¹ On the ego, especially Veyne, P., Roman Erotic Elegy: Love, Poetry and the West (1988)Google Scholar; on the puella, Wyke, M., ‘Written women: Propertius' Scripta Puella’, JRS 77 (1987), 47–61Google Scholar; also ‘Mistress and metaphor in Augustan elegy’, Helios 16 (1989), 25–47Google Scholar, and ‘Reading female flesh: Amores 3.1’, in Cameron, A. (ed.), History as Text (1989)Google Scholar. Cf. also Sharrock, A. R., ‘Womanufacture’, JRS 81 (1990), 36–49Google Scholar.

² Kennedy, D. F., The Arts of Love: Five Studies in the Discourse of Roman Love Elegy (1993), 13Google Scholar. Kennedy (13–21) discusses the representation of the first-person male authorial persona in Tibullus 1.1–3, 5, 6.

³ cf. e.g., Veyne, op. cit. (n. 1), 48. An exception is Wyke, M., ‘The elegiac woman at Rome’, PCPhS 33 (1987), 153–78Google Scholar; however, Wyke is concerned primarily with the diversity of female voices, and she does not consider the issue of Book IV's characters as representatives of a new, constructed discourse in the manner suggested here.

⁴ With the question of a ‘real’ Cleopatra already the ambiguity between representation and reality is highlighted; for Cleopatra, even before her death, was for the Romans both real and a construction to be manipulated for propagandist purposes. Cf. Wyke, M., ‘Augustan Cleopatras: female power and poetic authority’, in Powell, A. (ed.), Roman Poetry and Propaganda in the Age of Augustus (1992), 98–140Google Scholar.

⁵ For an analysis of the rhetorical use of ‘written fashion’ as a semiotic system, cf. Barthes, R., Système de la mode (1967)Google Scholar; also Barthes, S/Z (1974). Indeed, fashion metaphors are often an important means by which a text cloaks its existence as text to instil a sense of ‘reality’.

⁶ The additional characters of Book IV will be treated in a larger study of elegiac decorum in Book IV, now in progress.

⁷ Hercules’ attempt to reconstruct himself within the poem's discourse places him in the position of the poet (or here ego) as creator.

⁸ On the two poetic programmes and ‘new themes’ of Book IV, see Camps, W. A., Propertius Elegies Book IV (1965), 1–6Google Scholar; Pillinger, H. E., ‘Some Callimachean influences on Propertius Book 4’, HSCP 73 (1969), 171–99Google Scholar; Van Sickle, J., ‘Propertius (vates): Augustan ideology, topography and poetics in elegy IV, 1’, Dialoghi di Archeologia 8 (1974), 116–45Google Scholar; Hubbard, M., Propertius (1974), 116–56Google Scholar; Macleod, C. W., ‘Propertius 4,1’, PLLS 1 (1976), 141–53Google Scholar; Miller, J. F., ‘Callimachus and the Augustan aetiological elegy’, ANRW 11.30.1 (1982), 371–417Google Scholar; Stahl, H.-P., Propertius: ‘Love’ and ‘War’ (1985), 248–305Google Scholar; and Wyke, op. cit. (n. 3).

⁹ On which these lines are based in part; also important for this poem (and Book IV) is Tibullus 11.5.

¹⁰ Unless otherwise noted, citations from Propertius are from the Teubner edition of Fedeli (1984). All translations or paraphrases are my own and aim to be helpful rather than poetic.

¹¹ For a brief but very informative overview of Callimachus’ work, literary programme and use of imagery, cf. Hopkinson, N., A Hellenistic Anthology (1988), 83 ffGoogle Scholar. (on frg.1). Cf. esp. 98–101, for Callimachus’ influence on Latin programmatic poetry, including Propertius IV. 1.

¹² cf. esp. II.1, III.1–III.3, and III.9. The best discussion of the Augustan recusatio is still Wimmel, W., Kallimachos in Rom. Die Nachfolge seines apologetischen Dichtens in der Augusteerzeit (1960)Google Scholar.

¹³ While in his first books Propertius was enslaved by his love to Cynthia, it now seems he will exchange his servitium amoris for a servitium Romae.

¹⁴ cf. e.g., Ovid, Tristia 11.2.259.

¹⁵ For the idea of ‘rising’ as a particular aspect of the promise followed by hesitation in a recusatio, cf. e.g. II.10.11–12; III.9.52.

¹⁶ Wimmel, op. cit. (n. 12), 276–82, is the only one to emphasize this point; it is important, because it is this idea that the polarity has already been ‘resolved’ that makes the second part of IV.I so striking. Cf. also Pillinger, op. cit. (n. 8), 172–4, on these lines.

¹⁷ For comments on the ‘identity’ of Horos, his connection with Apollo in Alexandrian culture, and the similar relationship between Propertius-Horus in IV.I and Ovid-Apollo (through Germanicus) in the Fasti, cf. Barchiesi, A., ‘Discordant Muses’, PCPhS 37 (1991), 1–21Google Scholar.

¹⁸ Wyke, op. cit. (n. 3), 155.

¹⁹ cf. 125–6 and 65–6; here also is the imagery of ‘rising’ and ‘falling’ that always holds a hint of recusatio.

²⁰ For the labels amor and Roma, see Wyke, op. cit (n.3).

²¹ This could be said (to varying degrees) of most of those studies cited in n. 8 above, with the exception of Wyke. And of course, the ‘division’ is ostensibly a valid one.

²² cf. Nethercut, W. R., ‘Notes on the structure of Propertius Book 4’, AJPh 89 (1968), 449–64Google Scholar; Hutchinson, G. O., ‘Propertius and the unity of the book’, JRS 74 (1984), 100–3Google Scholar; also Wyke, op. cit. (n. 3).

²³ This shift began earlier, in Book III, but there the ‘favoured’ side was still that of love; on this similarity between Books III and IV, cf. Nethercut, W., ‘Recent scholarship on Propertius’, ANRW 11.30.3 (1984), 1849–50Google Scholar.

²⁴ Other examples of devices remanipulated in Book IV form the basis of a larger study. Cf. also DeBrohun, J., Hercules Belabored: Propertius 4.9 and the Discourses of Elegy (Ph.D. diss., University of Michigan, 1992)Google Scholar.

²⁵ Anderson, W. S., ‘Hercules Exclusus: Propertius IV, 9’, AJP 85 (1964) 1–12Google Scholar.

²⁶ For a summary of scholarship on Propertius IV.9 since Anderson, cf. esp. Cairns, F., ‘Propertius 4.9 : “Hercules Exclusus” and the dimensions of genre’, with a response by Anderson, in Galinsky, K. (ed.), The Interpretation of Roman Poetry: Empiricism or Hermeneutics (1992), 65–103Google Scholar. Cairns does not refute Anderson's original conclusions about the poem. See also DeBrohun, op. cit. (n. 24), 1–4 et passim.

²⁷ For the cult of the Bona Dea and its celebrations, see most recently H. H. Brouwer, Bona Dea. The Sources and a Description of the Cult, Etudes préliminaires aux religions orientales dans l'empire romain 110 (1989), although with caution. While Brouwer argues for the use of puellae to describe the worshippers of the Bona Dea, this is based solely on the evidence of Propertius IV.9 (pp. 256, 295, 371–2).

²⁸ cf. esp. Tibullus 1.1.56; 1.5.74; Ovid, AA III.581; Met. 14.717. These words are also customary for addressing the doorway of a temple, which makes the connection with the Bona Dea here especially effective; cf. Prop. in.III.17.37. Anderson does not bring out this aspect of the exclusus amator argument, but it is mentioned by Yardley, J. C., ‘The elegiac paraclausithyron’, Eranos 76 (1978), 19–34Google Scholar. Cf. also Pucci, P., ‘Lingering on the thresh-old’, Glyph 3 (1978), 52–73Google Scholar, with bibliography, and McParland, E., ‘Propertius 4.9’, TAPA 101 (1970), 349–55Google Scholar.

²⁹ cf. e.g. Tibullus 11.6.12.

³⁰ Anderson rightly considered these lines part of Hercules’ argument as amator, but he did not differentiate the two descriptions.

³¹ With Fedeli and Butler-Barber, I have accepted the postulation that 42 is an interpolation from 66. For a good presentation of the problem, see Butler, H. E. and Barber, E. A., The Elegies of Propertius (1969)Google Scholar, ad loc. In view of the ‘double’ argument here, it seems likely that the lost line 42 completed Hercules’ first plea, before he moved to his softer approach in 45–50.

³² This is perhaps in contrast to the more ‘bucolic’ description of the weapon as Maenalius ramus in 1. 15.

³³ For sources and a discussion of the Hercules-Omphale story, cf. Fantham, E., ‘Sexual comedy in Ovid's Fasti: sources and motivation’, HSCP 87 (1983), 185–216Google Scholar.

³⁴ This constitutes what has been called an Alexandrian footnote: ‘the same one’ you saw earlier (i.e. III. II, or, more generally, in ‘that other’ Hercules story). For this term, cf. Ross, D. O., Backgrounds to Augustan Poetry: Gallus, Elegy and Rome (1975), 78Google Scholar. Cf. also Conte, G. B., The Rhetoric of Imitation (1986), 52–69Google Scholar, for this type of reference.

³⁵ cf. on this poem Stahl, H.-P., Propertius: ‘Love’ and ‘War’. Individual and State under Augustus (1985), 234–47Google Scholar; Nethercut, W. R., ‘Propertius 3.11’, TAPA 102 (1971), 411–43Google Scholar; also Terence, Eun. 1027 (for Hercules-Omphale as exemplum).

³⁶ In particular, Propertius' poetry always comes out ‘soft’ (mollis) in II.1.1–2, in contrast to the hard verse (durus versus) he is unable to write (11.1.41). Cf. also III.1.19–20, III.3.1, 18. For durus, cf. esp. 11.1.41; for hirsutus, IV.1.61 (see n. 13 above), Ovid, Tristia 11.2.259; Horace, Odes 11.12. This opposition is so prevalent in the vocabulary of poetic conventions that Propertius can play on its ‘reversal’ already in 1.6, where the soft lover carries out his militia under a duro sidere while Tullus serves in mollis Ionia; cf. Kennedy, op. cit. (n. 2), 31–3.

³⁷ cf. Ovid, Ars Amatoria III.26, Propertius 11.34.83, for similar uses of minus; cf. also Latona's words in Ovid, Met. 6.368, which echo Propertius and are spoken in a similarly programmatic setting (I am indebted to K. Sara Myers for this point). Also, cf. on this Clauss, James J., ‘The episode of the Lycian farmers in Ovid's Metamorphoses’, HSCP 92 (1989), 297–314Google Scholar.

³⁸ Perhaps Hercules ‘remembers’ and expects his audience to remember that the lovely Hylas once tried for more with unhappy results — Propertius 1.20.43ft. Cf. also Ovid, Amores 1.6.3, where the poet asks for something exiguum. On the possible poetic nature of the water in IV.9, cf. DeBrohun, op. cit. (n. 24), 45–59.

³⁹ For Propertius' use of visual imagery, cf. esp. J.-P. Boucher, Etudes sur Properce. Problèmes d'inspiration et d'art (1965).

⁴⁰ On possible aspects of mime in IV.9, cf. McKeown, J. C., ‘Augustan elegy and mime’, PCPhS n.s.25 (1979), esp. 77 ff.Google Scholar; also, Zeitlin, F., ‘Travesties of gender and genre in Aristophanes' Thesmophoriazousae’, in Foley, H. P. (ed.), Reflections of Women in Antiquity (1981)Google Scholar, points out that role reversal is a typical comic device (cf. Plautus, Casina). What we do not see in Propertius' account is the woman who would be putting on his armour in the usual versions, thus completing the reversal.

⁴¹ Indeed, the recall works much as a mythological exemplum — only here the reference is to himself.

⁴² According to Horace, AP 179–88, those things which would be unacceptable or unbelievable should be removed from the eyes and only narrated to the audience. Cf. Heroides 9.119–22, where Deianira states that what she has heard is not nearly so painful as what she now sees with her own eyes.

⁴³ That Hercules serves Omphale because he has fallen in love with her is a reduction, perhaps by the Alexandrians, of the story as it appears in Sophocles’ Trachiniae, where Hercules is forced to perform his humiliating slave service to Omphale as punishment for killing Iphitus in anger; cf. Trachiniae 356–7. See on this Lyne, R.O.A.M., ‘Servitium Amoris’, CQ 29 (1979), 117–30CrossRefGoogle Scholar.

⁴⁴ This slippage between ‘real’ and verbal clothing is most explicit in IV.5.53–8. Here the lena warns her pupil not to accept words over ‘real’ goods; and the words she warns against most strongly are those with which the poet ‘clothed’ his puella in 1.2.

⁴⁵ This polarity begins in 25–6. Th e first word of 25 is femineae, the last of 26 viris. For this observation cf. McParland, op. cit. (n. 28).

⁴⁶ cf. here Tibullus 1.9.69–70: ‘istane persuadet facies auroque lacertos/vinciat et Tyrio prodeat apta sinu?’ where the clothing of the puella might also signify more than her love affair. Murgatroyd, P., Tibullus I: a Commentary (1980)Google Scholar, ad loc, failed to appreciate the implications of apta in IV.9 when he noted that Tibullus was the only writer to use apta with clothing. Of course, the Roman poets’ infatuation with the rhetorical possibilities of etymologies and wordplay generally makes the extension a natural one.

⁴⁷ Hercules’ oversized magni pedes could have generic implications within Ovid's poem, which is also an experiment in the expansion of elegy. Indeed, the same could be said for Ovid's depiction of the hero in Heroides 9 (see below). Cf. also Ars Amatoria 11.217–22.

⁴⁸ However, W. A. Camps, Propertius. Elegies, Book IV, ad loc. translates apta as deft, adducing Briseis’ words at Heroides 370.

⁴⁹ Note the ‘hard-soft’ opposition of duro and molli in these lines; and succumbere often has sexual connotations (cf. OLD s.v.2a).

⁵⁰ II.1.1 establishes Propertius' amores specifically as written discourse; and the mollis liber of 11.1.2 refers to his collection of poems. Cf. Wyke, op. cit. (n. 1, 1987).

⁵¹ For a detailed commentary on 11.1, cf. Wimmel, op. cit. (n. 12); on the generic implications of this poem, see Zetzel, J., ‘Re-creating the canon: Augustan poetry and the Alexandrian past’, Critical Inquiry 10 (1983), 83–105Google Scholar.

⁵² cf. on this especially Ross, op. cit. (n. 34), 58–9, 59 n. 2.

⁵³ The Propertian poet-lover often sets his activities beside those of ‘conventional men’ in lists such as this (1.6, III.9, III. 11). This rhetoric is also ‘turned against’ the poet by the lena of IV.5, who reminds the puella that it is precisely the ‘real’ soldier, sailor, or even ex-slave whose money should make him preferable to the poet, who has only verses.

⁵⁴ cf. also 1.6.29, where Propertius was born non idoneus armis.

⁵⁵ Wimmel, op. cit. (n. 12), 252.

⁵⁶ cf. on IV.2, Dee, J. H., ‘Propertius 4.2. “Callimachus Romanus” at work’, AJP 95 (1974), 43–55Google Scholar; Marquis, E. C., ‘Vertumnus in Propertius 4.2’, Hermes 102 (1974), 491–500Google Scholar; Shea, C., ‘The Vertumnus elegy and Propertius Book IV’, ICS 13 (1988), 63–71Google Scholar; also Pinotti, P., ‘Properzio e Vertumno : anticonformismo e restaurazione augustea’, Atti del Coll. Prop. III (1983), 75–96Google Scholar; Pillinger, op. cit. (n. 8); Wyke, op. cit. (n. 3); Deremetz, A., ‘L'élégie de Vertumne: l'oeuvre trompeuse’, REL 64 (1986), 116–49Google Scholar; P. Hardie, ‘Augustan poets and the mutability of Rome’, in Powell, op. cit. (n. 4), 74–5.

⁵⁷ Of course, the fact that Vertumnus is made of bronze should make him particularly dura. Dee, op. cit. (n. 56), 51–2, plays with the ambiguity well. Already, it seems women's clothing has a ‘softening’ effect.

⁵⁸ cf. esp. Shea, op. cit. (n. 56), 68–70.

⁵⁹ His index includes many ‘traditional’ jobs; cf. Ovid, Ars Amatoria 1.45–50. It is also noteworthy that except in IV.2.23, Vertumnus does not include women in his list, although they make up the majority of new characters in Book IV. In this respect, Vertumnus' list is more like those in II.1, III.9 or III. 11, where the poet-lover's occupation is opposed to those of other men.

⁶⁰ It is a natural development of the hard-soft imagery of Augustan poetics that the genres of epic and elegy are particularly gendered male (durus) and female (mollis). For this connection, see Propertius 1.7, and on 1.7 Kennedy, op. cit. (n. 2), 31–3, 58–9.

⁶¹ Shea, op. cit. (n. 56), 67.

⁶² There is one possible exception here. The desultor in IV.2.36 has a nature closest to that of Vertumnus. The trick rider is identified by his ability to jump from one horse to another. Like the wily Vertumnus, the desultor is never really ‘caught in the middle’. Or conversely, if he is suspended between two positions, one imagines him as perfectly balanced, not an unseemly mix of two opposites.

⁶³ Vertumnus is rather like the poet of Ovid, Amroes 11.4, whose ‘catalogue’ is introduced in 9–10, ‘non est certa meos quae forma invitet amores/centum sunt causae, cum ego semper amem’, and whose book ‘aptat omnibus historiis’ (44).

⁶⁴ Shea, op. cit. (n. 56), 64 suggested that Vertumnus wants us to mark his use of versus in IV.2.57, with this connotation of ‘verses’. Indeed, if this vertere/versus/versare pun works, the first line of IV.2 might recall more directly the beginning of poem III.11. In III.11.1–2, the reader presumably marvels at the fact that a woman ‘turns’ (versat) the poet's life (vitam), both ‘changing’ it from freedom to slavery and also ‘ruling’ its poetry (for vita = poetry, cf. 1.2.1). Cf. the similar play with versus and vertere in IV.5.46, 56–7; also versat in IV.5.63. And Cynthia's ghost in IV.7 holds the ‘change-ful’ (versutus) Nomas responsible for her death. Also, III.11.5–8 comprise another ‘traditional’ catalogue of decorum, including soldier, sailor, and slave.

⁶⁵ cf. also IV.2.1–2. III.25 and IV.2 are the only two occurrences of eventus in the Propertian corpus. Also, if the name (nomen) of Book 1 is Cynthia, Vertumnus is the fitting nomen for Book IV.

⁶⁶ On associations between Propertius 11.1 and Horace, Odes II.12, cf. Nisbet, R. and Hubbard, M., Horace: Odes. Book II (1978)Google Scholar, on 11.12. Cf. here also Davis, G., Polyhymnia (1991), 28–36Google Scholar (on 11.12, 30–3), for Horace's use of the recusatio as a device of ‘inclusion’ or ‘assimilation’ analogous to the rhetorical praeteritio.

⁶⁷ The text of Horace is cited from Shackleton-Bailey, D. R. (ed.), Q. Horati Flaccii Opera (1985, repr. 1991)Google Scholar.

⁶⁸ It is possible that Propertius was looking in part to Horace's poem. It is worth noting as well that in the second half of Horace's poem, in which he relates the themes which do fit his lyric poems, his poetry is described in terms of the beauty of the mistress Licymnia, who is ‘playing’ (cf. ludentem, 1. 20) with maidens at the festival of Diana, a scenario not unlike that encountered by Hercules in IV.9.

⁶⁹ In Cicero's example, the orator claims that Antony has done the converse of what Hercules attempt s here: through Antony's disgraceful behaviour, he has made his ‘male’ clothing ‘female’. Velleius Paterculus describes the literary patron Maecenas in similar terms in 11.88.2; on this see Kennedy, op. cit. (n. 2), 31.

⁷⁰ Cited from Winterbottom, M. (ed.), Quintiliani, M. Fabi, Institutionis Oratoriae Libri Duodecim (1970)Google Scholar.

⁷¹ Horace, AP, cited from Klingner (3rd edn, 1959).

⁷² These tears themselves are often representative of elegy. Cf. IV. 1.73, where Horos warns the ambitious poet that his new project will bring tears. For elegy as a poetry of ‘lament’, cf. Heinze, R., Ovids elegische Erzählung (1919), 19–20Google Scholar; for an etymological discussion of the term and its use by Roman poets, cf. esp. Hinds, S., The Metamorphosis of Persephone (1987), 103 ff.Google Scholar, with notes.

⁷³ It is worth noting that 11.5, like III.25, is also a renuntiatio amoris poem, in which the lover claims he will leave his mistress.

⁷⁴ Cited from Hollis, A. S. (ed.), Ovid Ars Amatoria Book 1. A Commentary (1977)Google Scholar.

⁷⁵ cf. here also Propertius 11.8.39–40, where the poet-lover reminds his audience that he is far inferior to Achilles.

⁷⁶ cf. e.g. 1.7.6, 11.1.78.

⁷⁷ Hercules' ‘conviction’ that his cross-dressed description will make him apta is itself a commentary on the puellae within the shrine. We never ‘see’ these girls. Do they look as he describes himself, or is his denied entrance in fact an assertion of his ‘difference’ in more than one respect?

⁷⁸ The sexual imagery of the limen as well as the poetic and political implications of its remanipulation in Book IV will be considered in my larger study of this book; cf. for now DeBrohun, op. cit. (n. 24), 17–37.

⁷⁹ Garber, M., Vested Interests: Cross-dressing and Cuttural Anxiety (1992), 16Google Scholar.

⁸⁰ Two articles which centre on an interpretation of these lines are McParland, op. cit. (n. 28) and Warden, J., ‘Epic into elegy: Propertius 4,9,70f.’, Hermes 110 (1982), 228–42Google Scholar (a response to McParland). Also, Hercules’ favoured position as ‘patron’ of Book IV can certainly be expected to have political as well as poetic implications. These further possibilities will be considered more fully in my larger study of Book IV.

⁸¹ Along with Fedeli, Camps, and Pasoli, I have accepted the emendation Sance in 72, as part of the cult title for Hercules, Semo Sancus; cf. esp. CIL IV.568, Sanco Sancto Semoni Dio Fidio sacrum.

⁸² While it is beyond the scope of this paper to deal with other examples, it is important to note that Vertumnus and Hercules are not the only characters to try to ‘have it both ways’ through this rhetoric of transformation. In a later study, I will demonstrate that the critical perception discussed here can be applied to all the characters and poems of Book IV.

⁸³ The issue of decorum in Augustan elegy has important political as well as poetic implications. In a larger study, I intend to demonstrate that the indecorous characters of Book IV have been exploited both for their value as poetic constructions and for their symbolic potential in the construction of Augustan hegemony.

⁸⁴ cf. Propertius 11.3.4, ‘et turpis de te iam liber alter erit’.