Nicola Clarke 510

‘They are the most treacherous of people’: religious difference in Arabic accounts of
three early medieval Berber revolts1

Nicola Clarke
Newclastle University

ʿAbd al-Malik b. Ḥabīb, a jurist and historian who died in the middle of the ninth
century, concluded his account of the eighth-century Muslim conquest of his native Iberia
with an extended dialogue scene, set at the court of the Umayyad caliphate (r. 661-750) in
Damascus. The dialogue is between Mūsā b. Nuṣayr, the commander of the conquest armies,
and Sulaymān b. ʿAbd al-Malik, who had recently succeeded his brother al-Walīd as caliph.
It takes a conventional form: a series of terse questions from the caliph (“Tell me about al-
Andalus!”) are met with responses that have the ring of aphorism. Here stereotypes dwell, not
least in the comments on Berbers:

[Sulaymān] said, “Tell me about the Berbers.” [Mūsā] replied, “They are the non-
Arabs who most resemble the Arabs (hum ashbah al-ʿajam bi-al-ʿarab) [in their]
bravery, steadfastness, endurance and horsemanship, except that they are the most
treacherous of people (al-nās) – they [have] no [care for] loyalty, nor for pacts.”
(Ibn Ḥabīb, 148)2

The conquest had taken place over a century before this narrative was written. By all
accounts, Berbers – recruited during the long conquest of North Africa – had made up the
overwhelming majority of the army that crossed the sea to Iberia. Subsequent decades
brought waves of additional Arab settlers to al-Andalus (Islamic Iberia), but Berbers
remained an important demographic force in the Peninsula. Politically, however, they were
out in the cold for long stretches of Andalusī history, and culturally their position was an
ambiguous one. Indeed, in the literary sources – the chronicles and geographies of the
Andalusī past and present – we really only hear of Berbers in four contexts: as members of
the army; as mercenaries or invaders from North Africa; as semi-idolatrous ethnographic
curios over the sea in North Africa; and as rebels, often sectarian ones. Indeed, the very term
‘Berber’ is a problematic one, imposed upon a great diversity of peoples and languages by
hostile observers: an expression of contempt derived from the Latin barbarus (barbarian,
itself derived from βάρβαρος in Greek)3 and associated with the verb barbara (to babble,
speak nonsense) in Arabic. As Michael Macdonald has noted about Roman discussions of the
Arabs, often the terminology tells us as much or more about the namers than the named (279-
80, 290-7). Since alternative terms are linked to a greater or lesser degree with modern
nationalist movements, however, and are often in their own way exclusionary, I will continue
to use ‘Berber’ as an (imperfect) umbrella term, while remaining aware that I am in some
measure contributing to the Othering discourse under discussion (Hoffman 2007, 13-22; Brett
and Fentress 4-7).
It is the image of the Berber as rebel that I intend to discuss in this article. I shall
examine the role that sectarian messages played in the development of Berber revolts, with a

1
I would like to thank the audience of the ‘Heresy and Orthodoxy in Late Antiquity’ panel at the International
Medieval Congress, Leeds, UK, 16th July 2009, for their questions and comments on an earlier (and shorter)
draft of this paper, Harry Munt for his comments on the expanded version, and Hannah-lena Hageman for
advice on the Khārijites.
2
The same passage appears some 450 years later in the North African historian Ibn ʿIdhārī’s (d. c. 1307) Bayān
al-mughrib (II, 21).
3
An onomatopoeic coinage meaning ‘those whose speech is blah, blah (bar, bar)’.

eHumanista 24 (2013)
Nicola Clarke 511

particular focus on how and why non-Berber Muslim authors used the language of
heterodoxy to describe these rebels. My goal is to investigate the literary category of ‘Berber’
in al-Andalus, and whether the use and presentation of this category changed over time,
notably in works written after the fall of the Iberian Umayyad caliphate in 1031. I shall begin
by discussing the present state of research on patterns of Berber settlement in al-Andalus and
medieval understandings of heresy in Islam, followed by an outline of the literary
stereotyping that medieval Arabic writing attached to the figure of the ‘Berber’. I shall
conclude by looking in detail at presentations of three early medieval revolts.

Berber settlement patterns

Medieval and modern scholars alike have long sought to trace a connection between the
shape of Muslim settlement in al-Andalus – where settlement was concentrated, and the
forms it took – and ethnic groupings within Andalusī society. Medieval commentators, such
as the easterner al-Yaʿqūbī (d. c. 900), portray an ethnic divide, with Arabs living primarily
in towns and Berbers in more remote, mountainous areas such as Extremadura (354-5). More
recent work, where it has not simply taken the division for granted, has attempted to build
evidence for this picture, with varying degrees of success (Guichard 1977; Ṭāha 166-82;
Kennedy 16-18; Cruz Hernández 61; Fierro 2005, 201, 206-7). Pierre Guichard has argued on
the basis of Arabic texts, possible Berber toponym survivals, material remains of household
organisation, and patterns of irrigation, that the medieval Berbers lived in segmentary clan
groups in rural areas like Valencia, with little day-to-day reference to wider political or social
organisation (1998, 148-9 and 151-5; 1977, 223-5 and 249-50).
This putative ethnic-territorial divide has sometimes been portrayed as a consequence
of Arab chauvinism driving the Berbers from political and economic centres, and from more
fertile land. Arabs, and increasingly also the Hispano-Roman Muslim converts known as
muwalladūn, dominated administrative posts and the cultural patronage associated with the
Umayyad capital of Córdoba (Oliver Pérez 2001, 329-3; Guichard 1992, 690-1; Fierro 2005,
216-17; Vallvé 1985, 576). But it is also possible that Berbers had compelling reasons to
choose such lands – that they were herders rather than farmers, for example. Furthermore,
Guichard has pointed out that mountainous areas were distant from not just urban life, but
centralised control (1977, 251-3); groups might, therefore, have opted to trade comfort for
autonomy. Many of the regions most strongly associated with Berber settlement were frontier
zones; when Córdoba’s reach was weakened by political circumstance, these were the regions
where ambitious individuals and families – whether Arabs, Berbers, or muwalladūn – could
carve out autonomy for themselves, or accumulate power for a strike at the centre. In the late
ninth century, a Berber clan, the Banū Tajīt, ruled outright in Guadiana (Guichard 1977, 198);
our eastern observer al-Yaʿqūbī seems to have been aware of this possible link between
separation and power when he noted that the Berbers of Valencia in his day, as he put it, “do
not offer obedience to the Umayyads” (355). The phrasing is an important one, and I shall
return to this point below.
It should be noted that other work, however, has challenged the picture of ethnic-
territorial – and lifestyle – division. Eduardo Manzano Moreno has argued that the Berbers
were not the nomadic, rural blank slate that medieval Arabic sources often portray them as; in
fact, large sections of the Berber population of North Africa were urbanised – and Romanised
– long before the Muslim conquests (1990, 417; Rushworth 86-8; Brett and Fentress 50-80),
and many may well have been Latin speakers (Wright 2012a, 2012b). Nor did the coming of
Islam to North Africa spell the end for Roman and Byzantine cities there, as has often been
suggested (Fenwick 10-11, 14-16, 20-30). Berber as a language disappeared quite quickly in
Iberia, for the most part (Vallvé 1985, 569), but Berber-Arabic bilingualism persisted in some

eHumanista 24 (2013)
Nicola Clarke 512

areas (Glick 2005, 203-4). Andalusī biographical dictionaries show us Berber families that
produced urban official and scholars aplenty in the early period – albeit often through
clientage links with Arab families, especially the Umayyads (Fierro 2005, 204-5). Guichard’s
arguments about Valencia have yet to be borne out conclusively by archaeological
investigation (Glick 1996, 193-9; Bazzana 1992, 187-202),4 and research by Manuela Marín
has been unable to confirm the predominance of endogamy on which much of Guichard’s
formulation rests (Marín 22; Manzano Moreno 2006, 139-46).
Finally, while the so-called ḥiṣn-qarya model of land ownership – that is, a
fortification (ḥiṣn) overseeing and protecting a neighbouring agricultural village (qarya) –
may seem to indicate decentralised social organisation along tribal lines (Bazzana 1998, 222-
3 and 232-3), there is no evidence that such arrangements were the sole province of Berbers,
nor that they were truly autonomous. The ḥiṣn functioned differently in different times and
places; some date from the ninth century, but most are from the tenth to the thirteenth
centuries, and their primary purposes seem to be defence in frontier zones (Bazzana 1998,
228-9) or the protection of shared irrigation sources (Glick 2002, 116-20). There is now
increasing acceptance that the qāʾid who controlled the ḥiṣn was an agent of the central
government, not an independent ‘lord’ or clan head (Boone and Benco, 61-2).

Literary stereotyping

Whatever the reality of the Berbers’ place in Andalusī society, a certain

‘Berberphobia’ within the Arabic-language literary tradition is undeniable: the dominant
culture drew an essentialising portrait of the subordinate group. Parallels between this
situation and Roman descriptions of the Arabs were noted above; the Arabs themselves, in
turn, used their pens to belittle a whole range of non-Arabs in the early Islamic period, not
just the Berbers.5 In our texts, individual Berbers may be lionised, particularly Mūsā’s deputy
Ṭāriq b. Ziyād, who launched the conquest of Iberia, her hero of most accounts (Clarke 25,
62-4, 98-100). But as a group, Berbers were subject to considerable hostile stereotyping in
literary texts, whether Andalusī, Egyptian, or eastern. The overriding portrayal is of the
Berbers as ʿajam: non-Arabic speaking barbarians who are barely controllable, barely
civilised, and barely Muslim. Speculations on the Berbers’ origins emphasised their lack of
close kinship with the Arabs (Shatzmiller 147-9),6 or else cast them as antagonists to the
grand march of monotheistic history: the Egyptian jurist Ibn ʿAbd al-Ḥakam in the ninth
century, and the eastern travel writer al-Masʿūdī in the tenth, for example, both name the
paradigmatic barbarian Goliath (Jālūt) as the ancestor of the Berbers, with the former
claiming the Berbers migrated westwards from Palestine after Goliath’s death (Ibn ʿAbd al-
Ḥakam 170; al-Masʿūdī §§ 93, 100-1, 1106).7 Al-Ḥamdānī (d. 945) wrote that the Berbers
had received nine-tenths of the world’s violent character (Glick 2005, 208). The very name
‘Berber’ could be used as an insult (García Gómez, 10). Even the Chronicle of 754 – an
Andalusī Christian work written in Latin, fifty years after the Muslim conquest – appears to

4
Glick also notes (201-4) that any community engaging in irrigation must have contact with its neighbours up-
and downriver. Bazzana highlights cultural parallels in the material evidence, like large communal serving
dishes and the organisation of living space (e.g. large central courtyards) in Andalusī houses, but notes that there
is no sign of separation of male and female space.
5
This sparked something of a backlash in the form of the pro-non-Arab literary movement known as the
shuʿūbīya, which primarily involved Persians (Gibb; Goldziher 1966; Mottahedeh; Savant), but did also create
some ripples in al-Andalus (Monroe).
6
Often they were said to be descended from Noah’s son Ham, not Shem like the Arabs (de Felipe 380-4; Norris
33-4); the Berbers responded with their own myths of kinship with the Arabs (Shatzmiller 145, 152).
7
The latter also notes (§1167), that some believe the Berbers came from Yemen, an argument that re-emerged
in Moroccan schools in the mid-20th century (Hoffman 2007, 14).

eHumanista 24 (2013)
Nicola Clarke 513

have picked up on Arab antipathy. Describing a skirmish during the Berber revolt of 740 –
the first of many (Berber) challenges to (Arab or Arabicised) state authority in the Muslim
west – its anonymous author writes that an Arab cavalry charge “recoiled instantly due to the
colour of the Moors’ [i.e. Berbers’] skin” (§84), drawing an explicit line of racial difference
and physical repulsion between Arab and Berber Muslims.
This trend of generalised hostility to the idea of Berbers became more pronounced
after the fall of the Córdoban Umayyad regime in 1031. The Umayyad family, once rulers of
the entire Muslim empire from their capital at Damascus, had found refuge in al-Andalus
during the 750s, after a coup back east saw a rival family, the ʿAbbāsids, seize power. This
distant western province was so far from Damascus, and so apparently unremarkable in the
grand scheme of Islamic empire, that the ʿAbbāsids never expended the effort to stop the
surviving Umayyads re-establishing themselves there. There is a certain irresistible
parallelism about Umayyad fortunes in al-Andalus: they arrived, in 756, in the wake of one
Berber revolt – that of the 740s, the first revolt discussed below – and their regime was
overthrown less than three centuries later by what has often been characterised as another.
The two decades leading up to their fall (1009-31) are widely labelled by the Arabic
chronicles as a fitna, a term used to denote major upheaval or civil war, and which has
explicitly religious connotations. The first century of Islam had been shaped and scarred by a
pair of fitnas – expressions of competing ideas about religio-political authority and
belonging, in the first of which lie the roots of the Sunnī-Shīʿī split – whose rights and
wrongs chroniclers and jurists continued to theorise about, well beyond the period under
discussion. A fitna was a trauma, a division at the heart of a community (the Muslim umma)
that should not be divided. To call something a fitna, then, was both to heighten its emotional
charge, and to make a claim for its religious significance.
In the wake of 1031, to judge from the surviving literary works, many blamed the
fitna on recent Berber immigrants: that is, the North African mercenaries introduced into the
Andalusī army by the Umayyad caliph al-Ḥakam II from the 970s (Scales 68-73, 165), as
distinct – perhaps – from the earlier waves of Berber settlers in al-Andalus. A number of
writers who lived through or after this period were extremely hostile to Berbers. Ibn Ḥayyān
(d. 1076) – on whose chronicle, the Muqtabis, we are dependent for much of our history of
the caliphate – lost everything in the fitna, for which he held the Berbers responsible (Scales
3), whose bad character he was not shy about tracing back into pre-fitna history. Ibn Ḥazm
(d. 1064) likewise castigated the more recently-arrived Berbers for their lack of assimilation
to polite Andalusī – that is, Arabic and Islamic – norms (Manzano Moreno 1990, 425). Even
a post-1031 work in praise of the Berbers – the anonymous Kitāb mafākhir al-barbar –
comments that they were widely seen as “the most contemptible of peoples (akhass al-
umam)” (Tres textos, 125-6). This last work only gave the label of Berber to tenth-century
and later arrivals, leading Joaquin Vallvé to question whether the earlier ‘Berbers’ of the
literary sources were actually Berbers at all (1992, 58-9).
Andalusī geographical works discussed Berbers’ past paganism and present
heterodoxy,8 and they used eastern models of describing barbarian Others to categorise,
explain, and condemn the Berbers. Just as the Andalusīs had seen their world upended by
‘barbarian’ outsiders, so, in 1055, had the Arabic and Persian writers of the east watched a
clan of Turkic nomads become the de facto rulers (‘sulṭāns’) in Baghdād. Turks, as
mercenaries and slave soldiers of the ʿAbbāsid caliphate for several centuries, had long been
scapegoated as a disruptive force, and many literary works linked these outsiders with the
eschatological horrors of Gog and Magog (Yājūj wa-Mājūj), by postulating a shared descent
from Noah’s son Yapheth (al-Bakrī §567). Andalusīs borrowed and adapted this idea; the

8
Al-Bakrī §§1092 and 1102 (worship of idols), 1274 (a note on the presence of Khārijites, on whom see below).

eHumanista 24 (2013)
Nicola Clarke 514

geographer al-Zuhrī, for example, writing in the latter half of the twelfth century, makes the
same link, but adds the Berbers to the family (§161). As with the idea of the Berbers’ kinship
with Goliath, we see writers in Arabic expressing their discomfort with Berbers by
constructing them as not only ethnic but scriptural Others.
Post-1031 geographical writing also features anecdotes about magical talismans,
erected to ward off Berber invasion. One comes from the Andalusī geographer al-ʿUdhrī (d.
1085):

It was related that, in the vicinity of the city of Elvira, [there was] a stone image
(ṣūra) of a horseman, which children used to ride. Part of it was broken. It was
said that in the year when the ṣūra was broken, the fitna overwhelmed Elvira, and
the Berbers entered it. That was the first year of its ruin. (88)

This is a re-working of a literary trope common in medieval Arabic accounts of the

Muslim conquest of Iberia, used here to explain why the fitna came to afflict this particular
part of al-Andalus: the statue is a protective seal that keeps the land inviolate until the day
when it is destroyed, whether through human folly or fate, just like the so-called ‘House of
Locks’ in Toledo is said to have been penetrated by the Visigothic king Roderick in 711, and
so brought about the Muslim invasion and his kingdom’s fall (Clarke 33-5, 149-50).
Another notable warding statue is said to have existed at Cádiz, as part of the
lighthouse there. Earlier Andalusī writers, like Aḥmad al-Rāzī (d. 955), had followed
Classical and Late Antique authorities in attributing this to Hercules (Lévi-Provençal), but
after the 1031 fitna, the story changed to fit a new context. Yāqūt (d. 1229), an enormously
influential eastern geographer, says that the Cádiz statue was built – by a suitor competing for
a local woman’s hand – with the aim of protecting Iberia against the Berbers (IV, 290). This
legend is an adaptation, recalling one that Graeco-Roman tradition associated with Sicily
(Fragmentary II, 177);9 and it was evidently constructed with hindsight, since al-Zuhrī
reports that the statue was destroyed in 1145-6, some time before Yāqūt wrote his account.
He adds, rather more prosaically than some of his fellow scholars:

The people of al-Andalus used to think that it was a talisman related to activity in
the sea, and that if it was destroyed no-one would be able to enter the sea. But
when it was destroyed, nothing changed. (§239)

In short, Berbers were a sinister Other for many of the historians, geographers and other
writers whose literary imagination shapes our knowledge of al-Andalus: a threat to the social
and political order. And in the medieval Islamic world, society and politics were inseparable
from religion.

Heresy in Islam

Before we turn to more detailed case studies, it is necessary to discuss, briefly,

conceptions of heresy and heterodoxy in Islam. It is frequently pointed out that heresy, in the
Christian sense, is not a concept that fits easily into an Islamic context (Lewis 52). There is
no single Muslim church, universally recognised as having the authority to make
pronouncements on correct and incorrect doctrine; nor is there a Sunnī, or a Shīʿī one,
although for a time various Shīʿī sects were led by lines of imāms, spiritual (and, where

9
Alaric, we are told, could make the crossing to Sicily because of a magical statue, built to keep away both
barbarians and ‘the fires of Etna’; once destroyed, its protection was removed and Sicily was ravaged.

eHumanista 24 (2013)
Nicola Clarke 515

possible, political) authorities by virtue of their descent from the Prophet Muḥammad’s
nephew and son-in-law, ʿAlī. After the death of the Prophet, the fledgling community was
united under the religio-political authority of the caliphs, but – as noted above – competing
ideas as to who was entitled to be a caliph, and on what grounds, soon challenged that unity.
For the group that would become the Shīʿa, the community’s rightful ruler – and the
illuminator of its doctrine and laws – could only come from the line of ʿAlī (Halm 2004); for
the Khārijtes, who will be discussed further below, those who ‘went out’ from the community
during the first civil war, the leader could be anyone, if they were a strict adherent of
scriptural law, and agreed upon communally.
What eventually became known as the Sunnī conception of authority took several
centuries to develop, and resided largely in the interpretation of a body of tradition: that is, in
the Qurʾān and the sunna of the Prophet (accumulated anecdotes of his words and deeds).
This knowledge was passed down by legal scholars, orally and in writing, and used by them
to make legal judgements. Some jurists had state-sponsored roles, but the training and
affiliation of jurists lay with their peers, who over time came to be grouped into four main
legal schools. But even a legal school (madhhab) did not operate as a body, issue corporate
judgements that it could expect to be universally accepted, or have the collective authority to
determine heresy as we understand it. Islamic law was about many individual authorities,
who not infrequently disagreed; Islamic orthodoxy, likewise, was in the eye of the beholder,
and endlessly mutable, even if it was always couched in the language of tradition. This is not
to say medieval Islam was a free-for-all; medieval Muslims certainly held that there were
correct and incorrect beliefs (Melchert 85-7, 89-93). Rather, the question of who was entitled
to circumscribe and enforce correct belief belonged in multiple hands and minds.
Nonetheless, individuals and groups were targeted by authorities for holding – or,
perhaps more pertinently, displaying and proselytising for – wrong beliefs (Fierro 1987, 101-
2, 113-18). States and rulers did, usually in conjunction with particular schools or
individually influential scholars, pick a definition of orthodoxy against which to judge their
opponents, who were duly rendered religious deviants (Judd). The most famous example of
this is the miḥna (‘inquisition’) of the early ninth century, when the ʿAbbāsid caliph al-
Maʾmūn attempted to use a disagreement over the nature of the Qurʾān to force the
burgeoning scholarly elite of Baghdad to acknowledge his authority to formulate doctrine; he
failed, and thereafter Muslim rulers largely had to approach such issues in conjunction with
scholars (Cooperson 113-21; Melchert 8-16; Turner). The Umayyad rulers of al-Andalus, for
their part, forged a relationship with the Mālikī school; Mālikī jurists provided the amīrs and
caliphs with legal advice and public support, in return for official backing and preferential
treatment for the school’s jurists: the appointment of Mālikīs to the post of chief judge (qāḍī),
for example.
Medieval Islam lacked a catch-all term like ‘heresy’, although a word like zandaqa –
meaning ‘Manichaean’, initially – came to cover that ground to some degree, and its use in
al-Andalus has been explored by Maribel Fierro (1987-88, 251-8). More common were
charges whose shades of meaning are indicative of the medieval Islamic community’s
priorities and preferred discourse, such as bidʿa (impermissible innovation, i.e. a deviation
from tradition, although one generation’s innovation could be the next’s accepted practice),
or ridda (apostasy, which could mean not only an active change of religion but also the denial
of core Islamic religious duties, like paying zakat [alms tax]) (Kraemer 36-7). Most important
for present purposes is the conflation of terminology for revolt, heterodoxy, and apostasy.
Since a Muslim ruler’s authority was ordained by God, rebellion against a ruler, a threat to
the established order, was a strike against God too. This was something the Umayyads in al-
Andalus were particularly keen to emphasise – through public ceremonial, literary patronage,
and regular military campaigns – given their breakaway independence (after 756) from the

eHumanista 24 (2013)
Nicola Clarke 516

wider Islamic caliphate, whose legitimacy they rejected, and the frequency with which they
faced revolt at home (Safran).
To medieval heresiographical scholars, too, perhaps the most famous of whom was
Shahrastānī (d. 1153), sectarian division was rooted in error, and error lay chiefly in failure to
obey rightful rulers (22-7).10 Obedience, to a ruler and to one’s communal duties as a
Muslim, was a cardinal virtue; disobedience threatened the unity of the community, insulting
God and threatening the well-being and salvation of all. In many ways it was the act, rather
than wrong belief, that was the issue; a degree of difference in opinion could be tolerated,
provided that communal duties were fulfilled, as is most famously exemplified through the
protection (dhimma) and freedom of worship afforded to Christians and Jews (known as
dhimmīs), in return for payment of a state poll tax. Being a medieval Muslim was a
communal endeavour: leaving the community was a religious statement, and legal writings
on apostasy and heterodoxy focus very much on encouraging apostates to “return to
obedience” and rejoin the community (Qurʾān 49:9; Kraemer 48-9). In theory, rebellion was
only a capital crime if an attempt was made to justify the disobedience using a taʿwīl (a
reading of the Qurʾān that was excessively different from accepted norms) (EI2 X, 390-2); in
practice, however, an Andalusī amīr would condemn any rebel as an apostate.
It has been argued in the past, by scholars such as Ignaz Goldziher (1963 91-6), that
al-Andalus was somehow a uniquely conservative Muslim state, and that the Mālikī law was
a uniquely intolerant, even “fanatical” one. This is a picture shaped largely by the likes of Ibn
Ḥazm – a detractor of the Mālikīs, with whom he had clashed, as well as of the Berbers – and
has been challenged in recent years. Al-Andalus was more diverse, religiously, than used to
be appreciated (Aguadé 54-60). Nevertheless, the amīrs and caliphs clamped down on any
heterodox movement they had the resources to deal with, seizing upon such movements as
way to display their orthodox piety and ability to protect the community from rogue
elements. The Umayyads of Córdoba, after all, had a point to prove, both at home and
abroad, since they did not offer allegiance to the new, usurping ʿAbbāsid rulers in the east;
and after 929, when the Umayyad ʿAbd al-Raḥmān III (r. 912-61) dropped the subordinate
title of amīr for the universalist one of caliph, he was laying claim to religious authority over
the whole Islamic world. In doing so, he raised the stakes at home, and made a statement
abroad: rebelling against a regional amīr was one thing; rebelling against a caliph was
something else entirely.

Revolt 1: The Berber Revolt

Unrest and conflict were endemic in al-Andalus in this period, although the levels
waxed and waned according to the strength of the regime, and perhaps also due to broader
trends, such as grain prices (Marín-Guzmán 186 and 218). It was not simply a case of Arabs
against Berbers (or either against muwalladūn). There were, as discussed above, social,
political and geographical distinctions between ethnic groups, but each group was also
internally divided along political and factional lines: the Arabs into baladīs (the earliest post-
711 settlers) and Syrian junds (settlers of the 740s wave) (Manzano Moreno 1998, 86-8), for
example, or the powerful old families resisting Córdoban centralisation, as in the late ninth-
century revolt of Seville, which resulted briefly (c. 899-913) in an autonomous amirate
(Carabaza 41-3; Guichard 1992, 683-4). Among the Berbers, there was conflict between
tribal elites – many of whom relied for their power and status at least in part on their
relationship with the Umayyads – and religious leaders (Manzano Moreno 1991, 256-7).

10
Shahrastānī did not advocated that a believer should follow a ruler blindly, without reflecting on what was
being asked, however (41).

eHumanista 24 (2013)
Nicola Clarke 517

Most the revolts of the eighth and early ninth centuries centred on groups of Arabs or
Berbers, although as the ninth century wore on, muwalladūn also became involved (Guichard
1977, 306-7). These rebels were often prominent families with a regional power-base, such as
the Banū Qasī (‘the sons of Cassius’), who dominated Aragón until the ninth century
(Guichard 1974, 1508; Kennedy 69-72).
The Berber uprising of the 740s is easily the most famous of the episodes under
discussion in this article, to the extent that it is known simply as ‘the Berber revolt’ in modern
scholarship (Aguadé 64-5; Brett and Fentress 87-8; Manzano Moreno 2010, 590-3), and even
in some medieval works (Ibn al-Qūṭīya 14: “thawra al-barbar”). I shall concentrate here on
two contrasting accounts of these events. Writing the tenth century, at the height of the
Córdoban Umayyad caliphate’s confidence and power, was the grammarian Ibn al-Qūṭīya (d.
977). As his name (‘son of the Gothic woman’) indicates, Ibn al-Qūṭīya had Visigothic
ancestors, and his chronicle is unique in the way he weaves the fortunes of his Christian
forebears into the narrative of Andalusī history – particularly in terms of the special
relationship they had enjoyed with the Umayyad dynasty since the conquest (Fierro 1989;
Christys 179-81; García Moreno 311-12). The second account comes from the Akhbār
Majmūʿa (‘Miscellaneous Reports’), an anonymous compilation that merges historical
material from multiple genres, including poetry; it contains tenth-century material, but the
date of its composition probably belongs to the eleventh century or later.11
Both accounts have broadly similar outlines: the revolt begins in Tangiers, before
spreading through the Maghreb – where it is led by one Maysara – and into al-Andalus,
threatening to destabilise the entire region and provoking the caliph to send troops from
Syria; the rebels are eventually defeated, but the Syrians become embroiled in a further with
the (Arab) governor of al-Andalus, representing the interests of the baladiyya, the peninsula’s
original Muslim settlers. The most immediately obvious way in which the accounts differ is
how they are structured: Ibn al-Qūṭīya spends less than two pages on the Berber revolt, and
six on the conflict between the caliph’s forces and the Andalusīs (14-15 and 15-20,
respectively); the Akhbār Majmūʿa, more prolix in general, devotes twelve pages to the
Berbers, but only six to the fight with the Syrians (28-40 and 406).
As this might suggest, Ibn al-Qūṭīya is dismissive of the Berber revolt; Maysara was
known as ‘the contemptible one’ (al-ḥaqīr, a word containing overtones of ignominy and of
lowly origins), he says, noting that before he became a rebel, Maysara was merely a water
vendor in the souq of Qayrawān (14). But for the Akhbār Majmūʿa’s compiler – and/or for
his sources – the revolt is a more serious episode, and Maysara gets no such contemptuous
nickname. A religious cast is given to the Berbers’ unrest; they were partisans, we are told, of
“the Ibāḍīya and Ṣufrīya” (28), two sub-sects of the Khārijites.
‘Khārijite’ – derived from the verb kharaja, ‘to go out’ – is an umbrella term for an
ideology that emphasised activist piety, and required its adherents to abandon any community
or ruler that did not measure up to its standards of behaviour (Watt ch. 1; Lewinstein;
Robinson 109-24; Kenney; Sizgorich 196-230; Gaiser; Hoffman 2011). Khārijites are viewed
with a fear and suspicion in the Arabic historical and heresiographical tradition that, for the
most part, far outweigh their actual significance (Shahrastānī 118-41); they are also presented
as being perhaps more numerous than they truly were, in the sense that scholars such as
Shahrastānī labelled anyone who rebels against “the legitimate ruler (imām)” a Khārijite,
regardless of what – if any – doctrine they actually espoused (118).
Nonetheless, they could be a force to be reckoned with in the Maghrib – an Ibāḍī
polity was established in the region only a few decades after the revolt (Rushworth 88-95)12 -
11
Molina (513) notes the impossibility of pinning it down, since it contains such a variety of reports and styles;
Manzano Moreno (1992, 43), suggests that there is at least a substantial amount of tenth-century material in it.
12
The Rustamids, centred on Tāhart, from 761 onwards.

eHumanista 24 (2013)
Nicola Clarke 518

and from the angry, Berber-distrusting perspective of the eleventh century it seems that
equating Maghribi Berber unrest with what was widely perceived to be the anarchism and
violence of Khārijī ideas was irresistible. The Akhbār Majmūʿa’s tone when describing these
rebels is not casual patrician dismissal, but outrage: they massacre the population of Tangiers,
including its children (28), and Maysara’s “uncountable” force of Berbers (a dehumanising
adjective that Arabic chronicles tend to reserve for animals, or loot) faces off against an army
whom the text calls simply, and repeatedly, “the Muslims” (32-3). The Berbers have left the
community of believers, and thus placed themselves beyond the civilised pale. Their
grievances are listed, but these are rendered invalid, we are told, by the fact that the Berbers
are rebelling against their rightful rulers (aʾimma, pl. of imām), and that they have chosen the
iconography of notorious earlier rebels, including shaving their heads, to express their
discontent (32).
Whereas Ibn al-Qūṭīya makes little connection between the later unrest in al-Andalus
and the revolt in North Africa, attributing the conflict with the Syrians largely to political
opportunism on the part of the Andalusī governor – and noting that “the Arabs and Berbers of
al-Andalus” are ranged on his side against the Syrians who might steal their land (14, 17-20)
– the Akhbār Majmūʿa paints the episode as primarily an ethnic and religious conflict. In the
latter text, the Arabs of al-Andalus send supplies to the beleaguered Syrians fighting in North
Africa, while the Berbers of al-Andalus shave their heads “in imitation of Maysara” (40)
while driving the Arabs out of several regions of the peninsula (37-40). At length, God gives
the Muslims victory over the Berbers (40).
Ibn al-Qūṭīya’s narrative was one of Umayyad authority being challenged but
ultimately triumphing in al-Andalus; his interest lay in portraying the Peninsula as chaotic,
and its governors untrustworthy, prior to the arrival of direct Umayyad control there in 756,
not in targeting the Berbers as specifically disruptive. But by the time the Akhbār Majmūʿa
was being compiled, Umayyad authority was a thing of the past, and the precarious new
world was the creation of North African Berbers. Ripples of the revolt were also felt farther
afield, although these were somewhat muted by the more successful regional revolt of the
same decade, the ʿAbbāsid ‘revolution’; nonetheless, the event, and the attendant notion of a
vast Berber insurrection in the Maghrib as a possible precursor to the End Times and the
coming of the Mahdī, left traces in apocalyptic literature as far away as Egypt and Syria
(Madelung 13-14, 20-1).

Revolt 2: Shaqyā, c. 768-77

The second of my chosen revolts came close on the heels of the first (Ibn al-Qūṭīya
32; Akhbār Majmūʿa 107-11; Fatḥ al-Andalus 102; Ibn al-Athīr V, 605-6, and VI, 9, 35, 42,
49-50; Ibn ʿIdhārī II, 54-5; Fierro 1987, 28-9; Aguadé, 65; Manzano Moreno 1991, 238-48).
The sources show some confusion over the dating (although they largely agree on the
sequence of events), but it seems to have lasted the better part of a decade. This revolt was
launched by one Shaqyā b. ʿAbd al-Wāḥid,13 a Miknāsa Berber and teacher (muʿallim) from
the district of Shanṭabariyya (Santaver, roughly corresponding to the present-day districts of
Guadalajara and Cuenca). It was a region strongly associated with Berber settlement
(Manzano Moreno 1991, 142-3; Ṭāha 174-7). In the course of the uprising, Shaqyā and his
followers killed the governor of Shanṭabariyya and took the town of Coria (some two
hundred km away to the west). They then led the authorities a merry dance for several years,
disappearing into the mountains – the almost canonical refuge of Berbers, as we have seen –

13
Or Shaqnā (Fatḥ al-Andalus 102) – there is no difference between this and ‘Shaqyā’ in unpointed Arabic
script – or Sufyān (Akhbār Majmūʿa 107).

eHumanista 24 (2013)
Nicola Clarke 519

whenever an expedition was sent to deal with him (Akhbār Majmūʿa 107; Fatḥ al-Andalus
102; Ibn al-Athīr V, 605, and VI, 9, 35, 42). Eventually he was killed by what seem to have
been some of his own followers, perhaps indicating a power struggle.
Again I want to consider the language in which the accounts of Shaqyā’s revolt are
couched. On the taking of Coria, for example, the Akhbār Majmūʿa – hostile to Berbers, as
we saw in the previous section – uses the verb ghalaba (to overcome), rather than fataḥa (to
conquer), the latter being more usual when discussing the military activity of Muslims; in
other words, it is being made clear that this is not a legitimate conquest, in which God works
through the warriors, but a usurpation of proper authority (107). The same text also says that
Shaqyā “spread corruption through the land” (107), wording that undoubtedly comes from a
verse of the Qurān (5:33) that declares it is obligatory for Muslims to put a stop to such
behaviour:

Indeed, the punishment of those who fight God and His Messenger and spread
corruption in the land is to be killed, crucified, have their hands and feet cut off
on opposite sides, or to be banished from the land. That is a disgrace for them in
this life, and in the life to come theirs will be a terrible punishment.

The wording may be a trace of legalistic justification used for the campaigns against
Shaqyā, but it is also an indication of how such revolts were conceptualised, as a source of
unacceptable disorder. Shaqyā is never called a zindīq (heretic), or an apostate; his rebellion
is a khurūj, or going-out, again linking him to the Khārijites, and emphasising again the way
that a rebellion involves leaving the community. Others are more dismissive: Ibn ʿIdhārī,
probably following the lost work of the Andalusī Aḥmad al-Rāzī (d. 955),14 calls Shaqyā’s
followers a ‘rabble’ (ghawghāʾ) (II, 54), a favoured term to describe the followers of Berber
religious or rebel movements.15
Language is also key to the religious dimension of this revolt’s presentation. A
recurring detail in the accounts is that Shaqyā’s mother was named Fāṭima, like the Prophet’s
daughter, and that Shaqyā called himself al-Fāṭimī, thus claiming ʿAlid descent. In other
words, he was – or was portrayed as – a proto-Shīʿite. As is so often the case with
unsuccessful (apparently) heterodox movements, it is difficult to be sure whether ‘al-Fāṭimī’
was a genuine self-identification by Shaqyā, or a label applied by his enemies. ʿAlid lineage
offered an alternative source of legitimacy to anti-Umayyad rebels, without requiring an
appeal to the ʿAbbāsids, and someone had made similar claims only a few years before
(Fierro 1987, 29). But the fact that most of the accounts agree about Shaqyā using it – with
the exception of Ibn al-Qūṭīya, who opts for a different phrasing16 – should not necessarily be
taken as corroboration; the agreement is a function not of multiple independent witnesses to
events, but of our various chroniclers drawing upon each other’s work, and each apparently
continuing to find the term meaningful for the effect they wished to create. As Maribel Fierro
has demonstrated, ‘Fāṭimī’ was for a long time a derogatory term applied to rebels and rivals
– especially by the ʿAbbāsids – rather than something that proto-Shīʿites claimed for
themselves: a way of belittling ʿAlid claims by linking them with a woman (1996, 153-5).
Even the dynasty that historians today call the Fāṭimids (r. 909-1174) – the Córdoban
Umayyads’ great regional rivals, who did claim descent from Fāṭima – did not use the term of

14
See Clarke (ch. 2) for further discussion of Ibn ʿIdhārī’s sources.
15
Followers of another teacher’s uprising in the marches, in 850-1, were also called al-ghawghā’ (Ibn al-Athīr
VII, 66; Ibn ʿIdhārī II, 90).
16
Ibn al-Qūṭīya does not name him, saying only that “a man claiming descent from ʿAlī [b. Abī Ṭālib] rebelled
(thāra) in al-Hawwārīyīn” (32). The place name indicates a strong Berber identification with the area; Hawwāra
Berbers were among those who settled in district of Shanṭabariyya after the conquest (Ṭāha 175).

eHumanista 24 (2013)
Nicola Clarke 520

themselves until over a century after they came to power (Fierro 1996, 144-7). It is therefore
possible that the accounts’ use of the title is simply a way of mocking Shaqyā, or else
retrospectively discrediting him by association with the external enemies of Umayyad al-
Andalus. It is notable, in this regard, that Ibn ʿIdhārī also calls Shaqyā a dāʿī (‘propagandist’),
a term specific to Fāṭimid missionaries active outside their own borders – anachronistic for
Shaqyā’s own time, but certainly meaningful to later audiences, as will be discussed below.

Revolt 3: Ibn al-Qiṭṭ

This article’s final revolt, which took place in the year 901, we know largely from one
of the surviving portions of Ibn Ḥayyān, who drew on two tenth-century sources, ʿĪsā b.
Aḥmad al-Rāzī (d. 989, son of the al-Rāzī mentioned above) and al-Shabīnasī (Ibn Ḥayyān
127, 133-9; Fierro 1987, 106-11, 123-4; Aguadé 66-7 and 70; Manzano Moreno 1991, 56 and
253-5; García-Arenal 90-1). The revolt’s leader, Ibn al-Qiṭṭ, was not a Berber but an
Umayyad prince (Ibn ʿIdhārī II, 140); his army, however, was overwhelmingly Berber, to the
tune of – it is said – 60,000 people (Guichard 1992, 684). With the Córdoban government
crippled by Umayyad infighting, the reign of ʿAbd Allāh (r. 888-912) was a period when
frontier clans flourished; furthermore, with the Christian kingdom of León pushing
southwards under the ambitious Alfonso III (r. 866-910), it was a period of regular warfare.
The shifting allegiances of the muwallad rebel Ibn Ḥafṣūn – who, in aid of his many
alliances, at various times converted to virtually every religious tradition going – are the best
known example of life in this period (Ibn al-Qūṭīya 90-4 and 103-15; Akhbār Majmūʿa 150-
2). From his fortified base at Bobastro, he took advantage of the regime’s disorder,
withholding tax and generally proving a headache for the Umayyads for several decades. His
motives have been much discussed (Oliver Pérez 1993; Acién Almansa 1994, 87, 111-13;
Marín-Guzmán; Fierro 1998, 299-328), in large part because he has seemed to some scholars
to be an example of ‘nativist’ reaction to Arab-Muslim rule, owing to a conversion to
Christianity accompanied by a genealogy detailing no fewer than seven generations of his
non-Muslim ancestors (Wasserstein, 270, 288-93). But his re-conversion to Islam ten years
later – specifically, to the Fāṭimid branch of Shīʿism – suggests that his career is better
understood as an example of how religious identity and allegiance could be exploited by
opportunistic individuals, than as a proto-Reconquista fightback from within.
Ibn al-Qiṭṭ was not nearly as successful as Ibn Ḥafṣūn, but accounts of his brief career
show a similar strategy at work: building up a following through a combination of religious
appeal, propaganda, and military campaigning. Unlike Shaqyā, Ibn al-Qiṭṭ’s life as a rebel did
not begin with ready-made credibility and kinship connections among the Berbers.
Accordingly, his religious message was more overt – he preached ḥisba (correct behaviour)
and jihad (holy war) – and it had a clearly-defined target, Zamora, a former Muslim city that
had been conquered and refortified by the Christians in 893 (EI2 XII, 842). Ibn al-Qiṭṭ also
had a publicist, one Abū ʿAlī al-Sarrāj, a wandering ascetic – portrayed as riding on a donkey
and wearing sandals (Ibn Ḥayyān 135) – with a history of meddling on the frontiers; he had,
earlier in his career, been associated with the two most prominent semi-autonomous Hispano-
Roman Andalusī clans, the Banū Ḥafṣūn and the Banū Qasī (García-Arenal 90). Having been
unsuccessful in forging an alliance between these two – which would have been a serious
threat to the Umayyads – Abū ʿAlī al-Sarrāj now lent his voice to Ibn al-Qiṭṭ’s cause, selling
him to the Berbers as the Mahdī (messiah). Rising to power on the back of a charismatic
preacher and a disaffected Berber clan or two was a time-honoured tradition in the medieval

eHumanista 24 (2013)
Nicola Clarke 521

Islamic west,17 and there are strong echoes of narratives of the origins of the Fāṭimid dynasty
in North Africa: they, too, had a public face, Abū ʿAbd Allāh, who recruited the Berber army
that brought his patrons to power in the same decade as Ibn al-Qiṭṭ’s revolt (Halm 1996, 38-
43, 134-5; Brett 73-100).18 In al-Andalus, however, Ibn al-Qiṭṭ’s jihād against Zamora was
apparently going well until those untrustworthy Berbers got tired of being merely followers
and deserted their leader (Ibn Ḥayyān 135-6, 139). Ibn al-Qiṭṭ was then defeated and had his
severed head displayed on the gate of Zamora for his trouble.
Again, the language used to describe all this is instructive. Ibn Ḥayyān calls Abū ʿAlī
al-Sarrāj the “dāʿī of the fitna” (127), identifying him with both the subversive agents of the
Fāṭimids and the traumatic chaos of civil war and communal division, as discussed above.
Both he and Ibn al-Qiṭṭ are accused of “tempting” people to follow them, and the revolt is
once again a khurūj (133, 138); in case the insinuations of Shīʿī practices were not enough to
condemn the rebels, the text’s audience is asked to recall the Khārijites at the same time.
Even though he and his followers were not directly threatening the regime, or attacking
fellow Muslims, the fact that Ibn al-Qiṭṭ set himself up as an alternative source of religious
authority (by claiming to be the Mahdī) and usurped the Umayyad ruler’s role in leading a
jihād meant that he – like Shaqyā – had “gone out from obedience” and was threatening the
unity of the community. Still, Ibn Ḥayyān and his sources save some of their invective for the
Christians: Zamora is a city of “unbelief”, and appeasement of the mushrikūn (polytheists, a
common name applied to Christians) is to be avoided (134).

Conclusion

The involvement of Berbers in revolts against state authority in al-Andalus has often
been seen as a testament to the cohesion of their shared identity – although by no means all
Berbers took part in all revolts – and to their sense of frustration at being politically and
socially disenfranchised within al-Andalus. Rebelling in the name of sectarian movements
gave religious legitimacy to both their separateness and their opposition to Arab rule. But it is
important not to overlook the nature of the accounts upon which we rely for information on
these events: with a few exceptions, Berbers were outsiders to Andalusī literary production, 19
leaving the task of discussing their moral character and political activities entirely to writers
with varying degrees of disdain or hostility towards them. The fact that the Berbers became
the scapegoats for the calamities of the early eleventh century only made matters worse;
whereas a number of ninth- and tenth-century writers in Arabic had portrayed the Berbers as
untrustworthy and unruly, these writers’ embittered post-1031 successors sharpened their
pens on the rebels of prior centuries, using Qurʾānic phrases and code words borrowed from
the Fāṭimid Shīʿī enemy to create an image of the Berbers as not only disorderly and
uncivilised, but heterodox threats to the very social and religious stability of al-Andalus,
perpetual fifth-columnists who were easily swayed into carrying out campaigns of violence
by demagogues and foreign agents. For these writers, every Berber revolt foreshadowed the
end of the Golden Age of al-Andalus.

17
This applies to both Khārijites (e.g. the Rustamids of Tāhart, the Midrārids of Sijilmāsa) and ʿAlids (e.g. the
Idrīsids) (Love 177–9; Abun-Nasr 42-59).
18
He met representatives of the Kutāma tribe while in Mecca for the ḥajj, and returned with them to North
Africa to lead the conquest of Ifrīqiya (modern-day Tunisia), before unveiling the Fāṭimid imām to the troops
once victory was achieved.
19
Some of the exceptions were not overly impressed by their fellow Berbers’ lack of culture, either: Ibn Mann
Allāh al-Hawwārī (d. c. 1102), a Berber from Qayrawān, mocked Berber efforts to claim superiority over the
Arabs who had at various times so thoroughly defeated them (Norris 9).

eHumanista 24 (2013)
Nicola Clarke 522

Works Cited

Primary texts

Al-Bakrī. Kitāb al-masālik wa-al-mamālik. A.P. van Leeuwen & A. Ferre eds. Beirut: Dār al-
Gharb al-Islāmī, 1992. 2 vols.
Conquerors and Chroniclers of Early Medieval Spain. K.B. tr. Wolf. Liverpool: Liverpool
University Press, 1999.
Crónica Mozarabe de 754. Ed. and tr. J.E. Lopez Pereira. Zaragoza: Anubar, 1980.
Fatḥ al-Andalus. Ed. L. Molina. Madrid: Consejo Superior de Investigaciones Científicas,
1994.
The Fragmentary Classicising Historians of the Later Roman Empire: Eunapius,
Olympiodorus, Priscus and Malchus. Ed. and tr. R.C. Blockley. Liverpool: Cairns,
1981-3. 2 vols.
Ibn ʿAbd al-Ḥakam. Futūḥ Miṣr. Ed. C.C. Torrey. New Haven: Yale University Press, 1922.
Ibn al-Athīr. Al-Kāmil fī ʾl-taʾrīkh. Ed. C.J. Tornberg. Beirut: Dar Sādir, 1965-7. 12 vols.
Ibn Ḥabīb. Kitāb al-taʾrīkh. Ed. J. Aguadé. Madrid: CSIC, 1991.
Ibn Ḥayyān, Muqtabis III. Ed. M.M. Antuña, Paris: P. Geuthner, 1937.
Ibn ʿIdhārī, Kitāb al-bayān al-mughrib. Ed. G.S. Colin and E. Lévi-Provençal, Leiden: Brill,
1948-51. 2 vols.
al-Masʿūdī. Murūj al-Dhahab. Ed. C. Pellat. Beirut, 1966-79. 7 vols.
Shahrastānī. Muslim Sects and Divisions: the section on Muslim sects in Kitāb al-milal wa ʾl-
niḥal. Ed. and tr. A.K. Kazi and J.G. Flynn. London: Kegan Paul International, 1984.
Tres textos árabes sobre beréberes en el occidente islámico. Ed. M. Yaʿla. Madrid: Consejo
Superior de Investigaciones Científicas, 1996.
Al-ʿUdhrī. Nusūs ʿan al-Andalus min Tarṣīʿ al-akhbār wa-tanwīʿ al-āṭār. Ed. ʿA.A. al-
Ahwānī. Madrid: Matbaʿat Maʿhad al-Dīrāsāt al-Islāmīyah, 1965.
al-Yaʿqūbī. Kitāb al-buldān. Ed. M.J. de Goeje. Leiden: Brill, 1892.
Yāqūt. Muʿjām al-buldān. Beirut: Dār Sādir, 1955-57. 5 vols.
Al-Zuhrī. Kitāb al-Jaʿrafīya. Ed. M. Hadj-Sadok. Bulletin d’Études Orientales 21 (1968): 9-
310.

Secondary works

Abun-Nasr, J.M. A History of the Maghrib in the Islamic Period. Cambridge: Cambridge
University Press: 1987.
Acién Almansa, M. Entre el Feudalismo y el Islam: ʿUmar Ibn Ḥafṣūn en los historiadores,
en las fuentes y en la historia. Jaén: Universidad de Jaén, 1994.
Aguadé, J. ‘Some remarks about sectarian movements in al-Andalus’. Studia Islamica 64
(1986): 53-77.
Bazzana, A. Maisons d’al-Andalus: habitat medieval et structures du peuplement dans
l’Espagne orientale. Madrid: Casa de Velázquez, 1992.
---. ‘Le ḥiṣn: modèle d’organisation du peuplement rural dans al-Andalus’. In R.-P. Gayraud
ed. Colloque international d’archéologie islamique, IFAO, Le Caire, 3-7 février
1993. Cairo: Institut français d’archéologie orientale, 1998. 217-44.
Boone, J.L. and N.L. Benco. ‘Islamic Settlement in North Africa and the Iberian Peninsula’.
Annual Review of Anthropology 28 (1999): 51-71.
Brett M. and E. Fentress. The Berbers. Oxford: Blackwell, 1996.
Brett, M. The Rise of the Fatimids: the world of the Mediterranean and the Middle East in
the fourth century of the hijra, tenth century CE. Leiden: Brill, 2001.

eHumanista 24 (2013)
Nicola Clarke 523

Carabaza, J.M. ‘La familia de los Banū Ḥaŷŷāŷ (siglos II-VII/VIII-XII)’. Estudios
onomástico-biográficos de al-Andalus v: Familias Andalusíes. Ed. M. Marín and J.
Zanón. Madrid: Consejo Superior de Investigaciones Científicas, 1992. 39-55.
Christys, A. Christians in al-Andalus (711-1000). Richmond: Curzon, 2002.
Clarke, N. The Muslim Conquest of Iberia: medieval Arabic narratives. London: Routledge,
2012.
Cooperson, M. al-Ma’mun. Oxford: One World, 2005.
Cruz Hernández, M. ‘The social structure of al-Andalus during the Muslim occupation (711-
55), and the founding of the Umayyad monarchy’. The Formation of al-Andalus, Part
1: history and society. Ed. M. Marín. Aldershot: Ashgate, 1998. 51-83.
Encyclopedia of Islam, Second Edition. Ed. P. Bearman, Th. Bianquis, C.E. Bosworth, E. van
Donzel, and W.P. Henrichs. Leiden: Brill, 1954-2004. 12 vols.
de Felipe, H. ‘Leyendes árabes sobre el origen de los Beréberes’. Al-Qanṭara 11 (1990): 379-
96.
Fenwick, C. ‘From Africa to Ifrīqiya: settlement and society in early medieval North Africa
(650-800)’. Al-Masāq 25 (2013): 9-33.
Fierro, M. La Heterodoxia en al-Andalus durante el period omeya. Madrid: Instituto
Hispano-Arabe de Cultura, 1987.
---. (as M.I. Fierro Bello). ‘Accusations of ‘zandaqa’ in al-Andalus’. Quaderni di Studi Arabi
5/6 (1987-1988): 251-8.
---. ‘La obra histórica de Ibn al-Qūṭiyya’. Al-Qanṭara 10 (1989): 486-512.
---. ‘On al-fātimī and al-Fāṭimiyyūn’. Jerusalem Studies in Arabic and Islam 20 (1996): 130-
161.
---. ‘Four questions in connection with Ibn Ḥafṣūn’. The Formation of al-Andalus, Part 1:
history and society. Ed. M. Marín. Aldershot: Ashgate, 1998. 299-328.
---. ‘Mawālī and muwalladūn in al-Andalus (second/eighth-fourth/tenth centuries)’.
Patronate and Patronage in Early and Classical Islam. Ed. M. Bernards and J.
Nawas. Leiden: Brill, 2005. 195-245.
Gaiser, A. Muslims, Scholars, Soldiers: the origin and elaboration of the Ibāḍī Imāmate
traditions. New York: Oxford University Press, 2010.
García-Arenal, M. Messianism and puritanical reform: Mahdis of the Muslim West. Leiden:
Brill, 2006.
García Gómez, E. Andalucía contra Berbería: reedición de traducciones de Ben Ḥayyān,
Šaqundī, y Ben al-Jaṭīb, con un prólogo. Barcelona: Departamento de Lengua y
Literatura Árabes, 1976.
García Moreno, L.A. ‘Spanish Gothic consciousness among the Mozarabs in al-Andalus
(VIII-Xth centuries)’. The Visigoths: studies in culture and society. Ed. A. Ferreiro.
Leiden: Brill, 1999. 303-323.
Gibb, H.A.R. ‘The social significance of the Shuubiya’. Studies on the Civilization of Islam.
Ed. S.J. Shaw and W.R. Polk. Princeton: Princeton University Press, 1962. 62-73.
Glick, T.F. ‘Berbers in Valencia: the case of Irrigation’. Iberia and the Mediterranean World
of the Middle Ages: essays in honor of Robert I. Burns S. J. Ed. P.E. Chevedden, D.J.
Kagay and P.G. Padilla. Leiden: Brill, 1996. 191-206.
---. ‘Tribal Landscapes of Islamic Spain: history and archaeology’. Inventing Medieval
Landscapes: sense of place in Western Europe. Ed. J. Howe and M. Wolfe.
Gainsville: University Press of Florida, 2002. 113-35.
---. Islamic and Christian Spain in the Early Middle Ages, Leiden: Brill, 2005 (2nd ed.).
Goldziher, I. ‘The Spanish Arabs and Islam’. The Muslim World 53 (1963): 5-18, 91-105,
178-184, 281-6, and 54 (1964): 27-38.
---. Muslim Studies. Ed. S.M. Stern. Chicago: Aldine, 1966.

eHumanista 24 (2013)
Nicola Clarke 524

Guichard, P. Structures socials ‘orientales’ et ‘occidentales’ dans l’Espagne musulmane,

Paris: Mouton, 1977.
---. ‘The social history of Muslim Spain from the conquest to the end of the Almohad régime
(early 2nd/8th-early 7th/13th centuries)’. The Legacy of Muslim Spain. Ed. S.K. Jayyusi.
Leiden: Brill, 1992. 679-708.
---. ‘The population of the region of Valencia during the first two centuries of Muslim
domination’. The Formation of al-Andalus, Part 1: history and society. Ed. M. Marín.
Aldershot: Ashgate, 1998. 129-81.
Halm, H. The Empire of the Mahdi: the rise of the Fāṭimids. Tr. M. Bonner. Leiden: Brill,
1996.
---. Shīʿism. Edinburgh: Edinburgh University Press, 2004 (2nd ed.).
Hoffman, K.E. We Share Walls: language, land and gender in Berber Morocco. Oxford:
Blackwell, 2007.
Hoffman, V.J. The Essentials of Ibāḍī Islam. Syracuse University Press, 2011.
Judd, S. ‘Muslim persecution of heretics during the Marwānid period (64-132/684-750)’. Al-
Masāq 23 (2011): 1-14.
Kennedy, H. Muslim Spain and Portugal: a political history of al-Andalus. London:
Longman, 1996.
Kenney, J. Muslim Rebels: Kharijites and the politics of extremism in Egypt. New York:
Oxford University Press, 2001.
Kraemer, J.L. ‘Apostates, rebels and brigands’. Israel Oriental Studies 10 (1980): 34-73.
Lévi-Provençal, E. ‘La «Description de l’Espagne» d’Aḥmad al-Rāzī’. Al-Andalus 18
(1953): 51-106
Lewinstein, K. ‘Making and unmaking a sect: the heresiographers and the Ṣufriyya’. Studia
Islamica 76 (1992): 75-96.
Lewis, B. ‘Some observations on the significance of heresy in the history of Islam’. Studia
Islamica 1 (1953): 43-63.
Love, P.M. ‘The Ṣufris of Sijilmasa: toward a history of the Midrarids’. Journal of North
African Studies 15 (2010): 173–188.
Macdonald, M.C.A. ‘Arabs, Arabias and Arabic before Late Antiquity’. Topoi 16 (2009):
277-332.
Madelung, W. ‘The Sufyānī between tradition and history’. Studia Islamica 63 (1986): 5-48.
Manzano Moreno, E. ‘Beréberes de al-Andalus: los factores de una evolucion histórica’. Al-
Qanṭara 11 (1990): 397-428.
---. ‘Oriental ‘topoi’ in Andalusian historical sources’. Arabica 39 (1992): 42-58.
---. ‘The settlement and organisation of Syrian junds in al-Andalus’. The Formation of al-
Andalus, Part 1: history and society. Ed. M. Marín. Aldershot: Ashgate, 1998. 85-
114.
---. Conquistadores, emires y califas: los omeyas y la formación de al-Andalus. Barcelona:
Crítica, 2006.
---. ‘The Iberian Peninsula and North Africa’. The New Cambridge History of Islam, vol. I.
Ed. C.F. Robinson. Cambridge: Cambridge University Press, 2010. 581-621.
Marín, M. Vidas de mujeres andalusíes. Málaga: Editorial Sarriá S.L., 2006.
Marín-Guzmán, R. ‘The causes of the revolt of ʿUmar ibn Ḥafṣūn in al-Andalus (880-928): a
study in medieval Islamic social history’, Arabica 42 (1995): 180-221.
Melchert, C. Ahmad ibn Hanbal. Oxford: One World, 2006.
Molina, L. ‘Los Ajbār Maŷmūʿa y la historiografía árabe sobre el periodo omeya an al-
Andalus’. Al-Qanṭara 10 (1989): 513-42.
Monroe, J.T. The Shuʿūbiyya in al-Andalus: the risāla of Ibn García and five refutations.
London and Berkeley: University of California Press, 1970.

eHumanista 24 (2013)
Nicola Clarke 525

Mottahedeh, R.P. ‘The shuʿūbīyah controversy and the social history of early Islamic law’.
International Journal of Middle Eastern Studies 7 (1976): 161-82.
Norris, H.T. The Berbers in Arabic Literature, London: Longman, 1982.
Oliver Pérez, D. ‘Una nueva interpretación de ‘árabe’, ‘muladí’ y ‘mawla’ como voces
representivas de grupos socials’, in Proyección histórica de España en sus tres
culturas, Valladolid: Junta de Castilla y León, Consejería de Cultura y Turismo, 1993.
143-55.
---. ‘Sobre el significado de mawlā en la historia omeya de al-Andalus’. Al-Qanṭara 22
(2001): 321-44.
Robinson, C.F. Empire and Elites After the Muslim Conquest. Cambridge: Cambridge
University Press, 2000.
Rushworth, A. ‘From Arzuges to Rustamids: state formation and regional identity in the pre-
Saharan zone’. Vandals, Romans and Berbers: new perspectives on late antique North
Africa. Ed. A.H. Merrills. Aldershot: Ashgate, 2004. 77-98.
Safran, J.M. ‘Ceremony and submission: the symbolic representation and recognition of
legitimacy in tenth-century al-Andalus’. Journal of Near Eastern Studies 58 (1999):
191-201.
Savant, S. ‘Isaac as the Persians’ Ishmael: pride and the pre-Islamic past in ninth and tenth-
century Iran’. Comparative Islamic Studies 2 (2006): 5-25.
Scales, P.C. The Fall of the Caliphate of Córdoba: Berbers and Andalusis in Conflict.
Leiden: Brill, 1994.
Shatzmiller, M. ‘Le mythe d’origine berbère: aspects historiographiques et sociaux’. Revue
de l’Occident Musulman et de la Méditerranée 35 (1983): 145-56.
Sizgorich, T. Violence and Belief in Late Antiquity: Militant Devotion in Christianity and
Islam, Philadelphia: University of Pennsylvania Press, 2009.
Ṭāha, A. D. The Muslim Conquest and Settlement of North Africa and Spain, London:
Routledge, 1989.
Turner, J.P. Inquisition in Early Islam: the competition for political and religious authority in
the Abbasid Empire. London: IB Tauris, 2013.
Vallvé, J. ‘Libertad y esclavitud en el Califato de Córdoba’, in Actas de las II Jornadas de
Cultura arabe e islamica (1980), Madrid: Instituto Hispano-Árabe de Cultura, 1985.
565-78.
---. El Califato de Córdoba. Madrid: Editorial Mapfre, 1992.
Wasserstein, D.J. ‘Inventing tradition and constructing identity: the genealogy of ʿUmar ibn
Ḥafṣūn between Christianity and Islam’. Al-Qanṭara 23 (2002): 269-97.
Watt, W.M. The Formative Period of Islamic Thought. Edinburgh: Edinburgh University
Press, 1973.
Wright, R. ‘Late and Vulgar Latin in Muslim Spain: the African connection’. Latin Vulgaire -
Latin Tardif IX. Ed. Frédérique Biville et al. Lyon: Maison de l’Orient et de la
Méditerranée, 2012. 35-54.
---. ‘Linguistic and ethnic identities in the Iberian Peninsula (400 - 1000 A.D.)’. Ed. Walter
Pohl and Bernhard Zeller. Sprache und Identität im Frühen Mittelalter. Ed. W. Pohl
and B. Zeller. Vienna: Verlag der Osterreichischen Akademie der Wissenschaften,
2012. 99-109.

eHumanista 24 (2013)