A review of In the Shadow of the Sword: The Birth of Islam and the Rise of the Global Arab Empire, by Tom Holland

The subtitle of Tom Holland's book In the Shadow of the Sword indicates its primary subject. But it is neither his only subject nor, more surprisingly, even the one to which he devotes the greatest number of pages. Rather, after introducing the subject of the birth of Islam and the rise of the global Arab empire, he spends the majority of his book weaving narrative and interpretative histories of antiquity—both classical and Near Eastern.

In his treatment of antiquity, Holland, an award-winning British historian, focuses mainly on Rome and Persia, and on the role of religious movements within each. Regarding Rome he considers the emergence of Christianity out of Judaism against the background of ancient paganism, the many early intra-Christian controversies, and Christianity's eventual establishment as Rome's state religion. Similarly, his history of successive Persian empires concentrates on the various religious movements—e.g., the religions of Zoroaster and Mazdak—that played a role in political developments.

Holland's ancient history culminates in the description of the last great conflict between Rome and Persia in the early part of the 7th century. Only then does he return to his ostensible primary subject: "the birth of Islam," which occurred shortly after this Roman-Persian conflict, and led very quickly to the establishment of the first great Muslim empire, "the global Arab Empire" of his subtitle. The rise of this empire and the eventual rule of its first great dynasty—the Umayyad—necessarily alter Holland's historical framework. For the first Muslim empire defeated, destroyed, and then incorporated the Persian Empire. Though it did not similarly destroy the late Roman or Byzantine Empire—two early attempts to take Constantinople failed—it did deal it a grievous blow, depriving it of the greater part of its territory in the Levant and North Africa. The Muslim empire replaced Persia as Rome's most potent, if not only, enemy and was for many centuries its superior as the most powerful political force in the world.

Holland's account of pre-Muslim history is elegantly told, as one would expect from this distinguished classical historian. It will offer much pleasure and profit to the reader. But what is it about Islam, its birth, and expansion that accounts for Holland's unusual approach to the subject? He has several answers, of which two predominate.

* * *

First, he is interested in the emergence of universal empire as an object of political endeavor, including the implicit notion that this would entail the proper ordering of humanity as a whole. Universal empire was—perhaps from the very beginning—the manner in which Islam defined its objectives. This is one reason Holland has left the confines of ancient history.

But Islam was preceded in this ambition by earlier empires—especially Rome and Persia—and may be reasonably seen in their context. Indeed, he understands the first Persian Empire, founded by Cyrus the Great in the 6th century B.C., to be the first concrete example of the ambition for universal empire and a comprehensive ordering of mankind. In this Cyrus went beyond his Near Eastern imperial predecessors. Although his own activities were limited to Asia, his ambitions were carried forward by his successors who sought to invade and conquer first Egypt—that is, Africa—and thereafter Greece and Europe. This "eastern" initiative prompted not only "western" self-defense—most famously at Salamis, Marathon, and Thermopylae; it also inspired Western imperial efforts beginning with Alexander the Great and culminating in the Roman Empire. By the era of what has come to be called, thanks to the work of Princeton historian Peter Brown, late antiquity—the 5th, 6th, and 7th centuries—we can see that neither Rome nor Persia, neither West nor East, had managed to prevail. Their ongoing and now millennial struggle took place along a relatively fixed border running from Egypt through Syria and the Levant.

This struggle was accompanied in due course by intellectual and especially religious developments that expressed and justified universal ambitions. The rise of Islam and its spectacular early military success continued, subsumed, and redefined that struggle. By Holland's lights, in so doing, it ended antiquity and created a new world order that in some essentials continues to the present day. Holland's history of antiquity is meant to clarify the dynamic of Islam's momentous struggle and its results. At the same time, and insofar as the emergence of Islam represents the culmination of the struggles of antiquity, it clarifies their dynamics as well.

* * *

But then we need clarity about the origins of Islam. Holland's second reason for recourse to antiquity derives from the difficulty of achieving such clarity. According to a venerable view this difficulty should not and did not exist with Islam as it did with other earlier religions. Islam's rise in the 7th century—an era in which it was supposed that myth no longer prevailed as the source of historical evidence and knowledge—meant that the facts about its origins were more readily accessible than those of earlier religions and polities. This accorded more or less with Muslim self-understanding, which put forward the Koran as a contemporaneous record of God's revelations to Muhammad, comprising teachings and instructions for the ordering of the Muslim polity he established—the polity which rapidly became a vast empire after his death. This was supplemented by a classical biography of Muhammad as well as the voluminous literature known as Hadith, which collected extra-Koranic sayings and doings of Muhammad as ostensibly reported by his companions, eyewitnesses to these events, who are known in Muslim tradition as as-Salaf as-Salih, the virtuous ancestors. In due course, Muslim historians arose who mined all this material to provide an account that met historical standards.

In recent years some non-Muslim scholars of Islam have called this understanding into question. The most immediate problem concerns the antiquity and thus reliability of the primary Islamic texts beginning with the Koran. The actual written texts presently available can not be dated earlier than the second century of Islam or still later. Indeed, the only strictly contemporaneous written accounts come from the rivals Islam defeated, especially Christian sources, and even these provide only limited information.

Hence several contemporary scholars have declined to see the Islamic texts as reliable evidence of Islam's origins. They have been inclined to see them, instead, as reflections of the interpretation of Islam's founding that arose in the course of the establishment of the empire, reflected its experience and requirements, and was then projected backward. This inclination is reinforced by the fact that the documents themselves, including the Koran, are not without considerable ambiguities about important matters—for example, the role of Mecca in early Islam, which is hardly mentioned in the Koran.

To be sure, subsequent Muslim tradition proposed resolutions of many of these ambiguities. But scholars ask whether these resolutions did not result from the incorporation of subsequent history, a doubt reinforced by the fact that Muslim historical and religious tradition itself is not unitary, above all due to the dispute between Sunnites and Shiites. Indeed, the dispute between them concerns some of the earliest events in Islam: the succession to Muhammad, and the principles that were or should have been applied to choose his successor.

* * *

Holland's account is heavily informed by contemporary scholarship as he indicates in the opening chapter of his book. He provides a description of the problem of the origins of Islam as he understands it as well as a very useful account of the work of well-known Islamic scholars and their admittedly diverse views of the matter, which will be very helpful to readers interested in the contemporary scholarly debate.

Holland, like some others, looks to shed light on the problem by coming to a proper understanding of the circumstances that prevailed on the eve of Islam's founding. This amounts to an attempt to understand Islam as a product of those circumstances—but only in part. For he does not deny that the founding of Islam entailed a distinctive teaching that was not simply reducible to its historical environment.

What are the results of this approach? Many and varied—too many and too varied to be fully addressed here. A few must suffice.

First, Holland, unlike some others, is inclined to credit the contemporaneity of much of the Koran and to regard it as providing the most reliable insight into Muhammad's experience, thought, and action. But even so, there remain ambiguities in the Koran's account, which he tried to resolve through proposals both historical and substantive.

Historically, for example, Holland adopts the view that Muhammad's point of origin and early career lay not in Mecca and the southerly Arab region but further north in the Arab region forming the borderland between the Roman and Persian empires. This region became divided between two Arab kingdoms: the Ghassanids were the allies of the Byzantines; the Lakhmids were allies of the Persians. More precisely, Muhammad hailed from the Ghassanid region whose Arabs were Christian. This provided him with ample knowledge of Christian teachings and of the disputes that raged among Christians. Subsequently Muhammad made his famous Hijrah or emigration to the more northerly city of Medinah (this Holland sees no reason to disbelieve) where, several Medinah tribes being Jewish, he also became familiar, if he was not already, with Jewish teachings.

* * *

This geographical reorientation reinforces, though it does not altogether explain, another more substantive proposal Holland offers regarding Muhammad's teaching about God. Here he focuses on the characteristic Koranic condemnation of those who would sully the pure unity of God through shirk—the association of partners with God. Typically this is understood to be a condemnation of a variety of apparently different things: idolatry, the ascription of divinity to Jesus, and more generally the Christian Trinitarian teaching. Holland does not deny that the Koran condemns these as violations of God's unity. Indeed the Koran's repudiation of Jesus' divinity is one of its verses that can be attested to be early inasmuch as it was inscribed on the Dome of the Rock in Jerusalem, built in the 7th century. But Holland proposes that these condemnations apply to a larger pejorative category—that of "intermediaries." The Koranic condemnation is thus not only an assertion of God's pure unity but a repudiation of the habit of seeking intermediation between God and man. He tends to regard this repudiation as a distinctive Islamic teaching, one that would apparently continue the universalistic dynamic of late antiquity, but would purify it.

A potential corollary of this proposal might be seen in another of the terms of Koranic condemnation. The Koran frequently condemns "the dividers," meaning Jews and Christians, whose separate existence shows that the ostensible community of believers in the one true God is not united. This, the Koran suggests, is a fatal flaw in the contemporary understanding of the requirements of monotheism, a flaw requiring rectification via the Koran and the establishment of Islam. Holland does not address this Koranic preoccupation, which is somewhat curious since it too points in a universalistic direction; more particularly, it points to universal empire, a theme central to his book.

In any event, Holland's emphasis on the essential character of the critique of intermediation would seem implicitly to raise another question: the theological-political structure of the early Muslim empire and its successors, the tradition of caliphal rule and the rise of Muslim jurists and other religious scholars as an authoritative class. He devotes the rest of his account of Islam to a description and interpretation of the first Muslim empire down to its end in 750, when the Umayyad dynasty was supplanted by the Abbasid.

* * *

As is well-known, the creation of the Muslim empire was the product of remarkably swift, expansive conquests. In less than a century, the empire stretched from the Pyrenees in Europe to the Indus in South Asia. This success was nonetheless attended by three major internal difficulties: the first concerned disputes about who were the empire's legitimate rulers, disputes that generated several civil wars and ultimately the Sunni-Shiite quarrel; the second derived from the vast numbers of non-Muslims who came to be subject to Muslim rule. According to Islam's own universal principles, its non-Muslim subjects should have been welcome into the Muslim fold. But initially entry was often only grudgingly granted and even when it was, the new Muslims were treated as second-class believers. Third, the governance of such a vast empire required institutions that went beyond those necessary for the initial polity established by Muhammad.

In Holland's view the relation between these factors—especially between the first two—served to produce the new Muslim order, which necessarily incorporated a new form of "intermediation." On the one hand the Umayyad dynasty, which had prevailed militarily and politically over its rivals, sought to reinforce its victory with greater legitimacy. Its most important method was to appropriate to itself the status Muhammad had enjoyed as God's representative on earth. Umayyad rulers thus adopted the title of the Deputy of God. But in so doing the dynasty set up a standard above itself—the standard of Muhammad—by which it might be judged and through which its authority could be contested. This standard provided an opportunity for other Muslims and especially new converts to make a claim to authority in their own right. One of their principal tools was the new Hadith literature and the new religious disciplines—including Muslim jurisprudence—built upon it. According to Holland, the vast majority of the practitioners of these disciplines were non-Arab Muslims—Persians, former Christians and Jews, and others—who regarded the new sciences as a form of "revenge" against their conquerors. In any case, these developments complicated the simplicity and unitary impulse of Islam's founding, and gave birth to the famous rivalry between the "men of the sword" and the "men of the pen."

Although these new religious disciplines did not deprive the Muslim empire of its extraordinary vitality, they complicated its efforts to eradicate the Roman or rather Christian Roman Empire known as Byzantium. By the time it finally achieved that with the Ottoman conquest of Constantinople in 1453, Christianity had other political bases. Subsequently, other and modern forms of universalism arose. Thus while the world may be informed by the vision of universalism, it remains divided still, as it was in antiquity, by that shared ambition.

* * *

Of course, Islam's adherents would largely reject the need for the kind of explanation of Islam that Holland offers or indeed the importance of the history that he adduces. For them Islam did not emerge from "history" but rather from a unique event—a decision by God to reveal His will and understanding to Muhammad. However inconvenient the lack of a copy of the Koran written in the time of Muhammad, or the lack of texts of the Hadith from their ostensible authors, the texts now available are faithful records of those events.

Today, too, Islam retains the notion of universal rule. This is especially true of the contemporary Muslim reform movement known as Islamism, which seeks to revitalize the prospect of universal empire by returning to the origins of Islam in their vaunted purity. Insofar as Islamism has been heavily informed by the Muslim trend known as Wahhabism, it places a special emphasis on the eradication of shirk—the impure and illegitimate veneration of "intermediaries" such as Sufi saints and Shiite imams. Oddly enough, there is a kind of agreement between these reformers and Holland.

But as he might retort, their efforts at purification and above all the achievement of "Muslim unity" remain bedeviled by "history," especially the history of the first Muslim empire. For although Sunni and Shiite Islamists agree about the need for unity and the identity of their main enemy—modern universalism—they disagree about the proper foundations of unity among Muslims. The current civil war in Syria is only the most visible expression of how fierce that disagreement is, and may continue to be.