Marriage in the Tribe of Muhammad

A Statistical Study of Early Arabic Genealogical Literature

Majied Robinson

• Publisher: De Gruyter
• Year: 2020
• ISBN:
  9783110624236
• Edition: 1
• Topic: Social Sciences
• Subtopic: Islamic Studies
• Index: Social Sciences

This study examines the marital data preserved within the Arabic genealogical works of the early ninth century CE in order to better understand the tribal relationships of the pre-Islamic Quraysh (the Arabic tribe to which Muhammad belonged). The research establishes the accuracy of the Nasab Quraysh ( Genealogy of the Quraysh ) and informs a more nuanced analysis of the politics of the Central Hijaz into which Islam was born.

1 Introduction

What follows is a demonstration of how prosopographical approaches can be used to turn the Arab genealogical literary tradition (Ar: nasab) into a social history of Muḥammad’s tribe for the period 500 – 750 CE. These approaches will primarily involve tracking the origins and statuses of the mothers of Qurashī tribespeople over time. Through this our study will confirm some elements of the existing historical narrative for the period while contesting others. The investigation will also tell us a great deal about the historiography of the nasab tradition in general, and the Nasab Quraysh of al-Zubayrī in particular.

Statistical analyses of the marriage data will show that the first Muslims married fewer Qurashī women than their non-Muslim fathers and instead married brides taken from a much wider geographical range. This correlates with elements of the traditional narrative such as the ostracisation of the early Muslims, their anti-tribal ideology and their conquest of the Arabian Peninsula. In the Umayyad era we see a further correlation between the results of our analysis and the traditional narratives; this is in the form of an increase in the numbers of children born to non-Arab slave women. The timing of this change matches exactly the period of the military conquests which brought with them an influx of captive women.

These concurrences are welcome because they tend to support the conclusion (arrived at separately on the basis of internal evidence and anthropological parallels) that our methodology is a sound means of extracting, structuring and interpreting the data. This means that where we find divergences between the prosopography and the traditional narratives we are potentially looking at original findings in what is normally considered a historiographically ‘dark’ period of history.

One of the most significant of these divergences concerns the pre-Islamic Quraysh, whose marriage behavior – as revealed by quantitative analysis – does not correlate with that of the Quraysh as they exist in many of our most familiar primary and secondary sources. It will be shown that the marriages made by the Quraysh in this period are not characteristic of a tribe at the heart of a far-flung trading empire, or even that of a tribe that enjoyed undisputed eminence in the Ḥijāz; in fact, our analysis highlights the parochiality of the Quraysh as they existed before and during Muḥammad’s lifetime.

Additionally, the data also reveal that concubinage was enthusiastically taken up by all sections of the Quraysh as soon as slave women became available; this was not a gradual change in behaviour that only became prevalent in the ʿAbbāsid era as previously believed. The data also go some way in showing that there is little evidence that the children of concubine unions were treated as being of lower status than their full-born siblings. Furthermore, it will be demonstrated that the Umayyad caliphs exhibited remarkable dynamism and creativity when it came to changing their marriage practices. Overall, it will repeatedly be shown that the prosopographical approach has the capacity to suggest a number of new avenues of inquiry unavailable to us had we relied solely on more traditional methodologies.

The structure of this research is shaped entirely by the nature of the sources. The Quraysh of the sixth to mid-eighth centuries have been chosen because the genealogical data available on this cohort in this time period enjoy a unique amount of detail – indeed, it will be argued that in terms of volume and nature these data are without comparison in all pre-modern human historiography. The Arabic historical sources record the names of approximately 3,000 Qurashīs of this period for whom we know the names of their fathers and at least the status of their mothers; no other tribal society known to this researcher preserved this level of genealogical information in anything like the same fashion.

The choice of marriages – specifically child-bearing marriages – as our principal data category is closely related to the nature of the sources. Genealogical memory can be very accurate and consistent in some respects, but it can also be highly divergent and contradictory in others. By drawing on anthropological studies of tribally organised societies we will show that there are good grounds for believing that the child-bearing marriage data for this period fall into the accurate and consistent category of genealogical memory. This will be further supported by a study on the internal and external consistency of the nasab data itself.

Finally, the nature of the data has led us to use prosopography as the primary methodological school of approaches. While the genealogical marriage data are extensive and arguably very accurate in general terms, it does not mean that we can be sure that any single record is true. The quantitative approaches familiar to prosopographers can overcome this problem as they mitigate the effects of small numbers of erroneous accounts. Prosopography is also useful because it can turn genealogy into a collection of individual choices (the decision to have children with a concubine rather than a wife, for example) which can then be studied in terms of different sub-groups and how these choices change over time. Through this we can understand how Qurashīs modified their marriage behaviour in response to the events of a particularly dramatic two and a half centuries of their history.

1.1 Prosopography and early Islamic history

Prosopography is taken here to mean the collective study of a group of historical actors.¹ The methodological approaches of most concern in this study are those that employ quantitative analysis to uncover changes in the social structure of a status group.²

The idea that these sorts of prosopographical techniques could be useful to historians of pre-modern Islam has over four decades of heritage. An early pioneer whose work led to published output was Richard Bulliet who published “A Quantitative Approach to Medieval Muslim Biographical Dictionaries” in 1970;³ he followed this with the 1972 work The Patricians of Nishapur⁴ and then Conversion to Islam in the Medieval Period,⁵ published in 1979. In all cases he applied quantitative techniques to biographical dictionaries of medieval Islamic ʿulamā. Some of the conclusions he drew based on this data analysis were bold – perhaps too bold – but the influence of the works is undeniable.⁶

The timing of the publication of these studies was in part related to the emergence of computers as research tools, but Bulliet’s work was also part of a wider debate on how Islamic historical sources can be used in reconstructing Islamic social history. His starting point was the hypothesis that the entries within the biographical dictionaries contained small amounts of stable information such as marriage behaviour and genealogy. By mining this information, the resulting database could then be used to construct a social history of medieval Islamic societies. Like the work in this present study, his research approach was shaped by the sources.

The methodologies used by Bulliet were not only of interest to social historians of the medieval era; those with interests in earlier times could also see the appeal of the quantitative approach. In 1980, Patricia Crone published her own prosopography of Umayyad and ʿAbbāsid governors in Slaves on Horses, declaring in the opening pages that “early Islamic history has to be almost exclusively prosopographical” (emphasis hers).⁷ Again, this declaration and the prosopographical study were shaped by the sources; in this case the governor lists which were extensive and (according to Crone) accurate, but limited in terms of what each individual piece of information could tell the historian. Crone’s use of prosopography was designed to turn this ‘shallow’ data into something more useable by giving it breadth.

In the background loomed larger but less successful projects. The 1960s saw the birth of the computerised (using punch cards) Onomasticon Arabicum which was based on the vast collection of biographical data that had been collected by Giuseppe Gabrieli and Leone Caetani in the first 14 years of the twentieth century. This project gathered pace in the 1970s, but constant changes to computing technologies (from the punch cards, to magnetic tapes, to Microsoft Access and to the website that exists today) along with the difficulties of getting early computer programmes to handle Arabic script has meant that the project has been perpetually beset by problems.⁸ We should also mention Charles Pellat’s “Peut-on connaître le taux de natalité au temps du Prophète? A la recherche d'une méthode” (published 1971) which was an ambitious but unconvincing attempt to calculate Prophetic-era birth rates using samples taken from the Nasab Quraysh of al-Zubayrī.⁹

But neither Bulliet’s work nor Crone’s Slaves on Horses, or even the Onomasticon Arabicum led to a glut of prosopographical works concerned with Late Antique¹⁰ or Early Medieval Islam. The Onomasticon ultimately yielded very little in terms of published output; it serves as a cautionary tale for the prospective prosopographer when one considers the enormous amount of effort that it has consumed. At the same time, Crone’s own prosopographical work did not extend far beyond Slaves on Horses, and very few scholars have followed Bulliet’s lead and taken a long-term commitment to prosopographical or statistical methods for this historical period (in this category we can only really point to Michael Lecker).¹¹ As for the successful use of these methodologies when occasionally deployed by established academics there are few other than Harald Motzki’s “The Role of non-Arab Converts in the Development of Early Islamic Law”¹² and Fred Donner’s “Tribal Settlement in Basra During the First Century A.H.”¹³

As for the younger generation of scholars, although there is a great deal of enthusiasm for the approach, this enthusiasm has only translated into a small number of recent works. Most notable of these are Fuʾad Jabali’s The Companions of the Prophet: A Study of Geographical Distribution and Political Alignments (from 2003) and Asad Ahmed’s The Religious Elite of the Early Islamic Ḥijāz (from 2011).¹⁴ Other academic studies for the Late Antique period include Salih Said Agha’s Revolution Which Toppled the Umayyads (from 2003)¹⁵ and Michael Ebstein’s “Shurṭa Chiefs in Baṣra in the Umayyad Period: A Prosopographical Study” (from 2010).¹⁶ This latter paper drew on data gathered and structured by the Jerusalem Prosopography Project run by Michael Lecker, but sadly is one of very few published works to have used this resource as a means for quantitative study.

The aspiring prosopographer therefore faces a significant problem: there are very few established methodologies to build on. While the works of Bulliet et al. mentioned in the section above are undoubtedly prosopographical studies of pre-modern Islam, there is little of their methodology that can be directly borrowed and applied to other sources describing different categories of people. For similar reasons, prosopographical studies of non-Islamic cultures are also of limited use. The tools of this trade therefore have to be fashioned en route and they must be accompanied with adequate explanation so that others may also make use of them.

In order for readers to be able to critically engage with the conclusions drawn from any novel research methods, the workings of these methods must first be shown and its tools fully understood. It is for these reasons that methodology and related issues account for a substantial portion of this book. While the conclusions drawn from this research on the subjects of marriage and identity are substantive, the methodology is at least as important; without detailing its underpinnings we are denying scholars the ability to easily replicate this research in the future.

1.2 The sources of early Islamic history

A convincing prosopographical methodology rests on the appropriateness of its source material and, as anyone with a degree of familiarity with the historiography of early Islamic history will know, this source material is very complicated. These issues have been discussed extensively in the secondary literature, but in essence the debate centres on what can be considered as primary evidence. Much of our extant historiographical data are presented in a primary, or even a documentary fashion (e. g. eyewitness statements, participant accounts, letters and contracts) but not only was the bulk of this data preserved centuries after the events described, they also bear the marks of repeated retellings and reformulations.¹⁷

The scholarly responses to this are typically schematised as falling into two broad categories. On the one side, some historians of the early Islamic period have seen enough of the contradictions and anachronisms of the traditional historical narrative t...

Table of contents

• Title Page
• Copyright
• Contents
• Preface
• List of Tables, Graphs and Maps
• Notes on Citations and Transliterations
• 1 Introduction
• Part I Data, methods and historiography
• Part II Data analysis and application
• Appendix One – Hajīn marriage
• Appendix Two – Marriages of the brothers of Muʿāwiya and Marwān I
• Bibliography
• Person Index