Preaching was central to the Muslim tradition and the Muslim experience from the very beginning. Muhammad is frequently depicted in the sources as preaching to his community – as, for example, in the famous ‘Farewell Sermon’, delivered shortly before his death, in which the Prophet is said to have communicated his final instructions to his followers. Muhammad did not, however, pioneer an entirely new genre of discourse. On the contrary, he and his contemporaries could draw on a long tradition of oratory and homiletics among the pre-Islamic Arabs (Serjeant 1983: 117–22).

Preaching became an important part of most Muslims’ experience of their faith and has remained so down to the present day. There are several reasons why this was the case. In the first place, Islam – like Christianity and especially Judaism – is an explicitly normative faith. Those who live in, or come from, the monotheistic traditions which emerged from the late antique Near East may take this for granted, but in fact not all religions are so keen on shaping the behaviour of their adherents. Judaism, as it emerged in the centuries before Christ, was constructed around a central paradigm: that God expects us to do (or not to do) certain things; and if we follow his will, we will be rewarded, and if we do not, we should expect punishment of some sort, either in this life or in the world to come. Christianity and then Islam followed the same paradigmatic structure. Islam especially became a religion of laws, expressed ultimately in the comprehensive shariʿa. But the impulse to shape the behaviour of its adherents is already deeply embedded in the Qurʾan, with its repeated injunction to ‘command what is good and forbid what is evil’. Islam is a normative faith, and so from the beginning a degree of exhortation, in which some Muslims would encourage other Muslims to behave in a certain way, was a part of the mix.

Islam is also a textual religion. This point should need no elaboration: without the Qurʾan, without the hadith, there would be no Islam. It is significant that the closest thing to a clergy in the Muslim tradition is the ʿulamaʾ, ‘those who know’ – that is, specialists on the texts on which the religion rests. But one consequence of the tradition's textual basis was to reinforce the role of ‘preaching’ in Islam.

In People of the Book: Canon, Meaning, and Authority, Moshe Halbertal described Judaism as a “text-centered community” (Habertal 1997). What he meant is that Judaism, as a community, is defined by the texts, principally the Torah and the Talmud, which it placed at the centre of its religious experience. In his analysis, it was not an idea about God which distinguished a Jew. As Halbertal pointed out, historically Jews have had radically different, even contradictory conceptions of the deity. Nor was Judaism defined by ritual, not even a ritual as central to Jewish identity as circumcision. It was rather the text, or certain texts, and the way in which those texts were employed both in religious discourse and in social life, which gave to the Jewish tradition its coherence. And not just Jewish identity, but Jewish divisions. Attitudes towards the foundational texts lie at the heart of the major sectarian divisions within the Jewish world. For example, it is a textual dispute which in the Middle Ages marked the difference between Rabbanites and Karaites: the Rabbanites accepting the authority of the Talmud in addition to the Bible; the Karaites rejecting the authority of anything but the canonical scriptures.

Change a key term or text, and these remarks also describe the situation in Islam. For the Torah and Talmud, substitute the Qurʾan and the hadith – the word of God and the words of his Prophet. The texts may be different, but Islam, too, is a “text-centered community.”

Of course, behind the simple focus on texts lies a more complex reality, not least because the text or texts in question are not easily defined. In the case of Judaism, the Torah alone is not the text: it is the Torah plus the Mishnah and the Talmud plus the entire universe of commentaries on the sacred texts which define the tradition. Islam constitutes a “text-centered community” in precisely analogous terms. It is not the Qurʾan and hadith alone which define the community, but a complex array of commentaries on those texts – especially those commentaries which came to define Islamic law.

As an observation about Islam, all this is, perhaps, self-evident. But it has very important consequences for Islamic history.

Islam, like any major religion, is a complex phenomenon. Diverse, at times even contradictory, it resists summary and categorical description. The religion was born among the inhabitants of the Arabian Peninsula in the early seventh century CE. Yet Islam as we know it is the product of many peoples and cultures: the Arabs, but also converts from among the Jewish, Christian and Zoroastrian communities of the Near East. By 1800 Islam had expanded into south-eastern Europe, central, south and South-East Asia, and sub-Saharan Africa. Indeed, Islam formed what many have called the first global civilisation. Its ecumenical reach finds inspiration in a famous verse from the Qurʾān: ‘Thus we have appointed you a middle nation, that you may be witnesses to mankind’ (Q 2:143). As a historical matter, however, Islam is best understood as an expression of the larger tradition of Near Eastern monotheism. It is distinct from its older cousins, Judaism and Christianity, but its origins and early development owe much to them.

The historical relationship of Islam and Christianity is especially fraught. From the beginning, Muslims were aware of Christianity. Several of their core beliefs were constructed as a response to or reaction against Christian doctrine. The Muslim state took shape in the context of an existential struggle against the Byzantine empire, which understood itself to be the defender of the Church. Over the ensuing centuries the competition between Christianity and Islam took ever sharper forms. In the Middle East, at least, most of those who eventually converted to Islam did so from Christianity, a fact that underlies the intense competition between the two religions. The competition only worsened, and grew more violent, with the rise of the Crusading movement in the late fifth/eleventh and sixth/twelfth centuries.

Some readers will take exception to the use of the term “medieval” to describe a phase of Islamic history. The term is borrowed from European history, where it signifies a period, the “Middle Ages,” distinguished from the “classical” one that preceded it and the “Renaissance” by which it was followed. In European history the term originally had something of a pejorative connotation – that the Middle Ages constituted a sort of valley between the peaks of classical and Renaissance culture and learning – although most historians would today describe the Middle Ages as considerably less “dark” than was earlier thought. The risks of abstracting the term from the European context that produced it, and applying it to the wholly different circumstances of the Islamic Near East, are obvious.

On the other hand, there were peculiar characteristics of the Islamic society and its religious institutions that took shape in the period between the beginning of the eleventh and the end of the fifteenth centuries. In the “Islam” which emerged over the course of these centuries are to be found various patterns of religious authority, affiliation, and relationship which distinguish it from what came before, which laid the foundation for the Islamic societies (particularly in the form of the Ottoman and Safavid empires) that followed, and which shaped the Islamic identities of those Muslims who suddenly found themselves faced with the changed circumstances of the modern period.

There were two basic patterns in medieval political life which had a profound impact on the religious life of the Muslim communities of the Near East. The first was a persistent and constantly shifting diffusion of power away from the center and towards more local and limited regimes. The central fact here was the decline in the power and authority of the cAbbasid caliphs. So much of early Islamic discourse and conflict had focused on the institution of the caliphate, yet in the Middle Period, despite moments of resurgence, its authority flickered and finally died. The jurists held a deep attachment to the office of the caliph as an integral part of the sharica; nonetheless they too were forced to confront the political realities. Toward the end of the Buyid period, the Baghdadi Shafici qadi al-Mawardi (d. 1058) wrote a treatise on the law of government, in the first chapter of which he outlined the position and powers of the caliph. His famous description is a classic treatment of the caliph as an active centerpiece of the unity of the Islamic umma, as the cornerstone of the administration of God's law, in the face of the growing political fragmentation of the medieval period. There is a certain irony here, as al-Mawardi wrote his treatise at a time when the caliphate had ceased to wield effective authority.

The master narrative of the two and a half centuries which followed the cAbbasid Revolution might be characterized as one that took the institution of the caliphate from revolution to autocracy, and thence to disintegration and the concomitant fragmentation of the umma – that, at least, was the political framework within which radical transformations in the society and religious identity of Muslims transpired. What follows is a very brief sketch of some of the political highlights of the period, from the accession of al-Saffah, the first cAbbasid caliph, to the end of the tenth century.

In 762, al-Mansur, the second cAbbasid caliph, established a new capital for the empire in Iraq. The foundation of Baghdad, which al-Mansur actually called the “City of Peace,” reflected the growing tensions between the cAbbasids and the supporters of cAli's family, who were especially strong in Kufa, the principal Muslim settlement in Iraq which had served as the cAbbasid caliphs’ first capital. In many ways the city can stand as a metaphor for the character of the Islamic empire in this period, and for its greatness. The city, like the state of which it was the capital, was an ambitious enterprise. Much of it was occupied by and organized around explicitly imperial structures – palaces, gardens, vast reception halls – with a domed room housing the caliph's throne at the very center.

There is an old and well-developed heresiographical tradition in Islam. In response to reported dicta of the Prophet to the effect that his community would ultimately fracture into seventy-three (or seventy-two – the precise number varies with the reports) sects, theologians composed extensive treatises identifying the different religious groups and their relation to what the authors identified as legitimate Islam. Of course, as Islam has generally lacked an institutional authority constituted to make definitive pronouncements about matters of religious interest, what exactly the parameters of Islamic legitimacy embrace has tended to shift with the perspective of the viewer. The very terms “heresy” and “orthodoxy” (since the former cannot exist without the latter) as they are used in the Western Christian tradition are in some ways misleading in an Islamic context. On the other hand, the flexibility of the Islamic tradition has not prevented Muslims from fervent denunciations of those who (in their view) falsely claim the mantle of Islam. Indeed, the very lack of an authoritative institutional structure has probably helped to make conflicts over issues of religious identity sharper and more intense. Those conflicts were central to the history of Islam in the first century and a half of the new religion when, as we have seen, it only gradually carved out for itself an identity distinct from those of the earlier Near Eastern monotheisms.

In a book entitled Medieval Islam: A Study in Cultural Orientation, the Orientalist G. E. von Grunebaum asserted that the difference between medieval and modern societies, in the Islamic Near East as well as in Christian Europe, could be located most clearly in the shift in the locus of social identity. In the “Middle Ages,” he said, religious affiliation was the fundamental component of an individual's outlook, and of his and others' understanding of his place in the world. Only on a secondary level would he think of himself as rooted in a local society, or bound to some local center of political power. And only as an afterthought might he think of himself as part of a larger national or ethnic community. “The gradual reversal of the strength of these loyalties,” von Grunebaum argued, “marks the close of the Middle Ages.” Like many grand themes of Orientalist scholarship, von Grunebaum's observation perhaps contains a certain truth. But it also obscures a great deal through over-simplification. As we have seen, the question of religious identity, even as late as what we have called the “Middle Period,” was in fact a complex matter, and its complexity did not evaporate in the years after its close. Moreover, as anyone who reads a newspaper in the early twenty-first century knows, it is by no means clear that, of religious, local, and national identities, the former has been relegated to a place of insignificance.

Muhammad's unexpected death in 632 threw his community into confusion, and the difficulty it had in simply surviving speaks volumes about the absence at this stage of a fully formed religious identity, or at least of the failure of that identity to claim the unremitting allegiance of many of those who had joined it. A number of points of tension surfaced, but probably no set of issues proved so contentious to Muslim posterity, or so critical in subsequent definitions of what it meant to be a Muslim, than that surrounding the question of leadership after the Prophet's death. Consequently this terrain is particularly dangerous for the historian. According to the standard Sunni account, Muhammad's friend and father-in-law Abu Bakr prevented the Medinese Muslims setting themselves up as a separate community from Muhammad's close circle of Meccan companions, and then was named through acclamation as the first caliph, or successor, of the Prophet. Shicis, however, have a different recollection, and stress a story according to which Muhammad, sometime prior to his death, identified his cousin cAli as his presumptive heir. Of course both the Sunni and Shici recollections in fact reflect the fully formed expectations of the later sectarian groups and political parties.

It is virtually certain that Muhammad had not made arrangements for the organization and leadership of his community before his death.