¹ Texts and commentaries:Rothstein, M., Die Elegien des Sextus Propertius(Berlin, 1920);Google ScholarButler, H. E. and Barber, E. A., The Elegies of Propertius(Oxford, 1933)Google Scholar;Barber, E. A., Sexti Properti Carmina(Oxford, 1960)Google Scholar;Camps, W. A., Propertius, Elegies Book III(Cambridge, 1966)Google Scholar;Richardson L. Propertius Elegies I-IV (Norman, Okla., 1977);Fedeli, P., Sexti Properti Elegiarum Libri IV(Stuttgart, 1984)Google Scholar and Properzio, , II Libro terzo delle Elegie(Bari, 1985)Google Scholar;Goold, G. P., Propertius, Elegies(Cambridge, MA, 1990)Google Scholar. References to these works will be by author’s name. This and all further citations of Propertius, unless otherwise noted, are from Fedeli 1984. All references in the text to the arrangements of 3.7 by Scaliger, Baehrens, Housman, Postgate, Richmond, and Tremenheere can be found in Smyth, W. R., Thesaurus Criticus ad Sexti Propertii Textum(Leiden, 1970), p.96Google Scholar. In addition, the following works will be cited by author's name:Hubbard, M., Propertius(London, 1974)Google Scholar;Schulz-Vanheyden, E., Properz und das greichische Epigramm(diss. Münster, 1969)Google Scholar;Walsh, T.,’Propertius′ Paetus elegy (3.7)’, LCM 12 (1987), 66–69.Google Scholar

² Goold, G. P., ’On Editing Propertius’, in Papers in Honour of Otto Skutsch, BICS suppl. 51 (1987), 27–38,Google Scholar at 37: ’Just as certain authentic words of the poet have been replaced by other words, so too the authentic position of certain verses has been falsified by deliberate removal of those verses elsewhere’, and Goold, p. 22. ’The chaos of 3.7 exceeds that of all other poems in Propertius. But the underlying reason is both simple and clear: from some pre-archetypal manuscript a leaf became detached and was replaced back to front’. See also Goold, , ’Problems in Editing Propertius’, in Grant, J. N.(ed.), Editing Greek and Latin Texts(New York, 1989), 97–119, at p. 108Google Scholar

³ See Butler-Barber, p. 276 on Vahlen’s 1883 article; Richardson, pp. 341–342 and Fedeli 1985, p. 240 both use Propertian precedent to defend the unprepared shift in addressee at 11–12. Fedeli appeals to the rapid succession of apostrophes throughout the poem, while Robertson, F., ’Lament for Paetus–Propertius 3.7’, TAPA 100 (1969), 377–386, at 379 reasons, ’The bewildering succession of apostrophes to different persons… are less troublesome when we consider the poem as a rhetorical theme where Paetus is less in the forefront of the poet's mind than the general reader who is being given a homily on ethics’.Google Scholar

⁴ Brief literary treatments can be found in Esteve–Forriol, J. V., Die Trauer und Trostgedichte inder romischen Literatur(diss. Miinchen, 1962), pp.119–120Google Scholar; Schulz-Vanheyden, pp. 58–69; F. Robertson, op. cit. (n. 3); Hubbard, pp. 83–87;Warden, J., Fallax Opus: Poet and Reader in the Elegies of Propertius(Toronto, 1980), pp.63–64,97–99;Google ScholarWilliams, G., Figures of Thought in Roman Poetry(New Haven, 1980), pp.112–113.Google Scholar

⁵ E.g.Rothstein, , pp. 50–51; Schulz-Vanheyden, pp. 58–69.Google Scholar

⁶ Hubbard, p. 84

⁷ Cf. the treatment of Propertius 1.1 as an expansion of Meleager A.P.12.101 in Camps, W. A., Propertius, Elegies Book I(Cambridge, 1961), p.42Google Scholar; Hubbard, pp. 14–20. On the affinities between epigram and Propertian elegy in general, see Boucher, J. P., Etudes sur Properce. Problemes dèinspiration et d'art(Paris, 1965), pp.410–417Google Scholar

⁸ This interpretation follows the transpositions of 17–24 to other locations in the poem. Lines 2–124 are transposed in the editions cited by Butler–Barber, p. 275, and the texts of Fedeli, Richardson, and Goold. Some, including Butler-Barber, guess that one or both of these couplets may be interpolated. Lines 17–20 are also transposed by the editions cited in Butler–Barber, by Richardson, Goold, and Walsh, p. 69.

⁹ There is no reason why Aquilo and Neptune together cannot be the plural subjects of ‘reddite’, even if they did cause the shipwreck. Walsh's suggestion, pp. 68–69, that ‘humus’ means ‘the sand at the bottom of the sea’ and thus out of Aquilo's jurisdiction, is outlandish. Goold 1987, op. cit. (n. 2), p. 37, supports the conjecture ‘aquae’ (vocative) for ‘humo’, but like others feels compelled to join these lines to the address to Thetis and the Nereids (67–70).

¹⁰ For examples, see Schulz-Vanheyden, pp. 61–62 and Fedeli 1985, pp. 237 and 247.1 disagree with Schulz-Vanheyden that this ‘sepulcrum’ could be a cenotaph. If a passerby had not found something to bury, Paetus would have no tomb at all, except possibly a cenotaph at home.

¹¹ Moreover, the next passage, beginning ‘ite, rates curvas et leti texite causas: / ista per humanas mors venit acta manus’ (29–30) should be read as a response to the blame of external forces and pity for Paetus’ miserable fate in 1–28 (cf. ‘causa’, 1; ‘per te’, 2).

¹² For other examples, see Schulz-Vanheyden, pp. 60–61. The literary background of the content of 1–4 is not epigrammatic.

¹³ The text of this epigram and of all others cited are from Gow, A. S. F. and Page, D. L., The Greek Anthology. Hellenistic Epigrams(Cambridge, 1965) [= HE]or The Greek Anthology. The Garland of Philip(Cambridge, 1968) [= GP]Google Scholar

¹⁴ Gow-Page, HE,Vol. II, p. 369; Schulz–Vanheyden, p. 59; Fedeli 1985, pp. 235–236.Google Scholar

¹⁵ Cf. the last line of Catullus 67:‘falsum mendaci ventre puerperium’ (48).

¹⁶ The paradox is enhanced by the fact that the speaker of the poem, the sea-tossed victim, is also not present, though he is able to point out ‘this stone’.

¹⁷ The same motif occurs in Leonidas A.P.7.654.5–6. Fedeli 1985, p. 239, explains that ‘adstare’ can signify ‘la presenza di congiunti accanto alia tomba o al rogo funebre’.

¹⁸ Commentators dispute the exact meaning of line 11. Do the birds hover or stand? Are Paetus′ bones still floating or have they drifted to shore? My position is that the birds are not flying, but standing (as Fedeli 1985, p. 239). Virgil's depiction of Iris coming down to Dido (‘devolat et supra caput adstitit’, Aen.4.702) means that Iris flies down and comes to a stop standing over Dido's head as she lies dying. Fedeli supposes that the birds are standing on floating bones, citing the Ovidian echo (‘ossa superstabunt volucres inhumata marinae?’Her.10.123), but as Gow-Page, HE, Vol. II, p. 287 recognize, Ovid, at least, portrays the bones lying unburied on the beach, for Ariadne was left behind on shore.

¹⁹ In addition to building a series of pictures concentrating on a single character, Propertius′ elegy differs from epigrams in that it never pretends to be words inscribed on stone. From the first line the elegy presents itself as an emotional outburst at the death of Paetus.

²⁰ Rothstein, p. 51; Butler-Barber, p. 287; Gow-Page, HE,Vol. II, p. 287; Schulz-Vanheyden, pp. 69–70; Fedeli 1985, p. 240.Google Scholar

²¹ Cf. the analogous use of TTOVin line 3 and in A.P.7.652.5, cited above. This interpretation is the reverse of Gow-Page, HE,Vol. II, p. 287: ‘G. plainly remembers Od. 1.161 and he and Propertius chose the second alternative’. Glaukos' words ()actually echo Telemachos' first alternative; the bones rot somewhere on land or a wave whirls them in the sea. Glaukos adds birds, who alone can tell where Erasippos' bones are rotting. Even if a human were to find the bones on land, they could not be identified, and still only the gulls would know the true location of Erasippos' remains. Propertius leaves out the rotting, but adds further that the birds are standing on the bones. While bones can be tossed in the sea, they do not float, as Fedeli and Gow-Page suggest.

²² Rothstein, p. 51, appropriately pictures the bones tossed by the sea onto a rock in 11, but explains 12, 'Schon im nachsten Verse lasst P. mit der ihm eigenen Freiheit in der Behandlung des Tatsachlichen dieses Bild durch ein anderes verdrangt werden; der Leichnam liegt nicht mehr auf einem Felsen im Meere, sondern auf dem Meeresgrunde, von dem Wasser bedeckt wie von einem Grabhiigel'. This is an over-literal interpretation of the line; pro probably means simply 'instead of.

²³ See also A.P. 121b,294, 506. On the difficulties of assigning authorship of epigrams to Antipater of Thessalonica (dated to the Augustan period) or to the earlier Antipater of Sidon, see Gow-Page, HE,Vol. II, pp. 31–33 and GP,Vol. II, pp. 20–21. This epigram is headed ’AvTindrpov,but its location in a ’small Philippan group’ suggests its attribution to Antipater of Thessalonica (GP,Vol. II, p. 68).

²⁴ Richardson, p. 341; Fedeli 1985, pp. 232–233.

²⁵ In the D.R.N.,’pabula praebens’ occurs at 1.229, ’pabula cum praebet’ at 2.996, ’pabula...praebebat’ at 5.991, and ’pabula laeta’ at 1.14 and 2.317, 364, 596, 875, 1159.

²⁶ Propertius uses both ’esca’ and ’pabula’ in one other place, again together and with sinister tones, as the food brought by trembling girls into the dark lair of Juno Sospita’s sacred serpent at Lanuvium (4.8.7 and 11).

²⁷ K. Morsley, ’Propertius 3.7’, CQ25 (1975), 315–318, at 316: ’This address [to ’pecunia’] clearly continues for the first 8 lines and there is no ground for supposing that the next couplet is not included as well’.

²⁸ See also the epigrammatic evidence adduced in Section I.

²⁹ For example, Warden, op. cit. (n. 4), p. 98, who does not concern himself with the text of 3.7, assumes, ’In line 9 the address switches to the drowned Paetus’. Walsh, p. 67, reverses the order of the two couplets to ensure that the ’tua’ and ’tibi’ of 11–12 apply also to Paetus’ mother and explains, ’it is pointless for Propertius to tell Money that Paetus’ mother is unable to give him proper burial’.

³⁰ 2.9.15 also has not escaped emendation. Goold accepts Housman’s ’cui turn’ for ’cum tibi’ (15), while Butler-Barber, p. 206, suggest a vocative, ’tantum corpus, Achille’, at 13. Others, such as Fedeli, Camps, and Richardson, do not deny Propertius his rhetorical freedoms, especially when there is no chance of confusion.

³¹ Cf. Tibullus 1.3.5–6 (’non hie mini mater / quae legat in maestos ossa perusta sinus’), where the poet dying far from home fears that his mother will not gather up his bones.

³² Moreover, 2.12.18 (’si puer est, alio traice puella tuo’) is certainly corrupt, and Goold’s ’si pudor est, alio traice tela, puer!’ explains the paradosis better than Fedeli’s (1984)’ si pudor est, alio traice tela tua’.

³³ These are 2.9.52, 3.4.4, a n d 3.13.42. 2.12.17 (see n. 32) a n d 4.11.42 should be moved t o this list: Camps obelizes 4.11.39–40 and Goold uses Heyne’s ’qui tumidas’ for ’quique tuas’ in 40. In 42, Goolds prints the variant ’nostros’ for ’vestros’ which lacks a second person plural antecedent, whether verb or vocative. If correct, ’vestros’ would have to refer back to ’maiorum cineres’ in 37. Though confusing, the issue is still not an unprepared shift in addressee of the same person and number.

³⁴ Fedeli 1985, p. 240.

³⁵ S. J. Heyworth, ’Propertius 2.13’, Mnemosyne45 (1992), 45–59 argues for the unity of the poem with a lacuna before line 17 and notes on 46–47 that ’the sudden change from third to second person without a vocative or a personal pronoun is one of those phenomena found most frequently in those authors whose manuscripts are late and corrupt’.

³⁶ The problem here is not so much the idea that the poet could not have spoken to Ulysses (’verbaque duxisset pondus habere mea’, 44), but rather the implication in the tense of’viveret’ that Ulysses would still be living content as a ’pauper’. For various arguments, see Goold 1987 and 1989, op. cit. (n. 2); Morsley, op. cit. (n. 27); Walsh, pp. 67–68.

³⁷ I owe this observation to Stephen Heyworth.

³⁸ Only Richardson, p. 345, fully confronts the problem of ’non habet unda deos’ (18) in the mouth of a poet who rebukes Aquilo and Neptune for causing the shipwreck (13–16) and Thetis and the Nereids for not saving Paetus (67–70). Cf. Camps, p. 84: ’Formally there is a contradiction between this and the poet’s apostrophe of Aquilo and Neptune in 14–15; but we do not feel this because beneath the stylized rhetorical manner the sentiments are consistent: wind and sea are cruel–prayer is useless, there is none to help you’. Fedeli 1985, p. 243: ’vista Pinutilita dei tentativi dell’incolpevole Peto il poeta si senta autorizzato a concludere che in realta non sono gli dei, ma una cieca violenza, a governare il mare’. Both dodge the problem.

³⁹ R. Lattimore, Themes in Greek and Latin Epitaphs(Urbana, 1942), p. 175.

⁴⁰ Lattimore, op. cit. (n. 39), p. 177.

⁴¹ Rothstein, p. 51: ’Unverkennbar is der Anklang an Cat. 68,97’. Fedeli 1985, p. 238, calls the verbal similarity ’la chiara allusione di Properzio ad un noto contesto catulliano’, but does not remark on the close structural correspondence.

⁴² Propertius’ postponement of’ nunc’ is a simple variation on its usual placement at the start of the account of the present which is contrasted with the past. Even in poem 68, Catullus does not use ’ n u n c ’ in 93–94, which describe the present, but waits until he describes the body’s situation in 97–100. So Propertius chooses not to emphasize the fact of Paetus’ non-burial with ’ n u n c ’, but rather to stress the actual present state of Paetus’ corpse. To reverse the two couplets, as Walsh does (pp. 67–68), disturbs the more significant construction’sed’. For another example of this construction in a Greek epitaph for a dead youth which also employs the contrast theme, see Lattimore, op. cit. (n. 39), p. 176.

⁴³ R. G. M. Nisbet and M. Hubbard, A Commentary on Horace,Odes Book I(Oxford, 1970), pp. 318 and 323. Many more examples of this theme are given in P. Fedeli, Properzio Elegie Libro IV(Bari, 1965), p. 247; cf. Lucretius’ mock lament, ’omnia ademit / una dies infesta tibi tot praemia vitae’(D.R.N.3.898–899).

⁴⁴ Hubbard, p. 86.

⁴⁵ The phrase, λπς…Tρου (4), suggests that Nikanor was young (Gow-Page, GP,Vol. II, p. 32). This epigram could be by either Antipater (see n. 23 and Gow-Page, GP,Vol. II, pp. 31–32).