Empire, monotheism and slavery in the Greater Mediterranean region from antiquity to the early modern era

Past and Present

Volumn 205, Issue 1, 2009, Pages 3-40

  Fynn Paul, Jeffrey ^a  

^a UTRECHT UNIVERSITY   (Netherlands)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 77952798455     PISSN: 00312746     EISSN: None     Source Type: Journal    
DOI: 10.1093/pastj/gtp036     Document Type: Article

References (234)

1. David Brion Davis, 'Looking at Slavery from Broader Perspectives', contribution to the forum 'Crossing Slavery's Boundaries', Amer. Hist. Rev., cv (2000).
2. For the older view, see, for example, Vincent Bakpetu Thompson, The Making of the African Diaspora in the Americas, 1441-1900 (London, 1987).
3. Herbert S. Klein, Slavery in the Americas: A Comparative Study of Virginia and Cuba (Chicago, 1967).
4. For an intermediate phase, which acknowledges the impacts of Islamic and European slavery on Africa, but still sees them as essentially independent, see Paul E. Lovejoy, Transformations in Slavery: A History of Slavery in Africa (Cambridge, 1983), ch. 1.
5. For a more recent synthesis, see Akosua Adoma Perbi, A History of Indigenous Slavery in Ghana from the Fifteenth to the Nineteenth Century (Legon, 2004)
6. See also Paul E. Lovejoy (ed.), Slavery on the Frontiers of Islam (Princeton, 2004), which comparatively addresses the impacts of the Muslim-pagan trade within Africa and the transatlantic slave trade.
7. David Brion Davis, Inhuman Bondage: The Rise and Fall of Slavery in the New World (Oxford, 2008)
8. Enrico Dal Lago and Constantina Katsari (eds.), Slave Systems: Ancient and Modern (Cambridge, 2008).
9. The latter volume, to my mind, still shows a tendency for writers on slavery to move too quickly from the ancient to the early modern period, without exploring many medieval continuities; this is in part due to the fact that the bulk of these continuities lie south and east of western Europe, but such a situation should increasingly prove to be less of a hindrance to exploration and synthesis
10. By 'Greater Mediterranean' is meant all of Asia west of the Indus, most of Africa, and all of Europe. It will be shown that these were all drawn into the civilization nexus which evolved around the Mediterranean (and the Fertile Crescent, which can be viewed as peri-Mediterranean) well before the early modern era. The Atlantic world, and particularly its slave system, should in this view be seen as an extension of this Greater Mediterranean slave system.
11. Michael McCormick's The Origins of the European Economy: Communications and Commerce, AD 300-900 (Cambridge, 2001) provides a true 'missing link' between slavery in the classical and high medieval periods, and also opens up a new understanding of the relationship between Islamic and Christian slaving activity during the early medieval period
12. The classic discussion on the varieties of indigenous slavery in Africa remains Suzanne Miers and Igor Kopytoff (eds.), Slavery in Africa: Historical and Anthropological Perspectives (Madison, 1977), esp. 12-14.
13. See also the newly re-edited study by Humphrey J. Fisher, Slavery in the History of Muslim Black Africa (New York, 2001), for an extensive discussion, albeit one which concentrates on the nineteenth century.
14. An excellent synthesis on the origins of racism against blacks can now be found in Davis, Inhuman Bondage, ch. 2.
15. See also David M. Goldenberg, The Curse of Ham: Race and Slavery in Early Judaism, Christianity, and Islam (Princeton, 2003)
16. Benjamin Braude, 'Cham et Noé: race, esclavage et exégèse entre islam, judaïsme et christianisme', Annales: histoire, sciences sociales, lvii (2002)
17. For one of the first attempts to trace the origins of racial prejudice back into Graeco-Roman society, see Benjamin Isaac, The Invention of Racism in Classical Antiquity (Princeton, 2004)
18. Michael Bush, in the introduction to M. L. Bush (ed.), Serfdom and Slavery: Studies in Legal Bondage (London, 1996), somewhat misleadingly implies that slave systems have only been evolving over the past 2, 500 years.
19. In fact, of course, slave systems existed from the earliest times of Sumer and Egypt. For an ageing but still valuable list of studies on ancient slavery in all these areas, see Bernard Lewis, Race and Slavery in the Middle East: An Historical Enquiry (Oxford, 1990), 3 n. 2.
20. But on the many ways to acquire slaves in traditional Africa, see Suzanne Miers and Igor Kopytoff, 'African "Slavery" as an Institution of Marginality', in Miers and Kopytoff (eds.), Slavery in Africa, 12-14.
21. For this phenomenon in general, see Orlando Patterson, Slavery and Social Death: A Comparative Study (Cambridge, Mass., 1982), esp. pt 2.
22. On the evolution of debt slavery in Babylonia, see Muhammad A. Dandamaev, Slavery in Babylonia: From Nabopolassar to Alexander the Great (626-331 BC) (DeKalb, 1984), 157-80.
23. For a brief mention of debt slavery amongst the ancient Celts, see H. D. Rankin, Celts and the Classical World (Portland, Oreg., 1987), 132.
24. For the poor submitting themselves to slavery in fifth-century Gaul, see R. Samson, 'Slavery, the Roman Legacy', in John Drinkwater and Hugh Elton (eds.), Fifth-Century Gaul: A Crisis of Identity? (Cambridge, 1992), 223.
25. For debt slavery and voluntary submission to slavery in high medieval Dalmatia, see Neven Budak, 'Slavery in Late Medieval Dalmatia/Croatia: Labour, Legal Status, Integration', Mélanges de l'École française de Rome: Moyen Âge, cxii (2000), 748
26. For economic and judicial slavery in Viking Scandinavia, see Eljas Orrman, 'Rural Conditions', in Knut Helle (ed.), The Cambridge History of Scandinavia, i, Prehistory to 1520 (Cambridge, 2003), 308.
27. On Tatar parents selling their children in the fourteenth century, see Peter Spufford, Power and Profit: The Merchant in Medieval Europe (New York, 2002), 340
28. Orlando Patterson discourses on these paths to slavery in Slavery and Social Death, esp. 105
29. There are of course many other variations. In some African cultures, certain children were considered supernaturally dangerous and would be left to die by their parents; these children, as well as naturally orphaned infants, would occasionally be brought up as slaves by those who found them: Miers and Kopytoff, 'African "Slavery" as an Institution of Marginality', 13.
30. Duby notes that Carolingian society utilized a similar combination of means for maintaining slave supplies: Georges Duby, The Early Growth of the European Economy: Warriors and Peasants from the Seventh to the Twelfth Century (Ithaca, 1974), 32
31. For a decisive argument against the assumption that Roman society relied on warfare as its primary means of replenishing slave supplies, see Keith Bradley, Slavery and Society at Rome (Cambridge, 1994), 31-56
32. It is now accepted that the Islamic Empire also used purchase as a primary means of gaining slaves, especially after the initial conquest was completed. Slaves were bought from agents who travelled throughout central Asia for the purpose of acquiring them from local chiefs and raiders: Daniel Pipes, Slave Soldiers and Islam: The Genesis of a Military System (New Haven, 1981), 146-8.
33. On a slave trade amongst Celtic tribes (and perhaps between these tribes and Rome) in pre-Roman Britain and Gaul, see Rankin, Celts and the Classical World, 132
34. On Gaulish slaves being traded for Roman wine, see Bradley, Slavery and Society at Rome, 36
35. For slave market mechanisms in the Roman Empire, see Bradley, Slavery and Society at Rome, 41-3
36. for the early Islamic Empire, see Pipes, Slave Soldiers and Islam, 142-8
37. For similar mechanisms in the early modern era, see William D. Phillips Jr, Slavery from Roman Times to the Early Transatlantic Trade (Minneapolis, 1985), 141.
38. For the overlap of slave and other trading networks in West Africa, see Philip Curtin, 'Africa North of the Forest in the Early Islamic Age', in Philip Curtin et al. (eds.), African History: From Earliest Times to Independence (London, 1995), 94.
39. For the overlap of furand slave-trading networks in medieval Russia, see H. R. Ellis Davidson, The Viking Road to Byzantium (London, 1976), 99.
40. This argument is made by Dmitrij Mishin in 'The Sq aqāliba Slaves in the Aghlabid State', Annual of Medieval Studies at the CEU, 1996-1997 (1998), 237
41. Early evidence for the price differential has now been presented for the eighth and ninth centuries in McCormick, Origins of the European Economy, esp. 754-9.
42. Hadud al-Alam, quoted in Claude Meillassoux, The Anthropology of Slavery: The Womb of Iron and Gold, trans. Alide Dasnois (Chicago, 1991), 44.
43. Where these slaves may have been headed is an interesting question
44. For raiders preying on children in ninth-century Sicily, see McCormick, Origins of the European Economy, 246
45. Olaudah Equiano was famously captured along with his sister despite the fact that their parents had attempted to barricade them within the family's walled compound as a precaution against slave-taking kidnappers: Olaudah Equiano, The Interesting Narrative of the Life of Olaudah Equiano (1789; New York, 2001), 32
46. For an argument regarding the impact of slaving on political centralization in early modern Africa, see Philip Burnham, 'Raiders and Traders in Adamawa: Slavery as a Regional System', in James L. Watson (ed.), Asian and African Systems of Slavery (Oxford, 1980), 59-64.
47. First-hand accounts of the devastation caused by imperial demand for slaves on slaving zones are comparatively rare, owing, amongst other things, to the fact that slaving zones have usually been pre-literate. Anglo-Saxon descriptions of Viking raids might count as slaving-zone destruction narratives, since it would seem as though Viking raids were in part motivated by Byzantine and Islamic demand for slaves.
48. On devastation caused in parts of Africa by slave demand during the later nineteenth century, see Ronald Segal, Islam's Black Slaves: A History of Africa's Other Black Diaspora (New York, 2001), 157
49. For a sense of the insecurity experienced by all slaving-zone societies, see Equiano, Interesting Narrative, 32-8
50. Bradley, Slavery and Society at Rome, 42-3
51. Note that the situation within the Dar al-Islam was somewhat complicated in so far as large numbers of Christians and Jews, who were usually protected as 'people of the book', were still liable to judicial slavery. The same went for Muslims living under Christian rule in the Middle Ages.
52. For an analysis of the manifold meanings and uses of slavery in a multi-religious setting, see Mark Meyerson, 'Slavery and the Social Order: Mudejars and Christians in the Kingdom of Valencia', Medieval Encounters, i (1995)
53. its companion piece, 'Slavery and Solidarity: Mudejars and Foreign Muslim Captives in the Kingdom of Valencia', Medieval Encounters, ii (1996)
54. Peter Garnsey, Ideas of Slavery from Aristotle to Augustine (Cambridge, 1996), 37-8, 108-10
55. But see Benjamin Isaac, 'Proto-Racism in Graeco-Roman Antiquity', World Archaeology, xxxviii (2006)
56. Benjamin Isaac, The Invention of Racism in Classical Antiquity (Princeton, 2004)
57. Garnsey argues that, contrary to what is often thought, the Stoics did not reject the idea of natural slavery: Garnsey, Ideas of Slavery from Aristotle to Augustine, 138-51
58. There is still debate on how comfortable emergent Christianity was with the idea of slavery
59. e s.)', Cahiers de civilisation médiévale, xxviii (1985).
60. Also crucial is Peter Brown, The Body and Society: Men, Women, and Sexual Renunciation in Early Christianity (New York, 1988), esp. 44-57, whose discussion of ideas of the body in the early Church touches on many aspects of slavery.
61. Jennifer Glancy reminds us that Christianity was born into a world where slavery was entirely normative: Jennifer A. Glancy, Slavery in Early Christianity (Oxford, 2002), esp. 3, 156.
62. Others have argued that the Christian ethos and slavery were more immediately at odds. See, for example, Rosemary Morris, 'Emancipation in Byzantium', in Bush (ed.), Serfdom and Slavery, 130-1.
63. The principal exception seems to be Gregory of Nyssa: Garnsey, Ideas of Slavery from Aristotle to Augustine, 80-5
64. Despite the knocking that many of Bonnassie's points have taken since the publication of From Slavery to Feudalism, there is still little reason to doubt his conclusions on the broad chronology of medieval slavery: Pierre Bonnassie, From Slavery to Feudalism in South-Western Europe (Cambridge, 1991), 51-6
65. Even William Chester Jordan, who is highly critical of the book, does not appear to argue with Bonnassie's chronology: see his review of Bonnassie in Slavery and Abolition, xiii (1992), 97-102
66. For the consensus that the transition to serfdom in the Frankish heartlands had occurred by the ninth century, see Alice Rio, 'Freedom and Unfreedom in Early Medieval Francia: The Evidence of the Legal Formulae', Past and Present, no. 193 (Nov. 2006), 7-11.
67. Much of the new consensus relies on Wendy Davies, 'On Servile Status in the Early Middle Ages', in Bush (ed.), Serfdom and Slavery.
68. Charles Verlinden long ago pointed out that slavery did persist during the otherwise slaveless period c.1000-1363, in the form of relatively small numbers of Muslim captives held in regions of Italy, southern France and eastern Spain: Charles Verlinden, L'Esclavage dans l'Europe médiévale, i, Péninsule Ibérique, France (Bruges, 1955)
69. ii, Italie, colonies italiennes du Levant, Levant latin, empire byzantin (Ghent, 1977)
70. This phenomenon has been examined for Barcelona in Stephen Bensch, 'From Prizes of War to Domestic Merchandise: The Changing Face of Slavery in Catalonia and Aragon, 1000-1300', Viator, xxv (1994)
71. Susan Mosher Stuard has also found evidence of continuing domestic slavery throughout this period: Susan Mosher Stuard, 'Ancillary Evidence for the Decline of Medieval Slavery', Past and Present, no. 149 (Nov. 1995), 4-5
72. see a reply by Jean-Pierre Devroey, 'Men and Women in Early Medieval Serfdom: The Ninth-Century North Frankish Evidence', Past and Present, no. 166 (Feb. 2000).
73. The institution of slavery persisted in England into the early twelfth century, where ecclesiastical institutions were some of the last holdouts: see David A. E. Pelteret, Slavery in Early Mediaeval England: From the Reign of Alfred until the Twelfth Century (Rochester, NY, 1995), 251-6.
74. In Sweden, where the process of Christianization and abolition happened later and can be traced more easily in written sources, farm slavery disappeared first, but household slavery lingered until all slavery was banned by royal decree in 1335
75. See Joan Lind, 'The Ending of Slavery in Sweden', Scandinavian Studies, l (1978)
76. Although slavery persisted in many parts of Europe up to about the year 1100, it is thought that the great Carolingian estates underwent a process of transformation from slavery to serfdom beginning perhaps as early as the third century, and which was certainly complete by the tenth century
77. e siècles), 2 vols. (Paris, 2003)
78. Chris Wickham, Framing the Early Middle Ages: Europe and the Mediterranean, 400-800 (Oxford, 2005)
79. Bonnassie was one of the first to see Christianization as a significant force in the decline of slavery in western Europe, and his view is still arguably not the majority opinion: Bonnassie, From Slavery to Feudalism, 37 ff
80. Devroey refines the argument when he writes: 'Without fundamentally challenging the institution of slavery, Christianity made a significant contribution to an increase in the subjective rights of the unfree, by restoring to slaves their human condition, and by recognizing in canon law their capacity to contract a legitimate marriage': Devroey, 'Men and Women in Early Medieval Serfdom', 8
81. Lind argues that the coming of Christianity led to an almost immediate cessation of slavery in Denmark and Norway, while being directly responsible for a more gradual erosion in Sweden: 'Ending of Slavery in Sweden', 60
82. Although Bonnassie followed Marxist paradigms in much of his work, it is thought that his consideration of the influence of Christianity in the ending of early medieval slavery was distinctly post-Marxist
83. See Rio, 'Freedom and Unfreedom in Early Medieval Francia', 9
84. McCormick makes this point in Origins of the European Economy, 734
85. See also Ross Samson, 'The End of Early Medieval Slavery', in Allen J. Frantzen and Douglas Moffat (eds.), The Work of Work: Servitude, Slavery, and Labor in Medieval England (Glasgow, 1994), 96.
86. Even the less Marxist Duby, for example, maintains that the Church dealt slavery 'barely a glancing blow': Duby, Early Growth of the European Economy, 32
87. McCormick, Origins of the European Economy, 748-52
88. This is especially true of enslavement of Latin Christians by Latin Christians; the early medieval incidence of Byzantines trading in Latins and vice versa has not been studied enough to allow any meaningful generalizations. Warren Treadgold suggests in passing that by 1200 most Byzantine slaves were pagans from the non-Christianized parts of Russia: Warren Treadgold, A Concise History of Byzantium (New York, 2001), 188.
89. The argument that points to slave-holding churchmen as proof that 'the Church' was either pro-slavery or effectively silent on the issue of slavery is, while still widely held, a gross oversimplification. For a cautious treatment which allows for a great deal of church influence in the decline of slavery, see Marc Bloch, Slavery and Serfdom in the Middle Ages, trans. William R. Beer (Berkeley, 1975), 9-17.
90. For a more recent dismissal of ecclesiastical influence on abolition, see Samson, 'Slavery', 223
91. Bjørn Myhre, 'The Iron Age', in Helle (ed.), Cambridge History of Scandinavia, i, 73.
92. Ammianus found the Alans remarkable for the fact that they did not practise slavery: Bradley, Slavery and Society at Rome, 19-20
93. Procopius records that the Slavs, for their part, were occasionally carrying large numbers of Byzantines into slavery by the sixth century: P. M. Barford, The Early Slavs (Ithaca, 2001), 58-9.
94. The Franks were raiding and probably trading for slaves by the early seventh century: McCormick, Origins of the European Economy, 739
95. Ellis Davidson, Viking Road to Byzantium, 99 ff., provides a good discussion of pagan Scandinavian slave raids, referencing Adam of Bremen's Gesta Hammaburgensis and the Vita Rimberti.
96. See also the older study by A. Mez, The Renaissance of Islam (London, 1937), 157 ff.
97. McCormick, Origins of the European Economy, 746-7
98. The power of the Church's teaching on marriage and sexuality should not be underestimated as a mechanism for reining in many of the excesses that traditionally accompany slave ownership. In the tenth century, for example, the king of the Bulgars could claim one in ten slaves brought to his country as a concubine. The Christianization of Bulgaria was accompanied by the demise of this system.
99. Ellis Davidson, Viking Road to Byzantium, 66
100. Bonnassie argues that Christianity slowly shut down the slave system, but, oddly, he writes about the various means of enslavement of Christians by Christians as though these processes did not diminish with increasing Christianization: Bonnassie, From Slavery to Feudalism, 35-6
101. Other evidence, however, suggests the pattern which probably occurred in most of Europe, that is, a gradual elimination of most of the traditional means of replenishing the slave supply
102. For this process in early medieval France, see Duby, Early Growth of the European Economy, 32-3, and in high medieval Scandinavia
103. Lind, 'Ending of Slavery in Sweden', 68
104. Lewis shows that a similar process happened under Islam: Lewis, Race and Slavery in the Middle East, 6-11
105. As has been stated, in frontier societies such as Spain, however, Jews and Muslims living under Christian rulers could be reduced to slavery through judicial means, even though enslavement of Christians by Christians was now prohibited
106. See Meyerson, 'Slavery and the Social Order'
107. In Dalmatia, a perfect no-slaving zone was not constituted until the early fourteenth century; this was then undermined by the post-Black Death resumption of a European slave-owning culture, when Dalmatia was partially incorporated into the Italians' slaving zone: Budak, 'Slavery in Late Medieval Dalmatia/Croatia', 752-7
108. Duby, Early Growth of the European Economy, 32-3
109. See also Devroey, 'Men and Women in Early Medieval Serfdom', 8
110. For the conflict between protecting church property and the pious ideal of manumission in the Iberian Visigothic Church, see Carles Buenacasa Pérez, 'Un exemple de la caritat cristiana a l'església primitiva: la manumisió dels esclaus a Hispània segons les fonts dels segles IV-VII', Annals de l'Institut d'Estudis Gironins, xxxviii (1996-7)
111. For a recent discussion of manumission in early medieval Europe, see Wickham, Framing the Early Middle Ages, 564-5
112. Mohamed Talbi suggests that the decrees of Gregory VII, and his pressure on the Venetian doges, played a significant role in the cessation of the Latin slave trade with the Islamic Mediterranean
113. See Mohamed Talbi, 'Law and Economy in Ifrīqiya (Tunisia) in the Third Islamic Century', in A. L. Udovitch (ed.), The Islamic Middle East, 700-1900: Studies in Economic and Social History (Princeton, 1981), 214.
114. Stephen Bensch describes two instances of high and late medieval popes rebuking Catalan kings and clergy over issues of slavery. In 1206, Innocent III rebuked the clergy of Barcelona for turning a blind eye while the city's burghers refused baptism to their Saracen slaves. By this time the Church taught that sincere conversion placed a strong moral obligation on owners to manumit their slaves, at least at the time of an owner's decease: Bensch, 'From Prizes of War to Domestic Merchandise', 83.
115. In another instance, in 1337 Pope Benedict XII rebuked King Pere III for the impropriety of having Saracen slaves serve at table
116. Bonnassie, From Slavery to Feudalism, 54
117. Morris, 'Emancipation in Byzantium', 133-4
118. As Charles Brand has pointed out, Justinian upheld the classical notion that slaves, as things, were not permitted to marry. But he also turned this law on its head, decreeing that anyone who had received the Christian sacrament of marriage was now legally free. This law was upheld until the time of Alexius I, who decreed that slaves could marry without altering their status. It seems as though Justinian's law had generally had the effect of denying marriage to slaves, while Alexius felt that all Christians, of whatever status, should be eligible for the matrimonial state.
119. Charles Brand, 'Slave Women in the Legislation of Alexius I', Byzantinische Forschungen: internationale Zeitschrift für Byzantinistik, xxiii (1996), 21
120. The argument presented here suggests that Islamic societies might have enforced the taboo against enslaving co-religionists before this became a common ideal within Christian lands
121. Patricia Crone has made the most influential case for the advent of the slave armies in particular, arguing that the system was implemented by al-Ma'mun (813-33) and greatly expanded by his successor al-Mu'tasim (833-42). It was soon widely imitated.
122. See Patricia Crone, Slaves on Horses: The Evolution of the Islamic Polity (Cambridge, 1980), 74-81
123. Matthew S. Gordon, in The Breaking of a Thousand Swords (Albany, 2001), narrativizes most of what is known about the importation of the first Turkish soldiers to Samarra and Baghdad under the caliphs al-Ma'mun and al-Mu'tasim.
124. Daniel Pipes suggested that Muslim citizens began to turn away from civil institutions for religious reasons, necessitating the recruitment of slaves: Pipes, Slave Soldiers and Islam, 70-5
125. Domestic slavery, and the concomitant restriction of women's freedoms, seems to have become fashionable from the time of al-Rashid (786-809): Albert Hourani, A History of the Arab Peoples (Cambridge, Mass., 1991), 32-7.
126. As is well known, the notion of 'slavery' in Islamic societies encompassed people of all social classes; provincial governors might technically be slaves or freedmen, while many slaves were agricultural drudges. Women slaves could range from politically powerful courtiers to the poorest maids or prostitutes.
127. For an overview of the types and conditions of slavery in Islamic countries, see Murray Gordon, Slavery in the Arab World (New York, 1989), 48-78
128. For the unusual case of the Zanj, black land-clearing slaves who revolted and successfully waged a guerrilla war in ninth-century Iraq, see Alexandre Popovic, The Revolt of African Slaves in Iraq in the 3rd/9th Century, trans. Leon King (Princeton, 1999).
129. For the large-scale presence of agricultural slaves in Iraq and North Africa, and an indication that the historiography concentrates perhaps too much on the exclusively domestic nature of Islamic slavery
130. see Michael G. Morony, 'Landholding in Seventh-Century Iraq: Late Sasanian and Early Islamic Patterns', in Udovitch (ed.), Islamic Middle East, 165
131. Talbi, 'Law and Economy in Ifrīqiya (Tunisia)', 221
132. See map in McCormick, Origins of the European Economy, 762
133. Mishin, 'Sq aqāliba Slaves', 237-8
134. Note that McCormick finds a pattern of slave imports to the Caliphate arising prior to the reign of al-Ma'mun (813-33)
135. The Venetian slave trade is summarized in Michael McCormick, 'New Light on the "Dark Ages": How the Slave Trade Fuelled the Carolingian Economy', Past and Present, no. 177 (Nov. 2002), 46-51.
136. See also McCormick, Origins of the European Economy, 759-77
137. McCormick, Origins of the European Economy, 748-52
138. Medieval Trade in the Mediterranean World: Illustrative Documents, ed. Robert S. Lopez and Irving W. Raymond (New York, 2001), 115.
139. Bonnassie, From Slavery to Feudalism, 48-51
140. McCormick was one of the first to assert the central importance of the slave trade in eighth-and ninth-century Europe, and to show its main directions. The suggestion that the Viking raids were spurred by slave demand from the Caliphate has not, to my knowledge, been made openly, although many authors hint at this conclusion.
141. That the Vikings were active slave takers and slave traders is wholly established: see Orrman, 'Rural Conditions', 308
142. H. R. Ellis Davidson long ago acknowledged that slaves were a central part of Viking trading with Constantinople: Ellis Davidson, Viking Road to Byzantium, 99-100.
143. Peter Sawyer notes that Baltic slaves were being exported to the Caliphate from about 790. He also suggests that the increasing sophistication of the Russian state probably led to the decrease in Viking slaving activity from the eleventh century: Peter Sawyer, 'The Viking Expansion', in Helle (ed.), Cambridge History of Scandinavia, i, 115-17.
144. Thomas Noonan writes: 'Between ca. 800 and 1020, millions of Islamic silver coins... were imported into European Russia from the Islamic world. A significant part of these dirhams were then re-exported from Russia to the lands around the Baltic Sea and especially Sweden... the overwhelming majority were the result of an active trade involving the Vikings and Russia': Thomas Noonan, 'The Vikings and Russia', in Ross Samson (ed.), Social Approaches to Viking Studies (Glasgow, 1991), 202.
145. For Noonan's explanation of why the Viking-Islamic trade began in the eighth century, see Thomas S. Noonan, The Islamic World, Russia and the Vikings, 750-900: The Numismatic Evidence (Aldershot, 1998), esp. ch. 2.
146. Note that the period of peak trade, which according to Noonan began about 840, corresponds with al-Mu'tasim's expansion of al-Ma'mun's mamlūk (slave soldier) programme, and its subsequent adoption by most Islamic rulers
147. See Crone, Slaves on Horses, 76-85
148. See also Michael Cooperson, Al Ma'mun (Oxford, 2005), 107-9
149. McCormick, Origins of the European Economy, 245-8
150. Talbi, 'Law and Economy in Ifrīqiya (Tunisia)', 224
151. This tendency is already noted by McCormick for the ninth century: McCormick, Origins of the European Economy, 769
152. See also Mishin, 'Sq aqāliba Slaves', 237
153. For a study on the redemption of Spanish Christian captives taken by Muslims in the later Middle Ages, see AndrésDíaz Borrás, El miedo al Mediterráneo: la caridad popular valenciana y la redención de cautivos bajo poder musulmán, 1323-1539 (Barcelona, 2001)
154. Talbi notes a paucity of slaves entering Tunisia from the tenth century onward, in contrast with a superabundance of slave imports during the ninth century: Talbi, 'Law and Economy in Ifrīqiya (Tunisia)', 214-27
155. The split between Latin and Orthodox Christendom led to an increased danger of enslavement of one type of Christian by another, but the danger to a Christian of enslavement by a Christian of the opposed Church was generally less than the danger of enslavement by Muslims or pagans, since diplomatic recourse could often be had, at least for wealthy or influential captives
156. For a ninth-century pope protesting against the sale by Byzantines of Latin Christians they had supposedly 'liberated', see McCormick, Origins of the European Economy, 745
157. On 'Greek' slaves in fourteenth-and fifteenth-century Spain, see Verlinden, L'Esclavage dans l'Europe médiévale, i, 321-30
158. Marc Bloch long ago noted the receding frontier with paganism as a possible source for the decline of western European slavery: Bloch, Slavery and Serfdom in the Middle Ages, 28
159. However, he and those following him have greatly preferred economic explanations: Bonnassie, From Slavery to Feudalism, 37-8
160. Bradley argues that manumissions were not excessive under the empire, and that the practice of manumission was employed primarily as a means of control. He also notes the continuing collection of the manumission tax in the fourth century as an indication that there were still substantial numbers of slaves at this time: K. R. Bradley, Slaves and Masters in the Roman Empire: A Study in Social Control (Brussels, 1984), 81-112.
161. For Jewish slave-owning and manumission under the Romans, see E. Leigh Gibson, The Jewish Manumission Inscriptions of the Bosporus Kingdom (Tübingen, 1999), esp. 56-94.
162. Again, for a recent discussion of manumission in early medieval Europe, see Wickham, Framing the Early Middle Ages, 564-5
163. For gradations of unfreedom in the Maliki school of Islamic thought (which was prevalent in North-West Africa and Spain), see Cristina de la Puente, 'Entre la esclavitud y la libertad: consecuencias legales de la manumisión según el derecho mālikí', Al-Qantq ara, xxi (2000)
164. For this view of auto-reproduction in the New World slave regions, see Herbert S. Klein, The Atlantic Slave Trade (Cambridge, 1999), 166-8.
165. For pagan Rome, see Bradley, Slavery and Society at Rome, 31-56
166. Bradley maintains that, generally speaking, ancient slave systems relied on external sources of supply as a significant maintenance mechanism
167. For the purposes of the present argument it does not much matter whether early Islamic slave-holding patterns were continuous with or fundamentally different from Sassanid or Byzantine patterns, since we are primarily concerned with the nature of slave supply and demand itself, and secondarily concerned to show that demand for slaves from non-Islamic areas probably increased from the late eighth and early ninth centuries. As noted above, scholars know that there was some continuity in agricultural slave-holding between pre-Islamic and Islamic periods, probably more than is usually acknowledged.
168. See Morony, 'Landholding in Seventh-Century Iraq', 165
169. Mohamed Talbi quite unequivocally states that slaves played a major role in Tunisian agriculture in the ninth century: Talbi, 'Law and Economy in Ifrīqiya (Tunisia)', 215
170. Certainly there is precedent for the institution of harem slavery in Sassanid sources, even though it did not become a dominant social institution until the time of al-Rashid; and, anyway, the institution of the mamlūk from the ninth century constitutes an entirely novel system which can only have increased the slave demand of Islamic societies. The high prices commanded for slaves from the early ninth century, as recorded by McCormick, are in any event evidence of a high demand within the Caliphate.
171. On the increasing importance of the harem and household slaves from early Abbasid times, see Hugh Kennedy, The Prophet and the Age of the Caliphates: The Islamic Near East from the Sixth to the Eleventh Century (London, 1986), 138-9
172. On increased numbers of slaves, see Claude Cahen, 'Tribes, Cities and Social Organization', in R. N. Frye (ed.), The Cambridge History of Iran, iv, The Period from the Arab Invasion to the Saljuqs (Cambridge, 1975), 328.
173. See Wickham, Framing the Early Middle Ages, 562-3
174. In Chris Wickham, 'The Other Transition: From the Ancient World to Feudalism', Past and Present, no. 103 (May 1984), he in fact argues that the bulk of this transformation occurred in the third century, prior to Christianization.
175. Just how many slaves remained in various parts of Europe in each century under scrutiny is still a thorny issue, in part because the vocabulary employed in the documents is often loosely applied
176. For the debate between A. H. M. Jones and others on the number of slaves in the late empire, see Wally Seccombe, A Millennium of Family Change: Feudalism to Capitalism in Northwestern Europe (London, 1992), 45-6.
177. For a good synopsis of the state of the early medieval slavery debate, see Rio, 'Freedom and Unfreedom in Early Medieval Francia', 8-12
178. See also the conclusions of Devroey in 'Men and Women in Early Medieval Serfdom'
179. For the abundance of slave markets in early medieval Europe, see McCormick, Origins of the European Economy, 748
180. For the role that Christianity might have played in suppressing these markets, see Devroey, 'Men and Women in Early Medieval Serfdom', 7-8, 15-19
181. See table 25.1 and its explanation in McCormick, Origins of the European Economy, 756 (reprinted and described in his 'New Light on the "Dark Ages"', 44).
182. Note also that the presumed impact of Islamic demand on Byzantium suggests a whole other avenue of research, even though McCormick's data suggests that the price differential between Byzantium and the Caliphate was virtually non-existent by the tenth century at the latest
183. Rio, 'Freedom and Unfreedom in Early Medieval Francia', 8
184. (But note, again, that the chronology of slavery differed in the various regions of Europe. Bonnassie, for example, found slavery in Catalonia up to the end of the tenth century: Bonnassie, From Slavery to Feudalism, 51-6.)
185. From the early modern period up to the nineteenth century, African slaves were sometimes marched over a thousand miles to reach coastal slave ports. Some slavers seem to have cared little if as many as four-fifths of their captives died on the way; the profits obtained at port made even losses of this magnitude economically sustainable. See below at n. 98.
186. Bensch notes that by 1300, Mediterranean markets had become far more integrated, and slave prices were comparable and relatively stable throughout the Mediterranean: Bensch, 'From Prizes of War to Domestic Merchandise', 78
187. Peter Spufford writes similarly: Spufford, Power and Profit, 338
188. Some slaves were, however, still kept in various parts of Europe throughout the high Middle Ages. See the sources cited in n. 25 above.
189. Verlinden, L'Esclavage dans l'Europe médiévale, i
190. Neven Budak provides a good chronology of the disappearance of early medieval slavery in his study of Dalmatia. Slave sales became quite rare in Dalmatia by the twelfth century, and slavery was almost entirely replaced by indentured servitude during the period of superabundant labour at the beginning of the fourteenth century. A late medieval revival of slavery, in both Dalmatia and Italy, began only in 1363, as a direct result of the second wave of plague. Split made its first late medieval laws restricting the export of slaves in 1375, after the third wave of plague.
191. See Budak, 'Slavery in Late Medieval Dalmatia/Croatia', 750-7
192. For Italy, see Iris Origo, 'The Domestic Enemy', Speculum, xxx (1955), 324-5
193. On the continual possibility that Saracen slaves would be ransomed, see Bensch, 'From Prizes of War to Domestic Merchandise', 72-4
194. Bensch notes that on the Catalan-Islamic frontier, ransom procedures had been regularized by the mid twelfth century at the latest. By the thirteenth century, the ransom of Christian captives, often using captured Saracens as a medium of exchange, was the legally sanctioned business of the Mercedarian Order
195. See Richard C. Hoffman, 'Outsiders by Birth and Blood: Racist Ideologies and Realities around the Periphery of Medieval Culture', Studies in Medieval and Renaissance Hist., vi (1983), 14-20.
196. This is also cited by Stuard in 'Ancillary Evidence for the Decline in Medieval Slavery', 15 n. 44, who observes the co-ordination between European frontiers and lingering slavery.
197. Phillips, Slavery from Roman Times to the Early Transatlantic Trade, 104
198. Bensch notes that, prior to the mid fourteenth century, Russian and Tatar slaves cannot be located in Catalonia with certainty: Bensch, 'From Prizes of War to Domestic Merchandise', 81
199. Phillips, Slavery from Roman Times to the Early Transatlantic Trade, 104-6
200. On the fourteenth-century figures, see Barbara F. Harvey, 'Introduction: The "Crisis" of the Early Fourteenth Century', in Bruce M. S. Campbell (ed.), Before the Black Death: Studies in the 'Crisis' of the Early Fourteenth Century (Manchester, 1991).
201. For the idea that a true labour shortage did not start until the plague of 1362-3, see A. R. Bridbury, 'The Black Death', Econ. Hist. Rev., 2nd ser., xxvi (1973).
202. For evidence of the same trend in Catalonia, see Jeffrey Fynn-Paul, 'Tartars in Spain: Renaissance Slavery in the Catalan City of Manresa, c.1408', Jl Medieval Hist., xxxiv (2008).
203. See also J. Fynn-Paul, 'The Catalan City of Manresa in the Fourteenth and Fifteenth Centuries: A Political, Social, and Economic History' (Univ. of Toronto Ph.D. thesis, 2005), 72-3.
204. Phillips, Slavery from Roman Times to the Early Transatlantic Trade, 105
205. Origo, 'Domestic Enemy', 324-5
206. For interest rates on interpersonal loans in Catalonia, see Fynn-Paul, 'Catalan City of Manresa', 138-9
207. For a threefold decrease on state interest rates in Florence and Catalonia
208. For servants' demands for high wages, in return for little service, during the course of the Black Death, see Giovanni Boccaccio, Decameron, trans. G. H. McWilliam (Harmondsworth, 1972), 54.
209. For a discussion of the 1351 English Statute of Labourers, which suggests that the lower classes had stepped beyond their former social and economic bounds, see Joseph P. Byrne, Daily Life during the Black Death (Westport, 2006), 251.
210. Spufford, Power and Profit, 338
211. Phillips, in Slavery from Roman Times to the Early Transatlantic Trade, ch. 5, does not address any northern European slavery in the later Middle Ages.
212. For the English diarist John Evelyn's too-brief description of a seventeenth-century slave market in Livorno, complete with 'Turks, Mores, and other Nations' (albeit we are left guessing as to who is selling whom), see The Diary of John Evelyn, ed. E. S. de Beer (London, 2006), 99.
213. The end of 'Renaissance slavery' in southern Europe is a topic that stands very much in need of further research. For the Venetian figures, see Phillips, Slavery from Roman Times to the Early Transatlantic Trade, 106.
214. Fynn-Paul, 'Tartars in Spain', 354
215. This is quoted in Richard Hellie, Slavery in Russia, 1450-1725 (Chicago, 1982), 22 n. 32.
216. There is a continuing debate on why and how early modern northern Europeans seemed to eschew slavery on their own soil, while becoming some of the greatest architects of the Atlantic slave system
217. See Davis, 'Looking at Slavery from Broader Perspectives', 458
218. Janet Martin, Medieval Russia, 980-1584 (Cambridge, 1995), 5-12
219. Spufford notes Pedro Tafur's statement that 'in this city [Kaffa] they sell more slaves, both male and female, than anywhere else in the world': Spufford, Power and Profit, 340
220. For the closing of European access to the Black Sea markets, see Phillips, Slavery from Roman Times to the Early Transatlantic Trade, 106
221. One of the first collections of essays which integrates the combined impact of Christian and Islamic slavery in West Africa from about 1450 right up to the twentieth century is Lovejoy (ed.), Slavery on the Frontiers of Islam.
222. See the introductory essay by Lovejoy, 'Slavery, the Bilād al-Sūdān and the Frontiers of the African Diaspora'
223. Phillips, Slavery from Roman Times to the Early Transatlantic Trade, 138-9
224. Lewis, Race and Slavery in the Middle East, 12
225. Hellie, Slavery in Russia, 695-708
226. A legacy of Byzantine Caesaropapism
227. T. H. Aston and C. H. E. Philpin (eds.), The Brenner Debate: Agrarian Class Structure and Economic Development in Pre-Industrial Europe (Cambridge, 1985).
228. Phillips notes that most of the roughly 1, 000 slaves brought to the Cape Verde Islands from Africa in the early 1500s were en route to other markets: Phillips, Slavery from Roman Times to the Early Transatlantic Trade, 141
229. Philip Curtin estimates that a total of 25, 000 African slaves were imported to the Atlantic islands between 1450 and 1600: Philip D. Curtin, The Atlantic Slave Trade: A Census (Madison, 1969), 268.
230. See Fisher, Slavery in the History of Muslim Black Africa, 18-32, 40-5
231. Segal, Islam's Black Slaves, 147-76
232. For an economic historian's take on the connection between British abolitionism and British expansionism, see David Eltis, Economic Growth and the Ending of the Transatlantic Slave Trade (Oxford, 1987)
233. Segal, Islam's Black Slaves, 157-9
234. Suetonius, The Twelve Caesars, trans. Robert Graves, revised Michael Grant (New York, 2003), 13.