Al-Ḥakam I in the Andalusi Sources: His Slaves, Eunuchs, and Concubines

ABSTRACT

Al-Ḥakam I (154-206 AH/771-822 CE) is one of the most controversial characters in the history of al-Andalus. Despite the fact that the extant chronicles of the Umayyad emirate and caliphate in al-Andalus are fundamentally pro-Umayyad, their authors nevertheless depict the emir as a cruel and pitiless character and are never at a loss in finding harsh terms to describe his political and military actions. The personality of the emir is especially evident in the descriptions that the chronicles make of his treatment of the slaves in his court. Firstly, this article seeks to underline the importance that slaves had in the court of al-Ḥakam and, secondly, it provides an analysis of the historians’ depictions of the emir based on the relationship he had with both his male and female slaves. Even though slaves in the Umayyad court have often been treated as secondary characters, their role is fundamental both in the narration of events and in the portrayal that Arab historians present of the sovereigns.

KEYWORDS:

• Al-Ḥakam I
• the Umayyad Emirate of al-Andalus
• slavery
• concubinage
• eunuchs
• violence
• chronicles
• Andalusi court

Introduction: Arab Chronicles and Slavery in al-Andalus

Slavery during the emirate of al-Andalus has been studied only partially. The most complete work to date is that of Mohamed Meouak in which he studies male slaves in the emiral and caliphal courts of the Umayyads.¹ This is the most extensive study both of the palace’s civil servants and of the male Andalusi slaves of the period. Some years prior to Meouak’s work, Joaquín Vallvé published an article in which he analyzed Andalusi social structures based on the information on the Umayyad family made available in the historical chronicles.² Finally, the eunuchs of the period have also been studied, with a focus on their social and family relationships and, above all, their involvement in the politics of the Umayyad court.³

The objective of this article is to study how the historical chronicles portray the third emir of the dynasty, al-Ḥakam I through numerous anecdotes that illustrate his relationship with his slaves – male, female and eunuchs. There is still no comprehensive work on slaves in the courts of al-Andalus. Nor has the use of slaves by Arab historians to characterize royal personages in other periods been studied. This article is therefore a first effort to begin to fill this gap in the secondary literature.

Scholarship has often regarded court slaves to be merely minor secondary characters in constructed stories whose sole function is to serve as testimony of the existence and exploits of rulers and other prominent free individuals. Despite this, a careful reading of the extant chronicles shows that Arab historians take seriously the role of court slaves, in some cases because they were politically important, in other cases because they help reveal the inner workings of the court and in others because they depict characters that we can consider central. Consequently, the relationship and shared experiences of the ruler with his slaves are fundamental to the image that the chronicler wants to convey of the protagonists of his works – the emirs and caliphs.

Although this is a phenomenon that occurs in all periods – a historiographical study of other kingdoms in al-Andalus would yield similar results – my focusing in this article on al-Ḥakam I is deliberate and for several reasons. First of all, there is the strength of his character: the chroniclers describe him as cruel and pitiless. Even though the extant Andalusi sources, which provide a documentary base for the later North African sources on the history of al-Andalus, are clearly pro-Umayyad, historians make liberal use of negative terms and expressions to describe his actions.⁴ Two events are the focus of particular attention: the ‘Day of the Moat’ in Toledo and the ‘Revolt of the Suburb (rabaḍ) of Cordoba’. On both occasions, according to the accounts, the emir acted with unusual violence, or so at least Arab historians perceive it, even though violence is pervasive in the chronicles.⁵

Secondly, the reign of al-Ḥakam can be seen as the turn to consolidation of the Umayyad court in al-Andalus. A set of reforms on his part allowed Cordoba to come into its own as a political centre, having previously been nothing more than the modest capital of an emirate dependent on Baghdad. His reforms allowed his son, ʿAbd al-Raḥmān II, to follow him in his imitation of the Abbasid court and in the introduction of eastern practices and customs in clothing, poetry and music.

Contributing further to the significance of al-Ḥakam’s reign was the fact that it marked a period of expansion of Mālikism in Andalusi territories. The chronicles, again despite their generally pro-Umayyad position, provide close detail regarding the conflicts that derived from the attempts to establish a legal doctrine as well as the behaviour of a despotic and arbitrary emir, whose conduct collided with the emergent and particular form of Muslim morality.

Finally, the reign of al-Ḥakam I has been studied in depth from both the political and the religious points of view but has rarely been delved into from a social point of view – a shortcoming that is common also to other periods of the history of al-Andalus. This is despite the fact that there does exist source material attesting to the social environment during his reign, largely in the form of scattered anecdotes and narratives that often involve his court slaves. A careful analysis of these stories – taken together with an awareness of their generic limitations – can yield valuable insights into social practices under al-Ḥakam’s rule.

Chronicles and Slavery

The Arabic chronicles provide valuable and often unique information on the history of al-Andalus, but their characteristics condition the results of research. For this reason, it is best to clarify some ideas regarding this type of literature and the information it provides on slavery. The Andalusi chronicles, mostly court records, were written for the dynasty and, in most cases, their authors were tied to the court in one way or another. Naturally, this influenced the authors’ selection of information and, especially, their interpretation of events. On some occasions, it even becomes evident how the author’s own position or situation shapes his ideology – what we could call today his own ‘agency’.⁶

We must also take into account that the chronicles do not speak of slavery per se, but rather of slaves. Although the terms may seem to be synonymous, there is a significant difference. Although chroniclers often mention that a specific person is a slave, they seldom provide either definitions of slavery or what it means to be a slave. Whereas legal sources detail the different degrees of slavery and the freedoms and rights of each individual, chroniclers assume that everyone has at least a basic understanding of what these differences are and therefore do not provide definitions, descriptions, or rankings.

The chronicles never use, for example, the words mudabbar or mukātab, preferring instead such general terms as ʿabd, mamlūk, fatā, ṣaqlabī or khādim.⁷ The same occurs with the vocabulary in the feminine gender, i.e. ama, jāriya, and mamlūka. Not even the expression umm walad (concubine-mother) is used very frequently, chroniclers preferring instead to speak of ‘the mother of emir x’ or ‘caliph x’. Some of these terms are ambiguous, as they refer to either a freed person or a slave. In addition, the sources do not pay attention, especially in the context of the court, to the differences between servants and slaves. The confusion is such that the only study we have on the mothers of the Umayyad emirs and caliphs, written by Joaquín Vallvé, at times confuses concubinage with marriage.⁸

Likewise, the terminology used in the Andalusi context must always be checked, since some words have a clearly different meaning from the one they had in the Eastern territories of the Abbasid Empire. For example, in al-Andalus eunuchs are not always called khādim as is the case in the East. Instead, they are generally called ṣaqlabī, but one should be aware that not everyone identified by this term was a eunuch because white slaves who held high positions in the Umayyad period are also referred to by that term.⁹

It is clear that the vocabulary used must be studied in the context of the original sources because the translations that we possess, even the good ones, distort the texts regarding slaves. In some cases, this is due to the use of euphemisms by Arabists, in other cases because the same word is used to translate both slave and serf or because the word ‘slave’ is used to refer to different categories of enslaved persons. On other occasions, Arabists have taken it upon themselves to ascribe certain characteristics to particular categories of slaves, without the sources providing any grounds for doing so. For example, early modern historiography noted that the ʿabīd in the service of the Umayyad court were Black men, although the sources say nothing about their skin colour; a more influential scholarly tradition, meanwhile, affirms that the ṣaqāliba formed troops, although it is known that many of them did not enter the army.¹⁰ It should also be borne in mind that secondary literature has largely bypassed the study of army troops. Confusing translations make that study difficult and acquire enormous weight in situations in which medieval authors assume on the reader’s part a degree of background knowledge that modern readers do not possess.

The slaves that appear in the chronicles are overwhelmingly either court slaves or servants of the court. The number of references we have to other kinds of slaves is negligible because the Andalusi chronicles are court records and focus overwhelmingly on court affairs. These slaves are often known by their proper names. The male slaves generally occupy high positions, while the female ones are favourite concubines or mothers of caliphs. Frequently sources allow us to reconstruct a brief biography of these slaves and inform us not only about the functions they performed, but also about their importance in political affairs.

Finally, in addition to these categories of slaves, the chronicles also mention captives. These individuals mostly appear anonymously in groups. References to them appear always to relate to the power of the ruler whose goal is, on the one hand, to capture and bring to al-Andalus as many captives as possible on the heels of military victories and, on the other hand, to rescue as many of his own captive subjects as possible.¹¹

Apart from the characteristics of the chronicles directly related to slavery, some limitations in the studies that come from the Arab Andalusi sources must also be mentioned. Although in the case of the emirate some texts from the ninth century are preserved, the vast majority of historical accounts come to us via the eleventh-century historian Ibn Ḥayyān.¹² This author selects, synthesizes and writes in a very personal way and, although some historians are very present in his work such as Aḥmad al-Rāzī (d. 344/955), it is impossible in most cases to know his sources. We must assume that Ibn Ḥayyān’s selection of narratives from the historians who preceded him is never random, but it is often impossible to follow the diachronic evolution of the stories in question because the original works were lost or because Ibn Ḥayyān does not cite his sources in a systematic way.

In this article I have taken from Arabic chronicles detailing the life and reign of al-Ḥakam I all the accounts that show some kind of relationship between the emir and his slaves. Naturally, I always note the sources of the stories, but the stories are almost always unique, lacking attestations or variations elsewhere in the historical record, which means that only exceptionally can a diachronic evolution be traced or parallels be drawn between different versions. Despite this limitation, which is common to much of the study of the history of al-Andalus, the collection and analysis of the content of these small narratives is of great interest for the study of the Andalusi court in general, the reign of al-Ḥakam I and, most especially, of the slaves who lived close to Andalusi royal power. Taking these constraints into consideration, it is nevertheless possible, using the wealth of information found in these sources, to write the history of slavery during a specific period.

Al-Ḥakam I, Master of Slaves

Al-Ḥakam acceded to the throne in the year 180/796 after the death of his father, Hishām I, who ruled al-Andalus for only twelve years. His mother was Zukhruf, a slave given as a present by Charlemagne to his grandfather, the first emir of al-Andalus, ʿAbd al-Raḥmān I (r. 138-172/756–788) with the signing of a peace treaty.¹³ Al-Ḥakam was not initially destined to reign, not because he was the son of an umm walad (concubine-mother), since almost all the Andalusi emirs and caliphs were sons of slave women, but because he had an older brother, ʿAbd al-Malik. Hishām I imprisoned his son ʿAbd al-Malik, however, for reasons not given by the sources, and this brought al-Ḥakam to office.

Several biographies of al-Ḥakam I, the third emir of al-Andalus, have been written, but his relationship with his slaves has never been analyzed nor has the importance of slaves in the great events that took place during his reign been assessed.¹⁴ Naturally, historians mention his slaves, but they often do so while confining themselves to describing their deeds and, above all, without considering the political or social consequences of their legal status. Not only are slaves present in the military actions he undertook, but they are also, to a greater or lesser extent, present in the reforms he carried out in restructuring government institutions. As is well known, al-Ḥakam undertook important reforms that remained in effect until the fall of the caliphate. The anonymous chronicle, Dhikr bilād al-Andalus, tells us, for example, that it was al-Ḥakam who introduced the maẓālim magistracy in al-Andalus and that the first to be in charge of it was Masarra, a eunuch.¹⁵ The sources also tell us that he was the first Umayyad to recruit a private body-guard and take into his service slaves (al-mamālik al-mustariqqīn).¹⁶ These and other scattered accounts demonstrate that slaves were numerous and relevant in the life and government of al-Ḥakam I, both in the public and private spheres.

It is true that slaves are present in the biographical accounts of all the Umayyad rulers and this subject deserves a broad study that is not limited to only one of them. In fact, historical texts usually begin the histories of kingdoms by pointing to the ability or inability of a given sovereign to keep the peace in his territories as well as his strength against his enemies. The best way to measure his strength is through the number of captives he was able to take and, more importantly, his ability to prevent Muslims from falling into enemy hands. Thus, Dhikr bilād al-Andalus states twice that at the time of the emir Hishām I there were no longer any Muslim captives in Christian territory.¹⁷ This topos, which is used frequently in the chronicles, is intended to express the idea that the military supremacy of al-Andalus over the Christian kingdoms was predominant and that, as a result, it was the Muslims who took captives in battles and not vice versa. The Dhikr uses another similar topos in relation to the reign of al-Ḥakam I, that of the woman who approaches the sovereign crying because she has relatives who are being held captive and he decides immediately to undertake jihād against the northern territories.¹⁸

This article will make a brief analysis of who the slaves of al-Ḥakam were and of the role they played in the public and private domains. It will consider male and female slaves separately because not only was their role at court different, but also because, as will be shown, the relationship that the emir had with them was of a completely different nature. The article concludes with an analysis of the presence of slaves in stories that illustrate the religiosity of the emir and of his way of dealing with the Mālikī jurists.

Al-Ḥakam I’s Male Slaves

Like his predecessors and successors, the emir had a significant number of slaves of both sexes. The palace’s male slaves basically performed two functions: one was to accompany the sovereign, guarding him, and delivering his messages, and the other was to fight in his army in which they held various ranks. These functions were not incompatible and we read of slaves who performed both.¹⁹ Among these slaves were eunuchs, whose number must have been small due to their high cost. Their functions were like those of the ‘whole’ slaves (faḥl), with some differences that their status as eunuchs granted them, among them the right to move throughout the entirety of the palace (qaṣr) in both the men’s and women’s quarters.²⁰

We also have information about male slaves who performed the functions of servants and messengers inside and outside the palace who are often called khādim khāṣṣ, the term khāṣṣ implying that they belonged to the private sphere of the sovereign and enjoyed his confidence.²¹ This word is ambiguous and can refer to free serfs or to slaves.

Thus, for example, the Muqtabis relates that it was one of these private servants (khādim khāṣṣ) who delivered a letter to the governor of Toledo, ʿAmrūs b. Yūsuf, instructing the latter to execute the rebellious nobles of that city.²² The text tells us that the slave, who was accompanying the emir’s son, the future ʿAbd al-Raḥmān II, managed to hand the letter directly to the governor without saying a single word.²³ It was this letter that gave rise to the bloody ‘Day of the Moat’ that took place in the early years of al-Ḥakam’s reign, in which a large number of notables of the city were deceived and then executed.²⁴ The anecdote attests both to the trust that could be placed in these servants and to the ability of some of them to go unnoticed at court. The text also shows how often the proximity of the slaves to the powerful made them vulnerable and that their missions as messengers could be very risky.

Other accounts show the familiarity some slaves enjoyed with the emir. In one account, he sends a certain Lorenzo, in whom he has great trust, to act as a spy in the suburb (rabaḍ) of Cordoba and allows himself the leisure to joke with him by calling him ‘son of an uncircumcised woman’, referring to the fact that his mother was a Christian, while at the same time joking about himself, since his own mother was also or had been a Christian.²⁵ On several occasions the chronicles mention Bernardo (Birnat) or Vicente (Bizint), another slave (fatā) with a Christian name, who was by the emir’s side at the most critical times.²⁶

The presence of slaves in the army during emir al-Ḥakam’s I’s rule is highly significant. Reports about the army, like those about slavery, are almost always brief and are scattered throughout the chronicles. Despite this, there is some interesting data regarding the reign of al-Ḥakam I, particularly because it explicitly relates to both institutions. Meouak drew attention to a valuable fragment of Ibn Saʿīd’s (d. 685/1286) Mughrib in which the chronicler refers to the novelties introduced by the emir in the army of al-Andalus.²⁷ Similarly, the historian Aḥmad al-Rāzī credits al-Ḥakam with having increased the number of professional soldiers and slaves among his troops, among whom there was a very significant number of ‘foreign’ slaves, most of them from Christian territories:

Al-Rāzī said: ‘He was the first to increase the number of professional soldiers and guards, to organise his horsemen in front of his door, in order to keep the powerful (al-mulūk) oppressors away from his affairs and he had five thousand mamālīk, three thousand of them horsemen, the “mute”, so called because they were foreigners’.

Meouak points out the value of this passage, both because it tells us when the slaves (mamālīk) became part of the Umayyad troops and because it offers an explanation for the use of the term ‘mute’, namely that it was used to designate troops who did not speak Arabic. In fact, at the head of those troops was a Christian officer, Rabīʿ ibn Teodulf, who must have been able to communicate with most of them in a Romance language.²⁸ The passage is also useful because it gives a clue to how the courts acquired their slaves in the first centuries of Andalusi history, drawing them from the Christian population of the northern kingdoms of the Peninsula.²⁹

The text is also interesting because it reveals why al-Ḥakam I decided to surround himself with trusted soldiers: ‘to keep the powerful oppressors out of his affairs’. Internal uprisings followed one after another during the first years of al-Ḥakam’s reign and this made him distrustful of the population of the capital. He therefore decided to create a professional army that was immune to internal strife and did so by resorting to slave troops. Al-Ḥakam’s reliance on individual slaves, in addition to the slave corps, reflects his suspicious nature and reluctance to rely on men from prominent families interested in promoting their interests and followers. The creation of a slave guard reflected al-Ḥakam’s desire to distance himself from the nobles of Cordoba, and his appointment of a Christian to command of the ḥasham seems similarly motivated.

Although in this article it is not possible to delve into the main events of his reign, which have been extensively studied, the events that led to the need to create the army must be taken into consideration as well as the part that different slaves took in those events.³⁰ It is known that the presence of slaves at court predated the reign of al-Ḥakam I and that he himself had slaves from the beginning. For example, the chronicles include the report that, after the conquest of Barcelona by the Franks in the year 185/801, al-Ḥakam sent a force to meet the Frankish advance. The army suffered a defeat in the port of Arganzon in which some of his commanders lost their lives, among them Manṣūr, a ṣaqlabī eunuch.³¹ Despite this, the sources also relate that there was a significant increase in the number of slaves during his reign. The civil unrest resulted in a growing need and desire for slaves as a means of maintaining order.

In the month of Jumādā II of the year 189/May 804, al-Ḥakam ordered the crucifixion of seventy-two Cordovan noblemen who had promoted a rebellion in the village of Shaqūnda (Secunda) in the hope of overthrowing him: among those executed was a eunuch named Masrūr.³² The following year, in reference to a new revolt in Cordoba, Ibn Ḥayyān states: ‘meanwhile the emir was striving to gather men, reinforce the army (jund) and purchase slaves, preparing weapons and supplies, without taking a break or relaxing’.³³ And, finally, years later, the bloody repression of the ‘Revolt of the Suburb’ took place on the fourteenth day of Ramadan of the year 202/30 March 818. The trigger to this event seems to have been a slave.

The reason was that one of those slaves of the Sultan, who was posted at the door of his palace and had dealings with the people, had given a rusty sword to one of the burnishers of the souk for him to spruce it up and repair it for a price that he had paid in advance asking for it to be done quickly, but the burnisher was extremely lazy, while the slave insisted that he ready it, complaining that he feared being punished by his superior if he did not carry the sword during an inspection, without the burnisher complying, considering it insignificant. Until the unfortunate moment when fate, deadly for those of the village whose time had come, made the slave insist, get angry with the burnisher and shout insults at him in his shop, the latter also becoming irritated and bringing out the worst in him, without hesitation, went for the sword, which was in a corner of his shop, drew it and killed the slave by striking him with it.³⁴

The truth is that any pretext would have sufficed to start the conflict, because the emir was already prepared to combat the rebels, relying on the services of his slaves to help him do so:

He [al-Ḥakam] feared their revolt because he was suspicious of their intrigues, he was cautious and ready, awaiting the assault with the cavalry posted at the gate of his palace, his slaves, who were ready for the clash, taking turns in the command, being fully armed, undergoing inspection twice a day, morning and afternoon, by the trustworthy quartermasters, all of which only increased the resentment, hatred and wickedness of the people until the culminating moment arrived and the people’s provocation against them took place, for which they deserved the punishment for their perversion, following the previous decree of the Almighty.³⁵

According to Ibn al-Qūṭiyya (d. 367/977), whose version of events survives in later chroniclers, al-Ḥakam achieved his objective:

After his victory over the ones from the Suburb, al-Ḥakam made the weight of his authority felt even more, making himself more inaccessible and feared, thus reaching with his subjects a situation that he bequeathed to his sons, whom no one dared to raise their sight to, but rather bowed to: they [his subjects] were educated as slaves who managed to survive in their world, within certain limits. After that, he dedicated himself to acquiring slaves in large numbers, selecting them and training them in horsemanship and the use of weapons; they yielded to him great support. He made them reside at the door of his palace, in continuous shifts, in rooms set up for them there, with compartmentalised stables for their beasts, turning them into a weapon against any incident, since he could order their immediate formation, without delay or need to wait, at any time of the day or night. He exalted [these slaves] by including free men among them and assigning salaries according to their service and merits, treating them well and giving them liberties, to the point that they attacked people and made them feel panic. He imposed a tithe taxation on all the people of the capital and the districts of the kingdom, as he wished and against their wishes, and they complied with that, without anyone daring to say a word from then on, a custom they have followed until today, while the sultan spent from it lavishly, as much as he wanted.³⁶

It is not surprising that al-Ḥakam acted generously after his victory and rewarded the slaves not only with gifts and money, but with the manumission of a small group of them whom he placed in a prominent position as his garde du corps.³⁷ Ibn Ḥayyān states that ‘following that episode, [al-Ḥakam] used to say: “No equipment is better for kings than men, no one protects them better than their slaves, and when they are in a predicament there is nobody to whom they can turn who can help them more immediately”.³⁸ In one of the fragments cited above, another award is mentioned that is relevant from a sociological point of view, namely the inclusion of free men among the slave troops and the assignment of wages. This small phrase shows, on the one hand, that it was considered an honour for a slave to share tasks with free men, which implies that the usual thing was that there was a division of labour according to the type of task to be performed and, on the other hand, that slaves could receive a salary.³⁹ More fundamentally, these interesting words also seem to indicate that, although slaves were commonly considered to have a low social status, there were nevertheless avenues for social mobility, in this case initiated by the emir. It is thus significant that slaves shared tasks with freemen and wage labourers, but the sources fail to specify what real power these slaves actually held within the troop and whether they commanded freemen.

As he notes in the letter that he sent to the inhabitants of Cordoba after the Revolt, al-Ḥakam opted to exercise some mercy: although he meted out harsh punishment on the rebels themselves, he decided not to enslave their women and children, in order, he says, to obtain the favour of God, who had led him to victory.⁴⁰ His decision perhaps also reflects ethical conventions: whereas the enslavement of women and children from other territories and religions was generally accepted practice, the enslavement of free Muslims, which would have been the case here, would have been a serious transgression of religious and legal ethical principles on slavery. But two details indicate that his decision in this instance to treat women and children with mercy cannot be explained simply by the fact that they were Muslims. Firstly, rebellion is regarded as such a serious crime that it nullifies the rights of individuals, even Muslims, who theoretically cannot be enslaved by another Muslim; secondly, al-Ḥakam actually enslaved and castrated some Muslim men after this revolt, which suggests that his decision to spare women and children was specific and intentional.

The Umayyad court subsequently distanced itself from the people and thereby became much less accessible. As is well known, there continued to be internal revolts in the Umayyad territories until the reign of ʿAbd al-Raḥmān III, but they no longer took place in the capital. Likewise, the people treated the sultans with much more respect and kept greater distance.

Concubines, ummahāt al-awlād and qiyān

No record of any legitimate wife of al-Ḥakam I is preserved. Following his custom, in the list he makes of the emir’s relatives, Ibn Ḥayyān first mentions his sons, then his daughters, and finally he refers to his relationship with his concubines. As is commonly the case in Ibn Ḥayyān’s work, he does not mention all of al-Ḥakam’s concubines, but rather just those who stand out because of their influence on the emir. As will be seen below, female slaves play a substantial role in the stories that the chronicles relate of the sovereigns.

Ibn Ḥayyān makes it clear from the start that al-Ḥakam had a special interest in the ‘quality’ of his women:

Muʿāwiya b. Hishām al-Shabīnasī recounts that emir al-Ḥakam was very selective when taking on concubines, preferring those who would bear him children, investigating their family origins and possible education, and giving preference to those who were most outstanding for their virtues. That is why the nobility of the origin in his children is still recognised.⁴¹

This care in the selection of concubines would be continued and intensified by his son and successor ʿAbd al-Raḥmān II.⁴² Discussion of this practice does not, however, seem to be a fixed topos, because in their references to other viziers the chronicles do not mention their wives in this same way. In the case of the emir ʿAbd Allāh, for example, Ibn Ḥayyān, in the third part of the Muqtabis, lists his viziers, judges and other characters related to the court, but he does not expressly refer to his concubines or wives.

However, even though the chronicles attach great importance to al-Ḥakam’s female slaves, the information that we have about them, though abundant, is very dispersed and fragmented, a problem that is common to other issues dealt with by the chroniclers. Among all his slave women, ʿAjab is mentioned most frequently in Arabic sources. She is known to have borne him at least one son, Abū ʿAbd al-Malik Marwān, and the Muqtabis describes her as the favourite of al-Ḥakam I, using the expression karīmat al-amīr al-Ḥakam. Manuela Marín explains that the term karīma was somewhat ambiguous because it could refer to the legitimate daughters or wives of a man as well as to his concubines. In the palatial context it is often used to refer to, as on this occasion, the sovereign’s favourite.⁴³

Others of al-Ḥakam’s female slaves are discussed in lesser detail. Little is known, for instance, of Ḥalāwa, the mother of ʿAbd al-Raḥmān II, other than the fact that she was an umm walad of Berber origin.⁴⁴ Fortunately, on other occasions the stories about slaves are more extensive. Ibn Ḥayyān relates an anecdote from the historian Ibn Ḥārith al-Khushanī (d. 361/971) referring to one of the emir’s concubines:

Ibn Waḍḍāḥ recounts that a concubine of emir al-Ḥakam, with whom she was spending the night, realised in the middle of the night that he was no longer there and could not find him, and she was very seriously offended, thinking that he had left her for another concubine; full of jealousy, she began to follow his tracks unto the garden, in one of whose corners she found him, praying and supplicating fervently. She went back to her place, still thinking about the ‘escape’, and when he came back to her bed, she told him what had happened and asked him, fearing that it was due to some event. He said to her: ‘It was not due to anything of what you have feared, but because I have learned that Muḥammad b. Bashīr has passed away and I am afflicted by losing him, as I cannot replace him with anyone, since I was comfortable with him and satisfied with his justice. As soon as I lost him, I felt something that has made me get out of bed, and that is why I have begun to pray, like someone who asks for refuge does, and imploring God, in the need that He give me good comfort and help me face this loss’.⁴⁵

Ibn Ḥayyān also presents a version of this story that was transmitted by the historian Ibn Ḥabīb (d. 238/853). Just as in Ibn Ḥārith al-Khushanī’s version, Ibn Ḥabīb has the emir give explanations to his concubine with tenderness (although, as Ibn Ḥayyān notes, the story mistakably identifies the deceased as Ibrāhīm b. al-ʿAbbās al-Marwānī).⁴⁶ What makes Ibn Ḥabīb’s version particularly noteworthy, however, is that the latter makes the concubine ʿAjab the protagonist of the story and that it is this enslaved woman herself who speaks in the first person. This is highly unusual in Arabic historical literature, in which slaves usually have no voice and mostly appear as servants of a story’s protagonist. In order to show its veracity, Ibn Ḥabīb points out that it was palace eunuchs who told him the story since they were the only male characters with access to the emir’s harem.⁴⁷ Because of the voice of the slave Ibn Ḥabīb purports to record, it is worth citing the words in full:

The night that it was my turn to be with the emir was the one in which he learned of the death of his cadi Ibrāhīm b. al-ʿAbbās and, after being with me in bed, I felt his absence, after a part of the night, and I started looking for him, finding him on a bench in the house, praying with his feet together, prostrating himself and getting up for long periods, and muttering the prayers insistently. I looked at him with worry until sleep overcame me, and then I was not aware of anything until he shook me up at the crack of dawn.⁴⁸

It is possible that Ibn Ḥārith al-Khushanī also took the story from Ibn Ḥabīb, although he does not acknowledge doing so, and that it thereby reached Ibn Ḥayyān in the eleventh century by both routes. As in other cases, it is difficult to trace the path of transmission because various textual details have been lost, including information about the informant who first related the story. For our present purposes it is significant that all the transmissions maintain the intimacy of the scene and use first-person narration, two features that are generally very uncommon in chronicles from medieval al-Andalus.

ʿAjab remained influential after the death of al-Ḥakam, and in the reign of ʿAbd al-Raḥmān II she was involved in a well-known episode, interceding on behalf of her sister’s son, Yaḥyā b. Zakarīyāʾ al-Khashshāb, who was accused of having committed blasphemy.⁴⁹ She appeared before the emir ‘with her clothes torn, and moaning’ seeking to save her condemned nephew from crucifixion. ʿAbd al-Raḥmān not only did not accept her intercession, but even reprimanded him for her request because he considered that in matters of religion intercession was not possible.⁵⁰ In this case, the words of the emir ʿAbd al-Raḥmān to ʿAjab enable the historian Ibn Ḥayyān to demonstrate the rectitude and religiosity of the sovereign, who does not take pity even before the pleas of a woman, possibly old at that time and devastated by sadness. It is with this intention that Ibn Ḥayyān describes ʿAbd al-Raḥmān’s words as harsh and transcribes them in full. It is also striking that the historian titles his chapter ‘The crucifixion of ʿAjab’s nephew’, granting more importance to the heretic’s relative rather than to the person himself.

Although ʿAjab failed in her attempt to save her relative’s life, she continued to be respected after the death of al-Ḥakam and was wealthy enough to do important pious works that allowed her name to survive on some monuments in Cordoba for centuries.⁵¹ The memory of her as a builder echoes that of another of al-Ḥakam’s concubines, Mutaʿ, who stood out for the same reason:

Among al-Ḥakam’s favourite concubines, of whom memory has remained due to their merit, in their time and later, are ʿAjab, mother of her son Abū ʿAbd al-Malik Marwān, who gave her name to the ʿAjab mosque in the western suburb of Cordoba and the orchard (al-mūnyā) by the river bank that bear the same name, given by her as a pious legacy for the sick; and Mutaʿ, mother of her son Abū ʿUthmān Saʿīd, namesake of her brother Saʿīd al-Khayr, who are sometimes confused because they bear the same name, her name was given to another mosque, towards the west of Cordoba, as well as to the adjoining cemetery, both bequeathed by her, together with many other gifts that she made for pious and charitable purposes, as she was one of the most generous women. Her son Saʿīd passed away in the days of his brother ʿAbd al-Raḥmān b. al-Ḥakam.⁵²

Marín points out that these are the two oldest pious legacies established by women in al-Andalus known to us.⁵³ She also mentions that thanks to some legal texts, it is known that the work on the orchard (al-mūnyā) of ʿAjab was carried out with slave labour.⁵⁴ This is valuable information because the Andalusi preserved sources do not tell us what type of labour was used on farms.⁵⁵

Emir al-Ḥakam’s Qiyān

We know of several female slave singers (qiyān) who belonged to al-Ḥakam I. This emir’s qiyān have been carefully studied by Dwight F. Reynolds who points out that that no qiyān are documented for the reigns of of ʿAbd al-Raḥmān I and Hishām I.⁵⁶ He notes that al-ʿUmarī (d. 749/1349) mentions that the first female slave singers of the Andalusi court were those belonging to al-Ḥakam I. We have the names of six qiyān from that period: two named ʿAzīz and four others named Baḥja, Muḥja, Fātin, and Fātik. The evident play on words in the naming of the slaves prompts Reynolds to consider that these names could have been the result of joking by the emir or that al-ʿUmarī could have duplicated the names taking them from different sources and that there really were only three qiyān and not six.⁵⁷ However, existing onomastic studies seem to support the idea of a play on words and of the pleasure owners felt in calling their slaves by beautiful or funny names.⁵⁸

There are numerous anecdotes that illustrate the pleasure the emir took in the company of educated female slaves and the celebration of nocturnal soirees in which they entertained him. The Dhikr affirms that the emir himself was a good poet and that some of his verses were dedicated specifically to his partying companions:

Al-Ḥakam was a magnificent poet and among his works stand out the verses that he dedicated to five of his female slaves whom he loved very much and preferred to the rest of his women, since he always tried to please them. One day he tried to get another one to come in with them, but they opposed this and got up angry and jealous; when they had gone, he recited: […].⁵⁹

The interesting thing about this passage is that it is the female slaves who allow themselves the liberty of getting angry with the emir, and he who yields to them out of the desire to ‘please them always’. Two of those verses even include the topos ‘slave of love’, he being, for a moment, the one who becomes their slave as they become his owners:

For excessive love he became a slave

the one who had been before the master⁶⁰

One of his qiyān, ʿAzīz, was rewarded after one of her performances with 10,000 dirhams and a length of cloth (qimāsh).⁶¹ Azīz received the award precisely for having consoled the emir from the nostalgia he was suffering from the absence of another of his slaves, called Hayn, whom he had left in the palace. The story is full of literary topoi about her great beauty, which allow us a glimpse at what may have been the nature of the relationship between the sovereign and his singing slaves – slaves who, as in this case, could also be his concubines and the mothers of his children.⁶²

A similar scene appears in two other anecdotes. In these stories the emir also resorts to his qiyān, the other ʿAzīz and Muḥja, to console him for the longing he felt for another of his concubines. In one of the cases, it is about a slave he had quarrelled with after having been greatly in love with her.⁶³ Here, the scene in which the emir ceases to be almighty in his own palace is repeated: letting the slave become angry, he falls victim to his own passion.

The theme of such anecdotes is consolation. But the emir does not use the singing slaves primarily for this purpose, but rather to provide entertainment and pleasure. The reward received by ʿAzīz is not exceptional and al-ʿUmarī leaves testimony in his text both to the verbal praise they received from al-Ḥakam for a job well done as well as to the generous amounts of money he bestowed on them for it.⁶⁴ These scenes show, on the one hand, an emir who is complacent, affectionate, and sensitive towards his slaves and, on the other, a court that had already reached a certain degree of sophistication. This use of slavery, holding enslaved women as personal entertainers, was new in al-Andalus and its introduction at the Umayyad court must be ascribed to al-Ḥakam I.

The fascination of some historians with the court of ʿAbd al-Raḥmān II and the role played by his musician Ziryāb has led them to think that sophistication in the court was a later phenomenon.⁶⁵ Nevertheless, although the musician Ziryāb is usually associated with ʿAbd al-Raḥmān II, whom he served and whose court he modified by importing oriental tastes, the fact that it was al-Ḥakam I who authorized him to move to al-Andalus with his family should not be forgotten. Ziryāb had written to him offering his services and the emir sent his servant Manṣūr al-Yahūdī al-Mughannī to receive him, but when it was time for the meeting to take place al-Ḥakam had already died.⁶⁶ Ziryāb therefore had to ask the newly appointed emir to accept him which he did.⁶⁷ Historians tend to overlook this fact because the musician’s biography has always been linked to the figure of ʿAbd al-Raḥmān II and his court, but it is very possible that Ziryāb felt attracted to al-Ḥakam’s I court by the news that had come to him about the emir’s tastes, including his delight in surrounding himself with concubines, amongst them beautiful and cultured singers. The ownership of qiyān must have become a custom, since it is documented that other sons of his, such as al-Mughīra or the future emir ʿAbd al-Raḥmān, also enjoyed these types of female slaves while their father was still alive.⁶⁸

Reynolds posits that the possible contrast between al-Ḥakam’s rude and cruel personality and the kind treatment he granted his female slaves may have resulted from the fact that he kept the latter aspect of his personality intimate and did not allow himself to be seen when enjoying the company of the qiyān, as stated by al-ʿUmarī.⁶⁹ This trait is however not unique to al-Ḥakam I since it is known that other rulers were also very jealous of their private space. Perhaps the answer is simpler, namely that he was a person who exhibited both facets, since we know of many characters in history capable of showing great sensitivity for love, music, or family who were at the same time hard and even very cruel in public life. Additionally, it must be noted that al-Ḥakam does not act cruelly with his male slaves either, and when he is tempted to do so, the jurists close to the court stop him. Although male slaves take part in various bloody events, they do so mostly as executioners and not as victims. As this article has shown, he placed enormous trust in them and conferred upon them the highest tasks.

Male and Female Slaves Kept in Order to Display the Religiosity of the Emir

Although there are not many anecdotes referring to the emir’s religiosity, the few that do exist sometimes feature slaves. Slaves are presented as the victims of the religious rigour of al-Ḥakam, who shows his desire to apply Muslim law fairly. The ninth century is the period of the consolidation of the Mālikī school as the majoritarian legal doctrine in al-Andalus and these stories of the pro-Umayyad and Mālikī chroniclers serve to attest to the will of the emir, whose violent and erratic behaviour often does not fit in well with the precepts of religious law.⁷⁰ From a historiographical point of view, the choice of the slave as a possible victim in these anecdotes seems not to be merely accidental, since the slave represents the weakest and most vulnerable stratum in the court’s hierarchy and, consequently, generosity towards those who deserve the least should be the most valued by readers of that period.

Ibn Ḥayyān recounts that on one occasion the jurist Shabṭūn, towards whom the emir al-Ḥakam I felt a special preference, had to intercede in favour of a slave (khādim khāṣṣ) whose hand the sovereign intended to cut off for having been the bearer of bad news:

He had a preference for the faqīh Ziyād b. ʿAbd al-Raḥmān, known as Shabṭūn. One day when he was present, having become angry with a servant (khāṣṣ) of his for having brought him an unpleasant message, as the sovereign gave orders to cut off his hand, Ziyād said to him: ‘God save the emir! Mālik b. Anas transmitted to me, according to information that he had, the following hadith: “Whoever represses his anger, being able to give it free course, on the day of Resurrection God will grant him safety and faith.”’ Then the emir’s anger calmed down and he said: ‘Do you swear by God that Mālik transmitted this hadith to you?’ Ziyād said: ‘I swear by God that Mālik transmitted it to me.’ And the emir gave orders to the servant to be left alone and pardoned him.⁷¹

The ire that characterizes al-Ḥakam is curbed by a prestigious faqīh, who makes him see the light through the teachings of the founder of the Mālikī school with whom he had studied himself. In order to understand this story in which the emir submits to the authority of the jurists, it must also be borne in mind that Ibn Ḥayyān often describes the characters based on anecdotes that show how far or how close they were to the good moral behaviour of a Mālikī Muslim of his time. He does so with all rulers, but there was a particular need to do so in this instance, since he was writing his chronicle 250 years after al-Ḥakam’s reign, and he needed to make his book understandable. His contemporaries lived in a period when the Mālikī legal school was official and its foundations were known to the majority of the population.

Other stories also show an emir who is compliant with the law and obeys the judgements of the jurists. Several of the stories suggest that al-Ḥakam adapted his conduct to Mālikī jurisprudence, and relate how the law of slaves was introduced and applied in al-Andalus:

Muḥammad b. ʿUmar b. Lubāba relates, based on the account by ʿAbd Allāh b. Wahb, the following: A certain slave and mother of a son of the emir al-Ḥakam turned out to be someone else’s property, according to the ruling of the cadi Muḥamamd b. Bashīr, who wrote to the emir informing him of the evidence before him of said property right and demanding that he hand her over to comply with the sentence. [The emir] submitted to the sentence, and the slave had to be inspected and her price paid to the holder of the right, according to a fatwā of Mālik b. Anas, to whom they resorted as a solution to the case and whose acceptance by the emir was considered one of his virtuous acts.⁷²

This passage in the Muqtabis follows immediately after another anecdote in which reference is made to the conduct of Muḥammad b. Bashīr with one of his concubines. It allows us to consider several relevant issues: first, that there were those who took the liberty to sue the emir himself for one of his possessions, in this case a very valuable one because not only was she his slave but also his umm walad; secondly, that the jurists allowed themselves the right to rule against the emir and in favour of the claimant; and, finally, that al-Ḥakam submitted to the Mālikī doctrine that, in a case like this, only permits one solution: financial compensation to the former owner for the loss of his property. A concubine-mother cannot be returned to her former owner, sold, or separated from her household. Motherhood also gives her the right to be manumitted after the death of her owner.

The scene has all the necessary elements to illustrate the authority of the jurists and the legal school they represented, even when confronting the power of the sovereign. From the legal point of view, the passage adapts literally to what is established by the jurisprudence in the matter of slaves. The text is equally interesting from a social point of view since it notes the accessibility that citizens had to the emir to the extent that they were able to lay claims against him. In addition to submitting to the sentence, we can imagine that the price that he had to pay for the slave must have been high, since she was the mother of one of his children.

In conclusion, there is a general tendency in the chronicles to characterize al-Ḥakam as a wrathful character, but one who is respectful of the sentences of the judges and therefore of Mālikī legal doctrine. Despite this, in other episodes al-Ḥakam’s brutality is not concealed and nor is the fact that he at times also disregarded transgressions of Islamic law. Ibn Ḥayyān cites a text by his contemporary Ibn Ḥazm (d. 456/1064) – who was not a Mālikī but a Ẓāhirī – in which the emir’s pitiless behaviour is reflected and in which the victims are also slaves:

The wise jurist ʿAlī b. Aḥmad b. Ḥazm omitted all the mentioned good deeds of this caliph, al-Ḥakam b. Hishām, by saying the following in his epistle on curious news called Naqṭ al-ʿarūs, which circulates among the people:

‘Among the bloodiest public sinners we have had are al-Ḥakam b. Hishām, the one from the Suburb, who, in his arrogance, used to castrate the children of his subjects who stood out for their beauty and take them to his palace as servants. One of them was Ṭarafah b. Laqīṭ, brother of ʿAbd al-Raḥmān b. Laqīṭ, whose name was given to the Ṭarafa mosque in the city of Cordoba, of Huwārī lineage, whose father and brother held various high positions; another was the eunuch Naṣr, who would become a favourite of his son, the emir ʿAbd al-Raḥmān, and who gave his name to Naṣr’s orchard, and whose father was a dhimmī from Carmona, a convert to Islam, who died a few days before his son; another one was Surāj, the one who the mosque in Cordoba is named after; and there were others.’⁷³

Ibn Ḥazm’s passage stands apart from references in other Arabic chronicles. As has been shown, the other authors agree in emphasizing the emir’s violent character, but simultaneously introduce texts attesting to his deference and respect for Mālikī jurists. By contrast, Ibn Ḥazm not only spares no negative adjectives to describe al-Ḥakam personally, but also accuses him of specific outrages, such as having ordered arbitrary castrations of his subjects. In this case, he does not even give as a reason the reprisals that followed the ‘Revolt of the Suburb’ but affirms that the emir simply opted to castrate those who stood out for their beauty without needing a better reason. Ibn Ḥazm knew that all legal schools prohibited castration in Islamic territory, both of slaves and free men, although not the purchase or the possession of eunuchs from other territories.⁷⁴ Consequently, if Ibn Ḥazm’s fragment is placed in the context of his time, the message is that al-Ḥakam represented an apogee of regnal cruelty.

Finally, we should bear in mind that these episodes in which an emir submits to Divine will by accepting the mandates of Muslim jurisprudence are not unique to the reign of al-Ḥakam. The episode of the crucifixion of ʿAjab’s nephew shows ʿAbd al-Raḥmān II as steadfast even in the face of pleas from a loved one. What is of interest in this period is that the sources represent the emir as having been the first to establish principles that would endure through the history of the entire dynasty and, indeed, characterize it.

Concluding Remarks

The main interest of al-Ḥakam’s reign in relation to slavery is that he was the first of his dynasty to increase the presence of slaves considerably both in public and private life, as well as being the first to entrust them with major tasks in an organized and large-scale manner. The importance of his reign to the slow process of Arabization and Islamization of al-Andalus has been repeatedly highlighted as well as its political importance in the consolidation of Umayyad power.⁷⁵ The social revolution carried out by the emir through the incorporation of slaves to his court and his army should also be emphasised. To date, scholars, noting that the slaves in question were usually Christians, have studied them from the point of view of religion and ethnicity, but the social relevance of their standing as slaves has been ignored. The chronicles show that the sovereign pursued a strategy of appointing slaves whom he trusted personally in place of personnel of Arab origin or muwallads (Christian converts to Islam). Their lack of family or tribal ties guaranteed him their loyalty; in return, he granted them certain privileges that allowed them to move around Cordoba with the same liberty as free courtiers.

Relations between the emir and his slaves are equally interesting historiographically. As we have seen in this article, slaves are not irrelevant characters in the historical narrative. The chroniclers want to portray al-Ḥakam’s slaves, but above all they want to portray the emir through his relationship with them. As in the Spanish theatre of Lope de Vega or Calderón de la Barca, the servants are for the author the measuring rod of the character they serve.

The anecdotes discussed here allow us to offer a more complicated picture of al-Ḥakam than has previously been possible: we see him in all his contradictions – contradictions, incidentally, of which he himself seems to have been well aware. His attitude towards slaves served as the measure of the emir’s qualities, good and evil. Ibn al-Qūṭiyya states in his work that al-Ḥakam was ill for seven years before he died, and that during this period he repented of his bad actions.⁷⁶ These beautiful lines serve to testify that he was also aware of the lights and shadows of his government, a reign in which his slaves played a predominant role.

Acknowledgements

The present work has been carried out through the Research Project ‘Gender, Family and Slavery: Slavery and Legal Status in the Structuring of Muslim Families, 8th–12th Centuries’ (PID2019-110663RB-100) financed by the Ministry of Science (Spain). I thank Jelle Bruning, Said Reza Huseini and the reviewers of the article for their valuable suggestions. I would also like to thank my colleague Matthew Gordon for his kind reading of the article. Finally, I thank the copyeditor for his thorough work.

Disclosure Statement

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

Additional information

Notes on contributors

Cristina de la Puente

Cristina de la Puente is a researcher at the Consejo Superior de Investigaciones Científicas (Spanish National Research Council), Calle de Albasanz 26–28, 28037 Madrid, Spain. Email: [email protected].

Notes

1 Mohamed Meouak, Ṣaqāliba, eunuques et esclaves à la conquête du pouvoir: Géographie et histoire des élites politiques ‘marginales’ dans l’Espagne umayyade (Helsinki: Annales Academiae Scientiarum Fennicae, 2004).

2 Joaquín Vallvé, ‘Sobre demografía y sociedad en al-Ándalus (siglos VIII-XI)’, Al-Andalus 42, no. 2 (1977), 323–40; Joaquín Vallvé, ‘Libertad y esclavitud en el califato de Córdoba’, Actas de las Segundas Jornadas de Cultura Árabe e Islámica (Madrid: Instituto Hispano-Árabe de Cultura, 1985), 565–84. D. Fairchild Ruggles has also studied the family ties of the Umayyad family of al-Andalus in ‘Mothers of a Hybrid Dynasty: Race, Genealogy, and Acculturation in al-Andalus’, Journal of Medieval and Early Modern Studies 34, no. 1 (2004), 65–94. This author does not mention Vallvé’s work and bases her study on secondary literature.

3 Cristina de la Puente, ‘Sin linaje, sin alcurnia, sin hogar: eunucos en al-Andalus en época omeya’, in Identidades marginales, ed. Cristina de la Puente (Madrid: CSIC, 2003), 147–94; and Cristina de la Puente, ‘Eunuchs in the Emirate of al-Andalus’, in Identity in the Middle Ages: Approaches from Southwestern Europe, ed. Flocel Sabaté (Leeds and Amsterdam: ARC Humanities Press, 2021), 148–65.

4 Regarding Muqtabis’s narration on the rebels and the loyalists and the adjectives and expressions used to define the ‘Other’, see Alberto Venegas Ramos, ‘Procesos de alteridad e identidad política en la Crónica de los emires Alhakam I y Abderramán I (sic) entre los años 796 y 847 [Almuqtabis II-1]: Leales e infieles’, Revista de Estudios Extremeños 72, no. 2 (2016), 907–36.

5 Regarding the Revolt of the Suburb, which has been the subject of so much discussion, see Al-Mulk II Época, n° 16 (2018), dedicated entirely to the subject. Regarding violence in al-Andalus, see for example Cristina de la Puente, ‘Cabezas cortadas: símbolos de poder y terror: al-Andalus s.II-VIII/IV-X’, in El cuerpo derrotado: cómo trataban musulmanes y cristianos a los enemigos vencidos, ed. Maribel Fierro and Francisco García Fitz (Madrid: CSIC, 2008), 319–47; and Cristina de la Puente, ‘Violencia y misericordia con los esclavos: regulación de derechos y deberes según la doctrina mālikí’, in Crueldad y compasión en la literatura árabe e islámica, ed. Delfina Serrano (Madrid: CSIC, 2011), 179–202.

6 See for example Luis Molina, ‘Técnicas de amplificatio en el Muqtabis de Ibn Ḥayyān’, Talia Dixit: revista interdisciplinar de retórica e historiografía 1 (2006), 55–79.

7 Regarding legal slave categories (such as mudabbar and mukātab), see Cristina de la Puente, ‘Entre la esclavitud y la libertad: consecuencias legales de la manumisión según el derecho mālikí’, Al-Qanṭara 21, no. 2 (2000), 339–60.

8 Joaquín Vallvé, ‘Sobre demografía y sociedad en al-Ándalus (siglos VIII-XI)’, Al-Andalus 42, no. 2 (1977), 327, 329.

9 Meouak, Ṣaqāliba, eunuques et esclaves, 221–32.

10 See Meouak, Ṣaqāliba, eunuques et esclaves, 223, citing data to disprove the idea that the ṣaqāliba formed an army unit, first published by Lévi-Provençal and then by Ayalon, Martinez-Gros and recently by Jessica Coop.

11 Ibn ʿIdhārī al-Marrākushī, Kitāb al-Bayān al-mughrib fī akhbār al-Andalus wa-l-Maghrib, vol. 2, ed. G.S. Colin and É. Lévi-Provençal (Leiden: E.J. Brill, 1951), 297–98; al-Maqqarī, Azhār al-riyāḍ fī akhbār ʿIyāḍ, vol. 4, ed. Saʿīd Aḥmad al-Aʿrabī and Muḥammad b. Tāwit (Rabat: Ṣundūq Iḥyāʾ al-Turāth al-Islāmī, 1978), 106–8.

12 On Ibn Ḥayyān and his work see Emilio García Gómez, ‘A propósito de Ibn Ḥayyān’, Al-Andalus 9 (1946), 395–423; María Jesús Viguera, ‘Cronista de al-Andalus’, in España, al-Andalus, Sefarad: síntesis y nuevas perspectivas, ed. Felipe Maíllo Salgado (Salamanca: Universidad de Salamanca, 1988), 85–98; Luis Molina, ‘Sobre la historia de al-Rāzī: nuevos datos en el Muqtabis de Ibn Ḥayyān’, Al-Qanṭara 1 (1980), 435–42; María Luisa Ávila Navarro, ‘La fecha de redacción del Muqtabis’, Al-Qanṭara 5 (1984), 93–108; and Luis Molina, ‘Ibn Hayyan’, in Diccionario Biográfico de la Real Academia de la Historia, https://dbe.rah.es/biografias/16692/ibn-hayyan.

13 Dhikr bilād al-Andalus = Una descripción anónima de al-Andalus, ed. and tr. Luis Molina (Madrid: CSIC, 1983), 124/133. Even though Zukhrūf was a slave-mother (umm walad), she is sometimes called a ‘Carolingian princess’, under the mistaken assumption that since she was given as a present by Charlemagne and must have been part of his family. See also Philippe Sénac, Les Carolingiens et al-Andalus (VIII^e-IX^e siècles) (Paris: Maisonneuve et Larose, 2002), 71–97.

14 See a comprehensive summary of his reign in Évariste Lévi-Provençal, Historia de España, vol. 4 (Madrid: Espasa Calpe, 1957), 91–128. Although some of Provençal’s rendition is outdated, the account given follows closely the information given by the Muqtabis by Ibn Ḥayyān. See also a comprehensive summary of his reign in Margaret E. Anderson, ‘The Relationship of the Amīr al-Ḥakam I with the Mālikī fuqahāʾ in al-Andalus’ (MA thesis, Institute of Islamic Studies McGill University, 1965), 12–41 https://escholarship.mcgill.ca/concern/theses/vm40xw422.

15 Dhikr bilād al-Andalus, 107/136. Regarding the implementation of the maẓālim in al-Andalus, see Anderson, ‘The Relationship of the Amīr al-Ḥakam I’, 56–7, and Mathieu Tillier, ‘The Mazalim in Historiography’, in The Oxford Handbook of Islamic Law, ed. Anver M. Emon and Rumee Ahmed (Oxford: Oxford University Press, 2015), 357–80.

16 Dhikr bilād al-Andalus, 104/133. See Hugh Kennedy, Muslim Spain and Portugal: A Political History of al-Andalus (London-New York: Routledge, 1996), 43.

17 Dhikr bilād al-Andalus, 100/129 and 101/130.

18 Dhikr bilād al-Andalus, 107/136.

19 Regarding tasks assigned to slaves during the Umayyad period, both during the emirate and the caliphate, see Meouak, Ṣaqāliba, eunuques et esclaves.

20 Regarding the characteristics of eunuchs in al-Andalus, see de la Puente, ‘Sin linaje, sin alcurnia, sin hogar’ In Ibn Ḥayyān, al-Sifr al-tānī min Kitāb al-Muqtabas, ed. and intro. Maḥmūd ʿAlī Makkī (Riyadh: Markaz al-Malik Fayṣal li-l-buḥūth wa-l-dirasāt al-islāmiyya, 1424/2003) (henceforth: Muqtabis II-1), 173r the historian opposes a eunuch horseman (fars khaṣī) to a horseman capable of fathering children (fars faḥl).

21 See, for example, Ibn Ḥayyān, Muqtabis II-1, 87v and 93v.

22 ʿAmrūs (Ambrosio) ibn Yūsuf was a Christian convert to Islam from Huesca, see María Jesús Viguera, Aragón musulmán (Zaragoza: Librería general, 1981), 87.

23 Ibn Ḥayyān, Muqtabis II-1, 93v-94r.

24 Regarding versions of this event in other sources, see María Crego Gómez, ‘La jornada del foso de Toledo según Ibn Faḍl Allāh al-ʿUmarī: Edición y traducción’, Al-Andalus-Magreb 14 (2007), 269–75.

25 Ibn Ḥayyān, Muqtabis II-1, 107r, 111v.

26 Ibn al-Qūṭiyya, Taʾrīkh iftitāḥ al-Andalus, ed. Ibrāhīm al-Abyārī (Cairo and Beirut: Dār al-kitāb al-Miṣrī and Dār al-kutub al-islāmiyya, 1982), 68; Ibn Ḥārith al-Khushanī, Taʾrikh quḍāt Qurṭuba, ed. and tr. Julián Ribera (Madrid: Ibérica, 1914), 48/58-9.

27 Meouak, Ṣaqāliba, eunuques et esclaves, 114, where he analyses Ibn Saʿīd al-Maghribī, al-Mughrib fī ḥulā al-Maghrib, ed. Shawqī Ḍayf, 2 vols. (Cairo: Dār al-Maʿārif, 1953-55), 1:39.

28 Meouak, Ṣaqāliba, eunuques et esclaves, 115; and Ángel Custodio López, ‘El conde de los cristianos Rabīʿ ben Teodulfo, exactor y jefe de la guardia palatina del emir al-Ḥakam I’, Al-Andalus-Magreb 7 (1999), 169–84. Margaret E. Anderson gives a different reason for the motives that al-Ḥakam could have had to place a Christian in charge of tax collection, see ‘The Relationship of the Amīr al-Ḥakam I’, 45–6.

29 Cristina de la Puente, ‘The Ethnic Origins of Female Slaves in al-Andalus’, in Concubines and Courtesans: Women and Slavery in Islamic History, ed. Matthew Gordon and Kathryn A. Hain (New York: Oxford University Press, 2017), 124–42.

30 For a study of the main events of his reign, see Lévi-Provençal, Historia de España, 91–128.

31 Ibn Ḥayyān, Muqtabis II-1, 95v.

32 Ibn Ḥayyān, Muqtabis II-1, 97r. For Masrūr, Meouak, Ṣaqāliba, eunuques et esclaves, 159–60.

33 Ibn Ḥayyān, Muqtabis II-1, 98v.

34 Ibn Ḥayyān, Muqtabis II-1, 105r. Makkī and Corriente in their translation of the text point out that Lévi-Provençal had understood the event the other way around, i.e. that it was the slave who had killed the smith.

35 Ibn Ḥayyān, Muqtabis II-1, 105r. Ibn Ḥayyān gives a different version of this story based on Aḥmad b. Muḥammad b. Khalaf al-Warrāq in Muqtabis II-1, 111r.

36 Ibn Ḥayyān, Muqtabis II-1, 110r.

37 Chroniclers rarely mention slave manumissions in the Andalusi courts. This is a question that would merit a separate study, as would the client networks of the freedmen with the sovereigns. On the other hand, I believe that the results of such a study would not affect the analysis of the relationship of the kings with their slaves, which is quite different from a legal and social point of view.

38 Ibn Ḥayyān, Muqtabis II-1, 107r.

39 We have very little data on labour in pre-modern Western Islam: see Cristina de la Puente, ‘Mano de obra esclava en al-Andalus’, Espacio, tiempo y forma (UNED) 23 (2010), 135–47.

40 Ibn Ḥayyān, Muqtabis II-1, 104r. See Maribel Fierro for an interpretation of this episode, ‘Violence against Women in Andalusi Historical Sources (Third/Ninth-Seventh/Thirteenth Centuries)’, in Violence in Islamic Thought from the Qur’an to the Mongols, ed. Robert Gleave and István Kristó-Nagy’ (Edinburgh: Edinburgh University Press, 2015), 155–74 at 160 and 173.

41 Ibn Ḥayyān, Muqtabis II-1, 116r.

42 Ibn Ḥayyān, Muqtabis II-1, 146r-146v.

43 Manuela Marín, Mujeres en al-Andalus (Madrid: CSIC, 2000), 37–8.

44 Ibn Ḥayyān, Muqtabis II-1, 116r, 139v; Dhikr bilād al-Andalus, 137/145; Ibn ʿIdhārī al-Marrākushī, Kitāb al-Bayān al-Mughrib fī akhbār al-Andalus wa-l-Maghrib, vol. 2, ed. G.S. Colin and É. Lévi-Provençal (Leiden: E.J. Brill, 1951), 2, 80.

45 Ibn Ḥayyān, Muqtabis II-1, 120r; al-Khushanī, Quḍāt Qurṭuba, 66–67, Ibn al-Qūṭiyya, Taʾrīkh iftitāḥ al-Andalus, 56–7.

46 Ibrāhīm b. al-ʿAbbās al-Marwānī did not live during al-Ḥakam I’s reign, but during that of his son ʿAbd al-Raḥmān II.

47 Regarding the role of eunuchs as go-betweens between the harem and the outside world, see De la Puente, ‘Eunuchs in the Emirate of al-Andalus’, 186.

48 Ibn Ḥayyān, Muqtabis II-1, 120v. Following that, al-Ḥakam I then replaced the jurist who had passed away by another jurist. In Akhbār majmūʿa it is also the female slave who relates the event in first person, 126 (Arabic)/114 (translation).

49 Regarding his sentencing for blasphemy, see Maribel Fierro, La heterodoxia en al-Andalus durante el período omeya (Madrid: Instituto Hispano-Árabe de Cultura, 1987), 57–63.

50 Ibn Ḥayyān, Muqtabis II-1, 175v/414. See Manuela Marín’s interpretation of this event in Mujeres en al-Andalus, 579. In Makkī and Corriente’s translation into Spanish there is an error in that they translate karīma as ‘daughter’ rather than ‘favourite’, which results in al-Ḥakam’s concubine being presented as his daughter.

51 Marín, Mujeres en al-Andalus, 340.

52 Ibn Ḥayyān, Muqtabis II-1, 116r-116v.

53 Marín, Mujeres en al-Andalus, 339. Even though in the text Marín mentions that she is a concubine, in the index she calls her ‘wife’.

54 Marín, Mujeres en al-Andalus, 340.

55 De la Puente, ‘Mano de obra esclava en al-Andalus’, 135–47.

56 See also Simone Prince-Eichner, ‘Embodying the Empire: Singing Slave Girls in Medieval Islamicate Historiography’, Claremont Colleges Library Undergraduate Research Award, paper 2 (April 2016), Chapter 4, http://scholarship.claremont.edu/cclura_2016/2.

57 Dwight F. Reynolds, ‘Qiyān in al-Andalus’, in Concubines and Courtesans: Women and Slavery in Islamic History, ed. Matthew Gordon and Kathryn A. Hain (New York: Oxford University Press, 2017), 104.

58 See Manuela Marín, ‘Notas sobre onomástica y denominaciones femeninas en al-Andalus (siglos VIII-XI)’, in Homenaje al prof. Darío Cabanelas Rodríguez, 2 vols. (Granada: Universidad de Granada, Departamento de Estudios Semíticos, 1987), 1:37–52 and Mohamed Meouak, ‘L’onomastique des personnages d’origine ‘slave’ et ‘affranchie’ en al-Andalus à l’époque califale (IVe/Xe siècle): premières approximations documentaries’, Onoma 31, no. 1-3 (1992–1993), 17–28.

59 Dhikr bilād al-Andalus, 105-6/134-6. The complete translation of the verses is omitted because they are not relevant.

60 Dhikr bilād al-Andalus, 106/136.

61 Marín, Mujeres en al-Andalus, 301, quoting al-ʿUmarī, Masālik al-abṣār fī mamālik al-amṣār, vol. 10, ed. Ghattas ‘Abd al-Malik Khashaba and Ḥusayn Naṣṣār (Cairo: Dār al-Kutub wa-l-Wathāʾiq al-Qawmiyya, 2005), 10, 369.

62 See the complete translation of al-ʿUmarī’s text and his interpretation in Reynolds, ‘Qiyān in al-Andalus’, 105–6.

63 Reynolds, ‘Qiyān in al-Andalus’, 107.

64 Ibid., 107–8.

65 On the complex historiography of Ziryāb see Dwight F. Reynolds, ‘Al-Maqqarī’s Ziryāb: The Making of a Myth’, Middle Eastern Studies 11, no. 2 (2008), 155–68. For the entry of Ziryāb in al-Andalus, see page 158.

66 Ibn Ḥayyān, Muqtabis II-1, 149r.

67 Ibid.; al-Maqqarī, Nafḥ al-ṭīb, ed. Iḥsān ʿAbbās, 8 vols. (Beirut: Dār Ṣādir, 1968), 3:125.

68 Reynolds, ‘Qiyān in al-Andalus’, 108–9.

69 Ibid., 104, 106.

70 Regarding the consolidation of the Mālikī school in al-Andalus during this period, see Anderson, ‘The Relationship of the Amīr al-Ḥakam I’, especially chapter 4, on ‘The Early History of the Mālikī School in al-Andalus’, 63–83. See also Alfonso Carmona, ‘The Introduction of Mālik’s Teachings in al-Andalus’, in The Islamic School of Law: Evolution, Devolution and Progress, ed. Peri Bearman, Rudolph Peters and Frank E. Vogel (Cambridge, MA: Harvard University Press, 2005), 41–56; and Maribel Fierro, ‘Proto-Malikis, Malikis, and Reformed Malikis in al-Andalus’, in The Islamic School of Law: Evolution, Devolution and Progress, ed. Peri Bearman, Rudolph Peters and Frank E. Vogel (Cambridge, MA: Harvard University Press, 2005), 57–70, esp. 62.

71 Ibn Ḥayyān, Muqtabis II-1, 87v-88r; al-Maqqarī, Nafḥ al-ṭīb, 1:3401; Ibn Saʿīd al-Maghribī, al-Mughrib, 1:39.

72 Ibn Ḥayyān, Muqtabis II-1, 124v.

73 Ibn Ḥayyān, Muqtabis II-1, 129r. Note that at the beginning Ibn Ḥayyān calls al-Ḥakam I ‘caliph’ (khalīfa) and not amīr.

74 De la Puente, ‘Eunuchs in the Emirate of al-Andalus’, 183.

75 For the Arabization and Islamization of al-Andalus, see, for example, Manuela Marín, ed., The Formation of al-Andalus, part 1: History and Society (London and New York: Routledge, 2016).

76 Ibn al-Qūṭiyya, Taʾrīkh iftitāḥ al-Andalus, 72.