In the middle was what can be described as a civic centre. The first building to be erected was the mosque, which sat in the middle of an open square. A mighty archer was called upon to stand at the centre and fire arrows in each direction: people were permitted to build their houses only beyond the places where the arrows had fallen. The interior of the square was left empty for people to meet.
The mosque itself seems to have been roughly square in shape, about 110 metres in each direction.
84
In its earliest phase it is said to have had no walls at the sides and a partial covering at one end. It was probably constructed very simply in reeds or mud brick. Sitting in the interior, you could look out and see the neighbouring Christian monastery of Hind and, further in the distance, the gate that led to the bridge of boats across the river.
85
Shortly after its construction, the treasury in the governor’s palace was robbed and Sa
c
d made the decision to bring the mosque right up to the palace so that they shared a common wall. The fact that the mosque was frequented day and night was felt to be the best protection against theft. This new mosque may have been rather more substantial. At one end there was a roofed area about 100 metres long, ‘whose ceiling resembled the ceilings in Byzantine churches’, by which he presumably meant open beams supported by columns of marble.
86
The columns are said to have come from Christian churches.
87
It was not until Ziyād’s governorate, in the time of the first Umayyad caliph Mu
c
āwiya, that the mosque was walled in. New pillars, 15 metres in height, were made of stone from Ahvaz, fixed together with lead centres and iron clamps.
If the mosque was simplicity itself, the palace was a more complex building, and it became the subject of a vigorous dispute. Sayf, as preserved in Tabarī, tells the story.
88
According to him, the citadel was built for Sa
c
d by a Persian from Hamadhan called Rūzbih b. Buzurgmihr, and it was made of fired bricks taken from an old palace of the pre-Islamic kings of Hīra. Because the palace lay in the centre of the city, in which there was a great deal of noise and commotion, Sa
c
d had constructed a wooden door with a lock on it. When the caliph Umar heard about this, he sent a man to burn the door down, abusing Sa
c
d for putting a barrier between himself and the ordinary Muslims, preventing them from entering any time they wished. The story is part of a polemical literature against rulers who attempted to separate themselves or put themselves above the rank-and-file believers. The story that Sa
c
d’s palace was made of reused bricks may well be true, however.
89
The primitive mosque of Kūfa lay on the site of the modern mosque of the town. This was the place where the caliph Alī was assassinated in 661, and it has long been a place of veneration to the Shia, so no archaeological excavation has been possible. The palace, however, was excavated in the 1950s and 1960s. Three main building phases were detected, all superimposed, an early one, an Umayyad one and an early Abbasid one. By the ninth century the building was essentially abandoned and occupied by squatters. The first phase was demolished to its foundations when the second Umayyad building was constructed. All that remains are outside walls with square bastions projecting at regular intervals. Was this the foundation of Sa
c
d’s palace, as the excavator thought, or the building constructed by Ziyād a generation later at the beginning of the Umayyad period, as the main historian of the city believed? It is impossible to tell.
We can, however, be certain that within a generation of the foundation of the city, it had acquired two public buildings, the mosque and the palace, which shared a common wall. In this way the classic central architectural layout of the Islamic city had been established, a layout that had no direct parallel in pre-Islamic architecture and which was to persist for centuries to come. To this official complex, a third element was added, the markets.
90
It is clear that Kūfa was provided with souks from the very beginning: after all, the victorious Arab troops had to spent the dirhams they had been given as booty somewhere. At a fairly early stage they were also being paid salaries, and these too they would have spent on both necessities and luxuries. It was the noise from the markets which is said to have induced Sa
c
d to strengthen the walls and gates of his palace. We know nothing, however, of the shape or form of the early souks except that they came to occupy the open spaces around the mosque and palace. They do not seem to have been built structures until the late Umayyad period, a century after the foundation of the city. Before this, they were probably flimsy shelters, built of wood and reeds and roofed with mats. Nonetheless, the presence of the souks, in the heart of the town, surrounding mosque and palace, set the fundamental pattern for subsequent Islamic urbanism.
The Muslims operating in southern Iraq also founded a city on the margins of the desert at Basra. The accounts of the early settlement of Basra are very confused, though the Khuzistān Chronicle clearly ascribes it to Abū Mūsā al-Ash
c
ari, commander of the forces that conquered his homeland. It was also much smaller than Kūfa, perhaps only 1,000 men, as the army in the south was much smaller.
91
The site of the first city of Basra is now known as Zubayr and lies about 20 kilometres from the centre of the modern city. It was some distance from the river bank and canals were required to bring water to it. Although the location of the site is well known and much of it is open semi-desert, there have been no published excavations and no serious survey. If conditions were more peaceful than they are as I write, it would present a wonderful opportunity for students of early Islamic urbanism to explore the archaeology of this early military settlement.
It was in these new cities that early Islamic fiscal administration developed most precociously.
92
The inhabitants lived off the receipts of taxation, paid in cash as salaries (
at
). At first this was supple mented by payments in kind, grain, oil and other foodstuffs (
rizq
), but this was gradually phased out and replaced by money. The names of those entitled to payments were entered in registers known as
dīwān
s. The administration of this system was very complex. In Basra, for example, there are said to have been 80,000 men by the end of Mu
c
a-wiya’s caliphate in 680, each of whom was entitled to at least 200 dirhams per year. This required the collection and payment of 16 million dirhams, a massive task demanding skilled workers. The Muslims were forced to employ accountants and officials who had worked for the defeated Sasanians, and they brought with them the old Persian traditions of financial administration and bureaucratic practice.
Both the new towns, Kūfa and Basra, played an immensely important part in the history of the early Muslim world, first as military bases from which armies set out for the conquest of Iran and the east and then as cultural centres. Kūfa was also politically important, a major centre of resistance to the Umayyad caliphs of Damascus and the centre of the movement of support for the family of the Prophet which was to develop into Shiism. The foundation of Baghdad, only a few kilometres to the north, in 762 dealt a fatal blow to the prosperity of the city. By the ninth century it was in full decline and only the status of the ancient mosque as a place of pilgrimage kept the city alive. Basra in contrast was far enough away to escape the gravitational pull of Baghdad and remained the major port at the head of the Gulf. Although the centre of the city has shifted, Abū Mūsā al-Ash
c
ari’s foundation has survived the centuries and is now the second-largest city of Iraq.
At about the same time, a force from Kūfa was marching up the Tigris towards the Jazira, accepting the surrender of towns and villages along the river banks and in the surrounding plains. When they came to the site where the city of Mosul now stands they found a castle, some Christian churches with a few houses near by and a settlement of Jews. Almost immediately after this small community was conquered, the Arabs set about developing a new town on the site, the origins of the modern city of Mosul. Plots for house building were distributed to the Arabs and the city grew rapidly to become one of the main urban centres of Iraq.
93
The absolute chronology of events is very difficult to ascertain, but we can be reasonably confident that by the end of 640 Muslim forces had taken control of the irrigated lands of Iraq from Tikrit in the north to the Gulf in the south and as far east as the foothills of the Zagros mountains. Muslim settlement remained very patchy and was largely concentrated in the newly founded garrison cities at Kūfa, Basra and, on a smaller scale, Mosul. There was a garrison holding the old Persian capital at Ctesiphon and there were probably others of which we know nothing. The numbers of the conquerors were very small to subdue and hold this large and populous territory. The 20,000 adult males who first settled in Kūfa were surrounded by a population in the surrounding countryside which is thought to number half a million men.
94
Although the number of Arabs was swelled by new immigrants, they were always a very small minority and cannot, in the first generation, have comprised more than 10 per cent of the total. Their problems would have been compounded by the nature of the terrain, criss-crossed as it was with irrigation ditches and canals. It would certainly not have been possible to conquer and hold the land if the Muslims had been faced with determined popular resistance. In the event, however, the only serious resistance came from the Persian royal army. For reasons that are not entirely clear, this army failed repeatedly to hold its own against the Arab forces. In field battles at Qādisiya and Jalūlā, and cities like Ctesiphon at Tustar, the Sasanian forces were decisively defeated. With the collapse of the Persian army, the Arabs were prepared to make fairly easy terms with the rest of the population - they did not massacre townspeople and villagers, they did not seize their houses or their lands, they did not interfere with their religions and customs, they did not even settle among them. They demanded only that taxes be paid and that the people did not aid their enemies. Whether the taxes were higher or lower than they had been under the previous administration we cannot tell, but we can be certain that most people in Iraq thought that it was a bargain well worth making.
4
THE CONQUEST OF EGYPT
The conquests of Syria and Iraq had followed on naturally from the conquest of Arabia. In Syria, and to a lesser extent in Iraq, there were already Arabs, both settled and nomad, either to be incorporated into the Muslim armies or subdued. It was logical, even unavoidable, to move on from there to conquer, as the Muslim armies did, the non-Arab peoples of the area.
Egypt was very different.
1
In the modern world we think of Egypt as an Arab country, in many ways a political and cultural centre of the Arab world. At the beginning of the seventh century, however, this was not the case at all. There seems to have been no substantial Arab settlement, no Arab tribes roamed the deserts and few Arab merchants did business in the towns. The earliest Muslims certainly knew of it but seem to have had few contacts there.
The story of the conquest is elaborated in the Arabic sources with a mass of confusing detail.
2
Egypt in the eighth and ninth centuries produced its own school of history-writing that was completely separate from the Iraqi tradition on which we depend for the history of the conquests of the Fertile Crescent and Iran. The great Baghdad-based historian Tabarī, who devotes hundreds of pages to collecting the stories of the conquests of Syria, Iraq and Iran, dismisses the conquest of Egypt in fewer than twenty.
3
A strong local tradition of history-writing developed early in Egypt, however. Stories about the Muslim conquest of the country were collected and written down by a historian called Ibn Abd al-Hakam (
c
. 805-71) in the mid ninth century.
4
He came from an Arab family whose ancestors had come over with the conquest, and he sought to record and preserve the memory of the great deeds of that time. He wrote at a time when the old Arab aristocracy of Egypt was being replaced as the ruling elite by Turkish soldiers brought in from the east, and his accounts are tinged with nostalgia for the days when his family, and families like them, had ruled the land. He derived his information from a variety of works, now lost, which were composed in Egypt in the eighth and early ninth centuries
5
and were probably themselves based on local oral tradition and reflect a real early Islamic social memory about the conquests. It is useful to consider these texts as a separate body of literature and I shall refer to this material as the Egyptian-Arab writing.
At the same time, the Muslim conquest is recorded in a contemporary Christian chronicle written by John, Bishop of Nikiu, a small city on the western margins of the Delta.
6
John was a near-contemporary of the events he describes, so his account is a reflection of attitudes of the time. He also provides us with some clear dates, which help to anchor the swirling confusion of the Arabic narratives in a chronological framework. The chronicle is not, however, without its problems. The Coptic original is long since lost and survives only in a single manuscript translation into Ge’ez (the ancient and liturgical language of the Ethiopian Church), made in the twelfth century. The translation is clearly confused in places and it is hard to know how accurately it reflects the original. There are also gaps at crucial points, such as the surrender of the fortress at Babylon. John does, however, give a reasonably coherent narrative and provides a useful check on the Egyptian-Arabic tradition.