by the state. 635 Alston (1998) 211 636 Richlin (1983) 140-1; Edwards (1993) 70, 74-5 220 citizen boy), 637 or a slave. 638 Alongside diminished social status, legal penalties could include fines and restricted political and legal participation. 639 A key component of elite men’s political capacity was public oratory. Bonnefond-Coudry cites several passages in which Cicero insists that an elite man should participate in public life. 640 Among them are these passages from the De Officiis, in which Cicero expresses the centrality of public speech to civic participation: “And if only the Republic had remained in the state in which it began and had not fallen into the hands of men desirous not so much of changing the state as of overturning it! For first, as I used to do when the Republic was standing, I would put more of my effort into public speaking than into philosophical writing, 641 and second, I would be committing to writing not these things, which I am now, but my speeches, as I often did. But when the Republic, on which all my care, my thought, my work used to be set, was entirely no more, naturally those speeches in the Forum and the Senate fell silent.” Atque utinam res publica stetisset quo coeperat statu nec in homines non tam commutandarum quam evertendarum rerum cupidos incidisset! Primum enim, ut stante re publica facere solebamus, in agendo plus quam in scribendo operae poneremus, deinde ipsis scriptis non ea, quae nunc, sed actiones nostras mandaremus, ut saepe fecimus. cum autem res publica, in qua omnis mea cura, cogitatio, opera poni solebat, nulla esset omnino, illae scilicet litterae conticuerunt forenses et senatoriae. (Cic. Off. 2.3). and Now that the Senate has been snuffed out and the courts done away with, what is there that is worthy of us that we can do in the Curia or the Forum? exstincto enim senatu deletisque iudiciis quid est quod dignum nobis aut in Curia aut in Foro agere possimus? (Cic. Off. 3.2) 637 Walters (1997) 31-6 638 Edwards (1993) 70-73 639 Edwards (1993) 71. E.g., disqualification from representing others in the praetor’s court 640 Bonnefond-Coudry (1989) 376ff cites Cic. Off. 2.3, 2.6, 3.2; De Or. 3.63; Rep. 1.10, 1.12 641 Because the word actiones in the next part of the sentence clearly refers to speeches, it makes sense to translate agendo as “public speaking” here. Similarly, because the words ea, quae nunc refer to philosophical writings and are opposed to the word actiones, it makes sense to translate scribendo as “philosophical writing.” The last sentence of this passage supports this reading: understood literally the verb conticuerunt suggests that the litterae were once spoken out loud in the Forum and the Senate (forenses et senatoriae). 221 In the first passage, Cicero singles out public speaking as the representative part of the service to the republic to which he used to devote all his energies (“on which all my care, my thought, my work used to be set,” in qua omnis mea cura, cogitatio, opera poni solebat). He also contrasts and valorizes public speech against philosophical writing, which he believes is a public service, too (Cic. Off. 1.1), but apparently a less valuable one. In the second passage, he says that the Senate and the courts, both institutions in which he spoke publicly, were the only worthy uses of the Curia and the Forum, and the verb “do” here is the same word (agere) that Cicero used for public speech in the first passage (agendo, actiones). Furthermore, this second passage mourns more for the plight of the former public speaker (“what is there that is worthy of us that we can do...?” quid est quod dignum nobis...agere possimus?) than for the Republic, the loss of which Cicero treats as the backdrop for his plight. Connolly argues that “though it is constructed as an elite domain, rhetoric operates as a discourse of citizenship in a broader sense, the collection of rights and obligations that endows individuals with a formal legal identity as free, male, and Roman” 642 and that, in the Rhetorica ad Herennium, the body of the orator is “a point of commonality between speaker and audience— a site at which republican virtue may begin to replicate itself.” 643 Civic participation and public speaking was crucial to viri as such, to their performance of virtus, and to their libertas. Various republican sources connect libertas closely with virtus. A fragment of Ennius’ Phoenix, a tragedy from the early 2 nd century B.C.E., 644 links the concepts vividly: It is proper that a man live animated by true virtus and that he bravely stand blameless against his adversaries. 642 Connolly (2007) 7. cf. Dupont (1997) 44: “during the Roman Republic, oratio, or public discourse, constitutes the means by which the ideal citizen enacts and confirms his status or dignitas within the socio-political hierarchy of the state. For the Roman nobleman, the opportunity to use language in such a manner and thereby gain access to high honors is the essence of libertas.” 643 Connolly (2007) 118-19 644 Boyle (2006) 58 222 This is libertas – he who carries a pure and steadfast heart; the other servile conditions skulk in dark night. sed virum virtute vera vivere animatum addecet fortiterque innoxium stare adversus adversarios. ea libertas est qui pectus purum et firmum gestitat; aliae res obnoxiosae nocte in obscura latent. (Phoenix frag. cxxvi Boyle, after Jocelyn) 645 It is difficult to say from this fragment whether virtus is a precondition for libertas or vice versa; it seems, rather, that they are mutually necessary and supporting conditions. Formally, the fragment virtually equates them. The phrase “this is libertas” immediately follows the elaboration in the second line 646 of what it means to “live animated by true virtus.” In performance, the aural effect is that “this is libertas” seems at first to be talking about “liv[ing] animated by true virtus” and its elaboration. This elaboration (“and that he bravely stand blameless against his adversaries,” fortiterque innoxium stare adversus adversarios) is thematically interchangeable with the description of libertas (“he who carries a pure and steadfast heart,” qui pectus purum et firmum gestitat). While innoxius and purus are not exact synonyms, purus can mean “stainless” and “faultless,” 647 which are close to “blameless” (innoxius). Meanwhile, the phrase “bravely 645 The speaker and context of this passage are not known, but Boyle suggests that the speaker may be Phoenix and that he may be responding to his father’s anger (Boyle 2006: 62). Both Jocelyn (Jocelyn 1967: 390) and Boyle (Boyle 2006: 62-3) think that the speaker of the fragment formulates changes in the concepts of virtus and libertas. Jocelyn writes, “Ennius’ first two verses make a statement about virtus, conventionally the successful exercise of a man’s physical powers in warfare, statecraft and procreation, his second two about libertas, conventionally the state of being a free man and not a slave. In both cases conventional notions are being corrected or reinterpreted but textual corruptions obscure the dramatist’s point” (Jocelyn 1967: 390). Jocelyn does not specify what changes in virtus and libertas he sees here, but on my reading this passage breaks down any boundaries between external peformances of virtus and libertas (stressed in Jocelyn’s account of the earlier understandings of these ideas) and internal qualities of character. If Boyle’s conjecture about the speaker and context is right, Phoenix is interpreting Roman aristocratic values to his father and to the audience in response to a false accusation (see Boyle 2006: 61-2 on aristocratic values in Ennian tragedy). The false accusation would create sympathy for Phoenix, which might incline an audience to receive his message favorably. Unfortunately, the place where the fragment was preserved— a passage of Aulus Gellius’ Noctes Atticae in which Gellius’ speaker is arguing with a grammaticus about the meaning of the word obnoxius (Gell. Noctes Atticae VI.xvii.1-13)—does not offer any help with the context or speaker of the fragment. 646 It is not necessary to read the second line as an elaboration of the phrase virtute vera vivere animatum, but this is a common use of the conjunction que (Lewis and Short 1879: que I.B). In any case, que connects the ideas closely. 647 Lewis and Short (1879) purus II.A suggest these translations for purus in this passage. It is worth noting that contactus (contactum nullis ante cupidinibus, Prop. 1.1.2) is nearly an antonym for purus. In fact, the speaker of 223 stand...against his adversaries” (fortiter...stare adversus adversarios) is one potential enactment of the more general phrase “carries a...steadfast heart” (pectus...firmum gestitat). The words animatum and pectus also link the first and third verses, as both words refer to inner qualities, frame of mind, and disposition, and animatus can even mean “stouthearted” in pre-classical poetry. 648 A passage from Cicero’s fifth Philippic makes a similar connection between virtus and libertas. Cicero writes of Julius Caesar, “Therefore, as he did not have respect for the Senate and the aristocrats, he opened for himself that path for increasing his power which the virtus of a free people (liberi populi) could not bear” (itaque cum respectum ad senatum et ad bonos non haberet, eam sibi viam ipse patefecit ad opes suas amplificandas quam virtus liberi populi ferre non posset, Cic. Phil. 5.49). Erotic Subjugation The beginning of Book 1 of Propertius’ Elegies sets up the speaker’s conflict over these ideas by establishing the theme of his slavery to love (servitium amoris), which was to become a primary trope of Propertius’ poetry. The speaker says that Cynthia and Love have physically dominated him: “Cynthia first captured wretched me with her eyes, me, infected with no Desires before. Then Love threw down my stubbornly arrogant gaze and pressed down my head with his feet placed upon it until he taught me to hate chaste girls, the jerk, and to live with no forethought.” Cynthia prima suis miserum me cepit ocellis, contactum nullis ante Cupidinibus. tum mihi constantis deiecit lumina fastus Propertius 1.1 claims conditions and characteristics, such as enslavement and domination, infection, passivity, and impropriety, which directly oppose those valued in this fragment of the Phoenix, and he continues to do so throughout the Elegies. The Propertian speaker 648 Lewis and Short (1879) animo I.c. See Plaut. Bacch. 4.9.18: milites armati atque animati probe and Att. ap. Non. p. 223.18: cum animatus iero, satis armatus sum. 224 et caput impositis pressit Amor pedibus donec me docuit castas odisse puellas improbus, et nullo vivere consilio. (Prop. 1.1.1-6) This humiliating treatment was unthinkable for a free Roman man according to the old, republican ways of thinking. In fact, it disqualifies the speaker from free status. Since free status really meant non-slave status, and no other possible statuses existed, 649 the speaker is telling us in this passage that Cynthia and Love made him their slave. Love, in particular, takes away the speaker’s ability to act of his own volition by physically controlling his gaze and his head and forcing him to hate chaste girls and to live without prudence or plan. This is a life of licentia, which is libertas that has “degenerate[d]” without the restraints of self-control and the law. 650 Cicero says of free men, “therefore we are all slaves of the laws so that we may be free” (legum idcirco omnes servi sumus ut liberi esse possimus, Cic. Cluent. 164). The seeming freedom of licentia is slavishness in another guise. Since a slave was not a juridical subject but rather was property without legal rights, 651 licentia as libertas stripped of the restraint of law can only be a state proper to a slave. In the vocabulary of speech, licentia is “the usual term for referring to the freedom of expression of slaves during Saturnalia,” 652 and Seneca the Younger reminds Lucilius and other readers of his De Providentia, “consider that we are delighted by our sons’ modesty but our slaves’ licentia, that the former are constrained by sterner discipline while 649 Roller (2001) 221-2: “The complementarity [of servitus and libertas] is implicit in almost any conjunction of these words, and is made explicit in certain legal texts, which in this case seem to reflect the ordinary usage of these terms. Thus the jurist Gaius writes that “all men are either free or slaves” (omnes homines aut liberi sunt aut servi, Inst. 1.9) and Florentinus (Dig. 1.5.4 pr. 1) likewise defines libertas and servitus but no other statuses, implying that there is no other category of personal status in this scheme.” Roller (2001) 218-33 argues that libertas was primarily defined negatively vis-à-vis servitude (servitus), that is, as non-slave status (‘freedom from…’), rather than positively (‘freedom to…’) and that even the examples of positive understandings of libertas that Wirszubski (1950) 1-12 adduces turn out, upon closer inspection, to define libertas as non-slavery. Connolly (2007) 158-62, Ando (2011) 86-7, and Arena (2012) 46-7 more or less agree with Roller on this. Arena (2012) 47 disagrees with Roller’s contention that this definition of libertas implies that libertas was not a political idea. 650 Wirszubski (1950) 7; Braund (2004) passim; Cogitore (2009) 8-9 651 Arena (2012) 15-16 652 Tovar (2005) 788: “término habitual para referirse a la libertad de expresión del esclavo en las Saturnales” 225 the boldness of the latter is fed” (cogita filiorum nos modestia delectari, vernularum licentia, illos disciplina tristiori contineri, horum ali audaciam, Sen. Prov. 1.6). I will not push the metaphor of slavery to love and to the elegiac mistress dogmatically, though, since it is characteristic of Roman elegy to deploy or subvert any particular meaning as is rhetorically convenient to the speaker in a given moment, and since the mistress “is situated as the upper half of an asymmetric master-slave arrangement, chiefly to be castigated as such.” 653 Instead I will examine how, in playing the metaphor out while ultimately retaining control, the speaker explores anxieties about libertas. The first couplet of Book 1 links domination (miserum me cepit) with desire (cupidinibus) figured as a disease (contactum). At the end of the poem, the speaker warns that love causes pain: “Our Venus stirs up bitter nights in me, and at no time is empty Love lacking. Avoid this evil, I warn you... But if anyone turns his ears to my warnings late, alas, with how much pain he will recall my words” (in me nostra Venus noctes exercet amaras, / et nullo vacuus tempore defit Amor. / hoc, moneo, vitate malum... quod si quis monitis tardas adverterit aures, / heu referet quanto verba dolore mea, Prop. 1.1.33-8). Poem 1.1 is programmatic: It sets out the speaker’s love for Cynthia and his sufferings on account of it as the main subjects of Book 1. Poem 1.22, the last poem of Book 1, redefines what pain (dolor) means to the speaker and so reprograms the book. In Poem 1.22, having spent most of the book discussing the vicissitudes of love, the speaker does not speak of love at all but rather of his Umbrian origins (Prop. 1.22.9- 10) and of the slaughter and unburied corpse of his relative at Perusia: “Let this in particular be my pain, Etruscan dust: / you have suffered the limbs of my relative to remain exposed, / you cover 653 Habinek (1998) 128 226 the poor man’s bones with no soil” (sit 654 mihi praecipue pulvis Etrusca dolor: / tu proiecta mei perpessa es membra propinqui, / tu nullo miseri contegis ossa solo), Prop. 1.22.6-8). In addition to the “Etruscan dust” (pulvis Etrusca), the speaker invokes the “Perusine graves of [his] fatherland” (Perusina…patriae…sepulcra, Prop. 1.22.3), so Poem 1.22 is generally understood to indicate that the speaker’s relative was one of those killed when Octavian immolated the senators and equestrians of Perusia in March of 40 BCE after the city surrendered to him. 655 Until this point in Book 1, love has seemed to be the speaker’s pain; now he uses a subjunctive (sit) to exhort the dust and the rest of his audience to understand that his pain is the fact that his relative’s corpse lies unburied. Poem 1.22 is a sphragis, a seal that binds the book of poems together, 656 so this exhortation suggests that the pain of love, as introduced in Poem 1.1 and returned to throughout the book, is the same for the speaker as the pain that the unburied corpse of his relative causes him. 657 The speaker links his amatory suffering with violence and politics in another part of Poem 1.1, as well. About two thirds of the way through the poem, the speaker begs his friends to bring aid for his infection and avers, “I will bravely suffer iron and savage fires; let there only be liberty to speak the things my anger wants” (fortiter et ferrum saevos patiemur et ignes, / sit modo libertas quae velit ira loqui, Prop. 1.1.27-8). Given the surrounding verses, it makes sense to read the “iron” (ferrum) and “savage fires” (saevos…ignes) of verse 27 as references to medical procedures, 654 Some editors emend sit to sic or sed (Heyworth 2007: 101). Whatever the merits of these changes, they do not substantially affect my reading of the poem: They simply replace an exhortation in the subjunctive with a constative declaration in the indicative. 655 Uggeri (2006). The Perusine War also appears in the list of Octavian/Augustus’ military victories that the speaker would discuss if he were to write epic (Prop. 2.1.29), from which he recuses himself. The first poem of Book 4 makes reference to the speaker’s Umbrian origins and the death and burial of his father (Prop. 4.1.121-7), which recalls the speaker’s discussion of his Umbrian origins and the death and non-burial of a male relative in Poem 1.22. 656 Putnam (1982) 177 657 As Nethercut (1971) 465 comments, “by concluding with two elegies which imply Caesar’s effect upon his formative years, the poet imparted symmetry to a book which had opened with a description of Cynthia’s influence in moulding his way of life.” 227 and, as Cairns notes, “The three remedies mentioned by Propertius can all be illustrated from Celsus as remedies for madness.” 658 Cairns, and Heyworth following him, also argues that the “anger” (ira) in verse 28 is “a virtual synonym of furor” 659 in verse 7, the “love-madness” 660 with which Cynthia has infected the speaker. Alongside this reading, we may see in the “iron” and “savage fires” the violence of the recent civil war, particularly because Poem 1.22 reprograms the “pain” (dolore, Prop. 1.1.38) of love that Poem 1.1 introduces, as discussed above. The libertas the speaker wishes for in verse 28 is precisely the liberty to speak freely, 661 but the terms in which he wishes for it concede that it is unattainable for him in his present condition. He asks for “liberty to speak the things [his] anger wants” (libertas quae velit ira loqui). If, with Cairns and Heyworth, we understand ira as furor, then the speaker asks to be able to say not what he wants but rather “what his frenzy demands.” 662 He is out of control and so would his prospective speech be; the libertas he wishes for is actually licentia. 663 The infection precludes libertas and free speech. By replacing the speaker’s libertas with licentia, its degenerate and slavish extreme, the infection enslaves him. The “iron” and “savage fires” of verse 27 also bring to mind branding, a way of marking a slave as a slave. 664 658 Cairns (1974) 75-6 discusses these verses, and the whole article discusses the medical imagery and terminology that pervade the poem. 659 Heyworth (2007) 11 660 Cairns (1974) 76 661 For the association of libertas with elite men’s freedom of speech in the late Republic, see Connolly (2007) 160: “When Cicero speaks among equals, he represents political participation, especially the freedom to say what one thinks, as key to the accrual of standing (dignitas), and this dignitas he associates with libertas. In a letter to Lentulus Spinther (55 BCE), for instance, Cicero laments the overthrow of the republic by the tyrannical machinations of Pompey and Caesar: “The entire system of the senate, the law courts, the whole republic, has been changed” (commutata tota ratio est senatus, iudiciorum, rei totius publicae, Fam. 1.8.3– 4). He explains that the members of the senate have lost the dignitas of deliberation (dignitas in sententiis dicendis) and the libertas they enjoyed in running political affairs (libertas in re publica).” 662 Heyworth (2007) 11 663 See above on libertas and licentia. 664 For branding of slaves see, for example, Cic. Off. 2.25: “Wretched man, who thought both a barbarian and a branded slave more faithful than his wife” (o miserum, qui fideliorem et barbarum et sigmatiam putaret quam coniugem). 228 The speaker links restrictions on his speech with violence against his body again in Poem 1.9. He warns, or perhaps gloats to, the epic poet Ponticus, “I told you love would come to you, mocker, / and your words would not be free forever” (Dicebam tibi venturos, irrisor, amores, / nec tibi perpetuo libera verba fore, Prop. 1.9.1-2). Amores may have been what Propertius’ friend and mentor, Gallus, called his elegies; the Propertian speaker calls his own love affair amores several times in Book 1 (Prop. 1.4.15, 1.7.5, 1.8.45, 1.12.5, perhaps 1.16.19, and indirectly 1.18.19) and was to begin his second book by calling his elegies amores. What has come to Ponticus, then, is elegiac love, as becomes clear in subsequent verses from the suffering the speaker predicts for Ponticus. Elegiac Love, personified as a god, has taken free speech from Ponticus and will also attack and penetrate his body with fire (igni, Prop. 1.9.17) and arrows shot into the marrow of his bones (arcum sentire medullis, Prop. 1.9.21). Love may at times seem to give Ponticus “easy wings” (faciles…alas, Prop. 1.9.23), an image of freedom, but never “without oppressing him with the other hand in turn” (ut not alterna presserit ille manu, Prop. 1.9.24). The word alterna here evokes the alternating hexameters and pentameters of elegiac couplets 665 and with them the trope that Love controls the elegiac lover’s speech even down to the meter of his verses. 666 As Heyworth writes, “the lover soars away in the epic hexameter, but then is always restricted by the alternating pentameter that the god has imposed on him.” 667 The speaker draws his warning to Ponticus from his own experience—“pain and tears have made me deservedly an expert: / would that, love set aside, I spoke as one inexperienced!” (me dolor et lacrimae merito fecere peritum: / atque utinam posito dicar amore rudis! Prop. 1.9.7-8)—so as he warns Ponticus he is also talking about the oppression of his own speech and body. 665 Heyworth (2007) 46 and cf. Cic. de Or. 3.50.193 and Arch. 10.25, for example. 666 See Chapter 4 for further discussion of this trope and particularly for the connection the trope draws between speech and the speaker’s body through the double meaning of “foot” (pes). 667 Heyworth (2007) 46 229 The speaker is never free anymore, because he is an elegiac lover. When he advises his friend Gallus that “he who will never be free with an empty chest will be able to stay happy with one girl” (is poterit felix una remanere puella, / qui numquam vacuo pectore liber erit, Prop. 1.10.29-30), he appears to be using the word liber in a narrow, emotional sense: “free with an empty chest.” But we have seen already how the speaker conflates his erotic and emotional servitude with a lack of libertas in its legal and political sense. Another sententia in Book 2 is more direct: “For indeed liberty remains for no lover; / no man will be free if he wants to love” (libertas quoniam nulli iam restat amanti, / nullus liber erit, si quis amare volet, Prop. 2.23.23-4). This thinking leads the speaker to ask Cynthia, “Have I ever seemed free to you?” (ecquandone tibi liber sum visus? Prop. 2.8.15). In contrast, a rival for Cynthia’s attention named Panthus remains free because he made love to Cynthia casually: “That handsome lover of yours has a wife! So many nights wasted; aren’t you embarrassed? Look, he sings freely; you, too credulous, lie alone, and now you’re the stuff of gossip between them, and arrogantly he says you were often at his house when he didn’t want you.” uxorem ille tuus pulcher amator habet! tot noctes periere: nihil pudet? aspice, cantat liber: tu, nimium credula, sola iaces. et nunc inter eos tu sermo es, te ille superbus dicit se invito saepe fuisse domi. (Prop. 2.21.4-8) Unlike the speaker or any other elegiac lover, Panthus has no long-term need for Cynthia, because he has a wife. 668 Furthermore, Panthus has left Cynthia, which the speaker will not manage to do until the end of Book 3, and even after that she will continue to haunt him and to remind him, “my reign in your books has been long” (longa mea in libris regna fuere tuis, Prop. 4.7.50), and then to demand that he replace what he has written about her with her own words (Prop. 4.7.77-86). 669 668 Prop. 2.7, and see James (2003) 36 and 43 for lack of interest in marriage as a generic quality of elegiac lovers. 669 See Chapter 1. 230 But Panthus did not love Cynthia elegiacally, and so he is free and can say whatever he wants about her. When the speaker does occasionally call himself liber, he is talking about how he used to be. In one poem, he says, “I was free, and it was my practice to live with an empty bed, / but though I had arranged my peace, Love tripped me up” (Liber eram et vacuo meditabar vivere lecto: / at me composita pace fefellit Amor, Prop. 2.2.1-2), and then he rhapsodizes about how Cynthia’s beauty has beguiled him ever since. Unlike his affair with Cynthia, his first love did not take away his freedom: “When the modesty of my bordered garment was put down and liberty given to know the way of love, she expertly 670 initiated my inexperienced soul through the first nights—ah me!—Lycinna, won with no presents.” ut mihi praetexti pudor est sublatus amictus et data libertas noscere amoris iter, illa rudis animos per noctes conscia primas imbuit, heu nullis capta Lycinna datis. (Prop. 3.15.3-6) These verses have been read as destabilizing the foundation of Propertius’ elegies by revealing that Cynthia was not, as the speaker claimed, “the first” (prima, Prop. 1.1.1, 12.20). 671 This reading is attractive for its poetic shock value, and it fits with Veyne’s understanding of the artifice of elegy, in which self-contradiction that reveals the artifice is a literary joke 672 and with James’s view that the elegiac lovers are insincere. 673 Another possibility is that Cynthia was the first, because the relationship with Lycinna was not elegiac love, and so the speaker “first came to know the true passion of love through Cynthia.” 674 For one thing, there is no suggestion that the speaker 670 For this translation of conscia see Heyworth (2007) 366: “Perhaps one should follow ThLL (373.43-5) in taking conscia in the sense of bene sciens, peritus, the contrast with rudes establishing this fresh development of the adjective’s meaning” (I have used Fedeli’s rudis instead of Heyworth’s rudes). 671 Yardley (1974) passim and Heyworth (2007) 366, for example. 672 Veyne (1988) 26-37 673 James (2003) 109, 133 674 Schulze (1910) 168: “die wahre Leidenschaft der Liebe lernte er erst durch Cynthia kennen.” 231 suffered from his relationship with Lycinna. He presents it as a tender, patient initiation that “filled [his] inexperienced soul” (rudis animos…imbuit 675 ). Cynthia fills the speaker, too, but she renders him “never free with an empty chest” (numquam vacuo pectore liber, Prop. 1.10.30) – that is, she enslaves and torments him by filling his chest. Verse 6 may also contrast Cynthia and Lycinna: Is the speaker lamenting that, as with Cynthia, he was never able to capture and possess Lycinna with gifts, or is he reminiscing fondly about a lover who, unlike Cynthia, demanded no gifts? 676 If it is the latter (and the speaker does make it clear elsewhere that he hates being shaken down for expensive presents 677 ), then Lycinna does not fit the character type of the elegiac mistress, who is economically dependent on expensive presents. 678 In the context of first love, the word “won” or “captured” (capta) recalls the very start of the Elegies, “Cynthia first captured wretched me with her eyes” (Cynthia prima suis miserum me cepit ocellis, Prop. 1.1.1). Cynthia exercises her power over the speaker abusively as a master over a slave, 679 whereas Lycinna used her power as a teacher for the speaker’s benefit and enjoyment. The love affair with Lycinna was not elegiac, and so the speaker could retain his liberty. The instability arising from inconsistency that Veyne perceives as characteristic of elegy is not lost in this reading, nor is the insincerity that James sees in elegiac lovers. While the relationship with Lycinna was not elegiac, it can still cast doubt on Cynthia’s primacy. In the context of erotic persuasion, ‘she was my first love, but you’re my first true love’ sounds glib and 675 The word for “initiated” (imbuit) has to do, in its basic sense, with liquid soaking into something (TLL 7.1.427.31-42). 676 The interjection heu could express regret for his wasted efforts or the bittersweet realization that his past with Lycinna was happier than his present with Cynthia. 677 Prop. 1.2 and 4.5, for example, and see James (2003) 73-107. 678 James (2003) 73-107 and Prop. 4.5 679 See above, and note that this presentation of the relationship is a misogynist rhetorical maneuver by the speaker (Habinek 1998: 128). 232 not especially convincing, hence Cynthia’s jealousy and the speaker’s haste to assure her that his relationship with Lycinna is in the past: “It’s been three years – not much less. / I scarcely remember ten words that passed between us. / Your love has buried them all” (tertius (haud multo minus est) cum ducitur annus, / vix memini nobis verba coisse decem. / cuncta tuus sepelivit amor, Prop. 3.15.8-9). Perceived Political Suppression In some poems, the speaker connects limitations on his speech and threats to his body with political events. As seen above, the speaker equates the “pain” (dolor) of elegiac love, which oppresses his speech and attacks his body, with the pain he feels over the unburied corpse of his relative, one of the senators and equestrians Octavian immolated at Perusia, in the final poem of Book 1. He begins Book 2 with a poem that associates his words with his body, 680 and as he lists the limitations on his poetry, one of the subjects he says he cannot sing about is “the overturned hearths of the ancient Etruscan race” (eversos focos antiquae gentis Etruscae, Prop. 2.1.29). Under the power of elegiac love, the speaker can only discuss the devastation that Octavian visited on his homeland with the land confiscations of 40 BCE through preterition. 681 Similarly, in Poem 4.1, the astrologer Horos ties the Propertian speaker’s personal history to restrictions on his speech. He begins his prophecy of the speaker’s life by advising him to “begin to prepare for fresh tears” (Prop. 4.1.120), evoking the dolor of Book 1 and its dual meanings. Horos then reinforces the association by referring to the speaker’s Umbrian origins (Prop. 4.1.121-6), which the speaker connects with his dolor in Poem 1.22. Next Horos reminds the speaker that at a painfully young age he “gathered the bones of [his] father” (ossaque legisti 680 See Chapter 4. 681 cf. Breed (2010) 233 233 non illa aetate legenda / patris, Prop. 4.1.127-8), which recalls the unburied bones of the speaker’s relative in Poem 1.22 and the pain the speaker felt over them. Horos adds that the speaker was “forced into a meager home” (in tenuis cogeris ipse lares, Prop. 4.1.128), because “the grim measuring rod stole [his] carefully cultivated wealth” (abstulit excultas pertica tristis opes, Prop. 4.1.130). This is another reference to Octavian’s land confiscations, 682 and just as the speaker could only discuss the confiscations through preterition in Poem 2.1, here he has Horos broach the subject. 683 Since a Roman man’s household was understood as an extension of his person, the confiscations were an incursion against the speaker’s person; because he had no recourse or remedy, they made him slavish rather than properly free. 684 Worse yet, the speaker was still a boy when they occurred (Prop. 4.1.129-32), so they stunted his development into a free man who could speak freely. He “assumed the free man’s toga before [his] mother’s gods” (matris et ante deos libera sumpta toga, Prop. 4.1.132) rather than his father’s, 685 and “then Apollo,” with whom Augustus aligned himself, 686 “dictated a little of his song to [the speaker] / and forbade [him] to thunder words in the crazed Forum” (tum tibi pauca suo de carmine dictat Apollo / et vetat insano verba tonare Foro, Prop. 4.1.133-4). The word “then” (tum) indicates that Horos is still describing past events, but the narrative present tense of the verbs suggests that Apollo’s prohibition is ongoing. Horos thus derives the speaker’s limited ability to speak from Augustus’ past actions and suggests that Augustus continues to limit his speech. 682 Richardson (1977) 422; Goold (1990) ad loc.; Hutchinson (2006) 83 683 Within the bounds of Poem 4.1, Horos’ speech is independent, but as part of the speaker’s poetry book it ultimately belongs to the speaker. 684 Cf. Janan (2001) 46 citing Tullus’ question at the start at Prop. 1.22.1 (“What sort of man are you, and where does your family come from…?,” qualis et unde genus…?) “as evidence that property ownership metaphorically distills social identity.” See also Janan (2007) 40: “However, the forceful appropriation of a persions region or estate could deal a deadly blow to a person’s potential for leaving a legacy. Once the connecting line between a man and his property was fractured, his standing in the community and his capacity for maintaining his status rapidly disintegrated.” 685 Cf. Ov. Fast. 3.771ff. 686 Zanker (1988) and see below. 234 In Poem 2.31, the speaker couches similar criticism of Augustus in praise for the princeps and the new temple complex of Apollo on the Palatine. He begins the poem in great excitement about the newly opened Portico of the Danaids, and he flatters Augustus by comparing him with Apollo: “This marble seemed to me, at least, more beautiful than Phoebus himself / [and seemed] to sing with a silent lyre” (hic equidem Phoebo visus mihi pulchrior ipso / marmoreus tacita carmen hiare lyra, Prop. 2.31.5-6 687 ). Barchiesi notes that “the iconography of Apollo had been converging with the official image of Octavian long before 28 BCE,” 688 and Heyworth adduces evidence from Servius and pseudo-Acro: “Augustus, to whom a likeness was made with all the statues of Apollo” (Augustum, cui simulacrum factum est cum Apollinis cunctis imaginibus, Serv. ad Ecl. 4.10) and “Caesar had put a statue of himself in the library with the attire and posture of Apollo” (Caesar in bibliotheca statuam sibi posuerat habitu ac statu Apollinis, ps.-Acro ad Hor. Ep. 1.3.17). 689 The flattery, then, is in the claim that this statue of Apollo, which resembled Augustus, seemed more beautiful than the proverbially beautiful god himself. But the praise sours when we recognize an allusion to Callimachus. Verse 6 alludes to verse 24 of the Hymn to Apollo: “and indeed the tearful rock puts off its pain, / the wet stone which is set in Phrygia, / marble in place of a woman gaping some pitiable utterance” (καὶ μὲν ὁ δακρυόεις ἀναβάλλεται ἄλγεα πέτρος, / ὅστις ἐνὶ Φρυγίηι διερὸς λίθος ἐστήρικται, / μάρμαρον ἀντὶ γυναικὸς ὀϊζυρόν τι χανούσης, Callim. Hymns 2.22-4). Heyworth writes, “Propertius has taken from a Greek hymn to the same god the unusual construction hiare (=χάσκειν) + internal accusative, and then confirmed the allusion by placing marmoreus at the start of his verse to echo 687 I find Heyworth’s (1994) 56-9 solution to the textual and interpretive problems of these verses (positing a lacuna before them) persuasive. See also his note on equidem here. 688 Barchiesi (2005) 284-5 689 Heyworth (1994) 58 235 μάρμαρον.” 690 These verses are about Niobe, whom Apollo and Artemis punished horribly for the hubris of comparing herself and her 7 sons and 7 daughters favorably with Leto and the twin gods: Apollo and Artemis slew Niobe’s children and in some versions her husband, and Niobe fled home to Phrygia, where she was turned to stone, which continued to weep just as she had in life. Niobe suffered violent punishment for her speech at the hands of the god with whom Augustus aligned himself, and verses 17-22 of Callimachus’ hymn indicate “that even Niobe has to stop her lament when the festival of Apollo comes,” 691 so she is silenced yet again. The Propertian speaker says that one door of the portico “mourned” (maerebat, Prop. 2.31.14) the deaths of Niobe and her children. 692 Barchiesi comments, “the allusion to the Greek model in literature complicates the reading of the Greek art in the temple: one quickly learns how to read Apollo the killer in Apollo the musician, how to listen to imaginary music but also to echoes of suffering and repression.” 693 The Propertian speaker thus suggests, amid the celebration of the temple, that Augustus’ iconography implicitly threatens violent punishment for anyone who might speak against him. Poem 3.11 expresses apprehension about Augustus and his regime even as it praises him. As discussed in Chapter 4, the speaker begins the poem by equating Cynthia’s domination of his body with her domination of his poetry. He then gives examples of mythical and historical women who held power over men: Medea, Penthesilea, Omphale, Semiramis, and Cleopatra. While Cleopatra, in his telling, succeeded in controlling individual Roman men, the greater horror is that “she demanded Rome’s / walls and senators be surrendered to her reign” (…Romana poposcit / moenia et addictos in sua regna Patres, Prop. 3.11.31-2). The poem celebrates her failure to 690 Heyworth (1994) 57 691 Barchiesi (2005) 285 692 In the phrase funera Tantalidos, the plural funera may mean Niobe’s death, corpse, or funeral rites, or it may mean her (plural) deaths, corpses, or funerals, i.e., the deaths, corpses, or funerals of her children. 693 Barchiesi (2005) 285 236 achieve this, but even as the speaker praises Octavian’s victory he reminds his audience that Octavian, now Augustus, succeeded where Cleopatra failed. Some features of the political dominance Cleopatra allegedly desired may also reflect anxieties about Augustus’ regime: repression of the Senate (32, 46); politically motivated executions of high-born citizen men such as Pompey (35); the monopolizing of opportunities for gloria such as triumphs (35); and the worry of a return to tyranny (47), with the image of Tarquin’s axes (secures) doing double duty – outside the Pomerium an ax was added to fasces, a symbol of imperium, and securis can also mean an executioner’s ax. Then in verses 55-6 the speaker tells how an inebriated Cleopatra offered two interpretations of Rome’s situation in one: “ ‘This woman, Rome, was not to be feared when you had so great a citizen!’ / her tongue, which was entombed by perpetual wine, had declared.” (‘Non haec, Roma, fuit tanto tibi cive verenda!’ / dixerat assiduo lingua sepulta mero, Prop. 3.11.55- 6 694 ). Verse 55 can be understood to mean that, with Octavian as its defender, Rome had no need to fear Cleopatra, but it can also be understood to mean that Rome should have feared not Cleopatra but rather its own greatest citizen. In some ancient portrayals, moreover, Cleopatra was like both an elegiac mistress and an elegiac lover. She and Antony were infamous for nequitia, “expending and squandering on pleasure the costliest outlay, as Antiphon put it, time” (ἀναλίσκειν καὶ καθηδυπαθεῖν τὸ πολυτελέστατον, ὡς Ἀντιφῶν εἶπεν, ἀνάλωμα, τὸν χρόνον, Plut. Ant. 28.1) and indulging in drunken feasts every day with their “Society of Inimitable Livers” (σύνοδος ἀμιμητοβίων, Plut. Ant. 28.2). 695 In Book 2 the speaker describes their love affair as elegiac, because it compelled Antony to forsake his soldiers and his duty to them at Actium and flee to Egypt to join Cleopatra 694 Following Heyworth, I prefer haec given by ς to hoc, which Fedeli prints, in verse 55. Both Fedeli (ad loc.) and Heyworth (2007) 341 replace dixit et with Housman’s dixerat in verse 56. 695 See also Zanker (1988) 46ff. on Antony’s alignment with Dionysus. 237 (Prop. 2.16.39-40). Here, in Poem 3.11, Cleopatra was drunk on wine, as the speaker likes to be, 696 and her tongue was entombed (lingua sepulta), which matches the restrictions that the speaker perceives on his own speech. The image of an entombed tongue may also have brought to mind Cicero’s tongue, if the story that Cassius Dio (47.8.4) reports—that Fulvia vindictively stabbed Cicero’s tongue with her hairpins before the head and right hand were mounted on the Rostra— had currency when Propertius was composing. In the rest of the poem, the speaker praises Augustus but in the process says that he occludes the great heroes of the Republic. As seen above, the speaker tells Cynthia five poems later, “your love has buried them all” (cuncta tuus sepelivit amor, Prop. 3.15.9) – that is, all the words he spoke in his previous love affair. The speaker begins Poem 3.11 by introducing as his topic Cynthia’s power over his body and words. To explain it, he then gives examples of women who dominated men, but when he turns to Cleopatra an example, he shifts from her failed bid for dominance over all Rome to Augustus’ successful bid. Conclusion The speaker of Propertius’ Elegies expresses considerable concern over the vulnerability of his body and the limitations on his speech. In the Propertian speaker’s fictional world, it is Love who attacks his body, restricts his speech, and enslaves him. In some poems, though, he points to Octavian/Augustus as a source of his repression as well. The question then arises, to what degree are Love and elegiac love affairs in Propertius’ Elegies allegories for Augustus and his regime? I hope to address this question more thoroughly in future research, in order to integrate historical sources and scholarship about Augustan politics into my discussion of pragmatics of authorship. For now, I will return to my question from the beginning of the chapter: If the text is the body or 696 Prop. 1.3.9ff.; 2.15.42-54; 2.34.59 (a hangover); 3.5.21; 3.10.21; 3.17; 4.6.73-6, 85; and 4.8.29-48 238 part of the bodies of the fictive Propertian speaker, a real person who performs Propertius’ poetry, and the historical Propertius himself, what do the Elegies say or imply that it means for their bodies when another user of the Elegies changes or destroys part of the text? The readings in this chapter have explored how the speaker of the Elegies thinks through several challenges to his identity as an elite, citizen man by comparing the literary vulnerability of his text and his body with the political precarity of his speech and his body. What these comparisons reveal will be the subject of my Conclusion. 239 Conclusion: The Value of Being Worthless The speaker of the Elegies wallows in worthlessness. The word for this condition is nequitia, and its locus classicus in Propertius’ Elegies is Poem 1.6. Here, the speaker recuses himself from service in the administration of his friend Tullus, who is leaving to take up a post as a provincial governor. The speaker cannot join Tullus, he says, because Cynthia is passionately and furiously entreating him to stay with her in Rome (Prop. 1.6.5-18). Having given his reason, the speaker contrasts Tullus’ future with his own: “Attempt to surpass the axes your uncle earned and bring back the old laws to forgetful allies. For your youth never tarried for love, and your care was always for your fatherland at war; and to you may that boy never bring sufferings like mine and all the things known to my tears! Let me, whom Fortune ever willed to lie down, give up this life to utter worthlessness. Many have gladly perished in drawn-out love, in whose number may the earth also cover me. Not suited to praise was I born, nor to arms: this is the service the fates wish me to undergo.” tu patrui meritas conare anteire securis et vetera oblitis iura refer sociis. nam tua non aetas umquam cessavit amori, semper et armatae cura fuit patriae; et tibi non umquam nostros puer iste labores afferat et lacrimis omnia nota meis! me sine, quem semper voluit Fortuna iacere, hanc animam extremae reddere nequitiae. 697 multi longinquo periere in amore libenter, in quorum numero me quoque terra tegat. non ego sum laudi, non natus idoneus armis: hanc me militiam fata subire volunt. (Prop. 1.6.5-18) 697 Several commenters have proposed alternate readings of this verse. I follow Shackleton Bailey (1954), Fedeli’s Teubner text, and Richardson (1977) in keeping the reading of the archetype, because I do not agree with Skutsch (1973) 318 and Heyworth (2007) 25 that the sense of the verse cannot be “to give up this life to utter worthlessness.” The verse need not be talking about an immediate, literal death, as Heyworth suggests that this reading entails – the speaker is giving up his life as he knows it and his social standing for a love affair with Cynthia, and this social death that will precede the speaker’s literal death is the point of the word nequitia, as the next couplet shows. 240 Tullus is setting off on a traditional career trajectory for a young man of the equestrian order, possibly with senatorial ambitions. 698 Never having allowed love to side-track him, he will now try to prove himself as a military leader, administrator, and politician and even to surpass the achievements of an illustrious uncle. The speaker, in contrast, will lie down: He will spend his life prone or prostrate. Whereas Tullus will position himself to conquer, the speaker will position himself to be conquered. Neither winning praise nor bearing arms, he will neglect what Roman society traditionally expected of a young man of his class. For the sake of his love affair with Cynthia, he will give himself up to idleness and wantonness, to being conquered, 699 abject, and of no use to his community. While Poem 1.1 does not contain the word nequitia, it describes how the disease of love has made the speaker worthless as an elite, Roman man. As discussed in Chapter 5, the disease of love has taken away his ability to speak freely, a crucial capacity of a Roman man. Gunderson proposes that, in Cicero’s letters from exile, Cicero’s health equals the health of his marriage equals the health of his social and political status, 700 and Varhelyi argues that “on Cicero’s reading, the concept of salus was closer to public welfare than to the health of an individual and, as such, it was associated with the achievements of the statesman who could secure it.” 701 If salus is associated with politically active men, then the illness of the Propertian speaker can be associated with his abdication of a political career. The illness has also isolated the speaker from his social peers and their communal identity- building. Varhelyi explains that attendance at the sickbeds of friends was part of how elite men 698 See Keith (2008) 143-5 for discussion. 699 As Keith (2008) 144 notes, the description of the speaker’s state as ‘military service’ (militia), along with the trope of militia amoris, is “heavily ironized.” 700 Gunderson (2007) 7-13. See below for further discussion of Gunderson’s arguments. 701 Varhelyi (2010) 86 citing Cic. Leg. 3.8 241 performed their status as such in the early Empire 702 and cites instances of prayers for health (vota pro salute) in Propertius as evidence for this practice in the Augustan period: “these my prayers and offerings, made on behalf of your health” (haec mihi vota tuam propter suscepta salutem, Prop. 2.9.25), and the prayer in Prop. 2.28. 703 Varhelyi writes, “However scattered, the evidence presented here for sharing health concerns in religious terms through mutual prayers is clear in its emphasis on the peer group and on the role of this group in shaping the self-understanding and behavior of individuals.” 704 The speaker’s illness has brought his friends to his bedside and involved them in trying to cure him (Prop. 1.1.25-30), but they are “too late” (sero, Prop. 1.1.25). Their communal identity-building has failed, and he tells anyone who has managed to stay safe in love to “stay home” (vos remanete) and watch out for their continued safety (Prop. 1.1.31-2). He reveals in Poem 1.5 that he is too ill to hold up his end of the reciprocity: “I will not be able to bring comfort to you when you ask, since there is no medicine for my own sickness” (non ego tum potero solacia ferre roganti, / cum mihi nulla mei sit medicina mali, Prop. 1.5.27-8). The speaker’s prayers for health in Poems 2.9 and 2.28 are for Cynthia. Love has redirected the speaker to focus on Cynthia instead of participating fully in homosocial identity-building with his peers. Modern readers have noted that the speaker engages in homosocial bonding and identity- building in poems such as 1.4, 1.9, 1.10, 2.34, and 3.22. 705 In these poems, it is through the media of Cynthia and his licentious poetry that the speaker bonds with his peers and joins them in constructing their identities as elite men. Some of the others may also be licentious: Ponticus in Poem 1.7, for example, will suffer the same fate in love as the speaker, and Bassus in Poem 1.4 “was accustomed to eagerly pursue sordid affairs” (consectari autem solebat res sordidas, Sen. 702 Varhelyi (2010) 78-90 703 Varhelyi (2010) 85 704 Varhelyi (2010) 90 705 Miller (2004) and Keith (2008) Chapters 5 and 6, for example. 242 Contr. 10.1.13) in his speech, if he is Seneca the Elder’s Julius Bassus. This must be a nontraditional kind of elite masculinity, then, as it is predicated on licentia and nequitia. The speaker links his nequitia directly to his poetry in Poem 2.24. In this poem, he has become the talk of the town (fabula, Prop. 2.24.1), and Cynthia has been read all over the forum (toto Cynthia lecta foro, Prop 2.24.2). The speaker admits he is embarrassed (Prop. 2.24.3) but then thinks better of it and lays claim to shamelessness: “A gentleman must keep quiet about either his respectability or his love life” (aut pudor ingenuis aut reticendus amor, Prop. 2.24.4). The speaker has not kept quiet about the latter and is not about to start. Cynthia, their bad romance, and the speaker’s poetry are the source and substance of his turpitude and ill-repute (Prop. 2.24.5- 8). On account of them, the speaker has come to be called “the wellspring of worthlessness” (nequitiae…caput, Prop. 2.24.6). Nequitia, then, is a standpoint from which the Propertian speaker compares the literary vulnerability of his speech and his body with their political precarity, as he perceives it. The perception of these vulnerabilities ran counter to traditional Roman ideas about manliness (virtus 706 ) and so rendered the traditional formulation of elite masculinity incoherent for the speaker in his circumstances. Practices of literary reception and composition that made Roman texts vulnerable presented analogous challenges to the manliness of the speaker as an author. He 706 See discussion of virtus and especially the legal inviolability of the body of a vir in Chapter 5. In addition, on virtus as the idealized, hegemonic masculinity in Roman society, see for example Cic. Tusc. 2.43, and see discussion of this idea in Edwards (1993), Williams (1999) 125-224, Skinner (2005) 195, and McDonnell (2006). On virtus and social class, see for example the elogium of L. Cornelius Scipio (CIL I 2 .8-9 [CLE 6]) and Cato’s De Agri Cultura, which pairs bonus with vir five times and never with homo, and see discussion in Santoro L’Hoir (1992) 1- 2 and 9ff.: “The nouns vir and femina came to signify the upper classes, while homo and mulier applied to everyone else.” On the association of virtus with high-level political activity and effectiveness, see for example a famous fragment of Ennius—“the Roman state stands on ancient customs and on viri” (moribus antiquis res stat Romana virisque, Enn. Ann. frag. 5.**i)—and Boyle’s (2006) 59 appraisal of Ennius’ influence: “Like his predecessors Ennius wrote an epic, in fact, the Roman epic of the republican period, Annales, and, although Ennius must have dominated the tragic stage for the first three decades of the second century..., it was primarily as poet of Annales that later Roman writers celebrated him... Written by a south Italian immigrant, Ennius’ epic yet played a constitutive part in defining for the Roman elite of the next 150 years what it was to be a Roman.” 243 thus presents nequitia as a less obviously incoherent alternative to republican virtus, much like he manages the literary vulnerability of his text and his body by incorporating it into his poetics. The vulnerable corpus of Propertius was an ad hoc technology of authorship and gender, to be adapted, adopted, or discarded as needed in the future. 244 Bibliography Adams, J. N. 1999. “The Poets of Bu Njem: Language, Culture and the Centurionate.” JRS 89: 109-34. Ando, C. 2011. Law, Language and Empire in the Roman Tradition. Philadelphia. Arena, V. 2012. Libertas and the Practice of Politics in Late Republican Rome. Cambridge, UK. Axelson, B. 1960. “Lygdamus und Ovid.” Eranos 58: 93-111. Barber, E. A. 1960. Sexti Properti Carmina. Oxford. Barchiesi, A. 1988 [1997]. “Ovid the Censor.” AJAH 13: 96-105. -----------. 1993. “Future Reflexive: Two Modes of Allusion and Ovid’s Heroides.” HSCPh 95: 333-65. -----------. 2001. Speaking Volumes. Narrative and Intertext in Ovid and Other Latin Poets. Eds. and Trs. M. Fox and S. Marchesi. London. -----------. 2005. “Learned Eyes: Poets, Viewers, Image Makers.” The Cambridge Companion to the Age of Augustus. Ed. K. Galinksy. Cambridge, UK. 281-305. Barwick, K. 1922. Remmius Palaemon und die römische Ars grammatica. Leipzig. Bejarano Sánchez, V. 1990. “Propercio en Carisio.” Anuari de filologia. Secció D, Studia graeca et latina 13.1: 25-32. Bell, C. 1992. Ritual Theory, Ritual Practice. Oxford. Ben Abdallah, Z. B. et al. 2005. “Carmina Latina Inedita Ammaedarae.” Zeitschrift für Papyrologie und Epigraphik 82: 89-113. Benefiel, R. R. 2010. “Dialogues of Ancient Graffiti in the House of Maius Castricius in Pompeii.” American Journal of Archaeology 144.1: 59-101. Bergmann, B. 1999. “Rhythms of recognition: Mythological encounters in Roman landscape painting.” Im Spiegel des Mythos: Bilderwelt und Lebenswelt. Eds. F. de Angelis and S. Muth. Wiesbaden. 81-107. -----------. 2007. “A painted garland: weaving words and images in the House of the Epigrams in Pompeii.” Art and Inscriptions in the Ancient World. Eds. Z. Newby and R. Leader- Newby. Cambridge, UK. 60-101. Bettenworth, A. 2016. ‘hoc satis in titulo’: Studien zu den Inschriften in der römischen Elegie. Münster. Bingham, S. 2013. The Praetorian Guard: A History of Rome’s Elite Special Forces. New York. Birley, A. R. 2006 [2014]. “Cassius.” Brill’s New Pauly. Eds. H. Cancik and H. Schneider. 15 March 2014. <http://www.brillonline.nl.libproxy.usc.edu/entries/brill-s-new- pauly/cassius-e228030> Bloomer, W. M. 1997. “A preface to the history of declamation: whose speech? whose history?” The Roman Cultural Revolution. Eds. T. Habinek and A. Schiesaro. Cambridge, UK. 199-215. -----------. 2011. “Transit admiratio: memoria, invidia, and the historian.” Velleius Paterculus: Making History. Ed. E. Cowan. Swansea. 93-119. Bonnet, G. 2009. “Remmius Palémon et la catégorie des adjectifs : Le sens de la leçon partio dans le texte de l’Ars de Charisius (146, 29 et 147, 1 B).” Revue de Philologie, de Littérature et d’Histoire Anciennes 83.1: 21-30. Bonadeo, A. 2004. “Tra ripetizione fonica e memoria poetica: l’eco, da imago vocis a icona intertestuale.” La citation dans l’antiquité. Ed. Catherine Darbo-Peschanski. Grenoble. 245-56. 245 Borbonus, D. 2014. Columbarium Tombs and Collective Identity in Augustan Rome. Cambridge, UK. Bormann. Inscr. Lat. Nov. unknown, cited at CLE 389. Boucher, J.-P. 1980. Études sur Properce. Problèmes d’inspiration et d’art. Paris. Braund, S. M. 2004. “Libertas or Licentia? Freedom and Criticism in Roman Satire.” Free Speech in Classical Antiquity. Eds. Sluiter, I and R. M. Rosen. Leiden. 409-28. Breed, B. W. 2010. “Propertius on Not Writing about Civil Wars.” Citizens of Discord. eds. B. W. Breed, C. Damon, and A. Rossi. Oxford: 233-48. Bremmer, J. 2006 [2016]. “Linus.” Brill’s New Pauly. Eds. H. Cancik and H. Schneider. 26 September 2016. <http://dx.doi.org.libproxy3.usc.edu/10.1163/15 74- 9347_bnp_e706110> Brizio, E. 1876. Pitture e sepolcri scoperti sull’Esquilino della Campagnia Fondaria Italiana. Rome. Brugnoli, G and F. Stok. 1997. Vitae Vergilianae Antiquae. Rome. Brunt, P. A. 1984. “The Role of the Senate in the Augustan Regime.” CQ 34.2: 423-44. Büchner, K. 1965. “Die Elegien des Lygdamus.” Hermes 93: 65-112. Butler, Judith. 1990. Gender Trouble: Feminism and the Subversion of Identity. New York. ————. 1993. Bodies That Matter. New York. Butler, S. 2011. The Matter of the Page: Essays in Search of Ancient and Medieval Authors. Madison. Butrica, J. L. 1981. “The Earliest Inaccurate Citation of Propertius.” AJP 102.3: 327-9. -----------. 1986. The Manuscript Tradition of Propertius. Toronto. -----------. 1993. “An Edition of the Consolatio ad Liviam: Henk Schoonhoven (ed): The Pseudo- Ovidian Ad Liviam de Morte Drusi (Consolatio ad Liviam, Epicedium Drusi): A Critical Text with Introduction and Commentary.” The Classical Review 43.2: 265-7. -----------.1996. “The ‘Amores’ of Propertius: Unity and Structure in Books 2-4.” Illinois Classical Studies 21: 87-158. -----------. 1997. “Editing Propertius.” CQ 47.1: 176-208. -----------. 2006. “The Transmission of the Text of Propertius.” Brill’s Companion to Propertius. Ed. Hans-Christian Günther. Leiden. 25-43. Cagnat, R. 1889. “Sur les manuels professionels des graveurs d’inscriptions romaines.” Revue de philologie 13: 51-65. Cairns, F. 1972 [2007]. Generic Composition in Greek and Roman Poetry. Ann Arbor. -----------. 1974. “Some Observations on Propertius 1.1.” CQ 24.1: 94-110. -----------. 1983. “Propertius 1,4 and 1,5 and the ‘Gallus’ of the Monobiblos.’ Papers of the Liverpool Latin Seminar 4.61-103. -----------. 2006a. Sextus Propertius: The Augustan Elegist. Cambridge, UK. -----------. 2006b. “Propertius and the Origins of Latin Love-Elegy.” Brill’s Companion to Propertius. Ed. Hans-Christian Günther. Leiden. 69-95. Calame, C. 2004. “Mode de la citation et critique de l’intetextualité : jeux énonciatifs et pragmatiques dans les Théognidea.” La citation dans l’antiquité. Ed. C. Darbo- Peschanski. Grenobles. 221-41. Caldelli, M. L. and C. Ricci. 1999. Monumentum familiae Statiliorum: Un riesame. Rome. Cameron, A. 1995. Callimachus and His Critics. Princeton. Cecere, M. G. G. 1994. “Il sepolcro della catella Aeolis.” Zeitschrift für Papyrologie und Epigraphik 100: 413-21. 246 Clarke, M. L. 1953 [1996]. Rhetoric at Rome: A Historical Survey. London. Clausen, W. 1994 [2003]. Virgil: Eclogues, With and Introduction and Commentary. Oxford. Cogitore, I. 2009. “Libertas et ses enjeux, entre littérature et politique.” Originally in Le poète irrévérencieux. Accessed online on 1 September 2016 at La Réserve: Livraison septembre 2015 1.2. <http://ouvroir-litt-arts.univ-grenoble- alpes.fr:8080/revues/reserve//revues/ reserve/122-libertas-et-ses-enjeux-entre-litterature- et-politique> Connolly, J. 2007. The State of Speech: Rhetoric and Political Thought in Ancient Rome. Princeton. Conte, G. B. 1986. The Rhetoric of Imitation: Genre and Memory in Virgil and Other Latin Poets. Ithaca. -----------. 1994a [1999]. Latin Literature: A History. Tr. J. B. Sodolow. Rev. D. Fowler and G. W. Most. Baltimore. -----------. 1994b. Genres and Readers. Tr. G. W. Most. Baltimore. Cooley, A. E. 2012. The Cambridge Manual of Latin Epigraphy. Cambridge, UK. Corbier, M. 2006. Donner à voir, donner à lire: mémoire et communication dans la Rome ancienne. Paris. Coutelle, E. 2015. Properce, Élégies, livre IV. Collection Latomus 348. Brussels. Cugusi, P. 2003. “ ‘Doppioni’ e ‘ritornelli’ epigrafici.” Bolletino di Studi Latini 33: 449-66. Cutolo, P. 1990. “The Genre of the Copa.” Proceedings of the Leeds Latin Seminar 6: 115-19. Darbo-Peschanski, C. 2004. “Les citations grecques et romaines.” La citation dans l’antiquité. Ed. Catherine Darbo-Peschanski. Grenoble. 9-21. Dalzell, A. 1955. “C. Asinius Pollio and the Early History of Public Recitation at Rome.” Hermathena 86: 20-28. de Libero, L. and D. Klose. 2006 [2017]. “Tresviri.” Brill’s New Pauly. Eds. H. Cancik and H. Schneider. Brill Online. Reference. University of Southern California. 4 January 2017. <http://dx.doi.org.libproxy1.usc.edu/10.1163/1574-9347_bnp_e1219840> Debray, R. 1996. Media Manifestos: On the Technological Transmission of Cultural Forms. Tr. Eric Rauth. New York. Dinter, M. 2011. “Inscriptional Intermediality in Latin Elegy.” Latin Elegy and Hellenistic Epigram: A Tale of Two Genres at Rome. Ed. A. Keith. Newcastle: 7-18. -----------. 2013. “Inscriptional Intermediality in Latin Literature.” Inscriptions and their Uses in Greek and Latin Literature. Eds. P. Liddel and P. Low. Oxford: 303-16. Dominik, W. J. 1993 [1996]. “From Greece to Rome: Ennius’ Annales.” Roman Epic. Ed. A. J. Boyle. London: 37-58. Dover, K. 1997 [2002]. Aristophanes: Frogs, Edited with Introduction and Commentary. Oxford. Dufallo, B. 2007. The Ghosts of the Past: Latin Literature, the Dead, and Rome’s Transition to a Principate. Columbus, OH. Dupont, Florence. 1997. “Recitatio and the reorganization of the space of public discourse.” Tr. Thomas Habinek and André Lardinois. The Roman Cultural Revolution. Eds. Thomas Habinek and Alessandro Schiesaro. Cambridge. 44-59. Ebner, C. 2006 [2014]. "Latrocinium." Brill’s New Pauly. Eds. H. Cancik and H. Schneider. 26 September 2014. <http://www.brillonline.nl.libproxy.usc.edu/entries/brill-s-new- pauly/latrocinium-e632540> 247 Eck, W. 2006 [2017]. “Haterius.” Brill’s New Pauly. Eds. H. Cancik and H. Schneider. Brill Online. Reference. University of Southern California. 10 March 2017. <http://dx.doi.org.libproxy2.usc.edu/10.1163/1574-9347_bnp_e504130> Eco, U. 1989. The Open Work. Tr. A Cancogni. Cambridge, MA. Eder, W. 2006 [2013]. “Social Wars.” Brill’s New Pauly. Eds. H. Cancik and H. Schneider. Brill Online. Reference. University of Southern California. 18 October 2013. <http://www.brillonline.nl.libproxy.usc.edu/entries/brill-s-new-pauly/social-wars- e221570> Edmunds, L. 2001. Intertextuality and the reading of Roman Poetry. Baltimore. Epigraphic Database Roma. 2011 [2015]. Entry for CIL VI 6592, Entry Number EDR111742. entry written by Sara Meloni. Accessed April 3, 2016. <http://www.edr-edr.it/edr_ programmi/res_complex_comune.php?id_nr=EDR111742> Farrell, J. 2004. “Ovid’s Virgilian career.” Re-presenting Virgil, special issue in honor of Michael C. J. Putnam, Eds. G. W. Most and S. Spence. Materiali e discussioni per l’analisi dei testi classici 52: 41-55. -----------. 2005. “Intention and Intertext.” Phoenix 59: 98-111. -----------. 2009. “The impermanent text in Catullus and other Roman poets.” Ancient Literacies: The Culture of Reading in Greece and Rome. Eds. W. A. Johnson and H. N. Parker. Oxford: 164-85. Fedeli, P. 1965. Properzio. Elegie. Libro IV, testo critico e commento. Bari. -----------.1984. Sexti Properti Elegiarum Libri IV. Stuttgart. -----------. 1985. Properzio: Il Libro Terzo delle Elegie: Introduzione testo e commento. Bari. -----------. 1989. “Il poeta lapicida.” Historia Testis: Mélanges d’épigraphie, d’histoire ancienne et de philologie offerts à Tadeusz Zawadzki. Eds. M. Piérart and O. Curty. Fribourg: 79- 96. Feldherr, A. 1998. Spectacle and Society in Livy’s History. Berkeley. Flaschenriem, B. L. 1998. “Speaking of women: ‘female voice’ in Propertius.” Helios 25.1: 49- 64. Fitzgerald, W. 1995 [1999]. Catullan Provocations: Lyric Poetry and the Drama of Position. Berkeley. Fredrick, D. 1998. “Reading Broken Skin: Violence in Roman Elegy.” Roman Sexualities. Eds. J. P. Hallett and M. B. Skinner. Princeton. 172-93. Gamel, M. K. 1998. “Reading as a Man: Performance and Gender in Roman Elegy.” Helios 25.1: 79-95. Gardner, H. 2013. Gendering Time in Augustan Love Elegy. Oxford. Geertz, C. 1973. The Interpretation of Cultures. New York. Gibson, R. 2014. “Starting with the index in Pliny.” The Roman Paratext: Frame, Texts, Readers. Ed. L. Jansen. Cambridge, UK. 33-55. Glock, A. 2006 [2015]. "Panegyricus Messallae." Brill’s New Pauly. Eds. H. Cancik and He. Schneider. 3 September 2015 <http://referenceworks.brillonline.com.libproxy2.usc.edu /entries/brill-s-new-pauly/panegyricus-messallae-e905760> Goldberg, S. M. 1993 [1996]. “Saturnian Epic: Livius and Naevius.” Roman Epic. Ed. A. J. Boyle. London. 19-36. Goold, G. P. 1989. “Problems in Editing Propertius.” Editing Greek and Latin Texts. Ed. J. N. Grant. New York. 97-119. -----------. 1990 [1999]. Propertius: Elegies, Edited and Translated. Cambridge, MA. 248 Gunderson, E. 2007. “S.V.B.; E.V.” Classical Antiquity 26.1: 1-48. Günther, H.-C. 1997. Quaestiones Propertianae. Leiden. Gurd, S. A. 2011. Work in Progress: Literary Revision in Classical Antiquity. New York Gutzwiller, K. J. 1998. Poetic Garlands: Hellenistic Epigrams in Context. Berkeley. Habinek, T. N. 1994. “Ideology for an Empire in the Prefaces to Cicero’s Dialogues.” Ramus 23: 55-67. -----------. 1998. The Politics of Latin Literature: Writing, Identity, and Empire in Ancient Rome. Princeton. -----------. 2005. The World of Roman Song: From Ritualized Speech to Social Order. Baltimore. -----------. 2011. “Fixity and Fluidity in Latin Literature and Roman Identity.” an unpublished paper delivered at the Centre for Canon and Identity Formation, University of Copenhagen. May 20-21, 2011. -----------. 2016. “At the Threshold of Representation: Cremation and Cremated Remains in Classical Latin Literature.” Classical Antiquity 35.1: 1-44. Hardie, P. 2002. Ovid’s Poetics of Illusion. Cambridge, UK. Heerink, M. 2015. Echoing Hylas: A Study in Hellenistic and Roman Metapoetics. Madison. Hollis, A. 2006. “Propertius and Hellenistic Poetry.” Brill’s Companion to Propertius. Ed. Hans- Christian Günther. Leiden. 97-125. Heyworth, S. J. 1986. “Notes on Propertius, Books III and IV.” CQ 36.1: 199-211. -----------. 1994. “Some Allusions to Callimachus in Latin Poetry.” Materiali e discussioni per l’analisi dei testi classici 33: 51-79. -----------. 1995. “Propertius: Division, Transmission, and the Editor’s Task. Roman Comedy, Augustan Poetry, Historiography. Papers of the Leeds International Latin Seminar 8. Leeds. 165-88. -----------. 2007. Cynthia: A Companion to the Text of Propertius. Oxford. Hinds, S. 1998. Allusion and Intertext: Dynamics of Appropriation in Roman Poetry. Cambridge, UK. Holliday, P. J. 2005. “The Rhetoric of ‘Romanitas’: The ‘Tomb of the Statilii’ Frescoes Reconsidered.” Memoirs of the American Academy in Rome 50: 89-129. Hollis, A. S. 1994. “Octavian in the Fourth Georgic.” CQ 46.1: 305-8. Hollywood, A. 2006. “Performativity, Citationality, Ritualization.” Bodily Citations: Religion and Judith Butler. Eds. E. Armour and S. St. Ville. New York. 252-275. Holzberg, N. 2002. Ovid: The Poet and his Work. Tr. G. M. Goshgarian. Ithaca. Horsfall, N. 2003. The Culture of the Roman Plebs. London. Houghton, L. B. T. 2013. “Epitome and Eternity: Some Epitaphs and Votive Inscriptions in the Latin Love Elegists.” Inscriptions and their Uses in Greek and Latin Literature. Eds. P. Liddel and P. Low. Oxford. 349-64. Housman, A. E. 1972. The Classical papers of A. E. Housman, Vol. 1. Ed. J. Diggle and F. R. D. Goodyear. Cambridge, UK. Hubbard, M. 1974. Propertius. London. Hutchinson, G. 2006. Propertius: Elegies Book IV. Cambridge, UK. James, S. L. 2003. Learned Girls and Male Persuasion: Gender and Reading in Roman Love Elegy. Berkeley. Janan, M. 2001. The Politics of Desire in Propertius IV. Berkeley. Jashemski, W. 1964. “A Pompeian Copa.” The Classical Journal 59.8: 337-49. ---------. 1967. “The Caupona of Euxinus at Pompeii.” Archaeology 20.1: 36-44. 249 Johnson, W. A. 2009. “Constructing Elite Reading Communities in the High Empire.” Ancient Literacies: The Culture of Reading in Greece and Rome. Eds. W. A. Johnson and H. N. Parker. Oxford. 320-332. Keith, A. 1992. “Amores 1.1: Propertius and the Ovidian Programme.” Studies in Latin literature and history 6. Ed C. Deroux. Brussels. 327-44. -----------. 1999. “Slender Verse: Roman Elegy and Ancient Rhetorical Theory.” Mnemosyne 52.1: 41-62. -----------. 2008. Propertius: Poet of Love and Leisure. London. Keil, H. 1874. Grammatici Latini. VI.1. Leipzig. Kennedy, D. F. 1993. The arts of love: Five studies in the discourse of Roman love elegy. Cambridge, UK. Knoche, U. 1936. “Zur Frage der Properzinterpolation.” Rheinisches Museum für Philologie 85.1. 8-63. Komp, M. 1988. Absage an Cynthia: Das Liebesthema beim späten Properz. Frankfurt am Main. Kristeva, J. 1969. “Le mot, le dialogue, et le roman.” Reprinted with slight changes in Σημειοτική: Recherches pour une sémanalyse. Paris. 143-73. Originally published in Critique 23 (1967): 438-65. -----------. 1980. “Word, Dialogue, and Novel.” Desire in Language: A Semiotic Approach to Literature and Art. Ed. L. S. Roudiez. Tr. T. Gora, A. Jardine, and L. S. Roudiez. New York. 64-91. Lachmann, K. 1816. Sex. Aurelii Propertii Carmina. Leipzig. Lausberg, H. 1960. Handbuch der Literarischen Rhetorik: Eine Gundlegung der Literaturwissenschaft. Munich. Lefèvre, E. 1990. “Die römische Literatur zwischen Mündlichkeit und Schriftlichkeit.” Strukturen der Mündlichkeit in der römischen Literatur. Ed. G. Vogt-Spira. Tübingen. 9- 15. Leonhardt, J. 2006 [2016]. "Caesius." Brill’s New Pauly. Eds. H. Cancik and H. Schneider. 19 February 2016. <http://referenceworks.brillonline.com.libproxy1.usc.edu/entries/brill-s- new-pauly/caesius-e224260> Lewis, M. 2016. “Catullus’ Intertextual Crossroads: Where Theocritus’ Idyll 15 meets Poems 64 and 36. AJP 137.2: 321-57. Lissberger, E. 1934. Das Fortleben der rӧmischen Elegiker in den Carmina Latina Epigraphica. Tübingen. Lord, A. B. 1951. “Composition by Theme in Homer and Southslavic Epos.” TAPA 82: 71-80. -----------. 1960 [2000]. The Singer of Tales. Cambridge, MA. Luck, G. 1955. “Das Acanthisgedicht des Properz.” Hermes 83.4: 428-38. -----------. Die römische Liebeselegie. Heidelberg. Malafouris, L. 2008. “Is it ‘me’ or is it ‘mine’? The Mycenaean sword as a body-part.” Past Bodies. Eds. J. Robb and D. Boric. Oxford. 115-23. Marini, G. 1785. Iscrizioni Antiche delle Ville de’ Palazzi Albani. Rome. Markus, D. 2000. “Performing the Book: The Recital of Epic in First-Century C.E. Rome.” Classical Antiquity 19.1: 138-79. Martelli, F. 2013. Ovid’s Revisions: The Editor as Author. Cambridge, UK. Mau, A. 1889. “Scavi di Pompei, 1886-1888.” Mittheilungen des kaiserlich deutschen Archaeologischen Instituts, Römische Abtheilung 4: 100-25. 250 Mauerhofer, K. 2004. Der Hylas-Mythos in der antiken Literatur. Beiträge zur Altertumskunde 208. Munich-Leipzig. Mayer, R. 1994. Horace: Epistles, Book 1. Cambridge, UK. Mayor, J. E. B. 1877 [1979]. Thirteen Satires of Juvenal, with a Commentary. New York. Mazzarino, A. 1955. Grammaticae Romanae Fragmenta Aetatis Caesareae Vol. 1. Turin. McCarthy, K. 1998. “Servitium Amoris: Amor Servitii.” Women and Slaves in Greco-Roman Culture. Eds. Sandra Joshel and Sheila Murnaghan. New York. 179-98. McKeown, J. 1987. Ovid: Amores, Volume I. Text and Prolegomena. Liverpool. McLuhan, M. 1964. Understanding Media: The Extensions of Man. New York. Menes, E. P. 1983. “External Evidence for the Division of Propertius, Book 2.” Classical Philology 78.2: 136-43. Miller, P. A. 2004. Subjecting Verses: Latin Love Elegy and the Emergence of the Real. Princeton. Milnor, K. 2014. Graffiti and the Literary Landscape in Roman Pompeii. Oxford. Montanari, F. 2012. “Introduction: The Homeric Question Today.” Homeric Contexts: Neoanalysis and the Interpretation of Epic Poetry. Berlin. Morgan, J. D. 1986. “Cruces Propertianae.” Classical Quarterly 36. 174-92. Morris, I. 1992. Death Ritual and Social Structure in Classical Antiquity. Cambridge, UK. Mynors, R. A. B. 1994. Vergil. Georgics. Edited with a commentary. Oxford. Nagy, G. 1990a. Pindar’s Homer: The Lyric Possession of an Epic Past. Baltimore. ------------. 1990b. “Early Greek Views of Poets and Poetry.” The Cambridge History of Literary Criticism. Vol. 1. Ed. G. A. Kennedy. Cambridge, UK. -----------. 2004. Homer’s Text and Language. Champaign, IL. Nappa, C. 2005. Reading after Actium: Vergil’s Georgics, Octavian, and Rome. Ann Arbor. Nelis-Clément, J. and D. Nelis. 2013. “Furor epigraphicus: Augustus, the Poets, and the Inscriptions.” Inscriptions and their Uses in Greek and Latin Literature. Eds. P. Liddel and P. Low. Oxford: 317-47. Nethercut, W. R. 1961. “Ille parum cauti pectoris egit opus.” TAPA 92: 389-407. Nisbet, R. G. M. and M. Hubbard. 1978 [2004]. A Commentary on Horace Odes, Book II. Oxford. Nisbet, R. G. M and N. Rudd. 2004 [2010]. A Commentary on Horace, Odes, Book III. Oxford. Noy, D. 2005. “Romans.” Encyclopedia of Cremation. Ed. D. Davies. London. 366-8. O’Gorman, E. 2016. “Sententiae and sensiculi: remediated sense(s) in Quintilian and Tacitus” [conference paper]. Classical Literature and Quotation Culture. July 21, 2016. Exeter, UK. Oliensis, E. 2014. “The paratext of Amores I: gaming the system.” The Roman Paratext: Frame, Texts, Readers. Ed. L. Jansen. Cambridge, UK. 206-23. Oliver, R. P. 1945. “The First Edition of the Amores.” TAPA 76: 191-215. Ong, W. J. 1982 [1991]. Orality and Literacy: The Technologizing of the Word. London. O’Rourke, D. 2014. “Paratext and intertext in the Propertian poetry book. The Roman Paratext: Frame, Texts, Readers. Ed. L. Jansen. Cambridge, UK. 156-75. Osgood, J. W. 2002. The Missing Years: Italy, 44-29 BC (diss.). New Haven. Page, D. L. 1962. Poetae Melici Graeci. Oxford. Papanghelis, T. D. 1987. Propertius: A Hellenistic Poet on Love and Death. Cambridge, UK. Parry, M. 1930. “Studies in the Epic Technique of Oral Verse-Making. 1. Homer and Homeric Style.” HSCPh 41: 73-147. Peiper, C. 1879. Quaestiones Propertianae. Kreuzberg. 251 Peirano, I. 2014. “‘Sealing’ the book: the sphragis as paratext.” The Roman Paratext: Frame, Texts, Readers. Ed. L. Jansen. Cambridge, UK. 224-42. Pérez Alonso, M. A. 2012. “Fusión de estilo directo e indirecto en la transmisión de doctrina en el Ars grammatica de Carisio.” Emerita: Revista de Lingüística y Filología Clásica 80.1: 149-70. Perkell, C. G. 1996. “The ‘Dying Gallus’ and the Design of Eclogue 10.” CP 91.2. 128-40. Petrain, D. 2000. “Hylas and Silva: Etymological Wordplay in Propertius 1.20. HSCPh 100: 409- 21. Pinotti, P. 1996. “The Pseudo-Ovidian Ad Liviam de Morte Drusi (Consolatio ad Liviam, Epicedium Drusi): A Critical Text with Introduction and Commentary.” Gnomon 68: 500-04. Pohlenz, M. 1913. De Ovidii carminibus amatoriis. Göttingen. Neumeister, C. 2006 [2014]. "Propertius." Brill’s New Pauly. Eds. H. Cancik and H. Schneider. 4 October 2014. <http://www.brillonline.nl.libproxy.usc.edu/entries/brill-s-new- pauly/propertius-e1010470> Ramsby, T. R. 2007. Textual Permanence. London. Rappaport, R. 1992. “Ritual, Time, and Eternity.” Zygon 27.1: 5-30. -----------. 1999. Ritual and Religion in the Making of Humanity. Cambridge, MA. Rees, R. 2014. “Intertitles as deliberate misinformation in Ammianus Marcellinus. The Roman Paratext: Frame, Texts, Readers. Ed. L. Jansen. Cambridge, UK. 129-42. Reitzenstein, E. 1936. Wirklichkeitsbild un Gefülsentwicklung bei Properz. Leipzig. Reynolds, L. D. and N. G. Wilson. 1968 [2013]. Scribes and Scholars: A Guide to the Transmission of Greek and Latin Literature. Oxford. Richardson, L, Jr. 1977. Propertius: Elegies I-IV. Norman, OK. Richlin, A. 1999. "Cicero's Head." Constructions of the Classical Body. Ed., James I. Porter. Ann Arbor. 190-211. Richmond, J. A. 2006 [2016]. "Consolatio ad Liviam (Epicedion Drusi)." Brill’s New Pauly. Eds. H. Cancik and H. Schneider. 02 March 2016. <http://referenceworks.brillonline.com.libproxy2.usc.edu/entries/brill-s-new- pauly/consolatio-ad-liviam-epicedion-drusi-e304300> Roller, M. 2001. Constructing Autocracy: Aristocrats and Emperors in Julio-Claudian Rome. Princeton. Roman, L. 2006. “A History of Lost Tablets.” CA 25.2: 351-88. Ross, D. O. 1975. Backgrounds to Augustan Poetry: Gallus, Elegy and Rome. Cambridge, UK. Rossi. 1877. Act. Inst. Arch. Unknown, cited at CLE 389. Rüpke, J. 2006 [2017]. “Fuscus, Arellius.” Brill’s New Pauly. Eds. H. Cancik and H. Schneider. 10 March 2017. <http://dx.doi.org.libproxy2.usc.edu/10.1163/1574-9347_bnp_e417110> Rust, E. M. 2009. Ex angulis secretisque librorum: Reading, writing, and using miscellaneous knowledge in the Noctes Atticae. Diss.: University of Southern California. Sailor, D. 2006. “Dirty Linen, Fabrication, and the Authorities of Livy and Augustus.” TAPA 136.2: 329-88. Schmitt, R. 1967. Dichtung und Dichtersprache in indogermanischer Zeit. Wiesbaden. Schoonhoven, H. 1992. The Pseudo-Ovidian Ad Liviam de Morte Drusi (Consolatio ad Liviam, Epicedium Drusi): A Critical Text with Introduction and Commentary. Groningen. Schrijvers, P. H. 1988. “A propos de la datation de la Consolatio ad Liviam.” Mnemosyne 41.3: 381-4. 252 Schulze, K. P. 1910. Römische Elegiker: Eine Auswahl aus Catull, Tibull, Properz. Berlin. Shackleton Bailey, D. R. 1952. “Echoes of Propertius.” Mnemosyne 5.4: 307-55. -----------. 1953. “Some Recent Experiments in Propertian Criticism.” Proceedings of the Cambridge Philological Society (New Series) 2. 9-20. -------------. 1956. Propertiana. Cambridge, UK. Shipley, F. W. 1924 [1992]. Velleius Paterculus: Compendium of Roman History and Res Gestae Divi Augusti. Cambridge, MA. Skutsch, F. 1901. Aus Vergils Frühzeit. Leipzig. -----------. 1906. Gallus und Vergil. Aus Vergils Frühzeit. Zweiter Teil. Leipzig. Skutsch, O. 1963. “The Structure of the Propertian Monobiblos.” CP 58: 238-9. -----------. 1973. “Readings in Propertius.” CQ 23.2: 316-23. -----------. 1975. “The Second Book of Propertius.” HSCPh 79: 229-33. Sogliano, A. 1888. “VII. Pompei – Degli edificii recentemente scoperti, e degli oggetti raculti negli scavi dal dicembre 1887 al giugno 1888. Relazione dell’ispettore prof. A Sogliano.” Notizie degli scavi di antichità Jan. 1888: 509-30. Solin, H. 1975. “Die Wandinschriften im sog. Haus des M. Fabius Rufus.” Neue Forschungen in Pompeji. Eds. B. Andreae and H. Kyrieleis. Recklinghausen. 99-124. Somerville, T. 2009. “The Pleonasm of the New Gallus, and the Gallus of the Monobiblos. Mnemosyne 62: 295-7. Spickermann, W. 2006 [2016]. "Brennus." Brill’s New Pauly. Eds. H. Cancik and H. Schneider. 5 April 2016. <http://referenceworks.brillonline.com.libproxy2.usc.edu/entries/brill-s- new-pauly/brennus-e219890> Squire, M. 2009 [2015]. Image and Text in Graeco-Roman Antiquity. Cambridge, UK. Stroh, W. 2006 [2014]. "Cornelius." Brill’s New Pauly. Eds. H. Cancik and H. Schneider. 04 October 2014 <http://www.brillonline.nl.libproxy.usc.edu/entries/brill-s-new- pauly/cornelius-e305350> Strothmann, M., Elvers, K.-L., and Stegmann, H. 2006 [2017]. “Caecilia.” Brill’s New Pauly. Eds. H. Cancik and H. Schneider. Brill Online. Reference. University of Southern California. 11 March 2017. <http://dx.doi.org.libproxy1.usc.edu/10.1163/1574- 9347_bnp_e222450> Sullivan, J. P. 1976. Propertius: A Critical Introduction. Cambridge, UK. Susini, G. 1973. The Roman Stonecutter: An Introduction to Latin Epigraphy. Ed. E. Badian. Tr. A. M. Dabrowski. Oxford. Syme, R. 1939 [1960]. The Roman Revolution. Oxford. ----------. 1959. “Livy and Augustus.” HSCPh 64: 27-87. -----------. 1978. “Mendacity in Velleius.” AJP 99.1: 45-63. Tarrant, R. J. 1987. “The Reader as Author: Collaborative Interpolation in Latin Poetry.” Editing Greek and Latin Texts. Ed., J. N. Grant. New York. 121-62. --------. 2006. “Propertian Textual Criticism and Editing.” Brill’s Companion to Propertius. Ed. H. C. Günther. Leiden. 45-65. Taylor, L. R. 1968. “Republican and Augustan Writers Enrolled in the Equestrian Centuries.” TAPA. 469-86. Thomas, R. 1979. “New Comedy, Callimachus, and Roman Poetry.” HSCPh 83: 179-206. Thorne, B. 1993. Gender Play: Boys and Girls in School. New Brunswick, NJ. Tovar, R. C. 2003. “Libertas en la Sátira: De Horacio a Juvenal.” Actas del XI Congreso de la Sociedad Española de Estudios Clásicos II: 785-93. 253 Tränkle, H. 1960. Die Sprachkunst des Properz und die Tradition der Lateinischen Dichtersprache. Hermes Einzelschriften 15. Wiesbaden. -----------. 1990. Appendix Tibulliana. New York. Uggeri, G. 2006 [2016]. “Perusia.” Brill’s New Pauly. Eds. H. Cancik and He. Schneider. 28 December 2016. <http://dx.doi.org.libproxy1.usc.edu/10.1163/1574-9347_bnp_e916000> Varhelyi, Z. 2010. The Religion of Senators in the Roman Empire: Power and Beyond. Cambridge, UK. Varone, A. 2002. Erotica Pompeiana: Love Inscriptions on the Walls of Pompeii. Tr. R. P. Berg. Rome. Veyne, P. 1988. Roman Erotic Elegy: Love Poetry and the West. Tr. David Pelauer. Chicago. Videau, A. 2010. La poétique d’Ovide, de l’élégie a l’épopée des Metamorphoses. Paris. Vogt-Spira, G. 1991. “VOX UND LITTERA: Der Buchstabe zwischen Mündlichkeit und Schriftlichkeit in der grammatischen Tradition.” Poetica 23.3: 295-327. Wachter, R. 1998. “ ‘Oral Poetry’ in ungewohntem Kontext: Hinweise auf Mündliche Dichtungstechnik in den pompejanischen Wandinschriften.” Zeitschrift für Papyrologie und Epigraphik 121: 73-89. Warmington, E. H. 1940. Remains of Old Latin Vol. IV: Archaic Inscriptions. Cambridge, MA. Weißenberger, M. 2006 [2015]. "Gorgias." Brill’s New Pauly. Eds. H. Cancik and H. Schneider. 9 February 2015. <http://www.brillonline.nl.libproxy.usc.edu/entries/brill-s-new- pauly/gorgias-e426320> Wellesley, K. 1969. “Propertius’ Tarpeia Poem (IV.4).” Acta Classica Univ. Scient. Debrecen. 5: 93-103. West, M. L. 2011. The Making of the Iliad: Disquisition and Analytical Commentary. Oxford. White, P. 1993. Promised Verse: Poets in the Society of Augustan Rome. Cambridge, MA. -----------. 2005. “Poets in the New Milieu: Realigning.” The Cambridge Companion to the Age of Augustus. Ed. K. Galinsky. Cambridge, UK. 321-39. Wilkinson, L. P. 1969. The Georgics of Virgil: A Critical Survey. Cambridge, UK. Williams, G. 1968. Tradition and Originality in Roman Poetry. Oxford. Wills, J. 1996. Repetition in Latin Poetry: Figures of Allusion. Oxford. Wimmel, W. 1960. Kallimachos in Rom: Die nachfolge seines apologetischen Dichtens in der Augusteerzeit. Wiesbaden. Wirszubski, C. 1950. Libertas as a Political Idea at Rome during the Late Republic and Early Principate. Cambridge, UK. Wyke, M. 2002. The Roman Mistress: Ancient and Modern Representations. Oxford. Yardley, J. C. 1974. “Propertius’ Lycinna.” TAPA 104: 429-34. Zanker, P. 1988. The Power of Images in the Age of Augustus. Tr. Alan Shapiro. Ann Arbor. Zeitlin, F. 1996. Playing the Other: Gender and Society in Classical Greek Literature. Chicago. Ziolkowski, J. M. 2005. “Oral-formulaic tradition and the composition of Latin poetry from andtiquity through the twelfth century.” New Directions in Oral Theory: Essays on Ancient and Medieval Literatures. Ed. M. C. Amodio. Tempe: 122-48.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

"ipsum principem cernere in publico": the visibility of the Roman emperor from 27BCE to 40CE

PDF

Cicero's Academica and the foundation of a Roman academy

PDF

One of many doors: sensing the literary sense through a cognitive poetics-inclusive reading approach

PDF

How does it feel to be a problem? Revisiting Cane and the life of Jean Toomer

PDF

The global market for wombs: a study of the transnational surrogacy industry in Mexico

PDF

Mercy and metaphor: the empathy era and its discontents

PDF

The metaphysical materiality of Vladislav Khodasevich and T.S. Eliot's poetry

PDF

The biopoetic: toward a posthuman epic

PDF

The magic mirror: representations of monsters in Chinese classical tales

PDF

The pantoum across the Pacific: the circumnavigation of a poetic form; and, Why can’t it be tenderness (poems)

Asset Metadata

Creator Matera, Robert Charles (author)

Core Title The vulnerable corpus of Propertius

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Classics

Publication Date 07/12/2017

Defense Date 04/21/2017

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag allusion,anaklasis,author,authorship,Bodies,Body,Echo,elegy,epigraphic poetry,epigraphy,gender,Graffiti,inscriptions,interpolation,intertext,intertextuality,Latin,libertas,masculinities,masculinity,nequitia,OAI-PMH Harvest,oral composition,oral formulas,poetic memory,Poetry,Pompeii,pragmatics of authorship,Propertius,Roman,Rome,Speech,textual tradition,verse inscriptions,virtus

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Habinek, Thomas (committee chair), Echols, Alice (committee member), Moatti, Claudia (committee member)

Creator Email matera@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-397970

Unique identifier UC11264731

Identifier etd-MateraRobe-5501.pdf (filename),usctheses-c40-397970 (legacy record id)

Legacy Identifier etd-MateraRobe-5501.pdf

Dmrecord 397970

Document Type Dissertation

Rights Matera, Robert Charles

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA