1 See Bartsch, H.-W., Die Anfänge urchristlicher Rechtsbildungen: Studien zu den Pastoralbriefen (ThF 34; Hamburg: Evangelischer, 1965) 112–43Google Scholar, who drew anachronistically on much later evidence for an order of widows. The presumption that the text regulates a pre-existing Stand or even Amt of widows has been continued in much scholarship: see e.g. Thurston, B. B., The Widows: A Women's Ministry in the Early Church (Minneapolis: Fortress, 1989) 36–55Google Scholar; Wagener, U., Die Ordnung des ‘Hauses Gottes’: Der Ort von Frauen in der Ekklesiologie und Ethik der Pastoralbriefe (WUNT ii/65; Tübingen: Mohr Siebeck, 1994)Google Scholar. For an early protest (arguing for, at most, a Gruppe or Ordnung), see Sand, A., ‘Witwenstand und Ämsterstrukturen in den urchristlichen Gemeinden’, Bibel und Leben 12 (1971) 186–97Google Scholar. Bassler, J. M., ‘The Widows’ Tale: A Fresh Look at 1 Tim 5:3–16’, JBL 103 (1984) 23–41Google Scholar refers, more cautiously, to a ‘circle’ of widows; cf. her more recent treatment of the text, ‘Limits and Differentiation: The Calculus of Widows in 1 Timothy 5.3–16’, A Feminist Companion to the Deutero-Pauline Letters (ed. A.-J. Levine; London/New York: T&T Clark, 2003) 122–46.

2 I take the three Pastoral letters to be closely related interpretations of the Pauline tradition dating from the late first/early second century. Roloff, J., Der erste Brief an Timotheus (EKKNT 15; Zürich/Neukirchen-Vluyn: Benziger/Neukirchener, 1988) 46Google Scholar dates our text ‘kaum sehr viel später als um das Jahr 100’. For full discussion of the dating (with a similar result), in light of the probable reception of the Pastorals in the letters of Ignatius and Polycarp (ca. 110–20 ce), see Merz, A., Die fiktive Selbstauslegung des Paulus (NTOA 52; Göttingen: Vandenhoeck & Ruprecht; Fribourg: Academic, 2004) 72–194CrossRefGoogle Scholar. I shall refer to our author as ‘the Pastor’.

3 For the inclusio of 5.3–4a and 5.16 (with key elements in inverse order), see Bassler, ‘Limits’, 134. Wagener's treatment of 5.16 as an unrelated addition to the preceding instructions is unconvincing on literary grounds (Ordnung, 124, 223–7).

4 For the literary structure of the text, founded on the four key imperatives of 5.3, 5.9, 5.11 (with its mirror instruction in 5.14) and 5.16, see Roloff, Der erste Brief, 284–5.

5 For the fifth commandment as envisaging material support for parents, see Philo, Decal. 106–20. The elderly, isolated widows to be supported in this context qualify as ‘virtual’ mothers (cf. 5.2).

6 In the phrase ἡ δὲ ὄντως χήρα καὶ μεμονωμένη (5.5), the καί is probably epexegetical; so Roloff, Der erste Brief, 289 n. 330; I. H. Marshall, The Pastoral Epistles (ICC; London: T&T Clark, 1999) 586–7.

7 See G. Stählin, ‘χήρα’, TDNT ix.440–65, at 440–1, with reference to the definition of Hesychius as either ἡ τὸν ἄνδρα στερηθεῖσα γυνή or ἡ μετὰ γάμον μὴ συνοικοῦσα ἀνδρί (the latter by either death or divorce).

8 In pagan Greek usage, see e.g. Aeschylus fr. 47a (Radt edition), of the ‘manless’, unmarried Danae; Euripides, Andr. 347–8 (coupled with ἄνανδρος, of a woman rejected by her husband). Cf. the verb χηρεύω in Demosthenes, Orat. 30.33 (of a woman who lives without her husband, though it is disputed if she is divorced); it is even used of a man (all alone) in Sophocles, Oed. Tyr. 479. In Josephus and Philo we find the noun used of a widow (e.g. Josephus, Ant. 16.221; Philo, Leg. 2.63) but also extended to a woman who is in other ways separated from a man (Philo, Spec. 3.64). Thus it is equivalent to ἐρήμη (‘bereft’, whether of a man or of other phenomena, Philo, Deus 136, 138), μονωθεῖσα (Philo, Spec. 1.105) and στερομένη (Philo, Virt. 114). Interestingly, χηρεία and χηρεύω can even be used of a woman before marriage (Josephus, Ant. 7.172), while a woman left alone after her marriage can be considered equivalent to a παρθένος in her condition ‘without a man’ (Philo, Spec. 1.129; QE 2.3b; cf. the parallelism in Spec. 3.26). See also on this topic Wagener, Ordnung, 127 n. 59.

9 The ἔκγονα here, juxtaposed with τέκνα, are probably ‘grandchildren’; women often bore children in their late teens, so they might well have grandchildren, via their daughters, by the age of forty. The reading of 5.4 followed here is adopted by most modern interpreters; see e.g. J. N. D. Kelly, A Commentary on the Pastoral Epistles (London: A & C Black, 1963) 113; M. Dibelius and H. Conzelmann, The Pastoral Epistles (Hermeneia; Philadelphia: Fortress, 1972) 74; N. Brox, Die Pastoralbriefe (Regensburg: Pustet, 1989) 188–9; Marshall, Pastoral Epistles, 583–5.

10 An alternative reading of this verse has gained some currency both in ancient and in modern times (e.g. Roloff, Der erste Brief, 287–8; Wagener, Ordnung, 149–54; J. M. Bassler, I Timothy, 2 Timothy, Titus (Abingdon NT Commentaries; Nashville, TN: Abingdon, 1996) 94–6; ancient commentators include John Chrysostom, PG 62.566–7): taking the χῆραι to be the subject of μανθανέτωσαν (cf. the shift in subject in 1 Tim 2.15), the Pastor, on this reading, here instructs them first to look after their own offspring. There are two main objections to this reading, which seem fatal: (i) it makes no sense for χῆραι to give return benefits (ἀμοιβαί) to their ancestors by looking after their own children. To give a return to your benefactor, you benefit them in exchange, but it is hardly a benefit to your ancestors to care for your own descendants; (ii) it seems nonsensical to tell a needy χήρα that she cannot be supported because she has children to look after: dependent children would make her more needy, not less so. This alternative reading only works on the assumption that (self-sufficient) χῆραι are being enlisted into a church office with a set of time-consuming duties, and the Pastor is instructing them not to neglect their families in their devotion to the church. But since, as we shall see, this passage prescribes no duties for χῆραι, this assumption is unfounded. An objection to the reading that I adopt is that προγόνοι can mean only dead ancestors and so cannot apply to the case of children being urged to look after their living parents/grandparents; but this is simply incorrect (see, e.g. Plato, Leg. 931e). The instructions in 5.8 and 5.16 confirm that the concern of this verse is how χῆραι are to be supported, not whom they are required to support.

11 The closing statement of 5.4 (τοῦτο γάρ ἐστιν ἀπόδεκτον ἐνώπιον τοῦ θεοῦ), like the matching clause in 1 Tim 2.3, makes clear that the moral standard invoked here is not just human, or natural, but divine.

12 At the same time, some of these factors might make them less attractive for remarriage (and with less power to bring about this outcome). Egyptian census records suggest that women were less likely to be married the older they got (only 48 percent by the age of fifty), while this is not true of men; see R. S. Bagnall and B. W. Frier, The Demography of Roman Egypt (Cambridge: Cambridge University Press, 1994) 121–6.

13 For life expectancy and mortality, see W. Scheidel, ed., Debating Roman Demography (Leiden: Brill, 2001); for age at marriage, R. P. Saller, ‘Men's Age at Marriage and its Consequences in the Roman Family’, CP 82 (1987) 20–35.

14 By contrast, rich widows are given literary representation in 1 Tim 5.6 and Jdt 8.7.

15 Whatever historical reality may or may not lie behind this text, the question of which network takes responsibility for widows is as urgent there as in 1 Tim 5.

16 J.-U. Krause, Witwen und Waisen im römischen Reich, vol. i:Verwitwung und Wiederverheiratung (HABES 16; Stuttgart: F. Steiner, 1994) 47–73, at 73. The three further volumes in this study (HABES 17–19) are packed with information, the fourth being devoted to Witwen und Waisen im frühen Christentum (1995). For a detailed review, suggesting Krause's figures are accurate chiefly among the sub-elite, see T. A. J. McGinn, ‘Widows, Orphans, and Social History’, JRA 12 (1999) 617–32.

17 See e.g. LXX Pss 4.6; 7.2; 15.1; cf. 2 Cor 1.10; 1 Tim 4.10; 6.17; Wagener, Ordnung, 132–4.

18 Herm. Sim. 2.6 (51.6): ‘the petition of the poor person is acceptable and rich before the Lord’. Clement of Alexandria indicates the same in Quis div. 34–5. For the confidence that God will listen to the prayers of the poor/widows, see Exod 22.22–3; Job 34.28; Sir 35.14; Jdt 9.4, 9. To continue in prayer night and day indicates persistence (born of necessity), not a church duty; cf. the ‘night and day’ prayers of widows in Luke 2.37 and 18.7. These prayers may also be for others (Polycarp expects widows to ‘pray without ceasing for everyone’, Phil. 4.3) and it is possible that this association of widows with intercessory prayer led later to the development of a special ecclesial role.

19 Cf. the discussion in Wagener, Ordnung, 155–61. The notion of a ‘living death’ is paralleled in Hermas (Sim. 6.5.4 (65.4); of a life of luxury), and is found in moral critique also in Philo, Fug. 55 and Publilius Syrus 47 (of a miser).

20 It is not wholly clear who is meant to be ἀνεπίλημπτοι in 5.7, but since no instructions so far have been issued to widows, this is probably directed at their children/grandchildren who were given a command in 5.4; 5.7 thus leads immediately into 5.8. Some of those who read 5.4 as directed at widows (see above, n. 10) regard 5.8 as also directed at widows with family responsibilities (e.g. Wagener, Ordnung, 149–54). But, again, this presupposes that widows need this instruction because they are inclined to prioritise their church responsibilities over their duties to their families, whereas our text neither presupposes nor creates responsibilities for widows that could compete with their family roles. The Pastor is particularly sensitive to criticism of apparently ‘disordered’ Christian households (1 Tim 5.14; 6.1; Titus 2.5, 8, 10), exemplified here by widows unsupported by their offspring.

21 For the assumption that the children of believers will be believers themselves, cf. 1 Tim 3.4–5, 12; Titus 1.6–7. The Pastor does not consider those whose offspring or wider family were pagan, though his comments here presuppose that even they could be expected to look after a widowed relative.

22 Cf. the quip of Antisthenes in response to the begging priests of Cybele (the mother of the gods): ‘I don't feed the mother of the gods; that is the gods’ job!’ (Clement, Protr. 7). B. Winter (Seek the Welfare of the City (Grand Rapids: Eerdmans, 1994) 62–78) notes the legal responsibilities of those who had control of a widow's dowry, but that duty does not surface in this text, and it is unlikely that there would be any remaining dowry for widows from poorer families.

23 For the attribution of familial labels to all believers, see 1 Tim 5.1–2; for the tensions this could cause within a household between slaves and their owners, see 1 Tim 6.1–2. The financial implications of familial designations in early Christianity are well discussed by T. J. Murray, Restricted Generosity in the New Testament (WUNT ii/480; Tübingen: Mohr Siebeck, 2018) 139–57.

24 Cf. Mark 3.31–5; Matt 8.20–2; 19.12. See G. Theissen, Social Reality and the Early Christians (Edinburgh: T. & T. Clark, 1993) 33–93.

25 Thecla is the most obvious, but by no means the only, example. See S. Davies, The Revolt of the Widows: The Social World of the Apocryphal Acts (Carbondale: Southern Illinois University, 1980); D. R. MacDonald, The Legend and the Apostle (Philadelphia: Westminster, 1983); V. Burrus, Chastity as Autonomy: Women in the Stories of Apocryphal Acts (Lewiston/Queenston: Edwin Mellen, 1987); P. Brown, The Body and Society: Men, Women and Sexual Renunciation in Early Christianity (London/Boston: Faber and Faber, 1988) 140–59. Lucian regards as absurd the Christians’ support of Peregrinus, which he relates to the Christian ideology of an alternate family (‘they are all brothers of one another’); the resultant pooling of possessions means that ‘any charlatan or trickster’ can profit from their gullibility and their reconfiguration of economic obligations (Peregr. 12–13).

26 For the church as macro-household, see D. C. Verner, The Household of God and the Social World of the Pastoral Epistles (Chico, CA: SBL, 1983). Horrell emphasises the Pastorals’ use of the household as a model of unequal relations (with corresponding diminution of sibling-language) in D. G. Horrell, ‘From ἀδελφοί to οἶκος θεοῦ: Social Transformation in Pauline Christianity’, reprinted in idem, The Making of Christian Morality: Reading Paul in Ancient and Modern Contexts (Grand Rapids: Eerdmans, 2019) 75–96; cf. his ‘Disciplining Performance and “Placing” the Church: Widows, Elders and Slaves in the Household of God (1 Tim 5,1–6,2)’, 1 Timothy Reconsidered (ed. K. Donfried; Leuven: Peeters, 2008) 109–34.

27 Contrast the ‘laying on of hands’ (1 Tim 4.14; 5.22; 2 Tim 1.6) and the use of καθίστημι in Titus 1.5.

28 For soldiers, see e.g. Josephus, J.W. 2.268, 576, 584; Philo, Abr. 232; Virt. 42.

29 Josephus, Ant. 2.180; 11.68; 18.142; 20.87; Ag. Ap. 1.131; Philo, Sacr. 122; Post. 110; Decal. 29; Aet. 114; Legat. 323.

30 For analysis, see C. Methuen, ‘Widows, Bishops and the Struggle for Authority in the Didascalia Apostolorum’, JEH 46 (1995) 197–213.

31 Despite arguments to the contrary (e.g. Thurston, Widows), I find no evidence of such an ‘order’ anywhere else in the NT, even where widows are singled out for special mention (e.g. in Acts 6.1–2; 9.36–43). For recent discussion (and cautious conclusions), see A. Standhartinger, ‘Witwen im Neuen Testament’, Frauen gestalten Diakonie, vol. i:Von der biblischen Zeit bis zum Pietismus (ed. A. M. von Hauff; Stuttgart: Kohlhammer, 2007) 141–54. The same holds for references in Ignatius (Smyrn. 6.2; 13.1; Poly. 4.1) and Polycarp (Phil. 4.3; 6.1), pace e.g. C. Back, Die Witwen in der frühen Kirche (Frankfurt am Main: Peter Lang, 2015) 242–50.

32 Tertullian, Ux. 1.7; cf. Praescr. 3; Pud. 13.

33 Pace, among others, L. Oberlinner, Die Pastoralbriefe, vol. i:Kommentar zum ersten Timotheusbrief (Freiburg/Basel/Vienna: Herder, 1994) 233–4, 245.

34 ἑνὸς ἀνδρὸς γυνή (5.9; cf. the equivalent for male leaders in 1 Tim 3.2, 5) could mean ‘not divorced/separated’ (or ‘not unfaithful’ to her husband), or it could mean ‘married only once’ (equivalent to the Latin univira), although that is more often expressed as μόνανδρος. See discussion in Roloff, Der erste Brief, 293–4 and S. Page, ‘Marital Expectations of Church Leaders in the Pastoral Epistles’, JSNT 50 (1993) 105–20. The latter option would clash with the instruction of 5.14 only if that concerns remarriage (see below), and even then, only if 5.9 is read as a legal ruling (rather than as an idealised image of a worthy widow).

35 For ἔργα καλά, see 1 Tim 5.25; 6.18; Titus 2.7, 14; 3.8, 14 (expressly in relation to urgent need); for ἔργα ἀγαθά and πᾶν ἔργον ἀγαθόν, see 1 Tim 2.10; 2 Tim 2.21; 3.17; Titus 1.16; 3.1; cf. ἀγαθοεργεῖν in 1 Tim 6.18 (with clear financial implications).

36 For the social realities, see R. S. Schellenberg, ‘Subsistence, Swapping, and Paul's Rhetoric of Generosity’, JBL 137 (2018) 215–34.

37 On footwashing as a distinctive early Christian practice (cf. John 13.1–16; Tertullian, Ux. 2.4), see J. C. Thomas, Footwashing in John 13 and the Johannine Community (Sheffield: JSOT, 1991) and R. Bauckham, The Testimony of the Beloved Disciple (Grand Rapids: Baker Academic, 2007) 191–206.

38 ‘Good works’ are associated especially with women both here and in 1 Tim 2.10; cf. Titus 2.5: οίκουργοὶ ἀγαθαί.

39 T. G. Parker, Old Age in the Roman World: A Cultural and Social History (Baltimore, MD: Johns Hopkins University Press, 2003) 280 estimates that at any one time only 7 per cent of the population would be over sixty and only 4 per cent over sixty-five.

40 Difficulty in supporting herself economically (because of the limitations of age) might be another factor here, but the insistence in 5.11–15 that ‘younger’ women should marry suggests that marriageability is the key issue.

41 S. Treggiari, Roman Marriage: Iusti coniuges from the Time of Cicero to the Time of Ulpian (Oxford: Clarendon, 1991) 60–80. Given the resistance to these laws, the age limit of fifty was perhaps as high as the legislators dared go in the hope of some level of compliance.

42 Since the benefits given by these women are said to extend through the church network, it is reasonable for the church, as such, to give them an appropriate return (cf. 5.4, ἀμοιβὰς ἀποδιδόναι). For the contrast with the ἀργαί (5.13), see below.

43 LSJ cites only our text for the compound verb. The use in Ps.-Ignatius, Antioch. 11 is clearly dependent on this passage. For στρηνιάω, see Rev 18.7, 9 and Isa 61.6 (Symmachus); στρῆνος is used in e.g. Rev 18.3; LXX 4 Kgdms 19.28.

44 Translations are various. NRSV (following BAGD) renders: ‘when their sensual desires alienate them from Christ’. But disloyalty may be the stronger connotation, rather than sexual infidelity, in relation to Christ. Wagener, Ordnung, 201 suggests ‘eine Kombination der Bedeutungsgehalts “Überschreitung von Ordnungsgrenzen” mit der sexuellen Konnotation’.

45 For representative examples, see Kelly, Pastoral Epistles, 117; C. Spicq, Les Épitres pastorales (Paris: Gabalda, 1969) 535–6; Roloff, Der erste Brief, 296–7; Marshall, Pastoral Epistles, 599–601 (though he takes the πίστις as faith, abandoned in marrying an unbeliever).

46 See, especially, C. Methuen, ‘The “Virgin Widow”: A Problematic Social Role for the Early Church?’, HTR 90 (1997) 285–98. For readings along such lines, though importing (unnecessarily) the notion of entry into a ‘circle’, ‘order’, or ‘office’, cf. Wagener, Ordnung, 200–11; Bassler, 1 Timothy, 94–8; Back, Witwen, 187; M. Tsuji, ‘Zwischen Ideal und Realität: Zu den Witwen in 1 Tim 5.3–16’, NTS 47 (2001) 92–104.

47 J. B. Lightfoot struggles unsuccessfully to read Ignatius’ phrase in the sense ‘widows who may be called virgins, on the grounds of their purity and devotion’, The Apostolic Fathers, Part ii: S. Ignatius, S. Polycarp, vol. ii (London/New York: Macmillan, 1889) 322–4. For other examples where Christian unmarried women are given this title, or closely associated with widows, see Acts Pet. 29 (‘virgins of Christ’); Tertullian, Virg. 9 (horrified at the practice of allowing virgins to sit with widows). Since a παρθένος was usually a teenage girl, it was natural to apply another label to older ascetic women that represented their status (without a man) and that encouraged the church to view them charitably.

48 The notion of virgins as ‘brides of Christ’ is attested only later (e.g. Tertullian, Or. 21–2; Virg. 16) but is a natural extrapolation from the Pauline texts cited above, and may be reflected in 1 Tim 5.12, if the language there has sexual connotations.

49 The Pastorals repeatedly warn against abandoning or undermining faith (ἡ πίστις); see 1 Tim 1.19; 4.1; 5.8; 6.10, 21; 2 Tim 2.18.

50 At a later date, Tertullian hails women who ‘set the seal on their virginity in their flesh at baptism’ (Ux. 1.6: statim a lavacro carnem suam obsignant; cf. Bapt. 18).

51 Pauline authority for this preference is explicitly evoked in Acts of Paul and Thecla 5–6.

52 Conversely, Roman historians have recently emphasised that even in marriage (indeed, even in a manus-marriage) and even under the (notional) oversight of a legal tutor, women of means enjoyed considerable economic independence; see S. Dixon, ‘Exemplary Housewife or Luxurious Slut: Cultural Representations of Women in the Roman Economy’, Women's Influence on Classical Civilization (ed. F. McHardy and E. Marshall; London/New York: Routledge, 2004) 56–74.

53 For scholarly discussion of the ascetic option in early Christianity, see n. 25. It is possible to read 1 Tim 2.15 as referring to women being brought safely through the ordeal of childbirth (τεκνογονία); I am indebted to Emily Gathergood for this suggestion.

54 Nothing in the context here suggests a fear lest they marry unbelievers, pace G. Fee, 1 and 2 Timothy, Titus (NIBC; Peabody, MA: Hendrickson, 1988) 121–2; Marshall, Pastoral Epistles, 598–601.

55 Wagener, Ordnung, 205–6. Cf. 2 Tim 2.14 (οὐδὲν χρήσιμον); Titus 1.16 (πρὸς πᾶν ἔργον ἀγαθὸν ἀδόκιμοι) and, for the alternative, Titus 3.14: μανθανέτωσαν δὲ καὶ οἱ ἡμέτεροι καλῶν ἔργων προΐστασθαι εἰς τὰς ἀναγκαίας χρείας, ἵνα μὴ ὦσιν ἄκαρποι.

56 The phrase is often interpreted as an indication of the widows’ pastoral responsibilities (‘pastoral house calls’, Dibelius and Conzelmann, Pastoral Epistles, 75). But there is no indication in this text that widows had such duties (rightly, Brox, Pastoralbriefe, 195). Going from place to place is precisely the itinerant lifestyle modelled by Jesus and his disciples (Matt 8.18–22; 10.5–14), and it is characteristic of Thecla and others in the apocryphal Acts. The Synoptic prohibition against moving from house to house in any one town (μὴ μεταβαίνετε ἐξ οἰκίας εἰς οἰκίαν, Luke 10.7; cf. Mark 6.10) is mitigated by the instruction in Did. 11.3–6 about limiting the time one should accommodate ‘apostles’ and ‘prophets’. For the use of the verb περιέρχομαι in the sense ‘live an itinerant life’, see Acts 19.13; Heb 11.37. Contrast the emphasis on the ἴδιος οἶκος in 1 Tim 3.4, 5, 12; 5.4.

57 See M. B. Kartzow, Gossip and Gender: Othering of Speech in the Pastoral Epistles (Berlin/New York: de Gruyter, 2009) 50–66, 142–59 (pointing out, however, uncertainties regarding translation and nuance). For parallel critiques of ‘useless’ speech, cf. 1 Tim 1.6; 6.20; 2 Tim 2.16; Titus 1.10; 3.9.

58 It is striking that the Pastor takes a motif used in that tradition (‘the workman is worthy of his hire’, Matt 10.10; Luke 10.7) and applies it not (as in the gospels) to itinerants, but to resident ‘elders’ (1 Tim 5.18).

59 The Pastor interprets the adoption of this alternative Christian model as a form of apostasy: cf. the use of ἐκτρέπω in 1 Tim 1.6; 6.20; 2 Tim 4.4. For Satan, cf. 1 Tim 1.20; and for ‘the snare of the διάβολος’, 1 Tim 3.6–7; 2 Tim 2.26.

60 Although γαμεῖν (5.14) can mean ‘remarry’ (cf. 1 Cor 7.39), the instructions of this verse give the impression that these are all first-time activities (not, e.g., bearing more children). Would this instruction apply also to young women who had been widowed? By the Pastor's logic, probably so (on the possible clash with 5.9, see above, n. 34), but they do not seem to be those primarily in view.

61 Both Ignatius and Polycarp indicate that, in Smyrna at least (and perhaps elsewhere), there arose a symbiosis of both models: the household (married Christians bringing up children) and celibate women (virgins called ‘widows’). The two are mentioned side by side not only in Ignatius, Smyrn. 13.1, but also in Ignatius, Pol. 5.1–2 (those liable to boast in their ‘purity’ are probably celibate women) and Polycarp, Phil. 4.2–3, where ‘widows’ are easy targets of slander (as in 1 Tim 5.14) as recipients of financial aid. There is an evident anxiety about supporting (virgin) ‘widows’ (cf. Ignatius, Pol. 4.1), but not, as in the Pastorals, an outright ban.

62 The alternative reading (εἴ τις πιστὸς ἢ πιστή) is less well attested (among uncials, only D Ψ K L) and assimilates the text to the form of later church orders (see Wagener, Ordnung, 118 n. 24).

63 A male fulfilling this role might invite scandal, to which the Pastor is very sensitive (5.14). The support of widows by Tabitha/Dorcas might be exemplary in this regard (Acts 9.36–9); at a much higher social level, note the patronage of Thecla by Tryphaena in Acts of Paul and Thecla. From the census records in Egypt, E. E. Hanson records a surprisingly high number of divorced or widowed women (39 out of 103) living in predominantly female households, ‘Widows Too Young in their Widowhood’, I Claudia II: Women in Roman Art and Society (ed. D. E. E. Kleiner and S. B. Matheson; Austin, TX: University of Texas Press, 2000) 149–65, at 152.

64 Cf. the church κοινόν mentioned in Ignatius, Pol. 4.3, and similarly protected; later, Justin, 1 Apol. 67.6–7. For similar limitations in generosity, see 2 Thess 3.10 and Murray, Restricted Generosity.

65 See Verner, Household, 161–6, with reference to earlier proponents of this view; Kidd, R. M., Wealth and Beneficence in the Pastoral Epistles (Atlanta: Scholars, 1990) 103–6Google Scholar.

66 Cf. the eighty-year-old widow, Regina, whose daughter made a point of insisting that she ‘never burdened the church’ (4th/5th cent., Rome, ILCV 1i.1581, cited and discussed by Elm, U. E., Women Officeholders in Early Christianity: Epigraphical and Literary Studies (Collegeville, MN: Liturgical Press, 2000) 145–6Google Scholar.

67 For this reason, it is a mistake to take this passage as a rule-book, and to worry, pedantically, over how the Pastor would treat aged widows who were childless or had been widowed twice: the portrait of 5.9–10 is idealised and hardly drafted with legal precision.

68 For the development of female asceticism in the church, see Brown, Body and Society; Elm, S., ‘Virgins of God’: The Making of Asceticism in Late Antiquity (Oxford: Clarendon, 1994)Google Scholar. I am grateful to all those who responded to versions of this paper at Durham, Münster, Vancouver, Exeter and Cambridge, and especially to those who kindly read this essay closely and offered detailed advice, including David Horrell, James Diggle, Paul Trebilco, Harry Maier, Rebecca Flemming, Tim Murray and Simon Gathercole.