BERNARD LEWIS
The research presented in this book was first undertaken as part of a group project on tolerance and intolerance in human societies, for which I was asked to prepare a paper on the Islamic world. I began by examining this topic in conventional terms and collected data on the relations between different religions-that is, in the sense in which the words "tolerance" and "intolerance" have in the past normally been used. At a certain point it occurred to me that such an enquiry need no longer be bound by the habits and concerns of a previous generation, concerns which have lost their sharpness for many of us at the present time, at least in the Western world. But if these concerns have diminished or disappeared, they have been replaced by others no less acute. In the present climate of opinion what matters most in this respect is not creed, or even class, but race. This is seen as the ultimate basis of identity and of difference; and it is in this area that, in much of the world, the crucial test of tolerance or intolerance is now applied. I therefore turned to this problem and, in the course of my work, began to examine certain assumptions hitherto unquestioningly accepted by Western as well as Islamic scholars.
The group project on tolerance and intolerance was never completed. The material on Islam, however, aroused some interest, and I was stimulated to pursue it further by an invitation to lecture on the subject at a combined meeting of the three institutes of Anthropology, International Affairs, and Race Relations in London. The lecture was delivered in December 1969 and published in a slightly expanded form in the London monthly Encounter in August 1970. This was, in turn, further expanded and published as a small book in New York in 1971, entitled Race and Color in Islamn.
The publication of a French translation in Paris in 1982 gave me the opportunity to make a number of substantial changes. In addition to correcting some errors, I added new documentation and discussed some further topics not touched upon in the earlier versions. I also appended a selection of relevant original sources, most of them translated from Arabic.
The study of race led inevitably, in the Islamic world as elsewhere, to the problem of slavery, by which both race relations and racial attitudes were profoundly affected. In preparing my successive earlier treatments of this topic, I was obliged to devote increasing attention to this aspect and to examine it in a Middle Eastern rather than in an Islamic context. In embarking, after an interval of years, on a new exploration of the theme, I decided to give the institution of slavery a more central position.
This immediately raised serious difficulties. One of them is the remarkable dearth of scholarly work on the subject. The bibliography of studies on slavery in the Greek and Roman worlds, or in the Americas, runs to thousands of items. Even for medieval Western Europe, where slavery was of relatively minor importance, European scholars have produced a significant literature of research and exposition. For the central Islamic lands, despite the subject's importance in virtually every area and period, a list of serious scholarly monographs on slavery-in law, in doctrine, or in practice-could be printed on a single page. The documentation for a study on Islamic slavery is almost endless; its exploration has barely begun.
Perhaps the main reason for the lack of scholarly research on Islamic slavery is the extreme sensitivity of the subject. This makes it difficult, and sometimes professionally hazardous, for a young scholar to turn his attention in this direction. In time, we may hope, it will be possible for Muslim scholars to examine and discuss Islamic slavery as freely and as openly as European and American scholars have, with the cooperation of scholars from other countries, been willing to discuss this unhappy chapter in their own past. But that time is not yet; meanwhile, Islamic slavery remains both an obscure and a highly sensitive topic, the mere mention of which is often seen as a sign of hostile intentions. Sometimes indeed it is, but it need not and should not be so, and the imposition of taboos on topics of historical research can only impede and delay a better and more accurate understanding. In this little hook, I have tried to deal fairly and objectively with a subject of great historical and comparative importance and to do so without recourse to either polemics or apologetics.
The present volume incorporates most of what was said in my early treatments. This has, however, been extensively revised, expanded, and recast, and a considerable body of new material added, including several new chapters, making it a new book dealing with a related but different topic. I have also added a documentary appendix, translated, where necessary, from the original languages. Some of the documents included in the French edition have been omitted, as they are already available elsewhere in English in print. Others have been added, including several translated for this purpose.
There remains the pleasant task of thanking those who have, in various ways, contributed to the completion of this book. My thanks are due to the authorities of the Public Record Office, the India Office Records, and the British Library in London; the Topkapi Saray Museum in Istanbul; the Bibliotheque Nationale in Paris; and Mr. Arthur A. Houghton, for permission to reproduce documents and pictures in their possession. Crown copyright mate rial in the Public Record Office and the India Office Records is reproduced by permission of the Controller of Her Majesty's Stationery Office.
I am indebted to David Goldenberg, John B. Kelly, Hava Lazarus-Yafeh, and Michel Le Gall for suggestions and help of various kinds. Above all, I would like to record my profound gratitude and appreciation to my assistant Leigh Faden, for her untiring and highly effective work in the preparation and numerous revisions of my text, to my research assistant Jonathan Berkey, whose scholarly knowledge and unstinting efforts in many ways lightened the labor of both research and writing and improved the quality of the results, and to both of them for preparing the index. Whatever faults and errors remain are entirely my own.
August 1989
1. Slavery, 3
2. Race, 16
3. Islam in Arabia, 21
4. Prejudice and Piety, Literature and Law, 28
5. Conquest and Enslavement, 37
6. Ventures in Ethnology, 43
7. The Discovery of Africa, 50
8. In Black and White, 54
9. Slaves in Arms, 62
10. The Nineteenth Century and After, 72
11. Abolition, 78
12. Equality and Marriage, 85
13. Image and Stereotype, 92
14. Myth and Reality, 99
Notes, 103
Documents, 141
Sources of Illustrations, 170
Index, 173
In 1842 the British Consul General in Morocco, as part of his government's worldwide endeavor to bring about the abolition of slavery or at least the curtailment of the slave trade, made representations to the sultan of that country asking him what measures, if any, he had taken to accomplish this desirable objective. The sultan replied, in a letter expressing evident astonishment, that "the traffic in slaves is a matter on which all sects and nations have agreed from the time of the sons of Adam . . . up to this day." The sultan continued that he was "not aware of its being prohibited by the laws of any sect, and no one need ask this question, the same being manifest to both high and low and requires no more demonstration than the light of day."'
The sultan was only slightly out of date concerning the enactment of laws to abolish or limit the slave trade, and he was sadly right in his general historic perspective. The institution of slavery had indeed been practiced from time immemorial. It existed in all the ancient civilizations of Asia, Africa, Europe, and pre-Columbian America. It had been accepted and even endorsed by Judaism, Christianity, and Islam, as well as other rel
igions of the world.
In the ancient Middle East, as elsewhere, slavery is attested from the very earliest written records, among the Sumerians, the Babylonians, the Egyptians, and other ancient peoples.` The earliest slaves, it would seem, were captives taken in warfare. Their numbers were augmented from other sources of supply. In pre-classical antiquity, most slaves appear to have been the property of kings, priests, and temples, and only a relatively small proportion were in private possession. They were employed to till the fields and tend the flocks of their royal and priestly masters but otherwise seem to have played little role in economic production, which was mostly left to small farmers, tenants, and sharecroppers and to artisans and journeymen. The slave popula tion was also recruited by the sale, abandonment, or kidnapping of small children. Free persons could sell themselves or, more frequently, their offspring into slavery. They could be enslaved for insolvency, as could be the persons offered by them as pledges. In some systems, notably that of Rome,.' free persons could also be enslaved for a variety of offenses against the law.
Both the Old and New Testaments recognize and accept the institution of slavery. Both from time to time insist on the basic humanity of the slave, and the consequent need to treat him humanely. The Jews are frequently reminded, in both Bible and Talmud.' that they too were slaves in Egypt and should therefore treat their slaves decently. Psalm 123, which compares the worshipper's appeal to God for mercy with the slave's appeal to his master, is cited to enjoin slaveowners to treat their slaves with compassion.5 A verse in the book of Job has even been interpreted as an argument against slavery as such: "Did not He that made me in the womb make him [the slave]? And did not One fashion us both?" (Job 31:15). This probably means no more, however, than that the slave is a fellow human being and not it mere chattel.' The same is true of the much-quoted passage in the New Testament, that "there is neither Jew nor Greek, there is neither bond nor free, there is neither male nor female; for ye are all one in Christ Jesus."' These and similar verses were not understood to mean that ethnic, social, and gender differences were unimportant or should be abolished, only that they conferred no religious privilege. From many allusions, it is clear that slavery is accepted in the New Testament as a fact of life.' Some passages in the Pauline Epistles even endorse it. Thus in the Epistle to Philemon, a runaway slave is returned to his master; in Ephesians 6, the duty owed by a slave to his master is compared with the duty owed by a child to his parent, and the slave is enjoined "to he obedient to them that are your masters, according to the flesh, in fear and trembling, in singleness of your heart, as unto Christ." Parents and masters are likewise enjoined to show consideration for their children and slaves.' All humans, of the true faith, were equal in the eyes of God and in the afterlife but not necessarily in the laws of man and in this world. Those not of the true faith-whichever it was-were in another, and in most respects an inferior, category. In this respect, the Greek perception of the barbarian and the Judeo-Christian-Islamic perception of the unbeliever coincide.
There appear indeed to have been some who opposed slavery, usually as it was practiced but sometimes even as such. In the Greco-Roman world, both the Cynics and the Stoics are said to have rejected slavery as contrary to justice, some basing their opposition on the unity of the human race, and the Roman jurists even held that slavery was contrary to nature and maintained only by "human" law. There is no evidence that either jurists or philosophers sought its abolition, and even their theoretical opposition has been questioned. Much of it was concerned with moral and spiritual themes-the true freedom of the good man, even when enslaved, and the enslavement of the evil freeman to his passions. These ideas, which recur in Jewish and Christian writings, were of little help to those who suffered the reality of slavery."' Philo, the Alexandrian Jewish philosopher, claims that a Jewish sect actually renounced slavery in practice. In a somewhat idealized account of the Essenes, he observes that they practiced a form of primitive communism, sharing homes and property and pooling their earnings. Furthermore,
not a single slave is to be found among them, but all are free, exchanging services with each other, and they denounce the owners of slaves, not merely for their injustice in outraging the law of equality, but also for their impiety in annulling the statute of Nature, who mother-like bore and reared all men alike, and created them genuine brothers, not in mere name, but in very reality, though this kinship has been put to confusion by the triumph of malignant covetousness, which has wrought estrangement instead of affinity and enmity instead of friendship."
This view, if it was indeed held and put into practice, was unique in the ancient Middle East. Jews, Christians, and pagans alike owned slaves and exercised the rights and powers accorded to them by their various religious laws. In all communities, there were men of compassion who urged slaveowners to treat their slaves humanely, and there was even some attempt to secure this by law. But the institution of slavery as such was not seriously questioned, and was indeed often defended in terms of either Natural Law or Divine Dispensation. Thus Aristotle defends the condition of slavery and even the forcible enslavement of those who are "by nature slaves, for whom to be governed by this kind of authority is beneficial";'' other Greek philosophers express similar ideas, particularly about enslaved captives from conquered peoples." For such, slavery is not only right; it is also to their advantage.
The ancient Israelites did not claim that slavery was beneficial to the slaves, but, like the ancient Greeks, they felt the need to explain and justify the enslavement of their neighbors. In this, as in other matters, they sought a religious rather than a philosophical sanction and found it in the biblical story of the curse of Ham. Significantly, this curse was restricted to one line only of the descendants of Ham, namely, the children of Canaan, whom the Israelites had subjugated when they conquered the Promised Land, and did not affect the others.14
The Qur'an, like the Old and the New Testaments, assumes the existence of slavery. It regulates the practice of the institution and thus implicitly accepts it. The Prophet Muhammad and those of his Companions who could afford it themselves owned slaves; some of them acquired more by conquest."
But Qur'anic legislation, subsequently confirmed and elaborated in the Holy Law,16 brought two major changes to ancient slavery which were to have far-reaching effects. One of these was the presumption of freedom; the other, the ban on the enslavement of free persons except in strictly defined circumstances.
The Qur'an was promulgated in Mecca and Medina in the seventh century, and the background against which Qur'anic legislation must he seen is ancient Arabia. The Arabs practiced a form of slavery, similar to that which existed in other parts of the ancient world. The Qur'an accepts the institution, though it may be noted that the word `abd (slave) is rarely used, being more commonly replaced by some periphrasis such as ma malakat aymanukum, "that which your right hands own." The Qur'an recognizes the basic inequality between master and slave and the rights of the former over the latter (XVI:71; XXX:28). It also recognizes concubinage (IV:3; XXIII:6; XXXIII:50-52; LXX:30). It urges, without actually commanding, kindness to the slave (IV:36; IX:60; XXIV:58) and recommends, without requiring, his liberation by purchase or manumission. The freeing of slaves is recommended both for the expiation of sins (IV:92; V:92; LVIII:3) and as an act of simple benevolence (11: 177; XXIV:33; XC:13). It exhorts masters to allow slaves to earn or purchase their own freedom. An important change from pagan, though not from Jewish or Christian, practices is that in the strictly religious sense, the believing slave is now the brother of the freeman in Islam and before God, and the superior of the free pagan or idolator (11:221). 17 This point is emphasized and elaborated in innumerable hadiths (traditions), in which the Prophet is quoted as urging considerate and sometimes even equal treatment for slaves, denouncing cruelty, harshness, or even discourtesy, recommending the liberation of slaves, and reminding the Muslims that his apostolate was to free and slave alike."
Though slavery was maintained
, the Islamic dispensation enormously improved the position of the Arabian slave, who was now no longer merely a chattel but was also a human being with a certain religious and hence a social status and with certain quasi-legal rights. The early caliphs who ruled the Islamic community after the death of the Prophet also introduced some further reforms of a humanitarian tendency. The enslavement of free Muslims was soon discouraged and eventually prohibited. It was made unlawful for a freeman to sell himself or his children into slavery, and it was no longer permitted for freemen to be enslaved for either debt or crime, as was usual in the Roman world and, despite attempts at reform, in parts of Christian Europe until at least the sixteenth century. It became a fundamental principle of Islamic jurisprudence that the natural condition, and therefore the presumed status, of mankind was freedom, just as the basic rule concerning actions is permittedness: what is not expressly forbidden is permitted; whoever is not known to be a slave is free.'9 This rule was not always strictly observed. Rebels and heretics were sometimes denounced as infidels or, worse, apostates, and reduced to slavery, as were the victims of some Muslim rulers in Africa, who proclaimed jihad against their neighbors, without looking closely at their religious beliefs, so as to provide legal cover for their enslavement. But by and large, and certainly in the central lands of Islam, under regimes of high civilization, the rule was honored, and free subjects of the state, Muslim and non-Muslim alike, were protected from unlawful enslavement.
Since all human beings were naturally free, slavery could only arise from two circumstances: (1) being born to slave parents or (2) being captured in war. The latter was soon restricted to infidels captured in a jihad.
These reforms seriously limited the supply of new slaves. Abandoned and unclaimed children could no longer be adopted as slaves, as was a common practice in antiquity,'" and free persons could no longer be enslaved. Under Islamic law, the slave population could only be recruited, in addition to birth and capture, by importation, the last either by purchase or in the form of tribute from beyond the Islamic frontiers. In the early days of rapid conquest and expansion, the holy war brought a plentiful supply of new slaves, but as the frontiers were gradually stabilized, this supply dwindled to a mere trickle. Most wars were now conducted against organized armies, like those of the Byzantines or other Christian states, and with them prisoners of war were commonly ransomed or exchanged.'' Within the Islamic frontiers, Islam spread rapidly among the populations of the newly acquired territories, and even those who remained faithful to their old religions and lived as protected persons (dhimmis) under Muslim rule could not, if free, be legally enslaved unless they had violated the terms of the dhimma, the contract governing their status, as for example by rebelling against Muslim rule or helping the enemies of the Muslim state or, according to some authorities, by withholding payment of the Kharaj or the Jizya, the taxes due from dhimmis to the Muslim state.
Race and Slavery in the Middle East: An Historical Enquiry Page 1