Navigation – Plan du site

AccueilNuméros54Acts of Protection in Early Islam...Introduction

Acts of Protection in Early Islamicate Societies


Acts of Protection in Early Islamicate Societies
Edmund Hayes et Eline Scheerlinck
p. 3-14

Texte intégral

  • 1 This special issue emerges from research done in the context of the project “Embedding Conquest: Na (...)
  • 2 For example, Grafe, Irigoin, 2012.
  • 3 For recent contributions to the conception of empire in the Islamic world, see, for example Nef, Ti (...)

1This special issue focuses on varied mechanisms of social protection in the context of social and political networks in the early Islamic empire up to and following the collapse of the caliphate in the ʿAbbasid period.1 Under the broad concept of “protection” we have brought together diverse, but comparable relationships within which risk is managed and problems are solved at different levels of society, and in difficult cultural and geographical spheres. When considering how societies functioned in the longue durée following the Arab‑Muslim conquests, it is necessary to understand which dynamics were local, and which were extensions of more widespread cosmopolitan imperial idioms. In this special issue, we have gathered contributions relating to diverse linguistic and cultural contexts based on a wide variety of source types in which protection dynamics can be studied. In doing so, we are intentionally bringing together conversations across different domains, with the conviction that society was not only ordered by governmental policy, but by the interaction of customs and norms which evolved organically from earlier models. Insights from the field of Empire Studies have long suggested that the old top-down model of imperial administration should be nuanced with an understanding of empires as constituted by a diverse coalition of “stakeholders”,2 albeit stakeholders whose interactions are structured according to the hegemonic categories and institutions of the imperial elites: in our case the Arab‑Muslim conquerors. In spite of the existence of hegemonic institutions in society, problems were solved, not only within these institutions, and not only through “official” channels. Official administrative procedures existed within a spectrum of problem-solving mechanisms from formal to non-formal which overlapped and were embedded in each other.3

1. General Definitions

2For the purposes of this publication, we understand “protection” as a mechanism to avoid a threat or danger, usually to life, property or one’s way of living. The protector and the protectee enter into an asymmetric relationship based on the protector’s greater ability to prevent a threat due to his or her higher position within society or administration, greater coercive power (whether legitimate or illegitimate), or possession of certain kinds of social positionality or symbolic capital which enable intervention. Mechanisms of protection can be seen throughout society, and operate within all societal, governmental and administrative structures. A request for protection can be to protect oneself or a third party, an individual or a group. Protection can be asked or given for one’s person, property, family, or social status. The protectees can thus be protected from impoverishment, but also from imprisonment, bodily harm or death. Acts of protection can have a physical aspect, performed by a protector or protectee, such as going to a place of asylum, or closing the door of a mosque. Places of protection or asylum are varied, and might include mosques; the middle of a square; one’s own neighborhood; under a woman’s clothes (see Van Gelder’s contribution); or on the “right” side of the border. On the other hand, acts of protection may manifest themselves in a verbal act, which can again be performed by protectors and protectees alike. This can include writing up documents such as letters, safe conducts, or responding to petitions. Documents do not just inform, but perform and channel the act of protection. Requesting or giving protection can also involve an oral act. In the law of war, the very utterance of a request for or offer of protection as an amān is conceived of as having legally binding consequences. Protection can, in some cases, involve secrecy or escape: fleeing or hiding from one’s oppressors, while in other cases the public nature of acts of protection (such as oaths) is crucial. While God was understood to be the greatest protector by the actors in this historical context, we do not, here, focus on understanding of God as protector in answer to prayers or invoked in magic items such as talismans. We have chosen to focus instead on protection offered by human actors. However, the social codes that trigger protection relationships are very often underpinned by religion, and as such God can be invoked as the ultimate arbiter of a protection relationship, for example in a sworn oath or a protection document (see the contributions of Huseini and Scheerlinck).

  • 4 The nearest is perhaps Lambert, Rollason, 2009.
  • 5 An excellent volume like Bernards, Nawas, 2005, provides an understanding of diverse instances of a (...)

3While there is no generally recognized field of “protection studies”,4 the concept of protection intersects with many existing discussions taking place across various fields. The most relevant areas of society where protection relationships can be effectively studied are in government and administration, legal institutions (including, but not limited to religious law), military contexts and other contexts in which violence is wielded expertly and asymmetrically, but also in the realm of social custom. Protection mechanisms are visible in institutions of patronage, military power, taxation regimes, and various different kinds of religious and political authority. As such, then, the understanding of protection mechanisms is relevant to studies of patronage, peace and security, social welfare, fiscal policy, risk management, and the philosophical discussion of what kinds of social contract underpin society. Why, then, not just study one of these contexts in its own right, rather than bringing together such varied contexts? The answer is that the very diversity of the early Islamic empire calls for a malleable term like “protection” in order to be able to juxtapose comparable developments occurring simultaneously in different places and at different levels of society. The languages, contexts and institutions present in the case studies in this issue are diverse and would not come to be placed side-by-side in a focused study on a particular institution represented by a single word in one of the languages of the early Islamic empire. Thus, for example, the study of an institution like walāʾ, clientage, while important for understanding how Islamic law mediates social relations, would stop short of enquiring into potential structural similarities between different kinds of protection relationship existing in different languages and legal traditions, or placing them into the global context of diverse milieus governed by a common Arab-Islamic administration.5

  • 6 A good example is found in Elaine Clark’s indication of the importance of land use as the crucial f (...)

4In determining how to approach protection relationships it is important to enquire into the overall social structures in which these relationships are forged. The power of the protector is not usually raw coercive power, but instead the powerful protector draws his or her power from the ability to intervene in one or more key social field, and manipulate social, political, economic or legal networks to achieve a result. Thus, the understanding of a protection relationship requires an understanding of the social structures which form its context, and the way in which the various relevant social fields interact to enable it to successfully achieve its intended aim.6

2. Pre-Islamic Empires and Protection

5The early Islamic empire was characterized by its fusion of different cultural spheres which converged under Arab‑Muslim rule: late Roman, Sassanian, Arabian, Islamic, and myriad local cultural forms. Each of these spheres came to overlap as the Arab‑Muslim conquerors brought their cultural and administrative norms with them into the conquered territories to influence and be influenced by the habits of the conquered populations. As such, the societies of the early Islamic empire were sustained by diverse sets of overlapping institutions and customs for ensuring individual and communal safety. Protection relations within the early Islamic empire are preceded by protection relations in its predecessors: the imperial regimes of the Sassanians and the Eastern Roman empire, both of which projected certain types of social models within which protection could be fostered or withheld.

  • 7 Garnsey, 2010.
  • 8 Lena Salaymeh’s conception of zakāt‑based “citizenship” is not applicable here, as it excludes the (...)
  • 9 Wickham, 2005, pp. 56–150.
  • 10 Wickham, 2005, p. 143; Crone, 1980, p. 51.

6In the case of the Roman empire, a key aspect influencing one’s ability to offer and receive protection was citizenship.7 In this respect, the Islamic empire was somewhat different,8 developing no codified measure to legally formalize an individual’s integration into a class of recognized stakeholders. In other ways, however, the social and political structures of the early Muslim empire represented continuity as well as rupture with the Roman system, certainly when compared with other post-Roman successor states like the Germanic kingdoms of post‑Roman Europe. Chris Wickham argues that Arab-Islamic administration was characterised by a high degree of continuity with regards to the Roman fiscal system: that is, the early Arabs maintained a complex and bureaucratic taxation‑based system, in contrast to the Germanic kingdoms in Europe which tended towards the distribution of land amongst the conquering elites who thereafter became the landed aristocracy upon whom rulers had to rely for support or fear for their ability to replace them.9 The macro‑structures of the Muslim empire would initially remain that of a bureaucratic state supported by agricultural taxation. Under the Umayyad and early ʿAbbasid administrations, the state more-or-less maintained its tax base, and thereby its ability to pay its army, obviating the need to rely on militarized landowners. This fact brought with it implications for the question of to whom one could turn for help in society if one’s family was terrorized, or if one’s tax burden was unreasonably high. Thus, as far as influence in the state was concerned, the key nodes of influence for matters of protection and patronage tended to revolve around bureaucratic, rather than aristocratic modes of intervention. Non‑bureaucratic modes of problem‑solving were, of course, also ubiquitous. Wickham sees a reduction in the social opportunities for forging protection relations under Muslim rule (or at least in Muslim Egypt where documentation is most abundantly preserved) due to the collapse of another Roman set of institutions: the patronage of great landed patrons, with the result that the majority of the population had fewer contacts with the powerful. Wickham approvingly cites Patricia Crone in her comment that, under the Umayyads, as a result of the weakening of local aristocracies, “because the countryside was thus denuded of its protective network, flight from the land replaced the traditional search for a rural patron as the primary mode of tax‑evasion”.10 Wickham goes on to suggest that the Arabs quarantined themselves from local population in such a way that limited the extension of Arabian social institutions to the entirety of society:

  • 11 Wickham, 2005, p. 143.

The Arabs did, actually, have an extremely elaborate conception of political patronage, but one usually had to become a convert to be a client (mawlā)… Unlike their predecessors, the Arabs set themselves up as a separated ‘state class’ of recipients of taxation, with no structural social links to taxpayers, whether rich or poor.11

7This idea of the self-quarantining of the Arab ruling class should, however, be seen as attenuated by the inevitable need for rulers to find local partners who themselves benefitted from their association with the military and fiscal overlordship of society. The contributions of Paul and Kennedy here both provide insights into how relationships between ruling elites and their subjects were forged. Whatever the fate of individual institutions of protection, the very success of ongoing Arab‑Muslim rule in Egypt and elsewhere indicates that mechanisms of social integration must have been employed. Certainly, the arrival of new rulers across the post‑Roman and post-Sassanian conquered territories generated new relations, and these relations had to fit within the overall modalities of power and wealth which were concentrated at different points. Wickham’s emphasis on the separation of the Arabs as a ruling class should be seen as the starting point in a transition: as power was concentrated in new hands embedded in new networks, inevitably it took time for new ways of mediating between rulers and ruled. The documents Weber has presented here give suggestions as to the emergence of mechanisms of social integration between Arabs and the conquered in Persia.

  • 12 Morony, 1984, p. 97.
  • 13 Morony, 1984, p. 97.
  • 14 Paul, 1996; Pourshariati, 2004; Pourshariati, 1995.
  • 15 Morony, 1984, p. 98.

8In the ex‑Sasanian domains, as in the ex‑Roman domains, Arab‑Muslim control did not totally dismantle earlier regimes, and local administrators carried some earlier administrative norms with them into the Arab-Muslim administration. Morony, in analysing the administration of early‑Islamic Iraq, notes that there were not only continuities, but also revivals of Sasanian practice in response to practical organizational needs, as well as the predilections of the administrative class.12 However, the destruction of the Sasanian monarchy radically re‑oriented society. Morony notes that “the administrative system in early Islamic Iraq was much less centralized that it had been under the last Sasanians, and most of the hierarchic structure of the royal court was eliminated.”13 Nonetheless many aristocratic families in the ex‑Sasanian east still retained their influence and their connection to the fiscal exploitation of the resources of the land.14 Distinctively Arab institutions that could not be assimilated to Sasanian models, including elements of the Arab tribal system, competed for relevance alongside Sasanian administrative traditions within the “creative adaptability of the Islamic regime.”15 Subjects then, had to navigate multiple frameworks in seeking protection: old, new, and emerging.

3. Arabian Protection Relationships

9Arabian institutions were initially hegemonic in the Islamic empire, but their influence was limited by the numerical minority of the Arabian conquerors. Thus, certain Arabian institutions (like walāʾ) proved ultimately to be not culturally useful to the very different social contexts of the conquered lands, and eventually fell into desuetude. Likewise, certain late-antique or Roman practises (like, for example, the very local Coptic protection letters studied by Scheerlinck) appear to have eventually fallen into disuse with the rise of the new Islamic society in which different procedures, norms and morals were emphasized. Equally, new norms and institutions developed to deal with new circumstances arising from the conquests, as seen here in Weber, Huseini and De la Vaissière’s articles.

10The pre‑Islamic Arabian society which produced the first generations of ruling elites of the early Muslim empire brought with it a number of institutions to regulate asymmetric protection relationships, all of which were implicated in broader structures of tribe and clan, and settled and nomadic lifestyles. Any discussion of the precise nature of pre-Islamic Arabian customs and institutions is fraught with the difficulty of authenticating the evidence for pre‑Islamic Arabian societies, the literary sources for which are heavily filtered through Muslim stereotypes about the non‑Muslim forerunners. Nonetheless, Islamic‑era material provides a fascinating window onto the ways in which pre‑Islamic customs (real or imagined) were navigated as society changed. Presumably, the anecdotes of the “bad old days” of pre‑Islamic Arabia also convey something of the experience of the social transitions made following the Islamic conquests, details which allow us to see something of the social institutions that were influential during the period of the formation of the early Islamic empire. Such anecdotes also tell us much about the construction of appropriate behaviour, and the conceivable limits of social interaction as formulated for a medieval Arab audience far removed from the social contexts in which such stories originated. In this special issue, Van Gelder refers to pre‑Islamic and Islamic‑era anecdotes in Arabic literary discourse, throwing light on the ways in which disguises could be used by those seeking protection. His contribution examines several cases of men dressing in women’s clothes to protect themselves, often to escape a life‑threatening situation. Women can take the initiative or play a (superficially) important role in these stories from the pre-Islamic and early Islamic era, but Van Gelder shows that although women and especially women’s clothes are crucial elements of the story, the protection relationships come down to powerful men protecting other men, through a disguise provided by women’s clothes.

  • 16 Evans‑Pritchard, 1940; Sahlins, 1961; Hoyland, 2001, p. 114.
  • 17 See also Peter Webb’s introduction to Webb, 2019.
  • 18 See Calder, 1988.

11Modern scholarly paradigms of pre-Islamic Arabia, and more generally, nomadic and tribal Arabia before the arrival of modernity, have been relatively consistent from early studies like that of Robertson-Smith, who synthesized the frameworks of comparative religion and comparative Semitics alongside more recent data from tribal Arabia since the age of colonialism and European exploration of Arabia. In the tradition of Robertson‑Smith, Arabian tribal society from pre-Islam to more recent pre-modernity has been conceived as being founded on “segmentary lineages”:16 family units which share a common ancestor (real or imagined), based on an appeal to whom, the various units can be united against a common enemy. In this model, in the absence of a state with a monopoly on violence, protection can only be offered by affiliation to a specific family, clan or tribal unit. Anyone outside of this system of mutual protection is understood to be deeply vulnerable, like the legendary saʿālīk brigands of pre‑Islamic and early Islamic poetry, referred to by Van Gelder in his contribution.17 Within this understanding of the need for affiliation with a particular familial or tribal unit to ensure one’s protection against potential enemies, we can see various key institutions that derive from the pre‑Islamic Arabian context, and which survive in discussions of early Islamic practice: in particular walāʾ, and ǧiwār or iǧāra and the various kinds of oaths and alliances which held society together in the absence of a strong state.18 Webb’s contribution in this special issue brings together Islamic legal discourse, historical narratives and poetry, to assess women’s position in early Islam in acting as the protector in these Islamic institutions of protection. Webb argues that representations of women protecting a man, regardless of their possible connections to historical events, are literary tropes emphasising the cowardly and immoral character of the male protectees.

4. The Islamic Empire and the Diversity of Post Conquest Societies

  • 19 Lecker, 2004; Arjomand, 2009.
  • 20 Hoyland, 2001, pp. 116–117.
  • 21 See, for example, Hayes, 2017.

12Muslims saw themselves as having forged new relationships out of the stereotyped feuding tribal society that preceded Islam. In the post‑hiǧra settlement of Yathrib/Medina, Muḥammad is famously said to have enforced his followers to relinquish old tribal and familial ties with those who did not embrace Islam. As a replacement, the new settlers were said to have sworn brotherhood to their new ‘kin’ within Islam. In the umma document (sometimes known as “The Constitution of Medina”) members of the new community, or umma, swore a pact of mutual protection.19 This mutual protection also had the flipside of enabling acts of aggression and expansion against the umma’s antagonists. Hoyland, invoking the model of segmentary lineage, sees the rise of Islam as an example of the high-level integration of the (normally distinct) segments of a tribal society, eventually leading to the evolution of the state based on the motivation of profit and wealth.20 In this reading, mutual protection is the pre-condition for aggressive expansion. However, in the long term, the relatively egalitarian, post-tribal formulation of the umma as the community of Muslims became a powerful ideological mechanism for the integration of the conquered peoples into the ruling classes, in spite of the counter-claims of Arabian solidarity.21

  • 22 See also Robinson, 2013.
  • 23 The understanding of Muslim relations with non-Muslim subjects as a relationship of protection (dhi (...)

13In the conquest period, pre-Islamic Arabian institutions evolved to meet new circumstances. Pre-existing protection institutions like amān provided mechanisms for offering terms to the conquered for the cessation of hostilities; walāʾ allowed conquered converts a way to integrate into the army and other key institutions,22 and ḏimma protection of non-converted conquered peoples was used to define the ongoing relationship between rulers and ruled in the transition from militarized conquest to a normalized rule which was often coercive, but also relied on the ongoing assent from the members of the conquered populations who collected taxes and provided concrete support to the conquerors.23 However, the classical Islamic formulation of such institutions should not be assumed to directly reflect their earliest Islamic expressions, many of which must have originated in the ad hoc application of pre-existing Arabian (or other) frameworks to new contexts. De la Vaissière’s article shows how a form of unofficial ḏimma protection was granted in Sogdiana upon the conquest of the region. Combining narrative sources (especially the letters contained in them) with archaeological evidence, he shows how Sogdiana effectively received the protection granted to ḏimmīs, albeit after a political annihilation following the conquest. De la Vaissière compares the governor al‑Ḥajjāj’s orders about the ḏimmī region of Sind, which protected the inhabitants’ choice to practice their religion, with the Sogdian archaeological evidence. Excavations of private houses in Panjikent show how conditional the protected status was: visual expressions of religion, through mural paintings, moved in the post‑conquest decades from very public spaces, such as entrance halls, to more private, inner rooms.

  • 24 For literature on the Islamic tradition’s take on the cessation of hostilities, and the provision o (...)

14In the following centuries, the Islamic tradition canonized the experience of the early conquests, organizing and editing the information transmitted about early practice to provide models for normative Muslim behaviour.24 This preserved, but also transformed, the information transmitted from the early years, as is clear from Webb’s contribution to this special issue.

  • 25 Thus, for example, Shi’i Imams and Sufi shaykhs came to be addressed by the term mawlāya.

15The kinds of relationships formed by conquest varied greatly according to region, both within the ruling classes, but especially between rulers and the conquered. The Pahlavi and Bactrian documents studied by Dieter Weber and Reza Huseini in their contributions to this special issue give a sense of the granular detail of protection systems at very specific times and places. Through his careful readings and re‑editions of a number of Pahlavi documents from seventh-century Iran, Weber shows how the local Arab conquerors who had arrived in the region of Qom in central Iran were integrated into the local economy of gift‑giving, no doubt reflecting their status as militarily powerful, but also the ways in which the exchange of goods helped cement mutual ties for protection from other potential threats. Thus, though relations within the ruling class of Arab‑Muslim conquerors continued to be defined by family, clan and tribe relationships, also new relationships arose relating to the management of the land and taxation, and management of the armies and the new relations that arose in the garrison towns. Relations within the ruling classes eventually changed massively due to conversion. The gradual extension of the meaning of walāʾ and mawālī is the most obvious as the meanings and possibilities of patronage and protection relations mutated to produce new meanings for these words.25

  • 26 Marsham, 2009.

16The old system of forging relations between Arabs through the making of oaths persisted, but changed in quality due to the fact of the state and other new institutions, as for example in the case of the bayʿa which became a key part of the performance of imperial legitimacy, explored, for example by Andrew Marsham.26 Oaths were, of course, key elements in middle eastern societies before and after the Islamic conquest. Mottahedeh’s Loyalty and Leadership is a landmark in the study of the way social coherency was established and maintained. Mottahedeh suggests that the 10th century CE Buyid context he investigates was characterized by a high degree of formality in links between individuals, and a low degree of formality in links between groups (guilds, societies, associations). Many of his insights can be extended to earlier periods. For Mottahedeh, among the most significant ties that bound Buyid society were those of “acquired loyalty”: oaths, vows, pledges of allegiance, and so on. He notes that,

  • 27 Mottahedeh, 2001, p. 96.

All these ties are alike in that they are individual ties… Behind all these acquired ties we see the individualistic presupposition that a man can accept or offer an obligation only on his own behalf, and not on behalf of a group. Spokesmen existed, of course, but… they were obliged to get the personal agreement of those for whom they spoke.27

17This bespeaks a transition from what is typically understood of pre‑Islamic society and the context of the early Muslims as attested in the umma document, in which relations are, indeed, constructed around blocs: tribes, clans and families, among whom, for example, punishment for a crime committed by one individual in a group can be exacted from any other member of that group. In the ʿAbbasid period and after, there was perhaps greater space in society for social mobility, with entrepreneurial individuals being able to move beyond the circumstances of birth. This is not to say that society was egalitarian: rather the reverse, and patronage was, as ever, a key factor to rise, especially through the process of adopting and training a protégé (iṣṭināʿ ), which established an asymmetric, reciprocal relationship between the neophyte and the patron through whose intervention he might rise to greater heights. In rural contexts, patronage and protection relationships often revolved around agricultural benefits for groups of farmers, rather than personal career paths.

18As the dynamics between different groups and political power changed, the words that referred to institutions of protection came to stand for different phenomena, especially in the case of those institutions which were not tied down to specific frameworks by the law of the jurists. For example, the changing fates of the word ḥimāya (ḥimāyat in Persian) are addressed here by Kennedy, Paul and Azad and Firoozbakhsh. Kennedy and Paul both engage with and move beyond the pioneering work of Claude Cahen on ḥimāya. Kennedy’s article focuses on acts of commendation of land (ilǧāʾ ) to a powerful protector in the Umayyad and early ʿAbbasid period, especially cases in which the caliph or his son acted as protectors of landowners. His different cases show how protective relationships could be initiated by protectors and protectees alike. A suitable protector was someone who commanded respect through power and prestige, but could also use incentives such as tax‑breaks for the protectees to enter into a protection relationship. In his contribution, Paul shows how the term ḥimāya was used to denote many different mechanisms of protection, for example, oscillating between protection money paid to nomadic groups with coercive power and the appointment of those same groups to act as agents of the state. Paul’s analysis of this term also provides important context for Azad and Firoozbakhsh’s presentation of a very specific instance of the usage of ḥimāyat in a Persian document of the twelfth century CE. In this case the ḥimāyat is an offer of protection to a fugitive official. In the decree presented, the authorities strongly signal that such protection is not to be extended, showing us how an analysis of protection can illuminate the contestation of spheres of authority between figures in power in different regions and capacities.

19As the contributions to this special issue indicate, protection relationships in the early Islamic empire may be couched in Islamic terms, but they may well be based on the administrative or customary formalization of immediate social requirements which have little to do with Islamic normativity. Institutions of protection that are expressed in Pahlavi (Weber), New Persian (Azad and Firoozbakhsh), Bactrian (Huseini) or Coptic (Scheerlinck) should not therefore be excluded from our understanding of the functioning of the Islamic empire. Scheerlinck’s Coptic protection letters on ostraca and papyrus hint at very local yet relatively long-lived administrative documentary practices which suggest local practices related to fiscal structures in the empire, while, however, falling beyond the direct remit of Arab‑Muslim administrators. Like Scheerlinck’s protection letters, the twelfth‑century decree fragment edited by Azad and Firoozbakhsh suggests a local, village context, and was likewise an attempt to make someone return to his proper place. While the Coptic protection letters used promises of protection as incentives for the fugitives to return, the Persian decree explicitly states that the addressee cannot be given protection, and only threatens with punishment if he does not return. The decree thus perhaps hints at a context in which one authority seeks to eliminate a potential multiplicity of authorities that might offer protection to people who are misbehaving. Even at a local level, then, we must assume that there could be multiple regimes offering protection. Another example of multiple competing regimes appears in the documents studied in Huseini’s contribution, which span the majority of the period of Bactrian documentation, from the fifth to the eighth century CE, and point to a local legal practice still in use after the Arab‑Muslim arrival to the region. One case study in his article shows how individuals experiencing legal family issues chose to bypass the Arab‑Muslim administrators present in his region, and rather brought his case before the more distant Turkic ruler. The lives of people living in a border area were thus characterised by the need to navigate competing local and imperial regimes of threat and protection.

20When seeking to understand the broader network of social relations that held the empire together, we should imagine a set of overlapping fields within which these relations were cast. Each came with its own institutions, roles, networks and vocabularies. If we limit our analysis of the social relations in the empire to only one of these fields, then we risk missing the overall implications of a relationship entered into. Attempting to map out all the intersecting fields that a given act of protection invokes, however, is none too easy in an academic context where humanists tend to work alone, and no scholar can become an expert in all fields of activity or in all the languages of the empire. By bringing together these diverse cases of analogous relationships, we hope to illuminate both the diversity and the connectivity of the societies which made up the early Islamic empire.

Haut de page


Abou El Fadl, Khaled, “Legal Debates on Muslim Minorities: Between Rejection and Accommodation”, The Journal of Religious Ethics 22, 1, 1994, pp. 127–162.

Ahmad, Muhammad M., “The Notions of Dār alḤarb and Dār alIslām in Islamic Jurisprudence with Special Reference to the Ḥanafī School”, Islamic Studies 47, 1, 2008, pp. 5–37.

Arjomand, Saïd A., “The Constitution of Medina: A Sociolegal Interpretation of Muhammad’s Acts of Foundation of the Umma”, International Journal of Middle East Studies 41, 4, 2009, pp. 555–575.

Bernards, Monique & Nawas, John (eds.), Patronate and Patronage in Early and Classical Islam, Brill, Leiden, Boston, 2005.

Bonner, Michael, Jihad in Islamic History: Doctrines and Practice, Princeton University Press, Princeton, 2006.

Calder, Norman, “Ḥinth, Birr, Tabarrur, Taḥannuth: An Inquiry into the Arabic Vocabulary of Vows”, Bulletin of the School of Oriental and African Studies 51, 2, 1988, pp. 214–239.

Clark, Elaine, “Social Welfare and Mutual Aid in the Medieval Countryside”, Journal of British Studies 33, 4, 1994, pp. 381–406.

Crone, Patricia, Slaves on Horses: The Evolution of the Islamic Polity, Cambridge University Press, Cambridge, 1980.

Crone, Patricia, Roman, Provincial and Islamic Law: The Origins of the Islamic Patronate, Cambridge University Press, Cambridge, 2002.

Evans-Pritchard, Edward E., The Nuer, A Description of the Modes Livelihood and Political Institutions of a Nilotic People, Oxford University Press, Oxford, 1940.

Garnsey, Peter, “Roman Patronage”, in S. McGill, C. Sogno and E. Watts (eds.), From the Tetrarchs to the Theodosians: Later Roman History and Culture, 284–450 CE, Cambridge University Press, Cambridge, 2010.

Grafe, Regina & Irigoin, Alejandra, “A Stakeholder Empire: The Political Economy of Spanish Imperial Rule in America”, Economic History Review 65, 2, 2012, pp. 609–651.

Hagemann, Hannah & Heidemann, Stefan (eds.), Transregional and Regional Elites – Connecting the Early Islamic Empire. The Early Islamic Empire at Work, Vol. 1, De Gruyter, Berlin, 2020.

Hayes, Edmund, “Early Islamic Cosmopolitanism? Constructing the ʾUmma of India in Pre‑Mongol Muslim Scholarship”, Comparative Islamic Studies 13, 1–2, 2017, pp. 75–120.

Hoyland, Robert, Arabia and the Arabs: From the Bronze Age to the Coming of Islam, Routledge, London, New York, 2001.

Lambert, Thomas B. & Rollason, David (eds.), Peace and Protection in the Middle Ages, Centre for Medieval and Renaissance Studies, Durham University, Durham; Pontifical Institute of Mediaeval Studies, Toronto, 2009.

Lecker, Michael, The “Constitution of Medina”: Muhammad’s First Legal Document, Princeton University Press, Princeton, 2004.

Marsham, Andrew, Rituals of Islamic Monarchy: Accession and Succession in the First Muslim Empire, Edinburgh University Press, Edinburgh, 2009.

Morony, Michael, Iraq after the Muslim conquest, Princeton University Press, Princeton, 1984.

Mottahedeh, Roy, Loyalty and Leadership in an Early Islamic Society, I.B. Tauris Publishers, New York, London, 2001 [1980].

Nef, Annliese & Tillier, Mathieu, “Les voies de l’innovation dans un empire islamique polycentrique. Introduction”, AnIsl 45, 2011, pp. 1–20.

Papaconstantinou, Arietta, “Between Umma and Dhimma: The Christians of the Middle East under the Umayyads”, AnIsl 42, 2008, pp. 127–156.

Paul, Jürgen, Herrscher, Gemeinwesen, Vermittler: Ostiran und Transoxanien in vormongolischer Zeit, Franz Steiner Verlag, Stuttgart, 1996.

Pourshariati, Parvaneh, “Iranian Tradition in Tus and the Arab presence in Khurasan”, PhD dissertation, Columbia University, 1995.

Pourshariati, Parvaneh, “Khurasan and the Crisis of Legitimacy”, in N. Yavari, L. G. Potter and J.‑M. Oppenheim (eds.), Views from the Edge: Essays in Honor of Richard W. Bulliet, Columbia University Press, New York, 2004, pp. 208–229.

Robinson, Majied, “Prosopographical Approaches to the Nasab Tradition: A Study of Marriage and Concubinage in the Tribe of Muḥammad, 500–750 CE”, PhD Thesis, The University of Edinburgh, 2013.

Safran, Janina M., Defining Boundaries in Al-Andalus: Muslims, Christians, and Jews in Islamic Iberia, Cornell University Press, Ithaca, 2013.

Sahlins, Marshall D., “The Segmentary Lineage: An Organization of Predatory Expansion”, American Anthropologist, New Series 63, 2, Part 1, 1961, pp. 322–345.

Salaymeh, Lena, “Taxing Citizens: Socio-Legal Constructions of Late Antique Muslim Identity”, Islamic Law and Society 23, 4, 2016, pp. 333–367.

Sijpesteijn, Petra M., Shaping a Muslim State: The World of a MidEighthCentury Egyptian Official, Oxford University Press, Oxford, 2013.

Tolan, J., “Breaches in the Conflict”, in J. Tolan, H. Laurens and G. Veinstein (eds.), Europe and the Islamic World: A History, Princeton University Press, Princeton, 2013.

Webb, Peter (ed.), AlMaqrīzī’s “alḪabar ʿan albašar”, Vol. V, Sections 1–2: “The Arab Thieves”, Brill, Leiden, 2019.

Wickham, Chris, Framing the Early Middle Ages: Europe and the Mediterranean, 400800, Oxford University Press, Oxford, 2005.

Haut de page


1 This special issue emerges from research done in the context of the project “Embedding Conquest: Naturalising Muslim Rule in the Early Islamic Empire (600–1000)” directed by Petra Sijpesteijn and funded by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 683194.

2 For example, Grafe, Irigoin, 2012.

3 For recent contributions to the conception of empire in the Islamic world, see, for example Nef, Tillier, 2011; Hagemann, Heidemann (eds.), 2020; Sijpesteijn, 2013; Wickham, 2005.

4 The nearest is perhaps Lambert, Rollason, 2009.

5 An excellent volume like Bernards, Nawas, 2005, provides an understanding of diverse instances of a particular legal and societal institution in the Muslim tradition, walāʾ, which is a standard topic both in Islamic law (fiqh) and in the historiography of the relations between conquered and conquerors. But the majority of early Islamic society was not Arab or Muslim, for all the importance of the hegemonic Muslim categories. Nor was all that was normative or structural to be found in the legal frameworks of fiqh. P. Crone (2002) sought the origins of Arab‑Islamic walāʾ in Roman provincial law.

6 A good example is found in Elaine Clark’s indication of the importance of land use as the crucial factor in securing protection in medieval English countryside: Clark, 1994.

7 Garnsey, 2010.

8 Lena Salaymeh’s conception of zakāt‑based “citizenship” is not applicable here, as it excludes the majority of the population of the early Islamic empire under study here, Salaymeh, 2016.

9 Wickham, 2005, pp. 56–150.

10 Wickham, 2005, p. 143; Crone, 1980, p. 51.

11 Wickham, 2005, p. 143.

12 Morony, 1984, p. 97.

13 Morony, 1984, p. 97.

14 Paul, 1996; Pourshariati, 2004; Pourshariati, 1995.

15 Morony, 1984, p. 98.

16 Evans‑Pritchard, 1940; Sahlins, 1961; Hoyland, 2001, p. 114.

17 See also Peter Webb’s introduction to Webb, 2019.

18 See Calder, 1988.

19 Lecker, 2004; Arjomand, 2009.

20 Hoyland, 2001, pp. 116–117.

21 See, for example, Hayes, 2017.

22 See also Robinson, 2013.

23 The understanding of Muslim relations with non-Muslim subjects as a relationship of protection (dhimma) as described in Islamic legal terms is a case in point. See, however, the cautions advised by Papaconstantinou, 2008.

24 For literature on the Islamic tradition’s take on the cessation of hostilities, and the provision of protection to the conquered, see Ahmad, 2008. A general survey of legal opinion on amān by Abou El Fadl, 1994; Bonner, 2006. A section on Aman in al‑Andulus in Safran, 2013; Tolan, 2013. Thanks to Robert Gleave for providing insights to this in his paper on amān at our workshop on protection in Leiden, and in subsequent communications.

25 Thus, for example, Shi’i Imams and Sufi shaykhs came to be addressed by the term mawlāya.

26 Marsham, 2009.

27 Mottahedeh, 2001, p. 96.

Haut de page

Pour citer cet article

Référence papier

Edmund Hayes et Eline Scheerlinck, « Introduction »Annales islamologiques, 54 | 2020, 3-14.

Référence électronique

Edmund Hayes et Eline Scheerlinck, « Introduction »Annales islamologiques [En ligne], 54 | 2020, mis en ligne le 27 octobre 2021, consulté le 27 mai 2024. URL : ; DOI :

Haut de page


Edmund Hayes

Radboud University

Eline Scheerlinck

Leiden Institute for Area Studies

Articles du même auteur

Haut de page

Droits d’auteur


Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search