Fordham University


Home | Ancient History Sourcebook | Medieval SourcebookModern History Sourcebook | Byzantine Studies Page
Other History Sourcebooks: African | East Asian | Global | Indian | IslamicJewishLesbian and Gay | Science | Women's



Subsidiary SourcebooksAfricanEastern AsianGlobalIndianJewishIslamicLesbian/GayScienceWomen

Special ResourcesByzantiumMedieval WebMedieval NYC
Medieval MusicSaints' Lives
Ancient Law
Medieval Law
Film: Ancient
Film: Medieval
Film: Modern
Film: Saints

About IHSPIJSP Credits

Patricia Crone. Meccan Trade and the Rise of Islam.

Princeton University Press. 1987. Beginning with pg. 231.


Having unlearnt most of what we knew about Meccan trade, do we find ourselves deprived of our capacity to explain the rise of Islam? If we take it that trade is the crucial factor behind the appearance of a prophet in Arabia, the spread of his message there, and the Arab conquest of the Middle East, then the answer is evidently yes. But, in fact, Meccan trade cannot be said ever to have provided a convincing explanation for any of these events.

The view that Meccan trade is the ultimate cause of the rise of Islam is Watt's. The reader may begin to feel that there has been enough polemic against Watt in this book, and this is a view which its author shares. But to disagree with the conventional account is of necessity to disagree with thefons and origo of this account: throughout the present work the reader can treat the name of Watt as a shorthand for "early Islamic historians in general" and take polemical attention as a backhanded compliment to him. It is thanks to the enormous influence exercised by his work that a general appraisal of the theories that dominate the field takes us back to Watt for a final round.

According to Watt, the Qurashi transition to a mercantile economy undermined the traditional order in Mecca, generating a social and moral malaise to which Muhammad's preaching was the response.I This hypothesis is clearly weakened by the discovery that the Meccan traded in humble products rather than luxury goods, but it is not necessarily inv-alidated thereby. Even so, however, there are other reasons why it should be discarded.

In the first place, it is unlikely that so brief a period of commercial the nineteenth century, for example, the town of Ha'il enjoyed a meteoric rise to commercial importance, comparable to that described for Mecca, without there being any indication of a correspondingly swift breakdown of traditional norms. Why should there have been? It takes considerably more than a century of commercial success to undermine the tribal order of a population that has been neither uprooted nor forced to adopt a different organization in connection with its economic activi- ties. Caravan trade is not capitalist in any real sense of that word, and Watt's vision of the Meccans as financiers dedicated to a ruthless pursuit of profit occasionaly suggests that he envisages them as having made a transition to the twentieth century .

In the second place, the evidence for a general malaise in Mecca is inadequate. According to Watt, the Qur'an testifies to an increasing awareness of the difference between rich and poor and a diminishing concern on the part of the rich for the poor and weak even among their own kin, orphans in particular being ill-treated; further, the Qur'anic stress on acts of generosity implies that the old ideal of generosity had broken down to the point that the conduct of the rich would have been looked upon as shameful in the desert, while at the same time the Qur'anic emphasis on man's dependence on God suggests that the Meccans had come to worship a new ideal, "the supereminence of wealth." But the Qur'an does not testify to an increasing awareness of social dif- ferentiation or distress: in the absence of pre-Qur'anic evidence on the subject, the book cannot be adduced as evidence of change. And charges of excessive attachment to wealth and neglect of others, especially the poor and the weak, are standard items in the repertoire of monotheist preachers, as is the theme of man's dependence on God: how different would Muhammad's preaching have been, one wonders, if he had begun his career in Medina, or for that matter elsewhere? It is not very likely that there should be a one-to-one correspondence between the objective factors that led to the appearance of a prophet in Arabia and Muhammad's subjective perception of his mission: prophets are heirs to a prophetical tradition, not to a sociological habit of viewing their society from outside.

Leaving aside the Qur'an, then, to what extent does the tradition corroborate Watt's diagnosis? Viewed as pagan enemies of Islam, the Meccans are accused of neglect of kinship ties and other protective relationships, as well as a tendency for the strong to "eat" the weak. But viewed as proto-Muslims, they are praised for their harmonious relations. The conduct of trade in particular is supposed to have been characterized by cooperation between rich and poor; indeed, by the time of the rise of IsIam there no longer were any poor. Both claims, of course, merely illustrate the point that what the tradition offers is religious interpretation rather than historical fact. If we go by the overall picture suggested by this tradition, there is, however, no doubt that Watt's diagnosis is wrong. In social terms, the protection that Muhammad is said to have enjoyed from his own kin, first as an orphan and next as a prophet, would indicate the tribal system to have been intact, as Watt himself concedes, adding that the confederate status of foreigners in Mecca would indicate the same. It was, as Abu Sufyan said, Muhammad who disrupted traditional kinship ties with his preaching. From the point of view of morality, traditional tribal virtues such as generosity were both esteemed and practised: wealthy Meccans such as 'Abdallah b. Jud'an would have been astonished to learn that their conduct would have been looked upon as dishonourable in the desert.

l In religious terms, the Meccans are depicted as zealots on behalf of their pagan shrine as well as devotees of a string of other deities by whom they swore, after whom they named their children, and whom they took with them in battle against the Muslims. Watt interprets the violations of the haram during the ars of Fijar as "probably a sign of declining belief." But obviously ho]y places and months were violated from time to time: Muhammad himelf is supposed to have violated a holy month without having lost belief in it and if the Meccans had come to regard such violations as unobjectionable, they would hardly have referred to the wars in question as huru-b al-ija-r, "the sinful wars. " The fact that the Meccans carried their pagan deities with them into battle does not mean that "the remnants of pagan belief in Arabia were now at the the level of magic'' we are hardly to take it that the remnants of Islam were similarly at the level of magic by the time of the battle of Siffin, in which the soldiers are said to have carried Qur'ans with them; or that Christians who wear crosses are mere fetishists. Watt concedes that "in view of the opposition to Muhammad at Mecca it is conceivable that some small groups there -- perhaps those specially concerned with certain religious ceremonies -- had a slightly higher degree of belief." But a slightly higher degree of belief among small groups with possibly special functions scarcely provides an adequate explanation for the magnitude of this opposition.

The fact is that the tradition knows of no malaise in Mecca, be it religious, social, political, or moral. On the contrary, the Meccans are described as eminently successful; and Watt's impression that their success led to cynicism arises from his otherwise commendable attempt to see Islamic history through Muslim eyes. The reason why the Meccans come across as morally bankrupt in the sources is not that their traditional way of life had broken down, but that it functioned too well: the Meccans preferred their traditional way of life to Islam. It is for this that they are penalized in the sources; and the more committed a man was to this way of life, the more cynical, amoral, or hypocritical he will sound to us: Abu Sufyan cannot swear by a pagan deity without the reader feeling an instinctive aversion to him, because the reader knows with his sources that somebody who swears by a false deity is somebody who believes in nothing at all.

In the third place, the Watt thesis fails to account for the fact that it was in Medina rather than in Mecca that Muhammad's message was accepted. In Mecca, Muhammad was only a would-be prophet, and if he had stayed in Mecca, that is what he would have remained. This makes sense, given the general absence of evidence for a crisis in Mecca: if Muhammad himself had conceived his monotheism as a blueprint for social and moral reform in Mecca, he must soon have changed it into something else. It was outside Mecca, first in Medina and then elsewhere in Arabia, that there was a market for his monotheism: the Meccans had to be conquered before they converted. It follows that the problems to which Muhammad's message offered a solution must have been problems shared by the Medinese and other Arabs to the exclusion of the Meccans. In short, they were problems that had nothing to do with Meccan trade.

Is this surprising? Ultimately, the Watt thesis boils down to the proposition that a city in a remote corner of Arabia had some social problems to which a preacher responded by founding a world religion. It sounds like an overreaction. Why should a blueprint for social reform in Mecca have caused the entire peninsula to explode? Clearly, we must concentrate on such factors as were common to Arabia, not on those that were peculiar to Mecca; the more unusual we consider Mecca to have been, the more irrelevant we make it to the explanation of the rise of Islam.

Watt is not, of course, unaware of the need to explain the success of Muhammad's message outside Mecca. But having linked its genesis with Meccan trade, he is forced to identify a second set of problems to account for its success in Medina; and having opted for problems arising from a transition to a settled life in Medina, he needs a third set of problems to account for its spread in Arabia at large, this time opting for a general spiritual crisis: "there was a growing awareness of the existence of the individual in separateness from the tribe, with the consequent problem of the cessation of his individual existence at death. What was the ultimate destiny of man? Was death the end?''

The changes and transitions in question would, however, seem to be largely of Watt's own making. As regards the feuds with which the Medinese had to cope, they did not arise from a transition to settled life, but simply from settled life in general. It is a mistake to regard tribal organization as peculiar to nomads and sedentarization as necessarily leading to alternative forms of organizations, norms, and beliefs. The settled people of pre-oil Arabia were tribally organized, like the Bedouin, and they subscribed to much the same norms and beliefs; both settled and nomadic life was typically life under conditions of statelessness. Watt is right that sedentarization created a greater need for authority, but the material resources required for the creation and maintenance of stable state structures simply were not available. Accordingly, Arabian settlements were usually plagued by feuds; those characteristic of Medina in the sixth century would appear to have been no different from those characteristic of most Arabian settlements, including Medina, in the nineteenth. The feuds to which Muhammad offered a solution were a constant of Arabian history, not a result of change. It was only the solution that was new. The novelty of the solution lay in the idea of divinely validated state structures; and it was Muhammad's state, not his supposed blueprint for social reform, which had such powerful effect on the rest of Arabia.

As for the spiritual crisis, there does not appear to have been any such thing in sixth-century Arabia, in the sense normally understood. There is no feeling in Muhammad's biography of burning questions and long- debated issues finally resolved. Instead, there is a strong sense of ethnogenesis. The message of this biography is that the Arabs had been in the peninsula for a long time, in fact since Abraham, and that they had finally been united in a state. Muhammad was neither a social reformer nor a resolver of spiritual doubts: he was the creator of a people.

The impulse behind Watt's attempt to identify social changes and spiritual crises in Arabia comes from his conception of religion as a set of ultimate truths concerning the nature and meaning of life: what is the destiny of man? Is death the end? When religion is thus conceived, it usually takes a fundamental change in people's way of life and outlook to make them abandon their beliefs, and the process tends to be accompanied by pangs of conscience and spiritual pain. If we assume that the pre-Islamic Arabs shared this conception of religion, it follows from the rapid spread of Islam in the peninsula that there must have been a fundamental change -- which to most of us conjures up an image of socio- economic change -- with accompanying spiritual crisis. All we need to do then is to identify the nature of this crisis. The immense appeal of Watt's work on the rise of Islam rests on the fact that he thought along these very intelligible lines and came up with a socioeconomic change of the requisite kind: the Meccans were making a transition to a capitalist economy and losing their faith in the process. How very familiar; the Meccans were just like us. But an explanation that credits our own experience to a simple society is unlikely to be right. What sort of socio- economic change and spiritual crisis preceded the Israelite adoption of Yahweh? How much thought about the ultimate destiny of man went into the Icelandic adoption of Christianity by vote of parliament? None, apparently. Similarly in the case of Islam. Islam originated in a tribal society, and any attempt to explain its appearance must take this fact as its starting point.

What, then, was the nature of religion in tribal Arabia? The basic point to note here is that tribal gods were ultimate sources of phenomena observable in this world, not ultimate truths regarding the nature and meaning of life. More precisely, they were ultimate sources of all those phenomena that are of great importance in human society, but beyond direct human control: rain, fertility, disease, the knowledge of soothsayers. the nature of social roups, and so forth. They were worshipped for the practical services they could render in respect of these phenomena. As Wellhausen noted, they differed from more spirits only in that they had names and cults devoted to them; without a name a deity could not be invoked and manipulated, and he very object of the cult was to make the deity exercise its power on behalf of its devotees. "Ilaha, regard the tribe of Rubat (with benevolence)," as a third-century inscription says. This being so, tribal gods neither required nor received emotional commitment, love, or loyalty from their devotees. Thus a famous story informs us that "in the days of paganism Banu Han-lfa had a deity made of dates mixed with clarified butter. They worshipped it for a long time. Then they were hit by a famine, so they ate it.'' In much the same pragmatic spirit a modern Bedouin vowed half of whatever he might shoot to God. Having shot some game, he ate half, left the other half for God and departed; but feeling hungry still, he crept back and successfully stole God's part, and ate it, boasting that "God was unable to keep his share, I have eaten his half as well as mine.'' Now if hunger could make a tribesman eat or cheat his god without remorse, then it is obvious that practical needs could likewise make him renounce or exchange this god for another without compunction. "We came to Sa'd so that he might get us together, but Sa'd dispersed us; so we have nothing to do with Sa'd," as a pre-Islamic tribesman is supposed to have said in disgust when his idol scared his camels away . In much the same fashion a whole tribe abandoned its native gods for Christianity when its chief was cured of childlessness by a Christian monk. And the numerous other Arabs who found the medical facilities of the Christian God suffi ficiently impressive to adopt Him as their own are unlikely to have found the act of conversion any more difficult. A god was, after all, no more than a powerful being, and the point of serving him was that he could be expected to respond by using his power in favour of his servants. A modern Tiyaha tribesman who was being swept away by a flood screamed in great rage at God, "I am a Tihi! I am a Tihi! God, if you don't believe it, look at the brand on my camels." Obviously, if a deity was so inefficient as to unleash floods against his own followers, or so weak as to be unable to protect them from famine, or to keep his own share of some game, or to work miraculous cures, then there was reason to eat, cheat, abuse, denounce, or abandon him. "What were two little words?" as Doughty was asked on one of the numerous occasions on which attempts were made to convert him, "pronounce them with us and it shall do thee no hurt." The idea that a believer might be personally committed to a deity, having vested the ultimate meaning of his life in it, did not occur to any of these men. Those who tried to convert Doughty were evidently thoroughly committed to Islam, but not to Islam as a saving truth of deep significance to them as individuals. Convert, settle, and we will give you palm trees, as they told Doughty; in other words, be one of ours. Allah was a source communal identity to them, not an answer to questions about the hereafter. And the numerous people who tried to convert him or to penalize him for his Christianity on other occasions were likewise people who neither knew nor cared much about Islam as a saving truth, but who were outraged by his open denial of the God who validated their society. Now, just as tribal gods did not articulate great spiritual truths, so also they were not deeply entrenched in everyday life. Pre-Islamic (or for that matter pre-modern) Arabia was strikingly poor in mythology, ceremonial, ritual, and festivals. Religious life was reduced to periodic visits to holy places, stones, and trees, to sacrifice and consultation of diviners; most Bedouin managed with even less than that; and these practices were not closely associated with belief in specific gods. The great annual pilgrimage was apparently not conducted in the name of any one deity, and the remaining practices could effortlessly be switched from one deity to another; all survived into modern times, among Muslim and Christian tribesmen alike. Renouncing one god for another thus did not require any change in either outlook or behaviour, unless the new deity carried with him a behavioural programme anti- thetical to tribal norms. In principle, the Christian deity did carry with him such a programme, though in practice the holy men active in Arabia were in no position to ensure that ccnversion amounted to more than two little words. But the Muslim deity did not. On the contrary, he en- dorsed and ennobled such fundamental tribal characteristics as mili- tance and ethnic pride. Despite the Qur'anic suspicion of Bedouin, it was only on the development of classical Islam in the Fertile Crescent that the celebrated antithesis between muruwwa and dm, manliness and religiosity, emerged.

It is thus clear that the mass conversion of Arabia to Islam does not testify to any spiritual crisis, religious decadence, or decline of pagan belief. Indeed, in behavioural terms, the better part of Arabia was still pagan in the nineteenth century. What the mass conversions show is that Muhammad's God had something very attractive to offer here and now. When Sa'd, the pre-Islamic deity, scared away the camels of his devotees, the latter concluded that "Sa'd is just a rock": the power that he was supposed to have exercised had proved unreal. But when Muhammad established himself, they concluded that "Allah is great." The Arabs converted to Islam because Allah was a greater power than any other spirit endowed with a name and a cult so far known in Arabia, and the problem is not the ease with which they could convert, but the inducement. What was it that Allah had to offer?

What he had to offer was a programme of Arab state formation and conquest: the creation of an umma, the initiation of jihad. Muhammad was a prophet with a political mission, not, as is so often asserted, a prophet who merely happened to become involved with politics. His monotheism amounted to a political programme, as is clear not only from non-Muslim accounts of his career, but also from Ibn Ishaq.

Thus Ibn Ishaq informs us that the turning point of Muhammad's career as a prophet came when he began openly to attack the ancestral gods of Quraysh and to denounce his own ancestors. This was a turning point because in so doing, he attacked the very foundations of his own tribe; and it was for this that he would have been outlawed or killed if his own kinsmen had not heroically continued to protect him -- not for the threat that his monotheist preaching allegedly posed to the pagan sanctuary or Meccan trade. He was, after all, no more than a local eccentric at the time, and Quraysh were quite willing to tolerate his oddities, including his minor following, as long as he confined his teaching to abstract truths about this world and the next. But they were not willing to tolerate an attack on their ancestors. By his they were outraged,and quite rightly so: a man who tries to destroy the very foundation of his own community is commonly known as a traitor. But Muhammad would scarcely have turned traitor without some vision of an alternative community. In denouncing his own ancestors, he had demonstrated that his God was incompatible with tribal divisions as they existed; and this incompatibility arose from the fact that his God, unlike that of the Christians, was both a monotheist and an ancestral deity. Allah was the one and only God of Abraham, the ancestor of the Arabs; and it was around ancestral deities that tribal groups were traditionally formed. It follows that it was around Allah, and Allah alone, that the Arabs should be grouped, all the ancestral deities that sanctioned current divisions being false. If we accept the traditional account of Muhammad's life, Muhammad was thus a political agitator already in Mecca, and it was as such that he offered himself to other tribes. "If we give allegiance to you and God gives you victory over your opponents, will we have authority after you?" an 'Amin is supposed to have asked, fully aware that ac- ceptance of Muhammad was acceptance of a ruler with ambitious plans. It was also as such, not merely as an otherworldly arbitrator, that he was accepted in Medina.

Assuming that Medinese society was rent by feuds, as opposed to united by proto-kings, it is not difficult to explain why the Medinese should have been willing to experiment with Muh. ammad's political programme; but given that Arabia had never been politically united before, and was never to be so again, it is certainly extraordinary that he and his successors should have succeeded in bringing this unification into effect. Why did the Arabs in Muhammad's time find the vision of state structures and unification so attractive?

It is customary to invoke Meccan trade in answer to this question. Quraysh, we are told, had in effect united most of Arabia already, numerous tribes having acquired an interest in the conduct of Meccan trade as well as in the maintenance of the sanctuary; inasmuch as the interests of Mecca and Arabia at large had come to coincide, Muhammad's conquest of Mecca amounted to a conquest of most of Arabia, though the process of unification was only to be completed on the suppression of the ridda. But though it is true that the suppression of the ridda completed the process, this is not an entirely persuasive explanation. If the interests of Mecca and the Arabs at large had come to coincide, why did the Arabs fail to come to Mecca's assistance during its protracted struggle against Muhammad? Had they done so, Muhammad's statelet in Medina could have been nipped in the bud. Conversely, if they were happy to leave Mecca to its own fate, why should they have hastened to convert when it fell? In fact, the idea of Meccan unification of Arabia rests largely on Ibn al-Kalbl's tl-tradition, a storyteller's yarn. No doubt there was a sense of unity in Arabia, and this is an important point; but the unity was ethnic and cultural, not economic, and it owed nothing to Meccan trade.40 Muhammad's success evidently had something to do with the fact that he preached both state formation and conquest: without conquest, first in Arabia and next in the Fertile Crescent, the unification of Arabia would not have been achieved. And there is no shred of evidence that commercial interests contributed to the decision, on the part of the ruling elite, to adopt a policy of conquest; on the contrary, the sources present conquest as an alternative to trade, the reward of conquest being an effortless life as rulers of the earth as opposed to one as plodding merchants. Nor is there any evidence that the collapse of Meccan trade caused an "economic recession" that contributed to the enthusiasm with which the tribesmen at large adopted this policy.44 It is, of course, legitimate to conjecture that trade may have played a role, but there is no need for such conjecture. Tribal states must conquer to survive, and the predatory tribesmen who make up their members are in general more inclined to fight than to abstain. "How many a lord and mighty chief have our horses trampled under foot . . . we march forth to war, the ever renewed, whenso it threatens," one pre- Islamic poet boasts. "We slew in requital for our slain an equal number lof them], and [carried away an uncountable number of fettered prisoners . . . the days have thus raised us to be foremost with our battles in warfare after warfare; men find in us nothing at which to point their finger of scorn," another brags. "When I thrust in my sword it bends almost double, I kill my opponent with a sharp Mashrafi sword, and I yearn for death like a camel overful with milk," a convert to Islam announced. Given that men of this kind constituted Muhammad's following, we do not need to postulate any deterioration in the material environment of Arabia to explain why they found a policy of conquest to their taste. Having begun to conquer in their tribal homeland, both they and their leaders were unlikely to stop on reaching the fertile lands: this was, after all, where they could find the resources which they needed to keep going and of which they had availed themselves before. Muhammad's God endorsed a policy of conquest, instructing his believers to fight against unbelievers wherever they might be found; and if we accept the testimony of non-Muslim sources, he specifically told them to fight the unbelievers in Syria, Syria being the land to which Jews and Arabs had a joint right by virtue of their common Abrahamic descent. In short, Muhammad had to conquer, his followers liked to conquer, and his deity told him to conquer: do we need any more?

The reason why additional motives are so often adduced is that holy war is assumed to have been a covr for more tangible objectives. It is felt that religious and material interests must have been two quite different things --an eminently Christian notion; and this notion underlies the interminable debate whether the conquerors were motivated more by religious enthusiasm than by material interests, or the other way round. But holy war was not a cover for material interests; on the contrary, it was an open proclamation of them. "God says . . . 'my righteous servants shall inherit the earth'; now this is your inheritance and what your Lord has promised you . . . ," Arab soldiers were told on the eve of the battle of Qadisiyya, with reference to Iraq; "if you hold out . . . then their property, their women, their children, and their country will be yours." God could scarcely have been more explicit. He told the Arabs that they had a right to despoil others of their women, children, and land, or indeed that they had a duty to do so: holy war consisted in obeying. Muhammad's God thus elevated tribal militance and rapaciousness into supreme religious virtues: the material interests were those inherent in tribal society, and we need not compound the problem by conjectu ing that others were at work. It is precisely because the material interests of Allah and the tribesmen coincided that the latter obeyed him with such enthusiasm.

The fit between Muhammad's message and tribal interests is, in fact, so close that there is a case for the view that his programme might have succeeded at any point in Arabian history. The potential for Arab state formation and conquest had long been there, and once Muhammad had had the idea of putting monotheism to political use, it was exploited time and again, if never on the same pan-Arabian scale. Had earlier adherents of Din Ibrahim seen the political implications of their own beliefs, might they not similarly have united Arabia for conquest? If Muhammad had not done so, can it be argued that a later prophet might well have taken his role? The conquests, it could be argued, turn on the simple fact that somebody had an idea, and it is largely or wholly accidental that somebody did so in the seventh century rather than the fifth, the tenth, or not at all.

But the fact that it was only in the seventh century that the Arabs united for conquest on a pan-Arabian scale suggests that this argument is wrong. If we choose to argue otherwise, we must look for factors which were unique to Arabia at that particular time, not constants such as the feuds of Medina, and which affected the entire peninsula, not just a single city such as Mecca. Given the fit between Muhammad's message and tribal interests, the factors in question should also be such as to accentuate the perennial interests of tribal society rather than to undermine them in the style of Meccan trade as conventionally seen. There is only one development which meets all three specifications, and that is the foreign penetration characteristic of sixth- and early seventh-century Arabia.

As mentioned already, the Persians had colonies throughout eastern Arabia, in Najd, and in the Yemen, as well as a general sphere of influence extending from the Syrian desert to the Hijaz. The Byzantines had no colonists to the south of Tabuk, but their sphere of influence was felt throughout western Arabia from the Syrian desert where they had client kings to the Yemen where their Ethiopian allies ruled until they were ousted by the Persians. Muhammad's Arabia had thus been subjected to foreign rule on a scale unparalleled even in modern times: where the Persians had colonists and fire-temples, the British merely had Philby. The scale on which Muhammad's Arabia exploded is equally unparalleled, the nearest equivalent being that of the Ikhwan. It seems unlikely that the two phenomena were unrelated.

If so, how? One model can be eliminated at once. It is well known that empires tend to generate state structures among their barbarian neighbours thanks to the ideas that they provide, the material sources that they pass on, and the resentment that their dominance engenders; and having generated such state structures, they will usually become targets of conquest, too. This is the pattern known from Central Asia and Europe; but it is not the pattern to which Arabia conforms. There was no incipient growth of state structures at the expense of tribal ties in Arabia, not even in Mecca. Muhammad's state in Medina-was formed by a prophet, not a secular statesman, by recourse to religious authority, not material power, and the conquests were effected by a fusion of tribal society, not by its disintegation. If the imperial powers contributed to the rise of Islam, they must have done so in a different way.

An alternative hypothesis would be that Islam originated as a nativist movement, or in other words as a primitive reaction to alien domination of the same type as those which the Arab conquerors were themselves to provoke in North Africa and Iran, and which European colonists were later to provoke throughout the Third World. If we accept the testimony of the non-Muslim sources on the nature of Muhammad's teaching, this interpretation fits extremely well.

Nativist movements are primitive in the sense that those who engage in them are people without political organization. Either they are members of societies that never had much political organization, as is true of Muhammad's Arabia,.or they are drawn from these strata of society that lack this organization, as is true of the villagers who provided the syncretic prophets of Iran. They invariably take a religious form. The leaders usually claim to be prophets or God Himself, and they usually formulate their message in the same religious language as that of the foreigners against whom it is directed, but in such a way as to reaffirm their native identity and values. The movements are almost always millenarian, frequently messianic, and they always lead to some political organization and action, however embryonic; the initial action is usually militant, the object of the movement being the expulsion of the foreigners in question. The extent to which Muhammad's movement conforms to this description can be illustrated with reference to a Maori prophet of the 1860s who practically invented Islam for himself. He reputedly saw himself as a new Moses (as did Muhammad), pronounced Maoris and Jews to be descended from the same father (as were the Jews and their Ishmaelite brothers), and asserted that Gabriel had taught him a new religion which (like that taught to Muhammad combined belief in the supreme God of the foreigners with native elements (sacred dances as opposed to pilgrimage). He proclaimed, or was taken to proclaim, the Day of Judgment to be at hand (as did Muhammad). On that day, he said or was taken by his followers to say, the British would be expelled from New Zealand (as would the Byzantines from Syria), and all the Jews would come to New Zealand to live in peace and harmony with their Maori brothers (as Jews and Arabs expected to do in Syria). This, at least, is how his message was reported by contemporary, if frequently hostile, observers.8 And though he may in fact have been a pacifist, his followers were not. Unlike the followers of Muhammad, howcver, they fought against impossible odds.

Like the Maori prophet, Muhammad mobilized theJewish version of monotheism against that of dominant Christianity and used it for the self-assertion, both ideological and military, of his own people. It is odd that what appears to have been the first hostilc reaction to alien domination, and certainly the most successful, should have come in an area subject to Byzantine rather than Persian influence, that of the Persians being more extensive. But Jewish-Arab symbiosis in northwest Arabia could perhaps account for this: according to Sebeos, the Byzantine victimization of Jews played a crucial role in the birth of Muhammad's movement. In any case, Muhammad was not the only prophet in seventh-century Arabia, and two of his competitors, Musaylima and Aswad, were active in areas subject to Persian influence, the Yamama and the Yemen, respectively, while a third, Sajah, was sponsored by tribes known to have participated in the celebrated battle against the Persians at Dhu Qar. The fact that the resistance to Islam in Arabia was led by imitators of Muhammad rather than by representatives of traditional paganism is thus unlikely to mean that traditional beliefs and values had lost force in Arabia; on the contrary, Muhammad would seem to have hit upon a powerful formula for the vindication of those values. And this formula was, of course, likely to be used against Muhammad himself when he began his subjection of Arabia.

A more serious objection would be that the foreign presence is unlikely to have affected thc majority of Arabs very deepl!. Unlike the Maoris, who were losing their land to the British, they certainly cannot have felt that their entire way of life was under threat; and unlike the Berbers, they were not exposed to forced conversion. Nor are expressions of dissatisfaction with foriegn domination very common in the sources. There is, admittedly, no lack of anti-Persian feeling in the poetry triggered by the battle of Dhu Qar, which the Prophet supposedly described as the first occasion on which the Arabs obtained revenge from the Persians, the conquests (by implication) being the second. But in historical fact this battle may not have represented more than a short-term disagreement between the Persians and their Arab subjects. Still, there were some who felt that "the Arabs were confined between the lions of Persia and Byzantium," as Qatada said in a passage contrasting the ignominious state of the Arabs in theJahiliyya with the grandeur achieved on the coming of Islam. "Other men trampled us beneath their feet while we trampled no one. Then God sent a prophet from among us . . . and one of his promises was that w e should conquer and overcome these lands," as Mughlra b. Shu'ba is supposed to have explained to a Persian commander. In general it is acknowledged that the Arab conquests were nothing if not "an outburst of Arab nationality."

To what extent, if at all, the nativist model can be applied to the rise of Islam is for future research to decide; no doubt there are other ways in which the interaction between Arabs and foreigners could be envisaged. But it is at all events the impact of Byzantium and Persia on Arabia that ought to be at the forefront of research on the rise of the new religion, not .leccan trade. Meccan trade may w ell turn out to throw some light on the mechanics behind the spread of the new religion; but it cannot explain why a new religion appeared at all in Arabia or why it had such massive political effect.