1 Richards, E. R., The Secretary in the Letters of Paul (Tübingen: Mohr Siebeck, 1991) 199–200Google Scholar. For earlier studies, see Roller, O., Das Formular der paulinischen Briefe: Ein Beitrag zur Lehre vom antiken Briefe (Stuttgart: Kohlhammer, 1933)Google Scholar; Longenecker, R. N., ‘Ancient Amanuenses and the Pauline Epistles’, New Dimensions in New Testament Study (ed. Longenecker, R. N. and Tenney, M. C.; Grand Rapids: Zondervan, 1974) 281–97Google Scholar.

2 Richards, The Secretary, 31, 45, 61–76, 123. Compare Cicero, Att. 6.9, 8.1, 8.13.1, 12.32.1, 13.28.

3 See Witherington, B., The Paul Quest: The Search for the Jew of Tarsus (Downers Grove, IL: InterVarsity, 1998) 99–109Google Scholar.

4 Richards, The Secretary, 77, cf. 24–52, 97–111; idem, Paul and First-century Letter Writing: Secretaries, Composition and Collection (Downers Grove, IL: InterVarsity, 2004) 80Google Scholar.

5 Richards, The Secretary, 3 et passim. Pliny's letters are sidelined as ‘rather artificial’, as if Cicero's published collections do not also offend in this way.

6 Richards, First-century, 103; Capes, D. B., Reeves, R. and Richards, E. R., Rediscovering Paul: An Introduction to his World, Letters and Theology (Downers Grove, IL: InterVarsity, 2010) 68Google Scholar.

7 Such as Phil 3.2; 1 Cor 14.33; see Richards, First-century, 108–9. On multiple letters, fragments and interpolations, see Welborn, L. L., ‘The Corinthian Correspondence’, ‘All Things to All Cultures’: Paul among Jews, Greeks and Romans (ed. Harding, M. and Nobbs, A.; Grand Rapids: Eerdmans, 2015) 205–42Google Scholar.

8 Richards, First-century, 155, 232 n. 1 (the ‘collapsing camel’ of deutero-Pauline theories!). See also Capes, Reeves and Richards, Rediscovering Paul, 18–19, 72.

9 Richards, The Secretary, 168–9 (cf. 107, 121 n. 40) is particularly dismissive of the arguments of Walker, W. O., ‘The Burden of Proof in Identifying Interpolations in the Pauline Letters’, NTS 33 (1987) 610–18CrossRefGoogle Scholar.

10 Richards, First-century, 35, 93, 152, 165. In his earlier work, he had allowed the questioning of the authenticity of the Pastorals (The Secretary, 182–93). Hebrews, though included in the Pauline corpus, remains barred.

11 Towner, P., The Letters to Timothy and Titus (NICNT; Grand Rapids: Eerdmans, 2006) 86–7Google Scholar.

12 Klauck, H.-J., Ancient Letters and the New Testament (Waco, TX: Baylor University Press, 2006) 58–60Google Scholar; Moon, J., Mark as Contributive Amanuensis of 1 Peter? (Berlin: LIT, 2009)Google Scholar; Ciampa, R. E. and Rosner, B., The First Letter to the Corinthians (Grand Rapids: Eerdmans, 2010)Google Scholar; McKnight, S., The Letter of James (NICNT; Grand Rapids: Eerdmans, 2011)Google Scholar. A more measured appreciation of Richards’ research is found in Weima, J. A. D., Paul the Ancient Letter Writer: An Introduction to Epistolary Analysis (Grand Rapids: Baker, 2016)Google Scholar.

13 See Ehrman, B., Forgery and Counter-forgery: The Use of Literary Deceit in Early Christian Literature (Oxford: Oxford University Press, 2013) 218–22Google Scholar. The arguments in this article address quite different issues from those raised by Ehrman.

14 Elmer, I., ‘I, Tertius: Secretary or Co-author of Romans’, ABR 56 (2008) 45–60Google Scholar; idem, ‘Setting the Record Straight at Galatia’, Religious Conflict from Early Christianity to the Rise of Islam (ed. Mayer, W. and Neil, B.; Berlin/Boston: de Gruyter, 2013) 21–37CrossRefGoogle Scholar, at 25–8; idem, ‘The Pauline Letters as Community Documents’, Collecting Early Christian Letters: From the Apostle Paul to Late Antiquity (ed. Neil, B. and Allen, P.; Cambridge: Cambridge University Press, 2015), 37–53CrossRefGoogle Scholar.

15 Richards, Rediscovering Paul, 73; Hull, R. F., The Story of the New Testament Text: Movers, Materials, Motives, Methods and Models (Atlanta: SBL, 2010) 10Google Scholar.

16 Elmer, ‘I, Tertius’, 54; see however C. E. B. Cranfield, The Epistle to the Romans (2 vols.; ICC; Edinburgh: T. & T. Clark, 1979) ii.806 for the difficulties of the prepositional phrase; also Bahr, G. J., ‘Paul and Letter Writing in the Fifth Century’, CBQ 28.4 (1966) 465–77Google Scholar, at 465–6.

17 Hull, Story of the New Testament Text, 10 suggests Gaius; cf. Murphy-O'Connor, J., Paul: A Critical Life (Oxford: Clarendon, 1996) 268Google Scholar. Elmer equivocates: ‘I, Tertius’, 54. Gerd Theissen places Tertius in Gaius’ house but stalls over whether he was a ‘Schreibsklave’ or an employee in the provincial government (Theissen, G., Studien zur Soziologie des Urchristentums (WUNT 19; Tübingen: Mohr Siebeck, 1979) 227, 253–4Google Scholar). Possible also is that Tertius may have been a slave-stenographer within Phoebe's household at Cenchreae, given her ‘benefaction’ (16.2). Nevertheless, this should not be inflated. Steven Friesen places both her and Gaius only at Level 4–5 and 4 respectively in his poverty scale, that is, either just above subsistence level or enjoying a moderate surplus. He finds no clear evidence that ‘any of the members of Paul's assemblies were rich’ (Friesen, S., ‘Prospects for a Demography of the Pauline Mission: Corinth among the Churches’, Urban Religion in Roman Corinth. Interdisciplinary Approaches (ed. Schowalter, D. N. and Friesen, S. J.; HTS 53; Cambridge, MA: Harvard University Press, 2005) 351–70Google Scholar, at 364–8). All such suggestions on the provision of Gaius are posited on a reading of ξένος in Rom 16.23 as ‘host’. The evidence however is clearly in favour of a translation as ‘guest’. See Last, R., The Pauline Church and the Corinthian Ekklēsia: Greco-Roman Associations in Comparative Context (SNTSMS 164; Cambridge/New York: Cambridge University Press, 2015) 62–71Google Scholar. Further, this resolves any difficulties in understanding the phrase ὁ ξένος καὶ ὅλης τῆς ἐκκλησίας – both Paul and the Corinthian church have welcomed Gaius as guest, implying that he was a visitor from Rome, not a resident of Corinth. See Kloppenborg, J. S., ‘Gaius the Roman Guest’, NTS 63 (2017) 534–49CrossRefGoogle Scholar.

18 Richards, First-century, 108, 151.

19 Richards, First-century, 109; idem, Rediscovering Paul, 74; Elmer, ‘I Tertius’, 58.

20 Compare Bahr, G. J., ‘The Subscriptions in the Pauline Letters’, JBL 87 (1968) 27–41Google Scholar.

21 Richards, First-century, 152.

22 Elmer, ‘I, Tertius’, 51.

23 Richards, The Secretary 170–2; idem, First-century, 77; Elmer, ‘I, Tertius’, 51, 54; idem, ‘Setting the Record Straight’, 26.

24 Richards, The Secretary, 171; idem, First-century, 77.

25 Elmer, ‘I, Tertius’, 56.

26 Elmer, ‘I, Tertius’, 54, citing Byrne, B., Romans (Sacra Pagina 6; Collegeville, MN: Liturgical Press, 1996) 459Google Scholar; Dunn, J. D. G., Romans 9–16 (WBC 38B; Dallas, TX: Word, 1988) 909–10Google Scholar; Witherington, B., Paul's Letter to the Romans: A Socio-Rhetorical Commentary (Grand Rapids: Eerdmans, 2004) 399Google Scholar.

27 Thiselton, A., The First Epistle to the Corinthians (NIGTC; Grand Rapids: Eerdmans, 2000) 164Google Scholar. Thiselton's translation is ‘professional’. Contra Richard, The Secretary, 3. I follow Winter in arguing that 1 Cor 1.20 is not referring to an expert in religious law apparent in the gospels: Winter, F., ‘Exkurs: Schreiber, Sekretäre, Schriftgelehrte’, 1. Korinther (ed. Arzt-Grabner, P., Kritzer, R. Elisabeth, Papathomas, A. and Winter, F.; PKNT 2; Göttingen: Vandenhoeck & Ruprecht, 2006) 88–9Google Scholar.

28 Bockmuehl, M., The Epistle to the Philippians (BMTC; London: A & C Black, 1998) 86Google Scholar. Bockmuehl considers that a co-sender does not mean a co-author. Similarly, Cranfield, Romans, i.4; Dunn, Romans 9–16, 910; Byrne, Romans, 460; Collins, R. F., Letters That Paul Did Not Write: The Epistle to the Hebrews and the Pauline Pseudepigrapha (Wilmington, DE: Michael Glazier, 1988) 74–5Google Scholar. Conversely, Elmer now holds that it no longer needs argumentation (‘Collecting Early Christian Letters’, 47).

29 This alone makes it improbable that Tertius was responsible for putting chapter 16 together (contra Richards, First-century, 152). The unusual descriptors added to many of those greeted hardly befits the secretary's invention – note especially the first-person possessives in vv. 3, 4, 5, 7, 8, 9, 11, 13, 21. See, generally, Mathew, S., Women in the Greetings of Romans 16.1–16: A Study of Mutuality and Women's Ministry in the Letter to the Romans (London: Bloomsbury, 2013) 21–45Google Scholar.

30 Richards, The Secretary, 55; Elmer, ‘I, Tertius’, 52–3. The assertion goes back to Ambrosiaster: Vogels, H. J., ed., Ambrosiastri qui dicitur Commentarius in epistulas Paulinas (CSEL 81; Vienna: Hoelder-Pichler-Tempsky, 1966) 491Google Scholar.

31 Elmer, ‘I, Tertius’, 52; Murphy-O'Connor, Paul: A Critical Life, 272.

32 The ‘rule of my writings’ in Cicero, Fam. 16.17.1 is taken as a literal recognition of Tiro's abilities, rather than the rhetorical hyperbole from which Cicero was rarely immune (cf. Att. 7.5); see Richards, The Secretary, 47–9; Elmer, ‘I, Tertius’, 51–2.

33 See White, P., Cicero in Letters: Epistolary Relations of the Late Republic (Oxford: Oxford University Press, 2010) 31–61CrossRefGoogle Scholar. Note however that he credits Tiro with a limited impact on the result.

34 This is a neglected factor in the study of ancient letters, though it enters the descriptions of ‘hands’ that scholars observe in papyri and other media: Youtie, H. C. and Winter, J. G., eds., Papyri and Ostraca from Karanis, vol. viii (Ann Arbor: University of Michigan Press, 1951) 74Google Scholar.

35 Cicero's concern that his work, the Academica, be well attested by Varro, a powerful leader in Roman society, drove the copy to Rome to receive the finest treatment – from the paper quality to, presumably, the calligraphy – with an accompanying letter dictated syllabatim to another specifically skilled secretary, Spintharus (Cicero, Att. 13.25). Compare Richards, The Secretary, 99.

36 Cicero, Att. 13.33.

37 Cicero, Att. 6.9.

38 Richards, Rediscovering Paul, 69. The absence of grace in the hand is taken as an indication of poor literacy and/or a poorly trained secretary: Longenecker, ‘Ancient Amanuenses’, 286.

39 Diocletian's Price Edict (s. 7) provides two grades of calligraphic scribe (§§41, 42) and a notary's fee (§43); it also recognises (§§70, 71) apprentice-training in writing documents and copying manuscripts, including palaeography (librarius sive antiquarius).

40 Elmer, ‘I, Tertius’, 55, this time citing J. A. Fitzmyer, Romans (AB 33; New York: Doubleday, 1991) 749. Fitzmyer also relies on secondary authorities for a single citation of Tertius and of Tertia, the feminine form.

41 Including all cases yields a far higher number. Iiro Kajanto found 1,042 instances of Tertius used as a cognomen. (He included Tertia.) Tertius therefore ranks as one of the eighteen most common cognomina. Kajanto, I., The Latin Cognomina (Helsinki: Societas Scientiarum Fennica, 1965) 30, 74Google Scholar.

42 The names can occur together – CIL iii.5086, iii.5387, v.7463; RIS 278 – but in each case, they are unrelated. Neither name has been found in inscriptions from Corinth.

43 Perhaps CIL vi.22993 = x.2776? AE 1999, 741 = AE 2001, 1073 is suggestive but incapable of resolution.

44 See Salomies, O., Die römischen Vornamen (Helsinki: Societas Scientiarum Fennica, 1987) 115Google Scholar.

45 See for example CIL v.480, v.6862, vi.6360, vi.36204, xi.786. Kajanto decides against the sequential number-naming of slaves (The Latin Cognomina, 77).

46 Kajanto, The Latin Cognomina, 73–5.

47 Steven Friesen takes the peculiar designation of Quartus as ὁ ἀδελφός as a contrastive from the previously named Erastus, arguing that Erastus was ‘not a participant in the churches’ but Quartus was his (Christian) slave (Friesen, S., ‘The Wrong Erastus: Ideology, Archaeology, and Exegesis’, Corinth in Context: Comparative Studies on Religion and Society (ed. Friesen, S. J., Schowalter, D. N. and Walters, J. C.; Leiden: Brill, 2010) 231–56CrossRefGoogle Scholar, at 251–3).

48 The suggestion is Roller's: Das Formular, 22.

49 As in CIL v.3554.

50 Note Theissen, Studien, 261, 305. Latin names show the influence of Romanitas: Winter, B. W., After Paul left Corinth: The Influence of Secular Ethics and Social Change (Grand Rapids: Eerdmans, 2001) 14–15Google Scholar.

51 Judge, E. A., ‘Greek Names of Latin Origin’, NDIEC 2 (1982) 106–8Google Scholar; idem, Rank and Status in the World of the Caesars and St Paul (Christchurch: University of Canterbury, 1982) 11–14Google Scholar; idem, ‘Latin Names around a Counter-cultural Paul’, The Bible and the Business of Life (ed. Holt, S. and Preece, G.; ATF Series 12; Adelaide: ATF, 2004) 68–84Google Scholar.

52 Judge, ‘Greek Names’, 81–2.

53 This occurs in 39 per cent of cognomina: Kajanto, The Latin Cognomina, 29.

54 See, for example, IG ii².7091, iv.602; I.Thesp 179.

55 Kajanto, The Latin Cognomina, 76–7.

56 Theissen problematised servile status: Studien, 254.

57 See Dutch, R. S., The Educated Elite in 1 Corinthians: Education and Community Conflict in Graeco-Roman Context (London/New York: T&T Clark, 2005) 280–92Google Scholar.

58 Kent, J. H., Corinth viii.3: The Inscriptions 1926–1950 (Princeton: American School of Classical Studies at Athens, 1966) 14–15Google Scholar §46; cf. SEG xliii.210, li.1526.

59 In the Ptolemaic period, the basilikos grammateus had a quite extensive list of functions: see Oates, J. F., The Ptolemaic Basilikos Grammateus (Atlanta: Scholars, 1995) 86–94CrossRefGoogle Scholar.

60 See T. Kruse, Der Königliche Schreiber und die Gauverwaltung, vol. ii (Leipzig/Munich: K. G. Saur, 2002) 773–82.

61 The term is uncommon and can be used for an important royal official as also for a slave; see P.Flor. i.50. Accordingly, the meaning ‘amanuensis’ (so LSJ s.v.) requires refinement.

62 Greg Horsley tentatively suggests these are shorthand writers (ταχυγράφοι): NDIEC i.70, following David Thomas, the editor of the papyrus. This is also the understanding of notarius in Diocletian's Price Edict s. 7 §70.

63 Compare P.Mich. v.326 (dated 48 ce), where, of eighteen slaves, one is mentioned as a muleteer and another as a barber.

64 J. D. Thomas, P.Oxy. xliv.3197, p. 169.

65 See P.Oxy. xliv.3197, Plate viii.

66 In P.Petaus 34 (184 ce) there is payment to an unnamed προχειροφόρος, in service to the komogrammateus, Petaus (l. 24). It is unclear, though likely, that this clerk is a slave. Payment is made to slaves (πετάρια for παιδάρια) of the basilikos grammateus (l. 26). Here the komogrammateus and basilikos grammateus, as holders of significant civic positions, are free, but in their service is an array of slaves, many with specialised skills. See Kruse, Der Königliche Schreiber, 775.

67 P.Oxy. xlii.3062. The hierarchy ascends from a συγγραμματεύς to γραμματεύς to εἰσαγωγεύς. A συγγραμματεύς is also found in BGU ii.451.14, cf. P.Oxy. xii.1427; see also IG i³.71 – suggesting that a συγγραμματεύς was a rank (that is, in a pool of junior clerks), not a fraternal expression.

68 As in P. Vind. Worp 22. Peter Arzt-Grabner considers the phrase ὑπέγραψα χειρὶ ἐμῇ to be a quasi-legal formula capable of attracting judicial recognition (private communication). I am grateful to Professor Arzt-Grabner for a number of suggestions.

69 See Haines-Eitzen, K., ‘“Girls Trained in Beautiful Writing”: Female Scribes in Roman Antiquity and Early Christianity’, JECS 6 (1998) 629–46Google Scholar, at 635–6 for evidence of servile, female scribes.

70 See above, n. 16 and the argument developed below regarding Tertius’ ability.

71 ⁴⁶ inserts a definite article before Τέρτιος; see Junack, K. et al. , eds., Das Neue Testament auf Papyrus, vol. ii: Die paulinischen Briefe Teil 1. Röm., 1. Kor., 2. Kor. (Berlin/New York: de Gruyter, 1989) 148Google Scholar.

72 Cicero, Att. 5.20.

73 Commonly the farewell formula, ἐρρῶσθαί σε εὔχομαι, is added: O.Did. 435; O.Claud. ii.258; P.Brem. 10; P.Herm. 6; P.Heid. ii.212; P.Kell. i.64; P.NYU i.25; see generally Mathew, Women in the Greetings, 22–4.

74 P.Brem. 51, 52.

75 Byrne, Romans, 455–8 considers vv. 17–20 inauthentic; by contrast Longenecker, ‘Ancient Amanuenses’ adds vv. 25–7 as from Paul. See the discussion in Käsemann, E., Commentary on Romans (trans. Bromiley, G. W.; Grand Rapids: Eerdmans, 1980) 422–8Google Scholar. The addition of a brief summary or further thought introduced by παρακαλῶ is attested in letters in the papyri P.Giss.Apoll. 18; P.Lond. ii.144; P.Sarap. 92, cf. P.Iand. vi.96.

76 C. H. Dodd sees Paul's own scribble in vv. 17–20 (psychological shift notwithstanding) with the pen returned to Tertius at v. 21 (Dodd, C. H., The Epistle of Paul to the Romans (London: Hodder & Stoughton, 1932) 241–2)Google Scholar.

77 BGU ii.601; P.Brem. 61; P.Ifao. ii.39; P.KölnGr. ii.109, cf. Chrest. Wilck. 483; P.Giss.Poll. 18; BGU i.261, which has three instances of ἀσπάζεται interrupted by one ἀσπαζόμεθα.

78 As, for example, in BGU iii.821; O.Claud. ii.258; P.Köln ii.108.

79 PSI xii.1247.

80 P.Brem. 48; cf. P.Mert. i.28 (on verso).

81 P.Giss.Apoll. 18. See also P.Oxy. xlix.3505 ll. 2425 ἀσπ̣άσο|μέ [read ἀσπάζομαί] σε Διονύσιος; P.Oxy. xlii.3057 ll. 28–9 ὑποφέρει Λεωνᾶς· ἀσπάζομαι σε might be another example (if the stop is placed after ὑποφέρει), though the editor (Peter Parsons) acknowledges the problems in punctuation and even capitalisation. Λεωνᾶς (lion) may even be a self-characterisation by (or second name of) the author of the letter, given its content about overcoming adversity.

82 Gamble, H., The Textual History of the Letter to the Romans (Grand Rapids: Eerdmans, 1977) 25–9Google Scholar, 101–14, 121–4, 129–32; Käsemann, Commentary, 422–8.

83 In minuscule 457, Rom 16.25–6 follows 14.23 (fol. 147r) and 16.26–7 follows 15.6 (fol. 147v)!

84 For example, Codex A in Rom 8.1 adds (from v. 4) μὴ κατὰ σαρκὰ περιπατοῦσιν; Codex 3 adds ἐν πολλοῖς τῶν ἀντιγράφων οὕτως εὕρηται (!) to δέξασθαι ἡμᾶς, a phrase itself added previously (6, 945). On the mechanics of the incorporation of scribal glosses, see Arzt-Grabner, P., ‘1 Cor. 4.6 – a Scribal Gloss?’, Biblische Notizen 130 (2006) 59–78Google Scholar.

85 For example, P.Princ. ii.19 (160–150 bce); see Clarysse, W., ‘The Archive of Euphron. Avec résumé en anglais’, AncSoc 35 (2005) 129–34Google Scholar; compare P Princ. ii.72 (third century ce).

86 Cicero, Att. 5.1.

87 For transverse writing in the right-hand margin, see P.Turner 18 (dated 84–96 ce).

88 Homann, M., ‘Eine Randerscheinung des Papyrusbriefes: Der versiculus transversus’, Archiv für Papyrusforschung 58 (2012) 67–80Google Scholar.

89 To be added to Homann's list: BGU iii.814, iii.845, O.Claud. ii.278, P.Giss. i.23 (= P.Giss.Apoll. 5), P.Kell. i.66 [?] (postscripts); O.Claud. ii.259 (personal greeting and a formulaic farewell prayer); P.Alex. 28 (continuation of the text with added greetings); P.Haun ii.18 (continuation of the text? with added greetings and postscript); P.Herm. 9 (Christian; prayer/exhortation); P.Mich. viii.504 (fragmentary); P.Abinn. 30, included in her list, also contains greetings and the farewell formula; P.Giss. 103 adds a postscript to the greetings; P.Kell. i.74 has a postscript and farewell formula. For P.Oxy. xvii, read P.Oxy. xvii.2151.

90 See Krebs, F., ed., Ägyptische Urkunden aus den königlichen [staatlichen] Museen zu Berlin, Griechische Urkunden, vol. ii (Berlin: Weidmann, 1898) §§423, 632Google Scholar; White, J. L., Light from Ancient Letters (Foundations & Facets; Philadelphia: Fortress, 1986) 159–60Google Scholar (§103). For an image, see the online Berliner Papyrusdatenbank P. 7950.

91 See, as a mere handful of examples, dated between the first and fourth centuries: BGU i.335, ii.384, Chrest.Wilck. 21 (ἀσπάζομαι alone, though sometimes multiply repeated: P.Abinn. 25; P.Ammon i.3 col. 6; P.Mil. ii.81; P.Paris 18); BGU i.247, P.Lond. iii.951 v. 483, P.Neph. 18 (ἀσπάζομαι; ἀσπάζεται); BGU iii.923, P.Oxy. xxxi.2593 (ἀσπάζομαι; ἀσπαζόμεθα); O.Claud. i.152 (ἀσπάζομαι; ἀσπάζεσθε); O.Claud. ii.279 (ἀσπάζομαι; ἄσπαζε; ἀσπάζεται); O.Did. 352, P Iand. vi.96 (ἀσπάζομαι; ἀσπάζεται; ἀσπάζου); P.Mert. ii.81, P.Oxy. xiv.1677 (ἀσπάζομαι; ἄσπασαι); P.Mert. ii.82 (ἀσπάζομαι; ἄσπασαι; ἀσπάζεται); P.Mert. ii.85 (ἀσπάζεται; ἀσπάζονται; ἀσπάζομαι); P.Mich. iii.208 (ἀσπάζομαι; ἀσπάζετε); P.Oslo ii.52, P.Rein. ii.118 (ἀσπάζομαι; ἀσπάζονται); P.Kell. i.66 (ἀσπάζομαι; ἀσπάσασθε); P.Oxy. iii.533 (ἀσπάσασθε; ἀσπάζονται). There is nothing special in the variety of forms in Romans 16, despite the suggestion of Richards, First-century, 152 n. 42.

92 The number of letters is based on the modern critical editions; whether Tertius used or copied iota adscripts (such as in P.Turner 18, l.2) and/or was content with variant morphology such as ἀσπάσομε (as in P.Oxy. xlix.3505 l. 24) cannot be known.

93 Homann's list contains eight papyri dated to the first century and six dated to the first-to-second centuries. A further forty, including ostraca, are dated to the second century (without specifying those that have provided dates). See Homann, ‘Der versiculus transversus’, 74–80.

94 As Kloppenborg has intimated, the recommendation for hospitable reception of Phoebe in 16.1–2 is balanced (at the end of the greetings list) by a demonstration of hospitality extended to one known to the Christ-followers in Rome (Kloppenborg, ‘Gaius the Roman Guest’, 549).

95 Elmer, ‘I, Tertius’, 54; Richards, The Secretary, 171; Murphy-O'Connor, J., Paul the Letter Writer: His World, his Options, his Skills (GNS 51; Collegeville: Liturgical Press, 1995) 6Google Scholar. Gordon Bahr even suggests that there is an implicit addition of μου after ἐν κυρίῳ, indicating, uniquely, the address of Paul as ‘master’ (Bahr, ‘Paul and Letter Writing’, 465). Richards offers a substantial critique of such a reading (Richards, The Secretary, 172 n. 202).

96 In addition to the above, see P.Mich. viii.504; P.Princ. ii.73; SB xxiv.16267.

97 An interruption or interpolation is the usual designation of his authored appearance in v. 22. This assumes of course that v. 23 is not an extension of Tertius’ own interests but rather a continuation of Paul's own letter.

98 Dunn, Romans 9–16, 910.

99 Dodd, Paul to the Romans, 244.

100 Richards, First-century, 152; The Secretary, 171.

101 Fitzmyer, Romans, 749; Cranfield, Romans, 806.

102 See Byrne, Romans, 460.

103 So Fitzmyer, Romans, 731; Dunn, Romans 9–16, 887.

104 Richards, First-century, 231.

105 See Arzt-Grabner, P., ‘Papyrologie und Neutestamentliche Wissenschaft: Einige Beispiele aus neueren Papyruseditionen’, Light from the East. Papyrologische Kommentare zum Neuen Testament (ed. Arzt-Grabner, P. and Kreinecker, C. M.; Wiesbaden: Harrassowitz, 2010) 11–26Google Scholar, at 17.

106 Pliny, Ep. 4.5, 7.2, 9.2, 9.13.

107 Murphy-O'Connor, Paul: A Critical Life, 268; Dunn, Romans 9–16, 910 (though allowing the possibility of slow dictation). Richards however provides a much more complex process (First-Century, 230–2).

108 Compare Parker, H. N., ‘Books and Reading Latin Poetry’, Ancient Literacies: The Culture of Reading in Greece and Rome (ed. Johnson, W. A. and Parker, H. N.; Oxford: Oxford University Press, 2009) 186–229Google Scholar, at 208–9, 215–17.

109 Pliny, Ep. 9.2.

110 Given Richards and Elmer's reliance on Cicero, another ancient voice may augment the picture.

111 Pliny, Ep. 6.16: tu potissima excerpes; aliud est enim epistulam aliud historiam, aliud amico aliud omnium omnibus scribere. The letter, from which selection was to be made according to genre and audience, has 717 words, a span that escaped an apology for length! Compare the recommended use of history and poetic excerpts within letters: Ep. 7.9, cf. philosophical exercises: Ep. 9.2.

112 Pliny, Ep. 1.6, 7.27, 9.36, cf. 8.15, 8.9, 9.6.

113 Pliny, Ep 1.10.

114 Pliny Ep. 8.21: lego enim omnia ut omnia emendem.

115 See Marchesi, I., The Art of Pliny's Letters: A Poetics of Allusion in the Private Correspondence (Cambridge: Cambridge University Press, 2008) 128CrossRefGoogle Scholar.

116 Pliny, Ep. 7.9, 7.17, 7.30, 9.28, cf. 7.13.

117 Pliny, Ep. 8.19: est autem mihi moris, quod sum daturus in manus hominum, ante amicorum iudicio examinare. For such amici, see Ep. 1.1, 6.33, 7.2, 8.7.

118 Pliny, Ep. 8.21.

119 Pliny, Ep. 7.17; see also 9.26.

120 Pliny, Ep. 9.26.

121 Pliny, Ep. 9.35, cf. 8.21.

122 Seneca, Ep. 40; Galen, In Hipp. De nat. hominis 1.42. This was a principle of textual criticism going back at least to the third-century (bce) Zenodotus of Ephesos, first librarian of Alexandria.

123 See Cadwallader, A. H., ‘Paul Speaks Like a Girl: When Phoebe Reads Romans’, Sexuality, Ideology and the Bible: Queer Readings from the Antipodes (ed. Myles, R. J. and Blyth, C.; Sheffield: Sheffield Phoenix, 2015) 69–94Google Scholar, at 72–6, 82, 85, 88.

124 Pliny, Ep. 7.17. Compare Ovid, Pont. 2.4.13–18, 4.2.35–8; Trist. 3.14.37–52.

125 Pliny, Ep. 9.34. There are occasions when Pliny would read the work himself: Ep. 8.21, 9.28. Compare the use of the slave Salvius as reader by Cicero (Att. 16.2), again preparatory to a final version released to the public. See Parker, ‘Books and Reading’, 209.

126 Pliny, Ep. 7.17, cf. 5.3.

127 Compare Pliny, Ep. 3.5, of his uncle's lector. Cornelius Nepos commented on the inclusion of lectors on Atticus’ staff: Att. 13.3–4.

128 Parker, ‘Books and Reading’, 199, 200.

129 See 1 Cor 15.33, citing Menander's Thaïs.

130 Pliny, Ep. 7.9.

131 Ep. Paul Sen. §7.

132 For Richards, Phoebe is the letter-bearer but is totally dependent on Tertius for the construction of a letter of recommendation (Richards, The Secretary 170, 171). She is not significant enough for an index entry.