Islamic Tombstones for Slaves from Abbasid-Era Egypt

ABSTRACT

This article studies tombstones from eighth- to tenth-century CE Egypt that are designed to mark the grave of a Muslim slave. These funerary inscriptions are unusual in that they do not marginalize the enslaved as much as do other early Islamic sources. Furthermore, they reveal otherwise undocumented attitudes towards persons who died as slaves. Offering a thick description of an unpublished tombstone for a ninth-century concubine-mother (umm walad), the present article analyses tombstones for slaves from two perspectives. It first studies the representation of the enslaved and the specific terminology that tombstones used to designate the deceased as enslaved. It then turns to the commemorative context of tombstones, arguing that tombstones of slaves served similar purposes and used similar illocutionary strategies to those used by contemporary tombstones for free Muslims. Despite these similarities between tombstones of free and enslaved persons, we see that deceased slaves were commemorated as members of the Muslim community as well as the legal property of their owners.

KEYWORDS:

• Abbasid-era Egypt
• tombstones
• commemoration practices
• slavery
• identification of slaves

Introduction

At some point after her death in Ramaḍān 245 AH/December 859 CE, unknown persons marked the grave of a woman named Qaṣṣāf or Qiṣāf in one of the graveyards of Fusṭāṭ, Egypt’s first Muslim capital, with an elegantly decorated tombstone. Like the roughly 4000 other tombstones dating from the first Islamic centuries that have been published (and the many others that remain unpublished), this tombstone offers intimate glimpses into the life and beliefs of an otherwise unknown Muslim individual. Even though this tombstone’s funerary inscription is nothing out of the ordinary and its epigraphy is similar to that on many other tombstones, it is nevertheless an invaluable source for the social history of burial and commemoration practices and the religious attitudes associated with them in the first centuries of Islamic history. Today, the woman’s tombstone is kept in the Museum of Turkish and Islamic Arts in Istanbul; see Figure 1.¹ It reads:

In the name of God, Merciful and Compassionate. Qaṣṣāf/Qiṣāf, the concubine-mother of Kaʿb b. Baḥīr, testifies to the following. She testifies that there is no god beside God alone, Who has no partner, and that Muḥammad is His servant and messenger – God’s prayers and peace be upon him. She testifies that death and the Raising (of the dead) are real, (that) Paradise and Hell are real, ‘and that the Hour is coming without doubt, and that God will raise up those who are in the graves.’² She died on Sunday with thirteen days of Ramaḍān remaining in the year two hundred and forty-five.³

As is the case with most inscriptions on early Islamic tombstones, the text on Qaṣṣāf/Qiṣāf’s tombstone is highly conventional, largely consisting of quotations from the Qurʾān, professions of faith, and prayers. As a result, such funerary epigraphy has mostly been studied for its religious content and the information that it offers about the development and spread of Islam.⁴ Nevertheless, tombstones may usefully be mined for other information as well.⁵ Qaṣṣāf/Qiṣāf’s tombstone stands out because it is one of the relatively few known tombstones erected at the grave of a Muslim slave: the inscription identifies her as her owner’s concubine-mother (umm walad), a female slave who had borne her master a child. The present article considers tombstones such as Qaṣṣāf/Qiṣāf’s, that is, tombstones explicitly dedicated to enslaved Muslims. As we will see, they offer unique windows onto slavery in Abbasid society. An edition of the inscription on Qaṣṣāf/Qiṣāf’s tombstone can be found in the appendix to this article.

Figure 1. Istanbul, Museum of Turkish and Islamic Arts, inv. 2525.

The existence of Egyptian tombstones designed to mark the grave of a Muslim slave has long been known. At present, 109 tombstones dating from the first four centuries of Islam that explicitly identify the deceased as enslaved have been published. With the exception of four second/eighth-century examples, all of these Egyptian tombstones for slaves date from the third/ninth and fourth/tenth centuries.⁶ They come from various different locations in Egypt. Twenty-nine of them come from an old cemetery located to the east of Aswan.⁷ Fourteen tombstones for slaves originate from the mostly Abbasid-era cemetery of ʿAyn al-Ṣīra, an area located between the city of Fusṭāṭ and Mount Muqaṭṭam.⁸ The exact provenance within Egypt of all remaining tombstones for slaves is unknown, with the exception of two grave markers from Ṭāfā, a locality just south of the First Cataract (now submerged in Lake Nasser) which in the early Islamic period was not part of the Realm of Islam.⁹ Even in those cases where we know to what graveyards these tombstones originally belonged, the archaeological context of many of these tombstones has not been documented. Many of the tombstones from Aswan were brought to the Egyptian Museum in Cairo in the early 1890s and later to the Museum of Arab (now Islamic) Art after heavy rainfall had dislocated them.¹⁰ Probably between 1912 and 1924, many of those from ʿAyn al-Ṣīra were also moved to the Museum of Arab Art after they had been excavated without any proper archaeological supervision in order to build the Museum’s collection at low cost.¹¹ At present, the majority of published Egyptian tombstones for slaves belong to this collection and have been edited in the ten volumes of the Catalogue général du Musée arabe du Caire: Stèles funéraires prepared by Hassan Hawary, Hussein Rached and Gaston Wiet between 1932 and 1942.

Despite their availability and the recent academic interest in the archaeology of slavery under Islam, tombstones for slaves have largely escaped the attention of historians of slavery in the first centuries of Islamic history.¹² This comes as no surprise, as there are no striking differences between the tombstones for enslaved and free Muslims. Like those for other Muslims, tombstones for slaves are highly formulaic and often present only the barest information about the deceased. They usually give the name of the deceased and that of his or her owner, record the date of the deceased’s death, and identify him or her as a Muslim.¹³ Although some have argued that these tombstones’ very existence in the only partially literate society of early Islamic Egypt may present information about the deceased independent of their inscription and may offer clues about the social milieu to which the deceased had belonged and the nature of his or her commemoration by those who visited the grave, one must bear in mind that, like Qaṣṣāf/Qiṣāf’s, Islamic tombstones in general do not indicate who commissioned them, and that this severely limits our view of the deceased’s social network.¹⁴ In addition, despite the large number of published early Islamic tombstones that once marked the grave of a Muslim individual (whether free or enslaved), the information they present resists statistical analysis because some tombstones are anonymous, lack information about the deceased’s legal status or are only partially preserved, having lost those parts that once identified the deceased. Naturally, any analysis of such funerary epigraphy reveals information only about persons whose graves were marked with texts; built in accordance with the opinions of some Muslim jurists, graves without any (textual) decoration existed as well.¹⁵ For this reason, tombstone inscriptions from early Islamic Egypt can hardly be used for studying demography, including the relative size of Egypt’s slave population at any given time.¹⁶ It is noteworthy, however, that the great majority of published tombstones were made to mark the grave of a free person, suggesting that social status influenced a deceased’s commemoration.

Despite these textual and methodological limitations, early Islamic tombstones open new windows onto slavery and slave ownership in second/eighth- to fourth/tenth-century Egypt. As inscribed archaeological artefacts that originally belonged to specific monumental settings, they uniquely complement other sources for the history of slavery in the first centuries of Islam, such as Abbasid-era belles lettres and legal texts or private letters and bills of sale. While tombstones were carefully designed to influence the thoughts and behaviour of the visitors to the graves they marked, they are in fact multi-layered texts which document otherwise unknown facets of the social integration that slaves could achieve and reveal how people regarded slaves as legally belonging within the wider Muslim community. These different layers are most visible in the way that tombstones represent the enslaved and use illocutionary strategies in commemorative contexts. This article explores each of these different layers of meaning in tombstones for slaves.

Identifying Slaves on Tombstones

After the conventional invocation of God, many tombstones, including those for slaves, start by identifying the deceased. Whereas occasionally a tombstone fails to give a deceased slave’s name, most slaves mentioned on tombstones are identified by their names, their legal status and the names of their owners.¹⁷ Qaṣṣāf/Qiṣāf’s tombstone identifies the deceased woman in exactly this way and calls her ‘Qaṣṣāf/Qiṣāf, the concubine-mother of Kaʿb b. Baḥīr’. Such a tripartite identity frequently highlights the deceased’s slave status in more than one way. It does so first through the deceased person’s name. Qaṣṣāf/Qiṣāf’s tombstone is a case in point. Its epigrapher executed the deceased woman’s name in a very clear way, leaving no doubt as to its individual letters, but a name with these letters cannot be found in biographical dictionaries listing women’s names.¹⁸ In all likelihood, the woman was known not by her original name but by a slave name given to her by her owner or a slave trader. As is well known, slave names, inherently humiliating, marked an enslaved person’s deracination and his or her subjection to a master. As many known Abbasid-era slave names reflect a slave’s physical characteristics or personal qualities, Qaṣṣāf (‘Used to luxury’) and Qiṣāf (‘Corpulent woman’) are likely interpretations of the name the epigrapher has carved on her tombstone (in Arabic, these names are homographs).¹⁹ Both are incidentally attested as anthroponyms when preceded by an article and Qiṣāf is known to have been used as a name for a horse.²⁰ Many of the slaves whose tombstones have been preserved bore slave names, such as Jawhar (‘Gem’), Maymūn (‘Lucky’), Fawz (‘Success’) and Kitmān (‘Secrecy’).²¹

The masculine gender of some names given to enslaved women who appear in our tombstone corpus deserves special attention. Qaṣṣāf, a grammatically masculine intensive adjective, is one such name. Enslaved women of the Abbasid period are regularly found bearing masculine names.²² Naʿīm/Nuʿaym, Jaʿfar, Muḥibb and Qāsim are examples of masculine names given to female slaves whose tombstones have been preserved.²³ Other examples of masculine names given to enslaved women can be found on tombstones for freedwomen, such as ʿĀdil, ʿAlī and Ḥusayn.²⁴ These names purposefully assign a masculine identity to these women, seemingly supressing their biological gender. One possible reason why these women had been given masculine names is that they were ghulāmiyyāt, ‘boyish girls’. These were female transvestites who performed sexual services to male owners who felt sexually attracted to adolescent boys but wished not to have homosexual relationships.²⁵ Like Qaṣṣāf/Qiṣāf’s, some tombstones of enslaved women with masculine names indicate that the deceased had borne her master a child. Many other female slaves with masculine names seem not to have performed such roles, however.²⁶ More likely, the reason for this remarkable name choice must be sought in the owner’s aspirations, such as the desire to have a son or to own a (generally more expensive) male slave.²⁷

Even though a slave name suppressed a slave’s origins, this person’s original name was not necessarily unknown to her owners. This seems especially to have been the case when someone had been born into slavery.²⁸ A deed recording the emancipation of a girl or young woman (ṣabiyya) in 393/1003, for example, identifies her by her Arabic slave name Ṣafrāt (‘Yellow one’) and her original Coptic name ‘Dajāsha, daughter of Aryana, the female slave of Isiṭūrhīwh’.²⁹ It is intriguing that in such instances slaves and freedmen retained the names given to them by their owners or former owners. Why they did so, even on occasion after emancipation or after having acquired high social status, remains unknown.³⁰ Perhaps a tombstone’s mentioning of the deceased’s slave name reflects the wish to emphasize the owner’s claim to possession, or former possession, over the deceased, and thus to highlight their own wealth and status. Semi-anonymous tombstones for slaves, which do not mention the deceased’s name, support this interpretation. They foreground the deceased’s owner when they mark a grave with such words as ‘the grave of the boy of Umayya b. al-Rabīʿ b. Sulaymān’ or ‘the grave of the concubine-mother of Umayya b. Yaḥyā b. Umayya b. Maymūn b. Yaḥyā b. Muslim b. al-Ashajj al-Zuhrī’ – the latter inscription clearly boasts about the owner’s descent.³¹

In addition to identifying the deceased by a slave name, tombstones emphasize the slave status of the deceased by denying this person a lineage. In general, a lineage was a prerogative of free persons and modern historians consider its absence to be yet another mark of deracination, stressing absence of kin and the irrelevance of someone’s origins.³² Like a slave’s original name, however, enslavement did not by definition erase knowledge of the enslaved person’s origins. One tombstone, for example, calls a deceased concubine-mother ‘Your (i.e. God’s) servant (ama), daughter of Your servant (ʿabda)’, thus indirectly indicating that she was born of a Muslim mother. At the same time, however, this tombstone tells us nothing about the identity of the woman from whom she was born.³³ Occasionally, however, tombstones do acknowledge a slave’s descent. Admittedly, the legal status of the deceased in most of these tombstones is ambiguous.³⁴ Such females as ‘Zaynab, daughter of Ismāyīl (sic), the boy of Yaḥyā b. Umayya b. Maymūn’, ‘Āmina, daughter of Rawāḥ, the boy of Muḥammad b. Hārūn b. Ḥalwā’ and ‘Jawhar, daughter of Iqbāl, the boy of ʿAlī b. al-Ḥasan b. ʿAlī b. ʿAbd Allāh’ may have been children born into slavery.³⁵ The absence of a profession of faith on the tombstone of some of them may confirm an early death.³⁶ It is equally possible, however, that they were the free offspring of free women married to male slaves, because children born in marriage inherited their mother’s legal status (irrespective of that of the father) whereas a lineage, to which they would then be entitled on account of their freedom, usually showed paternal descent.³⁷ But some persons who are ascribed a lineage in these tombstones – as well as in other epigraphic texts – were clearly born into slavery.³⁸ The slave status of a woman named Usāma bt. Abī al-Khayr, for example, who is identified as ‘the concubine-mother of Muḥammad b. Sulaymān b. Fātiḥ’, is beyond doubt.³⁹

In contrast to these examples, most tombstones for slaves do not present the deceased’s lineage. Instead, they solely indicate that the deceased had been the property of someone else, and that he retains this status even after death. Whereas Arabic has various terms to indicate someone’s slave status, most tombstones, like contemporary letters and legal documents, heed the jurists’ well-known discouragement of using such unambiguous terminology as ʿabd and ama (both meaning ‘slave’) to indicate someone’s slave status – these terms, the jurists hold, should be used only in descriptions of man’s relationship to God.⁴⁰ Among currently published early Islamic tombstones, only two use such terminology.⁴¹ By contrast, three other terms seem to have been much more popular for describing a deceased slave’s legal status. Forty-three tombstones, including Qaṣṣāf/Qiṣāf’s, report that the deceased had been an umm walad (‘mother of a child’, i.e. concubine-mother). As noted above, this is a legal term used for an enslaved woman who had borne her master a child whom he acknowledged as his own (irrespective of the children she had with another man).⁴² In addition, thirty men are identified as someone’s fatā (lit. ‘boy’). The use of this word is noteworthy. The word fatā appears infrequently as a euphemism for an enslaved male in letters and deeds contemporary with our tombstones. Instead, letters and legal documents prefer another euphemism, ghulām (‘adolescent boy’, ‘servant’).⁴³ Tombstones may well use fatā in order to convey a sense of social integration rather than a sense of social hierarchy, as the word ghulām may imply. In the few instances in which letters use fatā, the word seems to convey a sense of intimacy.⁴⁴ Lastly, in addition to three tombstones that use the feminine form of fatā (fatāt, ‘girl’) for enslaved women, thirty tombstones describe a deceased woman as being someone’s jāriya (also ‘girl’).⁴⁵ Note that the terms fatā(t) and jāriya do not imply a young age at death. Although other texts contemporary with our tombstones use ṣabī or ṣabiyya to indicate a slave’s minor status, the terms fatā(t) and jāriya have no age implications and may have been used for minors and adults alike.⁴⁶

Irrespective of these words’ principle meaning or connotations, on early Islamic tombstones they unequivocally indicate legal status. This is most clearly visible with regard to the umm walad. As is well known, according to most legal schools in Abbasid-era Egypt, a slaveholder’s legal relationship with an enslaved woman in his possession changed when he made her pregnant. The change included his inability to sell her, her unconditional emancipation upon his death, and the free status of the children he had with her. The legal change enhanced the woman’s social status.⁴⁷ More importantly for the present discussion, by the time our oldest tombstones for such women were made, most jurists agreed on the almost complete irreversibility of this change.⁴⁸ Other legal slave types, by contrast, offered a permanent change of legal status only after certain conditions had been met (sometimes in the distant future). For example, a mudabbar slave only gained freedom upon his master’s death if the slave’s value did not exceed the discretional third of the deceased owner’s estate or when it was not necessary to sell the slave in order to repay the deceased owner’s debts.⁴⁹ Likewise, a mukātab slave only received freedom upon having fully fulfilled his emancipation contract with his master. A mukātab’s failure to meet the contractual obligations rendered the contract void and reverted his legal status to that of an ordinary slave.⁵⁰ When a mudabbar slave died before his master or a mukātab before fulfilling his emancipation contract, they died as ordinary slaves. The status of an umm walad, however, was very secure. Only when she was convicted of having committed adultery could she revert to being an ordinary slave.⁵¹ Early Islamic tombstones, then, show two slave types: the umm walad or concubine-mother and the ordinary slave, using fatā for enslaved males and fatāt and jāriya for enslaved females irrespective of their legal history.

Commemorating Deceased Slaves

Qaṣṣāf/Qiṣāf’s tombstone originally belonged to a Muslim cemetery in Fusṭāṭ, perhaps located at ʿAyn al-Ṣīra, where it not only identified her as a concubine-mother but also served to commemorate her. It is important to note that tombstones were part of a cemetery’s monumental architecture and were designed to attract a visitor’s attention. In a third/ninth- and fourth/tenth-century Muslim cemetery excavated in Alexandria, for example, tombstones have been found set into one of the short sides of low plastered frames that enclosed the site of a grave.⁵² Because the height of such frames usually did not exceed forty centimetres, these tombstones rose above the frame and could easily be seen.⁵³ Recent archaeological research in Aswan, too, shows that tombstones formed central elements in the town’s largely second/eighth- to fifth/eleventh-century funerary architecture. There, tombstones were often set in the centre of an approximately one meter-high panel that separated a tomb’s superstructure from a narrow platform equipped with a small prayer niche indicating the direction of Mecca.⁵⁴ Similarly, in large funerary complexes, such as those excavated in a necropolis to the south-east of Fusṭāṭ and dated to the Abbasid and Fatimid periods, funerary and other inscriptions occupied highly visible places.⁵⁵ In what must have been one such funerary complex, the historian Ibn Yūnus al-Ṣadafī (d. 347/959) copied someone’s lineage from an inscription that he saw engraved on a marble slab attached to the superstructure of (lit. ‘above’) the graves of this person’s descendants in Fusṭāṭ.⁵⁶ From the late second/eighth century on, such funerary architecture was often whitewashed.⁵⁷ This included the face of a tombstone. The inscription itself and the tombstone’s decoration, however, were painted over with red or black ink, making these features stand out against a mostly white background.⁵⁸

Being publicly visible, these tombstones often invited visitors to a grave to commemorate the deceased, soliciting prayers for those buried there. They did so by offering a prayer for the visitor in return for one or more prayers for the deceased. On late second/eighth- and third/ninth-century tombstones, requests for prayers can regularly be found at the end (less frequently at the beginning or embedded in the text) of the inscription.⁵⁹ They usually amount to a simple ‘May God have mercy upon those who asked for mercy upon him/her (i.e. the deceased).’ Some of them, however, are quite elaborate and are meant to inspire the visitor’s prayer. The early third/ninth-century tombstone for one Harim b. ʿIyāḍ al-Qurashī, for example, has: ‘May God have mercy upon whomever paused here, sought forgiveness for him (i.e. the deceased), asked for mercy upon him, petitioned God for a good reward, and sought intercession for him against hellfire!’⁶⁰ Other such prayers were related to supplicatory prayers for the deceased already embedded in the tombstone’s inscription.⁶¹ The stele dedicated to a man named Ṣāliḥ b. ʿAbd Allāh al-Murādī, dated 185/801, illustrates well how funerary inscriptions shaped prayers during the commemoration of the deceased in this way. Linking the conditional prayer to the preceding prayer for the deceased by repeating the grammatical root gh-f-r, it reads towards the end:

O God! Forgive (ighfir) those of Ṣāliḥ b. ʿAbd Allāh al-Murādī’s sins that came before and those that came after (his passing); prompt him with his argument (on Judgement Day) and make his tongue firm in (making) a sincere declaration; and have him join his prophet Muḥammad, may God’s blessings and peace be upon him! May God have mercy upon those who have read (this text) and prayed for forgiveness (maghfira) for the occupier of this grave.⁶²

Funerary inscriptions that include such prayers make clear that a tombstone did not only mark a site of commemoration, but also that reading the tombstone constituted part of the commemoration itself. In fact, tombstones sometimes include hints at being designed for recitation, as was commonly the case with other epigraphy in the early medieval Near East.⁶³ For example, the words of some quotations from the Qurʾān found on tombstones address the living. A number of tombstones for slaves, for instance, include a quotation of Qurʾān 67:1-3, saying that God ‘created death and life to test you (pl.) and reveal which of you does best’.⁶⁴ Early Islamic tombstones belonged to a written funerary-monumental tradition largely consisting of a more or less fixed repertoire of prayers, creeds and Qurʾānic quotations.⁶⁵ Importantly, they were embedded in oral/aural commemorative practices as well, addressing the shared expectations of those present.

Funerary inscriptions dedicated to slaves should not be seen as being functionally different from those dedicated to free Muslims. They may contain prayers and Qurʾānic quotations similar or identical to those found on tombstones for free persons;⁶⁶ and their very existence reflects the same wish to have the deceased commemorated and to determine, or at least to influence, the commemoration’s textual form. Take, for example, the tombstone for a female slave named Ḥawrāʾ (d. 225/839), who is identified as the daughter of a concubine-mother – that is, a child not recognized as the offspring of the concubine-mother’s owner. It ends with the now familiar words ‘May God have mercy upon those who have read it (i.e. the inscription, possibly out loud) and asked for mercy upon its (i.e. the grave’s) occupier.’⁶⁷ It should be noted that many tombstones for slaves, including the one that marked Qaṣṣāf/Qiṣāf’s grave, lack prayers or include only a short and rather unexceptional prayer after mentioning the deceased’s name, such as ‘May God’s mercy, forgiveness and favour be upon him/her!’, writing out what the visitor is expected to say while reading the inscription.⁶⁸ This is not unusual, and these tombstones do not differ from many tombstones for free Muslims.

Regardless of the presence of prayer requests, many tombstones highlight the deceased’s sound Muslim beliefs. Again, Qaṣṣāf/Qiṣāf’s tombstone is a case in point. Beside identifying the deceased woman and mentioning her death date, it presents two professions of faith. Despite the seeming banality of this information in an Islamic funerary context, these professions of faith should not be understood as simply reporting that she had lived a pious life. Like many other funerary texts, each of Qaṣṣāf/Qiṣāf’s professions of faith begins with the word tashhadu, ‘she testifies’. In contrast with the perfect used for the verb tuwuffiyat (‘she died’), which refers to an event in the past, the imperfect of the verb tashhadu expresses the common belief that the deceased ‘lives on’ in the grave, waiting for Resurrection Day and the Final Judgement.⁶⁹ One tombstone, dated 217/832-3, even states that the deceased concubine-mother to whom it is dedicated ‘asks (tasʾalu) God to grant her an entrance [into Paradise] similar to that of the God-fearing, make her join the pious believers and have her gather with the rightly-guided.’⁷⁰ These parts of a tombstone’s text must be seen as speech acts formulated by the person(s) who commissioned or engraved the tombstone and performed by the tombstone’s reader. As such, these texts too belong to the oral/aural aspects of commemoration. When a visitor lent his voice to read such texts aloud, they became assertive speech acts that publicly acknowledged the deceased’s piety and membership of the Muslim community.⁷¹

Concluding Remarks

Read in these ways, early Islamic tombstones for Egyptian slaves convey multiple messages about a slave’s belonging and served different purposes. On the one hand, these tombstones record a little of the lives of persons whom other sources tend to marginalize, often reducing them to nameless individuals serving their protagonists. Tombstones may preserve the names of slaves, sometimes even their descent, and they may indicate that they were commonly known by slave names; they may refer to the ability of slaves to marry and have legitimate children; and, in case of concubine-mothers, they may record that these women had been used for sexual intercourse and procreation and that this had changed their legal status and, implicitly, had given them a limited amount of social prestige. Above all, however, tombstones are evidence of the emotional bonds that these slaves were able to develop with those who erected their tombstones. Because slaves did not leave behind an estate that could cover their burial costs, contemporary jurists prescribed that owners pay for their slave’s shroud and grave.⁷² A tombstone, however, was a voluntary expense not required by Islamic law. The existence of tombstones for slaves shows that people were willing to incur extra costs for these persons’ welfare after death. The tombstones that they erected were embedded in the same funerary tradition as those of freemen, using prayers, citations from the Qurʾān and illocutionary strategies that are similar or identical to those that can be found on tombstones for free Muslims.

On the other hand, each of these tombstones draws attention to the deceased’s owner and the rights of ownership he holds over the slave. By giving the deceased’s often intentionally chosen slave name and by stating the deceased’s legal identity, tombstones not only serve to commemorate an enslaved person but also to commemorate an owner’s loss of a valuable possession, thereby testifying to the owner’s wealth and status. As we have seen, in extreme cases tombstones might reduce their identification of the deceased slave to a bare minimum and name only the owner. Unlike sources that state that in the grave ‘a master (mawlā) cannot be distinguished from a slave (ʿabd)’ and that emphasize the equality of all Muslims before God, tombstones explicitly mention the deceased’s slave status and the master’s rights of ownership, presenting the deceased as still belonging to another despite his or her death.⁷³ This information allowed a visitor to know in what legal way the deceased had belonged to the Muslims’ tribal society, like a lineage usually identified someone as a freeborn Muslim and someone’s walāʾ showed his status as a client. In legal terms, slaves belonged to the Muslims’ tribal society because they were part of the property of a member of this society, even after having passed away.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Notes on contributors

Jelle Bruning

Jelle Bruning is a lecturer in the Department of Middle Eastern Studies, Leiden University, P.O. Box 9515, 2300 RA Leiden, The Netherlands. Email: [email protected]

Notes

1 For another image, a description and a partial transcription of the tombstone, see M. Murat Bozcu, ‘Türk ve İslam Eserleri Müzesi taş eserler koleksiyonu’ (PhD dissertation, Mimar Sinan Fine Arts University, 2017), 213–4.

2 Qurʾān 22:7.

3 17 December 859 CE.

4 Jonathan M. Bloom, ‘The Mosque of the Qarafa in Cairo’, Muqarnas 4 (1987), 7–20; Christian Décobert, ‘Sur l’arabisation et l’islamisation de l’Égypte médiévale’, in Itinéraires d’Égypte: Mélanges offerts au père Maurice Martin s.j., ed. Christian Décobert (Cairo: Ifao, 1992), 273–300; Werner Diem, The Living and the Dead in Islam: Studies in Arabic Epitaphs, vol. 1, Epitaphs as Texts (Wiesbaden: Harrassowitz Verlag, 2004); Leor Halevi, ‘The Paradox of Islamization: Tombstone Inscriptions, Qurʾānic Recitations, and the Problem of Religious Change’, History of Religions 44, no. 2 (2004), 120–52; idem, Muhammad’s Grave: Death Rites and the Making of Islamic Society (New York: Columbia University Press, 2007), 14–42; Paul Joüon, ‘Le sentiment religieux dans les plus anciennes épitaphes des musulmans d’Égypte’, Recherches de science religieuse 25 (1935), 513–30; and Solange Ory, ‘Aspects religieux des textes épigraphiques du début de l’Islam,’ Revue du monde musulman et de la Méditerranée 58 (1990), 30–9 are examples of studies into the development and spread of Islam using large numbers of tombstones. Among all tombstones, one dating from 71 AH and the first to refer to ‘the people of Islam’ and presenting an early non-literary reference to the prophet Muḥammad has attracted particular attention from historians of early Islam. See, for example, Jere L. Bacharach and Sherif Anwar, ‘Early Versions of the Shahāda: A Tombstone from Aswan of 71 A.H., the Dome of the Rock, and Contemporary Coinage’, Der Islam 89, no. 2 (2012), 60–9; Jonathan E. Brockopp, ‘Interpreting Material Evidence: Religion at the “Origins of Islam”’, History of Religions 55, no. 2 (2015), 121–47; and Ilkka Lindstedt, ‘Who is in, Who is out? Early Muslim Identity through Epigraphy and Theory’, Jerusalem Studies in Arabic and Islam 46 (2019), 147–246.

5 See Diem, The Living and the Dead in Islam for an elaborate analysis of the contents of Islamic tombstones. See also Mathieu Tillier, ‘Urban Populations in Early Islam: Self-Identification and Collective Representation’, in Civic Identity and Civic Participation in Late Antiquity and the Early Middle Ages, ed. Cédric Brélaz and Els Rose (Turnhout: Brepols, 2021), 333–61, who mines tombstones for information about how people identified themselves.

6 Second/eighth-century tombstones: Hasan Hawary and Hussein Rached, Catalogue général du Musée arabe du Caire: Stèles funéraires, vol. 1 (Cairo: Imprimerie Nationale, 1932), 1, no. 12; Gaston Wiet, Catalogue général du Musée arabe du Caire: Stèles funéraires, vol. 9 (Cairo: Imprimerie de la Bibliothèque Égyptienne, 1941), 6–7, nos. 3207–8 and 10, no. 3212. Note that George C. Miles, ‘Early Islamic Tombstones from Egypt in the Museum of Fine Arts, Boston’, Ars Orientalis 2 (1957), 215–26 at 225, no. 14 has been dated to the second/eighth or third/ninth century.

7 For Aswan’s old cemetery, especially its still existing southern part, see Sophia Björnesjö and Philipp Speiser, ‘The South Necropolis of the Fatimid Cemetery of Aswan’, Annales islamologiques 48, no. 2 (2014), 117–34 and Philipp Speiser, ‘Umayyad, Tulunid, and Fatimid Tombs at Aswan’, in The First Cataract of the Nile: One Region, Diverse Perspectives, ed. Dietrich Raue, Stephan J. Seidlmayer and Philipp Speiser (Berlin: Walter de Gruyter, 2013), 211–21.

8 Ayman F. Sayyid, La capitale de l’Égypte jusqu’à l’époque fatimide, al-Qāhira et al-Fusṭāṭ: Essai de reconstitution topographique (Stuttgart: Franz Steiner Verlag, 1998), 643–4.

9 See now Stefanie Schmidt, ‘The Problem of the Origin of Tombstones from Aswan in the Museum of Islamic Art in Cairo’, Chronique d’Égypte 96 (2021), 353–70, who identifies formularies, material characteristics and registration numbers that suggest an Aswan provenance of some Abbasid-era tombstones. For the mentioned tombstones from Ṭāfā, see Gaston Wiet, Catalogue général du Musée arabe du Caire: Stèles funéraires, vol. 4 (Cairo: Musée National de l’Art Arabe, 1936), 103–4 and 129–30, nos. 1429 and 1488.

10 Speiser, ‘Umayyad, Tulunid, and Fatimid Tombs’, 219–21.

11 Tanya Treptow, ‘Evolving Excavations: The Origins of a Practice of Islamic Archaeology in Egypt’ (PhD dissertation, The University of Chicago, 2013), 84–5. See also Yūsuf Rāġib, ‘Les pierres de souvenir: Stèles du Caire de la conquête arabe à la chute des Fatimides’, Annales islamologiques 35 (2001), 321–83 at 321.

12 Recent publications studying the archaeology of early Islamic slavery are Adam Gaiser, ‘Slaves and Silver across the Strait of Gibraltar: Politics and Trade between Umayyad Iberia and Khārijite North Africa’, Medieval Encounters 19 (2013), 41–70; Anne Haour, ‘The Early Medieval Slave Trade of the Central Sahel: Archaeological and Historical Considerations’, in Slavery in Africa: Archaeology and Memory, ed. Paul J. Lane and Kevin C. MacDonald (Oxford: Oxford University Press/London: British Academy, 2011), 61–78; Marek Jankowiak, ‘Wer brachte im 10. Jahrhundert die Dirhems in die polnischen Gebiete und warum?’, in Fernhändler, Dynasten, Kleriker: Die piatische Herrschaft in kontinentalen Beziehungsgeflechten vom 10. bis zum frühen 13. Jahrhundert, ed. Dariusz Adamczyk and Norbert Kersken (Wiesbaden: Harrassowitz Verlag, 2015), 41–54; idem, ‘What Does the Slave Trade in the Saqaliba Tell Us about Early Islamic Slavery?’, International Journal of Middle East Studies 49, no. 1 (2017), 169–72; and idem, ‘Tracing the Saqaliba: Slave Trade and the Archaeology of the Slavic Lands in the Tenth Century’, in The Archaeology of Slavery in Early Medieval Northern Europe: The Invisible Commodity, ed. Felix Biermann and Marek Jankowiak (Cham, Switzerland: Springer, 2021), 161–81. See also Paul J. Lane, ‘Slavery in Africa c. 500–1500 CE: Archaeological and Historical Perspectives’, in The Cambridge World History of Slavery, vol. 2: AD 500-AD 1420, ed. Craig Perry et al. (Cambridge: Cambridge University Press, 2021), 531–52. Occasionally, historians of slavery have used information found on tombstones; see, for example, Jelle Bruning, ‘Slave Trade Dynamics in Abbasid Egypt: The Papyrological Evidence’, Journal of the Economic and Social History of the Orient 63, no. 5–6 (2020), 682–742; Matthew S. Gordon, ‘Slavery in the Islamic Middle East (c. 600–1000 CE)’, in The Cambridge World History of Slavery, vol. 2, AD 500-AD 1420, ed. Craig Perry et al. (Cambridge: Cambridge University Press, 2021), 337–61; and Yūsuf Rāġib, ‘Esclaves et affranchis trahis par leur nom dans les arts de l’Islam médiéval’, in Les non-dits du nom: onomastique et documents en terre d’Islam: mélanges offerts à Jacqueline Sublet, ed. Christian Müller and Muriel Roiland-Rouabah (Beirut: Presses de l’Ifpo, 2013), 247–301. As far as I know, however, only Shaun Marmon, ‘Intersections of Gender, Sex, and Slavery: Female Sexual Slavery’, in The Cambridge World History of Slavery, vol. 2, AD 500-AD 1420, ed. Craig Perry et al. (Cambridge: Cambridge University Press, 2021), 185–213 embeds observations based on a study of (a very limited number of) tombstones for slaves in a broader discussion of early Islamic slavery.

13 Diem, The Living and the Dead, 9–10; Marco Schöller, The Living and the Dead in Islam: Studies in Arabic Epitaphs, vol. 2, Epitaphs in Context (Wiesbaden: Harrassowitz Verlag, 2004), 278.

14 Timothy Insoll, The Archaeology of Islam (Oxford: Blackwell Publishers, 1999), 188. See also Halevi, Muhammad’s Grave, 16; Diem, The Living and the Dead, 13–14.

15 Halevi, Muhammad’s Grave, 32–41; Schöller, The Living and the Dead, 270–73 and 278–86.

16 Gaston Wiet, ‘Stèles coufiques d’Égypte et du Soudan’, Journal asiatique 240 (1952), 274–97 at 273 and Bloom, ‘The Mosque’, 9–14 offer statistical analyses of tombstones. Christopher S. Taylor, ‘Reevaluating the Shiʿi Role in the Development of Monumental Islamic Funerary Architecture: The Case of Egypt’, Muqarnas 9 (1992), 1–10 at 6–7 and Yūsuf Rāġib, ‘La Mosquée d’al-Qarāfa et Jonathan M. Bloom’, Arabica 41, no. 3 (1994), 419–21 strongly criticise Bloom’s study.

17 Pace Rāġib, ‘Pierres de souvenir’, 343. Yūsuf Rāġib, Actes de vente d’esclaves et d’animaux d'Égypte médiévale, 2 vols. (Cairo: Ifao, 2002–6), 2:29, para. 71 notes that, like some tombstones, bills of slave sale sometimes omit the name of a slave as well.

18 Ibn Mākūlā, al-Ikmāl fī rafʿ irtiyāb fī al-muʾtalif wa’l-mukhtalif fī al-asmāʾ wa’l-kunā wa’l-ansāb, ed. ʿAbd al-Raḥmān b. Yaḥyā al-Muʿallimī, 10 vols. (Cairo: Dār al-Kitāb al-Islāmī, 1993–4), 10:860–6; Ibn Ḥajar al-ʿAsqalānī, Tahdhīb al-Tahdhīb, ed. Ibrāhīm al-Zaybaq and ʿĀdil Murshid, 4 vols. (Beirut: Muʾassasat al-Risāla, 1995), 4:663–705; and vol. 12 of al-Sakhāwī, al-Ḍawʿ al-lāmiʿ li-ahl al-qarn al-tāsiʿ, 12 vols. (repr. Beirut: Dār al-Jīl, 1412/1992).

19 For slave names, see Hekmat Dirbas, ‘Naming of Slave-Girls in Arabic: A Survey of Medieval and Modern Sources’, Zeitschrift für arabische Linguistik 69 (2019), 26–38 at 31 and 33 and Craig Perry, ‘The Daily Life of Slaves and the Global Reach of Slavery in Medieval Egypt, 969–1250 CE’ (PhD dissertation, Emory University, 2014), 76–8. For the meaning of Qaṣṣāf and Qiṣāf, see Ibn Manẓūr, Lisān al-ʿArab, ed. Amīn M. ʿAbd al-Wahhāb and Muḥammad al-Ṣ. al-ʿUbaydī, 18 vols. (Beirut: Dār Iḥyāʾ al-Turāth al-ʿArabī, 1416/1995-6), 11:196; Reinhart Dozy, Supplément aux dictionnaires arabes, 2nd ed., 2 vols. (Leiden: E.J. Brill/Paris: Maisonneuve, 1927), 2:359; A. de Biberstein Kazimirski, Dictionnaire arabe-français contenant toutes les racines de la langue arabe: Nouvelle édition, 2 vols. (Paris: Éditions G.-P. Maisonneuve, 1860), 2:754.

20 Ibn Ḥajar al-ʿAsqalānī, Tabṣīr al-muntabih bi-taḥrīr al-Mushtabih, ed. ʿAlī M. al-Bajāwī and Muḥammad ʿA. al-Najjār, 4 vols. (Cairo: al-Dār al-Miṣriyya li’l-Taʾlīf wa’l-Tarjama, 1964–7), 3:1170; Ibn Nuqṭa, Takmilat al-Ikmāl, ed. ʿAbd al-Qayyūm ʿAbd Rabb al-Nabī, 5 vols. (Mecca: Jāmiʿat Umm al-Qurā, 1410/1989-90), 4:635; Ibn al-Kalbī, Nasab al-khayl fī al-jāhiliyya wa’l-islām wa-akhbāruhā, ed. Nūrī Ḥ. al-Qaysī and Ḥātim Ṣ. al-Ḍāmin (Baghdad: Maṭbaʿat Jāmiʿat al-ʿIlmī al-ʿIrāqī, 1406/1985), 47.

21 Respectively Wiet, Catalogue général, 9:6, no. 3207; Hawary and Rached, Catalogue général, 1:67–8, no. 95; Wiet, Catalogue général, 9:119, no. 3392; and Gaston Wiet, Catalogue général du Musée arabe du Caire: Stèles funéraires, vol. 2 (Cairo: Imprimerie Nationale, 1936), 31, no. 460.

22 See the many examples in Dirbas, ‘Naming of Slave Girls’.

23 Respectively, ʿAbd al-Raḥmān M. ʿAbd al-Tawab, Stèles islamiques de la nécropole d’Assouan, rev. and annotated by Solange Ory, 3 vols. (Cairo: Ifao, 1977–86), 2:108, no. 256; Miles, ‘Early Islamic Tombstones’, 225, no. 14; Wiet, Catalogue général, 4:29 and 45, nos. 1267 and 1305.

24 Respectively, Thérèse Bittar, Pierres et stucs épigraphiés: Catalogue du Musée du Louvre, Département des Arts de l’Islam (Paris: Réunion des musées nationaux, 2003), 68, no. 15; Wiet, Catalogue général, 4:53, no. 1324; and Gaston Wiet, Catalogue général du Musée arabe du Caire: Stèles funéraires, vol. 8 (Cairo: Imprimerie de la Bibliothèque Égyptienne, 1941), 128, no. 3083. See Wiet, Catalogue général, 9:231, no. 3600; Gaston Wiet, Catalogue général du Musée arabe du Caire: Stèles funéraires, vol. 10 (Cairo: Imprimerie de la Bibliothèque Égyptienne, 1942), 155, no. 3894 and ʿAbd al-Tawab, Stèles islamiques, 3:64, no. 363 for freedwomen named Qāsim, Sāṭir/Shāṭir and ʿAzīz.

25 Dirbas, ‘Naming of Slave Girls’, 31. For the ghulāmiyya in Abbasid society, see Everett K. Rowson, ‘Gender Irregularity as Entertainment: Institutionalized Transvestism at the Caliphal Court in Medieval Baghdad’, in Gender and Difference in the Middle Ages, ed. Sharon A. Farmer and Carol Braun Pasternack (Minneapolis: University of Minnesota Press, 2003), 45–72 and Johannes Thomann, ‘Illusions of Androgyny: Cross-Dressing Women (Ghulāmiyyāt) in Abbasid Society’, in Sex and Desire in Muslim Cultures: Beyond Norms and Transgression from the Abbasids to the Present Day, ed. Aymon Kreil, Lucia Sorbera and Serena Tolino (London: I.B. Tauris, 2021), 47–65.

26 See, for example, the enslaved woman named variously ʿAzīz, Fātin and Fātik, whose role at the court of al-Ḥakam I in al-Andalus Cristina de la Puente discusses in her article in this special issue.

27 For the reflection of a slave owner’s aspirations in the names of his slaves, see Perry, ‘The Daily Life of Slaves’, 75.

28 Compare Rāġib, Actes de vente, 2:29, para. 70. Clément Onimus identified in papyri mawālī bearing Greek, Coptic or Turkic names and considers them to be freedmen who retained their original names; see his ‘Les mawālī en Égypte dans la documentation papyrologique, Ier-Ve s.H.’, Annales islamologiques 39 (2005), 81–107 at 83–4. This needs further study, however. It is unclear if the individuals in question are freedmen or clients of the walāʾ al-muwālāt type. The latter’s client status did not result from emancipation and did not require conversion to Islam. See EI³, s.v. ‘Client’.

29 P.Cair.Arab. I 37. The girl’s original name, Dajāsha, must be understood as an Arabic transcription of the Coptic Tegōshe or the like, for which see Monika Hasitzka, ‘Namen in koptischen dokumentarischen Texten’ (available online at https://www.onb.ac.at/fileadmin/user_upload/PDF_Download/1_PAP_kopt_namen.pdf), 97, 99–101. Like P.Cair.Arab. I 37, the second/eighth- or third/ninth-century documents P.Cair.Arab. IV 223 and 249 and P.Giss.Arab. 5 record the original and new names of mawālī. The type of walāʾ, through conversion or emancipation, that tied these persons to their patrons is not clear, however.

30 Gordon, ‘Slavery in the Islamic Middle East’, 341.

31 Wiet, Catalogue général, 2:184 and 196, nos. 769 and 794. Other examples of such semi-anonymous tombstones are Wiet, Catalogue général, 4:88, no. 1399; Suʿād Māhir, Madīnat Aswān wa-āthāruhā fī al-ʿaṣr al-islāmī (Cairo: al-Jihāz al-Markazī li’l-Kutub al-Jāmiʿiyya wa’l-Madrasiyya wa’l-Wasāʾil al-Taʿlīmiyya, 1977), 130, no. 99; and Gaston Wiet, Catalogue général du Musée arabe du Caire: Stèles funéraires, vol. 6 (Cairo: Imprimerie Nationale, 1939), 124, no. 2239. The latter two date from the early fifth/eleventh century.

32 Yūsuf Rāġib, ‘Les esclaves publics aux premiers siècles de l’Islam’, in Figures de l’esclave au Moyen-Âge et dans le monde moderne, ed. Henri Bresc (Paris: L’Harmattan, 1996), 7–30 at 7; Rāġib, ‘Esclaves et affranchis’, 248, 253; Perry, ‘The Daily Life of Slaves’, 78–82.

33 Hassan Hawary and Hussein Rached, Catalogue général du Musée arabe du Caire: Stèles funéraires, vol. 3 (Cairo: Imprimerie Nationale, 1939), 47, no. 896.

34 Even more ambiguous than the examples that follow are tombstones for boys or men whose names include a patronymic, such as ‘Faraj b. Ayman, Ibrāhīm b. Saʿīd’s boy’ (Wiet, Catalogue général, 2:168, no. 735). Who is identified as someone’s ‘boy’, the father or the son? See also Wiet, Catalogue général, 2: 94, no. 582; idem, Catalogue général, 4:129–30, no. 1488; Gaston Wiet, Catalogue général du Musée arabe du Caire: Stèles funéraires, vol. 7 (Cairo: Imprimerie de la Bibliothèque Égyptienne, 1940), 79, no. 2577.

35 Wiet, Catalogue général, 2:3, no. 405; idem, Catalogue général du Musée arabe du Caire: Stèles funéraires, vol. 5 (Cairo: Imprimerie de la Bibliothèque Égyptienne, 1937), 70, no. 1751; idem, Catalogue général, 6:67, no. 2130. Compare Ibn Ḥawqal, Kitāb Ṣūrat al-arḍ, ed. J.H. Kramers, 2 vols. (Leiden: E.J. Brill. 1938), 1:154, who writes that a slave (ʿabd) named Miṣbaḥ b. Maymūn had been born into slavery (muwallad).

36 See Diem, The Living and the Dead, 209–10. The following tombstones for persons with enslaved fathers lack a profession of faith: Wiet, Catalogue général, 2:3, no. 405; ibid., 2:94, no. 582; Hawary and Rached, Catalogue général, 3:123, no. 1052; Wiet, Catalogue général, 5:70, no. 1751; ibid., 5:72, no. 1756; Wiet, Catalogue général, 6:3, no. 2003; ibid., 6:67, no. 2130; Wiet, Catalogue général, 7:79, no. 2577.

37 For children receiving their mother’s legal status, see Rainer Oßwald, Das islamische Sklavenrecht (Würzburg: Ergon Verlag, 2017), 163–4. For (the socially unpopular) marriage between a free woman and a slave, see Mohammed H. Benkheira, ‘Un libre peut-il épouser une esclave? Esquisse d’histoire d’un débat, des origins à al-Shāfiʿī (m. 204/820)’, Der Islam 84 (2008), 246–355; Cristina de la Puente, ‘Esclavitud y matrimonio en al-Mudawwana al-Kubrā de Saḥnūn’, Al-Qantara 16, no. 2 (1995), 309–33 at 314–5; and Kecia Ali, Marriage and Slavery in Early Islam (Cambridge, MA: Harvard University Press, 2010), 169–71.

38 For examples of other epigraphic texts in which enslaved persons have a lineage, see Rāġib, ‘Esclaves et affranchis’, 248–50.

39 Hawary and Rached, Catalogue général, 1:11, no. 12. See Wiet, Catalogue général, 6:3, no. 2003 and idem, Catalogue général, 9:103, no. 3365 for other examples. According to the third/ninth-century stele published in Wiet, Catalogue général, 10:155–6, no. 3896, a deceased woman named Salāma bt. Yaḥyā al-Tujībī was commonly known as Umm Walad al-Tujībiyya (compare Rāġib, ‘Pierres de souvernir’, 343), a nickname that reveals that she had been a concubine-mother. Hawary and Rached, Catalogue général, 3:152–3, no. 1115 might well record another concubine-mother’s genealogy. Her name should probably be read as ‘U[māma bt. N.N. umm] walad of Ismāʿīl b. al-Qāsim’ because the lacuna has room for three words.

40 Franz Rosenthal, The Muslim Concept of Freedom Prior to the Nineteenth Century (Leiden: E.J. Brill, 1960), 30; Rāġib, Actes de vente, 2:23–4, paras. 57–9; Rāġib, ‘Esclaves et affranchis’, 251–3. An unambiguous exception is the fourth/tenth-century letter P.Vind.Arab. II 5, in which the author writes about circumcising an amīr’s slave (ʿabd). See Werner Diem, ‘Philologisches zu den arabischen Papyri der Hamburger Staats- und Universitäts-Bibliothek’, Zeitschrift für arabische Linguistik 45 (2006), 7–54 at 38 for a correction of line 10 of the third/ninth-century letter P.Hamb.Arab. II 38, reading ʿafīf (‘regarded highly’) instead of ʿubayd (‘young slave’). See Jelle Bruning, ‘Voluntary Enslavement in an Abbasid-Era Papyrus Letter’, Journal of the Royal Asiatic Society, 3rd series 33 (2023), 643–59 at 654–5 for the infrequent appearance in letters and deeds of other unambiguous terminology used for identifying persons as enslaved.

41 Wiet, Catalogue général, 4:52, no. 1320 and Wiet, Catalogue général 5:119–20, no. 1850; both report that the deceased woman had been an ama, ‘female slave’. See also the fifth/eleventh-century tombstone published as Wiet, Catalogue général, 6:124, no. 2239, which calls the deceased woman a mamlūka, ‘owned person’.

42 Compare note 39 above for a tombstone for a woman nicknamed Umm Walad al-Tujībiyya. The term is also found as a (nick)name in a second/eighth- or third/ninth-century private letter published in Jean David-Weill, ‘Papyrus arabes du Louvre, I’, Journal of the Economic and Social History of the Orient 8, no. 3 (1965), 277–311 at 304–8, no. 9 (line 8).

43 Papyrus and early paper letters tell of trade in ghulāms and of their emancipation: P.Berl.Arab. II 49 (second/eighth century), with the discussion in Jelle Bruning, ‘Arabic Documents for Slavery in Early Islamic Egypt’, Slaves and Dependents in Ancient Egypt: Sources in Translation, ed. Jane Rowlandson, Roger S. Bagnall and Dorothy J. Thompson (Cambridge: Cambridge University Press, in press); Chrest.Khoury I 80 (third/ninth century); Donald S. Richards, ‘Fragments of a Slave Dealer’s Day-Book from Fusṭāṭ’, in Documents de l’Islam médiéval: nouvelles perspectives de recherche, ed. Yūsuf Rāġib (Cairo: Ifao, 1991), 89–96 (fourth/tenth or early-fifth/eleventh century). ‘Servants’ with typical slave names, such as Maymūn, Faraḥ/Faraj and Ḥulbūb, are described as running errands and are the carriers of messages in, for example, P.JoySorrow 3 and 13 (both from the second/eighth century) and P.Vind.Arab. II 5 (fourth/tenth or fifth/eleventh century). Analogous to fatā, the word ghulām sometimes expresses dependent status in extended names, such as in the third/ninth-century Chrest.Khoury II 28 = CPR XXI 10, which documents the lease contract of one ‘Ḥusayn, Ibrāhīm b. ʿAlī’s servant (ghulām)’; see also CPR XXXII 4 (fourth/tenth century). Slightly beyond the chronological scope of this paper, the fifth/eleventh-century P.Heid.Arab. III 34 expresses a ghulām’s unfreedom most explicitly. Its sender orders the addressee to detain a servant (ghulām) ‘so that he will not hide from us’.

44 The second/eighth-century sender of P.JoySorrow 26, for example, one Rābiḥ (‘Profitable’), identifies himself to the addressee as ‘Rābiḥ, Saʿīd’s boy’. Although much of his letter conveys an angry complaint about not having received answers to earlier letters, Rābiḥ emphasizes the close relationship (‘brotherhood’, line 11) between him and his addressee. Similarly, in two second/eighth-century letters written on the same sheet of papyrus, two brothers who exchange information about their well-being and convey greetings on behalf of their associates identify each other as ‘the son of Saʿd, Qays b. ʿAbd Allāh’s boy’ (Yūsuf Rāġib, ‘Lettres arabes (II)’, Annales islamologiques 16 [1980], 1–29 at no. 9, recto and verso). And a third/ninth-century merchant named Muḥammad b. Wahb informs ‘his boy Fatḥ’, by (dictated?) letter, about the delivery of a very large quantity of limestone (Yūsuf Rāġib, ‘Lettres arabes (I)’, Annales islamologiques 14 [1978], 15–35 at no. 7). See also the third/ninth-century P.Marchands III 35.

45 Tombstones for women described as their owners’ fatāt: Wiet, Catalogue général, 9:88, no. 3341; Wiet, Catalogue général, 4:38, no. 1290 (has fatā instead of fatāt); and ibid., 4:38, no. 1290.

46 The use of ṣabī and ṣabiyya to indicate a slave’s young age can most clearly be seen in a slave dealer’s ledger from the fourth/tenth or early fifth/eleventh century (published in Richards, ‘Fragments’) which distinguishes between a jāriya, ghulām and ṣabī. Unambiguous are also the legal documents P.Cair.Arab. I 37 = P.World, 189 = Chrest.Khoury I 21 (393/1003) and P.Cair.Arab. II 119 (348/960). The sender of the third/ninth-century P.Khalili I 17 ends this letter with a request to convey greetings to a number of persons bearing typical slave names (Ṣafrāt, Dubāb, Ghazāl) and to ‘all the boys (ṣibyān)’. See also P.MuslimState 24 and 28 (first half of the second/eighth century) and P.Philad.Arab. 75 (241/855). Rāġib, Actes de vente, 2:25, para. 59 offers a very short discussion. I wish to thank one of the anonymous reviewers for pointing out that in two Genizah letters, which postdate the tombstones under study in this article, the word jāriya is used to refer to enslaved girls aged two and six. Note that Wiet, Catalogue général, 6:3, no. 2003, a tombstone for one Aḥmad b. ʿAtīq’s jāriya, lacks a profession of faith, which may indicate that the deceased had not yet reached adulthood (see note 36 above). See also Rāġib, Actes de vente, 2:23, para. 57.

47 Much has been written on the umm walad; for a recent study, with references to earlier literature, see Younus Y. Mirza, ‘Remembering the Umm al-Walad: Ibn Kathir’s Treatise on the Sale of the Concubine’, in Concubines and Courtesans: Women and Slavery in Islamic History, ed. Matthew S. Gordon and Kathryn A. Hain (New York: Oxford University Press, 2017), 297–323. See also Elizabeth Urban, Conquered Populations in Early Islam: Non-Arabs, Slaves and the Sons of Slave Mothers (Edinburgh: Edinburgh University Press, 2020).

48 For a discussion of the formation of Islamic laws surrounding the umm walad and the development of concubinage in Islamicate contexts, see, for example, Ingrid Mattson, ‘A Believing Slave is Better than an Unbeliever: Status and Community in Early Islamic Society and Law’ (PhD dissertation, The University of Chicago, 1999), 126–82; Majied Robinson, Marriage in the Tribe of Muhammed: A Statistical Study of Early Arabic Genealogical Literature (Berlin/Boston: Walter de Gruyter, 2020), 107–25.

49 Joseph Schacht, An Introduction to Islamic Law (Oxford: Clarendon Press, 1964), 129 and 169.

50 Jonathan E. Brockopp, Early Mālikī Law: Ibn ʿAbd al-Ḥakam and His Major Compendium of Jurisprudence (Leiden/Boston/Cologne: Brill, 2000), 174–5.

51 Cristina de la Puente, ‘Free Fathers, Slave Mothers and their Children: A Contribution to the Study of Family Structures in al-Andalus’, Imago Temporis, Medium Aevum 7 (2013), 27–44 at 36.

52 Emanuela Kulicka, ‘Remarks on the Typology of Islamic Graves from the Cemeteries on Kom el-Dikka in Alexandria’, Polish Archaeology in the Mediterranean 20 (2011), 483–98 at 485–8 and now Robert Mahler, Changing Life in Egyptian Alexandria: The Testimony of the Islamic Cemetery on Kom el-Dikka (Leuven: Peeters, 2021), 53–5.

53 Emanuela Kulicka, ‘Islamic Necropolis at Kom el-Dikka in Alexandria: Research in the 2010–2013 Seasons’, Polish Archaeology in the Mediterranean 24, no. 1 (2015), 62–72 at 65–7.

54 Speiser, ‘Umayyad, Tulunid, and Fatimid Tombs’, 214–9; Björnesjö and Speiser, ‘South Necropolis’, 122–4.

55 Roland-Pierre Gayraud, ‘Le Qarāfa al-Kubrā, dernière demeure des Fatimides’, in L’Égypte fatimide: son art et son histoire, ed. Marianne Barrucand (Paris: Presses de l’Université de Paris-Sorbonne, 1999), 443–64. See Yūsuf Rāġib, ‘Sur deux monuments funéraires du cimetière d’al-Qarāfa al-kubrā’, Annales islamologiques 12 (1974), 67–83 for a commemorative inscription that decorated the mausoleum of Taghrīd (d. 385/995), the mother of the Fatimid caliph al-ʿAzīz, in Fusṭāṭ.

56 Ibn Yūnus, Taʾrīkh Ibn Yūnus al-Miṣrī, ed. ʿAbd al-Fattāḥ F. ʿAbd al-Fattāḥ, 2 vols. (Beirut: Dār al-Kutub al-ʿIlmiyya, 1421/2000), 1:414.

57 Al-Kindī, Akhbār quḍāt Miṣr, ed. Rhuvon Guest, The Governors and Judges of Egypt (Leiden: E.J. Brill, 1912), 6–476 at 135 dates the beginning of the whitewashing of a tomb’s superstructure in Egypt to the late second/eighth century. See also al-Maqrīzī, Kitāb al-Muqaffā al-kabīr, ed. Muḥammad al-Yaʿlāwī, 8 vols. (Beirut: Dār al-Gharb al-Islāmī, 1411/1991), 1:180; compare Ibn al-Zayyāt, Kitāb al-Kawākib al-sayyāra fī tartīb al-ziyāra fī al-Qarāfatayn al-kubrā wa’l-ṣughrā, ed. A. Taymūr (Cairo: al-Maṭbaʿa al-Amīriyya, 1325/1907), 55–6. For archaeological attestations, see Björnesjö and Speiser, ‘South Necropolis’, 123; Roland-Pierre Gayraud, ‘Isṭabl ʿAntar (Fostat) 1992: rapport de fouilles’, Annales islamologiques 29 (1994), 1–27 at 7 and fig. 26. Kulicka, ‘Remarks’, 485 and Géza Fehérvári et al., The Kuwait Excavations at Bahnasā/Oxyrhynchus (1985–1987): Final Report (Kuwait: Dār al-Āthār al-Islāmiyyah/Kuwait Foundation for the Advancement of Sciences, 2006), 51 write that Abbasid-era grave-markers excavated in (respectively) Alexandria and Bahnasā were plastered. They do not indicate whether or not the tombs were whitewashed. Note that a page from a manuscript excavated in the Fayyūm and dated to the third/ninth or fourth/tenth century contains an illustration depicting two stepped grave-markers painted over with a floral design; see David S. Rice, ‘The Oldest Illustrated Arabic Manuscript’, Bulletin of the School of Oriental and African Studies 22, no. 2 (1959), 207–20 and Geoffrey R.D. King, ‘The Earliest Islamic Illustrated Manuscript, the Maqāmāt and a Graveyard at Suḥār, Oman’, in Arab Painting: Text and Image in Illustrated Arabic Manuscripts, ed. Anna Contadini (Leiden/Boston: Brill, 2007), 95–102, esp. 96–8.

58 Rāġib, ‘Pierres de souvenir’, 333.

59 Ibid., 348–9.

60 Wiet, Catalogue général, 8:170–1, no. 3186.

61 In addition to the example that follows, see also Hawary and Rached, Catalogue général, 1:12, no. 13; ibid., 1:13, no. 14; ibid., 1:18, no. 20; ibid., 1:25, no. 27; ibid., 1:54, no. 73; ibid., 1:137–38, no. 209; Gotthelf Bergsträsser, ‘Zur ältesten Geschichte der kufischen Schrift: Zwei altarabische Grabsteine im Leipziger Kulturmuseum’, Zeitschrift der Deutschen Vereins für Buchwesen und Schriftum 2, no. 5–6 (1919), 49–66 at 50, no. 1; Wiet, Catalogue général, 4:46, no. 490; Wiet, Catalogue général, 8:49, no. 2905; Wiet, Catalogue général, 9:65, no. 3299; ibid., 9:113, no. 3381.

62 Hawary and Rached, Catalogue général, 1:10, no. 10, lines 9–14.

63 In first/seventh- to sixth/twelfth-century southern Egypt and northern Nubia, supplicatory prayers on tombstones are likely to have been recited (Jacques van der Vliet, ‘“What is Man?”: The Nubian Tradition of Coptic Funerary Inscriptions’, in Nubian Voices: Studies in Christian Nubian Culture, ed. Adam Łajtar and Jacques van der Vliet [Warsaw: The Faculty of Law and Administration and the Institute of Archaeology of the University of Warsaw/Fundacja im. Rafała Taubenschlaga, 2011], 171–224 at 197). Some early Islamic graffiti, with statements that are sometimes very similar to funerary texts, occasionally hint at their oral performance (Ilkka Lindstedt, ‘Writing, Reading, and Hearing in Early Muslim-Era Arabic Graffiti’, International Qurʾanic Studies Association’s blog, 2 January 2017, https://iqsaweb.wordpress.com/2017/01/02/writing-reading-and-hearing-in-early-muslim-era-arabic-graffiti/). For the Byzantine realm, see Amy Papalexandrou, ‘Echoes of Orality in the Monumental Inscriptions of Byzantium’, in Art and Text in Byzantine Culture, ed. Liz James (Cambridge: Cambridge University Press, 2007), 161–87.

64 Hawary and Rached, Catalogue général, 3:141, no. 1091; Wiet, Catalogue général, 4:6, no. 1238; ibid., 4:114, no. 1454. ʿAbd al-Tawab, Stèles islamiques, 2:108, no. 257 cites a part of these verses. The translation of Qurʾān. 67:1–3 is Abdel Haleem’s.

65 See Diem, The Living and the Dead for prayers found on early Islamic tombstones and their variations.

66 Such as ʿAbd al-Tawab, Stèles islamiques, 1:19, no. 18; Hawary and Rached, Catalogue général, 1:53, no. 71; Hawary and Rached, Catalogue général, 3:20, no. 841; ibid., 3:46–7, no. 896; Wiet, Catalogue général, 2:99, no. 592; Wiet, Catalogue général, 4:88, no. 1399; ibid., 4:111, no. 1445; Wiet, Catalogue général, 7:79, no. 2577; Wiet, Catalogue général, 9:6–7, nos. 3207–8.

67 Wiet, Catalogue général, 9:199, no. 3365. See ʿAbd al-Tawab, Stèles islamiques, 2:107, no. 256 for another example.

68 For example, ʿAbd al-Tawab, Stèles islamiques, 1:138, no. 137; Hawary and Rached, Catalogue général, 3:53–4, no. 910; ibid., 3:99, no. 1005; Wiet, Catalogue général, 2:92, no. 579; ibid., 2:97, no. 588; ibid., 2:184, no. 769. Such a prayer also features in preserved letters of condolence dated to the first/seventh to third/ninth century. See Khaled Younes, ‘Arabic Letters of Condolence on Papyrus’, in New Frontiers of Arabic Papyrology: Arabic and Multilingual Texts from Early Islam, ed. Sobhi Bouderbala, Sylvie Denoix and Matt Malczycki (Leiden/Boston: Brill, 2017), 67–100, no. 1 (lines 1, 13 and 24; with commentary), no. 2 (lines 6–7) and no. 5 (lines 5–6); see also no. 3 (line 7) and no. 4 (line 4).

69 Schöller, The Living and the Dead, 107–45, esp. 131–4. Compare Rāġib, ‘Pierres de souvenir’, 340.

70 Bittar, Pierres et stucs, 46, no. 5.

71 For the assertive speech act, see John R. Searle, Mind, Language and Society: Philosophy in the Real World (New York: Basic Books, 1998), 148.

72 Al-Shāfiʿī, Kitāb al-Umm, ed. Rifʿat F. ʿAbd al-Muṭṭalib, 11 vols. (al-Manṣūra: Dār al-Wafāʾ, 1422/2001), 9:386; Ibn Abī Zayd al-Qayrawānī, al-Nawādir wa’l-ziyādāt ʿalā mā fī al-Mudawwana min ghayrihā min al-ummahāt, ed. Muḥammad Amīn Būkhubza, 15 vols. (Beirut: Dār al-Gharb al-Islāmī, 1999), 1:564–5; Ibn Rushd, al-Bayān wa’l-taḥṣīl wa’l-sharḥ wa’l-tawjīh wa’l-taʿlīl li-masāʾil al-Mustakhraja, ed. Muḥammad Ḥajjī et al., 20 vols. (Beirut: Dār al-Gharb al-Islāmī, 1984–7), 2:252; Saḥnūn, al-Mudawwana al-kubrā, 16 vols. (Cairo: Maṭbaʿat al-Saʿāda, 1323/1905), 14:11. For a discussion, see Oßwald, Sklavenrecht, 183–4.

73 Schöller, The Living and the Dead, 175–6.

74 For a description and partial transcription of the tombstone, see Bozcu, ‘Türk ve İslam’, 213–4.

75 For tombstone decorations, see Rāġib, ‘Pierres de souvenir’, 335 and Małgorzata Redlak, ‘Ornaments on Funerary Stelae of the 9th-12th Centuries from Egypt: Josef Strzygowski’s Publication Anew’, Polish Archaeology in the Mediterranean 20 (2011), 561–74 at 565.

76 Rāġib, ‘Pierres de souvenir’, 323–4.

77 It entered the Museum’s collection in 1909. See Can Kerametli, ‘Türk ve İslâm Eserleri Müzesinde Erken İslâm Devrine Ait Kitabeler’, Şarkiyat mecmuası 6 (1965), 19–22 at 21.

78 Diem, The Living and the Dead, 153–4.

79 Ibn Mākūlā, al-Ikmāl, 1:191, 196, 203–4; al-Dhahabī, al-Mushtabih fī al-rijāl, ed. ʿAlī M. al-Bajāwī, 2 vols. (Cairo: Dār al-Kutub al-ʿArabiyya, 1962), 1:46–8; Ibn Ḥajar al-ʿAsqalānī, Tabṣīr, 1:60–4.

 

Appendix: Qaṣṣāf/Qiṣāf’s Tombstone

Qaṣṣāf/Qiṣāf’s tombstone (see Figure 1) is a rectangular marble slab of 55 cm high and 37 cm wide and originates from one of Fusṭāṭ’s cemeteries.⁷⁴ Like most tombstones from the same period, it has an elegant appearance. A thin decorative plait consisting of S-shaped bends with a dot at both sides of an intersection – a typically third/ninth-century decoration – surrounds the tombstone’s text on all sides except the bottom.⁷⁵ The funerary inscription covers the surface of the entire slab and consists of thirteen lines. Although the epigrapher increased the script’s density and used smaller letters from the end of line 10, he lacked space for the inscription’s last word, which he added just right of middle in the narrow bottom margin. The inscription itself is carefully executed in a common and unadorned ‘Kūfic’ script style (compare the note to the text below), with very regular and often angular letter shapes and decorative serifs. The inscription dates the deceased’s death to 18 Ramaḍān 245/17 December 859, and in all likelihood the tombstone itself was made not long after this date.⁷⁶ At present, the tombstone is kept in the Museum of Turkish and Islamic Arts in Istanbul under the inventory number 2525.⁷⁷

Beside the invocation of God (line 1) and clauses that identify the deceased as a woman named Qaṣṣāf or Qiṣāf (lines 2-3) and record the date of her death (lines 11-13), the funerary inscription ascribes to the deceased a double profession of faith. The first profession of faith is the common shahāda and expresses the belief in God’s oneness and Muḥammad’s prophethood (lines 3-6) and is followed by a common prayer for the Prophet (lines 6-7). The second profession of faith, which cites Qurʾān 22:7, claims that the deceased believes in the existence of Heaven and Hell and in particular eschatological events: God’s raising of the dead from their graves on Resurrection Day (lines 8 and 10) and the Final Judgement (lines 9-10). Contemporary funerary inscriptions very often address these beliefs and use the same wording, including the citation of Qurʾān 22:7.⁷⁸ The tombstone’s contents are nothing out of the ordinary; similar tombstones for free persons have also been preserved.

Text

1 بسم الله الرحمن الرحيم

2 هذا ما تشهد به قصاف ا

3 م ولد كعب بن بحير تشهد

4 الا اله الا الله وحده لا

5 شريك له وان محمدا عبدہ

6 ورسوله صلى الله عليه

7 وسلم وتشهد ان المو

8 ت والبعث حق والجنة وا

9 لنار حق وان الساعة اتية لا ر

10 يب فيها وان الله يبعث من في القبو

11 ر توفيت يوم الاحد لثلاث عشر

12 ة بقين من رمضان سنة خمس واربعين و

13 ماتين

Note to the Text

The patronymic of Qaṣṣāf/Qiṣāf’s master, one Kaʿb, cannot be read with certainty. The patronymic’s first two letters are clear. It starts with a denticle followed by a jīm, ḥāʾ or khāʾ. Its last letter closely resembles a final rāʾ, although its part above the writing line is shorter than that of the rāʾs in al-raḥmān and al-raḥīm in line 1. The letter that the epigraphist wrote (or intended to write) between the first two letters and the final one is not clear. The very top of this letter is identical to the top of a dāl/dhāl and kāf and clearly shows a serif. What is visible of the upper half of this letter consists of a similar curve as is visible in dāls/dhāls or kāfs elsewhere on the tombstone. However, what seem like three scratches, resembling a half-palmette (unattested elsewhere on the tombstone), run through this part of the letter. It is unclear if they are added purposefully (perhaps to indicate that the line should not be read) or accidentally. There is also unclarity as to the lower part of the letter. The short diagonal line at the bottom of the letter ends too high in comparison to other dāls/dhāls and kāfs, creating an odd tip (see Figure 2.a). This short diagonal stroke could be a denticle if the scratches indicate that the letter’s upper half was added by mistake. Read without the remainder of the letter, the diagonal stroke strongly resembles other denticles on the tombstone (see Figure 2.b). These observations lead to the following possible interpretations of the name’s rasm: ىحكر and ىحىر. Rijāl works do not mention a name with the former rasm, but names on the rasm ىحىر, such as Baḥīr, Buḥayr, Buḥtur, Bujayr and Thujayr, feature in their overviews.⁷⁹ One of the latter is most likely the intended name.

Figure 2. Image and line drawings of the name in the owner’s lineage.