Arabs, Persians, and the advent of the Abbasids reconsidered.

Daniel, Elton L.

The Abbasid revolution, as R. Stephen Humphreys has observed, is "one of the very few topics in Islamic historical studies" to have "engendered a substantial scholarly literature."(1) That literature continues to expand rapidly, as evidenced directly or indirectly by a wealth of recent publications, but the issues involved in understanding the advent of the Abbasid era remain far from settled.

Roberto Marin-Guzman, a talented Arabist at the University of Costa Rica, has provided a concise overview of a number of questions involved in studies of the Abbasid revolution: Who supported the revolt and why? How were the Abbasids able to organize and popularize their opposition movement? Why did it succeed in toppling the Umayyad regime when so many other efforts had failed? After some introductory remarks categorizing the participants in the revolt, critiquing the source materials, and characterizing the secondary literature, the author proceeds in three short chapters to analyze the role of Arab tribes in the revolt, taxation and conversion as sources of discontent, and finally the "popular dimensions" of the revolution. (It might be noted that the author has reorganized and revised much of the material from this book into an article which has recently appeared in Islamic Studies.)(2)

For the most part, Marin-Guzman's handling of these three themes is limited to summarizing the findings of earlier studies. In the case of the tribal element, he portrays disputes between antagonistic "Northern" and "Southern" tribes, aggravated by competition for land, money and administrative posts, with the consequence that Marwanid partiality for the northerners finally drove the southerners in Khurasan into the arms of the Abbasid conspirators. He also parallels the work of M. A. Shaban in depicting the Qays as a faction interested in war and expansion and Yaman as one favoring consolidation and assimilation (p. 22). In the second chapter, the mawali are introduced as one of the most important dissident elements to be aligned with the Abbasids. Marin-Guzman accepts the notion that the mawali in Khurasan and elsewhere were victims of systematic discrimination and that their major grievance was the continued extraction of jizya from them - especially since the desire to escape such taxation was, in the view of the author and many others, a primary reason for their religious conversion in the first place. The perception that this religioeconomic injustice fueled various sectarian movements which in turn became enmeshed in the Abbasid struggle leads Marin-Guzman to a consideration of the Mukhtariyya, Kaysaniyya, and Hashimiyya. He also speculates that other "anti-Umayyad" religious movements, such as the Kharijites, Qadariyya, and Mutazila, may have "contributed to the popular dimensions of the Abbasid revolution" (p. 68) but prudently concedes that this "needs, evidently, further research" (p. 70). The third chapter of Marin-Guzman's book discusses the organization and ideology of the Abbasid propaganda mission, the history of the revolt under the leadership of Abu Muslim, the degree to which these events represented revolutionary change, and the difficulties the Abbasids faced in developing a theory of legitimacy for their rule. There are no real surprises in any of Marin-Guzman's conclusions at the end of the volume: the Umayyads were brought down by a combination of tribal discord, mawali discontent, burdensome taxes, and rivalry between Syria and Iraq. The Abbasids succeeded where others failed because of their efficient, determined organizational. skills and their clever exploitation of popular discontent and religious ideology.

Marin-Guzman notes that his book "is addressed not only to specialists on the topic, but also to students specializing in Middle Eastern history, as well as to the general public interested in this period and the projections of these issues into the modern Middle East" (p. xi). He is undoubtedly most successful in meeting the second of these objectives. Since his treatment of the subject is largely conventional and derivative, it is also relatively uncontroversial. The book could certainly be used as an introductory text in undergraduate classes and seminars on Islamic history. Although one gets the impression from much of the recent historiography on the Abbasid revolt that the subject is indeed becoming entangled in modern politics and assumptions about the nature of revolutionary change in Islamic societies, it is not at all apparent what Marin-Guzman would consider to be "the projections of these issues into the modern Middle East" or what "the general public" would find of interest in this volume. As for specialists, they are the most likely to be disappointed by Popular Dimensions of the Abbasid Revolution, which is open to criticism in several respects. One concern relates to the author's approach to documentation - the book is extensively annotated but the citations are occasionally imprecise and of questionable accuracy. For example, he says that al-Maqrizi "precisely pointed out the role taxes played in the Abbasid propaganda and ideology" (p. 60) and cites the Nizawa'l-Takhasum as his source. However, he gives no page reference to either the Arabic text or Bosworth's translation, and it is hard to see how this text supports such an argument. The Niza is, in fact, little more than a catalogue of outrages perpetrated by both the Banu Umayya and the Banul-Abbas against Muslims, in general, and the family of the Prophet, in particular, and grievances about taxation figure little if at all in al-Maqrizi's inventory of insults. As far as the Abbasid revolution is concerned, al-Maqrizi makes clear in this text his opinion that it was nothing more than a crude grab for power over the Muslim community and that the Abbasids "took it over . . . through the agency of the Persians, the men of Khurasan, and acquired the caliphate by a combination of sheer force. . . . "(3) There are also a few instances where a footnote might be expected, but none is given. Thus the author refers (p. 54) to "a great number of Persian peasants from Khurasan" who converted to Islam and visited al-Hajjaj in 700 to request exemption from the jizya, but al-Hajjaj "ordered them back to Khurasan and did not recognize them as Muslims." This is said to be "recorded in the major Arabic sources," but Marin-Guzman offers no supporting reference. The accusation that al-Hajjaj, as a matter of policy, tried to keep peasants tied to the land and refused to exempt converts from jizya is commonplace, but it would be useful to know exactly which text or texts Marin-Guzman has in mind for this specific case. A second and much more important problem is not entirely the author's fault but rather a reflection of how rapidly thinking about these contentious problems is developing and changing. He has repeated much of the received wisdom about the nature of the Abbasid revolution, but those ideas look increasingly unconvincing and are rapidly being superseded by new research, some of which has appeared subsequent to publication of this book. Certainly, the author's depiction of the oppressed and exploited mawali ought to have been revised in light of Patricia Crone's article on the subject in the Encyclopaedia of Islam; his acceptance of the Abbasid nature and orientation of the early da wa should have reflected the doubts expressed in her article on al-Rida; and his assessment of the tribal factor would have benefited greatly from access to her article pointing out the implausibility of viewing the Qays/Yaman as political parties.(4) He cites Madelung's Religious Trends in Early Islamic Iran in his bibliography, but he has clearly not given due weight to its implications for understanding the sectarian dimensions of the revolt (by failing, for example, to include the Murjia and the Hanafis in his religious equation or to note the well-known history of Abu Muslim's relations with Shayban and the Kharijites). In short, there may be few new insights to be gained from Popular Dimensions of the Abbasid Revolution, but it can nonetheless serve as a good general introduction to this complex and fascinating subject.

Moshe Sharon's Revolt: The Social and Military Aspects of the Abbasid Revolution is a much more sophisticated study which, along with its predecessor, Black Banners from the East, has earned a place at the cutting edge of scholarship on the Abbasid revolution. The introduction to Revolt provides a brief comment on the nature of revolutions and some intriguing remarks, perhaps, inspired by the ideas of Suliman Bashear and Yehuda Nevo, about the "myth of the unified beginning" of Islam and its role in shaping the world-view of the Shia. Like Popular Dimensions, Revolt then examines the history of Arab "tribal loyalties" and "tribes at war" in Khurasan, culminating in the conflicts between Harith b. Surayj, Juday b. Ali al-Kirmani, and Nasr b. Sayyar which opened the political window of opportunity for the Abbasid revolutionaries. The chapters which follow describe in considerable detail the early stages of the Abbasid revolt, the organization of the Abbasid forces, the consolidation of the revolution in Khurasan, the "westward march to victory" of the new Abbasid army from Khurasan to Iraq, and finally the establishment of the Abbasid caliphate. An appendix looks in more detail at the background and consequences of Abu Muslim's military reforms and the development of the Abbasid army.

Sharon's theory of revolutionary behavior comes down to this modest proposition (pp. 15-16): revolutions consist of two stages, a clandestine period of preparation and a decisive military phase which is "simpler and more assured of its success if the revolution is able to infiltrate the ranks of the standing army of the regime and attract it wholly or partially to its side." He then demonstrates how the Abbasid revolution fits this pattern. The preparatory phase was described in detail in Black Banners, and Revolt merely recapitulates its argument that there was a long period of organizational activity by the anti-Umayyad da wa but the movement did not "become Abbasid" until 125.(5) The main concern of Revolt is obviously with the overt military stage of the revolt. It begins by arguing that the tribal fighting in Khurasan "irreparably damaged" Umayyad power as "the allegiance of its traditional tribal support melted away." In the course of organizing the Abbasid forces, Abu Muslim introduced the supposedly radical reform of registering the members of the revolutionary army in the diwan according to their village of origin rather than their tribal affiliation. This blunted "the sense of tribal awareness and tribal loyalties in the army" (p. 99), and the da wa thus became a mass movement which indeed began to attract and incorporate existing military units, beginning with the warriors of Azd and then other tribal groups, into its army. From then on, the story of the Abbasid revolution is essentially that of how this "vibrant and amazingly well organized body," "full of vitality," and acting "with precision under the faultless guidance of talented political leaders, military commanders and professional administrators" rolled on to final victory (p. 93). Throughout, Sharon insists on the overwhelmingly Arab character of the revolution, leading up to this conclusion (p. 258): "We have proved beyond all doubt that in the decisive stage, the Dawah was a movement based primarily on the support of the Arabs who formed the bulk of its army and comprised the backbone of its command."

The technical merits of Sharon's work are considerable, especially in terms of the author's command of the Arabic sources and his sensitivity to the problems of conflicting accounts found in them. As a narrative history of the Abbasid revolt from the inception of its militant phase down to the proclamation of Abu'l-Abbas as caliph, there is little that could or need be added to it. As an interpretation of the "social and military aspects" of that revolt, however, it is less satisfactory, and three major reservations may be expressed about it.

First, the methodological foundation on which it rests is weak. Like so many studies of early Islamic history, it uses textual criticism to sift the source materials into the historical and the fictitious. This implicitly assumes that the modern scholar is in fact able to perceive the intention behind a text and assess its reliability accordingly. In some instances this may be possible, but in many others it would be more prudent to admit that we lack this ability. In the context of a short review, one example will have to suffice. The Akhbar al-Abbas, accepted by Sharon as one of the most important sources on the history of the revolution, quotes Nasr b. Sayyar and one of his lieutenants describing the Abbasid revolutionaries as "cat-worshippers." Sharon is reluctant to accept these statements at face value since they imply the revolutionaries were non-Muslims and thus non-Arabs; he dismisses them as propaganda and asserts (p. 109) that "such tales as that of Abu Muslim and his companions holding religious ceremonies in which cats were worshipped could easily capture the public imagination." To be effective, however, propaganda has to have an element of plausibility; if the rebels were, as Sharon claims, mostly Arabs, a description of them as "cat-worshippers" seems weirdly out of place and hardly credible. Moreover, it would not be a very convincing way of insinuating that the rebels were either Persians or Persianized and anti-Arab or anti-Islamic. It must be remembered that in Zoroastrian culture the cat was regarded as a ritually unclean, even satanic, animal (the exact opposite of its status in Islam). If anything, it was the dog which was venerated among the Persians and despised by Arab Muslims, so one would think that something like "dog-worshipper" would have been a much more useful charge to hurl at the rebels in such a propaganda campaign. Given the absurdity of anyone in Khurasan worshipping cats, as someone who had lived there as long as Nasr b. Sayyar must have known, why should we think that an inherently unbelievable accusation "could easily capture the public imagination"? What really ought one to make of such a strange comment? Was it intended as a clue to the reader that the charges were utterly slanderous? Is it evidence that this text was forged and devoid of historicity? Or does it, to the contrary, reflect a touch of absolute authenticity in that the statement is knowledgeable enough about the local culture to be using "cat-worshipper" as the worst possible insult? There must indeed be some kind of subtext involved in passages such as these, but the meaning is much more elusive than Sharon is usually willing to allow.

Second, Sharon's explanation of the importance of Abu Muslim's ostensible "military reforms" based on registering soldiers in the diwan without reference to their tribal affiliation seems somewhat exaggerated. For one thing, we do not know for certain that the soldiers being registered even had a tribal affiliation (Sharon merely assumes that they were Arabs). On the other hand, the soldiers were not registered only according to their village of origin but also "with the names of their fathers." This implies that lineage was not being disregarded in the new register, and it might still have been possible to deduce a tribal affiliation from the ancestry. It also appears possible that this "reform" had been used previously and was most likely not an original innovation by Abu Muslim.(6) Moreover, if tribal identity and asabiyya were as pervasive and powerful as depicted by Sharon, just exactly how could the mere act of recording names cause such sentiments to dissipate? Equally confusing is Sharon's assertion that while the omission of nisbas from the register was deliberate (p. 101), its radical effects were not: "By introducing his reforms to the system of the army registration, Abu Muslim unintentionally dealt a serious blow to the Arab supremacy and predominance in the Islamic state" (p. 99, emphasis added). It is thus ironic that Sharon's theory, if correct, greatly diminishes the significance of the Abbasid revolution itself, since it merely "replaced one Arab aristocracy with another" (p. 257) while pushing the Arabs themselves out of power, "contrary to its original aims and intentions" (p. 259).

The greatest difficulty with Sharon's interpretation, however, is its doctrinaire insistence on the Arab character of the Abbasid revolution. At one point (p. 75), he explains that the revolt was initiated in Marw "because Marw and its vicinity were mainly populated by Azd." It is hard to believe that the majority of the population of a major Iranian city such as Marw was Arab, much less Azdi, and Sharon presumably means that the bulk of the Arab inhabitants of the oasis were members of Azd. The statement, however, is typical of the extent to which Sharon consistently reduces the non-Arab population of Khurasan to historical invisibility. In his view, opposition movements against the Umayyads "had to arise from the Arabs, among the Arabs, and its leaders had of necessity to be Arabs" and indeed "it was impossible for any opposition whatsoever to exist unless it was made up of Arabs" (p. 32); moreover, "it was impossible for any opposition to the Arab rule to arise from any other than an Arab source" (p. 258). Apparently the numerous nativist revolts or resistance such as that led by Diwashani, Nizak and others do not count as "opposition." As for the Abbasid revolt itself, Sharon holds that it does not matter if it included non-Arab participants, since "names of non-Arabs, as many as they may be, can not [sic] obscure the fact that they were subordinate to Arab officers in the field and to Arab leaders who served the cause of an Arab dynasty" (p. 259 n. 84; emphasis added). One has to wonder if historians in the distant future, using such historical reasoning and methods, would conclude that World War II in Europe was a squabble among Germans, since the Allied commander was named Eisenhower, or that the French Revolution was neither French nor anti-monarchist, since it eventually produced a Corsican emperor.

There are in actuality a number of difficulties with this Arab constituency theory. First of all, it requires rationalizing away as propaganda or fabrications a considerable amount of information from diverse and independent sources - the Imam Ibrahim's instructions to Abu Muslim (pp. 52-57), the influx of "slaves" into the camp of Abu Muslim (pp. 58-62), Nasr b. Sayyar's depiction of the opposition (pp. 64-65, 108-9), accounts of large numbers of rebels from all parts of Khurasan bearing the clubs known as kafir kubat (p. 86), and Qahtaba's speech to an army implicitly regarded as largely Persian (p. 89). It also means that one must not attach much importance to the fact that Nasr b. Sayyar came very close to forging a broad coalition of all the major Arab interests in Khurasan against Abu Muslim until his efforts were disrupted by the mysterious murder of Juday b. al-Kirmani, or to the apparently systematic liquidation of Arab leaders in Khurasan, from both within and outside the movement, which accompanied the revolution (Sulayman b. Kathir, Lahiz b. Qurayz, Ali and Uthman al-Kirmani, Shayban al-Khariji, two dozens of Nasr's supporters, Yazid al-Nahwi, Ibrahim al-Saigh, etc.). Much of the evidence cited by Sharon is, in fact, ambiguous or susceptible of alternative interpretation. In describing the outbreak of the revolt, for example, he claims, on the basis of a passage from the Akhbar al- Abbas, that "warriors from the tribes of Khuzaa began pouring into the Abbasid camp at Saqidhanj (p. 76)" and later that "about seven hundred warriors from the region of Suqadim from all the qusur (fortified villages?) of Khuzaah were the first to arrive in Saqidhanj. We do not know how many of these were mawali and how many were Arabs, but they included Arabs of Khuzaah and even a number of fighters from the tribe of Tamim" (p. 87). There are two problems with this. First, it hinges on the assumption that when the sources refer to people coming from Khuzaah villages this means that they were Khuzaites. However, as was already apparent to Wellhausen, an "Arab village" in Khurasan was one owned by an Arab tribe, which is not necessarily the same thing as being inhabited by Arabs. Moreover, the source in question does give us very specific information about numbers - there were twentyfive men of Khuzaa, six of Tayy, twelve of Tamim, and thirty-one of "the nuqaba and shia" (who may or may not have been Arabs). If any of the other 626 had tribal affiliations, why are we not told what they were? And how does any of this add up to Khuzai (or Arab) warriors "pouring" into Saqidhanj? Finally, it must be noted that much of Sharon's argument is already being undermined by a new source and by new research. A letter from the famous Umayyad secretary Abd al-Hamid b. Yahya to Nasr b. Sayyar has now been identified and edited. If authentic, and both Ihsan Abbas and Wadad al-Qadi have accepted it as such,(7) it clearly implies that the Umayyad authorities were fully informed about the outbreak of the revolt and regarded it as essentially non-Arab in character. In addition, Saleh Said Agha has made an exhaustive prosopographical study of participants in the Abbasid organization and revolt and concluded that approximately thirty-seven percent were mawali, fortyfour percent were other non-Arab converts, and only eighteen and a half percent were Arabs; of "Abu Muslim loyalists" fully eighty-nine percent were non-Arabs and eleven percent Arabs.(8) Sharon's thesis is interesting and well argued, but for those who think evidence and numbers do matter more than dogma, it is hardly "beyond all doubt."

Mohsen Zakeri's Sasanid Soldiers in Early Muslim Society addresses a much broader subject than the Abbasid revolution but it has many interesting implications for problems such as these. It provides a remarkably clear, readable, and richly annotated study of how a pre-Islamic social class and its values may have survived to be transformed and incorporated into Islamic institutions and culture. After an outline of the problem and sources, Zakeri tackles the question of whether Sasanid society was "feudal." He argues that while it was not truly feudal, it nonetheless created a distinctive class of "land-holding nobility," the azadan (whom he variously describes as "small proprietors," "lesser nobility," or even "more substantial peasantry"). They were, as the name implies, free men and had independent bases of power, but they derived "prestige" from association with the monarchy and provided the ruler with the means of fiscal control over peasants and/or military service. The two most important representatives of the azadan were the cavalry (asbaran) and the village lords (dihqanan). This class of "gallant warriors" developed its own sense of solidarity and identity through the use of special dress, insignia, "chivalric" code of conduct, etc. Zakeri observes that much of later Sasanid history revolved around a struggle between the shahs, backed by the azadan, and the aristocratic families of large landholders. When the greater and lesser nobility joined forces against the Sasanid monarchy, the empire crumbled. He argues that while the "aristocrats" were largely swept away during the Arab conquests, the azadan survived because of their administrative and military usefulness, carrying out much the same functions as before and eventually converting and becoming integrated into Muslim society.

Zakeri then turns, in the third and lengthiest chapter of his work, to specific examples of the persistence of the azadan and their socio-cultural legacy in various parts of the Middle East. He examines the role of the dihqanan in the Muslim conquest; the asbaran in Basra and Kufa; the asbaran in Syria (referring to groups such as the Khadarima, Jaramiqa, and Mardaites); the asbaran and other Persians in provincial administration and the general bureaucratic structure of the caliphate; dihqans and asbaran in Iraq; the dihqanan and asbaran in Khurasan; and finally the abna (from those in pre-Islamic Yemen to Abbasid Baghdad). The final chapter considers the cultural influence of these influential individuals in the Shuubi controversies, the futuwwa and furusiyya traditions, and theological movements from the Qadariyya to the Mutazila. It should be obvious, then, that the title of his book is something of a misnomer. He does not limit his discussion to the Sasanids (finding, for example, Achaemenid origins for the Jaramiqa and Mardaites), nor to soldiers (including peasants, bureaucrats and other civilians), nor to the origins of the ayyaran and futuwwa (branching out into treatment of a wide variety of classes, institutions, and practices).

If Sharon is to be faulted for not seeing Persians anywhere, then it has to be admitted that Zakeri has a disconcerting tendency to see them and Persian influence everywhere, including some rather unlikely places. His search is made a great deal easier by the fact that so much depends on purely philological evidence, which is, to say the least, susceptible to multiple interpretations (as evidenced by the number of contrary theories he must rebut before introducing his own alternative). It is also facilitated by the ambiguity of the technical terms involved - ayyar could refer to anything from a valiant professional soldier to a common bandit to an adolescent urban thug - and Zakeri's tendency to stretch them even further by lumping related ones together as if they were virtually synonyms. In his view, for example, what would appear to be a discrete group of cavalry forces, the asbaran, suddenly comes to include in Islamic times groups as diverse as the banu'l-ahrar, asawira, ahamira, abna, jawannardan, ayyaran, ahdath, and fityan (p. xii). Although he wants to make the case for "Persian influence" in early Islamic society, he sometimes does so by using "Sasanid" to include tenuous political and territorial, as well as ethnic, connections. One has to feel a little uneasy about discussing the Nabateans, Zutt, and Siyabaja in this context, and it really seems to be stretching the limits of credulity to suggest that the Mardaites have anything to do with Sasanid military personnel and institutions in early Muslim society or that survivals of Sasanid garrisons in Syria might help explain the appearance of ahdath organizations in Damascus. Another likely objection to Zakeri's approach concerns his tendency to assume that the appearance of similar institutions implies an evolutionary relationship. Since any complex society is likely to have judges, police, and civic authorities, how certain can we really be, for example, that "the office of the qadi al-qudat . . . was an adaptation of the Persian institution of mobadhan mobadh" (p. 170), that the ma una was "the lingering sign of a Sasanid institution" (p. 175), or that the muhtasib "more probably had a precedent in the Sasanian institution of vacarapat" than the Greek agoranomos (p. 177)? Much often depends on the precise meaning of a single or perhaps dubious source, as with the tremendous weight placed on Abu'l-Faraj al-Isfahani's description of the asbaran or Ibn Durayd's etymology of Khadarima. And it may well be, for example, that the asbaran wore distinctive necklaces and earrings, but should we document this on the basis of Firdawsi's description of the battle of Qadisiyya (p. 75)? In any case, potential readers would be well advised to moderate Zakeri's shotgun approach to tackling the problem of the origins of the ayyaran and futuwwa by looking at the much more cautious views of Cahen and Hanaway now available in the Encyclopaedia Iranica (s.v. "Ayyar").

These concerns, however, should not obscure the fact that Zakeri raises some very significant issues. It never hurts to remember that Islamic civilization did not develop in a vacuum, and that deeply entrenched social institutions and practices are not likely to have evaporated overnight. As Islam spread and became a universal religion, it is certainly important to take into account how much cultural baggage the new converts may have brought along with them. Zakeri is especially effective in demonstrating how the interests of Arabs and Persians became closely intertwined in Khurasan because of the unique conditions that prevailed there (pp. 227-65). He reminds us that it is not necessary to think of the Abbasid revolt as exclusively an "Arab" or a "Persian" movement, nor in fact very useful to use racial categories at all in describing it; it might better be described as a Khurasani revolt that cut across ethnic lines and which was fighting an army that was more Syrian than tribal. What Sharon views as a radical military innovation by Abu Muslim appears in the context of Zakeri's work more like a revival or re-emergence of the traditional village-based militias of the Sasanid era; instead of tribal chiefs and warriors pouring into the Abbasid camp, we could just as easily and perhaps more accurately see alienated dihqanan and their lower-class infantry recruits (payadagan).

In conclusion, all these books provide valuable perspectives on the problems of early Islamic history and the Abbasid revolution, and they suggest some directions for future research. In particular, it is clear that much more work needs to be done to clarify the historiographical and methodological difficulties involved in utilizing the available source materials. There is also the crucial and still-unresolved mystery surrounding the figure of Abu Muslim - not just who he was but how he managed to take control of the Abbasid organization in Khurasan and how he was able to acquire his obvious military and administrative abilities. Similarly, a close study of al-Mansur and the way in which he was able to consolidate power is also badly needed (Sharon is apparently preparing a volume on this topic). Much remains uncertain about what really caused the revolt, the different stages of its development, and what changes it actually produced. Nonetheless, these authors deserve thanks both for the answers they have provided and for prodding us to ask still more questions.

1 R. Stephen Humphries, Islamic History: A Framework for Inquiry, rev. ed. (Princeton, 1991), 104.

2 Roberto Marin-Guzman, "The Abbasid Revolution in Central Asia and Khurasan: An Analytical Study of the Role of Taxation, Conversion, and Religious Groups in its Genesis," Islamic Studies 33 (1994): 227-52.

3 C. E. Bosworth, Al-Maqrizi's "Book of Contention and Strife" Concerning the Relations between the Banu Umayya and the Banu Hashim (Manchester, 1980), 88.

4 Patricia Crone, "Mawla," Encyclopaedia of Islam, new ed., 6: 874-82; idem, "On the Meaning of the Abbasid Call to al-Rida," in C. E. Bosworth et al., The Islamic World from Classical to Modern Times (Princeton, 1989), 95-111; idem, "Were the Qays and Yemen of the Umayyad Period Political Parties?" Der Islam 71 (1994): 1-57.

5 Somewhat confusingly, Sharon does seem to accept Abbasid involvement in the dawa at a very early date as genuinely historical. For example, he notes the story that Muhammad b. Ali sent Abu Ikrima as a propagandist to Khurasan, ordering the latter to conceal his identity behind a different kunya (p. 98). This would have been in 104/722, and if we accept that at that date Muhammad b. Ali was in touch with the organization and designating its leaders, what exactly does it mean to say that the da wa was "not Abbasid"?

6 See the comments below on the Zakeri volume. Continuities between the Umayyad and Abbasid periods in military policy and other respects are also emphasized in an interesting article by I. Bligh-Abramski, "Evolution versus Revolution: Umayyad Elements in the Abbasid Regime 133/750-320/932," Der Islam 65 (1988): 226-43.

7 Ihsan Abbas, Abd al-Hamid b. Yahya al-Katib (Amman, 1988); Wadad al-Qadi, "Earliest 'Nabita' and Paradigmatic 'Nawabit,'" Studia Islamica 78 (1993): 27-61.

8 Saleh Said Agha, "The Agents and Forces that Toppled the Umayyad Caliphate" (Ph.D. diss., Univ. of Toronto, 1993), 404, 462. Some of the percentages have been rounded off.