V The Fatimid Khalifs of Kairawan
Led by religious enthusiasms, the Berber tribes had succeeded in sweeping away the Arab government of the province of Ifrikiya. To a very large extent, however, this was as much a racial and anti-Arab movement of the Berbers as a religious one: of course, very much the same has been true of every Mahdist movement in Africa. The history of Islam is full of similar revolts, for the most part either with a religious motive, or at least a religious pretext. Now the destructive work was finished and the Mahdi settled at Kairawan, having damped or perhaps quenched the religious fervour of his followers by the execution of Abu ʿAbdullah and the implied shelving of the miraculous powers which his earlier followers had associated with him, was faced with the task of constructing an orderly and stable principality out of what must be confessed to have been rather unpromising materials. More than once the Semitic and Berber tribes have shewn themselves quite capable of nation-building, and their work has not always been short-lived. The religious motive was effective in arousing the enthusiasm of fighting men, the task of framing political institutions demanded different qualities. At this time, no doubt, we must regard the Mahdi as primarily a political adventurer: that he had any serious regard for Shiʿite principles is incredible; that he was the missionary of an enlightened philosophy which would deliver men from the fetters of religion,—a position which may have been true of his ancestor ʿAbdullah,—is extremely improbable in his case. Unexpected circumstances had given him an exceptional opportunity as the founder of a dynasty, and we have now to see how he used this opportunity.
Towards religion the Mahdi’s attitude had been at first one of rabid Shiʿism, though he, as one of the fully initiated, could not have been sincere: no doubt he was acting up to what he expected to be the feelings of his subjects so far as he had observed the Katama and the immediate followers of Abu ʿAbdullah: closer acquaintance with the people of Kairawan showed him that he had been mistaken, the people generally were quite ready for a Mahdi, or anyone else, who could establish and maintain an orderly government, but as Muslims they were orthodox by a large majority, and by no means willing to accept the rather fantastic theories of incarnation and transmigration which appealed to the Persian mind. As soon as this was made clear the Mahdi formulated a definite policy in religion, enforcing strictly all the outward observances of Islam, rigidly punctilious in the prohibition of forbidden food and drink, and punishing severely those of the Ismaʿilian sect, who tried to practice the freedom of the higher grades of the initiated. It was no doubt possible for the initiate to disregard the rites of religion in their private life, but any external neglect, likely to cause scandal amongst the populace at large, was treated as a criminal offence: there was none of the open lawlessness of the Qarmatians tolerated in Ifrikiya: the inner grades of the sect were distinguished from other Muslims only by their reverence for the family of ʿAli, whom all revered to some extent, by their repudiation of the first three Khalifs, which was offensive to the orthodox but not intolerable, and by a few minor differences in the ritual of prayer, and in the treatment of the problems of the canon law.
The most difficult problem demanding the new ruler’s immediate attention lay in the lands to the west, for the Mahdi claimed to control all the territory to the Atlantic, over which the Aghlabid princes had pretended to rule. The first difficult task came in the revolt of Tiharet.
For long past the Berber lands of North Africa had afforded a refuge for every persecuted sect and dynasty of Islam. The earliest sect, the Kharijites, the wild men of the desert who adhered to the oldest form of purely Arab Islam, had entered Africa after they had been hunted down and slaughtered in Asia by the Umayyad Khalifs. In the days of the Mahdi they still held their own in the district of Tiharet in the mountainous country of Central Maghrab. They threw off all allegiance to the ruler at Kairawan and invited Muhammad b. Khazar to be their Emir. The Mahdi sent the Katami Aruba b. Yusuf against them: after three days siege the city was taken, plundered, and some 8,000 of the inhabitants slain.
The Umayyads who had put down the Kharijites in Asia had been compelled by the course of events to seek a refuge in Africa for themselves, and thence had passed over to Spain which was regarded as the remotest of the western parts. At this time they were ruling at Cordova (they did not assume the title of Khalif until A.H. 317), and held also some possessions in Africa about Oran. The same Karmati leader who had taken Tiharet was able to seize Oran.
The Idrisid dynasty, descendants of ʿAli by Hasan, expelled from Madina in 169, had founded a state in the remoter part of Morocco where they were still ruling. This state also was attacked by Aruba and reduced, so that all the western lands to the Atlantic coast was brought under the control of the Mahdi (Ibn Khald. i. 244-5, 267-8, etc.).
This course of consolidation of the most loosely held part of the Muslim world speaks well for the organising ability of the general Aruba, and established the Mahdi’s authority upon a sound foundation. It was, however, disturbed by domestic difficulties in the capital. Kairawan was an Arab colony, but under the Mahdi the Berbers were in the ascendant, and racial disputes were inevitable. One day a Katama tribesman treated a city merchant with insolence; a riot ensued, and some 1,000 of the Katama were slain. After this had been repressed the governor rode through the city and ordered the dead bodies of the Berbers to be removed. The workmen who carried out this order threw the bodies into the channel which served as the city sewer. At this the Katama tribesmen removed from the city in indignation, and declared that they would no longer submit to the Mahdi’s rule, and chose a youth named Kadu as their emir. Very soon this rebel was in possession of the whole province of Zab, and the Mahdi sent several generals against him without result. Some of these generals, indeed, deserted to the enemy, for the Berbers were the main fighting force in Africa, and there was a general indignation amongst them at the way in which the Katama rebels in Kairawan had been treated, and there were many followers of Abu ʿAbdullah still who threw in their lot with the revolted Berbers. At length ʿUbayd Allah sent his son Abu l-Qasim, and he, with some difficulty, managed to reduce the tribesmen.
In 300 the colony of Tripoli revolted. There, as in Kairawan, there had been riots between the Berbers and Arabs. When Abu l-Qasim returned from punishing the tribes he advanced to attack Tripoli, whilst the Mahdi at the same time sent a fleet against it, and after some delay it was reduced. Then Sicily revolted, and this proved to be a permanent loss to the Fatimid Khalifs. At first the Sicilians invited Ahmad, a son of Ziadat Allah, the former emir of Kairawan, to take charge. He refused, but after some time, as the invitation was repeated, he consented to be recognised as emir of Sicily. As soon as he was established he sent a letter to the Khalif of Baghdad professing loyalty and asking to be confirmed as emir by the Khalif. Thus Sicily broke away from the Fatimid dominions and became once more a part of the empire of the ʿAbbasid Khalif.
In 301 the Mahdi founded a new city on the coast near Kairawan, and gave to it the name of al-Mahadiya. The site was very badly chosen, and the place afterwards decayed completely, although it served as the Fatimid capital for some generations. At the same time he commenced building a fleet, by the help of which he hoped to make an attack upon Egypt in due course; no doubt he was by this time convinced that his kingdom in North Africa was not likely to be a stable one, just as it had been held precariously by the Arab rulers who preceded him: in fact it was an unsettled and savage country, which could be under control only so long as under actual military occupation. Probably, also, he hoped that the prospect of conquering Egypt would attach the Berbers to him more successfully. The weak point in these plans was that the building and manning of a fleet depended almost entirely on what Greek help he could hire. Soon afterwards he sent his general Khubasa eastwards and extended his authority, somewhat precariously, to Barqa. In the summer of 302 he made his first attempt against Egypt, sending forces by land under his son Abu l-Qasim, and Khubasa against Alexandria. The inhabitants of that city were obliged to take refuge in the ships in the harbour, whilst the invaders plundered their houses. The invading army then passed southwards to the Fayyum, but here they were met by an Egyptian army strongly reinforced from Baghdad, and compelled to retire. The effort, however, had brought the invasion of Egypt within the sphere of practical politics, and the plunder of Alexandria raised much enthusiasm amongst the Mahdi’s followers. At that time the ʿAbbasid Khalifate was in its decline: in Baghdad the government was in the hands of the military guard, the commander of that guard was the real ruler, the Khalif being no more than a figure head liable to be deposed and replaced at the will of the soldiery. The provinces were semi-independent, in most cases ruled by hereditary emirs who paid no more than a formal tribute of respect to the Khalif; indeed, in many cases it meant simply that his name was mentioned in the Friday prayer. Of all the provinces Egypt was, perhaps, the worst administered, and the ripest for falling away from the ʿAbbasid dominions. It was on the verge of disintegration by natural decay, whilst the Fatimid state which coveted it, though outwardly strong and efficient, had already showed that it had the seeds of internal weakness in the tribal jealousies of Berbers and Arabs.
In 307 the Mahdi’s armies made another attempt on Egypt, this time supported by a fleet of 85 ships, which passed along the coast from al-Mahadiya and anchored in the harbour of Alexandria. The Khalif’s officers at Baghdad could only get together 25 ships which were assembled at Tarsus and sailed over to Alexandria. But those twenty-five ships were manned by experienced Greek mariners, and inflicted a decisive defeat on the Mahdi’s fleet.
As Egypt now enters very directly into the affairs of the Fatimids, it will be necessary to consider its condition. For the last four years it had been governed by the Emir Dhuka ar-Rumi, i.e., Ducas the Roman (or Greek). Before the defeat of the Mahdi’s fleet Dhuka resolved to check the invaders who had followed their former route to the Fayyum, and were laying waste and plundering at will. He had great difficulty in inducing the Egyptian army to move at all, but at last marched out to Giza and encamped on the same side of the Nile as the Mahdi’s army. Soon afterwards he died, and the governorship was taken over by Tekin al-Khassa, who had been governor before from 298 to 303 and had been associated with the former victory over the Shiʿites. Immensely popular with the soldiery, his resumption of office made an immediate change, and he was able to take the offensive and inflict a serious check upon the invaders, about the same time as the naval victory at Alexandria. Although the Fayyum was cleared the Fatimid forces were still in control in Upper Egypt, whither their cavalry had pressed on whilst others stayed in the Fayyum. There the extreme narrowness of the Nile valley and the exposed condition of the Bahariya and the other oases always meant a minimum of defence, and the invaders were able to hold their own until the next year. That meant that the whole area was infested by bands of light cavalry, rapidly moving Bedwin, both Berber and Arab, always able to retreat at will into the neighbouring desert and very difficult to be restrained by any ordinary military force. In our own dealings with the Sanusi in 1916 we had experience of such difficulty. The only possible solution is a system of organised military patrol, which takes some little time to dispose efficiently. That Egypt was cleared after a few months’ interval shows that Tekin had considerable ability in handling the military task with which he was confronted.
These attacks of the Shiʿites revealed another weakness in Egypt. There was strong reason to suspect that they had many sympathisers there. There was an active branch of the Ismaʿilian propaganda at work in the country, and all who were initiated in the sect were of necessity spies and helpers of the invaders. Two at least of the leading officials, the Qadi and the Treasurer were in correspondence with the Mahdi and in his employ: this does not mean that they were converts to the Shiʿite sect, but simply that they were disloyal to their own service as the result of personal jealousies and rivalries, the perennial bane of all oriental governments. It was only the support of the army which maintained Tekin, and even so he was not in a position to attack his rivals in the government. When he had been successful in clearing the country of the Mahdi’s forces, he had his reward in dismissal from the governorship in which he was succeeded by Muhammad b. Hamal, but three days later he was restored, to be deposed again soon afterwards as the result of more palace intrigues. The two following governors, Hilal b. Badr and Ahmad b. Kayghalagh, held office, the first for two years, the other for one, and then in 312 Takin was restored and remained governor until his death in 321. Conditions indeed were such that only a military leader with the support of the army could exercise any effective control in the country. The reinforcements sent from Baghdad in 302 had done more harm than the Shiʿite invaders; they had totally demoralised the native soldiery, and the army was now no more than a large troop of brigands who lived on the plunder of the country. At his appointment in 312 Takin established the army in camps around his own palace as well as in quarters in the building itself, and, more by the force of his own personality than anything else, managed to keep them fairly in hand until his death. It was no small feat, for he was utterly unable to provide them with their pay, which was many years in arrear. At his death the governorship was assumed by his son Muhammad, but he had not his father’s power or popularity, and was soon mobbed and driven out by the discontented soldiers clamouring for their pay. The Treasurer Madaraʿi, who was in the Mahdi’s employ and largely responsible for the disorder in the finances, was obliged to hide himself. Several ambitious officers assumed the title of Governor and tried the expedient of raising funds by brigandage organised on a larger scale than usual, and the country had relief only in the fact that these were soon occupied in war against one another. It is not difficult to understand that the eyes of many Egyptians were turned longingly towards Kairawan, where the Mahdi, in an efficient though somewhat brutal manner, was administering a firm and well ordered state, maintaining civil law and peace. This is the easier to appreciate when we remember that Ismaʿilian missionaries were busy in Egypt, and the orderly government at Kairawan would naturally form one of their arguments.
At this juncture, when Egypt was plunged in anarchy, the Khalif at Baghdad intervened and appointed as governor Muhammad b. Tughj the Ikhshid, son of the Emir of Syria, who had himself been governor of Damascus since 318. As his name denotes, this new governor was of Turkish birth. For some time now the Khalifs, seriously alarmed at the growing independence of the various dynasties of hereditary governors, especially in Persia and the neighbouring lands, had been introducing Turkish mercenaries, reckless of the inevitable consequences.
The conditions of the Khalifate at this time show a close parallel with those prevailing in Europe under the later Karlings, when “the governor,—count, abbot, or bishop—tightened his grasp, turned a delegated into an independent, a personal into a territorial authority, and hardly owned a distant and feeble suzerain” (Bryce: Holy Rom. Empire, p. 79). So each governor appointed by the Khalif became the founder of an independent dynasty, barely conceding the mention of the suzerain’s name in the khutba and on the coinage. Such were the Tahirids who ruled in Khurasan from 205 to 259, the Saffarids in Persia from 254 to 290, the Samanids in Transoxiana and Persia from 288 to about 400, the Hamdanids who established themselves at Mosul in 292, at Aleppo in 333, and ruled there until 394, and the Aghlabids whom we have seen in Kairawan.
The Ikhshids claimed to be descended from the ancient kings of Ferghana on the Jaxartes, a district inhabited by fighting races, from whom the Khalif al-Muʿtasim (218-227) drew many mercenaries. The first of the Ikhshids to serve the Khalifs was a mercenary named Juff, and he continued in the Khalif’s employ until his death in 247. One of his sons named Tughj was in the service of Luʿluʿ, who acted as squire to Ibn Tulun in Egypt, and, when his master died, in that of Ishaq b. Kundaj, and afterwards in that of Ibn Tulun’s son, Abu l-Jaysh Khumasawaih, who regarded him with great favour and formed a very high opinion of his military abilities, in consequence of which he procured for him the governorship of Damascus and Tiberias. At his patron’s death Tughj offered himself to the Khalif al-Muktafi, who considered this an act of marked loyalty, and was greatly pleased with him, and made him one of his confidential officers. These favours provoked the jealousy of the wazir al-ʿAbbas, and he succeeded in getting Tughj cast into prison where he died. He left two sons, Muhammad and ʿAbdullah, who burned to avenge their father’s death, and their resentment was gratified when they saw al-ʿAbbas executed by the Hamdanid al-Husayn.
After this the elder son, Muhammad, went to Syria and joined himself to Takin, who was governor of Syria as well as of Egypt. In this service he prospered and was made governor of Amman. Then in 316 he was appointed to Ramla, in 318 he was transferred to Damascus, which led the way to his appointment as Emir of Egypt. This last charge was given him in 321, but the state of Syria did not allow his immediate departure, and Egypt was left for a while in the hands of Ahmad b. Kayghalagh, who returned to office temporarily. By 324 Syria had been reduced to order, and Muhammad the Ikhshid went over to Egypt to assume his governorship in person, leaving his brother ʿAbdullah in Syria.
There were some in Egypt who did not like the prospect of this new governor, and amongst these was the Treasurer Madaraʿi, who induced the acting governor, Ibn Kayghalagh, to take up arms to resist his entry. The Egyptian army marched to the frontier and engaged the Syrians and Turks under the Ikhshid at Farama, the ancient Pelusiun, now more generally known as Tineh (Arabic tîn = Greek πηλός “mud”), near the Egyptian end of the “short desert route,” via al-Arish from Syria. The result was a complete defeat of the Egyptians, and so Muhammad the Ikhshid continued on his way to the capital Fustat (“Old Cairo”) without further opposition. Meanwhile the Syrian fleet had sailed up the Nile and anchored off Giza, thus commanding the city until the Ikhshid marched up his forces and took possession. The arrival of the new governor and his army, largely Turkish in composition, established a firm and efficient government in Egypt again until his death in 335. At their first arrival indeed the Turkish troops began plundering the city, but they were soon called to order and then, although the new governor was severe and exacted heavy contributions, this stern rule was welcomed as it recalled the peace and prosperity of the golden days of Ibn Tulun. The resultant peace very soon opened up the way to literary activity and scholarship, and Egypt began to follow, though at a distance, the culture of ʿIraq. This literary development, as well as theological discussion and debates on jurisprudence, centered in the “Old Mosque,” which was also the scene of the most important state functions.
Although the establishment of the Ikhshid rule in Egypt gives the appearance of supreme power to the Khalif at Baghdad, who seems thus able to dispose of provinces and appoint governors at discretion, his position at the time was really very precarious. The Buwayhid dynasty of governors had established itself in ʿIraq in 320 (= A.D. 932), but Baghdad itself remained under the Khalif until 334, though generally he was only a tool in the hands of the commander of the garrison. These Buwayhids claimed descent from Buwayh, a prince in the hill country of Daylam, and so ultimately from the ancient kings of Persia. They appeared as rivals of other Daylamites led by Bajukin, who was Emir al-Umara or “Supreme Prince,” and had control of the government under the Khalifs ar-Razi and al-Muttaqi. Alarmed at the progress of the Buwayhids, Bajukin took up arms against them in 327, but was compelled to abandon his efforts by the report of disorders in Baghdad. Soon afterwards Bajukin was killed by a band of Kurdish marauders, and the capital was left in a state of anarchy. Then Baridi became Chief Emir, but was expelled a few weeks later: then the Daylamite Kurtakin, who turned out to be a tyrant. At this the Khalif appealed to Ibn Raiq, the Emir of Syria, and he expelled Kurtakin. Not long afterwards Baridi attacked Baghdad and Ibn Raiq had to flee, taking the Khalif with him to Mosul, which was in the hands of the Hamdanids. As champions of the Khalifate the Hamdanids marched against Baghdad, took it, and ruled there for a short time, until the Turk Tuzun drove them out and made himself Emir al-Umara in 331. Then another revolt drove him out, and the Khalif appealed again to the Hamdanids and escaped to Mosul; but when peace was concluded between Tuzun and the Hamdanids the Khalif remained in their hands. At this time, indeed, the Khalifate was very far from showing the character of an absolute monarchy. All over the Muslim world the Sunni, or orthodox party, recognised the Khalif as the Commander of the Faithful, except of course in Spain where the Umayyads of Cordova assumed the title of Khalif in 317. Enjoying great dignity and prestige in an office which combined many of the characteristics of the Pope and Emperor in the West, he was in fact no more than a puppet, a valuable asset in the hands of any one of the warring dynasties of Asia, but possessing no real authority. Yet his formal recognition was eagerly sought as a precious endorsement of de facto rights by Muslim rulers, and even princes in far-off India humbly begged his approval of their titles. It seems indeed as though the office of Khalif gained in spiritual influence as it lost in political authority.
Whilst in exile and in Hamdanid’s hands, the Khalif appealed to the Ikhshid whom he had set over Egypt, and Muhammad visited him at Riqqa and invited him to take refuge in Egypt; but al-Muttaqi, though anxious for help to recover the external symbols of authority at Baghdad, was not willing to put himself so entirely in the Ikhshid’s hands; he knew that Ikhshid and Hamdanid alike only desired to possess his person as a kind of imperial regalia, and so he preferred to entrust himself to the Turk Tuzun, who at least could establish him in the capital. He reigned in Baghdad in name only until 333, when Tuzun deposed him, put out his eyes, and enthroned al-Mustakfi in his place. But this was followed by a period of anarchy in Baghdad, until in 334 the Buwayhid prince took the city. A few months later al-Mustakfi was deposed and replaced by al-Muʿti, whose position under the Buwayhid princes was parallel to that of the Frankish kings under the “Mayors of the Palace,” with the aggravated condition that the Khalifs were spiritual pontiffs and the Buwayhids were, like the Hamdanids, Shiʿite heretics of the “Twelvers” sect. Buwayhid’s rule over Baghdad lasted from 384 to 447, when the Emir was displaced by the Saljuk Turks under Tughril Beg. Throughout this period the Buwayhids were content with the title of Emir al-Umara; they never assumed that of Sultan.
It has been necessary for us to turn aside to note the position of the Khalifate at the time, for otherwise we should have some difficulty in understanding the course of events in Egypt, which now takes the foremost place in the policy of the African Shiʿites. It is often possible to ignore the contemporary history of Spain and of North Africa when following the course of events in Egypt, but Egypt forms so integral a part of the world of Islam that it is never possible to treat its history, even during the comparative isolation of the Fatimid period, without some passing note of the contemporary history of the Baghdad Khalifate.
Whilst these changes were taking place in Asia and the Ikhshid was consolidating his power in Egypt, the Mahdi continued ruling at Kairawan, and, though North Africa was one of the most turbulent and the least civilized parts of the Islamic world, his rule was stable and orderly. In 312 he added a suburb to the city which he called al-Muhammadiya, and which served as a kind of royal cantonments closed against the ordinary citizens, and used only as an official settlement of those engaged in the public administration and as the site of the various public offices. Such official suburbs were very frequent in oriental capitals, and become a regular feature of the great Muslim royal cities. The Mahdi’s later years were somewhat clouded by his relations with the Qarmatians, who were still active in Asia, and who caused the whole Ismaʿilian movement to be regarded with grave suspicion by the Muslim world at large.
Since 311, as we have seen, the Qarmatians had occupied Basra. In 317 they had spread down into the Hijaz, and on the 8th of the month of the pilgrimage in that year the pilgrims who had come up to Mecca were attacked by them. The Sherif of Mecca, many of his attendants, and many of the pilgrims, were killed: the sacred spring of Zamzam was choked up with the bodies of the slain which were tumbled in: the door of the “House of God” was broken open, the veil which covered the House was torn down, and the sacred black stone was removed from the Kaʿaba and carried away to the Qarmatian headquarters at Hajar. Never in the history of Islam has there been sacrilege at all comparable to this, and never before had the Qarmatians advertised so boldly their contempt for the Muslim religion. Begkem, the Emir of Baghdad, offered them a reward of 50,000 dinars to restore the sacred stone, but the offer was refused.
According to Ibn Athir, quoted by Ibn Khallikan (i. 427, etc.) the Mahdi then wrote to them from Kairawan: “By what you have done you have justified the charge of infidelity brought against our sect, and the title of ‘impious’ given to the missionaries acting for our dynasty; if you restore not what you have taken from the people of Mecca, the pilgrims and others, if you replace not the Black Stone and the veil of the Kaʿaba, we shall renounce you in this world and the next.” This letter was more effectual than Begkem’s proffered reward, and the Qarmatians restored the Black Stone with the statement, “We took it by order, and by order we return it.” It was restored either in Dhu l-Kaada or Dhu l-Hijja of 339. Of the year there seems no question, and Ibn Khallikan points out that the Mahdi died in 322. He suggests, therefore, that the letter and the Qarmatian reply were fabrications, presumably for the purpose of throwing the odium of sacrilege on the Mahdi. But it is not necessary to suppose that the Black Stone was returned immediately in response to the Mahdi’s request. A more likely interpretation is given by Macdonald, who accepts the letter as genuine and comments: “When an enormous ransom was offered for the stone they (i.e., the Qarmatians) declined—they had orders not to send it back. Everyone understood that the orders were from Africa. So ʿUbayd Allah found it advisable to address them in a public letter, exhorting them to be better Muslims. The writing and reading of this letter must have been accompanied by mirth, at any rate no attention was paid to it by the Qarmatians. It was not till the time of the third Fatimid Khalifa that they were permitted to do business with that stone” (Macdonald: Muslim Theology, pp. 46-47). This suggests a plausible explanation, that the letter was sent by the Mahdi, but was only intended to disclaim any responsibility for the taking of the stone on his part; that it was not intended to be heeded, and was not taken seriously, the stone being detained until long after the Mahdi’s death. This theory would fit in with the policy of the Fatimids at Kairawan, which carefully avoided anything likely to offend the orthodox, and would dispose of Ibn Khallikan’s objection, which is based on the supposition that the date of the return of the stone was shortly after the writing of the latter. The letter assumes that the Qarmatians and the Fatimids were members of the same sect. Undoubtedly they had been so originally, but later on they definitely separated, and we are not clear as to the time of this division. It seems probable that the external quasi-orthodoxy of the Fatimids in Africa was the cause of its separation from the Qarmatians, who had made more open profession of the destructive elements of their religion.
The Mahdi died in 322 (A.D. 933), and was succeeded by his son Abu l-Kasim, who assumed the name of al-Qaʾim.