¹ Wili, Walter, Horaz und die augusteische Kultur (Basel 1948) 182Google Scholar, Nisbet, R.G.M. and Hubbard, Margaret, A Commentary on Horace: Odes Book One (Oxford 1970) 72–80Google Scholar, Jones, C.P., ‘Tange Chloen semel arrogantem’, HSCP 75 (1971) 81–83Google Scholar, Deitmer, Helena, Horace. A Study in Structure (Hildesheim-Zürich-New York 1983) 140–155Google Scholar, Santirocco, Matthew, Unity and Design in Horace’s Odes (Chapel Hill-London 1986) 143Google Scholar, Lefèvre, Eckard, Horaz, Dichter im augusteischen Rom (München 1993) 233Google Scholar. Quinn, Kenneth, ‘A Reading of Odes 1.5Google Scholar’, Arion 2 (1963) 68–69Google Scholar and West, David, Reading Horace (Edinburgh 1967) 100–107Google Scholar leave the similarity between these odes out of their consideration.

² I read deae at 1.5.16Google Scholar, following Thaddäus Zieliński, ‘Marginalien’, Philologus 60 (1901) 2.Google Scholar Nisbet-Hubbard in a lengthy note ad locum argue in favour of emending, citing as the main reason the appropriateness of Venus in her double function as goddess of the sea and of love at the end of Horace’s extended metaphor of shipwreck on the ‘sea of love’. The use of Venus in this double capacity was a source of wit in Hellenistic epigram (Nisbet-Hubbard compare AP 5.11, 9.143Google Scholar and Philodemus AP 10.21Google Scholar.5 ff.), on which Horace obviously drew throughout this poem (see below). The reading deae is, however, rejected by Kinsey, T.E., ‘Horace, Odes 1.5.16’Google Scholar, Lato-mus 23 (1964) 502–505Google Scholar, Jones, (n.l above) 83 and Syndikus, Hans-Peter, Die Lyrik des Horaz 1 (Darmstadt 1972) 83.Google Scholar

³ Stressed by Williams, Gordon, Tradition and Originality in Roman Poetry (Oxford 1968) 747.Google Scholar

⁴ Fraenkel, Eduard, Horace (Oxford 1957) 14, 335, 422, 434–435Google Scholar draws particular attention to this characteristically Horatian trait which he rightly distinguishes from ‘“Irony” last expedient of despairing commentators’ (457).

⁵ Cairns, Francis, Generic Composition in Greek and Roman Poetry (Edinburgh 1972) 80–82Google Scholar. Jones (n.l above) makes most of the parallelism between the poems but does not consider 1.5 a renuntiatio amoris. Stinton, T.C.W., ‘Horatian Echoes’, Phoenix 31 (1977) 164–165CrossRefGoogle Scholar, now in Collected Papers on Greek Tragedy (Oxford 1990) 375Google Scholar, disagrees: ‘in both of them Horace resigns from the works of Venus.’ I agree with Stinton in considering both poems renuntiationes amoris, while the difference between them lies in the fact that in 1.5 Horace has given up Pyrrha but in 3.26 he is giving up all love affairs before re-entering in the game just once more.

⁶ R.G.M. Nisbet and Margaret Hubbard (n.l above) 72-73.

⁷ Nisbet-Hubbard, compare Mimnermus AP 5.160Google Scholar and PhilodemusAP 5.112Google Scholar, both addressed to the lover’s rivals.

⁸ Williams, Gordon, The Third Book of Horace’s Odes (Oxford 1969) 133Google ScholarNikias, comparesAP 6.127Google Scholar where a dedicated shield tells its story and Syndikus, Hans Peter, Die Lyrik des Horaz 2 (Darmstadt 1973) 224Google Scholar n.2 citesAP 6.1Google Scholar, PhilippusAP 6.5,6.38, 6.94Google Scholar, AntiphilusAP 6.95Google Scholar. See also Reitzenstein, Richard, Aufsätze zu Horaz. Abhandlungen und Vorträge aus den Jahren 1908-1925 (Darmstadt 1963) 11–13Google Scholar, Pasquali, G., Orazio lirico (Firenze 1920) 498–499Google Scholar. On the problem of the evidence of later epigram see Hubbard, Margaret, Propertius (London 1974) 13–14.Google Scholar

⁹ Williams, (see previous note) also compares LeonidasAP 6.13Google Scholar. The mechanism of combining a number of related epigrams in one ode has been recognised as ‘die Charakteristik des horazischen Dichtens’ (Syndikus [see previous note] 225 n.11).

¹⁰ Most notably Jacoby, Felix, ‘Zur Entstehung der römischen Elegie’, RhM 60 (1905) 38–105Google Scholar, Smith, Kirby Flower, The Elegies of Albius Tibullus (New York 1913) 23Google Scholar with older literature in n.l, Archibald Day, The Origins of Latin Love-Elegy (Oxford 1938) 106–111Google Scholar, Puelma, Mario, ‘Die Vorbilder der Elegiendichtung in Alexandrien und Rom’, MH 11 (1954) 101–116Google Scholar, Fraenkel (n.4 above) 182 n.4, 197, 201, Wimmel, Walter, Der frühe Tibull (München 1968) 245–248Google Scholar, Williams, Gordon (n.3 above) 181-182, 192-193, 207-208, Christoff Neumeister, Tibull. Einßhrung in sein Werk (Heidelberg 1986) 150–152Google Scholar, Nisbet-Hubbard (n.l above) xiii-xiv, 72-73, 262-263, 290-291,318.

¹¹ Comager, Steele, The Odes of Horace. A Critical Study (New Haven-London 1962) 67–69Google Scholar, Poschl, Viktor, ‘Die Pyrrhaode des Horaz (c. 1,5)’, in Hommages à J. Bayet, Collection Latomus 70 (1964) 581–582.Google Scholar

¹² Syndikus (n.2 above) 82 n. 12.

¹³ There may be some irony in the fact that Pyrrha’s name implies the very opposite of ‘sea’, namely fire, Greek πυρ. Horace also seems to be playing with this etymology when he calls her flava (4); see Nisbet-Hubbard on 1.5.4.

¹⁴ This in turn contrasts with aurea (9) and nites (13).

¹⁵ On this see Verdenius, W.J., ‘Semonides über die Frauen. Ein Kommentar zu Fr. 7’, Mnemosyne 21 (1968) 132–158CrossRefGoogle Scholar; other instances of the motif are found at Trag, adesp. 187 Nauck² (possibly Euripides), Cercid. 5 Powell, MeleagerAP 5.156Google Scholar; cf. Syndikus (n.2 above) 82 n.14.

¹⁶ Meleager, AP 5.190Google Scholar, 12.84, 157, 167, Philodemus AP 10.21Google Scholar.6, Prop. 2.12.7-8, 2.25.27, 3.24.15-16 with Fedeli’s note, Ov. Am. 2.9.33Google Scholar; cf. Penna, A. La, ‘Note sul linguaggio erotico dell’elegia latina’, Maia 4 (1951) 202–205Google Scholar, Syndikus (n.2 above) 82 n.14.

¹⁷ Cf. McKeown, James C., Ovìd. Amores. Text, Prolegomena and Commentary in four volumes 2 (Leeds 1989) 259.Google Scholar

¹⁸ Syndikus (n.8 above) 225.

¹⁹ The lover, of course, often finds himself locked out by the beloved, e.g. Callimachus, AP 5.23Google Scholar(HE 1327-1332), Meleager AP 5.191Google Scholar (HE 4378-385), Hor. Carm. 1.25, 3.7, 3.10, Prop. 1.16, Tib. 1.2, Ov. Am. 1.6; cf. Copley, Frank O., Exclusus Amator. A Study in Latin Love Poetry (Ann Arbor 1956).Google Scholar In this situation force is often applied in love-poetry to try to open the door: Theocr. 2.128, Tib. 1.10.54 with Smith and Murgatroyd, Prop. 2.5.22, Ov. Am. 1.6.57-58 with McKeown.

²⁰ et arcus has been under suspicion since Bentley conjectured securesque, which Shackleton Bailey in his 1985 Teubner glosses as fartasse recte, also recording the conjectures et uncos (Bisconi) and aduncos (Giangrande). Bentley’s securesque on my analysis is preferable to Giangrande’s aduncos because it keeps the military imagery. Bisconi’s et uncos also preserves this.

²¹ This feature has been demonstrated to be a characteristic of a number of Horatian love-poems by Stinton, Collected Papers (n.5 above) 376–377.Google Scholar

²³ On this general problem in Horace see Rudd, Niall, Lines of Enquiry (Cambridge 1976) 145–181CrossRefGoogle Scholar and in the Cambridge History of Classical Literature 2 (Cambridge 1982) 370–404.Google Scholar. Prof. Niall, Rudd also suggests to me that there may be a double entendre here on aura meaning ‘breeze’ (OLD 1, 2)Google Scholar or ‘fragrance’ (OLD 6).Google Scholar That would put the perfume-drenched puer (2) on a par with Pyrrha.

²⁴ Dettmer (n.l above) 153 sees an ‘ironic twist’ but thinks that Horace undercuts his role as praeceptor amoris in 1.5. There is nothing in 1.5 that indicates that Horace is playing the role otherwise well exemplified in Tib. 1.4, Prop. 2.4, 4.5 and Ov. Am. 1.8 and Ars and Rem. passim.

²⁵ This also applies to Ode 1.33 where he is part of the never-ending love-circuit at vv. 13-16.