¹ For the charge of misogyny, see e.g. Highet, G., Juvenal the Satirist (1954), 103Google Scholar; Rogers, K. M., The Troublesome Helpmate: A History of Misogyny in Literature (1966), 41Google Scholar. Critics do not agree upon the theme of Satire 6. Those who see it as a ‘catalogue of women’, include: Ferguson, J., Juvenal: The Satires (1979), 185Google Scholar; Lefkowitz, M. R. and Fant, M. B., Women's Life in Greece and Rome (1982), no. 157Google Scholar; J. E. Carr, ‘The view of women in Juvenal and Apuleius’, CQ 58 (1982), 61; Anderson, W. S., Essays on Roman Satire (1982), 275Google Scholar; D. S. Wiesen, ‘The verbal basis for Juvenal's satiric vision’, in ANRW 11.33.1 (1989), 733. For the satire as a dissuasion from marriage: Highet's chapter-heading (above), 91: ‘Advice to Those About to Marry’; L. I. Lindo, ‘The evolution of Juvenal's later satires’, CPh 69 (1974), 25; D. Singleton, ‘Juvenal 6.1–20 and some ancient attitudes to the golden age’, G&R 19 (1972), 151–64, following Mason, H. A., ‘Is Juvenal a classic?’, in Sullivan, J. P. (ed.), Critical Essays on Roman Literature 2: Satire (1963), 137Google Scholar. Both views in: Coffey, M., Roman Satire (1976), 127Google Scholar; Winkler, M. M., The Persona in Three Satires of Juvenal (1983), 147Google Scholar; Henderson, J., ‘…When satire writes “Woman”’, in Braund, S. H. (ed.), Satire and Society in Ancient Rome (1989), 89–125Google Scholar; idem, ‘Satire writes “woman”: Gendersong’, PCPhS n.s. 25 (1989), 68. Neither view is followed by Courtney, E., A Commentary on the Satires of Juvenal (1980), 252Google Scholar, who seems to regard the poem as a ‘one-off’.

² On the poem's structure no consensus has emerged, as indicated by Anderson's comment (op. cit. (n. 1), 255 with 275 n. 2): ‘Scholars have been divided in their proposed solutions: the brave have assumed a coherent organization; the prudent have abandoned what seemed a thankless and futile effort, denying any structural unity.’ Disagreement about the structure of the poem relates to disagreement about the theme of the poem, see n. 1.

³ For the claimed ordinariness of satire, see e.g. Hor., Sat. 1.4.38–42, Juv. 1.79–80.

⁴ On the relationship of satire, parody, and irony, see Hutcheon, L., A Theory of Parody (1985), esp. 52–68Google Scholar. As M. Fusillo observes (‘Il testo nel testo: la citazione nel romanzo greco’, MD 25 (1990), 27), there is a risk of reducing parody to synonymity with intertextuality.

⁵ e.g. on the relationship between Petronius' Satyrica and the Odyssey, see Cameron, Averil, ‘Myth and meaning in Petronius: some modern comparisons’, Latomus 29 (1970), 400Google Scholar; Richlin, A., The Garden of Priapus. Sexuality and Aggression in Roman Humor (1983), 192Google Scholar. On Horace, Satires 11.5 as parody of the scene in which Odysseus consults Tiresias, cf. Rudd, N., The Satires of Horace (1966), 228Google Scholar. Horace, Satires 11.4 recalls the backdrop of Platonic dialogues, in particular, the Phaedrus: see S. H. Braund, Beyond Anger: A Study of Juvenal's Third Book of Satires (1988), 144 and 247, n. 67. Juvenal's Satire 3 is illuminated by Virgil's first Eclogue: see Witke, C., Latin Satire (1970), 133–4Google Scholar. Satire 4 reworks Statius' panegyrical epic poem, De Bello Germanico, now lost, on Domitian's German campaign. Parody of an epic topos, the storm at sea, appears in Satire 12: see Scott, I. G., The Grand Style in the Satires of Juvenal (1927), 83–8Google Scholar. For an analysis of the ‘literariness’ of Satire 6, see Wiesen, op. cit. (n. 1).

⁶ De Decker, J., Juvenalis Declamans (1913)Google Scholar.

⁷ M. P. O. Morford, ‘Juvenal's thirteenth satire’, AJPh 94 (1973), 26–36; A. D. Pryor, ‘Juvenal's false consolation’, AUMLA 18 (1962), 167–80.

⁸ Cairns, F., Generic Composition in Greek and Roman Poetry (1972), 75Google Scholar.

⁹ Treggiari, S., Roman Marriage (1991), 223Google Scholar.

¹⁰ cf. Stobaeus in Ecl. IV.22 πεϱὶ γάμου (‘concerning marriage’), item 2 ὅτι οὐϰ ἀγαθὸν τὸ γαμεῖν (‘that marriage is not good’). Thus Schuetze, R., Juvenalis Ethicus (1905), 35Google Scholar: ‘noster amicum monet ut matrimonio desistat’. Courtney, op. cit. (n.:), 252, raises this as a possibility, but then pulls back: ‘no firm links can be forged with the rhetorical tradition as it survives’.

¹¹ For a feminist reading of Semonides, see Loraux, N., Les enfants d'Athéné (1981), 95–117Google Scholar.

¹² e.g. Semonides, passim; Hesiod, Theogony 594–9 (women as drones); Phocylides quoted in Stobaeus IV.22.1 (bitch, bee, sow, and mare).

¹³ The trite comparison with a bereaved tigress; orba tigride (270); the woman addressed as a most savage viper, saeuissima uipera (641); Messalina in the brothel calling herself Lycisca (123, an allusion, via the Greek word for wolf, to lupa, a prostitute); and gannit (64), a word typically used to describe a dog's whimper, see OLD gannio 1. I exclude ‘rara auis in terris nigroque simillima cyeno’ (165): the expression is proverbial (Otto, A., Die Sprichwörter und sprichwörtlicker Redensarten der Römer (1890)Google Scholar, auis 2), used by Seneca in a similar context, ‘si bona fuerit et suauis uxor, quae tamen rara auis est’ (Jerome, Adv. Iouin. 1.47 = Sen. fr. 56).

¹⁴ e.g. comparing a woman's breasts to a mare's teats, Horace, Epode 8.7–8. On invective against women which uses animal comparisons, see Richlin, A., ‘Invective against women in Roman Satire’, Arethusa 17 (1984), 70–1Google Scholar and esp. Martial 3.93, using comparisons with animals and insects in an attack on an old woman.

¹⁵ e.g. Ar., Lysis. 195–208, Thesmo. 556–7, 733–62, Eccl. 43–5, 132–43, 1112–24.

¹⁶ Pliny, NH XIV. 89–90, for examples of traditional disapproval of drinking by women, cf. Valerius Maximus VI.3.9. Livy 1.57.9 for Lucretia as the epitome of abstemiousness. Pliny (loc. cit.) and Gellius X.23.1–3 record Cato's view that it was male kinsmen's wish to check whether or not a woman had indulged in secret drinking that caused them to kiss close female relatives on the lips. Cf. too Plut., Mor. 265b on the ius osculi.

¹⁷ Gellius X.23.3; cf. ibid. 4–5; Richlin, op. cit. (n. 14), 78 n. 10.

¹⁸ On women's consumption of wine, see Wissowa, G., Religion und Kultus (2nd edn, 1971). 217Google Scholar; de Cazanove, O., ‘Exesto: L'incapacité sacrificielle des femmes à Rome (à propos de Plutarque Quaest. Rom. 85)’, Phoenix 41 (1987), 159–61Google Scholar. In the Bona Dea rites we are told that the wine-bowl was referred to as a honey-pot and the wine as milk, Plut., Mor. 268d-e, cf. 20, Macr., Sat. 1.12.25. I am grateful to Nicholas Purcell for this observation.

¹⁹ On seruitium amoris, see F. O. Copley, ‘Seruitium Amoris in the Roman Elegists’, TAPhA 68 (1947), 285–300, amplified by Murgatroyd, P., ‘Seruitium Amoris and the Roman elegists’, Latomus 40 (1981), 589–606Google Scholar.

²⁰ See Howell, P., A Commentary on Book One of the Epigrams of Martial (1980)Google Scholar, on Mart. 134.7 for references.

²¹ Similarly in Satire 5 amicitia is announced as the central topic at 1. 14: both poems allege the disappearance and destruction of their central concept. On the prominence of the theme of amicitia in Juvenal, see R. Seager, ‘Amicitia in Tacitus and Juvenal’, AJAH 2 (1977), 40–50 and LaFleur, R. A., ‘Amicitia and the unity of Juvenal's First Book’, Illinois Classical Studies 4 (1979), 158–77Google Scholar.

²² See Lattimore, R., Themes in Greek and Latin Epitaphs (1942), 295–6Google Scholar; cf. Virg., Georg. 2.524, ‘casta pudicitiam seruat domus’, part of the praise and idealization of country life, see R. Thomas, F., Virgil Georgics (1988)Google Scholar, ad loc.; cf. Hor., Od. III.5.41; Liv. III.45.6; Sen., Ag. 110; for the inverse, cf. Sall. B.C. 13.3, ‘mulieres pudicitiam in propatulo habere’, on the decadence of modern morals; and the curse uttered by Ovid, Ibis 349–50, ‘nee tibi contingat matrona pudicior’. For pudicitia as mulieris propria uirtus, see Jerome's quotation of Seneca, adv. Iovin. 1.49.

²³ e.g. 6.193, sermo pudicus; 137, pudicam; 49, capitis matrona pudici; pudor at 252 and 357; cf. 287, castas.

²⁴ Well shown by Singleton, op. cit. (n. 1).

²⁵ Juvenal evokes Propertius 11.32, complaining initially about Cynthia's suspected infidelities but then accepting the fact that Roman girls long ago gave up pudicitia. Propertius' poem also names Lesbia (45) and refers to the Saturnian Age (52); see Mason, op. cit. (n. 1), 136–7.

²⁶ See Hesiod, WD 197–201 and Aratus 96–136; cf. Ovid., Met. 1.149–50.

²⁷ See Hesiod, WD 106–201 and Aratus 96–136.

²⁸ cf. Hor., Sat. 1.3.104–6: ‘abhinc absistere bello, oppida coeperunt munire et ponere leges, ne quis fur esset, neu latro, neu quis adulter’ (‘Thereafter they began to avoid war, to build towns, and to pass laws making it an offence for any person to engage in theft, armed robbery, or adultery’).

²⁹ Lucilius provides a satiric precedent on the madness of marrying: ‘qua propter deliro et cupidi officium fungor liberum’ (646W) (‘wherefore do I go mad and do the duty of a man eager for children’).

³⁰ The translation of Juvenal is Niall Rudd's (1992). On the Julian law see below, Section IV. The meat-market is portrayed as taking the place of the human legacy-hunters who feature in Roman satire — most obviously in Hor., Sat. 11.5 and Juv., Satire 12; also Juv. 5.98, 10.202.

³¹ cf. Sen., Ep. 94.26, ‘scis improbum esse qui ab uxore pudicitiam exigit, ipse alienarum corruptor uxorum’ (‘you know that he who demands chastity from his wife but is himself the seducer of others' wives is unreasonable’).

³² See McKeown, J. C., ‘Augustan elegy and mime’, PCPhS n.s. 25 (1979), 71–84Google Scholar. On the dangers allegedly incurred by adulterers caught in the act, see Richlin, A., ‘Approaches to the sources on adultery at Rome’, in Foley, H. P. (ed.), Reflections of Women in Antiquity (1981), 394Google Scholar.

³³ On the univira, see Pomeroy, S. B., Goddesses, Whores, Wives, and Slaves (1975), 161Google Scholar; Lattimore, op. cit. (n. 22), 296 n. 251 and Williams, G., ‘Some aspects of Roman marriage ceremonies and ideals’, JRS 48 (1958), 23–4Google Scholar. Certain rituals were reserved to uniuirae: Gagé, J., Matronalia (1963), 59–60, 120–2Google Scholar. Cf. n. 74 below.

³⁴ On the Romans' horror of adultery between women of high status and men of lower status, see Treggiari, op. cit. (n. 9), 308 and cf. n. 85 below. For an example, Plin., Ep. VI.31.4–6.

³⁵ Thus Livy X. 23.

³⁶ Juvenal here incorporates the classic Roman explanation of the decline in morals, cf. Sall., B.C. 10; Earl, D. C., The Moral and Political Tradition of Rome (1967), 17–19Google Scholar.

³⁷ For ‘in … uices equitant’ of the sexual act, here between women: Adams, J. N., The Latin Sexual Vocabulary (1982), 166Google Scholar.

³⁸ For the legal framework, see Gardner, J. F., Women in Roman Law and Society (1986), 77–8Google Scholar.

³⁹ Not surprisingly, the words for husband (uir, maritus) and for adulterer/adulteress (adulter and moechus/moecha) occur frequently in the poem: uir fifteen times; maritus eighteen times; adulter five times; moechus five times out of a total of eleven instances in Juvenal's poems and moecha once from a total of twice in Juvenal's poems.

⁴⁰ puluinar denoted the bed of the emperor, as quasi-divine being, OLD puluinar b, e.g. Suet., Dom. 13.1, Sen., Dial. XI.16.4.

⁴¹ e.g. Suet., Ner. 26.

⁴² cf. 6.330, 8.145 nocturnus adulter disguised Santonico … cucullo.

⁴³ cf. 14.30 where the cinaedi of the household abet the lady's adultery by carrying messages.

⁴⁴ The wording recalls 2.8, ‘frontis nulla fides’. In his description of the disguised adulterer (O 21–2), Juvenal reworks several details from Satire 2.93–8, a passage which portrays passive homosexuals and effeminates staging their own Bona Dea rites, e.g. fuligine as eye makeup O 21, cf. 2.93–5; reticulatus O 22, cf. 2.96; note the incidence of yellow fabrics in both (croceis, galbina).

⁴⁵ adulter/adulterium was derived by Romans from ad + alter: e.g. Papinian, D. XLVIII.5.6.1 ‘proprie adulterium in nupta committitur, propter partum ex altero conceptum composite nomine’ (‘strictly speaking adultery is committed with a married woman, the name being derived from children conceived by another’), cf. R. Maltby, A Lexicon of Ancient Latin Etymologies (1991), s.v.

⁴⁶ cf. e.g. Plaut., Miles Gl. 703–15.

⁴⁷ See Hor., Od. III.24.19–20; Eur., Phaethon 158–9, with the commentary of J. Diggle; Arist., NE 8.1161a1; Mart. VIII. 12. The uxor dotata was a stock character-type in Roman comedy, Hunter, R. L., The New Comedy of Greece and Rome (1985), 90–2Google Scholar, E. Schuhmann, ‘Der Type der uxor dotata in den Komodien des Plautus’, Phil. 121 (1977), 45–65.

⁴⁸ Donatus says that Terence departs from usual practice in presenting on stage a mother-in-law who is a noble character (ad Hecyram 198 and 774).

⁴⁹ Σ: ‘simulat aegritudinem socrus, ut habeat facultatem ad se filia ueniendi causa adulterii’ (‘the mother-in-law feigns illness, so that her daughter has an opportunity of visiting her for the purpose of adultery’); cf. Ov., Am. 11.2.21 and Mart. XI.7.7, both of visits of a friend.

⁵⁰ I suggest that the following sequence of sections be adopted: cinaedi (Oxford fragment), eunuchs (366–78), Ogulnia (350–65), singers (379–97): the Ogulnia section introduces discussion of singers by its mention of public entertainments. Contrast the OCT, which has the sequence Ogulnia (350–65), cinaedi (Oxford fragment), eunuchs (366–78), singers (379–97); contrast too Martyn's text (J. R. C. Martyn, D. IVNI IVVENALIS SATVRAE (1987)), where the sequence is cinaedi (Oxford fragment), Ogulnia (350–65), eunuchs (366–78), singers (379–97). Martyn and I agree, however, in placing the Oxford fragment after 345 and on the excision of 346–8 (following Ribbeck and Clausen) as a doublet of O 30–2.

⁵¹ On this aspect of Dido, see Williams, op. cit. (n. 33), 23–4.

⁵² For cosmetics as a standard topic, cf. Ov., Rem. Am. 351–6, Medic. Fac., Lucian, Ἔϱωτες, 38–41.

⁵³ cf. Liv. XXXIV. 1–8, Val. Max. ix.1.3, Tac., Ann. III. 34, Orosius IV. 20. 14, Zonaras IX. 17.1; G. Rotondi, Leges publicae populi Romani (1912, repr. 1966), 254, Culham, P., ‘The Lex Oppia’, Latomus 41 (1982), 786–93Google Scholar and idem, ‘Again, what meaning lies in colour!’, ZPE 64 (1986), 235–45.

⁵⁴ cf. Lucil. 534–5W, ‘cum tecum es, quiduis satis est; uisuri alieni sint homines, spiram pallas redimicula promit’ (‘when she is with you, anything will do; should other men be coming to see her, she brings out her chinribbons, her mantles, her headbands’).

⁵⁵ Indicated by the location, the temple of Isis (489); cf. Sat. 9.22–5, note moechus 25.

⁵⁶ cf. Mart. 2.66, Ov., Am. 1.14.16, A.A. 3.239.

⁵⁷ The language here recalls Domitian's consilium in Satire 4: sententia, 498, cf. 4.136; censebunt, 500, cf. censes, 4.130; ‘tamquam famae discrimen agatur aut animae’ (500–1), cf. ‘tamquam …’ 4.147–8.

⁵⁸ The expression ‘uiolato … cadurco’ (the sex act has ‘profaned the coverlet’ 537) seems to imply adultery.

⁵⁹ Only a few topics in the poem are not explicitly linked with adultery: the beautiful wife so expert at spending her husband's money (142–60); the proud wife (161–83, though ‘intactior omni … Sabina’ 163–4 introduces the theme of chastity: the Sabine women were examples of pudicitia, cf. Juv. 10.297–9); the woman who uses lewd speech (184–99); women who enter the law-courts (242–5, unless the charge on which women are defendants in court is adultery).

⁶⁰ Syme, R., Roman Papers III (ed. Birley, A. R., 1984), 1134Google Scholar.

⁶¹ Cairns, op. cit. (n. 8), 75; cf. Clark, D. L., Rhetoric in Greco-Roman Education (1957), 177ffGoogle Scholar and D. A. Russell, ‘Rhetors at the wedding’, PCPhS n.s. 25 (1979), 106 on progymnasmata; the latter discusses the epithalamium specifically.

⁶² [Dion. Hal.] Ars 261 (translated in Russell, D. A. and Wilson, N. G. (eds), Menander Rhetor (1981), Appendix 362–81)Google Scholar; cf. Russell and Wilson ad Men. Rhet. 400.32ff. for some other references.

⁶³ e.g. the title of Sen., Contr. 6.6; adultera uenefica (‘the adulteress who was a poisoner’); in Suas. 2.21 a controuersia is mentioned ‘about the woman who argued before matrons that children should not be reared and is therefore accused of harming the state’; the outline of Contr. 2.7, ‘A man with a beautiful wife went off abroad. A foreign trader moved into the woman's neighbourhood. He three times made her propositions of a sexual nature, offering sums of money. She said no. The trader died, leaving her all his wealth in his will, to which he added the clause: “I found her chaste.” She took the bequest. The husband returned and accused her of adultery on suspicion.’ See De Decker, op. cit. (n. 6), 23–9 for some detailed congruences between the Controversiae and Satire 6.

⁶⁴ On pudicitia in Valerius Maximus and Livy, see Fantham, E., ‘Stuprum: public attitudes and penalties for sexual offences in Republican Rome’, Echos du Monde Classique n.s. 10 (1991), 273–82Google Scholar. On Livy's political engagement, see Treggiari, op. cit. (n.9), 212, Richlin, op. cit. (n. 32), 383, and on the value of Lucretia as an exemplum, see Richlin, op. cit. (n. 14), 68.

⁶⁵ On the progymnasmata of Theon, see Russell and Wilson, op. cit. (n. 62), xxv–xxvii.

⁶⁶ cf. Sulpicius Victor, Inst. Or. 3, Hermogenes, Prog. 11; and Aphthonius (late fourth/early fifth century), Prog. 13 who discusses the desirability of marriage as a theme for rhetoric.

⁶⁷ For other, shorter, passages on the theme of town and country in Roman satire, see Braund, S. H., ‘City and country in Roman satire’, in Braund, S. H. (ed.), Satire and Society in Ancient Rome (1989), 43–7Google Scholar.

⁶⁸ One might surmise that the theme of Juvenal's incomplete sixteenth satire was similarly a comparison of the soldier's life with the lawyer's life.

⁶⁹ Similarly Cairns, op. cit. (n.8), 38–49 categorizes Satire 3 as an ‘inverse’ syntaktikon (the farewell of a departing traveller); within this framework the poem delivers a dissuasion from city life, cf. Braund, op. cit. (n. 67), 23–8.

⁷⁰ For the use of Menander Rhetor to illuminate the genres of rhetoric at earlier periods, see Cairns, op. cit. (n. 8), 34–75, cf. DuQuesnay, I. M. LeM., ‘Vergil's First Eclogue’, in Cairns, F. (ed.), Papers of the Liverpool Latin Seminar III (1981), 53ff.Google Scholar

⁷¹ cf. Hesiod, WD 235 with West, M. L., Hesiod. Works and Days (1978)Google Scholar, ad loc.; Cat. 61.221–5, Hor., Od. IV.5.23, Mar. VI.27.3–4, Chanton II. 11.2, and in epitaphs EG 243b and CE 387.8–11 cited by Lattimore, op. cit. (n. 22), 276–7; contrast Mart. VI.39 ‘in grabatis tegetibusque concepti materna produnt capitibus suis furta’ (‘creatures conceived on truckle-beds and mats betray by their features their mother's adulteries’, ll. 4–5). A fragment of Seneca praises the pudica as not spoiling her ancestors’ blood by clandestine offspring, Jerome adv. Iovin. 1.49).

⁷² The inversion is most obvious at 597–8 where the husband is urged to administer an abortion-inducing drug to his wife, to prevent him from being presented with a child who does not resemble him.

⁷³ See in general Berger, P. L. and Luckmann, T., The Social Construction of Reality (1967)Google Scholar and on Roman antiquity Foucault, M., The History of Sexuality 3 The Care of the Self (trans. Hurley, R., 1988)Google Scholar and Veyne, P. ‘The Roman Empire’, in Aries, P. and Duby, G. (eds), A History of Private Life I (trans. Goldhammer, A., 1987), 5–234Google Scholar.

⁷⁴ cf. also Lattimore, op. cit. (n. 22), 295–7; Libanius, Decl. 26.9 (6 p. 516 Foerster) on the praises of a prospective wife. Note Juvenal's reference here (6.230) to funerary monuments, ‘titulo res digna sepulchri’ (‘a feat which should be carved on her tombstone’).

⁷⁵ Important material is preserved through the copious quotations and references in Jerome's treatise Aduersus Iouinianum 1.41–9; see Schuetze, op. cit. (n. 10), 35–44, also Epicurus fr. 19 Usener. Foucault, op. cit. (n. 73), 145ff., for the pronouncements on marriage of the philosophical schools. Seneca's De Matrimonio presumably supported marriage; its themes and arguments were probably inverted by Juvenal in Satire 6 (on his likely acquaintance with other works of Seneca e.g. De Ira, see Anderson, op. cit. (n. 1), 293–361, esp. 315, 341). Many points in the γαμιϰὰ παϱαγγέλματα addressed to the bride and groom by Plutarch (Moralia 1383–1463) appear in Juvenal's poem in inverted form, as allegations against women.

⁷⁶ Conveniently listed by Courtney, op. cit. (n. I), 261; cf. van Wageningen, J., ‘Seneca et Iuvenalis’, Mnemosyne 45 (1917), 417–29Google Scholar.

⁷⁷ Courtney, op. cit. (n. 1), 252.

⁷⁸ Dion. Hal. 11.25.6, Plut., Rom. 22.3.

⁷⁹ Gellius X. 23.4–5.

⁸⁰ cf. Juv. 2.40, ‘tertius e caelo cecidit Cato’, ‘a third Cato has dropped from the sky’; 3.314 where Umbricius speaks wistfully of the days when Rome experienced so little crime that it was ‘satisfied with a single prison’ (‘uno contentam carcere’); 5.108–12 where the speaker longs for the ordinary courtesy of patron–client relationships which (allegedly) pertained in the days of Senaca, Piso, and Cotta. On the old morality presented here see n. 112 below.

⁸¹ Lex Iulia de adulteriis coercendis (18 B.C.) and Lex Papia Poppaea (A.D. 9). For full discussion see Gardner, op. cit. (n. 38), 127–31, Treggiari, op. cit. (n. 9), 277–98 and Edwards (see n. 82).

⁸² C. Edwards in ch. I of her forthcoming book, The Politics of Immorality in Ancient Rome (Cambridge), discusses the Augustan legislation, in particular its function as symbolic discourse and its ambivalence. I am most grateful to Catharine Edwards for making her manuscript available to me. Cf. also Richlin, op. cit. (n. 32), 381 and now Fantham, op. cit. (n. 64), 267–91, an examination of attitudes to adultery.

⁸³ On patria potestas, see Gardner, op. cit. (n. 38), 5–11 and on marriage cum manu and sine manu, see A. Watson, The Law of Persons in the Later Roman Republic (1967), 10–27 and Gardner, 11–14.

⁸⁴ Treggiari, op. cit. (n. 9), 299–309; cf. Simone de Beauvoir, The Second Sex (trans. Parshley, H. M., 1988), 221–2Google Scholar.

⁸⁵ cf. n. 34 above. Cf. Richlin, op. cit. (n. 32), 385 on the extra opprobrium in such cases.

⁸⁶ Levick, B., ‘The Senatus Consultum from Larinum’, JRS 73 (1983), 114Google Scholar connects legislation on marriage and adultery with that on public performance and infamia, perceiving ‘a nexus of measures in the early Principate to …strengthen the existing social structure and keep its strata distinct… and to demonstrate acceptable canons of behaviour’.

⁸⁷ A. Wallace-Hadrill, ‘Family and inheritance in the Augustan marriage-laws’, PCPhS n.s. 27 (1981), 59; cf. Veyne, P., ‘La famille et l'amour à Rome’, Annales 33 (1978), 39–40Google Scholar.

⁸⁸ cf. Gardner, op. cit. (n. 38), 77–8.

⁸⁹ See Mart. 6.2, 6.4, ‘censor maxime… plus debet tibi Roma, quod pudica est’ (‘greatest of censors … yet more Rome owes you in that she is chaste’), 6.7.

⁹⁰ See BMCRE p. 355, nos 911, 912, 913, p. 537 nos 1877 and 1878, p. 540, no. 1899.

⁹¹ Syme, R., Tacitus (1958), 500Google Scholar, convincingly argues that Juvenal's poems were written during the years A.D. 115–130 and later reiterates, op. cit. (n. 60), 1125 n. 37, that ‘there are no valid reasons for supposing that Juvenal had published anything before 117’.

⁹² See E. S. Ramage, ‘Juvenal and the establishment: denigration of predecessors in the ‘Satires’, in ANRW 11.33.1 (1989) 640–707. For the role played by satire in articulating paradigmatic imperial ideology, see S. H. Braund, ‘Paradigms of power: Roman emperors in Roman satire’, in K. Cameron (ed.), Humour and History (forthcoming).

⁹³ On the Hadrianic context behind criticism of Domitian in Satire 7, see A. Hardie, ‘Juvenal and the condition of letters: the Seventh Satire’, in F. Cairns (ed.), Papers of the Leeds International Latin Seminar VI (1990), esp. 179–90. Hardie in an unpublished paper which he has generously shown me mounts a similar argument for a Hadrianic back-drop to Satire 3; if he is right, this has important implications for Satire 6, given that Satire 3 is of earlier or contemporary date; cf. Syme, op. cit. (n.91).

⁹⁴ On persona theory Anderson's work is central, op. cit. (n. 1), esp. 3–10; for an excellent restatement of this approach, see the forthcoming study of Horace by K. Freudenberg (Princeton). On narrative, see Winkler, J. J., Auctor and Actor: A Narratological Reading of Apuleius' Golden Ass (1985)Google Scholar.

⁹⁵ The range of tones available are set out in for example Rhet. ad. Herenn. 111.23–7 and the technique of character delineation (notatio) and appropriate dialogue (sermocinatio) at IV.63–5.

⁹⁶ See Braund, op. cit. (n. 5), 197–8. In the case of Juvenal, there is a broad homogeneity within Books I and II (i.e. Satires 1–6), where the persona is essentially an indignant character, while the later books develop an increasingly ironic, detached and cynical persona.

⁹⁷ ‘If’ implies where ‘since’ would assert; evidently the speaker does not dare utter such an assertion (cf. Courtney, op. cit. (n. I), ad 6.166); nevertheless the condition is tacked on at the end so that in Tacitean mode it reverberates longest, quotiens ‘whenever’ at 180 is similar; cf. Wiesen, op. cit. (n. 5), 726.

⁹⁸ Anderson, op. cit. (n. 1), 278–84: angry rhetorical questions and the theme of ‘enduring’ are two classic marks of indignatio.

⁹⁹ Satire 6 is by far the longest satire in extant Roman verse satire: it is nearly 700 lines long (661 + 34 lines in the Oxford fragment) and occupies the whole of Juvenal's second book on its own.

¹⁰⁰ Juvenal uses the same technique of the broken programme at Sat. 1.127, conspicuously not followed by an account of the daily round.

¹⁰¹ L. F. Raditsa, ‘Augustan legislation concerning marriage, procreation, love affairs and adultery’, ANRW 11.13 (1980), 317; Gardner, op. cit. (n. 38), 38–41; Treggiari, op. cit. (n. 9), 398–403. For an example of the age gap, see Quint. IO 6 pr. 5.

¹⁰² Discussed in Braund, op. cit. (n. 5), 3–6.

¹⁰³ On the speaker's assertion that satire can replace tragedy because modern wives exceed the wicked wives of tragedy in their cold-bloodedness, see Bramble, J. C., Persius and the Programmatic Satire (1974), 165Google Scholar and W. S. Smith, ‘Heroic models for the sordid present: Juvenal's view of tragedy’, in ANRW 11.33.1 (1989) 811–23.

¹⁰⁴ The simile has epic antecedents in Homer (Hector: Iliad XIII.136–46) and Virgil (Turnus: Aen. XII.684ff.); here, however, the movement does not come to a halt.

¹⁰⁵ On the failure to allow for hyperbole by those seeking to use satire as source material for Roman social history, see Braund, op. cit. (n. 67), esp. 1–2, 26 with nn. 6–8.

¹⁰⁶ The analogy proposed by O. Weinreich, Römische Satiren (1949), LXI–11 and picked up by Coffey, op. cit. (n. 1), 246 n. 63 and Winkler, op. cit. (n. 1), 148 with the sequence of scenes on Trajan's Column is helpful, not only in drawing attention to the paratactic sequence but also in suggesting an underlying principle in the ordering of those scenes. In what follows, I adapt the theory suggested by W. S. Smith Jr., ‘Husband vs. wife in Juvenal's sixth Satire’, CW 73 (1980), 323–32.

¹⁰⁷ LI. 133–5, a praeteritio, do not fit here and seem to belong between 626 and 627, where they pick up the mention of poison used to befuddle the husband and transfer the topic to step-children. See Highet, op. cit. (n. 1), 267: ‘133–5 are obviously misplaced and must follow 626’. Read nimia for minimo, with Martyn, op. cit. (n.50).

¹⁰⁸ cf. on the structure of Satire 5, M. Morford, ‘Juvenal's Fifth Satire’, AJP 98 (1977), 219–45, esp. 233–7 and 245, in which the two menus follow the sequence of dishes at a cena; Smith, op. cit. (n. 106), 323–4 also draws this broad analogy, although the detailed comparison of common elements is not convincing.

¹⁰⁹ Other examples of such characters include Damasippus in Horace, Satires 11.3, whose sudden fervent missionary zeal for Stoicism sits ill with his previous life-style; Catius in Satires 11.4 who inappropriately elevates gourmandise to the level of philosophy; and Naevolus of Juvenal's ninth Satire who complains angrily about his ex-patron but appears to have earned the bad treatment which he has received.

¹¹⁰ On the satire of out-groups, see Richlin, op. cit. (n. 14), 67.