The Great Arab Conquests

In the same summer a small merchant ship was making its way up the west coast of Asia Minor. As it passed through the narrow and often stormy channel that nowadays separates the Greek islands of Kos and Kalymnos from the Turkish mainland, it struck an underwater reef by the little outcrop known as Yassi Adi (‘Flat Island’).¹ Whether because the crew did not know of the reef or because the small boat was trying to shelter from the fierce Meltemi winds, the ship sank in 30 metres of water. She must have gone down quickly because they did not have time to remove the gold and copper coins they had put in a locker for safe-keeping or the kitchen utensils from the galley. Unless they were strong swimmers who could make the 50 metres to the shore, they perished with their vessel.

The Yassi Adi wreck is of key importance in our understanding of Mediterranean shipping at the end of antiquity. From 1961 to 1964 it was the subject of a major underwater excavation which recovered a vast amount of information about the ship and its cargo. It was not a large boat, just under 21 metres long with a capacity of about 60 tons. It was a cargo ship, laden with some 900 large amphorae, which were probably filled with wine. The sailors intended to travel in some comfort, for there was an elaborate, tiled galley towards the stern, well equipped with cooking utensils and fine tableware.

It has been speculated that the ill-fated vessel belonged to a church and was being used to transport supplies to the Byzantine army, but the truth is that we do not know who was sailing it or why. The ship, dating as it does from the years immediately before the beginning of the Muslim conquests, tells us much about the coastal trade of the eastern Mediterranean in the last years of antiquity. The waters through which it sailed were stormy and dangerous, to be sure, but they were largely free from piracy and hostile attack, as they had been for the long centuries when the waters of the Mediterranean were the Byzantine ‘Mare Nostrum’. Within two decades all that was to change and the peaceful waters of the Levant were to become the theatre for a fierce and destructive naval confrontation.²

There was a tradition of seafaring among the Arabs. In pre-Islamic times, Arabs did put to sea and the Koran (30: 46) tells the faithful that God sent the winds ‘so that the ship may sail at His command and so that you many seek of His bounty’, and that ‘It is He who makes the ship sail on the sea so that you may seek of His bounty’ (17: 66). These and other references make it clear that some Arabs at least were used to making trading voyages.³ There was also a tradition of distrust of the sea among the early Muslims. The caliph Umar in particular is said to have been deeply suspicious of the sea, holding it to be a danger for the Muslims. This caution was short lived. One of the most astonishing aspects of the early Muslim conquests was the speed with which the Muslims, or more exactly fleets under Muslim command, were able to challenge the well-established naval power of the Byzantine Empire. In part this was forced on them by the need to defend the coasts of Syria and Egypt against raids by the Byzantine navy, which retained its capacity to mount seaborne assaults on the coastal towns throughout the first three Islamic centuries. If the Byzantines were allowed unchallenged command of the sea, no one along the coasts of Syria, Palestine or Egypt could be considered safe.

The Muslims soon began to see the possibility of using ships for offensive purposes as well. The island of Cyprus, lying as it does only 100 kilometres from the coast of Syria, was an obvious target.⁴ In 649 the governor of Syria, Mu^cāwiya, later to be the first Umayyad caliph, sent a naval expedition against the island. Interestingly the date of the invasion is confirmed by a Greek inscription commemorating the restoration of a basilica at Soli, which had been damaged by the raid, by Bishop John in 655.⁵ This is an almost unique contemporary reference to destruction and rebuilding at the time of the first Muslim conquests.

According to the tradition preserved in the Muslim sources,⁶ Umar had refused to allow Mu^cāwiya to venture on the sea, but his successor, Uthmān, gave his permission with the curious provision that Mu^cāwiya should take his wife with him, presumably to encourage him not to take unnecessary risks. He, and a number of other prominent Muslims, were duly accompanied by their women. After this first successful raid, the people of Cyprus were obliged to pay an annual tribute to the Muslims. They already paid a tribute to the Byzantines, so the island came under a sort of joint rule, both sides receiving some money but neither maintaining a permanent garrison. In 654 Mu^cāwiya invaded again because, the Muslims claimed, the Cypriots had offered ships to help the Byzantines against them, so breaking the terms of the treaty. The Muslim fleet is said to have consisted of 500 ships and carried a force of 12,000 regular soldiers (that is, men whose names were entered in the dīwān). At that time Mu^cāwiya is reported to have erected mosques and built a new city on the island in which he settled men from Ba^calbak as a garrison and gave them salaries. This Muslim outpost lasted until Mu^cāwiya’s son Yazīd withdrew the men and demolished the city, presumably because he did not consider that it was worth the expense of paying the garrison.

Throughout the late seventh, eighth and ninth centuries, Cyprus enjoyed a unique position between the Muslim and Christian worlds. It was not always easy. The Muslim jurists were unhappy about a treaty that seemed not to conform to Islamic law in many ways. From a military point of view, too, there was always suspicion that the Cypriots were aiding the Byzantines. The Umayyad caliph Walīd II deported many of the Cypriots to Syria because he suspected them of aiding the Byzantines, but they were allowed to return by his successor Yazīd III. Problems continued under the Abbasids, and in 806, during the reign of Hārūn al-Rashīd, the people of the island are said to have caused disturbances and an expedition was launched to bring them into line; 16,000 prisoners are said to have been taken to Raqqa, Hārūn’s base in northern Syria, where they were ransomed or sold as slaves - one Cypriot bishop fetched 2,000 dinars.⁷ Despite these setbacks, Greek Christian culture survived in Cyprus when it had effectively disappeared from the nearby mainland. In the second Council of Nicaea (in the Byzantine Empire) held in 787, the bishops of the churches under Muslim rule were unable to attend, but no less than five bishops came from Cyprus, showing that contacts with the Byzantine world were still close.

The first raid on Cyprus was followed by other attacks on Mediterranean islands, Rhodes and Kos being pillaged, probably in 654.⁸ Up to this point the Muslims had not directly engaged the Byzantine navy, which still commanded the seas of the eastern Mediterranean. The first real naval engagement between the Muslims and the Byzantines was the so-called Battle of the Masts (Dht al-sawārī) or battle of Phoenix off the Lycian coast in 655.⁹ Descriptions in Ibn Abd al-Hakam, the Greek Chronicle of Theophanes and the later Arabic chronicle of Ibn al-Athīr¹⁰ mean that we have more information about this encounter than any other naval engagement of the period. According to the Arabic sources, the campaign began when Emperor Constans II (641-68) assembled a naval expedition to oppose the Muslim conquest of North Africa. He set out with a fleet of 500 or 600 ships, ‘and more men than the Byzantines had ever collected since the coming of Islam’. Mu^cāwiya sent Ibn Abī Sarh, governor of Egypt, who was also ‘in charge of the sea’^r to intercept them. The two navies met off the Lycian coast. The wind was against the Muslims when they first saw the Byzantines, but then it dropped and the two fleets anchored. The two sides agreed to a truce for the night; the Muslims read the Koran and prayed and the Byzantines rang their bells (nawqīs). The next morning the two fleets closed together and the Muslims grappled with the Byzantines. The fighting was with swords and daggers and many men on both sides were killed. In the end God favoured the Muslims, the emperor was wounded and fled the scene and only a few Byzantines escaped with their lives. Ibn Abī Sarh remained at the site for a few days and then returned to Syria.

The longest account of the battle we have is given by Ibn Abd al-Hakam, who used sources from Egypt, presumably collected there because many of the men in the Arab fleet came from Egypt and returned there. The account is largely formulaic, however, and a disappointingly large amount of space is given to discussion of who married whose daughter after the event and other matters of little use to the naval historian. According to what can be gleaned from this account, the sea battle was part of a combined operation and half the ships’ crews (shihna) were on land at the time. The Byzantines had 1,000 ships compared with the Muslims’ 200. The commander, Ibn Abī Sarh, held a council of war at which one of the speakers said in an encouraging way that a small group could win over a much larger one if God supported them. With Muslim morale thus bolstered, the two fleets approached each other and the fighting began with bows and arrows (nabl wa nushāb). The emperor¹¹ sent messages to find out how the fighting was going. When he heard that they were fighting with bows and arrows, he said that the Byzantines would win; when he next heard that they were hurling stones, he again said that the Byzantines were winning; but when he heard that the boats had been tied together and the men were fighting with swords, he predicted that the Arabs would be victorious.

Theophanes’ Greek account gives a somewhat different background. According to him, Mu^cāwiya was preparing a fleet for an attack on Constantinople. While the fleet was being prepared in Tripoli (Lebanon) two ‘Christ-loving brothers, the sons of Bucinator [the Trumpeter]’, broke into the prison in Tripoli and released a large number of Byzantine captives there. They then sacked the town and killed the governor, before escaping to Byzantine territory. Mu^cāwiya, however, was not deterred and the fleet, commanded by one Abū’l-Awar, duly set out. The emperor Constans joined battle at Phoenix in Lycia woefully ill prepared. The sea was soon full of Byzantine blood and the emperor threw off his imperial robes to make his escape undetected. He was saved only by one of the sons of Bucinator, who rescued him from the water and was killed in his place.

The most rudimentary rules of naval warfare were grossly neglected by both parties, partly because of the Byzantines’ underestimation of their enemy. The two fleets faced each other the whole night before their engagement without any plan. No projectiles were thrown between each other, either with arrows or stones launched from special machines. No ram was used by any ships of either party. Since boarding practice required great skill, the Arabs found an easier solution; they managed to tie the ships to those of the enemy and thus they changed naval warfare into land warfare ... None of the parties took into consideration the wind.¹²

The sources are really too thin to know whether this castigation is justified. It is clear, however, that the Muslim navy remained generally inferior to the Byzantine forces. This was especially apparent during the attack on Constantinople, which began in 674.¹³ The Muslims understood from the beginning that it was impossible to take the city without first dominating the waters around it. A large Arab fleet, commanded by the caliph Mu^cāwiya’s son and eventual successor Yazīd, entered the Sea of Marmara. For four years it blockaded the city all summer and then retired to Cyzicus, on the south side of the sea, for the winter. Despite this relentless pressure, the defence held firm. The Byzantines were helped by the deployment, apparently for the first time, of the celebrated ‘Greek fire’, invented by one Callinicus, a refugee from Arab-held Ba^calbak in Syria. Greek fire was a combination of crude oil and other substances to make it adhere to wood. It was lit and propelled from a siphon at enemy ships. Given that the Byzantines received the formula from a native of Ba^calbak, however, it is certainly not impossible that the technology originated in the Middle East. There is indeed some evidence (see poem below) that the Muslims had the fire during this first siege of the city.

For behold just as you, Lord of All, saved your city from the crashing waves of the filthy and most evil Arabs, you stole away fear of them and the trembling and their returning shadows ...

Where now, O cursed ones are your shining bright ranks of arrows? Where now the melodious chords of the bow strings? Where is the glitter of your swords and spears, your breast plates and head-borne helmets, scimitars and darkened shields?

Where are the twin decked, fire throwing ships, and again, the single decked ships, also swift in the battle step?

What do you say, miserable and voracious Ishmael? Christ was mighty in the work of salvation and He rules as God and Lord. He gives strength and supports the battle. He shatters the bow and grinds down human power . . . Therefore, Ocean, you who displayed the murderers broken to pieces, applaud the Lord! And Earth who has shown forth and applauded the God of all, raise a chorus of hymns to whom honour and glory and power are proper through the unceasing aeons of aeons and long years.¹⁴

The second major naval expedition against Constantinople took place in the years 716-18. Once again the Greek chronicler Theophanes is our main witness since the Arabic sources are very brief. According to the Greek monk, the conflict began with a struggle over the timber resources so vital to shipbuilding. The Byzantines became aware that the Arabs of Egypt were going on an expedition to Lebanon to collect timber. The emperor Artemios decided to intercept them and collected swift sailing ships to do so. The Byzantine fleet assembled at Rhodes under the command of a deacon of the great church of Hagia Sofia called John, who was also minister of finance. Their orders were to raid Lebanon and burn the timber. The expedition did not work out as planned. As so often in the Byzantine Empire in this period, there was a mutiny, the imperial commander was murdered and the troops set out for the capital to overthrow Artemios, leaving the Arabs free to carry on their shipbuilding.

In 716 a massive land army commanded by Maslama b. Abd al-Malik set out to march to Constantinople. At the same time a fleet was collected. Its main function seems to have been to support and supply the land army with which Maslama was attacking the city. The winter of 716-17 was spent on the Cilician coast. In the spring the ships sailed west, then north. They anchored at Abydos on the Hellespont before entering the Sea of Marmara. On 15 August Maslama began to lay siege to the city and on 1 September a huge fleet, said to have comprised 1,800 ships, dropped anchor below the walls of the city, some by the suburbs on the Asian side of the Bosporus, others on the European coast north of the Golden Horn. Theophanes says that the Arab ships were useless because they were weighed down by their cargo. The weather was fine and they pushed on up the Bosporus. This was a big mistake. The emperor Leo III, observing and directing operations from the Acropolis, sent fire ships among Arab vessels, which turned them into blazing wrecks: ‘Some of them still burning smashed into the sea wall, while others sank in the deep, men and all and others still, flaming furiously, went as far as the islands of Oxeia and Plateia [the modern Princes’ Islands in the Sea of Marmara].’ The citizens were greatly cheered by this while the attackers shivered in terror, ‘recognising how strong the liquid fire was’. Some Arab ships survived the conflagration and the emperor tried to lure them into the Golden Horn by lowering the chain that stretched across between the city and Galata. The Arab commanders feared that if they went in, the chain would be raised and they would be completely trapped. Instead they went on up the Bosporus, where they wintered in a bay on the European coast where the great Ottoman fortress of Rumeli Hissar now stands.

The winter was very hard. Snow lay on the ground for a hundred days, and the Muslim forces on land suffered terribly from hunger and cold. The next spring reinforcements arrived, 400 food-carrying merchantmen from Egypt commanded by Sufyān, followed by 260 merchantmen from North Africa with both arms and supplies. Both commanders had now heard of the dangers of Greek fire and, rather than approaching close to the city walls, they kept their ships well hidden out of harm’s way on the Asiatic shore of the Sea of Marmara.

Many of the sailors in both the Muslim fleets were Coptic Christians from Egypt and at least some of them decided that their real loyalties lay with their fellow Christians in the Byzantine Empire. One night they took the light boats from the merchant ships and went to the city, proclaiming their allegiance to the emperor. They told the emperor about the fleets hidden along the southern shores of the sea and he prepared the fire-carrying siphons and put them on board warships and ‘two-storeyed ships’. ‘Thanks to the help of God,’ wrote the pious chronicler, ‘through His wholly immaculate Mother’s intercession, the enemy was sunk on the spot. The goods and supplies from the Arab fleets were seized.’

while their expedition was on the way back, a furious storm fell on them: it came from God at the intercession of His Mother. God drowned some of them by Prokonessos [an island in the Sea of Marmara famous in antiquity for its marble quarries] and others on Apostrophoi and other promontories. Those who were left had got through the Aegean Sea when God’s fearful wrath attacked them; a fiery shower descended on them, making the sea’s water foam up [this may have been connected with the earthquake in Syria at this time]. Once their caulking pitch was gone, the ships went to the bottom, men and all. Only ten survived to tell us and the Arabs the magnitude of what God had done to them.¹⁵ ¹⁶

With the foundation of Tunis, the Arabs began to develop a naval base in North Africa. The foundation of the city was probably begun by the governor Hassān in about 700 in the immediate aftermath of the fall of Carthage. The reason for choosing the new site, rather than simply using the Byzantine harbour of Carthage, is not clear. It may be that the earlier harbour had silted up or was becoming unusable for other reasons, but it is most likely that the attraction of Tunis was that it was not on the open sea, vulnerable to Byzantine naval attacks, but on a lagoon that was then connected to the sea by a short canal. This made it much easier to fortify. The city throve as the main naval base in Africa, though the centre of government remained at inland Qayrawān.

It was shortly after this that the Muslims made their first conquests in the Mediterranean islands with the taking of Pantelleria, probably in 700. A few years later, probably in 703, a large Egyptian fleet under the command of Atā b. Rāfi arrived in North Africa.¹⁸ It was already autumn and storms were to be expected. The governor, Mūsā b. Nusayr, warned against undertaking a campaign that year but Atā had his eye on the potential booty that the islands could offer and was not prepared to wait. They decided to raid Sardinia. All went well until the return journey. When they had almost reached their home port of Tunis, a sudden storm struck and most of the fleet was wrecked. On the nearby shore the governor’s son Abd al-Azīz collected the corpses of the drowned and the remains of their ships and cargoes. The surviving ships and their crews took shelter in Tunis where Mūsā looked after them. Perhaps as a result of the charity he showed these men, they were to form the basis of the naval force with which Mūsā invaded Spain nine years later.

This maritime disaster has left an interesting echo in the Egyptian papyri. Among a number of letters from the Arab governor to the pagarch (local landowner and official) of Aphrodito in Upper Egypt is one in which the governor enquired what had happened to the sailors, probably all Copts, from the town who had joined the fleet. With a fairly heavy-handed bureaucratic inquisitiveness, he wants to know how many have returned home and how many have stayed in the Maghreb.¹⁹ He also wants more details of those who have not returned, of who has died and why some have remained in Africa. We have only the governor’s letter, not the pagarch’s reply, but the papyrus letter shows two points very clearly: how closely the fleet was supervised by the governor and how even Aphrodito, some 500 kilometres from the sea, was obliged to send men for the expedition.

After the foundation of the arsenal at Tunis, the fleet of North Africa was essentially independent of the Muslim fleets in the eastern Mediterranean and was under the command of the local governor. It was essentially a band of corsairs, independent sailors operating in effect as pirates, raiding the islands and vulnerable coastlines of the central Mediterranean for booty and slaves. As we have seen, the North African fleet could provide 360 armed ships to aid the Muslims in the attack on Constantinople in 718. Sometimes the corsairs encountered naval opposition. In 733 they were caught off Sicily by a Byzantine flotilla which used Greek fire to burn many of the Arab ships²⁰ and the next year another group encountered Byzantine ships and lost its stock of prisoners. In 740 a much larger-scale campaign was undertaken. This time the objective was the capital of Byzantine Sicily at Syracuse, and the Arabs brought horses with them on campaign. This might have marked the real beginning of the Arab conquest of Sicily, except that the next year, 741, saw the massive Berber revolt in North Africa against Arab tax gatherers and slavers. The Arabs were temporarily driven out of much of North Africa and were certainly in no position to launch any offensive raids.

Naval organization was centred on arsenals. The English word, which comes from the Italian, is ultimately derived from the Arabic Dar al-Sinūa or House of Manufacturing. It is a term that was already in use in the ninth century, if not before, to describe the naval bases used by the Muslim fleets. The first naval bases were in Syria and Egypt. The earliest one in Syria seems to have been at Acre, but it was moved to Tyre by the caliph Hishām (723-41) because the local landowner in Acre refused to sell the required property to the caliph: no question of compulsory purchase here. In Tyre he built a hotel (funduq) presumably to house the workers, and a granary ²¹ (mustaghal). At about this time the Anglo-Saxon St Willibald visited Tyre twice in the course of the pilgrimage to the Holy Land in 724-6, and it was from Tyre that he took ship on his way home. He recorded with glee how he was able to take some of the precious and holy balsam of Jericho through the Arab customs by disguising it in a flagon of mineral oil. He also noted that the port was in a security zone and anyone visiting without permission would be arrested. ²² We have several descriptions of Tyre from Arab geographers in the ninth and tenth centuries. One geographer describes it as ‘the chief of the coastal cities, housing the arsenal. From here the government ships sail on expeditions against the Greeks. It is beautiful and well-fortified’.²³ Another writes: ‘Tyre is a fortified city on the sea and one enters through one gate only, over a bridge, and the sea lies all around, the rest of it is enclosed by three walls which rise straight out of the sea. The ships enter every night and then a chain is drawn across ... there are workmen there, each with his own speciality.’²⁴

Acre is a fortified city on the sea-coast ... the defences of which were greatly strengthened after Ibn Tūlūn visited it. He had already seen the fortifications of Tyre where the harbour was protected by an encircling wall and he wanted to fortify Acre on similar lines. Engineers [sunū^c ] were brought from all over the province but when the plan was described to them they all responded that no one could lay foundations under water. Then someone mentioned my grandfather Abu Bakr the architect [bin’] and said that if it were possible to do such a thing, he was the man who could undertake it. So Ibn Tūlūn ordered his governor of Jerusalem to send my grandfather to him. When he arrived they asked his opinion ‘No problem,’ he replied. ‘Bring big strong sycamore beams!’ They were floated on the surface of the water as you would for a castle built on the land and tied together. A big gate was left on the west [sea] side. He then raised a structure of stone and cement [shayyid] on them strengthening it by inserting great columns every five courses [dawmis]. The beams began to sink under the weight. As soon as they rested on the sandy bottom of the harbour, he stopped building for a year to allow the structure to settle. Finally he connected these defences to the old walls of the city and built a bridge across the entrance to the port. Whenever there were ships in the harbour, a chain was stretched across the entrance as at Tyre. Before this was done, the enemy [the Byzantines] used to do serious damage to the ships collected there. My grandfather is said to have been given one thousand dinars besides robes of honour, horses and other gifts as his reward and his name was inscribed over the work.

In about 780 another naval base was established at Tarsus in Cilicia. Tarsus had been an important Byzantine city and the original home of St Paul. It seems to have been ruined and deserted in the immediate aftermath of the Muslim conquests when it was in the no man’s land between Byzantine and Arab territory. The caliph Hārūn al-Rashīd ordered that it should be fortified and it became a centre for volunteers from all over the Muslim world who came to join in the jihād against the Byzantines. The ships were probably moored in the estuary of the river which connected Tarsus to the sea, and there is no record of any built harbour. In 900 the then caliph ordered that all the ships should be burned, apparently because he was told that the inhabitants were of doubtful loyalty. ‘About fifty ships, on which large sums of money had been spent and which could never be replaced at that time, were destroyed. The loss endangered the Muslims, lessened their power and increased that of the Greeks who were now safe from attack by sea.’²⁶ Despite this pessimistic assessment, Tarsus soon recovered its role because in 904 the Muslim ships raided along the Mediterranean coast of Anatolia to Antalya. The city was taken by force, about five thousand prisoners were taken and four thousand Muslim prisoners of war released. Sixty Byzantine ships were taken and loaded with booty, including gold, silver, goods and slaves. Every Muslim who took part in this raid received about a thousand dinars. The Muslims rejoiced at the news.²⁷ At a time when the Byzantine army was increasingly effective against Muslim overland raids, this sort of booty must have made naval warfare look very attractive.

Naval bases were established in Egypt very soon after the Muslim conquest and, as we have already seen, Coptic sailors were in action in the Sea of Marmara and in North Africa at the beginning of the eighth century. As on the Syrian coast, the naval bases in Egypt were developed in Byzantine ports. The most famous of these was of course Alexandria. This certainly remained a port in the years after the Muslim conquest. The pilgrim Arculf arrived there after a voyage of forty days from Jaffa in Palestine. He found a city so large that it took a day to walk across, surrounded by walls and towers. He also describes the ancient lighthouse, the Pharos, as still being in operation.²⁸ Unfortunately Arabic sources tell us almost nothing about the city and its port. We know that an Arab garrison was maintained there but there is no mention of naval forces.²⁹ The other important base on the Mediterranean coast was Faramā. But again the sources have little to say about it. There were also bases at Rosetta and Damietta. A letter written on papyrus and dated to 710 contains orders for supplies to be sent to Damietta ‘for the raiding fleet’, but our fullest information about the city comes from an account of a Byzantine raid in the early summer of 853. It was the time of the feast that marked the end of Ramadan and the governor of Egypt had incautiously ordered the local garrison to go to the capital at Fustāt to join the celebrations. While they were gone a Byzantine fleet of a hundred shalandiya vessels, each carrying between fifty and a hundred men, attacked. They burned the Friday mosque and the churches. They took furnishings, candy (qand) and flax, which were waiting to be transported to Iraq. They also found military and naval equipment, 1,000 lances on their way to the Arab forces fighting in Crete, and they burned the storehouse containing ships’ sails. Some six hundred women, both Muslim and Copt, were taken captive and many more women and children drowned as they tried to escape across the shallow lake. The marauders then moved on towards the island city of Tinnis but found the lake was too shallow for their heavily laden ships. They had to content themselves with sacking the little town of Ushtum, which had recently been fortified with a wall and iron gates on the caliph’s orders. Here they found and burned an arsenal of siege engines, both manjanīq and arrādat. Then, unmolested by any Muslim forces on land or sea, they returned home. We hear of fortified towns and military and naval equipment but there do not seem to have been any Muslim ships in the area to defend it.

The island of Roda in the Nile at Fustāt was a major centre of shipbuilding and in the early Arab sources the island is simply called ‘Jazirat al-Sina^ca’ or the Island of the Arsenal. This seems to have been established after a Byzantine raid on the Egyptian coastal town of Burullus in 673, presumably because the site, well upriver from the coast, would allow ships to be built and repaired safe from any raider. Papyrus documents of 709 shows the governor demanding that carpenters and other tradesmen be sent to the superintendent of the arsenal at Fustāt to help in the construction of ships.³⁰

Further indications of what went on in an early Muslim arsenal can be found in a form letter of appointment from the caliph (unnamed) to the (also unnamed) governor of a frontier area, recorded in a tenth-century source.³¹ Like most such documents, much of it is taken up with general exhortations and common sense. It begins with a whole series of pious commands to obey God, favour good people over bad and so on, but it does give some orders directly connected with ports and ships. The governor is urged to spend money to keep the ships and their equipment in good order and to bring the ships up out of the water in the winter. He should send out spies and keep himself well informed. He should not allow any Greek fire experts (naffatīn), sailors, throwers of projectiles (qadhdhfīn) or any other tradesmen into the ships unless they are properly qualified and capable of working well. Only the best troops were to be employed. He is to inspect the shipbuilding yards and make sure that there are adequate supplies of wood, iron, flax, pitch (zift) and other things so that the ships are properly built and well caulked and supplied with oars and sails (qulu ū ^c ). Reliable and experienced sailors are to be selected. Merchants are to be watched in case they are spies. He should also keep an eye on the harbours to make sure that no ships go in or out without his knowledge. Everything in the dockyards should be kept clean and well maintained, ready for action. He should check that there are adequate supplies of oil (naft), balsam and ropes, all in good order.

WARSHIPS ³²

Both the Arabs and their Byzantine opponents drew on a common legacy of ship design. The great triremes and quinquiremes of the Hellenistic and early Roman period had long since disappeared from the waters of the Mediterranean to be replaced by small, lighter galleys. No wrecks of warships from this period have been identified so we are dependent on scanty references in literary sources and a small number of inadequate drawings and graffiti to reconstruct what the warships of the period may have looked like. A great deal remains uncertain. The nature of the source material, both textual and in visual representations, means that we know slightly more about Byzantine ships in the early Middle Ages than about Arab ones, but there is little evidence that the warships used by each side differed in any significant way.

The standard Byzantine warship of the period was called the dromon or chelandion and the Arabs adopted the same types, calling them shinī or shalandi. Merchant ships in this period relied exclusively on wind power but warships were propelled by oars, using sails only when cruising in suitable weather or as a supplementary power source. Oars were essential to provide speed and manoeuvrability during combat. It has been estimated that an average dromon would have been about 30 metres in length and, given a 1:8 ratio of beam to length, a breadth of between 3 and 4 metres. Muslim ships were probably similar. The largest dromon crew known from Byzantine sources was 230 oarsman and 70 marines on one ship, but most probably carried between one and two hundred men.

The early Middle Ages saw a number of important changes in the way in which warships were designed and built.³³ The first was the change in hull construction. In the ancient world, hulls had been built using planks laid edge to edge and held together by pegged mortise-and-tenon joints. As reconstructed from the preserved wood, the Yassa Adi ship of 626 was built in the modern way, using a frame of ribs to which the planks are then attached; it made for a lighter, more economical but less robust style of vessel. We do not know whether the navies took advantage of the new techniques of hull construction that we find in the Yassa Adi ship, but they probably did, because these were cheaper and lighter. The second was the change from underwater rams to above-water spurs at the bows of the ships. Classical ships had used underwater rams as an important weapon in naval warfare, but these had been phased out by late antiquity and the lighter hull constructions would have been strained by a direct impact.³⁴ The third innovation was a change in the shape and rigging of sails. Late Roman ships had used square sails rigged across the beam of the ship, but at some unknown time in the early Middle Ages these came to be replaced by triangular lateen sails, which made tacking close to the wind easier. Arab ships seem to have used lateen sails from the start. Another characteristic development of the period was the use of wooden, deck-top ‘castles’ to give a height advantage to the marines when fighting at close quarters. In late antiquity ships had been steered with two large oars at the stern, and this seems to have continued until the tenth or eleventh century, when such steering oars were replaced by a single stern rudder.

In many ways naval warfare was little more than land warfare fought on ships. Byzantine treatises on naval warfare do suggest arranging the fleet in crescent formations with the commander and the strongest ships in the centre. One of them also suggests that if battle is joined off the enemy coast, it is better to be near the shore so that their sailors will be tempted to abandon ship and swim for it! Beyond these there seem to have been few guides for the tactical deployment of ships. Battle was usually begun with the throwing of projectiles, arrows, stones and inflammable materials. In addition to siphons for the Greek fire, usually mounted in the bows, ships would carry catapults for propelling stones and pots of Greek fire. One of the more fanciful ideas was to hurl containers of scorpions or vipers on to the decks of enemy ships, an idea that may seem more attractive in theory than it does in the practical circumstances of fighting from ship to ship.³⁵ The main weapons were bows and cross-bows and in the end naval battles, like the Battle of the Masts, were probably decided by hand-to-hand fighting between soldiers, much as on land.

The crews were made up of two elements, the oarsmen and sailors on one hand and the soldiers or marines on the other. The evidence suggests that in Byzantine ships the two groups were not entirely separate and that sailors could also become fighters if required. In early Muslim navies, by contrast, there seems to have been a fairly strict distinction between the soldiers, who were Arab Muslims, and the seamen, who were Coptic or Syrian Christians. Such distinctions must have become irrelevant by the ninth and tenth centuries, especially in corsair ships.

Rowing in galleys, especially war galleys belonging to an alien ruling class, can never have been a popular career option, but the letters suggest that, although service was in theory compulsory if you were on the list, you did at least get paid for it. These were not galley slaves as had been used in ancient Rome. Furthermore, it is clear that sometimes, but by no means always, it was possible to make a money payment instead of doing the service in person. One papyrus even contains a requisition for cushions and it has been suggested, perhaps over-optimistically, that these were for the rowers’ benches.³⁷ We have already noted how Qurra wrote to find out the fate of those men from Aphrodito who had joined the unsuccessful raiding fleet of Atā b. Rāfi. Some had died, others had returned home but some had remained in Africa, and the governor wanted to know why. Was it possible that service in the navy offered at least some men an opportunity to escape from the restrictions of village life and make a new start for themselves?

If the navy needed people, it also needed materials for shipbuilding. Again the landowners of Aphrodito were called upon to help out. Timber was clearly the most important of these. Some timber came from the ancient forests in the Lebanese mountains but Egypt itself produced some good wood. There was the lebbek tree, of which it was said that ‘if two pieces were firmly joined together and left in the water for a year, they would become as one’, the acacia tree, whose wood was hard as iron, and the palm tree. One letter from Qurra requires that the pagarch of Aphrodito send beams of palm and fig-tree wood for building ships ‘on the island of Babylon [Fustāt]’ to be delivered this year for building ships for next year’s raid.

As well as wood, iron for nails was required and, again, the people of Aphrodito were required to take scrap or rough iron from the government store, make it into nails and send them to the chief of shipbuilding operations in Fustāt. Egypt itself produces no iron so these must have been imports, perhaps from Spain, or perhaps reused iron from Byzantine buildings. Finally there were ropes, and it is interesting to note that the English word cable is ultimately derived from the Arabic habl, meaning rope. Egypt was well supplied with hemp for this purpose.

NAVAL ORGANIZATION

WARSHIPS ³²

THE EVIDENCE OF THE EGYPTIAN PAPYRI

NAVAL ORGANIZATION

WARSHIPS32

THE EVIDENCE OF THE EGYPTIAN PAPYRI

WARSHIPS ³²