Hostname: page-component-74656f9884-8r8nv Total loading time: 0 Render date: 2026-11-08T13:58:50.859Z Has data issue: false hasContentIssue false
  • English
  • Français

OVID, TRISTIA 2.7–8 REVISITED*

Published online by Cambridge University Press:  12 December 2016

Barak Blum
Barak Blum*
Affiliation:
Wolfson College, Oxford
*
barak.blum@wolfson.ox.ac.uk
Rights & Permissions [Opens in a new window]

Extract

At the beginning of Tristia 2, a single long apologetic elegy, Ovid struggles with making sense of his continued occupation with poetry, despite the disaster it has brought upon him. In lines 7–8 he broods over Augustus’ displeasure, roused by the Ars Amatoria, which led to his reproach. The passage is also sometimes adduced as a reference to the removal of the Ars from Rome's public libraries.1

Information

Type
Research Article
Information
The Classical Quarterly , Volume 66 , Issue 2 , December 2016 , pp. 598 - 604
Check for updates
Copyright
Copyright © The Classical Association 2016 

At the beginning of Tristia 2, a single long apologetic elegy, Ovid struggles with making sense of his continued occupation with poetry, despite the disaster it has brought upon him. In lines 7–8 he broods over Augustus’ displeasure, roused by the Ars Amatoria, which led to his reproach. The passage is also sometimes adduced as a reference to the removal of the Ars from Rome's public libraries.Footnote 1

In the following piece I shall argue that the beginning of line 8 has been corrupted, review several emendations of the text, and suggest one of my own. Finally, the broader implication of the passage for the issue of the Ars’ involvement with Ovid's punishment shall be briefly discussed.

Tr. 2.1.7–8 Footnote 2

carmina fecerunt, ut me moresque notaret
iam demi iussa Caesar ab Arte meos.

8Footnote 3 iam demi iussa TR L: iam demum uisa A G H P V: iam pridem uisa D T: iam pridem inuisa K Pa: iam pridem iussa Ob: iam pridem emissa Luck: tum demum inuisa Traill

The poems have brought it about that Caesar reproved both me and my conduct, because of the Ars, already ordained to be removed.

Let us first look carefully at the paradosis of line 8. Owen and many other modern editorsFootnote 4 adhere to EhwaldFootnote 5 and print iam demi iussa. This reading is based on both the Fragmenta Trevirensia (TR) and the codex Laurentianus (L), two of the oldest manuscripts. Yet, there are textual variants that merit attention.Footnote 6 Five major manuscripts (A G H P V)Footnote 7 read iam demum uisa. Of the modern editors, it seems that very few print it.Footnote 8 Other notable manuscripts either have iam pridem uisa (D T),Footnote 9 or iam pridem inuisa (K Pa).Footnote 10 We should note that the distribution of the variants of 2.8 basically corresponds with that of the general transmission of the Tristia. Though stemmatics, owing to the preponderance of contamination, are of limited value in the case of the Tristia,Footnote 11 it is still possible to identify broadly three main branches of the textual tradition: TR L, A G H P V, and D T K.

None of these variants, however, seems to be satisfactory. Regarding iam demi iussa, Magnus, Luck and Traill all reject it as a corruption of the genuine reading.Footnote 12 I would add that it is prudent to reject this variant, as it is supported by only one branch of the paradosis. The reading iam pridem uisa is discarded by OwenFootnote 13 on account of ‘poor manuscript support’, while iam pridem inuisa is objectionable because the meaning would remind the reader that the Ars was loathed, something which Owen finds unacceptable. Magnus upholds iam pridem uisa,Footnote 14 but rejects iam pridem inuisa, claiming it is a mediaeval interpolation. LuckFootnote 15 follows Magnus's lead in rejecting iam pridem inuisa, though he differs in his formulation of that interpolation. Also, in contrast to Magnus, he dismisses iam pridem uisa, considering uisa a corruption. TraillFootnote 16 rejects both pridem readings on the basis of their provenance, saying they stem from ‘inferior manuscripts’. On the inadequacy of these two variants, I concur with Owen, Luck and Traill, and the reasoning of the latter seems closest to the mark. I believe that the stemma, contaminated as it may be, renders pridem improbable, as it is backed by only one major branch, against the other two. Finally, iam demum uisa is shunned by the aforementioned scholars, as well as by most of the editors (I shall go back to the reasons later on).

Two significant emendations have been proposed. LuckFootnote 17 suggests iam pridem emissa, which enjoys some popularity.Footnote 18 Indeed, emissa could have been corrupted to uisa, and pridem has some manuscript support. However, since it is given by only one branch of the paradosis, Traill suggests tum demum inuisa.Footnote 19 In fact, the change from iam to tum had already been proposed by Biensfeld.Footnote 20 Traill favours tum instead of iam owing to the relative frequency of the combination tum demum, and the common confusion of tum with iam. I agree with Traill that demum should be kept. It has conceivably given rise to the variant demi and may thus account for two of the three major branches of the tradition. However, there is simply no trace of Traill's tum in the manuscript tradition, just as there is no evidence for Luck's emissa.

I would, therefore, like to propose a slightly, yet significantly different emendation: iam demum inuisa. Thus, the distich would read: ‘The poems have brought it about that Caesar reproved both me and my conduct, because of the Ars, only nowFootnote 21 odious.’ The process leading to the different corruptions in the tradition might be inferred as follows. In the first branch (TR L), demum might have been corrupted into demi. One or several of the following factors may be responsible: (a) demum is graphically very similar to demi. (b) In the next line there is an inflection of the verb demĕre,Footnote 22 which the scribe might have previewed and mistakenly thought to be the case in line 8 as well. (c) Elision of –um in demum inuisa, in accordance with scansion, resulted in dem + inuisa. The ‘i’ sound was subsequently interpreted as belonging to the first word, thus producing demi. A passive infinitive naturally required a verb or participle to govern it, and in the context of Caesar as the subject of the consecutive clause, a command of some sort must have seemed sensible. Indeed, in two nearby instances the word Caesar appears alongside inflections of the verb iubere, together with a complementary infinitive.Footnote 23 That is why the second word, inuisa, or what was left of it after the removal of the letter ‘i’ (if it did not retain its place, which is also possible), was changed into the visually similar iussa, generating a perfectly reasonable phrase, so no particular suspicion should have arisen. The corruption of demum inuisa to demum uisa in the second branch (A G H P V) can be easily accounted for: as Traill suggests,Footnote 24 a scribal convention, by which demum inuisa was abbreviated to demūinuisa, may have graphically been mistaken for demum uisa. As for the third branch (D T K), the corruption of demum to pridem may have been the result of an intrusion of a gloss into the text, or perhaps an intended ‘correction’, by a person who was awareFootnote 25 that the Ars was written long before the composition of the Tristia, yet missed the ironic tone.

The emendation iam demum inuisa thus has a solid textual basis. Still, an objection already levelled against the variant iam demum uisa may be raised, that the combination iam demum is ‘impossible Latin’Footnote 26 (which is why Traill suggested tum instead of iam), or at least unparalleled.Footnote 27 But is this really the case? iam demum occurs once in Quintilian, Inst. 8.2.21, but, following Spalding,Footnote 28 editors generally suspect it.Footnote 29 Spalding remarked that it was a scribal error.Footnote 30 This has been takenFootnote 31 to mean that the ‘error’ was basically diplography, as there is another instance of iam in Quintilian, one line prior to iam demum. I suggest, however, that Spalding may himself have succumbed to an error, owing not to any visual confusion but to contemporary conventions regarding word repetition. Thus, the text of Quintilian should perhaps stand as transmitted. The combination iam demum might not be unique in Latin after all.

Even if iam demum is very rare, the combination is far from ‘impossible’. It is true that iam is commonly followed by other adverbs (such as dudum, pridem),Footnote 32 and that demum is commonly preceded by other adverbs (such as nunc, tunc, tum).Footnote 33 Yet, nunc and iam can have the same temporal sense,Footnote 34 and they are also interchangeable in certain combinations.Footnote 35 Moreover, iam is frequently followed by tandem, to convey the meaning of ‘at last, only now’.Footnote 36 The same meaning may be given by the combination nunc demum.Footnote 37 Both tandem and demum, on their part, can have the sense of ‘at last, at length’.Footnote 38 To sum up, iam and nunc can be synonymous, and so can demum and tandem. The combinations nunc demum and iam tandem are well documented. Therefore, it is quite possible that Ovid here purposefully utilizes an unusual combination to express a similar meaning, giving it a new ‘spin’, as it were, by supplanting nunc with iam (or tandem with demum). This would certainly be both within the bounds of poetic license and easily comprehended by the audience. In fact, Ovid is known for his inventive and sometimes idiosyncratic use of Latin, both in syntax and semantics.Footnote 39

It is noteworthy that part of the older editorial tradition was not uncomfortable with iam demum uisa,Footnote 40 nor did it hold iam demum to be insufferable Latin. The crucial factor for the widespread later rejection of iam demum uisa must have been, then, not its problematic language but merely the discovery of the manuscript TR. In support we can adduce the case of Ehwald. In his 1884 editionFootnote 41 he printed iam demum uisa, apparently having no irreconcilable qualms about it. His change of heart and adoption of iam demi iussa was only subsequent to the discovery of TR, which he himself had publishedFootnote 42 in his 1892 treatise, a reading reprinted in his joint 1922 edition with Levy.Footnote 43 Another telltale instance is Owen. While in his 1889 editio maior of the Tristia Footnote 44 he printed iam demum uisa, in his 1915 edition he altered the text to accommodate iam demi iussa, explicitly referring to Ehwald's work of 1892 in the apparatus.Footnote 45 Other editors, as we saw above, followed suit.

The pronounced influence of the TR reading on most modern editors is understandable. The discovery of TR by Ehwald brought to light a manuscript of the Tristia earlier than all others previously known, as it was dated to the tenth century. Its publication at the end of the nineteenth century must have been quite a sensation, surely heightened in an era when scholarly communications were beginning to be facilitated by the advance of technology. Both of these factors, that is the manuscript's early date and the excitement over its discovery, may have contributed to the influence of TR. It is no wonder, then, that iam demum uisa was discarded in favour of iam demi iussa.

However, the excitement over that once-new finding should not cloud our judgement of its quality. While TR indubitably offers some good readings, it also suffers from extensive corruption,Footnote 46 just like all the other manuscripts of the Tristia,Footnote 47 not only owing to scribal errors, but also because of well-intended yet ill-advised attempts at correction, apparently starting from Late Antiquity.Footnote 48 Consequently, venerable as it may be, not every reading of TR should automatically be adopted, even if it correlates with that of L, the other member of its branch (and prized for its age, sections of it being dated to the eleventh century, though it too, like TR, suffers from extensive corruptionFootnote 49 ). The testimony of the other manuscripts (such as the A G H P V branch) should be taken into account, especially when they all present readings which are different from TR and L,Footnote 50 as they do in the case of Tr. 2.8. Therefore, iam demi may not rest on such firm ground after all. iam demum, on the other hand, may be vindicated.

My emendation iam demum inuisa slightly differs from Traill's tum demum inuisa. Traill, too,Footnote 51 seems to grasp the irony that may lie in Tr. 2.8, which conveys a not-quite-veiled remark that the Ars has suddenly (referring to the time in which Ovid was relegated, in contrast with the time in which the Ars was published) begun to be hated by Augustus. The distich 2.7–8 suggests that Augustus had either never actually read the Ars Amatoria up until he decreed Ovid's relegation,Footnote 52 or that only at that stage he began to despise this work.Footnote 53 The irony is, of course, directed at the considerable time-interval (roughly a decade) between the publication of the Ars and Ovid's sentencing: Augustus must have been aware of it, yet chose to ignore it for a long period. This would support the notion that the Ars Amatoria, perhaps the only official charge against the poet, was used merely as a pretext for his relegation.Footnote 54 My emendation is consistent with this interpretation while retaining iam, thereby amplifying the irony.

This emendation also means that the passage has nothing to do with libraries, and cannot, therefore, be taken to refer to the ban on the Ars Amatoria. In this respect, my emendation joins in with the implications of the other emendations proposed,Footnote 55 as well as with those of the readings of the majority of manuscripts.Footnote 56

Lastly, the emendation has the added advantage of involving the least change to the paradosis, as all its individual components enjoy a good measure of manuscript support. Moreover, it supplies a reasonable and parsimonious explanation for the process by which the corruption may have taken place. Thus, iam demum inuisa presents itself as the more plausible solution to the text of Tristia 2.7–8.

Footnotes

*

I would like to thank Professors A. Rotstein, J. Price, M. Finkelberg and S.J. Harrison for reading and commenting on earlier versions of this article. Thanks are also due to the CQ editor Prof. B.J. Gibson and to the anonymous referee for their comments.

References

1 E.g. Ehwald, R., Ad historiam carminum Ovidianorum recensionemque symbolae II (Gotha, 1892), 10 Google Scholar; Owen, S.G. (ed.), Ovidi Nasonis Tristium Libri quinque, Ibis, Ex Ponto Libri quattuor, Halieutica, Fragmenta (Oxford, 1915 [reprint: 1959])Google Scholar, critical apparatus ad loc.; id., P. Ovidi Nasonis Tristium liber secundus (Oxford, 1924 [reprint: Amsterdam, 1967]), 45, 123; Wheeler, A.L. (ed.), Ovid: Tristia, Ex Ponto (London and New York, 1924 [reprint: Cambridge, MA, 1975])Google Scholar, 56 n. 1 (though he is aware that ‘the text is not certain’), and the revised, corrected second edition of Wheeler by G.P. Goold (Cambridge, MA, 1996), 56 n. 1 reproduces the same note; W. Kraus, in RE 18.2 (1942), s.v. ‘Ovidius Naso P., der Dichter’, cols. 1910–86, at 1919; André, J. (ed.), Tristes / Ovide (Paris, 1968), xvi n. 2, 36 n. 2Google Scholar; Luisi, A. and Berrino, N.F., Culpa silenda: le elegie dell'error Ovidiano (Bari, 2002), 148 Google Scholar; White, P., ‘Ovid and the Augustan milieu’, in Weiden Boyd, B. (ed.), Brill's Companion to Ovid (Leiden, Boston and Köln, 2002), 1–25, at 17Google Scholar.

2 Passages of the Tristia appearing in this article are cited from Owen (n. 1 [1915]). Translations from Latin are mine.

3 The manuscript sigla employed in this critical apparatus mostly correspond to Luck, G. (ed.), Tristia / P. Ovidius Naso, Band I: Text und Übersetzung (Heidelberg, 1967)Google Scholar. Variant readings and corruptions at the end of line 8 do not appear in the apparatus and the discussion of the tradition below, since they have no substantial bearing on the present argument. For further specifications of sigla and variants, see below and notes 6–7, 9–10.

4 Owen (n. 1 [1915]), ad loc. For other modern editions which print the same reading, see e.g. Landi, C. (ed.), P. Ovidii Nasonis Tristia (Turin, 1917)Google Scholar; Ehwald, R. and Levy, F.W. (eds.), Tristium libri V, Ibis, Ex Ponto libri IV (Leipzig, 1922)Google Scholar and cf. R.S. Deferrari, M. Inviolata Barry and R.P. Martin McGuire, A Concordance of Ovid (Washington, 1939 [reprint: Hildesheim, 1968]), amongst the verses s.v. demo, -ere; Wheeler (n. 1); André (n. 1); Hall, J.B. (ed.), P. Ovidi Nasonis Tristia (Stuttgart, 1995). Also cf. Luisi and Berrino (n. 1), 60Google Scholar.

5 Ehwald (n. 1).

6 A more detailed apparatus and sigla can be found in Owen, S.G. (ed.), P. Ovidi Nasonis Tristium Libri V (Oxford, 1889)Google Scholar, ad loc. and cxi-cxiii, 48; Luck (n. 3), ad loc. and 21–6; Hall  (n. 4), ad loc. and v-viii.

7 The Angeli Politiani Marcianus (A); Guelferbytanus Gudianus lat. 192 (G); Holkhamicus 322, also called Londiniensis Bibl. Brit. Add. 49368 (H); Palatinus, also called Vaticanus Palatinus, lat. 910 (P); and the Vaticanus lat. 1606 (V), as well as one minor manuscript, the Guelferbytanus extr. 76.3.

8 Possibly only Ripert, E. (ed.), Ovide: Les Tristes, Les Pontiques, Ibis, Le Noyer, Halieutiques (Paris, 1957)Google Scholar, ad loc. For earlier editorial support, see the discussion below and notes 40–1, 44.

9 Gothanus membr. II 122 (D); Turonensis 879 (T). Also some minor manuscripts, e.g. Londiniensis Mus. Brit. Add. 18,384; Berolinensis lat. Octau. 67.

10 Leidensis, bibl. publ. 177 (K); Vaticanus, Palatinus 1668 (Pa). In addition, some minor ones, e.g. Leidensis bibl. publ. 191; Londiniensis, Mus. Brit. Add. 18,384 [manus altera]; Bodleianus Auct. F. 1.18; Berolinensis Lat. Octau. 67 [manus altera]. Also all the ueteres editiones. A similar reading, iam pridem iussa is found in one manuscript: Vaticanus Ottob. lat. 1469 (Ob); cf. Tarrant, R.J., ‘Ovid’, in Reynolds, L.D. (ed.), Texts and Transmission: A Survey of the Latin Classics (Oxford, 1983), 257–84, at 283 n. 8Google Scholar.

11 See e.g. Luck (n. 3), 18, who recognizes the problem, yet supplies a tentative stemma.

12 Magnus, H., ‘P. Ovidius Naso, Vol. III Fasc. I Tristium libri V Ibis Ex Ponto libri IV. Ediderunt Rudolphus Ehwald et Fridericus Waltharius Levy Leipzig 1922’, Philologische Wochenschrift 44 (1924)Google Scholar, cols. 244–53, at 250; Luck, G., ‘Textprobleme der Tristien’, Philologus 103 (1959), 100–13CrossRefGoogle Scholar, at 101; Luck, G. (ed.), Tristia / P. Ovidius Naso, Band II: Kommentar (Heidelberg, 1977), 170 Google Scholar; Traill, D.A., ‘Ovid, Tristia 2.8, 296, and 507: happier solutions’, Hermes 120 (1992), 504–7Google Scholar, at 504.

13 Owen (n. 1 [1924]), 123.

14 Magnus (n. 12), cols. 251–2. For editions in favour of iam pridem uisa, see e.g. Loers, V. (ed.), P. Ovidii Nasonis Tristium Libri V (Trier, 1839), 138 Google Scholar; Riese, A. (ed.), P. Ovidii Nasonis Carmina. Vol. III: Fasti. Tristia. Ibis. Ex Ponto. Halieutica. Fragmenta (Leipzig, 1874), 134 Google Scholar.

15 Luck (n. 12 [1959]), 102; id. (n. 3), 14.

16 Traill (n. 12), 504.

17 Luck (n. 3), ad loc. See also Luck (n. 12 [1959]), 102; Luck, G., ‘Notes on the language and text of Ovid's Tristia ’, HSPh 65 (1961), 243–61, at 244Google Scholar. Note that there is a precedent for this emendation, raised, according to Ehwald (n. 1), 8, by E. Baehrens in 1881.

18 Adopted by, for example, Diggle, J., ‘Notes on Ovid's Tristia, Books I-II’, CQ 30 (1980), 401–19, at 401CrossRefGoogle Scholar; Green, P., Ovid. The Poems of Exile: Tristia and the Black Sea Letters (Berkeley and London, 2005), 222 Google Scholar; Ingleheart, J., A Commentary on Ovid, Tristia Book 2 (Oxford, 2010), 32, 68Google Scholar.

19 Traill (n. 12), 504.

20 Binsfeld, J.P., Observationes Ovidianae criticae (Bonn, 1860), 7 Google Scholar, on the grounds of a parallel with Ov. Met. 13.391. Binsfeld was given credit (though not assent) for this emendation by Owen (n. 6), 224, as well as Ehwald (n. 1), 8. Traill (n. 12), 504 n. 6 does not refer to Binsfeld (and is perhaps unaware of him), but mentions that Magnus had already considered tum instead of iam. Magnus, H.R., ‘R. Ehwald, Ad historiam carminum Ovidianorum recensionemque symbolae. II. III. Programm des Gymn. zu Gotha, 1892’, Berliner Philologische Wochenschrift 12 (1892), cols. 1485–7, at 1486Google Scholar, in turn, mentions that this had been proposed before, meaning perhaps Binsfeld.

21 I.e. ‘just / only at that point in time’, more vividly expressed by virtue of the use of iam, which like nunc (L&S s.v. nunc II.A.1; OLD s.v. 3.a) can be used to describe actions taking place in the past (L&S s.v. jam I.B). Moreover, iam used in this way would increase the irony, for which see below.

22 2.9: deme.

23 Tr. 2.23–5: Caesar… | … dicere iussit … | iusserat … dici …. Cf. also Tr. 1.3.5–6: discedere Caesar | … iusserat …; Tr. 1.3.85: iubet … discedere Caesaris ira.

24 Traill (n. 12), 504–5.

25 Perhaps also ‘inspired’ by Tr. 2.539: nos quoque iam pridem scripto peccauimus isto.

26 See Traill (n. 12), 504, explicitly concurring with Ehwald (n. 1) and Magnus, H., ‘Magdalena Schmidt, De Ovidii Tristium libro II. 1923’, Philologische Wochenschrift 43 (1923), cols. 817–21, at 820Google Scholar, who in turn mentions his debt to Ehwald in this point.

27 See Binsfeld (n. 20), 7: ‘ipsa … dicendi formula neque Ovidius umquam, neque aliorum quisquam scriptorum usus est.’

28 Spalding, G.L. (ed.), M. Fabii Quintiliani De Institutione Oratoria Libri duodecim, Volumen III: Continens Libros VII-IX (Leipzig, 1808 [reprint: Hildesheim, 1969]), 208. Cf. Binsfeld (n. 20), 7Google Scholar; Ehwald (n. 1), 8; Owen (n. 1 [1924]), 123.

29 Mostly by enclosing iam with incriminating square brackets, e.g. Radermacher, L. (ed.), M. Fabi Quintiliani Institutionis Oratoriae Libri XII, Pars Secunda: Libros VII-XII continens (Leipzig, 1959), 77 Google Scholar; Winterbottom, M. (ed.), M. Fabi Quintiliani Institutionis Oratoriae Libri duodecim, Tomus II: Libri VII-XII (Oxford, 1970), 429 Google Scholar; Rahn, H. (ed.), Marcus Fabius Quintilianus: Ausbildung des Redners, Zweiter Teil, Buch VII-XII (Darmstadt, 1975), 148 Google Scholar; Cousin, J. (ed.), Quintilien: Institution Oratoire, Tome V, Livres VIII et IX (Paris, 1978), 59 Google Scholar. Sometimes by omitting iam from the text altogether, as e.g. in Russell, D.A. (ed.), Quintilian: The Orator's Education, Books 6–8 (Cambridge, MA, 2001), 338 Google Scholar. Editions printing iam demum in their text without any qualifying brackets are very few, e.g. Halm, C. (ed.), M. Fabi Quintiliani: Institutionis Oratoriae Libri duodecim, Pars posterior (Leipzig, 1869), 66 Google Scholar, who none the less remarks in his critical apparatus that ‘id demum probabiliter coni. Spalding’, and Butler, H.E. (ed.), Quintilian: Institutio Oratoria, Books VII-IX (Cambridge, MA, 1921 [reprint 1996]), 208Google Scholar.

30 Spalding (n. 28), 208.

31 E.g. by Owen (n. 1 [1924]), 123.

32 OLD s.v. dudum 1b, 2b and 3, and s.v. pridem 3 and 4. Cf. TLL 7.1.112.66, 7.1.114.80–1, s.v. iam I.D.3.a, and 5.1.2177.62–5.1.2180.57, s.v. dudum II. For the distribution of those two adverbs after iam in Ovid, see also Deferrari, Inviolata Barry and McGuire (n. 4), s.v. iam.

33 L&S s.v. demum I.B.a-c; OLD s.v. demum 1b. Cf. TLL 5.1.513.45–5.1.514.71, s.v. demum I.A.1.c, d. For those combinations specifically in Ovid, see Deferrari, Inviolata Barry and McGuire (n. 4), s.v. demum.

34 TLL 7.1.102.74–8, s.v. iam I.B.

35 E.g. the combination iam iam has the same meaning as that of iam nunc (L&S s.v. jam I.A.1.b.α and I.A.1.b.β respectively).

36 L&S s.v. jam I.A.2.b; OLD s.v. tandem 2b.

37 L&S s.v. demum I.B.a.

38 L&S s.v. demum I and tandem; OLD s.v. demum 1a and 3a, and s.v. tandem 2.

39 On Ovid's linguistic innovations, see e.g. De Jonge, T.J., Publii Ovidii Nasonis Tristium Liber IV, Commento exegetico instructus (Groningen, 1951), 3–4 Google Scholar; Booth, J., ‘Aspects of Ovid's language’, ANRW 2.31.4 (1981), 2686–700, esp. at 2688, 2700Google Scholar; E.J. Kenney, ‘Ovid's language and style’, in Weiden Boyd (n. 1), 27–89, at 29, 42, 44, 45, 65–6.

40 See e.g. Burmann, P. (ed.), Publii Ovidii Nasonis Opera omnia. Tom. 3, Pars 2: Tristium Lib. V. Ex Ponto Lib. IV (Amsterdam, 1727), 503 Google Scholar; Güthling, O. (ed.), P. Ovidi Nasonis carmina in exilio composita, Tristium libri, Ibis, Epistulae Ex Ponto, Halieutica (Leipzig, 1884), 20 Google Scholar. Traill (n. 12), 504 is also aware that this variant was ‘often adopted’.

41 R. Ehwald (ed.), Ovidii Tristia, in P. Ovidius Naso, ex iterata R. Merkelii recognitione, Vol. III: Tristia. Ibis. Ex Ponto Libri. Fasti (Leipzig, 1884), 22 ad loc.

42 Ehwald (n. 1). And thus he was, in all probability, highly motivated to maintain the superiority of its readings over others.

43 Ehwald and Levy (n. 4).

44 Owen (n. 6), ad loc.

45 Owen (n. 1 [1915]), ad loc.

46 Luck (n. 12 [1959]), 103; Luck (n. 3), 13.

47 On the poor textual tradition of the Tristia in general, see Luck (n. 12 [1959]), 100; Luck (n. 17), 243; Luck (n. 3), 11; Luck (n. 12 [1977]), 9; Diggle (n. 18), 401.

48 Cf. Magnus (n. 12), col. 252.

49 See Riese (n. 14), vii-viii; Luck (n. 3), 12; Tarrant (n. 10), 283 and n. 6; J. Richmond, ‘Manuscript traditions and the transmission of Ovid's works’, in Weiden Boyd (n. 1), 443–83, at 477.

50 Owen himself (n. 1 [1924], 83), while maintaining that TR and L preserve the ‘true reading’ in line 2.8, acknowledged that these two manuscripts also have ‘corruptions peculiar’ to them, and that A G H P V, by virtue of partial kinship with them, ‘are valuable for elucidating the corruptions of those manuscripts’. Perhaps, then, Owen should have applied this insight here, especially as he identified corruptions in the reading of TR and L in the two immediately adjacent lines (2.7 and 2.9).

51 Traill (n. 12), 504.

52 For which, also see Tr. 2.219–40, and cf. S.G. Nugent, ‘Tristia 2: Ovid and Augustus’, in Raaflaub, K.A. and Toher, M. (eds.), Between Republic and Empire: Interpretations of Augustus and his Principate (Berkeley and LA, 1990), 239–57, at 251–2Google Scholar; Barchiesi, A., Il poeta e il principe: Ovidio e il discorso Augusteo (Rome and Bari, 1994), 22 Google Scholar; Ingleheart (n. 18), 207.

53 Or, indeed, both senses at the same time.

54 Cf. e.g. Green (n. 18), xxiv; Volk, K., Ovid (Chichester and Malden, MA, 2010), 31, 101CrossRefGoogle Scholar.

55 Luck (n. 12 [1959]), 101 and (n. 3), 14, following Magnus (n. 12), col. 250, does not attribute a special meaning connected with libraries to Tr. 2.8. Luck (n. 12 [1959]), 102 presents emissa as meaning ‘published’. Traill (n. 12) briefly mentions that iam demi iussa (which he does not accept, but maintains is superior to iam demum) ‘… was held to be a reference to Augustus’ removal of Ovid's “Ars” from all public libraries’. However, he does not discuss the ramifications of his emendation (tum demum inuisa) on the issue.

56 I.e. all except TR and L.