The domestic origins of imprisonment: an inquiry into an early Islamic institution.

Author:Anthony, Sean

    In the Futuh al-buldan of Ahmad b. Yahya al-Baladhuri (d. 279/892), we read a rather short notice on the origins of a prison, well known in his day, located in Medina. It reads:

    Al-'Abbas b. Hisham al-Kalbi related to me, "An individual of the tribe of Kinda wrote to my father (1) asking him about the prison of Ibn Siba' in Medina and after whom it is named (ila man nusiba). ..." So [my father] wrote to him: "As for the prison of Ibn Siba', it used to be the house (kana dar) of 'Abd Allah b. Siba' b. 'Abd al-'Uzza b. Nadla b. 'Amr b. Ghubshan al-Khuza'I, and Siba' used to be called Abu Niyar and his mother was a midwife in Mecca. ..." (2) The account goes on to relate the death of Siba' at the hands of a believing uncle of the Prophet Muhammad, Hamza b. 'Abd al-Muttalib, who taunted Siba' as "a son of a female circumciser" before taking his life, but it says no more concerning the origins of the prison. A surface reading would see in this passage a preoccupation with the father of the man after whom the prison had been named rather than the circumstances leading to the conversion of Ibn Siba's house (Ar. dar: pl. dur) into a prison and the conditions of the subsequent use of the prison itself. Indeed, the historian by training is predisposed to treat such folk traditions about eponymous personages who leave their names on the sundry locations of a given region with reticence. However, the above passage, I would like to contend, offers more of value than one might initially suspect, for it contains a subtle instance of an oft-repeating pattern that remains salient wherever one stumbles across mentions of prisons in the period contemporaneous with and immediately following the first Islamic conquests.

    The pattern this passage conveys is a simple one: the origin of this Medinese prison is to be found in the conversion of a large house, dar, into a structure employed for incarceration--presumably by a state authority. As this study hopes to show, this pattern appears with surprising regularity, particularly in the early conquest period, although it continues to manifest itself, albeit much more sporadically, even in the Umayyad and Abbasid eras as well. Thus, for example, Mecca's first prison also allegedly resulted from such a conversion: serving as governor of Mecca during the caliphate of 'Umar b. al-Khattab (r. 13-23/634-44), Nafi' b. 'Abd al-Harith al-Khuza'i allegedly purchased a house (dar) from the Qurashite Safwan b. Umayya for use as the city's prison. (3) At least as early as the third/ninth century, early medieval historians such as 'Umar b. Shabba (d. 262/878) attested to the longevity of this Meccan prison's use, claiming that this house-turned-prison later became the same place that the so-called counter-caliph 'Abd Allah b. al-Zubayr (d. 73/692) employed as a prison, during which time it would come to be known infamously as Sijn 'Arim. (4) Besides Ibn Sibac's prison, Medina also housed yet another prison that was once a house. (5) In the early 'Abbasid period one frequently encounters a certain Dar Ibn Hisham referred to as the city prison. (6) This dar was once the house of the former governor of Medina, Ibrahim b. Hisham al-Makhzumi, under Hisham b. 'Abd al-Malik (r. 105-25/724-42). (7)

    These examples, as well as others to follow below, raise a number of questions regarding the social origins and the implications of the laconic, historical mentions of the conversion into prisons and/or the usage for incarceration of various buildings referred to as "dar so-and-so" in the early conquest period. Why do sources so often mention prisons either being contained in or originating from the conversion of a dar, or other like structures? Much of the rationale behind such conversions certainly drew from pragmatic concerns: the utilization of pre-existing buildings for newer, more public uses, of which imprisonment was surely merely one among many, represents an efficient means of swiftly addressing the rapidly expanding and pressing needs of the early conquest polity. (8) This practice also reveals more, however, for it points to important socio-historical facets of early carceral institutions and not merely their scale, scope, management, connection to authorities, etc. In effect, such conversions, and the carceral practices that precede and anticipate them, potentially offer profound insights into the socio-historical origins of the first Islamic prisons.

    While what is meant by dar is not always so easy to determine, its most salient connotations are that of a residence, even though the word dar appears in a number of protean usages, making the exact meaning of the word imprecise, especially as time progresses. (9) This also applies to the myriad of words that can be deemed roughly synonymous with dar such as: palace (qasr; pl. qusur), fortification (hisn; pl. husun), castle (utum: pl. atam), and stronghold (sisiya; pl sayasi). Many of the latter terms appear most often associated with the hilltop fortifications of the Jews of pre-Islamic Yathrib and its neighboring towns, which perhaps drew their inspiration from Yemeni architecture. (10) These terms, furthermore, also appear to be interchangeable, as one can glean from their Qur'anic usage (cf. Q 33:26 and 51:2). Lawrence Conrad has cogently argued in his study of early qusur that many of these structures may be understood as residential compounds of eminent persons, i.e., larger or more extravagant variations of the same sort of dar complex. (11) Hence, a dar may .simply be designated by the English "house," but in many cases one must also gloss the term with words such as "court" and "complex" or even envision a dar as an entire compound of buildings.

    Our conception of what a dar may or may not have been need not assume an overly rudimentary understanding of the structures being converted--perhaps only in part--into prisons. One may, therefore, conceive of a dar in the early Islamic period simply as either a nondescript, multi-purpose structure or as a domicile owned by a family household, which the household maintains for manifold domestic purposes. Originally, the examples of dars converted into prisons adduced above were all primarily thought of as dwellings in the broadest sense of the word. Later, such buildings, or parts thereof, would be dedicated for the purposes of incarceration. In what follows, we delve deeper into the cultural and political roots of this transformation.

    Despite the pluriformity of dars in the earlier period, the most salient feature they share is their rootedness in the domestic sphere prior to their conversion into prisons for use by the state. Properly speaking, such structures leave the domestic sphere once the household's claims to the structure are relinquished to the state and/or its representatives--thus transforming the building into a commodity for communal, social, and/or political usage. (12) Such conversions of domiciles into prisons thus reflect not merely pragmatic considerations but also the "domestic" origins of the carceral practices associated with imprisonment in the pre-and early Islamic period. The gist of what is meant by the "domestic" origins relies on the observation of a profound sea change transpiring in the course of the emergence of the Islamic state. This sea change transpires in three stages. At one end of the spectrum, one encounters informal carceral practices of confinement and detention as undertaken by the immediate family, kinsmen, tribesmen, etc. of the incarcerated individual--i.e., what amounts to "domestic imprisonment" or, more properly speaking, "domestic incarceration." By the designation "domestic incarceration" I do not mean the common practice of "house arrest" per se--i.e., confinement in and prohibition to leave one's own home by the state--but, rather, confinement in a house or similar domestic structure by one's own family members and/or kinsmen. At the other end of the spectrum, one encounters the transformation of such informal practice into the carceral practices formally appropriated and institutionalized by the burgeoning state, all the while maintaining vestiges of earlier practices. This process can be most clearly perceived in the intermediate stages between the informal carceral practices of the pre-conquest era and the formalization of the prison by the conquest-era state authorities--predominantly as represented by the caliph and his administrators.

    Inasmuch as carceral practices, as well as their accompanying customs and institutions, remain one among many neglected facets of the arena of early Islamic social-history, (13) this article aims to map out how and why carceral practices exited the private, domestic sphere and became the public prerogative of the state given the available anecdotal data present in the sources. To do so, our study begins with an investigation into the informal carceral practices of the Hijaz in the pre-conquest era and then moves on to demonstrate the continuity between these early practices as more formalized carceral institutions (i.e., those rooted in state, rather than private or domestic, authority) emerge throughout the early conquest era. The process, it will be argued, explains why the cases of the Meccan and Medinan prisons recorded above were not atypical but represented a widespread phenomenon that sheds considerable light on the earliest prisons of Islamicate society.


    An examination of the scattered mentions of the punitive and informal carceral practices of the Hijaz preceding the Arab conquests provides important insights into the precursors of the process of house-to-prison conversion seen above. Virtually no evidence survives attesting to the existence of formal prisons in this region before the Islamic conquests, whether among the settled populations of Mecca and Yathrib or in the hinterlands of either. What scant...

To continue reading