Although I have misgivings about adding to the already vast bibliography on poem 68 (see esp. H. W. Prescott, TAPA 71 (1940) 473–500, T. P. Wiseman, Cinna the poet (1974) 77–103, C. J. Tuplin, CQ 31 (1981) 113ff.), I nevertheless venture to hope that my ‘permutation of answers’ (Wiseman's phrase) is sufficiently different from others to justify its appearance. In the interests of brevity I have preferred to be dogmatic and to restrict my references to earlier scholars. I am not concerned with the vexed question whether 68 is a single poem (for a recent survey of scholarly opinion see G. W. Most, Philol. 125 (1981) 116 n. 31), but it ought to be clear from my discussion that in my opinion it most certainly is not.

I am very grateful to Professor T. P. Wiseman for commenting on an earlier draft of this paper. I have also tried to meet some points which were kindly raised by Professors G. P. Goold and E. J. Kenney.

1. See Wiseman 88–90.

2. Kinsey, T. E. (Latomus 26 (1967) 38 n. 1)Google Scholar divides into 1–14, 15–26, 27–40; accepted by M. B. Skinner in her elaborate treatment of the structure of 68 A (TAPA 103 (1972) 495–512). It is of course the case that the brother's death occupies the centre of the poem, with 14 lines on either side of it; but whatever the significance of this numerical arrangement, it should not in my opinion be allowed to obscure the argument of the poem.

3. Or rather, Catullus writes as if Manlius had described himself in these terms. Naturally we cannot ever know whether Manlius really used this metaphor (nor even whether Manlius wrote a letter in the first place); but what we must deal with is the context which is suggested by Catullus' poem.

4. So e.g. Quinn ad loc. A variation on this view is that Manlius had indeed made two requests but Catullus chose to interpret them as one and correspondingly to make a single response in the form of a love poem, viz. 68 B (so Ellis ad loc., Prescott 497–500). Skinner (500 n. 13) adheres to the view put forward by Jachmann, (Gnomon 1 (1925) 211)Google Scholar that two separate types of poetry were requested, viz. learned and erotic.

5. R. G. M. Nisbet has emended utriusque to hucusque on the grounds that ‘the gifts of the Muses and Venus are one and the same’ (PCPS 24 (1978) 105); but this seems to me to beg the question. In his recent edition of Catullus (Duckworth, 1983) G. P. Goold accepts Nisbet's emendation.

6. So Prescott 478–9, Most 121.

7. Fordyce and Quinn ad loc. both stress that cum = ‘when’ but draw no conclusions.

8. According to Ellis on line 33 this is the ‘common view’; but it has been held, so far as I know, by no modern scholar except Kinsey (38–40), whose arguments are different from mine.

9. It might be objected that it is unflattering to ask a poet to send poetry which is not his own, and it cannot be denied that there existed a familiar convention whereby writers professed that they wrote their work(s) in response to a particular request (see e.g. Sussman, L. A., The Elder Seneca (1978) 53–4)Google Scholar. To this objection I can only reply that, while lines 7–8 are certainly capable of meaning that Manlius had asked Catullus to write him a new poem, they are also capable of the meaning which I have put forward and that this latter meaning is confirmed by Catullus' response to Manlius' request in line 33 (which I proceed to discuss in the remainder of the above paragraph). In other words, we have here an example of an initial ambiguity being resolved by a statement later in the poem (for such e sequentibus praecedentia see Williams, G., Figures of thought in Roman Poetry (1980) 102ff.)Google Scholar.

10. So e.g. Fordyce ad loc. Wiseman (90 n. 15) takes nam to introduce a new argument and refers to Poyser, G. H., CR 2 (1952) 8–10Google Scholar. But I do not think there is any essential difference between the usages to which Fordyce and Poyser allude: nam explains the difference between listed items which are apparently similar: in the present passage the similarity is that Catullus declines both requests, the difference is that in the first case the reason is his brother's death whereas in the second it is the lack of books. For nam see also Pease on Cic. ND 1.27.

11. In addition, Kinsey well observes (39): ‘The absence of books … is a poor excuse for not writing a poem and since Catullus had a perfectly good excuse to hand, namely the death of his brother, which he had used before (65, 1–4), why should he use a poor excuse here? It is more reasonable to suppose that Catullus is giving this as an excuse for not doing something for which it is a good excuse, namely not lending … a book’.

12. So e.g. Wiseman 94–6, with earlier references. Kinsey (41–2) and Most (122) suggest that Manlius had requested Catullus as a homosexual partner. There is some attractiveness in this thesis if the munera Musarum are taken to mean Catullus' own poetry, for then the poet would be asked to be personally responsible for both requests. But if the request is not for Catullus' own poetry, as I believe, the theory of homosexuality becomes in my opinion less plausible.

13. For a comparable use of turpis, but in a serious context, see e.g. Cic. TD 4.68 turpes sunt qui ecferunt se laetitia tum, cum fruuntur Veneriis uoluptatibus. The interpretation of lines 27–30 which I put forward here, and which I have not seen recorded elsewhere, can be derived either from the paradosis of 27–9 as punctuated by Wiseman 96–7 (with id in 30 picking up quod … tepefactat in 28–9 rather than quod scribis in 27):

quare quod scribis ‘Veronae turpe, Catulle,

esse', quod hic quisquis de meliore nota

frigida deserto tepefactat mebra cubili…

or from the emended text as found in the OCT:

quare quod scribis Veronae turpe Catullo

esse, quod hie quisquis de meliore nota

frigida deserto tepefactet membra cubili…

The advantage of the latter is that the subjunctive tepefactet leaves the reader in no doubt that quod hic … cubili is part of what Manlius had said; and Catullo would presumably represent Catullus' own rendering into or. obl. of some putative original such as tibi (for a similar case cf. 72.1). But if we prefer the former alternative, quod hic … tepefactat … cubili (indic.) represents Manlius' words in the same way as quem neque … perpetitur nec … oblectant at 5–8, and for the direct speech after quod scribis see Wiseman 97 n. 52. [The third way of presenting the passage is to regard Veronae … cubili as all being in direct speech (so e.g. Quinn); but this means that hic cannot refer to Verona.]

Whether my interpretation of the passage is itself acceptable is naturally something which readers must decide for themselves; but I do not think there is any less plausibility in my concessive rendering of deserto … cubili than in either of the two traditional versions (viz. ‘It is disgraceful for Catullus to be in Verona because in Verona all the best people sleep alone’ [= OCT] or ‘It is disgraceful, Catullus, to be in Verona because in Rome all the best people warm their limbs in the bed you have left’ [= the paradosis as punctuated in Quinn]). On the contrary, the similarity of wording in lines 6 (desertum in lecto caelibe) and 29 (deserto … cubili) suggests that related points are being made in each case (cf. the repetition of the shipwreck metaphor, discussed above); and if Manlius had indeed asked Catullus in a joking fashion to send him a girl (line 10), a humorously unfavourable comparison between his own and Catullus' circumstances, such as that suggested here, would be very much ad rem and is underlined by the frequentative verb tepefactare. It might be objected that it is absurd for Manlius to write from Rome, where girls were no doubt plentiful, to ask Catullus to send him a girl from Verona. But this objection takes no account of the humorous nature of Manlius' request, and there is in any case no indication that Manlius is in Rome: indeed lines 34–5 may well be taken to suggest that he is not (it would be different if istic, not illic, were present in 35).

14. The commonest interpretation is ‘obligation to my host’, the word ‘host’ being explained by reference to 68B.68–9, where we learn that the addressee of that poem lent Catullus a house in which he could meet Lesbia: so e.g. Kroll ad loc. and Williams, G. (Tradition and originality in Roman poetry (1968), 229–30)Google Scholar, who appears to regard this as one of the ‘decisive arguments’ in favour of the unity of poem 68 as a whole. In Figures of thought in Roman poetry (1980) 50 n. 3Google Scholar, Williams says that nothing he has read on the structure and interpretation of poem 68 has persuaded him to abandon his view of its unity, which I find particularly surprising in the light of Wiseman's discussion.

15. For this technical meaning see OLD s.v. 3.

16. So e.g. Kroll ad loc., Most 121.

17. The only possible evidence to the contrary is the phrase delicias animi in line 26: thus Quinn ad loc. says that Catullus ‘adds animi to exclude the meaning “making love”’. Yet most other scholars, no matter what their views on other matters, seem agreed that the expression refers to sensual pleasures, and Prescott even says that the addition of animi ‘removes any ambiguity’ (481 n. 15). I take the phrase to mean something like ‘emotional delights’. For Skinner 501 n. 15 and Macleod, C. W. (CQ 24 (1974) 84–5)CrossRefGoogle Scholar the vocabulary throughout this passage is intentionally ambiguous.

18. This view of haec is mentioned but rejected by Most 121 n. 48 (on the false grounds that nam in 33 precludes it).

19. See the discussion in Prescott 486–7.

20. Kroll ad loc. not unreasonably remarks that the context is decisive in favour of this interpretation.