1 Harnack, A. von, Die Mission und Ausbreitung des Christentums in den ersten drei Jahrhunderten, vol. ii (Leipzig: J. C. Hinrich, 1906 ²) 1–16Google Scholar, Eng. trans.: The Mission and Expansion of Christianity in the First Three Centuries, vol. ii (London: Williams and Norgate/New York: G. P. Putman's Sons, 1908², rev. and enlarged edn) 1–22.

2 Origen, Comm. ser. Matt. 39 on Matt 24.9: multi enim non solum barbararum, sed etiam nostrarum gentium usque hunc non audierunt Christianitatis verbum.

3 Harnack, Mission and Expansion, ii.23: ‘They tell us hardly anything about [Christianity's] actual spread, though they certainly bear witness to its energetic character and to the fact that the gospel had already reached barbarians, Greeks and Latins.’

4 Pliny, Ep. 10.96.9: multi enim omnis aetatis, omnis ordinis, utriusque sexus etiam vocantur in periculum et vocabuntur. neque civitates tantum, sed vicos etiam atque agros superstitionis istius contagio pervagata est (‘For there are many of all ages, all ranks, and even of both sexes who are or will be summoned to justice. The infection of this superstition has penetrated not only the cities, but the villages and the countryside as well’). On this, see Hopkins, K., ‘Christian Number and Its Implications’, JECS 6 (1998) 185–226Google Scholar.

5 Eck, W., ‘Das Eindringen des Christentums in den Senatorenstand bis zu Konstantin d. Gr’, Chiron 1 (1971) 381–406Google Scholar; Barnes, T. D., ‘Statistics and the Conversion of the Roman Aristocracy’, JRS 85 (1995) 135–47Google Scholar; Weiß, A., Soziale Elite und Christentum: Studien zu ordo-Angehörigen unter den frühen Christen (Millennium-Studien / Millennium Studies 52; Berlin/New York: de Gruyter, 2015) 189–200Google Scholar.

6 See Mckechnie, P., ‘Christian City Councillors in the Roman Empire Before Constantine’, Interdisciplinary Journal of Research on Religion 5 (2009) 2–20Google Scholar; Öhler, M., ‘Graeco-Roman Associations, Judean Synagogues and Early Christianity in Bithynia-Pontus’, Authority and Identity in Emerging Christianities in Asia Minor and Greece (ed. Breytenbach, C. and Ogereau, J.; AJEC 103; Leiden: Brill, 2018) 62–88CrossRefGoogle Scholar.

7 Mullen, R. L., The Expansion of Christianity: A Gazetteer of its First Three Centuries (VC Suppl. 69; Leiden/Boston: Brill, 2004)CrossRefGoogle Scholar; Trombley, F., ‘Overview: The Geographical Spread of Christianity’, The Cambridge History of Christianity, vol. i: Origins to Constantine (ed. Mitchell, M. M. and Young, F. M.; Cambridge: Cambridge University Press, 2006) 302–13CrossRefGoogle Scholar; Tabbernee, W., ed., Early Christianity in Contexts: An Exploration across Cultures and Continents (Grand Rapids: Baker Academic, 2014)Google Scholar.

8 Rothschild, C. K. and Schröter, J., eds., The Rise and Expansion of Christianity in the First Three Centuries of the Common Era (WUNT 301; Tübingen: Mohr Siebeck, 2012)Google Scholar.

9 Breytenbach, C., ed., ‘The Rise and Expansion of Christianity in Asia Minor: First Steps Towards a New Harnack’, EC 2 (2011) 547–52CrossRefGoogle Scholar. Two volumes are in print: Huttner, U., Early Christianity in Asia Minor, vol. i:Early Christianity in the Lycus Valley (AJEC 85; Leiden/Boston: Brill, 2013)Google Scholar; Breytenbach, C. and Zimmermann, C., eds., Early Christianity in Asia Minor, vol. ii: Early Christianity in Lycaonia and Adjacent Areas: From Paul to Amphilochius of Iconium (AJEC 101; Leiden and Boston: Brill, 2017)CrossRefGoogle Scholar.

10 B. Kowalzig, ‘Cults, Cabotage, and Connectivity: Experimenting with Religious and Economic Networks in the Greco-Roman Mediterranean’, Maritime Networks in the Ancient Mediterranean World (ed. J. Leidwanger and C. Knappett (Cambridge: Cambridge University Press, 2013) 93–131; C. Bonnet and L. Bricault, Quand les dieux voyagent. Cultes et mythes en mouvement dans l'espace méditerranéen antique (Geneva: Labor et Fides, 2016).

11 M. Malaise, ‘La diffusion des cultes égyptiens dans les provinces européennes de l'Empire romain’, ANRW ii.17.3 (1984) 1615–91; T. Hegedus, ‘The Urban Expansion of the Isis Cult: A Quantitative Approach’, Studies in Religion / Sciences Religieuses 27 (1998) 161–78; L. Bricault, Atlas de la diffusion des cultes isaiques (iv^e s. av. J.-C. – iv^e s. apr. J.-C.) (Mémoires de l'académie des inscriptions et belles-lettres 23; Paris: Éditions de Boccard, 2001); L. Bricault, ‘La diffusion isiaque, une ésquisse’, Fremdheit - Eigenheit: Ägypten, Griechenland und Rom. Austausch und Verständnis (ed. P. C. Bol, G. Kaminski and C. Maderna; Städel-Jahrbuch, Neue Folge 19; Munich: Prestel, 2004) 548–56.

12 Peter Marshall Fraser, ‘Two Studies on the Cult of Sarapis in the Hellenistic World’, Opuscula Atheniensia 3 (1996): 1–54.

13 A. Collar, ‘Military Networks and the Cult of Jupiter Dolichenus’, Asia Minor Studies 64 = Dolichener und Kommagenische Forschungen iv (2011) 217–45, Plates 1–3; ‘Commagene, Communication and the Cult of Jupiter Dolichenus’, Iuppiter Dolichenus: vom Lokalkult zur Reichsreligion (ed. M. Blömer and W. Engelbert; Orientalische Religionen in der Antike 8; Tübingen: Mohr Siebeck, 2012) 99–109; Religious Networks in the Roman Empire: The Spread of New Ideas (Cambridge: Cambridge University Press, 2013). See also the review by Mihály Loránd Dészpa, ‘Rez.: Michael Blömer – Engelbert Winter (hrsg.), Iuppiter Dolichenus. Vom Lokalkult zur Reichsreligion’, Klio 96.2 (2014): 749–56.

14 On the application of social network theory to antiquity, see in general S. Graham and G. R. Ruffini, ‘Network Analysis and Greco-Roman Prosopography’, Prosopography Approaches and Applications: A Handbook (ed. K. S. B. Keats-Rohan; Prosopographia et Genealogica; Oxford: Oxford University Press, 2007) 325–36; A. Collar, ‘Network Theory and Religious Innovation’, Greek and Roman Networks in the Mediterranean (ed. I. Malkin, C. Constantakopoulou and K. Panagopoulou (London/New York: Routledge, 2009) 144–57. For particular applications: G. R. Ruffini, Social Networks in Byzantine Egypt (Cambridge/New York: Cambridge University Press, 2008) (Byzantine Egypt); A. M. Schor, Theodoret's People: Social Networks and Religious Conflict in Late Roman Syria (Transformation of the Classical Heritage 48; Berkeley: University of California Press, 2011); Collar, Religious Networks; C. Taylor and K. Vlassopoulos, eds., Communities and Networks in the Ancient Greek World (Oxford: Oxford University Press, 2015); E.-M. Becker, The Birth of Christian History: Memory and Time from Mark to Luke-Acts (ABRL; New Haven: Yale University Press, 2017) 82–5 (Luke-Acts); and J. Fousek et al., ‘Spatial Constraints on the Diffusion of Religious Innovations: The Case of Early Christianity in the Roman Empire’, PLoS ONE 13.12 (2018) e0208744.

15 M. S. Granovetter, ‘The Strength of Weak Ties’, American Journal of Sociology 78 (1973) 1360–80, at 1361: ‘the strength of a tie is a (probably linear) combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and the reciprocal services which characterize the tie. Each of these is somewhat independent of the other, though the set is obviously highly intracorrelated.’

16 Granovetter, ‘Weak Ties’; see also M. S. Granovetter, ‘The Strength of Weak Ties Revisited’, Sociological Theory 1 (1983) 201–33. ‘Strong’ and ‘weak’ are of course relative to one's position in networks. A weak tie (acquaintance) in one network might have strong ties to other networks, and mutatis mutandis with strong ties.

17 Granovetter, ‘Weak Ties Revisited’, 205. Granovetter reports other empirical studies that supported the ‘strength of weak ties’ thesis, but noted differences between administrative and managerial employees, office workers, semi-professionals and blue collar workers, and differences indexed to educational levels.

18 Granovetter, ‘Weak Ties Revisited’, 202. D. J. Watts, Six Degrees: The Science of a Connected Age (New York: Norton, 2003) 230 observes that an isolated and tightly organised cluster – for example, the Branch Davidians – could maintain a set of completely implausible beliefs and practices and resist more reasonable ideas as long as they maintained a network of mutually reinforcing influences (strong ties) and prevented nodes from interacting with those outside the cluster (weak ties). The point is that strong ties resist innovation (or new ideas).

19 D. J. Watts, ‘Networks, Dynamics, and the Small-World Phenomenon’, American Journal of Sociology 105 (1999) 493–527; D. J. Watts and S. H. Strogatz, ‘Collective Dynamics of “Small-World” Networks’, Nature 393 (4 June 1998) 440–2. The classic study is J. Travers and S. Milgram, ‘An Experimental Study of the Small World Problem’, Sociometry 32 (1969) 425–43.

20 D. Centola, How Behavior Spreads: The Science of Complex Contagions (Princeton Analytical Sociology Series; Princeton: Princeton University Press, 2018) 27–9.

21 For the diffusion of elective cults, epidemiological models have been endorsed by Collar, Religious Networks and G. Woolf, ‘Only Connect? Networks and Religious Change in the Ancient Mediterranean World’, Hélade 2 (2009) 43–58, at 51.

22 Agency plays virtually no role in the transmission of measles. Because measles is a simple contagion with a very high reproduction rate or ‘R-nought number’ (R₀), an unvaccinated person coming in close contact with an infected person will in all probability become infected. The R₀ for measles, which is transmitted by aerosol contact, is 12–20, meaning that an infected subject will infect on average 12–20 susceptible persons. By contrast, the R₀ for cholera, which has a faecal-oral pathway, is in many locales < 3.0.

23 P. Fine, K. Eames and D. L. Heymann, ‘“Herd Immunity”: A Rough Guide’, Clinical Infectious Diseases 52 (2011) 911–16.

24 Centola, How Behavior Spreads, 40–2.

25 See also D. McAdam and R. Paulsen, ‘Specifying the Relationship between Social Ties and Activism’, American Journal of Sociology 99 (1993) 640–67, at 646, who observe in relation to activism, ‘any major decision we are contemplating will likely be mediated by a significant subset of [close] relationships. This, of course, would apply to participation in any significant forms of activism, especially those of the “high-risk” variety.’

26 See IG xi/4.1299 (Delos, before 166 bce) and IG x/2.1.255 (Thessaloniki, 1st–2nd cent. ce) = GRA i.77.

27 R. S. Ascough, ‘The Thessalonian Christian Community as a Professional Voluntary Association’, JBL 119 (2000) 311–28. See also J. S. Kloppenborg, Christ's Associations: Connecting and Belonging in the Ancient City (New Haven: Yale University Press, 2019) 88–91.

28 Acts 17.1 mentions only Paul and Silas (= Silvanus?) entering Thessaloniki, but Acts 17.14, 15 and 18.5 suggest that Timothy was also present.

29 See M. Gladwell, The Tipping Point: How Little Things Can Make a Big Difference (Boston: Little, Brown, 2002); Watts, Six Degrees, 239–41; C. Kadushin, Understanding Social Networks: Theories, Concepts, and Findings (New York: Oxford University Press, 2012) 153–5. See also I. Malkin, A Small Greek World: Networks in the Ancient Mediterranean (Greeks Overseas; New York: Oxford University Press, 2011) 36: the phrase transition occurs ‘when a network reaches a point when nodes appear in clusters with only a few links connecting them. The addition of a relatively small number of links can result in isolated groups suddenly coalescing into a giant cluster … The giant cluster engenders multidirectional ripple effects.’

30 R. S. Ascough, ‘Re-Describing the Thessalonians’ “Mission” in Light of Greco-Roman Associations’, NTS 60 (2014) 61–82 is correct, I think, to interpret 1 Thess 1.8 to mean not that the Thessalonian Christ devotees had themselves become ‘missionaries’, but rather that the news of the coming of Christ to Thessaloniki had become known. Indeed, one of the obligations of a newly arrived deity is for the adherents to broadcast that arrival.

31 Given the spatial clustering of trades in an ancient city, it was relatively easy for a newcomer to a city to locate his trade, and once connected, a ready-made network was available. A guild of, say, awning makers (σκηνοποιοί, Acts 18.3) might be located adjacent to other guilds associated with textile manufacture. The Tosephta says that the great synagogue in Alexandria was structured in order to facilitate travelling artisans being able to connect with persons of similar trades (t. Sukk. 4.6). Other diaspora assemblies may similarly have functioned to connect artisans and merchants with each other.

32 Fortunatus is one of the most common slave names. See T. Frank, ‘Race Mixture in the Roman Empire’, AHR 21 (1916) 589–708, at 692, and H. Solin, Die stadtrömischen Sklavennamen: Ein Namenbuch, vol. i (Forschungen zur antiken Sklaverei, Beiheft 2; Stuttgart: Franz Steiner, 1996) 95–8 (Fortunatus/-a: 225 occurrences). Achaicus is not otherwise attested in Solin, but since it means ‘the Achaian’, we might presume that Achaicus was as some point the Greek slave of a Roman owner.

33 According to the Orbis site (http://orbis.stanford.edu/), travel from Corinth to Ephesus would take 3.1 days by sea, and almost 50 by land.

34 CIL xi.5748, a bronze tablet, 57 × 40 cm., Musei Capitolini, inv. 7342. Sala della Colombe.

35 See J. Nicols, ‘Tabulae patronatus: A Study of the Agreement between Patron and Client-Community’, ANRW ii.13 (1980) 535–61.

36 CIL xi.5737; ILS 4215; CIMRM i.688.

37 CIL xi.5749. They were also patrons of the dendrophores, but we lack epigraphical evidence for the dendrophores at Sentium.

38 See now a reconstruction of cultic networks in Ostia: J. S. Kloppenborg, ‘Occupational Guilds and Cultic Associations in Ostia Antica: Patronage, Mobility, Connectivity’, Roman Imperial Cities in the East and in Central-Southern Italy (ed. N. Andrade, et al.; Ancient Cities; Rome: L'Erma di Bretschneider, 2019) 413–36 and the earlier work of D. Rohde, Zwischen Individuum und Stadtgemeinde: Die Integration von collegia in Hafenstädten (Studien zur Alten Geschichte; Mainz: Verlag Antike, 2012).

39 Measles, for example, is transmitted through aerosol contact and has an incubation period of about two weeks; but the virus is vulnerable to high temperatures, sunlight and low (acidic) pH environments. This means that measles propagates best in temperate environments and has the highest incidence in late winter/early spring or after rainy seasons. Of course, measles is not able to spread effectively in a population that has achieved the threshold for ‘herd immunity’, since vaccinated (non-susceptible) links in the network break the pathways for the diffusion of the infection.

40 D. Mitrofan, ‘The Antonine Plague in Dacia and Moesia Inferior’, Journal of Ancient History and Archeology 1 (2014) 9–13. J. Liu (‘Group Membership, Trust Networks, and Social Capital: A Critical Analysis’, Work, Labour, and Professions in the Roman World (ed. K. Verboven and C. Laes; Impact of Empire 23; Leiden: Brill, 2017) 203–26, at 200–12) suggests by contrast that the failure of the group is due to an economic downturn that resulted from the Marcomannic War, which meant that fewer members could afford the yearly or monthly fees, and fewer were in a position to make special contributions, coupled with the inability of the collegium to enforce (with fines) its behavioural rules.

41 I.Beroia 22 (Beroia, 7 bce) = SEG 48.751: ἔτους α’ καὶ μ’ καὶ ρ’ Ἀρτεμισίου· | Παράμονος Θεογένους vac. | ἀγορανομήσας τοῦ θιάσου | ἐκ τοῦ ἰδίου Διονύσωι. vac. || τὸ κοινὸν τῶν θιασ[ω]|τῶν Παράμονον | Θεογένου. <crown> (‘Year 441, (month of) Artemesion. Paranomos son of Theogenes, having served as agoranomos of the association at his own expense, (dedicated this) to Dionysos. The association of thiasōtai  (honoured) Paranomos son of Theogenes’).

42 On the role of the ἀγορανόμος in Egyptian towns, see R. Alston, The City in Roman and Byzantine Egypt (London/New York: Routledge, 2002) 190–2.

43 For the supply of wood for roasting sacrifices, see IG xii/5.606.6 (Iulis, Cos); P.Cair.Zen. ii.59154 (Philadelphia, 256 bce), 59191.3–4 (Philadelphia, 255 bce); P.Oxy. viii.1144.15 (Oxyrhynchus, 75–125 ce); SEG 31.122.42, 45 = GRA i.50 (Liopesi, ca. 100 ce).

44 Since Ptolemaios does not mention any others in the letter apart from δύο φίλοι (both nameless), the network in Arsinoe is unlikely to be a family network of persons known to Kastor.

45 K. Vandorpe, W. Clarysse and H. Verreth, Graeco-Roman Archives from the Fayum (Collectanea Hellenistica - KVAB vi; Leuven/Paris/Bristol: Peeters, 2015) 373–8 and TM archive at www.trismegistos.org/arch/detail.php?arch_id=109. The archive consists of at least thirty papyri (letters, declarations, census returns) and perhaps as many as fifty papyri. See also P. van Minnen, ‘House-to-House Enquiries: An Interdisciplinary Approach to Roman Karanis’, ZPE 100 (1994) 227–51; S. Strassi, ‘Le carte di Σωκράτης Σαραπίωνος, πράκτωρ ἀργυρικῶν a Karanis nel ii sec. d.C’, Atti del xxii Congresso internazionale di papirologia, Firenze, 23–29 agosto 1998, vol. ii (ed. I. Andorlini et al.; (Florence: Istituto papirologico G. Vitelli, 2001) 1215–28; D. Hagedorn, ‘Sokrates und Asklepiades, Praktoren in Karanis’, ZPE 167 (2008) 149–50.

46 The archive includes P.Mich. ix.536 (Karanis, 185 ce), a nomination list for liturgies of assisting in tax collection, prepared by Kastor. The list includes the net worth of each individual and declares that all are εὔποροι καὶ ἐπι‹τή›δειοι, ‘financially able and suitable’.

47 See further discussion of the recruitment of Ptolemaios in J. S. Kloppenborg, ‘Social Networks and the Dissemination of Elective Cults’, Early Christianity 19 (2019) 121–56.

48 Kloppenborg, Christ's Associations, 209–39.

49 U. Fellmeth, ‘Die römischen Vereine und die Politik: Untersuchungen zur sozialen Schichtung und zum politischen Bewußtsein in den Vereinen der städtischen Volksmassen in Rom und Italien’ (Diss., Stuttgart, 1987); ‘Politische Bewußtsein in den Vereinen der städtlichen Massen in Rom und Italien zur Zeit der Republik und der frühen Kaiserzeit’, Eirene: Studia Graeca et Latina 27 (1990) 49–71 analysed the records of twenty-nine Italian associations (more than 2,300 names) and pointed out that while occupational guilds often had patrons/members of senatorial or equestrian ranks, none of the cultic associations had members or patrons above the rank of local magistrate. In a wider-ranging study (Christ's Associations, 192–6) I have suggested that the same was more generally true, with a very few notable exceptions. It is not until the very late second or third century ce that Christ associations were able to attract members of rank.

50 See e.g. I.Beroia 27 (Beroia, before 212 ce), a poorly cut dedication by a diverse group of thirty-seven handworkers and slaves, and including two Roman citizens (neither of senatorial or equestrian rank). On the dedications of the cult of Theos Hypsistos, see S. Mitchell, ‘Further Thoughts on the Cult of Theos Hypsistos’, One God: Pagan Monotheism in the Roman Empire (ed. S. Mitchell and P. Van Nuffelen (Cambridge: Cambridge University Press, 2010) 167–208, at 178–9.

51 See CIRB 1283 (Tanais, 228 ce), and P. A. Harland, Greco-Roman Associations: Texts, Translations, and Commentary, vol. ii:North Coast of the Black Sea, Asia Minor (BZNW 204; Berlin/New York: de Gruyter, 2014) 63–70.

52 On signalling costs, see R. Sosis and C. Alcorta, ‘Signaling, Solidarity, and the Sacred: The Evolution of Religious Behavior’, Evolutionary Anthropology 12 (2003) 264–74; R. Sosis, ‘The Adaptive Value of Religious Ritual: Rituals Promote Group Cohesion by Requiring Members to Engage in Behavior That is Too Costly to Fake’, American Scientist 92 (2004) 166–72; J. W. Dow, ‘The Evolution of Religion: Three Anthropological Approaches’, MTSR 18 (2006):67–91; J. Bulbulia and R. Sosis, ‘Signalling Theory and the Evolution of Religious Cooperation’, Religion 41 (2011) 363–88.

53 L. W. Hurtado, Destroyer of the Gods: Early Christian Distinctiveness in the Roman World (Waco, TX: Baylor University Press, 2016) 118–19. See earlier his The Earliest Christian Artifacts: Manuscripts and Christian Origins (Grand Rapids: Eerdmans, 2006).

54 Cicero, Fin. 5.52–3: quid, cum fictas fabulas, e quibus utilitas nulla elici potest, cum voluptate legimus? (‘But what, when we read for pleasure fictitious stories from which no utility can be had?’). Cicero continues by noting (with some contempt) that persons of ‘the humblest station (homines infima fortuna), who have no expectation of participating in public life, even mere artisans (opifices)’ read history.

55 Cavallo, G., ‘Between volumen and Codex: Reading in the Roman World’, A History of Reading in the West (ed. Cavallo, G. and Cartier, R. (Cambridge: Polity, 1999) 65–89Google Scholar, at 76. G. Bazzana (‘“You Will Write Two Booklets and Send One to Clement and One to Grapte”: Formal Features, Circulation, and Social Function of Ancient Apocalyptic Literature’, Scribal Practices and Social Structures among Jesus Adherents: Essays in Honour of John S. Kloppenborg (ed. W. E. Arnal et al.; BETL 285; Leuven: Peeters, 2016) 43–70) suggests that some of the earliest copies of apocalyptic literature circulated among such ‘free readers’. See also Schwendner, G., ‘Literature and Literacy at Roman Karanis: Maps of Reading’, Proceedings of the xxiv International Congress of Papyrology (ed. Frösén, J., Purola, T. and Salmenkivi, E.; Helsinki: Societas Scientiarum Fennica, 2007) 991–1006Google Scholar and Bagnall, R. S., ‘An Owner of Literary Papyri’, CP 87 (1992) 137–8Google Scholar.

56 Respectively, LDAB 4764; LDAB 2643; and LDAB 15. Other administrative archives likewise include literary texts. See Vandorpe, Clarysse and Verreth, Graeco-Roman Archives, 375.

57 Luijendijk, A., ‘A New Testament Papyrus and its Documentary Context: An Early Christian Writing Exercise from the Archive of Leonides (“P.Oxy.” ii 209/ P¹⁰)’, JBL 129 (2011) 575–96Google Scholar. For the non-canonical gospels (P.Oxy. ii.210), see Smith, G. S. and Landau, B. C., ‘Canonical and Apocryphal Writings Copied by the Same Scribe: P.Oxy. ii 209 (=P¹⁰) and P.Oxy. ii 210, and the Archive of Aurelius Leonides’, ETL 95 (2019) 143–60Google Scholar.

58 E. M. Rogers, Diffusion of Innovations (New York: Free Press, 1995⁴) 254–8. Most of the studies of innovation-adoption have been done on North American and European subjects, but Rogers reports a study on the adoption of agricultural innovations by Colombian peasants that followed the same pattern as that of Western subjects.

59 See Kloppenborg, Christ's Associations, 108–11: the mean size of a cultic association is 29.29 members.

60 Rogers, Diffusion, 263–5.

61 It is striking that in Romans 15, when Paul conveys greetings to his addressees, neither Stephanas nor Crispus is mentioned.

62 Stark, R., The Rise of Christianity: A Sociologist Reconsiders History (Princeton: Princeton University Press, 1996) 15–21Google Scholar, at 16–17; Cities of God: The Real Story of How Christianity Became an Urban Movement and Conquered Rome (New York: HarperSanFrancisco, 2006) 8–13. One study cited by Stark, , Kox, W., Meeus, W. and ‘t Hart, H., ‘Religious Conversion of Adolescents: Testing the Lofland and Stark Model of Religious Conversion’, Sociological Analysis 52 (1991) 227–40Google Scholar, in fact concludes that in the case of Dutch adolescents, the appeal of a recruiting group is twofold, both ‘ideological, by offering a new perspective on life, and social, by providing a satisfactory social network’, although the overwhelming majority of new converts (80 per cent) had established affective bonds with other members of the group in the process of conversion.

63 Stark's model has been adopted in K. Eshleman's study of conversion to philosophy, ‘Affection and Affiliation: Social Networks and Conversion to Philosophy’, CJ 103 (2008) 129–40. In relation to Epicureanism, she concludes, ‘it is worth observing that both the teachings and the social ties play a part in full conversion; it would certainly be wrong to emphasize social networking at the expense of ideological conviction. Yet to focus on ideology to the exclusion of social bonds, as is more often done, is to ignore a crucial component of the process by which people came into contact with philosophers’ teachings and were convinced of their truth’ (139).

64 Monson, A., ‘The Ethics and Economics of Ptolemaic Religious Associations’, Ancient Society 36 (2006) 221–38CrossRefGoogle Scholar. See also P. F. Venticinque, Honor among Thieves: Craftsmen, Merchants, and Associations in Roman and Late Roman Egypt (New Texts from Ancient Cultures; Ann Arbor: University of Michigan Press, 2016); W. Blockmans, ‘Inclusiveness and Exclusion: Trust Networks at the Origins of European Cities’, Theory and Society 39.3/4, Special Issue in Memory of Charles Tilly [1929–2008]: Cities, States, Trust, and Rule (2010) 315–26.

65 Details in Kloppenborg, Christ's Associations, 151–9.

66 Sosis, R. and Bressler, E. R., ‘Cooperation and Commune Longevity: A Test of the Costly Signaling Theory of Religion’, Cross-Cultural Research 37 (2003) 211–39CrossRefGoogle Scholar. See also Sosis, R., ‘Religion and Intragroup Cooperation: Preliminary Results of a Comparative Analysis of Utopian Communities’, Cross-Cultural Research 34 (2000) 77–88CrossRefGoogle Scholar and ‘Religious Behaviors, Badges, and Bans: Signaling Theory and the Evolution of Religion’, Where God and Science Meet: How Brain and Evolutionary Studies Alter our Understanding of Religion, vol. i: Evolution, Genes and the Religious Brain (ed. P. McNamara; Westport, CT: Praeger, 2006) 61–85.

67 For example, IG ii².1297.3–4 = GRA i.24 (Athens, 236/235 bce): ἐπειδὴ Σώφρων καλῶς καὶ φ[ιλ]οτί|μως συνήγαγε τὸν θίασον (‘since Sophron honourably and with public spirit had convened the thiasos’). See also LSAM 48.3–4 (Miletos, 276/5 bce); IG ii².1343 = GRA i.43 (Athens, 37/36 or 36/35 bce); IG ii².1369.25–6 = GRA i.48 (Liopesi, ca. 100 ce); IG ii².1366.21 = GRA i.53 (Laurion, ca. 200 ce).

68 SEG 52.761.9–10 = G. Siebert, ‘Sur l'histoire du sanctuaire des dieux syriens à Délos’, BCH 92 (1968) 360 (Delos, 2nd cent. bce): τὸ κοινὸν τῶν θιασιτῶ[ν] … οὓς συνήγαγε || ἡ θεός (‘the association of the thiasitai … whom the goddess has convened’). Similarly, ID 2225.3–5 (Delos, end of 2nd cent. bce): ἀ[πὸ τῶν? θι]|ασιτῶν Ἁγνῆς Θεοῦ οὗς συνήγα[γε] (‘the thiasitai whom she has convened’). See M.-F. Baslez, ‘Entre traditions nationales et intégration. Les associations sémitiques du monde grec’, La questione delle influenze vicino-orientali sulla religione greca: stato degli studi e prospettive della ricerca (ed. S. Ribichini, M. Rocchi and P. Xella (Monografie Scientifiche. Serie Scienze Umane e Sociali; Rome: Consiglio nazionale delle ricerche, 2001) 235–47, esp. 242.

69 Welborn, L. L., ‘How “Democratic” Was the Pauline Ekkles̄ia? An Assessment with Special Reference to the Christ Groups of Roman Corinth’, NTS 65 (2019) 289–309CrossRefGoogle Scholar, at 309. See also A. C. Miller, Corinthian Democracy: Democratic Discourse in 1 Corinthians (Princeton Theological Monograph Series 220; Eugene, OR: Pickwick Publications, 2015) and Kloppenborg, J. S., ‘Associations, Christ Groups, and their Place in the Polis’, ZNW 108 (2017) 1–56CrossRefGoogle Scholar.

70 S. T. Payne, ‘Spiritual Bodies and the Afterlives of Ancient Democracy in Early Paulinism’ (PhD diss.; New York: Fordham University Press, 2019).

71 Jones, N. (The Associations of Classical Athens: The Response to Democracy (London/New York: Oxford University Press, 1999)Google Scholar) argues that the rise of deme-based associations compensated for the de facto exclusion of most of the Athenian population from democracy. (With a citizen population of about 30,000, and an assembly that could hold at most 6,000 persons, the majority of Athenians from the inland and coastal demes were excluded from the practice of democracy). J. Ustinova (‘Orgeones in Phratries: A Mechanism of Social Integration in Attica’, Kernos 9 (1996) 227–42) has proposed something parallel, but at an earlier stage of Athenian history. Immigrants in the eighth and seventh centuries who were incorporated as citizens into the Athenian demes but not into the clans (the centre of cultic activity) formed cultic associations called orgeōnes, which were ‘effective in liberating the immigrants from the sense of inferiority and defenselessness, both in the cultic and the social spheres’ (241). K. Vlassopoulos (‘Free Spaces: Identity, Experience and Democracy in Classical Athens’, CQ 57 (2007) 33–52) and A. Gottesman (Politics and the Street in Democratic Athens (Cambridge/New York: Cambridge University Press, 2014)) have likewise argued that the formation of small groups in Athens provided alternative forms of sociality for those de facto excluded from Athenian democracy.

72 I have elsewhere argued that the Pauline collection for the poor, patterned on a civic epidosis, can be seen as a performance of citizenship in a fictive translocal polity that embraced not only the Pauline groups of Achaia, Macedonia and (perhaps) Asia, but also a non-Greek community in Palestine. See Kloppenborg, J. S., ‘Fiscal Aspects of Paul's Collection for Jerusalem’, Early Christianity 8 (2017) 153–98CrossRefGoogle Scholar.

73 I am grateful to Ms Christina Gousopoulos, who has assisted me in producing the network visualisations in this paper.