Hostname: page-component-7c8c6479df-p566r Total loading time: 0 Render date: 2024-03-28T20:00:54.331Z Has data issue: false hasContentIssue false

Paul, Philemon and the Dilemma of Christian Slave-Ownership

Published online by Cambridge University Press:  05 February 2009

Extract

By any standards, discussion of Paul's reaction to slavery and slave-ownership is exceptionally complex and controversial. There are more than enough difficulties in attempting to grasp the true character and significance of slavery in the Graeco-Roman world of the first century, due not merely to the fragmentary and one-sided nature of our evidence but also to the deep-rooted political and philosophical commitments which influence almost every significant treatment of the topic. But to discuss Paul in relation to slavery is to add further complications. If, with the majority of scholars, we bracket off Col 3. 22–4. 1, Eph 6. 5–9, 1 Tim 6. 1–2 and Tit 2. 9–10 as Deutero-Pauline, we are left with comparatively few texts which refer directly to the institution of slavery: Gal 3. 28 (cf. 1 Cor 12. 13), 1 Cor 7. 21–24 and the letter to Philemon are all that remain. But even these texts contain sufficient complexities and ambiguities to render an assessment of Paul's view of slavery far from straightforward. Moreover, varying ideological commitments play a significant role in interpretation here too.

Type
Articles
Copyright
Copyright © Cambridge University Press 1991

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 The best survey of these issues is by Finley, M. I., Ancient Slavery and Modern Ideology (London: Chatto & Windus, 1980) 1166Google Scholar; cf. Vogt, J., Ancient Slavery and the Ideal of Man (Oxford: Blackwell, 1974) 170210Google Scholar. For a fuller description of the debate between Marxist and non-Marxist see Brockmeyer, N., Antike Sklaverei (Erträge der Forschung 116; Darmstadt: Wissenschaftliche Buchgesellschaft, 1979) 373Google Scholar; Yavetz, Z., Slaves and Slavery in Ancient Rome (New Brunswick: Transaction, 1988) 115–75Google Scholar.

2 Overbeck, F., Studien zur Geschichte der alten Kirche(Schloss-Chemnitz: E. Schmeitzner, 1875) 158230Google Scholar; Kehnscherper, G., Die Stellung der Bibel und der alten christlichen Kirche zur Sklaverei (Halle: Max Niemeyer, 1957) 7996Google Scholar; Schulz, S., Gott ist kein Sklavenhalter (Zürich: Flamberg Verlag; Hamberg: Furche Verlag, 1972) 167193.Google Scholar

3 Lightfoot, J. B., Saint Paul's Epistles to the Colossians and to Philemon (5th ed.; London: Macmillan, 1880) 323Google Scholar; Wright, N. T., Colossians and Philemon (TNTC; Leicester: Inter-Varsity, 1986) 150, 169.Google Scholar

4 Preiss, T., ‘Life in Christ and Social Ethics in the Epistle to Philemon’, in Life in Christ (SBT; London: SCM, 1954) 3242.Google Scholar

5 Moule, C. F. D., The Epistles to the Colossians and to Philemon (Cambridge: Cambridge University, 1957) 11.CrossRefGoogle Scholar

6 Scott, E. F., The Epistles to the Colossians and to Philemon and to the Ephesians (MNTC; London: Hodder and Stoughton, 1930) 100Google Scholar. Cf. Wallon, H., Histoire de l'esclavage dans l'antiquité (3 vols.; 2nd ed.; Paris: Hachette, 1879)Google Scholar and Allard, P., Les esclaves chrétiens, depuis les premiers temps de l'Église (Paris: Gabalda, 1876).Google Scholar

7 Conzelmann, H., 1 Corinthians (Hermeneia; Philadelphia: Fortress, 1975) 126Google Scholar. Lohse, Cf. E., Colossians and Philemon (Hermeneia; Philadelphia: Fortress, 1971) 203, 205Google Scholar and Luther's famous statement in introducing Galatians: Christus nos liberavit, non e servitute aliqua humana aut vi tyrannorum, sed ira dei aeterna', WA 40/2, 3Google Scholar. Cf. Schweizer, E., ‘Zum Sklavenproblem im Neuen Testament’, EvTh 32 (1972) 502–6.Google Scholar

8 Most of the literature on slavery is listed in Bibliographie zur antiken Slaverei ed. J. Vogt and H. Bellen (reworked by E. Herrmann with N. Brockmeyer; Bochum: Studien-verlag Dr. N. Brockmeyer, 1983)Google Scholar. On slavery in early Christianity see e.g. Gülzow, H., Christentum und Sklaverei in den ersten drei Jahrhunderten (Bonn: Habelt, 1969)Google Scholar; Gayer, R., Die Stellung des Sklaverei in den paulinischen Gemeinden und bei Paulus (Bern/Frankfurt: Lang, 1976)Google Scholar; Laub, F., Die Begegnung des frühen Christentums mit der antiken Sklaverei (SB 107; Stuttgart: Katholisches Bibelwerk, 1982).Google Scholar

9 The current consensus places Paul in Ephesus in the mid-50s; Lohse, , Colossians and Philemon, 188Google Scholar; Gnilka, J, Der Philemonbrief (HTKNT; Freiburg/Basel/Vienna: Herder, 1982) 45.Google Scholar

10 Knox, J., Philemon among the Letters of Paul (London: Collins, 1960, but first published in 1935)Google Scholar. See Lohse, , Colossians and Philemon, 190–1.Google Scholar

11 Philemon, 15.

12 Winter, S. C., ‘Paul's Letter to Philemon’, NTS 33 (1987) 115, at 3 (emphasis hers)CrossRefGoogle Scholar; cf. her earlier essay, ‘Methodological Observations on a New Interpretation of Paul's Letter to Philemon’, USQR 39 (1984) 203–12Google Scholar. See also Houlden, J. L., Paul's Letters from Prison (London: SCM, 1970) 226Google Scholar and Bruce, F. F., The Epistles to the Colossians, to Philemon, and to the Ephesians. (Grand Rapids: Eerdmans, 1984) 197.Google Scholar

13 It is likely that Paul is playing on the meaning of the slave's name (Onesimus means ‘useful’) and it is possible that there is a further play on words in that ἄχρηστος would be pronounced much like ἄχριστος (i.e. without Christ); see e.g. Lohse, , Colossians and Philemon, 200–1 with references to Justin and Tertullian.Google Scholar

14 Bruce, , Epistles, 219.Google Scholar

15 Most take the reference to debt as signalling a theft, but Lohse, Colossians and Philemon, 204 rightly canvasses other possibilities; if Philemon had had to pay for the performance of Onesimus' tasks in his absence, that too would constitute a debt which the owner would expect to recover from his slave's peculium. For a slave to run away was itself a form of theft.

16 Runaway slaves frequently sought asylum, but it is unlikely that Paul in prison would qualify for this, pace Lohmeyer, E., Die Briefe an die Philipper, an die Kolosser und an Philemon (Meyers; 8th edn.; Göttingen: Vandenhoeck und Ruprecht, 1930) 172Google Scholar. And if Onesimus had met Paul as a fellow prisoner (imprisoned as a runaway), it would be the authorities, not Paul, who sent him back to his master. But given that prisoners like Paul would need the services of friends bringing food etc. to him in prison, one can imagine how Onesimus might be encouraged by Christians in Ephesus to visit Paul. If he regretted his flight, Onesimus might be eager to seek out a potentially sympathetic friend of his master who could act as an intermediary.

17 Theissen, G., The Social Setting of Pauline Christianity (Edinburgh: T. and T. Clark, 1982) 83–7, 92–4Google Scholar. For a broader discussion of the social status of Paul's converts and their house-church setting see Meeks, W., The First Urban Christians (New Haven/London: Yale University, 1983) 5173Google Scholar; Klauck, H.-J., Hausgemeinde und Hauskirche im frühen Christentum (SB 103; Stuttgart: Katholisches Bibelwerk, 1981).Google Scholar

18 According to Tacitus, Annals 14. 42–5, there were as many as 400 in the house of L. Pedanius Secundus, the urban prefect in Rome.

19 In the fourth century A.D. the orator Libanius pleads on behalf of impoverished lecturers who have to live in lodgings, are in debt and have only two or three slaves, Oratio 31.11 (on which see Jones, A. H. M., The Later Roman Empire A.D. 284602 [Oxford: Blackwell, 1964] 851)Google Scholar. In both Greek and Roman worlds it could be considered the mark of utter poverty to own none at all: Lysias 24. 6 (a disabled man claiming a state pension is so poor that he cannot afford to buy a slave to work for him); Xenophon, Memorabilia 2.3.3; Catullus 23–24. See Finley, M. I., ‘Was Greek Civilization Based on Slave Labour?’ in Finley (ed.), Slavery in Classical Antiquity. Views and Controversies (Cambridge: Heffer and Sons, 1960) 5372, esp. 58Google Scholar. Slave-owning was particularly important as a status-indicator, quite apart from its economic value.

20 On the likely size of Gaius' house in the light of the excavations at Corinth see Murphy-O'Connor, J., St. Paul's Corinth (Wilmington: Glazier, 1983) 153–61.Google Scholar

21 Buckland, W. W., The Roman Law of Slavery (Cambridge: Cambridge University, 1908) 1072Google Scholar on the slave as res. Finley rightly notes that ‘the uniqueness of slavery … lay in the fact that the labourer himself was a commodity, not merely his labour or labour-power. His loss of control, furthermore, extended to the infinity of time, to his children and his children's children…’, Ancient Slavery, 74–5. The Delphic manumission inscriptions clarify the four basic elements of slavery: lack of legal rights, liability to seizure, inability to choose one's activities and lack of freedom to determine one's residence; see Westermann, W. L., ‘Slavery and the Elements of Freedom in Ancient Greece’, in Finley, M. I. (ed.), Slavery in Classical Antiquity, 1732.Google Scholar

22 E.g. Vedius Pollio's intention to throw his careless slave to his man-eating lampreys, Seneca, De Ira 3.40 (the slave begs for mercy – i.e. that he might be put to death some other way!).

23 See Vogt, J., Ancient Slavery, 103–21Google Scholar. Many masters recognised that if they avoided excessively harsh treatment of their slaves they would get better and longer service from them; see Philo, , Spec Leg 2. 83, 90–1.Google Scholar

24 Prov 29.19; LXX Ecclus 33. 24–28; Seneca, De Ira 3.24, 32, 35.

25 See the texts cited by Wiedemann, T., Greek and Roman Slavery (London: Croom Helm, 1981) 167–9CrossRefGoogle Scholar and the discussion by Finley, Ancient Slavery, 94–5 and Buckland, Roman Law, 86–97.

26 Juvenal 11. 152–3 notices the homesick slave-boy. The papyri from Egypt indicating the sale of slaves suggest the frequency of familial disruption in this procedure: see Bradley, K. R., Slaves and Masters in the Roman Empire. A Study in Social Control (Bruxelles: Latomus, 1984) 4780.Google Scholar

27 De Beneficiis 3.19.Google Scholar

28 See Petronius, e.g., Satyricon 75.11Google Scholar; Horace, , Satires 1.2.116–19Google Scholar; the elder Seneca, , Controversiae 4 preface 10; Dio Chrysostom 15. 5.Google Scholar

29 Seneca, , Epistle 47. 5Google Scholar; cf. Macrobius, , Saturnalia 1.11. 13Google Scholar. The memories of the (admittedly rare) slave revolts did not fade quickly and in Rome in particular there is evidence for a certain paranoia; in a Senate debate in A.D. 61 Cassius expresses a basic insecurity: conluviem istam non nisi metu coerceris, Tacitus, Annals 14. 42–5.

30 Pliny, Epistle 3.14 notes: ‘no master can feel safe because he is kind and considerate’. On the psychologically devastating insecurity of the slave, see Finley, Ancient Slavery, 74, and on the role of fear in the social control of slaves see Bradley, Slaves and Masters, 113–37.

31 Dio Chrysostom 14.1Google Scholar; cf. Philo, , Spec Leg 2.84Google Scholar; Seneca, De Beneficiis 3.19.Google Scholar

32 E.g. to clear debts, Dio Chrysostom 15. 23; Petronius, , Satyricon 57. 4Google Scholar; cf. 1 Clement 55. 2 (Christians selling themselves into slavery in order to ransom others).

33 Epictetus, Diss 4.1.33.Google Scholar

34 See the discussions by Treggiari, S., Roman Freedmen during the Late Republic (Oxford: Clarendon, 1969) 1120Google Scholar and Duff, , Freedmen in the Early Roman Empire (Oxford: Clarendon, 1928) 1521.Google Scholar

35 Slaves manumitted so they can marry their masters: see ILS 1519,1552 etc. Gaining a reputation for generosity: Dionysius of Halicarnassus, Roman Antiquities 4.24. 6. To provide an incentive for good behaviour: Philo, , Spec Leg 2. 67Google Scholar; cf. Aristotle, Ps., Oikonomikos 1.5. 6Google Scholar; Aristotle, , Politics 1330a 32–3.Google Scholar

36 Dio Chrysostom 15.22Google Scholar; Seneca, , Epistle 80.4Google Scholar; Hopkins, K., Conquerors and Slaves (Cambridge: Cambridge University, 1978) 131Google Scholar: ‘For the masters, manumission was economically rational, partly because it tempted slaves to increase their productivity and lowered the cost to the master of supervising his slaves at work, and partly because the slave's purchase of freedom recapitalised his value and enabled the master to replace an older slave with a younger one.' In the records of 1237 slaves manumitted at Delphi, only two are said to have been given their freedom free of charge (167).

37 The terms for release of a Jewish slave in Ex 21.1–6 presuppose that any children born during the period of slavery will remain in the master's possession. Hopkins, Conquerors and Slaves, 155–8 and 164–8 discusses Greek examples.

38 Hopkins, , Conquerors and Slaves, 117–18, 128–32, illustrates how ‘humanity was complemented by self-interest’ (118).Google Scholar

39 On obsequium and operae see Duff, , Freedmen, 3649Google Scholar; Waldstein, W., Operae Liber-torum. Untersuchungen zur Dienstpflicht freigelassener Sklaven (Stuttgart: F. Steiner, 1986).Google Scholar

40 Hopkins, , Conquerors and Slaves, 133–71.Google Scholar

41 S. Bartchy, Μᾶλλον Χρῆσαι. First-Century Slavery and the Interpretation of 1 Corinthians 7:21(SBLDS 11; Missoula: Society of Biblical Literature) rightly insists that the terms of manumission were entirely under the control of the owner. A master could always revoke a promise of manumission or change its conditions at the last minute (Tacitus, Annals14. 42).

42 See Xenophon, , Memorabilia 2.1. 16Google Scholar (chains) and ILS 8726–8733 (collars); Bellen, H.,Studien zur Sklavenflucht im römischen Kaiserreich (Wiesbaden: F. Steiner, 1971).Google Scholar

43 For ‘wanted’ notices see P. Oxy. 1423 and 1643 and the papyrus cited by Moule, Epistles, 34–7. For the employment of fugitivarii see Finley, Ancient Slavery, 111–12 and W. L. Westermann, The Slave Systems of Greek and Roman Antiquity (Philadelphia: American Philosophical Society, 1955) 77.

44 Under Roman law he would be, but this would probably only apply if Philemon was a Roman citizen, Coleman-Norton, P. R., ‘The Apostle Paul and the Roman Law of Slavery’, in Studies in Roman Economic and Social History (Princeton: Princeton University, 1951) 155–77Google Scholar. On the distinction between Roman law and the various Greek laws practised in the provinces see Crook, J. A., Law and Life of Rome (London: Thames and Hudson, 1967) 2930, 284Google Scholar; and on the impossibility of generalising about Greek law see Finley, M., The Use and Abuse of History (London: Chatto & Windus, 1975) 134–52.Google Scholar

45 See Bellen, , Sklavenflucht, 1731. Branding: PetroniusGoogle Scholar,Satyricon 103.4Google Scholar; Juvenal 14.24. In the third century A.D. a Christian master punished a runaway Christian slave (Callistus) by sending him to a treadmill (Hippolytus, , Refutatio 9.12. 4).Google Scholar

46 Bradley, , Slaves and Masters, 97Google Scholar notes the exceptional generosity of an anonymous Egyptian master in providing for his slaves' freedom although they had run away.

47 Note Petronius, , Satyricon 107.4.Google Scholar

48 This is well noted by Petersen, N. R., Rediscovering Paul. Philemon and the Sociology of Paul's Narrative World (Philadelphia: Fortress, 1985) 99, although it is not necessary to suppose that Philemon might even face excommunication.Google Scholar

49 Petersen, , Rediscovering Paul, 74–8.Google Scholar

50 Knox, Philemon, 24. The argument (19–27) includes an interpretation of παρακαλῶ σε περί in v. 10 as meaning ‘I am asking for’ Onesimus (which he also takes to be implied by v. 20). Cf. Lohmeyer, , Briefe, 188Google Scholar; Gayer, , Stellung, 241–4.Google Scholar

51 Lohmeyer seems to assume that Onesimus will be manumitted, as does Knox (‘the release of a slave’, 27), although the latter also talks in terms of a ‘transfer of ownership’, 24. Dibelius, M., An die Kolosser, Epheser, an Philemon (HNT 12; Tübingen: J. C. B. Mohr [P. Siebeck] 1953) 107Google Scholar and Stuhlmacher, P., Der Brief an Philemon (EKK; Zürich: Benziger; Neukirchen: Neukirchener, 1975) 40–1Google Scholar rightly note that Onesimus could serve Paul either as a slave or as a freedman. Stuhlmacher is unique in discussing the practical implications of these options (54 n. 104).

52 Moule, Epistles, 146 relates αίώνιον to the LXX translation of Ex 21. 6 είς τόν αίῶνα. Contrast e.g. Gnilka, Philemonbrief, 50–1 and J. F. Collange, L'Épître de Saint Paul à Philémon (CNT 2nd series 11c; Geneva: Labor et Fides, 1987) 62–3. This question is made slightly more complex by the problem of the meaning of άπέχω: see Knox, Philemon, 23 and BAG s.v.

53 See Wright, , Colossians and Philemon, 185Google Scholar. Others suggest that vv. 13–14 and v. 15 represent two different options for Philemon and that Paul wants to keep both open, Moule, , Epistles, 21,146–7Google Scholar; Stuhlmacher, , Brief, 40–3Google Scholar; Gnilka, , Philemonbrief, 49.Google Scholar

54 Lightfoot, , Epistles, 343.Google Scholar

55 Vincent, M. R., A Critical and Exegetical Commentary on the Epistles to the Philippians and to Philemon (ICC; Edinburgh: T. and T. Clark, 1897) 188Google Scholar: ‘Ώς denotes the subjective conception of Onesimus’ relation to his master, without reference to the external relation.’ One might compare v. 14 ώς κατά άνάγκην.

56 Pace e.g. Vincent, , Epistles, 189Google Scholar and Lohse, , Colossians and Philemon, 203.Google Scholar

57 Stuhlmacher, , Brief, 42Google Scholar; Gnilka, , Philemonbrief, 52.Google Scholar

58 See Caird, G. B., Paul's Letters from Prison (Oxford: Oxford University, 1976) 222Google Scholar and Bruce, , Epistles, 216.Google Scholar

59 Lohmeyer, , Briefe, 189Google Scholar; Bruce, , Epistles, 217.Google Scholar

60 Wright, , Colossians and Philemon, 185Google Scholar; Lightfoot, , Epistles, 342–3Google Scholar; Scott, , Epistles, 110Google Scholar (Paul does not even imagine that Onesimus should be exempt punishment); Lohse, , Colossians and Philemon, 203Google Scholar; Gnilka, , Philemonbrief, 51.Google Scholar

61 Collange, , L'Épître, 63–4Google Scholar; Stuhlmacher, , Brief, 43 and n. 108.Google Scholar

62 Lightfoot, , Epistles, 345Google Scholar: ‘the idea [of emancipation] would seem to be present to his thoughts, though the word never passes his lips’.

63 Collange, , L'Épître, 72Google Scholar; Gnilka, , Philemonbrief, 88.Google Scholar

64 Dibelius, , An die Kolosser, 107Google Scholar; Lohse, , Colossians and Philemon, 206.Google Scholar

65 Knox, , Philemon, 17.Google Scholar

66 Barton, S., ‘Paul and Philemon: A Correspondence Continued’, Theology 90 (1987) 97101CrossRefGoogle Scholar imagines a puzzled reply by Philemon, who says that it was only on a second reading that he understands Paul to be expecting him to manumit Onesimus. I suspect that on a third reading Philemon might have changed his understanding of the letter again!

67 Collange, , L'Épître, 72.Google Scholar

68 /Lohse, , Colossians and Philemon, 202Google Scholar; cf. Stuhlmacher, , Brief, 41, 43, 57.Google Scholar

69 Schrage, W., Die konkreten Einzelgebote in der paulinischen Paränese (Gütersloh: Gütersloher, 1961).Google Scholar

70 Therapeutae: Philo, , De Vita Contemplativa, 70Google Scholar; Essenes: Philo, , Quod Omnis Probus Liber sit 79Google Scholar; Apologia 11. 4Google Scholar; Josephus, , Antiquities 18. 21Google Scholar (but cf. CD 11.12 and 12.10 on the use and sale of slaves). In fact, it was not at all uncommon to declare slavery contrary to nature: see, e.g., the Sophists countered by Aristotle, , Politics 1253bGoogle Scholar and further texts discussed by Gayer, , Stellung, 2638Google Scholar. It was another matter to renounce the institution in practice!

71 Although Philo considered all humanity equal and naturally free (Spec Leg 2.69, 84), he thought there were thousands of jobs which required slaves (2.123). On utopian dreams of automation and slavelessness see Vogt, , Ancient Slavery, 2638.Google Scholar

72 Caird, , Letters, 216Google Scholar; cf. Gülzow, , Christentum und Sklaverei, 40.Google Scholar

73 See the dismissal of a critical slave at the end of Horace, Satire 2. 7.Google Scholar

74 Those who identify the Onesimus of Philemon with the later bishop of Antioch assume that he must have been given his freedom: see Stuhlmacher, , Brief, 53–4Google Scholar. However, it appears that the ministrae interrogated by Pliny (Epistle 10.96. 8) were ancillae.

75 On the Saturnalia see Pauly-Wissowa, , RE, s.v.Google Scholar

76 Seneca notes that it is commonly thought degrading to eat with one's slaves (Epistle 47. 28)Google Scholar. In Aristotle, Ps., Oikonomikos 1.5.2 masters are warned about giving wine to slaves; it easily makes them insolent.Google Scholar

77 See Theissen, , Social Setting, 145–74.Google Scholar

78 Perhaps Philemon could get the best of both worlds by manumitting Onesimus with a contract binding him to carry on living and working in his house. This would enable Philemon to remain as host of the church, but the personal relations between Philemon and Onesimus in this sort of restricted freedom would be little different from that of master and slave.

79 For the subsequent Jewish interpretation of such passages see Urbach, E. E., ‘The Laws regarding Slavery as a Source for Social History of the Period of the Second Temple, the Mishnah and the Talmud’ in Weiss, J. G. (ed.), Papers of the Institute of Jewish Studies, London (Jerusalem: Magnes, 1964) 194, esp. 26–8.Google Scholar

80 The milder comments could refer to the special situation in which the master has only one slave (so the Hebrew and Syriac versions, but not the Greek).

81 Cf. Seneca, , Epistle 44.1; 95. 52–3Google Scholar; De Beneficiis 3.28.13.Google Scholar

82 Chrysostom, Dio, 1415Google Scholar; Epictetus, , Diss 4.1Google Scholar; Seneca, , Epistle 47.17 etc.Google Scholar

83 For a fuller discussion see Richter, W., ‘Seneca und die Sklaven’, Gymnasium 65 (1958) 196218Google Scholar and Griffin, M. T., Seneca. A Philosopher in Politics (Oxford: Clarendon, 1976) 256–85.Google Scholar

84 Petersen's assertion, Rediscovering Paul, 289Google Scholar that ‘it is logically and socially impossible to relate to one and the same person as both one's inferior and as one's equal' requires some qualification. Key questions are in what respects and in what context one understands the brotherly relationship of equality to apply. If the relationships ‘slave’ and ‘brother’ relate to two entirely different spheres (as for the Stoics) it is possible to be superior in one sphere but equal in another. It is when the two relationships operate in the same sphere (e.g. in everyday behaviour and the personal relationships of the home) that the tension arises.

85 See the full discussion by Bartchy, Μᾶλλον Χρῆσαι, and the important comments by Stuhlmacher, , Brief, 44–5.Google Scholar

86 For a discussion of these factors see Gayer, , Stellung, 212–22Google Scholar. Schulz, Like, Gott, 159–67Google Scholar, Gayer suggests that Paul is contending against an ‘enthusiast’ theology which advocates a more radical policy concerning slaves; but the evidence for such an hypothesis is not strong.

87 Even Aristotle recognised that one could have a friendship with a slave as a human being (Nicomachean Ethics 1161b 5–6) and large tracts of Roman law deal with the slave as a human being (see Buckland, Roman Law). The deep sympathy with slaves shown in Euripides' plays and in Seneca's philosophy highlights the human dimensions of slavery.

88 See e.g. Aristides, , Apology 15. 6Google Scholar; Lactantius, , Institutes 5. 16Google Scholar; Apostolic Constit 4. 12Google Scholar: master and slave are to love each other as fellow Christians, but the distinction in status must be observed.

89 Homily 40 on 1 Corinthians (PG 61, 384–5).Google Scholar

90 See Wallon, , Histoire, 3. 389443.Google Scholar

91 Finley, , Ancient Slavery, 121–2Google Scholar. Cf. Troeltsch, E., The Social Teaching of the Christian Churches (London: G. Allen and Unwin, 1931) 1. 132.Google Scholar

92 Croix, G. E. M. de Ste., ‘Early Christian Attitudes to Property and Slavery’, in Baker, D. (ed.), Studies in Church History 12 (Oxford: Blackwell, 1975) 138Google Scholar, here at 20; cf. his The Class Struggle in the Ancient Greek World (London: Duckworth, 1981) 418–25.Google Scholar

93 See the careful conclusions by Westermann, , Slave Systems, 149–62Google Scholar and the comments on Wallon by Finley, Ancient Slavery, 12–17. For a sober assessment of the role of Christian ideas in the abolition of slavery one has first to acknowledge the role of Christianity in sanctioning Negro and Indian slavery in the West Indies and the Americas and the cogent biblical arguments that could be mounted against abolition: see Swartley, W. M., Slavery, Sabbath, War and Women. Case Studies in Biblical Interpretation (Scottdale, Pa.: Herald, 1983) 3164.Google Scholar

94 In the interests of space I have deliberately refrained from comparing the practice of slavery in other cultures and more recent periods, although if handled with care such comparisons can be illuminating. See e.g. Stampp, K. M., The Peculiar Institution. Slavery in the Ante-Bellum South (New York: Knopf, 1956)Google Scholar; Patterson, O., Slavery and Social Death. A Comparative Study (Cambridge, Mass.: Harvard University, 1982).Google Scholar

I wish to thank Mr J. K. Riches (Glasgow) and Dr S. C. Todd (Keele) for their comments on an earlier draft of this essay.