We live in an age of identity politics. We define ourselves by one or more objective measures: measures of race, ethnicity, gender, politics, religion, sexual orientation, to name just a few. Those measures then define who we are to others. They determine our place in society, the communities with which we identify, our attitudes towards others and other communities.
The politics of identity are fraught, and they interact in ways that both liberate and confine. On the one hand we prize diversity. On the whole, this is a good thing, since it reflects a larger transformation in American life. Like it or not, the fact is that we are becoming, have become, a “multi-cultural society.” No matter what terms we use to define diversity—racial, ethnic, religious, sexual, gender, whatever—we are more diverse now than we have ever been, and we are destined to grow more so. Multi-culturalism is not an option; it is the future. The only question is how, and how well, we are going to deal with it.
On the other hand, the politics of identity can at times stoke tension between the different identities which make up our social mosaic. Moreover, our celebration of diversity masks a contradictory truth: that we are at the same time caught up in a larger, deeper historical process of cultural homogenization. This process of homogenization is the product of the same historical forces which have encouraged us to embrace diversity. Globalization has brought us together—both individuals and entire societies. Proximity can breed contempt, but it has also contributed to a spirit of tolerance which transforms diversity from something to be feared into something to be embraced. But globalization is also undermining the structural foundations of that very cultural diversity. One need only think, for example, of the alarming disappearance of distinct human languages. One language disappears, on average, every fourteen days: it disappears, that is, in that the last remaining speaker of the language dies, and carries with him the cultural legacy of his spoken tongue. There are currently 7,000 languages in daily use; by the next century that number will have been cut in half. What sort of diversity will be possible when English, or some barbaric mutation of English, is the only language the world’s billions of humans will speak?
This tectonic process of cultural homogenization is important for us because it lies behind much of the tension and violence that has been endemic in the modern Middle East. Some years ago, at the invitation of a commercial press, I began writing a narrative history of the Middle East from the rise of Islam in the seventh century down to the present day. One of the reasons I took up the challenge was the opportunity the book presented to play with an idea that had been piquing me for some time. As a medieval historian, when I look at the modern Middle East, what I see is a region which has for the last two centuries suffered from a series of political movements which have, in different ways, embraced the forced homogenization of cultural difference. This was true of nationalism, an ideology which gathered steam toward the end of the nineteenth century, and then dominated the politics of the region for much of the twentieth. Whether in the form of Arabism, or Zionism, or any of its other manifestations, nationalism encouraged its adherents to embrace a particular expression of cultural and political identity to the exclusion of others. And nationalism has had no monopoly on this exclusivist vision of cultural and political identity. It is certainly characteristic of the radical religious ideologies which are now eclipsing nationalism in much of the region.
The inexorable homogenizing tendencies of these modern political developments stand out to me, as a medieval historian, because they present a sharp contrast to the relatively tolerant atmosphere of the pre-modern Islamic world. I am not trying to depict classical Islamic society as a happy utopia which knew no discrimination, as some apologists will do. There was plenty of discrimination in pre-modern Islamic societies, which we will come to shortly. But the discrimination was a reflection of the fact that the Middle East was extraordinarily diverse. We in the twenty first-century United States are becoming multi-cultural. The Islamic Middle East was multi-cultural from the very beginning, and its peoples of necessity had to work out mechanisms for dealing with its diversity.
One of the first things to note is that some of those things by which modern humans frequently define their personal, social, and political identities were not important to the inhabitants of the pre-modern Middle East—or at least were not important to them in the way they are important to us now. A salient example is that of class. Americans don’t like to talk about class—or, more precisely, we are often told that we don’t like to talk about class. But in fact class has been one of the most important markers of political identity for the last century and a half, especially in Europe but also here in the United States. Of course, some of those things which help to define class as an analytical concept—things like wealth, occupation, property—were naturally present in the pre-modern Middle East, and sometimes they had political implications. But class was generally not important to social and political identity for the inhabitants of the region before the modern period.
There were exceptions, although they may be exceptions which prove the rule. An interesting example was a group known as the Qarmatians. The Qarmatians were sectarian Shi`is who, in the tenth century, rebelled against the Islamic caliphate and established a utopian regime in northeast Arabia. Most accounts locate their origins in peasant communities and associate their rebellion with efforts to overthrow the authority of oppressive landowners. Some accounts of the Qarmatians describe them as creating a sort of classless society, in which property was shared in an egalitarian manner—their property, and their women, too. Those accounts may reflect less what the Qarmatians actually did than what their Sunni enemies believed they did. But even if there was a kind of levelling, a flattening out of social distinctions based on wealth, this did not necessarily result from what a Marxist would call class consciousness. Rather it was driven, and justified, by a radical religious creed grounded in millenarian expectations—that is, in expectations of a looming end time when the chosen instruments of God’s will would overthrow a corrupt social and political order.
By contrast, one of the principal markers of personal identity in the pre-modern Islamic world is (theoretically, at least) absent from our own: namely, the distinction between those who were slaves and those who were free. Slavery was a widespread phenomenon in the Islamic Middle East. In Istanbul, the capital of the Ottoman Empire, at the turn of the seventeenth century, approximately 20% of the population held slave status. Slavery took many forms in the pre-modern Middle East, but in general meant something very different than what it did, say, in the antebellum United States. There was very little brutal plantation-style slavery—a few notable exceptions, such as the Zanj of southern Iraq in the ninth century, aside. Most slaves served in some sort of domestic capacity—as cooks, cleaners, household servants—and as such were frequently treated in effect as members of the owners’ families. Certain types of slavery carried with them an almost exalted status. Concubines, for example, female slaves purchased specifically for the sexual pleasure of their masters, often held a position in the household not at all inferior to free-born wives. And if a concubine bore her master a child, that child was free and fully legitimate—no less so than the offspring of a man and his wife.
Other markers of identity which are common in our world were also common in medieval Islamic societies, although the experiences of them in those two settings differ significantly. Take, for example, ethnicity. Ethnicity—that is, social distinctions rooted in cultural and especially linguistic differences, distinctions which may under some circumstances have political consequences—was a meaningful marker of identity in the pre-modern Middle East, and a comparison to our own conceptions of ethnicity is therefore a useful exercise. On the one hand, there is an important principle of Islamic law that ethnicity should not matter. “We have created you male and female,” says the Qur’an, “and [have] made you peoples and tribes that you may know one another. Indeed, the most noble of you in the sight of Allah is the most righteous of you.” The usual interpretation of this verse is that what we would call the ethnic diversity of the human race is a sign of the splendor of God’s creation. No one—no ethnic group, no tribe—is to be preferred to any other: we are all equal in the eyes of God. The only thing which meaningfully and legitimately distinguishes one human from another is the degree of individual piety. This egalitarianism became a distinct and, for most Muslims, a normative element in Islamic juristic discourse.
Nonetheless, ethnic identity mattered and shaped political experience in many ways. Perhaps the most important example is the way in which Arab identity shaped the contours of the early Islamic polity. By the time a consensus had been reached about the substance of Islamic law in the ninth and tenth centuries, Islam had been largely detached from its Arab roots, and Muslim discourse had come to insist upon the priority of religious rather than ethnic identity. But that was the end result of a long and contentious process. There is much debate among historians over these matters, but in broad terms it is fair to say that, in its origins, Islam was tied very closely to Arab identity. Islam was probably conceived of by its earliest practitioners as a monotheistic faith for the Arab people. The Jews had their religion, and the Christians theirs; Islam was a monotheistic faith for a people, the Arabs, who had not previously been given their own revelation. For some decades after the rise of Islam, in order to convert to the new faith, a non-Arab couldn’t simply embrace Islam. It was necessary for a Muslim Arab or his tribe to embrace him as a “client” (mawla, pl. mawali)—a sort of adoptive Arab status. Over the first century of Islam, several caliphs actively discouraged the conversion of non-Arabs to Islam—for complex reasons, not least because the conversion of non-Arabs would undermine the tax basis of the early Islamic state.
Eventually, the pietistic view—that, as the Qur’an said, “the most noble of you in the sight of Allah is the most righteous of you”— won out. Nonetheless, the preeminence of the Arabs in early Islam left its residue on later, more cosmopolitan versions of the faith. So, for example, there are plenty of statements from the early Islamic period, some of them attributed (probably inaccurately) to the Prophet, that Arabs should refrain from marrying non-Arabs. As late as the ninth century, an Arab poet could write a satirical poem comparing the miscegenation of Arabs and non-Arabs to Arab women fornicating with donkeys. The ethnic hierarchy of early Islam survived in a rather arcane doctrine of Islamic law which allowed a woman’s male guardian to object to her marriage to a socially unequal male—and one recognized ground for such inequality is the preeminence of pure-blooded Arab families. More important is the consensus of Islamic political theorists that a legitimate caliph can only be chosen from among the descendants of the Prophet’s own tribe of Quraysh. This principle has played a role in recent political developments, as the so-called Islamic State has gone to some lengths to establish that its current “caliph,” Abu Bakr al-Baghdadi, is in fact an Arab of Qurashi descent.
To be sure, there was nothing in the pre-modern Middle East resembling ethnic nationalism. Nationalism as an ideology is a peculiarly modern phenomenon. It is a product of historical contingencies—for example, the rise of a politically active middle class—which were not in place before the nineteenth and twentieth centuries. Before the nineteenth century, it would not have occurred to most Arabs, or Egyptians, or Jews, or Turks, that what we would call their “ethnic identity” should be the fundamental basis of political community and legitimate government. Nonetheless, people were aware of ethnicity, and sometimes ethnic identities clashed or found themselves in competition. Other than the complex relationship of Arab and Muslim identity, the most important example of an ethnic identity with political ramifications was that of the Turks. Beginning in the late ninth century, growing numbers of Turks from Central Asia began to infiltrate the Islamic Middle East. Turks were praised for their martial abilities, and soon they constituted the core of the Muslim empire’s armies. Before long, those same Turks became politically dominant, eclipsing and eventually replacing the authority of the established Muslim governments. From the eleventh century down into the modern period, the ruling elites in most states in the Middle East consisted of groups who were, in some sense, “Turkish.” Consequently, Turks as an ethnic group became associated in people’s expectations with government, with ruling. According to an apocryphal statement in an eleventh-century text, Muhammad advised his followers to “Learn the language of the Turks, for their dominion will be long.” That association of Turks with government was probably a factor in perpetuating the long drawn-out twilight of Ottoman rule in the Middle East—although ultimately it was also a factor in the rise of Arab nationalism.
Another important marker of identity in the pre-modern Middle East was gender. Gender probably shaped an individual’s experience more firmly than any other marker of identity. After all, a slave might always be freed, and a Jew or Christian might convert to Islam. By contrast, a woman was a woman, and a man a man—the notion of a flexible construction of gender being incomprehensible to the inhabitants of the pre-modern Middle East.
The importance of gender as a marker of identity is apparent to anyone with even a superficial knowledge of Islamic law. It was not simply that the lawyers spent much time outlining the responsibilities of women regarding matters such as marriage, the family, sex, childrearing, and other matters which might have special resonance in women’s lives. It was also that one’s gender defined one’s rights and responsibilities in a variety of more public arenas: for example, where and how one should pray, or whether and what a woman might expect to inherit from her father, or what value would be accorded to her testimony in a court of law.
So important was gender identity to the Islamic lawyers that they went to great lengths to resolve those rare cases in which an individual’s gender really was—that is, was objectively—ambiguous. Hermaphrodites, or “inter-sexed” individuals, who had external genitalia of both male and female, posed an almost existential problem for the classical Muslim lawyers. As the Qur’an made clear, gender was a part of the fabric of the universe, so every individual had to be either male or female—a person could not be both. More immediately, one’s social role was largely defined by gender. For example, in a mixed congregation, men should pray at the front, women behind; the prayers of a man who prays while standing behind a woman are invalid. So what were the lawyers to do with an individual whose sexual identity, on the basis of his (or her) external genitalia, was ambiguous? What they did was to go to great lengths to establish criteria for determining gender—crafting, for example, elaborate rules for observing whether a hermaphrodite urinated as a male or as a female (or, in cases in which it did both, measuring the quantity of urine which emerged from male and female organs).
In Islamic constructions of political authority, however, there was no ambiguity whatsoever. The jurists were virtually unanimous in insisting that politics was an exclusively male arena—this despite the well-known political roles played by some of Muhammad’s wives. Some especially pietistic jurists were willing to dispense with the ethnic requirement that a caliph must be a Qurashi Arab. Their emphasis on piety and competence was reflected in their dictum that anyone could be a caliph, “even a slit-nosed Abyssinian slave,” so long as he ruled justly and administered the shari`a. But even they, for the most part, could not countenance the possibility that a woman might serve as caliph. There were only three instances in which women ruled over Muslim states in their own names—as “sultans,” a title adopted by most medieval dynasties in place of the earlier title “caliph.” The most famous of those involved a woman named Shajar al-Durr, the concubine of a sultan of Egypt in the mid-thirteenth century who was briefly raised to the throne, and even had coins minted in her own name—until a message arrived from the caliph in Baghdad that emphatically rejected the right of a woman to rule.
Of course, formal rule is not the same thing as political power. Human relations being what they are, there were episodes when women might play an important political role behind the scenes. Not infrequently, these episodes provoked the wrath of male jurists and historians. Prompted by the influence wielded by the women of the imperial household, an eleventh-century Persian vizier warned his monarch about the wiles of women. Ottoman observers condemned a long period in the sixteenth and seventeenth centuries when the mothers and concubines of the reigning sultans wielded considerable power—a period the critics dismissed as the “sultanate of the women.” Such episodes aside, however, politics more than any other sphere of life was one in which gender identity formally and definitively circumscribed public behavior.
Of all possible markers of identity, the most important and far-reaching was that of religion. Islam was born into a world of diverse faiths: Judaism and Christianity, of course, but also other historically significant, although perhaps now less well known traditions, such as that of the Zoroastrians in Iran. Indeed, the very idea of “religion” as we know it—that is, “religion” as adherence to a discrete and mutually exclusive body of convictions and practices—this notion of religion as a marker of identity is a product of the religious competition which characterized the Middle East in late antiquity. That means that Islam from the very beginning had to do what American Christians are only now learning to do: to live in a world in which theirs is not necessarily the dominant faith. How did they do so?
We should start by remembering that Islam is a diverse phenomenon, and that the Muslim experience has been very different in different times and different places—just as, one might say, the Christian experience in twenty first-century America is rather different than, say, Spain during the Inquisition. God is One, or so, at least, Muslims, along with Jews and Christians, proclaim. But even if God is One, humans have never been able to agree about very much more concerning the divine, or how God should be worshipped. That is just as true of Muslims as it is of others, and their differences of opinion have had a profound impact on how Muslims have understood their own religious identity.
The major division within the Muslim world is that between the Sunni and Shi‘i branches of Islam. In the wake of the Islamic revolution in Iran in 1978-9, many Americans came to hold a number of mistaken perceptions about Shi‘i Islam. For example, because Iran is overwhelmingly Shi‘i, and because the Iranian revolution was waged on specifically Shi‘i terms, many Americans came to think of Shi‘ism as a specifically Iranian form of Islam, as something that distinguished Iranian (or Persian) Muslims from Arab Muslims. It is true, of course, that the vast majority of Iranians are Shi‘is, although that has only been true for the last 400 or 500 years: before the year 1500, that is, in the pre-modern period, most Iranians, like most Muslims generally, were Sunnis. But in no meaningful sense is Shi‘ism a specifically Iranian form of Islam. Nor is it a particularly radical form of Islam; there are plenty of “radicals” on the Sunni side, too, as well as large majorities of both who are generally more “moderate.”
What, then, is the difference between Sunni and Shi‘i identities? The root of the difference is a historical one, and a political one. Basically, it has to do with how a believer feels about certain events that took place almost fourteen hundred years ago. This is an immensely complicated question, but in a nutshell, the story is this:
When Muhammad died in the year 632, he left no instructions about who was to succeed him as leader of the Muslim community. Some Muslims came to feel that that leadership should have passed—as it apparently did in fact—to some qualified individual who was chosen and recognized by the community at large. There was no question about this individual inheriting Muhammad’s status as a prophet. But most Muslims came in time—this process took several hundred years—to feel that the political leadership of the community should be vested in some qualified individual who would be chosen through a process of consensus. Religious authority, by contrast, passed to what the tradition calls the ulama—literally, “those who know,” that is, the religious scholars. Over time—and this process of separating Muslim sectarian identities took time, quite a bit of time, well over a century—these Muslims came to be known as “Sunnis.” Today, they constitute perhaps 90% of Muslims worldwide.
But other Muslims felt differently. For them, leadership of the community, and absolute authority over both its political and religious affairs, should have passed after Muhammad’s death to his cousin and son-in-law, a man named `Ali, and after `Ali’s death to his sons and descendants: that is, to the descendants of Muhammad himself. This group came to be known in Arabic as the shi‘at ‘ali, the “party of ‘Ali,” that is, the Shi‘a. They set themselves apart from other Muslims by their conviction that the community made a terrible mistake in not ensuring that ‘Ali and his descendants held effective rulership, and that the community would be acting in contravention of the will of God until, through some political revolution, the rightful heir of Muhammad, through his son-in-law ‘Ali, was recognized as Imam, or “leader.” In other words, the difference between Sunni and Shi‘i identities has nothing to do with being Iranian, or radical, or anything like that; rather, it is a fundamentally historical difference, and also a political one.
The second issue concerning religious identity I want to address is perhaps an even more interesting one: namely, the historical relationship between Islam and the diverse religious communities of the Middle East. It is important to start by reminding ourselves that the recurring image that many in the West have—of Islam spreading through “the sword”—is very misleading. As a general rule, forced conversion is repugnant to the Islamic tradition. Of course, there have been exceptions, but by and large Muslims have adhered to the Qur’anic principle that “there is no compulsion in religion.” Religious decisions, that is, must be made freely.
This is not to say that Islam has not had a violent side. The seventh century of the Common Era was one of the most decisive in human history, precisely because it was then that the newly-converted Muslim Arabs swept out of Arabia, and within one hundred years had conquered all the territory between the Atlantic Ocean and the borders of China. But during that period of conquest, when Islam really did in a sense “spread by the sword,” there was little effort to convert those who came under the rule of the Muslim Arabs. For almost a century, as we have seen, the Muslims thought of Islam as the monotheistic religion of the Arabs. Hence those efforts to discourage conversion by non-Arab peoples. It was only later, from the eighth century on, and in part through competition with the universalist imperative of Christianity, that Muslims overwhelmingly came to think of their religion as one that was addressed to all of humankind.
That points to a second element of the Islamic history of inter-faith relations which is also important: namely, that for Muslims the question of religious identity has also had a political dimension. Our notion that the sacred and the secular can and should be separated, that church and state represent distinct spheres of practice and authority, is in many ways an odd one in human history. Some have argued that it has roots in Christianity itself. Jesus, after all, is quoted in the Gospel as urging his followers to “render unto Caesar that which is Caesar’s, and unto God that which is God’s.” More historically, it is true that, for its first three centuries, the church grew up, not so much in opposition to the state, but in a separate sphere from the Roman state. But after the conversion of the Roman emperors to Christianity in the fourth century, Christians learned very quickly how to wield political power, and for the next thousand years consistently defined their political institutions in explicitly religious terms. The separation of church and state, the sacred and the profane, is in fact a product of the Enlightenment of the eighteenth century. It is the Enlightenment, together with the Enlightenment’s skepticism regarding absolute religious truth, which has left a secularist legacy on our own society. This separation of religion and politics is thus a very recent thing.
From the very beginning, by contrast, Muslims have claimed to find in Islam political as well as religious authority. And ever since the seventh century, Islam has been the religion of those who wielded political power in the Middle East. For the whole of Islam’s history, in other words, Muslims have been in charge. Indeed, in the medieval period, Muslim jurists debated among themselves whether it was even possible to live as a Muslim in a land that was not ruled by Muslims.
That suggests a third point: namely, that the very principle around which many of us might frame the question of inter-faith relations—that is, equality—simply was not historically an issue for Muslims, at least until the irruption on the scene of Western, secular political ideas in the last century and a half. There is no question that, in pre-modern Islamic societies, non-Muslims were treated as second-class citizens. Jews and Christians and Zoroastrians living under Muslim rule were subject to certain restrictions: on what they could wear; how they were to treat Muslims; whether they could carry arms or ride horses; the degree to which they could build or repair their houses of worship; what special taxes they had to pay. All that seems to us, committed as we are to the principle of equality, to be unfair. But it may be that equality is the wrong way to think about the problem, at least as far as the pre-modern Middle East is concerned. In the first place, it is certainly the case that Jews and Christians in the pre-modern Middle East were treated much better than were religious minorities (Jews, mostly) living in Christian Europe. For the most part, pre-modern Muslim history is devoid of the kind of pogroms and massacres which medieval and early modern Christians often inflicted on Jews. So, for example, when the Spanish monarchs Ferdinand and Isabella drove the Jews out of Spain in 1492, many of them ended up in the Ottoman Empire, which they chose as a place of refuge because of the relatively high degree of tolerance and freedom they found there. In the second place, as Bernard Lewis has observed, second-class citizenship is a form of citizenship—not perfect, perhaps, by modern standards, but respectable for its day. And so while Jews and Christians suffered from certain restrictions, and even from certain humiliations, they were also guaranteed the protection of the Muslim state, and were allowed a fair degree of autonomy, to order their lives and direct their own communities as they saw fit.
And so, for the inhabitants of the pre-modern Middle East, as for us today, identity was a fraught and complicated matter. Identity shaped who they were, and what sort of communities they belonged to, and how they related to the body politic. What was fundamentally different was that, for most, identity involved little or nothing in the way of choice. For us, of course, that is an entirely different matter.
 Russ Rymer, “Vanishing Voices,” National Geographic (July, 2012).
Shattered Mosaic: The Middle East Since the Rise of Islam, to be published shortly by W. W. Norton.
 Richard Bulliet, Islam: The View from the Edge (Columbia University Press, 1994), 213; Jonathan Berkey, The Formation of Islam: Religion and Society in the Near East, 600-1800 (Cambridge University Press, 2003), 118.
 Bernard Lewis, Race and Slavery in the Middle East: An Historical Enquiry (Oxford University Press, 1990), 85-8.
 There are for example frequent reports of tension between Arabs and Berbers in North Africa. A famous case of ethnic rivalry was that between Arabs and Persians during what was known as the Shu`ubiyya movement.
 Bernard Lewis, Islam from the Prophet Muhammad to the Capture of Constantinople, vol. 2: Religion and Society (Oxford University Press, 1987), 197.
 Paula Sanders, “Gendering the Ungendered Body: Hermaphrodites in Medieval Islamic Law,” in Women in Middle Eastern History: Shifting Boundaries in Sex and Gender, ed. Nikki Keddie and Beth Baron (Yale University Press, 1991), 74-95.
 Patricia Crone, God’s Rule: Government and Islam (Columbia University Press, 2004), 57.
 See Leslie Peirce, The Imperial Harem: Women and Sovereignty in the Ottoman Empire (Oxford University Press, 1993).
 For a useful survey of Shi`i identity and history, see Bernard Lewis, “The Shi`a in Islamic History,” Islam and the West (Oxford University Press, 1993), 155-65.
 There is an excellent introduction to the history of relations between Muslims and non-Muslims living under Muslim rule in Bernard Lewis, The Jews of Islam (Princeton University Press, 1984), 3-66.