Abstract
At some moment in time, the patronate system that had been introduced as a way to incorporate non-Arab Muslims into Arab society, allowed the client of a patron to have clients of his own. Using this phenomenon of mawālī of mawālī as focal point, this article pinpoints when changes in the patronate system occurred and sketches the process of islamization of society during the first four centuries of Islam.
* I would very much like to dedicate this article to Patricia Crone, a pioneer of Islamic History. My thanks go to Meia Walravens for reading and commenting on an earlier version.
Introduction
Islam was born into a society that was Arab and tribal. An individual’s place in society was determined by his or her affiliation to an Arab tribe. As Islam expanded, more and more non-Arabs came into contact with the Arab Muslims. The manner in which non-Arab Muslims entered the newly evolving Arab-Islamic polity was — despite the affirmation in Islam that all Muslims are equal — by becoming a mawlā (pl. mawālī, English “client”) of an Arab.1 By way of this patronate, the neophyte became a “member” of the patron’s tribe and thus, by extension, a member of society.2 Along these lines, the primordial Arab-tribal society was gradually transformed into a much larger cosmopolitan-Islamic society. This article deals with the emergence of an Islamic polity from the original society in which Islam was born, thus creating a new society that had not existed before, a society that would soon become a great world civilization.3
This process of acculturation or “islamization” has been described as follows. Shortly after the first conquests, prisoners of war — who were too many to keep as slaves — were amongst the first mawālī.4 A second tide of mawālī came from peasants who sought to improve their own personal position by affiliating themselves with Arab tribes as mawālī.5 Three researchers propose that the process of mawālī joining the Islamic polity was probably very fast, resembling a snowball effect. Bulliet, for instance, argues from a theoretical viewpoint that
[l]ogically, given the disproportion between the small, scattered Arab populations and the immeasurably larger conquered populations, this form of conversion should have given rise to a larger and larger population of non-Arab converts whose conversion had been accomplished through presentation before other non-Arab mawālī (or mawālī of mawālī) rather than through actual contact with Arabs.6
Crone also states that patrons were not only Arabs: “. . . mawālī rapidly acquired freedmen and other clients of their own, who acquired freedmen and clients in their turn, and so on.”7 Indeed, Juda goes even further by stating that “the mawālī were able to dominate crafts and trade primarily because they owned slaves and were the patrons of other mawālī, who acted as craftsmen, salesmen and traders for their mawālī masters or patrons.”8
The focus of this article is on this group of mawālī of mawālī. The literature cited up to now suggests that the phenomenon of mawālī of mawālī had been there almost from the outset. To help organize our thinking, I will speak of three phases of “islamization”: Phase 1 was when non-Arab converts to Islam had to find an Arab patron; in Phase 2, mawālī could also have a mawlā as patron instead of an Arab patron. The third phase was when a convert did not need a patron at all and thus had no need for the patronate relationship in becoming a Muslim. Pinpointing the transitions from Phase 1 to Phase 2 to Phase 3 provides us with more insight into this process of islamization, which will be studied here through the mawālī of mawālī. However, before we move on to the mawālī of mawālī, the material used for this article will be described.
The empirical data to be analyzed were taken from the main database of the Ulama Project (up).9 This main database consists of a random sample drawn in such a way as to allow generalizations about the entire group of ulama as recorded in the classical Arabic biographical dictionaries.10
The timespan of this study consists of the first four centuries of Islam (approximately the seventh up to the tenth centuries ce). The beginning of Islam is one temporal boundary. The end-point is when the Islamic college or madrasa came into being and with it the professionalization of the ulama as regularly paid ulama-teachers as well as the canonization of Islamic knowledge to be passed on and about which some form of consensus existed by the time the madrasa arrived on the horizon.11 Stated differently, the period studied is when the initial elements of the Islamic polity started to emerge, then expanded as did the Islamic religion and, finally, developed into what that polity was when the madrasas started to populate the Islamic world.
To create a context in which we can study the mawālī of mawālī, the overall percentage of mawālī and how they compare with the Arabs during the first four centuries of Islam will first be presented. The goal of this exercise is to provide a general view of the process of the integration of the mawālī. Secondly, the patrons they had and how large the proportion of mawālī patrons was in comparison with Arab patrons as well as how this proportion changed in time will be discussed. The subsequent section provides us with a qualitative analysis of the mawālī of mawālī from the main database in an attempt to discover who these men were and if they had any common features that set them apart from other mawālī. The article closes with a discussion of what the findings tell us about the process of islamization in early and classical Islam.
The Integration of the Mawālī in Society
Our first task, then, is to obtain a general view of the gradual integration of mawālī amongst the Arab Muslims. As such, we are creating a context which offers us the possibility of scrutinizing the mawālī of mawālī within this particular framework. Figure 1 displays the percentage distribution drawn from the sample, per Islamic century, of Arabs, mawālī and a third group that did not receive such an explicit designation from the compilers of the biographical dictionaries, labelled in Figure 1 as “Unknowns.”12
All in all and irrespective of the distribution per century, there were 245 Arabs, 252 mawālī and 552 unknowns, totaling 1,049. Turning to the aspect of time depicted in Figure 1, the highest percentage in the first Islamic century is that of the Arabs who, at that time, made up more than 80%: of the 93 sampled ulama who died during the first century, 77 were Arabs, followed by a small group of mawālī (9 out of 93, or 9.7%) and then those ulama who were not classified as Arab or mawlā (7 of the 93, 7.5%, labeled in the graph “Unknown”). The very rapid decrease of the share of Arabs is just as dramatic as the fast increase in the number of mawālī during the second Islamic century, tailed by the “Unknowns” — that is, 37.1% Arab, 39.1% mawālī and 23.8% “Unknown.” However, in the course of the third Islamic century and continued in the fourth, the trend was that both percentages of Arabs and mawālī diminished, while simultaneously the percentage of the “Unknowns” steadily increased.
Percentage Distribution of Ethnicity across the First Four Islamic Centuries (N = 1,049).
Citation: Journal of Abbasid Studies 1, 2 (2014) ; 10.1163/22142371-12340009
Percentage Distribution of Ethnicity across the First Four Islamic Centuries (N = 1,049).
Citation: Journal of Abbasid Studies 1, 2 (2014) ; 10.1163/22142371-12340009
Percentage Distribution of Ethnicity across the First Four Islamic Centuries (N = 1,049).
Citation: Journal of Abbasid Studies 1, 2 (2014) ; 10.1163/22142371-12340009
These features have led to the conclusion that what we call the “Unknowns” were, in all likelihood, in the earliest period ulama about whose ethnicity no information was available, but as time passed by — and the ethnic designation became less important as society became increasingly “Islamic” rather than Arab or otherwise — these “Unknowns” were simply Muslims with no patronate affiliation. We can therefore take the continuous increment of the number of “Unknowns,” especially when compared with the Arabs and mawālī, as an overall gauge of the degree of islamization of society. The Arab character of the new Islamic polity was at the outset overwhelming while in the second century more and more non-Arab mawālī entered society and participated in making it Islamic in its own right. As Figure 1 shows, this process was fully underway by the third Islamic century and was completed in the fourth century, when society was chiefly defined in terms of being Muslim or not: The ethnic designation of being an Arab or a mawlā Muslim had been relegated to social irrelevancy.
As noted earlier, the secondary literature suggests that, due to the sheer size and speed with which the Arab Muslims conquered other areas and peoples at the beginning of Islam, the emerging society had no choice but to incorporate the system of mawālī having mawālī of their own. This theoretical depiction implies that Phase 1 — when only Arabs could have mawālī — ended early as Phase 2 commenced, that is, when the mawālī were able to acquire mawālī themselves, just like their Arab patrons. Phase 3 was defined as the moment in time when a convert to Islam no longer needed a patron in order to join Muslim society. If we focus on Phase 2, when it began and when it ended, we will obtain more insight into the process of islamization as the Arabs and the mawālī together incorporated new members of the emerging society. It is with this thought in the back of our mind that we now turn to the mawālī whose patrons were other mawālī, not Arabs.
Mawālī of Mawālī
Our first step is to take a look at how many of the 1,049 ulama of our sample have an explicit listing of patron. As we saw earlier, there are 252 mawālī in the dataset. Of these 252, information on the ethnicity of their patrons is available for 95, or about 40%. The distribution of ethnicity of these 95 patrons is given in Table 1.
Patron Ethnicity
It is at this stage important to stress that we are using data from a representative sample. Therefore, we can generalize our findings to say something about the entire population — in our case, the ulama of the first four centuries of Islam. Consequently, the estimate of the percentage of mawālī of mawālī found in the sample is the best estimate for the entire population from which the sample was drawn — all ulama with an explicit year of death before 400 AH. Some 10.5% (10 of the 95 mawālī, see Table 1) of those ulama, for whom the ethnicity of the patron is known, had a mawlā as patron — a minority.13 The percentage of some 10% is hardly the reflection of any snowball effect — supposedly the result of sheer necessity — of mawālī increasing at such an exponential demographic rate that mawālī had to also have other mawālī of their own. More insight is gained if we situate these ten mawālī of mawālī in time, which is done in Figure 2. Figure 2 displays the birth year (if known) and the year of death of the ten mawālī of mawālī taken from the sample.
Lifespan (Birth and Death Years) of Mawālī of Mawālī across the First Four Islamic Centuries.
Citation: Journal of Abbasid Studies 1, 2 (2014) ; 10.1163/22142371-12340009
Lifespan (Birth and Death Years) of Mawālī of Mawālī across the First Four Islamic Centuries.
Citation: Journal of Abbasid Studies 1, 2 (2014) ; 10.1163/22142371-12340009
Lifespan (Birth and Death Years) of Mawālī of Mawālī across the First Four Islamic Centuries.
Citation: Journal of Abbasid Studies 1, 2 (2014) ; 10.1163/22142371-12340009
The majority of mawālī of mawālī are found in the second and early third centuries — their years of death range from 145 AH to 231 AH with two later in time, one who died in 264 and a loner who died as late as the middle of the fourth Islamic century. We do not know when the mawālī of mawālī became mawālī — were they so at birth or did they acquire this status in the course of their (adult) lives? Undoubtedly, both possibilities occurred in reality but our source material does not allow us to determine this for our ten mawālī of mawālī.
Inasmuch as some researchers have suggested that the Abbasid Revolution may have occurred in part because non-Arab Muslims wanted the same “rights” as Arab Muslims, let’s adopt the year 132/750 as a bench mark and take another look at Figure 2.14 Prior to 132/750, five mawālī out of the ten were born and may have been a mawlā of a mawlā before the Abbasid Revolution. This finding suggests that the mawālī had the possibility of acquiring mawālī of their own before the Abbasid Revolution.15
In terms of size, the phenomenon of mawālī of mawālī was marginal — about 10.5% were mawālī of mawālī, no exponentially increasing majority, as says the secondary literature above. For our purposes, however, the group of mawālī of mawālī from the sample is small enough to allow a more in-depth, qualitative analysis of these mawālī. Before we continue with this transition from a quantitative analysis to a qualitative one, we must first “clean up” the sample we will be using. For purely technical reasons that have to do with guarding the randomness of a sample to ensure its generalizability, some ulama were, by sheer chance, drawn twice: for instance, once as someone active in ḥadīth and once as someone active in exegesis, as dictated by our research design referred to earlier.16 Translating this to concrete steps, of the ten mawālī, one was randomly drawn twice and therefore our qualitative analysis will total nine mawālī who had a mawlā as patron.17 The names of the men, together with a brief and succinct biographical description of each, chronologically listed according to year of death, in order to search for anything that may set these men apart from others, constitute the subject of the next section.
Biographies of the Mawālī of Mawālī
(1) Ashʿath b. ʿAbd al-Malik,18 from Basra, was called al-Ḥumrānī because he was the mawlā of Ḥumrān who in turn was a mawlā of the caliph ʿUthmān. His birth date is not recorded and he died around 145/762. He was a scholar of ḥadīth and Islamic law, described as being a ṣāḥib sunna and cited in four of the six Sunnite canonical ḥadīth collections (not in al-Bukhārī and Muslim).19
(2) Maʿmar b. Rāshid, born in Basra in ca. 95/713-14, was a resident of Yemen where he also died in 153/770; he was a mawlā of the Azd tribe — a mawlā of ʿAbd al-Salām b. ʿAbd al-Quddūs to be more precise, who in turn was the mawlā of ʿAbd al-Raḥmān b. Qays al-Azdī. As a pre-canonical ḥadīth collector, he is mentioned in all six canonical collections.20 He was furthermore the first to comprise a muṣannaf of ḥadīth in the Yemen (awwalu man ṣannafa bi-l-Yaman).
(3) ʿAbdallāh b. al-Musayyab b. [Abī] al-Sāʾib al-ʿAbīdī al-Qurashī, died in 170/786-7.21 He was the mawlā of ʿAmr b. al-ʿI[A?]jlān who, in turn, was a mawlā of the caliph ʿUmar. He is only cited in the canonical ḥadīth collection of Abū Daʾūd.
(4) ʿAbd al-Razzāq b. Hammām al-Ṣanʿānī was born in 126/744 and died in 211/827.22 He was, according to al-Mizzī a mawlā of the Ḥimyar23 or, as Ibn ʿAsākir puts it, “a mawlā of mawālī of an Arab tribe” (mawlā li-mawālī qawmin min al-ʿarab).24 He lived in Sanaa, the Yemen, and was possibly originally of Persian extract. This scholar was a traditionist who composed pre-canonical ḥadīth collections and also wrote exegesis. It is said that he had memorized 17,000 ḥadīths and he is cited in all six Sunnite canonical ḥadīth collections.
(5) ʿAbdallāh b. ʿAbd al-Ḥakam b. Aʿyan was a leading legalist scholar of Egypt, associated with the Malikites, who was born in around 155/772 (Alexandria) and died in 214/829 (Cairo).25 He is cited in al-Nasāʾī’s canonical ḥadīth collection. Al-Mizzī quotes from al-Kindī’s Aʿyān al-mawālī bi-Miṣr to explain that he was a mawlā of Rāfiʿ, a mawlā of the caliph ʿUthmān.26
(6) Saʿīd b. al-Ḥakam, known as (Saʿīd) Ibn Abī Maryam.27 Born in about 144/761, died in 224/838-9, he was an important ḥadīth-scholar in Egypt who is cited in all six Sunnite canonical collections. He was the mawlā of Abū l-Sabīgh, who in turn was a mawlā of the Banū Jumaḥ.
(7) Yaḥyā b. ʿAbdallāh, Ibn Bukayr,28 born in 154/771, died in 231/845-6, was a scholar of ḥadīth who is cited in three of the six canonical collections and who also wrote about history. According to Ibn Ḥajar al-ʿAsqalānī, Ibn Bukayr transmitted from ʿAbdallāh b. al-Musayyab listed above. Al-Mizzī explains that he was the mawlā of ʿAmra bint Ḥunayn, a woman who was a mawlāt (!) of the Banū Makhzūm.29
(8) Aḥmad b. ʿAbd al-Raḥmān, known as Baḥshal, died in 264/877-8.30 He was a transmitter of ḥadīth, cited in Muslim’s canonical ḥadīth collection, and active in Egypt. He was a descendent of a mawlā of Yazīd b. Rammāna, who was a mawlā of the Companion Abū ʿAbd al-Raḥmān al-Fihrī.31
(9) ʿAbdallāh b. Idrīs, an otherwise unknown legal scholar from Cordoba who is associated with the Malikites and who died in 344/955-6.32 He was an ancestor of the mawlā Abān b. ʿUmar who was a mawlā of the caliph ʿUthmān.
The most prominent details from the preceding biographies are summarized in Table 2.
Except for the earliest mawlā mentioned, all hailed from outer regions of the Islamic empire, Yemen, Egypt and al-Andalus. The image this finding brings to mind is that the mawālī of mawālī were primarily found in the periphery of the Islamic world, at any rate, in regions where the Arab presence was not as prominent as it was in the central provinces of Iraq or Syria (al-Shām).
About half of these men (5 out of 9) were a mawlā of a mawlā — or in one single case the mawlā of a female mawlā — who was affiliated with a member of an Arab tribe.33 The other half (4 out of 9) were mawālī of mawālī of one of two of the so-called Orthodox/Rāshidūn Caliphs, either ʿUmar or ʿUthmān (three men, Ashʿath, ʿAbdallāh b. al-Ḥakam and ʿAbdallāh b. Idrīs are described as being the mawālī of mawālī of ʿUthmān, and one, ʿAbdallāh b. al-Musayyab, a mawlā of a mawlā of ʿUmar). Taking into account when these men lived and died, in most cases, the mawlā relationship was one that must have been passed on from generation to generation: not these men themselves (in some cases), but an ancestor of theirs reportedly had this special legal tie with either the third or the fourth Orthodox Caliph.
Discussion and Conclusions
The research reported in this article is based on a representative sample of ulama who lived and died during the first four centuries of Islam. We have generalized from the sample to the entire population of ulama. Yet the question remains to what extent can we extrapolate these results to the rest of society. In his work on conversion to Islam in the early and classical periods, Bulliet directly tackles the issue of the generalizability of the data about ulama to society as a whole. Bulliet cogently demonstrates how the aggregate of the thousands of biographies of ulama in the classical Arabic biographical dictionaries approximately follows the general course of political Islamic history. Due to the lack of a pre-imposed hierarchical structure, like the church in Catholicism, the only requisite for anyone who wanted to join the ulama, Bulliet continues, was an environment that was conducive to study Islam — in other words, the ulama came from different walks of life. The environment they worked in was an urban environment. Therefore, the findings contained in this study are generalizable — at least generalizable for urban society.35
This issue of validity is germane to our discussion since the results call into question a number of assumptions encountered in the secondary literature. We have seen that, in general, the first Islamic century was primarily an “Arab” century; the second century witnessed an astonishing increase of mawālī amongst the ulama. Limiting our discussion here to the phenomenon of mawālī of mawālī as a gauge for islamization, we must conclude that the amalgamation of the mawālī with and their influx into the (Arab) ulama in terms of being able to have mawālī of their own, only started to take place during the second and early third Islamic centuries.
Furthermore, the phenomenon was not a common one in two completely different senses. First, a mere 10.5% of the mawālī were mawālī of other mawālī, a far cry from some unyielding snowball effect that has been suggested by other studies. The marginality of the phenomenon can perhaps in part be explained by the fact that, as the third Islamic century proceeded, the patronate became more and more an obsolete social category as society was becoming “universally” Muslim. Secondly, we have also discovered that the majority of the mawālī of mawālī probably lived in the “periphery” of the Islamic empire — most of them in Egypt, some in Yemen and al-Andalus. This result suggests that the mawālī of mawālī phenomenon was primarily found in the outer regions of the empire, not in the core. Very intriguing is furthermore the observation that we did not encounter mawālī of mawālī in the eastern part of the empire. An essential question for further research is (the lack of) mawālī of mawālī in the eastern part of the Islamic realm or, why was the eastern part of the empire different from the western part in this aspect?
Intriguingly, almost half of the mawālī of mawālī studied here were, or claimed descent from, a mawlā of a mawlā of ʿUthmān or ʿUmar, two of the four Orthodox Caliphs. From a Sunnite purview, noting descent from someone who was affiliated with men like ʿUthmān or ʿUmar could be used as a means for religious legitimization of their status as a religious scholar — why else would a scholar who died halfway the fourth Islamic century, in which the patronate had passed away, claim affiliation by way of a mawlā to ʿUthmān? The claimed reference to being a mawlā of one of the mawālī of an Orthodox Caliph was most probably used to buttress the status and social standing by implying a continuous link all the way back to the Prophet and his Companions. Rather than carrying a negative connotation of having to be somebody else’s client, the reference by mawālī of mawālī here to their affiliation with an Orthodox Caliph was meant to be positive, in the sense that this association implied some form of religious authority vis-à-vis other Muslims who could not make such a claim. Further research should look into this shift in perception of the mawlā status across time for it, too, tells us something about the pace of islamization.
In closing, let us refer back to our model of Phase 1, 2 and 3 which we adopted to help organize our thinking in order to summarize the main conclusions with regard to the islamization of society during the first four centuries of Islam. In Phase 1, only Arabs acquired mawālī; based on what we have seen, this Phase 1 ended toward the end of the first Islamic century when Phase 2 started, that is, when mawālī themselves were able to acquire mawālī of their own. The analyses additionally showed that the process of the acculturation of non-Arab Muslims amongst the ranks of Arab Muslims — the process of the islamization of society — ended in the course of the third and very especially the fourth Islamic century, ushering in Phase 3 when the patronate system was no longer required and converts had no need for a patron — neither Arab nor mawlā.
1 More general works on the mawālī include Crone, Mawlā; Juda, Die sozialen und wirtschaftlichen Aspekte; idem, al-Awḍāʿ al-ijtimāʿiyya; Bi-l-nūr, Dawr al-mawālī; al-Badrāwī, Al-Mawqif al-umawī.
2 The most elaborate study of the patronate to date is Bernards and Nawas (eds.), Patronate and Patronage. For legal aspects of the term see Crone, Roman, Provincial and Islamic Law who stresses the continuation and/or adaptation of the Roman patronate system within the new Islamic Near Eastern setting; also Mitter, Das frühislamische Patronat.
3 To be sure, Arab-tribal society did not disappear altogether. On a smaller scale, Arab-tribal society naturally also continued to exist.
4 Crone provides a smattering of evidence for the early period that some mawālī may not have converted to Islam, emphasizing the point that, according to her, the issue was more of a master/client relationship than of winning new souls for the new religion. However, the phenomenon of non-Muslim mawālī was on the whole very marginal and Crone herself is not categorical as she carefully qualifies the first example given (“. . . presumably a Christian freedman”), Crone, Slaves on Horses, 237 n. 358. While collecting data from the biographical dictionaries — the main source for this article, see below — we did not come across one single non-Muslim mawlā; if the category existed in reality, it did not make it to the biographical dictionaries.
9 This project was formerly known as the Netherlands Ulama Project when it started with a grant from the Dutch government in the 1990s. There are now two databases, a general one and a specific one. The general one is the main database used in this research and which is a representative sample of the entire group of ulama during the first four centuries of Islam. The other, specific database includes all known linguists/grammarians who lived in the same period and additionally has all lines of transmission of knowledge amongst these grammarians. For more information about the set-up of the research project and some results, see Nawas and Bernards, A Preliminary Report; Bernards and Nawas, The Geographic Distribution; Nawas, The Development of the Islamic Religious Sciences; idem, The Contribution of the Mawālī; idem, The Birth of an Elite.
10 A so-called “stratified sample,” each stratum being one of the five main Islamic religious sciences (ḥadīth, tafsīr, qirāʾa, naḥw, and fiqh); on sampling techniques see Blalock, Social Statistics. Almost all published classical Arabic biographical dictionaries were used to collect as much information as possible for each sampled individual. These dictionaries range from the earliest, Ibn Saʿd’s (d. 230/845) Ṭabaqāt, to the much later Shadharāt al-dhahab compiled by Ibn al-ʿImād (d. 1089/1679).
12 Our research design stipulated that the researchers only record what the compilers of the biographical dictionaries explicitly state, keeping guesswork, reading between the lines and other forms of speculation to a minimum.
13 We can use this one estimate, the percentage of 10.5, to generalize about the population but more information is given by using a so-called 95% Confidence Interval, which is customary practice in statistics. What this interval tells us — in order to explain the principle — is that, purely theoretically speaking, 95% of all possible samples drawn from a population will each give an estimate whose value will be between percentage x and percentage y, in 95 out of 100 samples drawn. The corresponding 95% Confidence Interval for our data lies between 4.2% and 16.8% which means that the estimate for the percentage of all mawālī who had mawālī as patrons during the first four centuries of Islam very probably lies between 4% to 17%, which still renders the phenomenon peripheral.
14 Some secondary literature suggests that the Abbasid Revolution of 132/750 was, at least in part, due to the mawālī who wanted equal rights for themselves. Humphreys, Islamic History, 104-127, discusses the literature about the pros and cons of the hypothesis that the mawālī were primarily responsible for the Abbasid Revolution.
15 The finding also indicates that the “revolution” brought about political change — one dynasty being supplanted by another — rather than social. Consequently, the Abbasid “takeover” or “coup d’état” is perhaps more appropriate than using the term “revolution” since societal changes were limited, at least on the aspect of the patronate. Rejecting the idea that the Abbasid “Revolution” brought about equality for all fits much better with the data displayed in Figure 1 that indicate that real societal change occurred in the later second and especially third Islamic centuries, not at the beginning of the second.
16 This procedure, which allows duplicates, is the correct method when generalizing from sample to population based on strata (in our case, the Islamic religious sciences, each being a stratum). Of the 1,049, forty-six ulama were drawn twice. See Blalock, Social Statistics.
17 The mawlā of a mawlā who was drawn twice by chance was ʿAbd al-Razzāq b. Hammām al-Ṣanʿānī, first as someone active in ḥadīth and another time as someone active in tafsīr.
18 Ibn Ḥajar, Tahdhīb, i, 312-313 (entry no. 652); Dhahabī, Siyar, vi, 278-280 (121); Ibn Ḥibbān, Thiqāt, vi, 62; Ibn al-ʿImād, Shadharāt, i, 217; Ibn Abī Ḥātim, Jarḥ, ii, 275-276 (990); Bukhārī, Taʾrīkh, i, 431-432 (1388); Mizzī, Tahdhīb, iii, 277-286 (531); Khalīfa, Ṭabaqāt, 220.
20 Ibn Ḥajar, Tahdhīb, x, 218-220 (441); Mizzī, Tahdhīb, xxviii, 303-312 (6104); Dhahabī, Taʾrīkh, Year 152 ah, 625-631; Dhahabī, Siyar, vii, 5-18 (1); Ibn Abī Ḥātim, Jarḥ, viii, 255-257 (1165); Ibn Ḥibbān, Thiqāt, vii, 484; Ibn Saʿd, Ṭabaqāt, v, 546; Jaʿdī, Fuqahāʾ Yaman, 66; Bukhārī, Taʾrīkh, vii, 378-379 (1631); Khalīfa, Ṭabaqāt, 520; (2665); ʿIjlī, Thiqāt, 435 (1611); Ibn Ḥibbān, Amṣār, 192 (1534); Ibn al-ʿImād, Shadharāt, i, 235; Ibn Qutayba, Maʿārif, 506.
21 Ibn Ḥajar, Tahdhīb, vi, 30 (54); Ibn Ḥibbān, Thiqāt, vii, 18; Mizzī, Tahdhīb, xvi, 144-145 (3573); Ibn Abī Ḥātim, Jarḥ, V, 173 (807); Bukhārī, Taʾrīkh, v, 202 (838).
22 Ibn Ḥajar, Tahdhīb, vi, 278-281 (611); Ibn ʿAsākir, Tārīkh, xxxvi, 160-193 (4039); Ibn Abī Ḥātim, Jarḥ, vi, 38-39 (204); Mizzī, Tahdhīb, xviii, 52-62 (3415); ʿIjlī, Thiqāt, 302 (1000); Jaʿdī, Fuqahāʾ Yaman, 67-68; Ibn Abī Yaʿlā, Ḥanābila, i, 209 (280); Ibn Ḥibbān, Thiqāt, viii, 412; Dhahabī, Siyar, ix, 563-580 (220); Bukhārī, Taʾrīkh, vi, 130 (1933); Ibn Saʿd, Ṭabaqāt, v, 548; Khalīfa, Ṭabaqāt, 521 (2673).
25 Qāḍī ʿIyāḍ, Tartīb, i, 523-528 (97); Ibn Ḥibbān, Thiqāt, viii, 347; Mizzī, Tahdhīb, xv, 191-194 (3371); Dhahabī, Taʾrīkh, Year 214 ah, 220-222 (210); Ibn Khallikān, Wafayāt, iii, 34-35 (323); Dhahabī, Siyar, x, 220-223 (57); Ibn Ḥajar, Tahdhīb (i), v, 289-290 (489); Bukhārī, Taʾrīkh, v, 142 (428); Ibn Saʿd, Ṭabaqāt, vii, 518.
27 Ibn Ḥajar, Tahdhīb, iv, 16-17 (23); ʿIjlī, Thiqāt, 182-183 (537); Ibn al-ʿImād, Shadharāt, ii, 53-54; Mizzī, Tahdhīb, x, 391-395 (2235); Ibn Ḥibbān, Thiqāt, viii, 266-267; Ibn Abī Ḥātim, Jarḥ, iv, 13-14 (49); Dhahabī, Siyar, x, 327-330 (80); Bukhārī, Taʾrīkh, iii, 512-513 (1703).
28 Ibn Ḥajar, Tahdhīb, xi, 208-209 (388); Qāḍī ʿIyāḍ, Tartīb, i, 528-529; Mizzī, Tahdhīb, xxxi, 401-404 (6858); Ibn Ḥibbān, Thiqāt, ix, 262; Dhahabī, Siyar, x, 612-615 (210); Ibn Abī Ḥātim, Jarḥ, ix, 165 (682); Bukhārī, Tārīkh, viii, 285 (3019).
30 Ibn Ḥajar, Tahdhīb, i, 47-48 (91); Mizzī, Tahdhīb, i, 387-391 (68); Ibn Ḥibbān, Majrūḥīn, i, 149; Dhahabī, Siyar, xii, 317-323 (122); Ibn al-ʿImād, Shadharāt, ii, 147; Subkī, Shāfiʿiyya, ii, 26 (5); Ibn Abī Ḥātim, Jarḥ, ii, 59-60 (91).
31 Ibn Ḥajar, Iṣāba, vii, 236, who discusses the name and states that this Companion had witnessed the conquest of Egypt.
33 Crone, Mawlā, reminds us that the patronate was between individuals; one could not become the mawlā of a tribe sec. Nonetheless, only two of the nine men (Maʿmar and Baḥshal) have straightforward listings of being the mawlā of a specific other individual mawlā.
References
Primary Sources
Al Bukhārī Kitāb al Taʾrīkh al kabīr 1941 5 Hyderabad Maṭbaʿat Dāʾirat al Maʿārif al Niẓāmiyya 9 vols. repr. Hyderabad: Maṭbaʿat Dāʾirat al Maʿārif al ʿUthmāniyya, 1377/1958; repr. Beirut: Dār al Kutub al ʿIlmiyya, n.d.
Al-Dhahabī Shams al-Dīn Muḥammad b. Aḥmad b. ʿUthmān Shuʿayb al- Arnaʿūṭ Siyar aʿlām al-nubalāʾ 1993 Beirut Muʾassasat al-Risāla 25 vols.
Al-Dhahabī Shams al-Dīn Muḥammad b. Aḥmad b. ʿUthmān ʿUmar b. ʿAbd al-Salām Tadmurī Taʾrīkh al-Islām wa-ṭabaqāt al-mashāhīr wa-l-aʿlām 1989-2000 Beirut Dār al-Kutub al-ʿArabī 52 vols.
Ibn Abī Ḥātim al Jarḥ wa l taʿdīl 1360/1941 1371/1952 Hyderabad Jamʿiyyat Dāʾirat al Maʿārif al ʿUthmāniyya 9 vols. repr. Beirut: Dār Iḥyāʾ al Turāth al ʿArabī, n.d.
Ibn Abī Yaʿlā Abū l-Ḥusayn M. al-Faqī Ṭabaqāt al-Ḥanābila Cairo Maṭbaʿat al-Sunna al-Muḥammadiyya 2 vols. (+ 2 vols. dhayl) n.d.
Ibn ʿAsākir ʿUmar b. Gharāma al-ʿAmrawī Tārīkh madīnat Dimashq 1995-2000 Beirut 80 vols.
Ibn Ḥajar, Shihāb al-Dīn Aḥmad b. ʿAlī b. Ḥajar al-ʿAsqalānī Kitāb Tahdhīb al-tahdhīb 1984 Beirut Dār al-Fikr 14 vols.
Ibn Ḥajar, Shihāb al-Dīn Aḥmad b. ʿAlī b. Ḥajar al-ʿAsqalānī ʿAlī Muḥammad al-Bajāwī al-Iṣāba fī tamyīz al-ṣaḥāba 1992 Beirut Dār al-Jīl 8 vols.
Ibn Ḥibbān, Muḥammad b. Ḥibbān b. Aḥmad Abī Ḥātim Kitāb al-Thiqāt 1973 Hyderabad 10 vols.
Ibn Ḥibbān, Muḥammad b. Ḥibbān b. Aḥmad Abī Ḥātim Fleischhammer Manfred Mashāhīr ʿulamāʾ al amṣār 1959 Wiesbaden Franz Steiner [Bibliotheca Islamica 22]
Ibn Ḥibbān, Muḥammad b. Ḥibbān b. Aḥmad Abī Ḥātim Maḥmūd Ibrāhīm Zāyid Kitab al-Majrūḥīn min al-muḥaddithīn wa-l-ḍuʿafāʾ wa-l-matrūkīn Mecca Dār al-Bāz 3 vols. n.d.
Ibn al-ʿImād ʿAbd al-Qādir al-Arnāʾūṭ & Maḥmūd al-Arnāʾūṭ Shadharāt al-dhahab fī akhbār man dhahab 1986-1995 Beirut Dār al-Āfāq al-Jadīda 8 vols. n.d.; ed. by ʿAbd al-Qādir al-Arnāʾūṭ and Maḥmūd al-Arnāʾūṭ, 11 vols., Damascus and Beirut: Dār Ibn Kathīr
Ibn Khallikān Iḥsān ʿAbbās Wafayāt al-aʿyān Beirut 8 vols. n.d.
Ibn Qutayba Tharwat al-ʿUkkāsha al-Maʿārif 1992 Cairo
Ibn Saʿd Iḥsān ʿAbbās Ṭabaqāt al-kubrā Beirut 9 vols. n.d.
Al-ʿIjlī Abū l-Ḥasan Aḥmad b. ʿAbdallāh b. Ṣāliḥ ʿAbd al-Muʿtā Qalʿajī Tārīkh al-thiqāt 1984 Beirut Dār al-Kutub al-ʿIlmiyya
Al-Jaʿdī ʿUmar b. ʿAlī b. Samura Fūʾād Sayyid Fuqahāʾ Yaman 1981 Beirut Dār al-Kutub al-ʿIlmiyya
Khalīfa b. Khayyāṭ Akram Ḍiyāʾ al-ʿUmarī Ṭabaqāt 1966 Baghdad
Al-Mizzī Jamāl al-Dīn Abū l-Ḥajjāj Yūsuf Bashshār ʿAwwād Maʿrūf Tahdhīb al-kamāl fī asmāʾ al-rijāl 1993 Beirut Muʾassasat al-Risāla 35 vols.
Al-Qāḍī ʿIyāḍ Aḥmad Bukayr Maḥmūd Tartīb al-madārik wa-taqrīb al-masālik li-maʿrifat aʿlām madhhab Mālik Beirut Dār Maktabat al-Ḥayāʾ 3 vols. n.d.
Al-Subkī Maḥmūd Muḥammad al-Ṭanāḥī & ʿAbd al-Fattāḥ Muḥammad al-Ḥilw Ṭabaqāt al-shāfiʿiyya al-kubrā Cairo Dār Iḥyāʾ al-Kutub al-ʿIlmiyya 10 vols. n.d.
Secondary Sources
Al-Badrāwī Riyāḍ ʿAbd al-Ḥusayn Al-Mawqif al-umawī min al-mawālī: Bayn al-shubahāt wa-l-wāqiʿ al-tārīkhī 2011 Damascus Tammūz li-l-Ṭibāʿa wa-l-Nashr wa-l-Tawzīʿ
Bernards Monique & Nawas John The Geographic Distribution of Muslim Jurists during the First Four Centuries a.h. Islamic Law and Society 2003 10 2 168 181
Bernards Monique & Nawas John Patronate and Patronage in Early and Classical Islam 2005 Leiden Brill
Bi-l-nūr, Īmān ʿAlī Dawr al-mawālī fī suqūṭ al-dawla al-umawiyya (41-132 AH/661-750 AD) 2008 Benghazi, Libya Dār al-Kutub al-Waṭaniyya
Blalock Hubert M. Social Statistics. Second Edition 1972 New York McGraw-Hill
Bulliet Richard W. Conversion to Islam in the Medieval Period: An Essay in Quantitative History 1979 Cambridge, MA Harvard University Press
Bulliet Richard W. Bernards Monique & Nawas John Conversion-Based Patronage and Onomastic Evidence in Early Islam Patronate and Patronage in Early and Classical Islam 2005 Leiden Brill 246 262
Crone Patricia Bernards Monique & Nawas John Mawālī and the Prophet’s Family: An Early Shīʿite View Patronate and Patronage in Early and Classical Islam 2005 Leiden Brill 167 194
Crone Patricia Mawlā Encyclopaedia of Islam, Second Edition 1954-2005 Leiden Brill 874 882 13 vols.
Crone Patricia Roman, Provincial and Islamic Law: The Origins of the Islamic Patronate 1987 Cambridge Cambridge University Press
Crone Patricia Slaves on Horses: The Evolution of the Islamic Polity 1980 Cambridge Cambridge University Press
Gilbert J. Institutionalization of Muslim Scholarship and the Professionalization of the Ulama in Medieval Damascus Studia Islamica 1980 52 105 135
Humphreys R. Stephen Islamic History. A Framework for Inquiry 1991 Princeton, NJ Princeton University Press
Juda Jamal Die sozialen und wirtschaftlichen Aspekte der Mawālī in frühislamischer Zeit 1983 PhD, Tübingen Arabic transl., al-Awḍāʿ al-ijtimāʿiyya wa-l-iqtiṣādiyya li-l-mawālī fī ṣadr al-islām, 1989
Juda Jamal Bernards Monique & Nawas John The Economic Status of the Mawālī Patronate and Patronage in Early and Classical Islam 2005 Leiden Brill 263 277
Makdisi George The Rise of Colleges: Institutions of Learning in Islam and the West 1981 Edinburgh Edinburgh University Press
Mitter Ulrike Das frühislamische Patronat. Eine Untersuchung zur Rolle von fremden Elementen bei der Entwicklung des islamischen Rechts 1999 PhD, Katholieke Universiteit Nijmegen
Nawas John A. The Birth of an Elite: Mawālī and Arab Ulama Jerusalem Studies in Arabic and Islam 2006 31 74 91
Nawas John A. Günther Sebastian The Contribution of the Mawālī to the Six Sunnite Canonical Ḥadīth Collections Ideas, Images, and Methods of Portrayal: Insights into Classical Arabic Literature and Islam 2005 Leiden Brill 141 151
Nawas John A. The Development of the Islamic Religious Sciences in Early and Classical Islam: A Comparative Approach al-Masāq 1999 11 159 169
Nawas John A. How Sunnism Came to Be: The Appellation Ṣāḥib Sunna in Classical Islam forthcoming
Nawas John A. & Bernards Monique Vermeulen U. & Van Reeth J. M. F. A Preliminary Report of the Netherlands Ulama Project (nup): The Evolution of the Class of ʿulamā’ in Islam with Special Emphasis on the Non-Arab Converts (Mawālī) from the First through Fourth Century a.h. Law, Christianity and Modernism in Islamic Society: Proceedings of the Eighteenth Congress of the Union Européenne Des Arabisants Et Islamisants 1998 Leuven Peeters 97 107