At the end of Amores 3.9, Ovid's lament for Tibullus, there is a description of the dead poet's shade in Elysium. Kenney's Oxford Classical Text is as follows (Am. 3.9.59–68):

The modern editionsFootnote ¹ list no variants or conjectures for lines 59–60 and the text has not, to my knowledge, been doubted. But if umbra has what would be its natural meaning in the context of death, the dead person's ‘shade’ (OLD s.v. 9a), then we encounter a contradiction. Ovid imagines something enduring after death ‘other than a name and a shade’, and then proceeds to identify that something as Tibullus’ shade (implicitly in line 60 and explicitly with umbra … umbra in line 65).

If the text is to stand, umbra in line 59 evidently must mean something other than a dead person's shade.Footnote ² Whatever this other meaning might be, it would not only violate the expectation produced by the context of death but also go against Ovid's practice elsewhere. I find three other Ovidian variations on the theme of lines 59–60, with umbra in every case named as what survives the death of the body (used synonymously with Manes, anima and/or spiritus). None of these passages uses umbra with any meaning other than the expected ‘shade’:Footnote ³

The final passage, beginning with a close echo of our lines, is particularly noteworthy. Perhaps between the composition of the Amores and that of the Tristia Ovid came to regret his confusing formulation of umbra as ‘something other than umbra’. Or perhaps in Tristia 4.10 he reworks a text of the Amores that never contained this confusion in the first place.

Most translators of Amores 3.9 simply reproduce the difficulty of the Latin by rendering umbra as ‘shade’ in lines 59 and 65.Footnote ⁴ The verse translations of Lee and Green understand nomen et umbra as a kind of hendiadys: ‘a haunting name’ or ‘a ghostly reputation’.Footnote ⁵ Such a daring formulation, with the striking idea that a name could be (or could have?) a ghost, would seem to have required some elaboration by Ovid, or some reference to the idea in a previous text (which I have not been able to find). Indeed, Ovid's more than one hundred uses of umbra all have conventional meanings for the word: a shadow or shade cast by an object in the day (OLD s.v. 1, 3), the darkness of night (OLD s.v. 6) or of the underworld (OLD s.v. 7b), the false semblance of a body (OLD s.v. 9: Ov. Met. 4.434 [Narcissus], 14.362 [a boar], Fast. 3.702 [Caesar], Pont. 3.3.3 [Amor]), the false semblance of a concept (OLD s.v. 10: Ov. Met. 9.460 [pietas]), and most commonly the shade or ghost of a dead person in the underworld or on earth (over three dozen instances; OLD s.v. 7a).

Ovid does not juxtapose umbra with nomen anywhere else. In other authors, the collocation describes a diminished person who is a ‘shadow’ of his former self (OLD s.v. 7: Livy 5.18.4 me iam non eundem sed umbram nomenque P. Licini relictum uidetis; Luc. 1.135 stat magni nominis umbra, a phrase reworked at Luc. 8.449 and [Sen.] Oct. 71) or the empty ‘semblance’ of freedom (OLD 10: Luc. 2.302–3 tuumque | nomen, Libertas, et inanem persequar umbram [cf. 3.144–6]; Plin. Ep. 8.24.4 reliquam umbram et residuum libertatis nomen). In the case of Luc. 1.135, the metaphor is aided by an extended simile of a dead tree casting sterile shade (1.136–43); explicitly at Luc. 2.302–3 and (I think) implicitly in Livy and Pliny, the metaphor is of a dead person's ‘shade’. The difficulty thus remains that a Roman reader of our passage envisioning a ‘shadow’ or ‘semblance’ of a person after death would inevitably picture what umbra cannot mean here: the dead person's ‘shade’.

Abandoning special pleading, we may suspect instead that the transmitted umbra has supplanted something at the end of line 59. The corruption could have been prompted by the couplet's depiction of Tibullus in Elysium, which brought the idea of a shade to the scribe's mind. The scribe may also have been influenced by umbra at the end of line 65, in a similar grammatical construction (conditional clauses in lines 59 and 65). Perhaps the most likely scenario is an intrusive gloss (a common route of corruption), with umbra originally identifying the aliquid that survives death. One possibility for what this gloss displaced is inane: death has left behind a nomen that is ‘empty’ in the absence of its owner, whose shade has descended to Elysium. This would produce a thought similar to Tr. 4.10.85–6 (see above), where the umbra is the aliquid nisi nomina which survives death. nomen inane is used a few times in Ovid and elsewhere to describe some concept or value (uirtus, fides, amicitia, conubium, imperium) that has been emptied of its usual force (Hor. Epist. 1.17.41; Ov. Ars am. 1.740, Her. 10.118; Luc. 2.342–3, 5.389–90). At Tr. 3.3.50, however, the phrase describes Ovid's wife calling his name after his imagined death: clamabis miseri nomen inane uiri. As we will see, Tristia 3.3 repeatedly reworks ideas from our poem. Corruption from inane to et umbra could have occurred via the intermediary in umbra, a common line ending (Verg. 4x, Prop. 2x, Ov. 5x). But because in Tristia 3.3 Ovid describes dying in exile, his name would be ‘empty’ perhaps not so much because of its owner's death as because of his absence, which has made the ritual of conclamatio devoid of its usual significance.Footnote ⁶

In any case, attention to the intertextual models for Amores 3.9 suggests a more likely original text: nomen et ossa, referring primarily to a gravesite with a funerary inscription. As has been well established, Tibullus’ poetry is the primary model for our poem, in particular Tibullus 1.3, where the poet imagines his death far from home.Footnote ⁷ There are allusions to Tibullus earlier in Ovid's poem, such as the futile piety of the poet's girlfriend(s) (Am. 3.9.33–4, 37–8 ~ Tib. 1.3.23–6). But the passage immediately preceding our lines is a particularly overt refashioning of Tibullus’ poem. Tibullus fears death on the foreign shores of Phaeacia, where his mother, his sister and Delia would be unable to perform his funeral rites (Tib. 1.3.3–10);Footnote ⁸ Ovid replies that, because Tibullus did not die as a stranger on Phaeacia, his mother, his sister, Delia and Nemesis were able to perform his funeral rites (Am. 3.9.47–54). This ‘correction’ of Tibullus leads to a second correction in lines 55–8, since Tibullus died when he had both Delia and Nemesis to mourn him, not just Delia as in Tibullus 1.3. To underscore this fact, Ovid borrows a Tibullan description of dying in Delia's arms for his description of Tibullus’ death in Nemesis’ arms: Am. 3.9.58 me tenuit moriens deficiente manu ~ Tib. 1.1.60 te teneam moriens deficiente manu.

Ovid's next lines continue his reworking and (as we will see) his correction of Tibullus, as he imagines the poet's shade surviving his funeral and residing in Elysium:

The allusion is recognizable even with the transmitted text of Amores 3.9, but becomes more pointed with my emendation to nomen et ossa. Tibullus requests a funerary epigram including his name (hic iacet … Tibullus) to be inscribed on a stone above his ossa, and then imagines his shade in Elysium; Ovid similarly refers to Tibullus’ nomen and ossa before imagining his shade in Elysium.Footnote ⁹ With nomen et ossa, in fact, Ovid would reproduce a collocation which is common in epigraphic funerary epigramsFootnote ¹⁰ (for example CLE 965.1–2 quandocumque leuis tellus mea conteget ossa | incisum et duro nomen erit lapide, 966.1–2, 980.1–2 hospes consiste et Thoracis perlege nomen: | immatura iacent ossa relata mea, 1085.1–2 si quis forte legit titulum nomenue requirit, | Dorchadis inueniet ossa sepulta loco, 1086.1–2), and which also appears in Virgil and Ovid to describe graves with inscribed epitaphsFootnote ¹¹ (Aen. 7.3–4 ossaque nomen … signat,Footnote ¹² Met. 2.337–8 repperit ossa tamen peregrina condita ripa | incubuitque loco nomenque in marmore lectum … [referring to the epitaph just given at lines 327–8 hic situs est Phaethon …], 11.705–7 et tibi nunc saltem ueniam comes, inque sepulcro | si non urna, tamen iunget nos littera, si non | ossibus ossa meis, at nomen nomine tangam).

Ovid therefore imagines Tibullus’ umbra in Elysium surviving all that remains of him on earth: his bones marked by his name on a tombstone. But this kind of epigraphic nomen offers a limited and contingent memorial in comparison (the poets claim) to the immortal name afforded by poetry. Thus, for instance, Catullus writes his elegy for Allius to immortalize his name so that cobwebs will not gather on his neglected tombstone (68b.49–50 nec tenuem texens sublimis aranea telam | in deserto Alli nomine opus faciat), or Horace claims that his Odes are a more permanent monument than the tombs of the Pharaohs (Carm. 3.30.1–5).Footnote ¹³ The immortal nomen afforded by poetry becomes a frequent theme in Ovid's works (for example in Am. 1.3.21–6, 1.15.20 and passim, 3.1.65), including in our poem, where he repeatedly contrasts poetic fame with the stark, physical reality of death (3.9.5–6, 17–32 [especially 31 longum … nomen], 39–40).

Particularly instructive is Tristia 3.3, a poem which reworks both Amores 3.9 and Tibullus 1.3 to show how the death far from home that Tibullus only feared has become an imminent reality for Ovid.Footnote ¹⁴ As we have seen, Ovid worries that the umbra which survives the death of his body will be doomed to an eternity among Sarmatian umbrae (lines 59–64). Reversing the succession of thought in Tib. 1.3.53–8 and Am. 3.9.59–60 (as I have emended it), Ovid then turns to the fate of his earthly remains. He requests that his bones be repatriated (3.9.65 ossa tamen facito parua referantur in urna) and marked with an epitaph carved in stone (3.9.73–6):

As in Amores 3.9, the poet's name is juxtaposed with his bones (Nasonis … ossa). But Ovid then claims for himself a more lasting nomen afforded by his books of poetry, quos [sc. libellos] ego confido, quamuis nocuere, daturos | nomen et auctori tempora longa suo (3.9.79–80). This, then, brings us to Ovid's final ‘correction’ of Tibullus 1.3 in Amores 3.9. Tibullus claims that his dedication to love is what will qualify his shade for Elysium, which he describes as a paradise for devoted lovers (1.3.57–66). Ovid, in contrast, imagines Tibullus in Elysium as a love poet, where he will join the famous ranks of Catullus, Calvus and Gallus.Footnote ¹⁵ Implicitly, then, Ovid adds Tibullus’ name to the rolls of immortal poets, as he had done explicitly in Am. 1.15.27–8. Both Tibullus and Ovid know that the earthly fate of everyone, even great poets, is bones and a name on a gravestone. Tibullus dared to imagine his soul surviving these remains in Elysium. Ovid goes even further in his tribute to his predecessor, immortalizing his soul in Elysium and his name in poetry.