The rule of empires : those who built them, those who endured them, and why they always fall (15 page)

son-in-law, blamed Mu’awiya for displacing the Prophet’s family.

In their eyes, the Umayyads were secular kings rather than righteously guided successors to the Prophet. Their sins included moving

the political seat of Islam from Mecca to Damascus and aspiring to

become emperors.8

Ruling from 661 to 750, the Umayyads indeed played the central

role in transforming the caliphate into an empire. Just as the Roman

emperor Claudius sought legitimacy through empire building, Caliph

Abd al-Malik launched a new phase of imperial expansion that added

Sind, Transoxiana, North Africa, and Spain to the caliphate. Lacking

extensive administrative experience, the Umayyads relied on local

bureaucrats who had served the Byzantine and Sassanid empires

to manage their conquered territories. Their central bureaucracy in

Damascus was initially quite rudimentary. Originating as the keepers

of the payroll (
diwan
) that shared out plunder to the Arab soldiery,

the fi rst caliphal offi cials expanded this military record-keeping system into simple departments for correspondence and tax collection.

Dhimmi
and clients writing in Greek and Persian staffed the original

bureaus, but Arabic-speakers took over in the 690s under Caliph Abd

al-Malik. His successors in the eighth century further expanded the

imperial bureaucracy to include more formal departments for security, taxation, military pay, and the caliphal household.

In terms of sovereignty, the Umayyad Caliphate was theoretically

a unifi ed imperial state under the absolute authority of the caliph.

Yet the limits of medieval travel and communication meant that the

Umayyad caliphs’ direct authority rarely reached beyond Syria. Initially, there was no standardized system of provinces on the Roman

model, and the Arabs drew their territorial boundaries around their

original
amsar
, or garrison cities. This created de facto viceroyalties encompassing Syria, Jazira (the territory between the Tigris and

Euphrates rivers in northern Iraq), Kufa and Basra in southern Iraq,

Khurasan, and Egypt and North Africa. The caliphs usually ruled

Syria and Jazira directly and appointed governors to control the

Muslim

Spain 75

rest of the empire. Each viceroyalty consisted of provinces (
wilayat
)

divided into districts and subdistricts.

The viceroyal capital cities were also recruiting bases for the Arab

regiments (
ajnad
; sing.
jund
) who settled within their borders. The

tribal organization of the original Arab armies thus shaped settlement

and governance. When necessary, governor-commanders divided each

jund
into “quarters” and “fi fths” to create new administrative units.

This tribally based system of civil and military administration

evolved as the caliphate expanded into North Africa and Central

Asia. By the time of Abd al-Malik, the Umayyad military consisted

of the viceroyal fi eld armies of Syria, Jazira, southern Iraq (Kufa and

Basra), Khurasan, and Egypt. The southern Iraqi army carried out the

caliphate’s early eastern conquests, but disputes over the division of

plunder led the caliphs to demobilize it in the late seventh century.

The Jazira force almost exclusively fought the Byzantines, while the

relatively small Egyptian army operated in North Africa. But the real

seat of Umayyad political and military power was in Syria. Divided

into fi ve separate military districts, the province fi elded a force of

approximately 175,000 men who underpinned Umayyad rule and

fought as expeditionary forces throughout the caliphate.

The Umayyad soldiery’s regular wages came from provincial taxation, but pay rates were not uniform. The privileged Syrians earned

three times more than troops in the discredited Iraqi fi eld army,

and non-Arab
mawali
clients earned even less. But inclusion on the

rolls of a victorious Umayyad army was much more important than

wages, for participation in a successful campaign entitled even the

lowest-ranking soldier to a share of the plunder. When Muhammad

ibn al-Qasim conquered Sind in 712 he duly turned over one-fi fth of

the loot to the caliphate. This amounted to 120 million dirhams, the

equivalent of the entire land tax of Iraq. Simple calculation reveals

that the general’s entire haul in Sind was six hundred million dirhams. Further raids into India by his successors produced another

four hundred million dirhams. These warrior-administrators became

fabulously wealthy, but lower-ranking offi cers and common soldiers also received a generous share that far surpassed their regular

wages.

Predictably, confl icts over the division of imperial spoils became a

central point of tension in the caliphate. Maneuvering to monopolize

76 THE RULE OF EMPIRES

the wealth of empire, the Umayyad military split into coalitions claiming descent from the tribes of northern (Qays/Mudar) and southern

(Yaman) Arabia. The Yamani colonists who settled in southern Syria

got the lion’s share because they made up the core of the Umayyad

armies. Although these supratribal factions acquired real political

meaning, they were essentially genealogical inventions that emerged

as commanders modifi ed or fabricated tribal identities to reorganize

their armies. Persian units that defected en masse from the Sassanid

Empire became a fi ctive Qays subtribe, and succeeding groups of converts similarly formed subordinate components of the larger tribal

regiments. In this way, the Qays and Yaman factions became the primary medium through which Arab soldiers and their clients staked

their claims to wealth, prestige, and political power.

From a military standpoint, it clearly paid to become a Muslim. As

with Roman tribal auxiliaries, military service offered a path to citizenship and a chance to share in the imperial spoils. This explains why

Berber converts to Islam were a signifi cant portion of the Umayyad

forces that conquered North Africa. Abandoning their initial resistance, they joined the Umayyads in despoiling the Byzantine settled

regions. After that, the Berber general Tariq’s conquest of Iberia was

the next logical step.

It was relatively easy to turn conquered peoples into expendable

shock troops, but mass popular conversions were problematic. Seeking

to shore up the boundaries of subjecthood, the Umayyads manipulated the tenets of early Islam to fudge the distinction between citizen and subject. Islam was the caliphate’s de facto state religion, but

it took more than a century for it to develop into a text-based faith

with an explicitly proselytizing obligation. The Umayyads claimed to

be God’s secular deputies, but they did not initially push their subjects to convert. Religious men were still debating what exactly it

meant to be a Muslim during the fi rst century of Islam, and a clerical

class (the
ulama
) and a fully codifi ed religious law (the
shari’a
) did

not emerge until around the turn of the eighth century. Still, empire

building forced scholars and intellectuals to defi ne Islam more precisely in a futile attempt to ensure that the much larger and more

culturally sophisticated non-Muslim subject majority did not absorb

and assimilate Arab soldiers and colonists. This codifi cation of Islam

facilitated imperial domination and extraction, and after the reign of

Muslim

Spain 77

Abd al-Malik the Umayyad caliphs began to style themselves more

explicitly as Islamic rulers. They also became less willing to tolerate

public expressions of non-Islamic religions and cultures.

Yet the Umayyads still shied away from undertaking a systematic

effort to convert non-Muslims. Uncertainty over how to incorporate

outsiders into the
umma
and the desire to exploit imperial subjects

were powerful counterweights to Muhammad’s injunction to spread

the faith. As a result, the rates of conversion under the Umayyads were

generally low. Muslims were less than 10 percent of the Persian population when the Abbasids seized the caliphate in 750, and most Egyptians remained Coptic Christians until well into the ninth century.

Nevertheless, caliphal imperial rule created real incentives for certain groups to embrace Islam. Excepting Byzantine and Sassanid prisoners of war, most early converts in Iraq and Iran were peasants and

tenant farmers, for, at least in theory, conversion brought a signifi cant

tax reduction. Islamic law required all Muslims to pay the
zakat
, a

religious tax assessed at differing rates based on property and wealth.

Non-Muslims, in contrast, paid a land tax (
kharaj
) and a head tax

(
jizya
). Taken together, these obligations were considerably heavier

than the
zakat
and became a major source of funding for the caliphal

state. Seeking relief from tax collectors and obligations to local landlords, the Umayyads’ subjects had an incentive to convert to Islam.

Anxious to protect an important imperial revenue stream,

Umayyad administrators often refused to recognize the conversion

of their taxpaying non-Muslim subjects. Iraqi offi cials sent wouldbe peasant converts back to the countryside, and the governor of

Khurasan used circumcision, a practice required under Islam, to determine if new Muslims were sincere. Taxation became more equitable

as Muslims became liable for the
kharaj
when they acquired land

owned by
dhimmi
and pagans, but the Umayyads’ need to tax their

subjects at higher rates perverted their obligation to treat converts as

equals. Ultimately, military service was the only sure way for common men to become Muslims in good standing, thereby going from

taxpayers to tax recipients.

Converts usually became clients (
mawali
) of powerful Arab

patrons. Under the Umayyads only Arabs were “true” Muslims, and

mawali
adopted the tribal identities of their sponsors. Even so, it was

diffi cult to escape the stigma of their non-Arab origins, and most

78 THE RULE OF EMPIRES

still paid the non-Muslim poll tax. Initially, most new Muslims were

slaves, peasants, tenants, and foot soldiers. In the west, commanders

converted entire communities of Berbers by promising them equal

status within the Arab armies, but they used them as shock troops and

denied them full shares of the spoils. The
mawali
responded to this

institutionalized discrimination by inventing fi ctional Arab genealogies and insisting that a sincere profession of faith was all that it took

to be a true Muslim. Although Islam legitimized Arab empire building, its inherent inclusiveness complicated the business of imperial

rule. Given a choice, caliphal offi cials probably would have preferred

not to convert their subjects, for they were much easier to govern and

exploit while they remained non-Muslims.

Consequently, the Umayyads did not meddle too deeply in the

daily lives of the subject populations. Like the Roman emperors, they

lacked the means to rule directly at the local level, which meant that

their demands on their non-Muslim subjects were no greater than

those of the Byzantines or Sassanids. Although Islamic law obligated

Muslim conquerors to either convert pagan “idolaters” or put them

to death, the Umayyads generally left these communities alone if

they acknowledged Muslim sovereignty and paid their taxes. Furthermore, Islamic law explicitly protected Jews and Christians from

forced conversion because they shared a spiritual heritage with Islam

as “peoples of the book.”

Although these rules seem reasonably tolerant, there is no disguising the imperial nature of the caliphal rule. Jews and Christians

could not testify against Muslims in court, nor could they marry

Muslim women. The Umayyads’ increased reliance on Islam for their

legitimacy in the early eighth century made the status of
dhimmi

even more precarious. In 717, the famously pious Caliph Umar II

issued decrees banning Christians from dressing like Muslims and

limiting their right to maintain churches and display religious symbols. Conversion offered the tempting promise of social advancement,

but even the
mawali
could not fully escape subjecthood.

The Umayyad Caliphate was also as dependent on slave labor as

any other ancient or medieval empire. Arab armies took captives by

the tens, if not hundreds of thousands, and victorious generals treated

these prisoners as booty to be shared out among their troops. The

conquest of Sind netted almost seven hundred thousand saleable

Muslim

Spain 79

captives, and commanders in North Africa often paid their Berber auxiliaries in slaves rather than treasure. Islamic law generally

granted slaves greater legal protection and social status than Roman

slaves, and Muhammad praised owners who manumitted captives as

an act of piety. But plenty of Muslim masters still treated slaves as an

exploitable and disposable resource. Indeed, the ninth-century rebellion of desperate African slaves (the
zanj
) in the marshes of Basra

continued the long tradition of mass slave uprisings against uncaring

imperial masters. Conversion to Islam did offer captives more rights