3 The
authorities
are those of the last note.
Cambridge Medieval History - v5 - Contest of Empire and the Papacy
His minority was a time of chaos, when
the chiefs of the Berber and Turkish guards fought and schemed for
supremacy. The native historians relate strange and incredible stories of
his personal government, out of which it is nearly impossible to make a
coherent picture. He is represented as arbitrary and cruel beyond measure
and as the persecutor of every class in turn. He kept his position only
by unscrupulous assassination and by playing off against one another
the Arab, Turkish, Berber, and Negro factions which mingled in his
court. On the other hand, measures are attributed to him which have
been interpreted as the conceptions of a would-be reformer and unpractical
idealist. In part of his reign he seems to be a rigid Muslim, persecuting
Jews and Christians against all tradition and in spite of the fact that his
mother was a Christian and his uncle at one time Patriarch of Jerusalem.
At another period his conduct suggests that he was influenced by the
esoteric doctrine of the Ismāʻīlian sect to which his ancestors belonged.
Towards the end of his life he seems to have countenanced sectaries who
proclaimed him to be an incarnation of deity. The mystery of his death
was a fitting close to a mysterious life. He left his palace one dark night
(13 February 1021) never to return; the presumption is that he was
assassinated. But some declared that he would yet return in triumph as
the divine vice-gerent, and the Druses of Lebanon are said to maintain
this belief to the present day.
The revolts in southern Syria at the beginning of Hākim's reign
reflect the strife of parties in Egypt and did not threaten the authority
of the Caliphate itself. This distinction helps to make intelligible the
maze of revolts and depositions and revolutions in which the governorship
of Damascus was now involved. In twenty-four years and a half there
were at least twenty changes in the occupancy of the post. Two governors
between them held office for nine years, so that the average term of the
remainder was less than ten months each. More than one was deposed
within two months of his appointment. Generally the only cause of change
was the arbitrary disposition of the Caliph or an alteration in the balance
of power amongst the emirs of his court. Sometimes the new governor
had to establish his authority by force of arms.
On one occasion in these years there was a revolt of a more serious
character. Early in 1011 the Arab chief Mufarrij ibn Daghfal ibn
Jarrāḥ, having defeated the Caliph's representative, became ruler of inland
Palestine for the second time. He failed to occupy any of the coast-towns
but held possession of the interior for two years and five months, until
his death (1013). A peculiar feature of this revolt was the acknowledgment
by Ibn Daghfal of the sharīf of Mecca, Hasan ibn Jaʼfar, as “commander
of the faithful. ” This personage was a descendant of the Prophet and so
possessed one outstanding qualification for the Caliphate. But his only
supporter was Ibn Daghfal, and his phantom authority lasted less than
two years. Ibn Daghfal's sons were defeated by Hākim's troops im-
cH. VI.
## p. 254 (#300) ############################################
254
Ruin of the Holy Sepulchre
mediately after their father's death, and the control of Palestine passed
again to the governor of Damascus.
An event of special interest to Christendom occurred in Jerusalem
during Hākim's Caliphate, namely, the profanation and ruin of the
church of the Holy Sepulchre (commencing 27 September 1009). It is
unlikely that the fabric of the church was seriously injured. Hākim
ordered its relics to be taken away and its monuments, including the
Holy Sepulchre, to be destroyed. The portable furnishings of the church
and its treasures were carried into safety before the Caliph's agents arrived.
But the Holy Sepulchre and other venerated shrines were destroyed as
completely as possible. The interior must have been left in a very
mutilated condition. Mufarrij ibn Daghfal began the work of restora-
tion when he was ruler of southern Palestine (i. e. between 1011 and 1013).
Saʻīd-ad-Daulah of Aleppo having died early in January 1002, Lūlū
banished the surviving members of the Hamdān family to Egypt and
assumed the emirate. He acknowledged the Fățimite Caliph, Hākim,
and also continued to pay tribute to the Greek Emperor. His rule is
praised as having been wise and just. After his death (August 1009)
1 The intention of the Caliph and the extent of the destruction of the church are
to be ascertained chiefly from the narratives of Abu Ya'là (pp. 66-68) and Yahyà
(p. 195 sq. ). The former conveys the impression that a primary object of the Caliph
was to plunder the treasures of the church. His version of the Caliph's order par-
ticularly specifies the destruction of the monuments (athar, which, however, is an
ambiguous word and might signify relics). When the historian says “its structures
were ruined and uprooted stone by stone,” he refers, presumably, to such buildings
as the shrine of the Holy Sepulchre. Yahya's account would harmonise completely
with this view but for his summary statement "it was thrown down completely to
the foundations, except what could not be ruined and was too difficult to uproot. ”
If the subject of the first verb is the church, then obviously the expression “to the
foundations” cannot seriously be pressed, and the amount of damage to the walls
and fabric remains obscure. But the subject in the original is quite vague, and the
reference may be principally to the monuments and the interior structures. The
brief statements of other writers, Eastern and Western, are no proof of such a
complete destruction of the church as modern writers have assumed. The exact
date is supplied by Yahyà (Tuesday, five days from the end of Şafar, A. H. 400).
This agrees so far with the statement of Ademar (Lequien, 111, 478) that the Holy
Sepulchre was destroyed on 29 September (A. D. 1010). Yahyà (p. 201, lines 9 sqq. )
relates that Ibn Daghfal began the work of restoration; he observes particularly
that “he restored the places in it” (i. e. the shrines ? ). Possibly Hākim also
authorised its repair, under the influence of his Christian mother and uncle (cf.
Glaber). In A. H. 411 (A. D. 1020) the church is described by Yahyà (p. 230) as still
in a ruined condition. The relevant extracts from the Latin historians are given by
Lequien, Oriens Christianus, III, 475 sqq. Other references will be found in Ibn al-
athir, ix, 147, Maqrīzi, 11, 287 (Kitāb al-khițaț, Bulak edit. , 1853–1854), Abu 'l-maḥā-
sin, 11, 64 (from Adh-dhahabi), and 11, 101, Cedrenus (Corp. Script. Byzant. ), Vol. 11,
p. 456. Forty-five years before this, in A. D. 966, this church had been seriously injured
by the local Muslim and Jewish population (Yahyà, p. 125 sq. ). The dome of the
Holy Sepulchre and the roof of the adjoining main church were then both de-
stroyed, and the necessary repairs had been completed only a short time before the
events of 1009.
## p. 255 (#301) ############################################
Egypt and Syria
255
Manşūr his son, although unpopular, held the emirate for some years
against the Hamdan family and the attacks of the Bani Kilāb under
Şāliḥ ibn Mirdās. Finally he was expelled from Aleppo by an insurrec-
tion (6 January 1016), headed by the governor of the castle, Mubārak-
ad-Daulah Fataḥ, and, having escaped to Antioch, became a pensioner of
the Greeks. These events increased the authority of the Egyptians in
northern Syria. About a year later, Mubārak-ad-Daulah was made
governor of Tyre, Sidon, and Beyrout by Hākim, and ‘Azīz-ad-Daulah
Fātik, an Armenian, was installed as governor of Aleppo (3 February
1017). As so often happened in such cases, the new governor began to
act as an independent emir, and his assassination (13 June 1022)
was probably instigated by Sitt-al-mulk, Hākim's sister, now regent.
During the next two years and a half an Egyptian garrison held the
citadel of Aleppo, and a series of Egyptian governors controlled the city.
The seventh Fāțimite Caliph was Abu 'l-ḥasan 'Ali az-Zāhir. He
was a boy when he succeeded his father and he never exercised much
influence in the government of his dominions (1021-1036). For the
first three years of his reign Hākim's sister, Sitt-al-mulk, was regent.
Soon after her death the Arab tribes on the borders of Syria made a
league against the Caliph, hoping to conquer and rule the country
(1024). The leaders of the revolt were Şāliḥ ibn Mirdās, chief of the
Bani Kilāb, who lived in the neighbourhood of Aleppo, Sinān ibn
«Ulyān, chief of the Bani Kalb, near Damascus, and Hassān, a son of
Mufarrij ibn Daghfal, whose home was in southern Palestine. The con-
federates were at first successful both in Palestine and northern Syria.
Aleppo was captured by Sāliḥ ibn Mirdās (January 1025)', and Hims,
Baalbek, and Sidon soon acknowledged his authority. Thus a
dynasty, that of the Mirdāsites, was established in Aleppo (1025–1080).
In Palestine the Caliph's representative, Anūshtakīn ad-dizbirī, was more
than once defeated and was driven out of Syria. The least successful of
the allies was Sinān ibn 'Ulyān. After his death in July 1028, his
successor deserted the alliance and submitted to the Caliph. In the
following year a decisive battle was fought at Uqḥuwānah, south of Lake
Tiberias, between Sāliḥ and Hassān on the one side, and the Egyptians
and their allies on the other (14 May 1029). Şāliḥ was killed and
Hassān's power was completely broken. From now onwards Anushtakīn
was governor of Damascus and the most powerful emir in Syria
(1029-1041).
i The citadel did not surrender until five months later (June 1025). Cf. note
supra, p. 246. As illustrating the textual criticism that must always be applied to
the dates of Arabic historians, it may be pointed out that 13 Dhu 'l-qa'dah 415 in
Kamal-ad-Din's text (date of the capture of Aleppo) should be 23 Dhu 'l-qa'dah 415,
and that 1 Jumădà ii 416 (date of the surrender of the citadel) should be 1 Jumādà i
416. These are typical errors. The correct dates are given by Yahyà (pp. 246, 248),
along with the week-days, which provide the necessary test of accuracy. See
Stevenson, Crusaders in the East, appendix on Chronology.
new
CH. VI.
## p. 256 (#302) ############################################
256
The Greeks in Syria
During the period of this rebellion, in 1027 (A. H. 418), an in-
teresting treaty of peace was made between the Fāțimite Caliph and the
Emperor Constantine VIII. It was provided that the Caliph's name
should be mentioned in the public prayers of the mosques throughout
the Empire, to the exclusion of his Abbasid rival. This arrangement
was continued until the year 1056, when it was reversed at the instance
of the Turkish Sultan Tughril Beg. A further recognition of the re-
presentative character of the Fātimite Caliph, and another concession to
Islām, was contained in the provision that the Caliph might restore the
mosque in Constantinople and appoint a muezzin to officiate there. The
counterpart of these provisions gave the Emperor the right to restore
the church of the Holy Sepulchre in Jerusalem. It is not to be assumed
that the church had lain in ruins since its profanation by Hākim's
orders in 1009, nor, perhaps, that much was actually done at this
time in the way of restoration'. Another concession made by the
Caliph was that those Christians who had become Muslims by compul-
sion in the time of Hākim might again profess Christianity without
penalty. It may be assumed that the treaty of peace, as usual, was valid
for a limited period only; but the term is not specified by the only
source that mentions the treaty.
Nașr Shibl-ad-Daulah, son of Şāliḥ ibn Mirdās, was permitted to
succeed his father as ruler of Aleppo on the condition that he acknow-
ledged the Fāțimite Caliph in the customary manner, on his coinage and
in the public prayers of Friday. His emirate did not include Hims or
Hamāh, but extended north-eastward to the Euphrates. The Greeks,
who had recently been losing ground in Syria, now seized what seemed
to them an opportunity of improving their position. The territory of
Aleppo was twice invaded (1029 and 1030), both times unsuccessfully.
The Emperor Romanus shared in the second invasion, a very ill-judged
attempt. The Greek army suffered so much in the neighbourhood of
“Azāz from the hot season, lack of water, and fever that it was com-
pelled to retreat in a few days and lost heavily as it retired (August
1030). The Emir of Aleppo, reckoning his triumph an occasion of con-
ciliation and not of defiance, at once opened negotiations for peace.
A treaty was signed on terms that were distinctly unfavourable to the
Muslim city. Aleppo again became tributary to the Empire, and a Greek
deputy was allowed to reside in the city and wateh over the due per-
formance of the conditions of peace (April 1031).
1 Maqrizi (Khițaț, p. 355, lines 9 sqq. ) is here the chief authority. He does not
name the Emperor, but does mention ļāhir. William of Tyre also refers to a
restoration of the church by permission of “Daher” (= Zāhir), and Cedrenus, 11,
501, implies that the restoration was permitted by a son of the Caliph who destroyed
the church. The brief statement of Cedrenus is obviously very confused and in-
accurate (cf. infru, p. 257, n. 3). If Maqrizi were the only authority it might be con-
jectured that his date A. H. 418 was an error for A, H. 428 (i. e. A. D. 1037). See p. 257.
## p. 257 (#303) ############################################
The Greeks in Syria
257
At this date the territory of the Greeks in Syria extended eastward
from Antioch to Hārim and southwards along the coast as far as
Maraqīyah. The hillmen of the Jabal Anşarīyah, who adjoined this
territory, were partially held in check by strong castles such as Bikisrāyil,
but still maintained their independence. After the defeat of Romanus,
one of the chiefs of the hill tribes, Nașr ibn Mushraf, captured Bikisrāyil
and a general rising took place. Maraqīyah was besieged by Ibn Mushraf
and the Emir of Tripolis. Nicetas, the new governor of Antioch, took
prompt action against a very dangerous situation. He raised the siege
of Maraqiyah (December 1030), and during the next two years syste-
matically besieged and reduced the castles of the hillmen (1031-1032).
Balātunus, Bikisrāyil, and $āfīthā were among the fortresses now garri-
soned and held by the Greeks.
These events brought about a resumption of hostilities between the
Empire and the Egyptian Caliph. Anūshtakin of Damascus and the
Emir of Tyre had given a timorous support to the mountaineers in their
struggle with Nicetas. Rafanīyah was therefore attacked and captured
by Greek troops. A Byzantine fleet threatened Alexandria and the
mouths of the Nile. Both parties desired a stable peace, but the task
of settling the matters in dispute proved to be long and difficult. The
chief obstacle to a settlement was the demand of the Emperor that
Aleppo should be treated as a Greek dependency? . The negotiations
were continued, or resumed, after the death of Romanus (April 1034),
and peace was signed, perhaps in the autumn of 1037? . Each party
pledged itself not to assist the enemies of the other, and their respective
spheres of influence in northern Syria were defined. The Greek deputy
whom Romanus had stationed in Aleppo had been driven out soon after
that Emperor's death, so that Aleppo probably secured its independence.
The right of the Emperor to renovate the church of the Holy Sepulchre
was acknowledged, and possibly the privilege of appointing the Bishop
of Jerusalem. In return Michael IV set free 5000 Muslim prisoners.
The duration of the peace was fixed at thirty years. The Emperor sent
builders and money to Jerusalem, but the repairs to the church were not
completed until the reign of his successor Constantine IX 3.
1 Full details of the negotiations are given by Yahyà (p. 270 sq. ). His account
throws light on the terms of the treaty of 1037, and has been used above to sup-
plement the meagre details of that treaty preserved by others. Unfortunately
Yahya's narrative, at least as printed, breaks off before the year 1037 is reached.
? This date is derived from Yaḥyà, who seems to fix it as being three and a
half (Muslim) years after the death of Romanus. Barhebraeus gives A. H. 427, but
his narrative, on the whole, is not strongly against A. H. 428, which would agree
with Yahyà. On the other hand, Cedrenus (vol. 11, p. 515) gives A. M. 6544
(=A. D. 1035–1036) and Abu 'l-fidā (111, 96) and Ibn al-athir (1x, 313) both A. H. 429.
3 The authorities are those of the last note. The renovation of the church must
have been considerable and not merely a repair of the damage done by Hākim.
Cedrenus (11, 501) wrongly makes Romanus commence the work and Michael com-
17
C. MED. H. VOL. V. CH. VI.
## p. 258 (#304) ############################################
258
Caliphate of Mustanșir
The eighth Fāțimite Caliph, Abu tamīm Maʻadd al-Mustanşir,
was only seven years old when his father died (June 1036), so that his
reign began with a succession of regencies. The Caliph's mother, an
African woman, exercised a considerable amount of influence. The con-
temporary Persian traveller Nāşir-i-Khusrau records very favourable
impressions of the prosperity and tranquillity of the country while the
Caliph was a minor.
Early in this reign peaceful relations between Aleppo and Egypt
were broken off. Nașr ibn Şāliḥ was defeated and slain in battle with
Anūshtakīn (May 1038), and Aleppo was captured and garrisoned by
Egyptian troops for a few years (1038-1042). The disgrace of Anushtakin,
followed immediately by his death (January-February 1042)', weakened
the Fāțimite dominion all over Syria. Aleppo was recovered by Nașr's
brother, Mu'izz-ad-Daulah Thumāl (March 1042). He resumed payment
of tribute to the Greeks and so secured himself in that direction. The
terms of the rulers of Egypt were not so easily satisfied. Envoys came
and went between the parties. Attacks were launched against Thumal
by the Emirs of Hims and Damascus, acting in the name of the Caliph
(1048–1050). At length, in 1050, an agreement satisfactory to both sides
was arrived at.
Two isolated events, which are a part of the history of the Fāțimite
Caliphs, deserve mention here. In 10492 Mu'izz ibn Bādīs, the Zairite
Emir of Tunis, ceased to pay tribute to Mustanşir and transferred his
allegiance to the Abbasid Caliph. His family had ruled in Qairawān,
in practical independence, since 973, when the Fāțimite Caliph of the
day made Cairo his residence and capital. But the formal separation,
signalised by the acknowledgment of the Caliph of Baghdad, took place
only now. On the other hand, for the greater part of the year 1059 the
Caliphate of Mustanşir was acknowledged in Baghdad itself. Such
acknowledgments were now symbols of the triumph of political parties
and alliances. The Turkish Sultan Țughril Beg identified his cause with
that of the Abbasid Caliphs, with the result that his enemies in
plete it. William of Tyre, also, says that Romanus received permission to restore
the church. Possibly the explanation of these statements is that during the nego-
tiations of 1031-1034 the article regarding the restoration of the church was agreed
to (so Yaḥyà), although the treaty of which it was part was not signed until the
reign of Michael. William of Tyre seems to imply that not much progress was
made with the repairs until the reign of Constantine. He gives 1048 as the date
when the work was completed; similarly Abu 'l-fidā (A. H. 440). There was a severe
earthquake in Jerusalem in January 1034 (Yaḥyà, p. 272; cf. Cedrenus, 11, p. 511),
which may possibly have caused some of the injury that was afterwards repaired.
1 1042 according to Kamāl-ad-Din and Ibn al-athir, 1041 according to Abu
Yaʻlà.
2 So Ibn 'Adhāri (A. H. 440), who further states that Ibn Bādis struck new
coinage in Shaʻbān 441 (end of 1049 or beginning of 1050). Ibn al-athir is incon-
sistent with himself, giving both A. H. 435 and A. H. 440. Abu 'l-maḥāsin gives both
A. H. 435 and A. H. 443, preferring the latter.
## p. 259 (#305) ############################################
Emergence of the Turks
259
Mesopotamia were disposed to favour recognition of the Fāțimite Caliphs
in those districts and cities where they triumphed. In 1059 Baghdad
was occupied by a Turkish emir, Arslān al-Basāsīrī, who, being an enemy
of the sultan, acted in the manner just des ped. The occasion was
hailed in Egypt as an extraordinary triumph, and in fact probably
marked the highest point of superiority to the Abbasids ever reached
by the Fāțimite Caliphs.
When Mustanşir came of age he shewed such feebleness and in-
capacity that he was treated by all parties as a cypher in the government.
The ministry of Hasan al-yāzūrī (1050-1058) was still, on the whole,
prosperous and considerate of the general welfare. But after his death
there recommenced a bitter struggle for power between the leaders of the
Turkish and those of the negro troops. The country was devastated and
impoverished by civil war, and finally lay at the mercy of the unscrupulous
and cruel Turkish leader Nāsir-ad-Daulah ibn Hamdān (1062-1073). Pro-
longed drought and famine increased the miseries of the unhappy people.
The influence of Egypt upon foreign affairs fell to its lowest ebb. It was
in no way able to share in the defence of Syria against the Seljūq Turks.
The rule of Mu'izz-ad-Daulah Thumāl in Aleppo was mild and
generous, and therefore popular. His greatest troubles were caused by
the unruly Arabs of the district, the Bani Kilāb, and latterly by the
Seljūq Turks, already planted at Raḥabah on the Euphrates. In January
1058, feeling no longer equal to the tasks of his position, he abdicated
and left an Egyptian governor and garrison once more in power. These
were soon expelled by the citizens assisted by the Bani Kilāb (September
1060), and shortly afterwards Mu'izz-ad-Daulah was persuaded to return
to his former post (April 1061). During his second brief emirate the
Greeks provoked hostilities by repairing some border castles, and Artāḥ
was taken from them. Peace with them was renewed during the civil
war that followed Mu'izz-ad-Daulah's death (November 1062). Artāḥ
appears to have returned to its former owners.
Thumāl's brother, Asad-ad-Daulah “Atīyah ibn Şāliḥ, was his suc-
cessor. His title to succeed was challenged by a nephew, Maḥmūd ibn
Naşr, and the brief period of his emirate was one of civil war (1062–1065).
It was at this date, just before the Norman conquest of England, that
the Seljūq Turks entered Syria.
From the ninth century onwards, Turkish governors and Turkish
generals and Turkish mercenaries play an important part in the history
of Syria and especially of Egypt. The Tūlūnites were a Turkish family
and were served by Turkish officers and soldiers. So also were the Ikh-
shīds. In Mesopotamia, from which these viceroys came, Turkish slaves
held the highest place, subject only to the nominal authority of the
Caliphs. In Egypt the Fāțimite dynasty retained and added to the
Turkish household troops of their predecessors. Turkish, Berber, and
Negro factions struggled for supremacy, and the Fāțimite governors of
CH. 1.
17-2
## p. 260 (#306) ############################################
260
Turkish conquest of Syria
Syrian towns in the tenth and eleventh centuries were often Turkish
Mamlūks.
Before the middle of the eleventh century, a new wave of Turkish
migration, under the great Sultan Țughril Beg (1037–1063), swept into
Lower Mesopotamia from the north and threatened Armenia and Upper
Mesopotamia. It was the precursor of the conquest of Syria by the
Seljūq Turks. The manner of their conquest is representative of many
other periods in Syrian history. Bands of horsemen, a few hundred
strong-seldom as many as a thousand—rode under adventurous leaders
who sought their fortune and lived by their swords. They took service
with any ruler for money or for lands, and gained their chief advantage
where local feuds were being waged. Some novelty in their arms or in
their way of fighting might give them an advantage in battle. In any
case they were always on the war-path, and so could finally wear down
the resistance of cities which depended upon the cultivation of the land
or upon peaceful industry. The inland towns of Syria -- Aleppo, Hims,
Baalbek, Damascus, Jerusalem-yielded first and most completely to
the Turks. Once established, the way of the conquerors was smoothed by
their being Muslims. Their introduction of the nominal authority of the
Caliphs of Baghdad was almost a matter of indifference to their subjects.
The rule of Turkish emirs was already familiar in Syria. The invaders
were backed by the prestige of the Seljūq sultans, but only to a slight
extent occasionally by their armies.
A conquest of the character just described implies, of course, that
Syria was in its normal state of political disintegration. It was, in fact,
even less united than it had been for some time past. Aleppo was an
independent territory and was rent by civil war. The Arabs hung loosely
on the borders. The hillmen of the Jabal Anşarīyah took no interest in
the fate of the neighbouring plains. Antioch and its dependencies were
under the rule of foreigners. Damascus and the coast towns from Tripolis
southwards had cut themselves adrift from Egypt, which was in the
throes of revolution. They were governed by independent emirs, anta-
gonistic to one another. Only the south-west of Palestine was still closely
attached to Egypt. After the great defeat of the Greeks at Manzikert
(1071), Antioch was almost left to its own resources. Even the Arme-
nians, who had long given soldiers to the Greeks on the eastern borders
of the Empire and in Syria, now preferred to make terms with the
Turks.
Hārūn ibn Khān was the first of the Seljūq Turks to gain a footing
in Syria. About the end of 1064 he and his thousand followers turned
the scale in favour of 'Ațīyah ibn Şāliḥ against his rival Maḥmūd. When,
however, ‘Ațīyah and the citizens of Aleppo rose against their deliverer
and massacred his followers, he made off with the survivors to Maḥmūd
and helped him to victory at the battle of Dābiq (16 June 1065). After
the surrender of Aleppo to Maḥmūd (13 August 1065), Hārūn was given
## p. 261 (#307) ############################################
Recognition of the Abbasid Caliphs
261
the little township of Ma'arrat-an-Nuʻmān in fief, and settled there
with a mixed following of Turks, Kurds, and Dailemites.
In the summer of 1067 another Turkish leader, ‘Afshin by name,
raided the territory of Antioch and carried off great booty. His prisoners
were so many that “a girl was sold for two dinars and a boy for a set of
horseshoes. ” In the following year ‘Afshin besieged Antioch and was
bought off by the payment of a large sum of money (1068). At the
same time there was war between Aleppo and Antioch, and Artāḥ was
captured by Hārūn ibn Khān after a five months' siege (1068). In the
following year a Greek army, under the Emperor himself (Romanus
Diogenes), recovered Artāḥ and captured Mambij. Before the close of
the year the Armenian governor of Antioch (Kachatur) made peace with
Mahmud on terms that were favourable to the latter.
In 1070 a Turkish leader, known as Zandiq, entered Syria at the
head of large forces and ravaged the territories of Aleppo, Hamāh, Hims,
and Rafanīyah. This was the first devastation of Muslim Syria by the
Turks. It decided Maḥmūd to seek the protection of the Sultan Alp
Arslān (1063-1072), and at the same time, in consequence, to transfer his
allegiance from the Fāțimite to the Abbasid Caliph. Prayers were said
in the mosques of Aleppo for the new Caliph and for the sultan on Friday
30 July 10702
Alp Arslān now demanded that Mahmūd should engage in war with
Antioch and with the Fāțimite emirs. Maḥmūd having at first refused,
the sultan invaded Syria (spring of 1071). Two months were spent in
negotiations, and during another month Aleppo was blockaded. Then
Mahmud submitted and became the sultan's vassal. The historian of these
events comments especially upon the discipline of Alp Arslān's army.
The persons and the property of the country people were respected. Even
the forage that the soldiers used was often paid for. Aleppo was neither
ruined nor pillaged. Fasdiq, where Alp Arslān pitched his tent during
the expedition, was henceforth known as the Sultan's Hill (Tell-as-sulțān).
Maḥmūd does not seem to have shewn much zeal in the fulfilment of his
pledge to the sultan during the remainder of his emirate (ob. 10 January
1074). His sons Nașr (1074-1076) and Sābiq (1076-1080) were the last
of the Mirdāsites to rule Aleppo. Fresh bands of Turks were pouring into
Syria. Rafanīyah was occupied by Jawāli ibn Abaq (1075), who raided
the territory of Aleppo until he was severely defeated by Aḥmad Shāh,
another Turkish leader, in the service of Nasr ibn Maḥmūd and after-
wards of his brother Sābiq. The assassination of Nașr and the acces-
sion of Sābiq illustrate the influence now exercised by the Turks over
the internal affairs of Aleppo. Sābiq was opposed by two of his brothers
and by the Bani Kilāb, but defeated his enemies with the help of Aḥmad
1 Fifteen dinars may have been a normal price.
? The same change was made in Mecca in this same year. But the acknowledg-
ment of the Fățimites was resumed there again in 1074 or 1075.
CH. VI.
## p. 262 (#308) ############################################
262
The Turks in Palestine
Shāh and other Turks (July 1076). Nașr and Sābiq both waged war inter-
mittently with the Greeks. In 1075 Mambij was recovered by the former.
The principal Seljūq emirs of the north of Syria were ‘Afshin, Zandiq,
and Muḥammad ibn Dimlaj. In the summer of 1077 they were ordered
by Alp Arslān's successor, Malik Shāh (1072-1092), to unite under the
command of his brother Tāj-ad-Daulah Tutush. In the spring of 1078
Tutush attacked Aleppo at the head of a large force, which included the
Bani Kilāb and the soldiers of Sharaf-ad-Daulah Muslim of Mosul (1061-
1085). The siege lasted four months and its failure was attributed to the
action of Sharaf-ad-Daulah, an old ally of the Turks, who was now
turning against them. Next year (1079) Tutush resumed his operations
in Syria, with some success. Mambij, Buzā‘ah, and other places sur-
rendered or were captured. Then an invitation from the Turkish Emir
of Damascus, At-siz ibn Abaq, drew his attention southwards.
The first mention of the presence of Seljūq Turks in Palestine be-
longs to the year 1070". The authority of Nāșir-ad-Daulah, governor
of Egypt, did not extend at that time beyond the south of Palestine.
Acre and Sidon were governed by an Armenian, Badr-al-jamālī, who
had played a prominent part in Syrian affairs since 1063. Damascus,
Tyre, and Tripolis were in the hands of other independent emirs. The
Arab tribes on the southern and eastern borders were their own masters.
After the assassination of Nāșir-ad-Daulah (10 May 1073), Mustanşir
appealed to Badr-al-jamāli to end the régime of the Turkish slaves in
Egypt. At the head of his Syrian troops he occupied Cairo (February
1074), and in a few years restored unwonted peace and order to the
country. He was the all-powerful ruler of Egypt for twenty years
(1074–1094).
Several Turkish leaders shared in the conquest of southern Syria, but
they all, in a measure, seem to have obeyed At-siz ibn Abaq? His
first acquisition was ‘Ammān, an Arab stronghold in the Balqā (1071? )*.
From there he became master of the south of Palestine, including Jeru-
salem and Ramlah. Jerusalem capitulated on terms, and suffered nothing
from its change of rulers. For several years At-sīz, having marked Da-
mascus as his prey, ravaged its territory, especially at harvest time, and
levied contributions from the coast-towns as the price of their immunity.
In 1075 he captured Rafanīyah and gave it over into the charge of
his brother Jawālī. In the summer of 1076 Damascus at last sur-
rendered to him. After this he ventured to invade Egypt and was
severely defeated in the neighbourhood of Cairo (January 1077). His
1
A. H. 462 (Ibn al-athir, x, 40; Kamāl-ad-Din, p. 77).
? His name first appears in the year 1071 (A. H. 463). Qarlū is the only leader
mentioned by name in the previous year (A. A. 462).
3 Quatremère, p. 413 f.
4 The date is from Sibt ibn al-Jauzi. The general sequence of events suggests a
year later. Cf. infra, p. 263, n. 1.
## p. 263 (#309) ############################################
Turkish dissensions
263
bold challenge prompted Badr-al-jamāli to seek the recovery of Palestine
and Damascus. At-siz, fearing the issue of the conflict he had provoked,
invited Tāj-ad-Daulah Tutush to his aid. The result might have been
expected. Tutush took possession of Damascus and put At-sīz to death
(September 1079)". Badr-al-jamālī withdrew his forces from Palestine.
The emirs of the coast-towns, for the most part, paid tribute to Tutush
rather than submit to their ancient rival, the governor of Egypt.
Finding himself secure in Damascus, Tutush at once sent most of his
army back into northern Syria. “Afshīn, his general, laid waste the
country from Baalbek to Aleppo and ravaged the territory of Antioch.
In consequence of this attack Sabiq and the citizens of Aleppo surrendered
the town to Sharaf-ad-Daulah Muslim of Mosul (June 1080). Sābiq re-
tired to Raḥabah, and Muslim and Tutush stood opposed as well-matched
antagonists.
As matters turned out, there was little actual fighting between the
rivals. For two or three years Muslim strengthened his position in
northern Syria and Upper Mesopotamia, held communications with Badr-
al-jamālī, and sought to divert the tribute of Antioch from the sultan to
himself.
the chiefs of the Berber and Turkish guards fought and schemed for
supremacy. The native historians relate strange and incredible stories of
his personal government, out of which it is nearly impossible to make a
coherent picture. He is represented as arbitrary and cruel beyond measure
and as the persecutor of every class in turn. He kept his position only
by unscrupulous assassination and by playing off against one another
the Arab, Turkish, Berber, and Negro factions which mingled in his
court. On the other hand, measures are attributed to him which have
been interpreted as the conceptions of a would-be reformer and unpractical
idealist. In part of his reign he seems to be a rigid Muslim, persecuting
Jews and Christians against all tradition and in spite of the fact that his
mother was a Christian and his uncle at one time Patriarch of Jerusalem.
At another period his conduct suggests that he was influenced by the
esoteric doctrine of the Ismāʻīlian sect to which his ancestors belonged.
Towards the end of his life he seems to have countenanced sectaries who
proclaimed him to be an incarnation of deity. The mystery of his death
was a fitting close to a mysterious life. He left his palace one dark night
(13 February 1021) never to return; the presumption is that he was
assassinated. But some declared that he would yet return in triumph as
the divine vice-gerent, and the Druses of Lebanon are said to maintain
this belief to the present day.
The revolts in southern Syria at the beginning of Hākim's reign
reflect the strife of parties in Egypt and did not threaten the authority
of the Caliphate itself. This distinction helps to make intelligible the
maze of revolts and depositions and revolutions in which the governorship
of Damascus was now involved. In twenty-four years and a half there
were at least twenty changes in the occupancy of the post. Two governors
between them held office for nine years, so that the average term of the
remainder was less than ten months each. More than one was deposed
within two months of his appointment. Generally the only cause of change
was the arbitrary disposition of the Caliph or an alteration in the balance
of power amongst the emirs of his court. Sometimes the new governor
had to establish his authority by force of arms.
On one occasion in these years there was a revolt of a more serious
character. Early in 1011 the Arab chief Mufarrij ibn Daghfal ibn
Jarrāḥ, having defeated the Caliph's representative, became ruler of inland
Palestine for the second time. He failed to occupy any of the coast-towns
but held possession of the interior for two years and five months, until
his death (1013). A peculiar feature of this revolt was the acknowledgment
by Ibn Daghfal of the sharīf of Mecca, Hasan ibn Jaʼfar, as “commander
of the faithful. ” This personage was a descendant of the Prophet and so
possessed one outstanding qualification for the Caliphate. But his only
supporter was Ibn Daghfal, and his phantom authority lasted less than
two years. Ibn Daghfal's sons were defeated by Hākim's troops im-
cH. VI.
## p. 254 (#300) ############################################
254
Ruin of the Holy Sepulchre
mediately after their father's death, and the control of Palestine passed
again to the governor of Damascus.
An event of special interest to Christendom occurred in Jerusalem
during Hākim's Caliphate, namely, the profanation and ruin of the
church of the Holy Sepulchre (commencing 27 September 1009). It is
unlikely that the fabric of the church was seriously injured. Hākim
ordered its relics to be taken away and its monuments, including the
Holy Sepulchre, to be destroyed. The portable furnishings of the church
and its treasures were carried into safety before the Caliph's agents arrived.
But the Holy Sepulchre and other venerated shrines were destroyed as
completely as possible. The interior must have been left in a very
mutilated condition. Mufarrij ibn Daghfal began the work of restora-
tion when he was ruler of southern Palestine (i. e. between 1011 and 1013).
Saʻīd-ad-Daulah of Aleppo having died early in January 1002, Lūlū
banished the surviving members of the Hamdān family to Egypt and
assumed the emirate. He acknowledged the Fățimite Caliph, Hākim,
and also continued to pay tribute to the Greek Emperor. His rule is
praised as having been wise and just. After his death (August 1009)
1 The intention of the Caliph and the extent of the destruction of the church are
to be ascertained chiefly from the narratives of Abu Ya'là (pp. 66-68) and Yahyà
(p. 195 sq. ). The former conveys the impression that a primary object of the Caliph
was to plunder the treasures of the church. His version of the Caliph's order par-
ticularly specifies the destruction of the monuments (athar, which, however, is an
ambiguous word and might signify relics). When the historian says “its structures
were ruined and uprooted stone by stone,” he refers, presumably, to such buildings
as the shrine of the Holy Sepulchre. Yahya's account would harmonise completely
with this view but for his summary statement "it was thrown down completely to
the foundations, except what could not be ruined and was too difficult to uproot. ”
If the subject of the first verb is the church, then obviously the expression “to the
foundations” cannot seriously be pressed, and the amount of damage to the walls
and fabric remains obscure. But the subject in the original is quite vague, and the
reference may be principally to the monuments and the interior structures. The
brief statements of other writers, Eastern and Western, are no proof of such a
complete destruction of the church as modern writers have assumed. The exact
date is supplied by Yahyà (Tuesday, five days from the end of Şafar, A. H. 400).
This agrees so far with the statement of Ademar (Lequien, 111, 478) that the Holy
Sepulchre was destroyed on 29 September (A. D. 1010). Yahyà (p. 201, lines 9 sqq. )
relates that Ibn Daghfal began the work of restoration; he observes particularly
that “he restored the places in it” (i. e. the shrines ? ). Possibly Hākim also
authorised its repair, under the influence of his Christian mother and uncle (cf.
Glaber). In A. H. 411 (A. D. 1020) the church is described by Yahyà (p. 230) as still
in a ruined condition. The relevant extracts from the Latin historians are given by
Lequien, Oriens Christianus, III, 475 sqq. Other references will be found in Ibn al-
athir, ix, 147, Maqrīzi, 11, 287 (Kitāb al-khițaț, Bulak edit. , 1853–1854), Abu 'l-maḥā-
sin, 11, 64 (from Adh-dhahabi), and 11, 101, Cedrenus (Corp. Script. Byzant. ), Vol. 11,
p. 456. Forty-five years before this, in A. D. 966, this church had been seriously injured
by the local Muslim and Jewish population (Yahyà, p. 125 sq. ). The dome of the
Holy Sepulchre and the roof of the adjoining main church were then both de-
stroyed, and the necessary repairs had been completed only a short time before the
events of 1009.
## p. 255 (#301) ############################################
Egypt and Syria
255
Manşūr his son, although unpopular, held the emirate for some years
against the Hamdan family and the attacks of the Bani Kilāb under
Şāliḥ ibn Mirdās. Finally he was expelled from Aleppo by an insurrec-
tion (6 January 1016), headed by the governor of the castle, Mubārak-
ad-Daulah Fataḥ, and, having escaped to Antioch, became a pensioner of
the Greeks. These events increased the authority of the Egyptians in
northern Syria. About a year later, Mubārak-ad-Daulah was made
governor of Tyre, Sidon, and Beyrout by Hākim, and ‘Azīz-ad-Daulah
Fātik, an Armenian, was installed as governor of Aleppo (3 February
1017). As so often happened in such cases, the new governor began to
act as an independent emir, and his assassination (13 June 1022)
was probably instigated by Sitt-al-mulk, Hākim's sister, now regent.
During the next two years and a half an Egyptian garrison held the
citadel of Aleppo, and a series of Egyptian governors controlled the city.
The seventh Fāțimite Caliph was Abu 'l-ḥasan 'Ali az-Zāhir. He
was a boy when he succeeded his father and he never exercised much
influence in the government of his dominions (1021-1036). For the
first three years of his reign Hākim's sister, Sitt-al-mulk, was regent.
Soon after her death the Arab tribes on the borders of Syria made a
league against the Caliph, hoping to conquer and rule the country
(1024). The leaders of the revolt were Şāliḥ ibn Mirdās, chief of the
Bani Kilāb, who lived in the neighbourhood of Aleppo, Sinān ibn
«Ulyān, chief of the Bani Kalb, near Damascus, and Hassān, a son of
Mufarrij ibn Daghfal, whose home was in southern Palestine. The con-
federates were at first successful both in Palestine and northern Syria.
Aleppo was captured by Sāliḥ ibn Mirdās (January 1025)', and Hims,
Baalbek, and Sidon soon acknowledged his authority. Thus a
dynasty, that of the Mirdāsites, was established in Aleppo (1025–1080).
In Palestine the Caliph's representative, Anūshtakīn ad-dizbirī, was more
than once defeated and was driven out of Syria. The least successful of
the allies was Sinān ibn 'Ulyān. After his death in July 1028, his
successor deserted the alliance and submitted to the Caliph. In the
following year a decisive battle was fought at Uqḥuwānah, south of Lake
Tiberias, between Sāliḥ and Hassān on the one side, and the Egyptians
and their allies on the other (14 May 1029). Şāliḥ was killed and
Hassān's power was completely broken. From now onwards Anushtakīn
was governor of Damascus and the most powerful emir in Syria
(1029-1041).
i The citadel did not surrender until five months later (June 1025). Cf. note
supra, p. 246. As illustrating the textual criticism that must always be applied to
the dates of Arabic historians, it may be pointed out that 13 Dhu 'l-qa'dah 415 in
Kamal-ad-Din's text (date of the capture of Aleppo) should be 23 Dhu 'l-qa'dah 415,
and that 1 Jumădà ii 416 (date of the surrender of the citadel) should be 1 Jumādà i
416. These are typical errors. The correct dates are given by Yahyà (pp. 246, 248),
along with the week-days, which provide the necessary test of accuracy. See
Stevenson, Crusaders in the East, appendix on Chronology.
new
CH. VI.
## p. 256 (#302) ############################################
256
The Greeks in Syria
During the period of this rebellion, in 1027 (A. H. 418), an in-
teresting treaty of peace was made between the Fāțimite Caliph and the
Emperor Constantine VIII. It was provided that the Caliph's name
should be mentioned in the public prayers of the mosques throughout
the Empire, to the exclusion of his Abbasid rival. This arrangement
was continued until the year 1056, when it was reversed at the instance
of the Turkish Sultan Tughril Beg. A further recognition of the re-
presentative character of the Fātimite Caliph, and another concession to
Islām, was contained in the provision that the Caliph might restore the
mosque in Constantinople and appoint a muezzin to officiate there. The
counterpart of these provisions gave the Emperor the right to restore
the church of the Holy Sepulchre in Jerusalem. It is not to be assumed
that the church had lain in ruins since its profanation by Hākim's
orders in 1009, nor, perhaps, that much was actually done at this
time in the way of restoration'. Another concession made by the
Caliph was that those Christians who had become Muslims by compul-
sion in the time of Hākim might again profess Christianity without
penalty. It may be assumed that the treaty of peace, as usual, was valid
for a limited period only; but the term is not specified by the only
source that mentions the treaty.
Nașr Shibl-ad-Daulah, son of Şāliḥ ibn Mirdās, was permitted to
succeed his father as ruler of Aleppo on the condition that he acknow-
ledged the Fāțimite Caliph in the customary manner, on his coinage and
in the public prayers of Friday. His emirate did not include Hims or
Hamāh, but extended north-eastward to the Euphrates. The Greeks,
who had recently been losing ground in Syria, now seized what seemed
to them an opportunity of improving their position. The territory of
Aleppo was twice invaded (1029 and 1030), both times unsuccessfully.
The Emperor Romanus shared in the second invasion, a very ill-judged
attempt. The Greek army suffered so much in the neighbourhood of
“Azāz from the hot season, lack of water, and fever that it was com-
pelled to retreat in a few days and lost heavily as it retired (August
1030). The Emir of Aleppo, reckoning his triumph an occasion of con-
ciliation and not of defiance, at once opened negotiations for peace.
A treaty was signed on terms that were distinctly unfavourable to the
Muslim city. Aleppo again became tributary to the Empire, and a Greek
deputy was allowed to reside in the city and wateh over the due per-
formance of the conditions of peace (April 1031).
1 Maqrizi (Khițaț, p. 355, lines 9 sqq. ) is here the chief authority. He does not
name the Emperor, but does mention ļāhir. William of Tyre also refers to a
restoration of the church by permission of “Daher” (= Zāhir), and Cedrenus, 11,
501, implies that the restoration was permitted by a son of the Caliph who destroyed
the church. The brief statement of Cedrenus is obviously very confused and in-
accurate (cf. infru, p. 257, n. 3). If Maqrizi were the only authority it might be con-
jectured that his date A. H. 418 was an error for A, H. 428 (i. e. A. D. 1037). See p. 257.
## p. 257 (#303) ############################################
The Greeks in Syria
257
At this date the territory of the Greeks in Syria extended eastward
from Antioch to Hārim and southwards along the coast as far as
Maraqīyah. The hillmen of the Jabal Anşarīyah, who adjoined this
territory, were partially held in check by strong castles such as Bikisrāyil,
but still maintained their independence. After the defeat of Romanus,
one of the chiefs of the hill tribes, Nașr ibn Mushraf, captured Bikisrāyil
and a general rising took place. Maraqīyah was besieged by Ibn Mushraf
and the Emir of Tripolis. Nicetas, the new governor of Antioch, took
prompt action against a very dangerous situation. He raised the siege
of Maraqiyah (December 1030), and during the next two years syste-
matically besieged and reduced the castles of the hillmen (1031-1032).
Balātunus, Bikisrāyil, and $āfīthā were among the fortresses now garri-
soned and held by the Greeks.
These events brought about a resumption of hostilities between the
Empire and the Egyptian Caliph. Anūshtakin of Damascus and the
Emir of Tyre had given a timorous support to the mountaineers in their
struggle with Nicetas. Rafanīyah was therefore attacked and captured
by Greek troops. A Byzantine fleet threatened Alexandria and the
mouths of the Nile. Both parties desired a stable peace, but the task
of settling the matters in dispute proved to be long and difficult. The
chief obstacle to a settlement was the demand of the Emperor that
Aleppo should be treated as a Greek dependency? . The negotiations
were continued, or resumed, after the death of Romanus (April 1034),
and peace was signed, perhaps in the autumn of 1037? . Each party
pledged itself not to assist the enemies of the other, and their respective
spheres of influence in northern Syria were defined. The Greek deputy
whom Romanus had stationed in Aleppo had been driven out soon after
that Emperor's death, so that Aleppo probably secured its independence.
The right of the Emperor to renovate the church of the Holy Sepulchre
was acknowledged, and possibly the privilege of appointing the Bishop
of Jerusalem. In return Michael IV set free 5000 Muslim prisoners.
The duration of the peace was fixed at thirty years. The Emperor sent
builders and money to Jerusalem, but the repairs to the church were not
completed until the reign of his successor Constantine IX 3.
1 Full details of the negotiations are given by Yahyà (p. 270 sq. ). His account
throws light on the terms of the treaty of 1037, and has been used above to sup-
plement the meagre details of that treaty preserved by others. Unfortunately
Yahya's narrative, at least as printed, breaks off before the year 1037 is reached.
? This date is derived from Yaḥyà, who seems to fix it as being three and a
half (Muslim) years after the death of Romanus. Barhebraeus gives A. H. 427, but
his narrative, on the whole, is not strongly against A. H. 428, which would agree
with Yahyà. On the other hand, Cedrenus (vol. 11, p. 515) gives A. M. 6544
(=A. D. 1035–1036) and Abu 'l-fidā (111, 96) and Ibn al-athir (1x, 313) both A. H. 429.
3 The authorities are those of the last note. The renovation of the church must
have been considerable and not merely a repair of the damage done by Hākim.
Cedrenus (11, 501) wrongly makes Romanus commence the work and Michael com-
17
C. MED. H. VOL. V. CH. VI.
## p. 258 (#304) ############################################
258
Caliphate of Mustanșir
The eighth Fāțimite Caliph, Abu tamīm Maʻadd al-Mustanşir,
was only seven years old when his father died (June 1036), so that his
reign began with a succession of regencies. The Caliph's mother, an
African woman, exercised a considerable amount of influence. The con-
temporary Persian traveller Nāşir-i-Khusrau records very favourable
impressions of the prosperity and tranquillity of the country while the
Caliph was a minor.
Early in this reign peaceful relations between Aleppo and Egypt
were broken off. Nașr ibn Şāliḥ was defeated and slain in battle with
Anūshtakīn (May 1038), and Aleppo was captured and garrisoned by
Egyptian troops for a few years (1038-1042). The disgrace of Anushtakin,
followed immediately by his death (January-February 1042)', weakened
the Fāțimite dominion all over Syria. Aleppo was recovered by Nașr's
brother, Mu'izz-ad-Daulah Thumāl (March 1042). He resumed payment
of tribute to the Greeks and so secured himself in that direction. The
terms of the rulers of Egypt were not so easily satisfied. Envoys came
and went between the parties. Attacks were launched against Thumal
by the Emirs of Hims and Damascus, acting in the name of the Caliph
(1048–1050). At length, in 1050, an agreement satisfactory to both sides
was arrived at.
Two isolated events, which are a part of the history of the Fāțimite
Caliphs, deserve mention here. In 10492 Mu'izz ibn Bādīs, the Zairite
Emir of Tunis, ceased to pay tribute to Mustanşir and transferred his
allegiance to the Abbasid Caliph. His family had ruled in Qairawān,
in practical independence, since 973, when the Fāțimite Caliph of the
day made Cairo his residence and capital. But the formal separation,
signalised by the acknowledgment of the Caliph of Baghdad, took place
only now. On the other hand, for the greater part of the year 1059 the
Caliphate of Mustanşir was acknowledged in Baghdad itself. Such
acknowledgments were now symbols of the triumph of political parties
and alliances. The Turkish Sultan Țughril Beg identified his cause with
that of the Abbasid Caliphs, with the result that his enemies in
plete it. William of Tyre, also, says that Romanus received permission to restore
the church. Possibly the explanation of these statements is that during the nego-
tiations of 1031-1034 the article regarding the restoration of the church was agreed
to (so Yaḥyà), although the treaty of which it was part was not signed until the
reign of Michael. William of Tyre seems to imply that not much progress was
made with the repairs until the reign of Constantine. He gives 1048 as the date
when the work was completed; similarly Abu 'l-fidā (A. H. 440). There was a severe
earthquake in Jerusalem in January 1034 (Yaḥyà, p. 272; cf. Cedrenus, 11, p. 511),
which may possibly have caused some of the injury that was afterwards repaired.
1 1042 according to Kamāl-ad-Din and Ibn al-athir, 1041 according to Abu
Yaʻlà.
2 So Ibn 'Adhāri (A. H. 440), who further states that Ibn Bādis struck new
coinage in Shaʻbān 441 (end of 1049 or beginning of 1050). Ibn al-athir is incon-
sistent with himself, giving both A. H. 435 and A. H. 440. Abu 'l-maḥāsin gives both
A. H. 435 and A. H. 443, preferring the latter.
## p. 259 (#305) ############################################
Emergence of the Turks
259
Mesopotamia were disposed to favour recognition of the Fāțimite Caliphs
in those districts and cities where they triumphed. In 1059 Baghdad
was occupied by a Turkish emir, Arslān al-Basāsīrī, who, being an enemy
of the sultan, acted in the manner just des ped. The occasion was
hailed in Egypt as an extraordinary triumph, and in fact probably
marked the highest point of superiority to the Abbasids ever reached
by the Fāțimite Caliphs.
When Mustanşir came of age he shewed such feebleness and in-
capacity that he was treated by all parties as a cypher in the government.
The ministry of Hasan al-yāzūrī (1050-1058) was still, on the whole,
prosperous and considerate of the general welfare. But after his death
there recommenced a bitter struggle for power between the leaders of the
Turkish and those of the negro troops. The country was devastated and
impoverished by civil war, and finally lay at the mercy of the unscrupulous
and cruel Turkish leader Nāsir-ad-Daulah ibn Hamdān (1062-1073). Pro-
longed drought and famine increased the miseries of the unhappy people.
The influence of Egypt upon foreign affairs fell to its lowest ebb. It was
in no way able to share in the defence of Syria against the Seljūq Turks.
The rule of Mu'izz-ad-Daulah Thumāl in Aleppo was mild and
generous, and therefore popular. His greatest troubles were caused by
the unruly Arabs of the district, the Bani Kilāb, and latterly by the
Seljūq Turks, already planted at Raḥabah on the Euphrates. In January
1058, feeling no longer equal to the tasks of his position, he abdicated
and left an Egyptian governor and garrison once more in power. These
were soon expelled by the citizens assisted by the Bani Kilāb (September
1060), and shortly afterwards Mu'izz-ad-Daulah was persuaded to return
to his former post (April 1061). During his second brief emirate the
Greeks provoked hostilities by repairing some border castles, and Artāḥ
was taken from them. Peace with them was renewed during the civil
war that followed Mu'izz-ad-Daulah's death (November 1062). Artāḥ
appears to have returned to its former owners.
Thumāl's brother, Asad-ad-Daulah “Atīyah ibn Şāliḥ, was his suc-
cessor. His title to succeed was challenged by a nephew, Maḥmūd ibn
Naşr, and the brief period of his emirate was one of civil war (1062–1065).
It was at this date, just before the Norman conquest of England, that
the Seljūq Turks entered Syria.
From the ninth century onwards, Turkish governors and Turkish
generals and Turkish mercenaries play an important part in the history
of Syria and especially of Egypt. The Tūlūnites were a Turkish family
and were served by Turkish officers and soldiers. So also were the Ikh-
shīds. In Mesopotamia, from which these viceroys came, Turkish slaves
held the highest place, subject only to the nominal authority of the
Caliphs. In Egypt the Fāțimite dynasty retained and added to the
Turkish household troops of their predecessors. Turkish, Berber, and
Negro factions struggled for supremacy, and the Fāțimite governors of
CH. 1.
17-2
## p. 260 (#306) ############################################
260
Turkish conquest of Syria
Syrian towns in the tenth and eleventh centuries were often Turkish
Mamlūks.
Before the middle of the eleventh century, a new wave of Turkish
migration, under the great Sultan Țughril Beg (1037–1063), swept into
Lower Mesopotamia from the north and threatened Armenia and Upper
Mesopotamia. It was the precursor of the conquest of Syria by the
Seljūq Turks. The manner of their conquest is representative of many
other periods in Syrian history. Bands of horsemen, a few hundred
strong-seldom as many as a thousand—rode under adventurous leaders
who sought their fortune and lived by their swords. They took service
with any ruler for money or for lands, and gained their chief advantage
where local feuds were being waged. Some novelty in their arms or in
their way of fighting might give them an advantage in battle. In any
case they were always on the war-path, and so could finally wear down
the resistance of cities which depended upon the cultivation of the land
or upon peaceful industry. The inland towns of Syria -- Aleppo, Hims,
Baalbek, Damascus, Jerusalem-yielded first and most completely to
the Turks. Once established, the way of the conquerors was smoothed by
their being Muslims. Their introduction of the nominal authority of the
Caliphs of Baghdad was almost a matter of indifference to their subjects.
The rule of Turkish emirs was already familiar in Syria. The invaders
were backed by the prestige of the Seljūq sultans, but only to a slight
extent occasionally by their armies.
A conquest of the character just described implies, of course, that
Syria was in its normal state of political disintegration. It was, in fact,
even less united than it had been for some time past. Aleppo was an
independent territory and was rent by civil war. The Arabs hung loosely
on the borders. The hillmen of the Jabal Anşarīyah took no interest in
the fate of the neighbouring plains. Antioch and its dependencies were
under the rule of foreigners. Damascus and the coast towns from Tripolis
southwards had cut themselves adrift from Egypt, which was in the
throes of revolution. They were governed by independent emirs, anta-
gonistic to one another. Only the south-west of Palestine was still closely
attached to Egypt. After the great defeat of the Greeks at Manzikert
(1071), Antioch was almost left to its own resources. Even the Arme-
nians, who had long given soldiers to the Greeks on the eastern borders
of the Empire and in Syria, now preferred to make terms with the
Turks.
Hārūn ibn Khān was the first of the Seljūq Turks to gain a footing
in Syria. About the end of 1064 he and his thousand followers turned
the scale in favour of 'Ațīyah ibn Şāliḥ against his rival Maḥmūd. When,
however, ‘Ațīyah and the citizens of Aleppo rose against their deliverer
and massacred his followers, he made off with the survivors to Maḥmūd
and helped him to victory at the battle of Dābiq (16 June 1065). After
the surrender of Aleppo to Maḥmūd (13 August 1065), Hārūn was given
## p. 261 (#307) ############################################
Recognition of the Abbasid Caliphs
261
the little township of Ma'arrat-an-Nuʻmān in fief, and settled there
with a mixed following of Turks, Kurds, and Dailemites.
In the summer of 1067 another Turkish leader, ‘Afshin by name,
raided the territory of Antioch and carried off great booty. His prisoners
were so many that “a girl was sold for two dinars and a boy for a set of
horseshoes. ” In the following year ‘Afshin besieged Antioch and was
bought off by the payment of a large sum of money (1068). At the
same time there was war between Aleppo and Antioch, and Artāḥ was
captured by Hārūn ibn Khān after a five months' siege (1068). In the
following year a Greek army, under the Emperor himself (Romanus
Diogenes), recovered Artāḥ and captured Mambij. Before the close of
the year the Armenian governor of Antioch (Kachatur) made peace with
Mahmud on terms that were favourable to the latter.
In 1070 a Turkish leader, known as Zandiq, entered Syria at the
head of large forces and ravaged the territories of Aleppo, Hamāh, Hims,
and Rafanīyah. This was the first devastation of Muslim Syria by the
Turks. It decided Maḥmūd to seek the protection of the Sultan Alp
Arslān (1063-1072), and at the same time, in consequence, to transfer his
allegiance from the Fāțimite to the Abbasid Caliph. Prayers were said
in the mosques of Aleppo for the new Caliph and for the sultan on Friday
30 July 10702
Alp Arslān now demanded that Mahmūd should engage in war with
Antioch and with the Fāțimite emirs. Maḥmūd having at first refused,
the sultan invaded Syria (spring of 1071). Two months were spent in
negotiations, and during another month Aleppo was blockaded. Then
Mahmud submitted and became the sultan's vassal. The historian of these
events comments especially upon the discipline of Alp Arslān's army.
The persons and the property of the country people were respected. Even
the forage that the soldiers used was often paid for. Aleppo was neither
ruined nor pillaged. Fasdiq, where Alp Arslān pitched his tent during
the expedition, was henceforth known as the Sultan's Hill (Tell-as-sulțān).
Maḥmūd does not seem to have shewn much zeal in the fulfilment of his
pledge to the sultan during the remainder of his emirate (ob. 10 January
1074). His sons Nașr (1074-1076) and Sābiq (1076-1080) were the last
of the Mirdāsites to rule Aleppo. Fresh bands of Turks were pouring into
Syria. Rafanīyah was occupied by Jawāli ibn Abaq (1075), who raided
the territory of Aleppo until he was severely defeated by Aḥmad Shāh,
another Turkish leader, in the service of Nasr ibn Maḥmūd and after-
wards of his brother Sābiq. The assassination of Nașr and the acces-
sion of Sābiq illustrate the influence now exercised by the Turks over
the internal affairs of Aleppo. Sābiq was opposed by two of his brothers
and by the Bani Kilāb, but defeated his enemies with the help of Aḥmad
1 Fifteen dinars may have been a normal price.
? The same change was made in Mecca in this same year. But the acknowledg-
ment of the Fățimites was resumed there again in 1074 or 1075.
CH. VI.
## p. 262 (#308) ############################################
262
The Turks in Palestine
Shāh and other Turks (July 1076). Nașr and Sābiq both waged war inter-
mittently with the Greeks. In 1075 Mambij was recovered by the former.
The principal Seljūq emirs of the north of Syria were ‘Afshin, Zandiq,
and Muḥammad ibn Dimlaj. In the summer of 1077 they were ordered
by Alp Arslān's successor, Malik Shāh (1072-1092), to unite under the
command of his brother Tāj-ad-Daulah Tutush. In the spring of 1078
Tutush attacked Aleppo at the head of a large force, which included the
Bani Kilāb and the soldiers of Sharaf-ad-Daulah Muslim of Mosul (1061-
1085). The siege lasted four months and its failure was attributed to the
action of Sharaf-ad-Daulah, an old ally of the Turks, who was now
turning against them. Next year (1079) Tutush resumed his operations
in Syria, with some success. Mambij, Buzā‘ah, and other places sur-
rendered or were captured. Then an invitation from the Turkish Emir
of Damascus, At-siz ibn Abaq, drew his attention southwards.
The first mention of the presence of Seljūq Turks in Palestine be-
longs to the year 1070". The authority of Nāșir-ad-Daulah, governor
of Egypt, did not extend at that time beyond the south of Palestine.
Acre and Sidon were governed by an Armenian, Badr-al-jamālī, who
had played a prominent part in Syrian affairs since 1063. Damascus,
Tyre, and Tripolis were in the hands of other independent emirs. The
Arab tribes on the southern and eastern borders were their own masters.
After the assassination of Nāșir-ad-Daulah (10 May 1073), Mustanşir
appealed to Badr-al-jamāli to end the régime of the Turkish slaves in
Egypt. At the head of his Syrian troops he occupied Cairo (February
1074), and in a few years restored unwonted peace and order to the
country. He was the all-powerful ruler of Egypt for twenty years
(1074–1094).
Several Turkish leaders shared in the conquest of southern Syria, but
they all, in a measure, seem to have obeyed At-siz ibn Abaq? His
first acquisition was ‘Ammān, an Arab stronghold in the Balqā (1071? )*.
From there he became master of the south of Palestine, including Jeru-
salem and Ramlah. Jerusalem capitulated on terms, and suffered nothing
from its change of rulers. For several years At-sīz, having marked Da-
mascus as his prey, ravaged its territory, especially at harvest time, and
levied contributions from the coast-towns as the price of their immunity.
In 1075 he captured Rafanīyah and gave it over into the charge of
his brother Jawālī. In the summer of 1076 Damascus at last sur-
rendered to him. After this he ventured to invade Egypt and was
severely defeated in the neighbourhood of Cairo (January 1077). His
1
A. H. 462 (Ibn al-athir, x, 40; Kamāl-ad-Din, p. 77).
? His name first appears in the year 1071 (A. H. 463). Qarlū is the only leader
mentioned by name in the previous year (A. A. 462).
3 Quatremère, p. 413 f.
4 The date is from Sibt ibn al-Jauzi. The general sequence of events suggests a
year later. Cf. infra, p. 263, n. 1.
## p. 263 (#309) ############################################
Turkish dissensions
263
bold challenge prompted Badr-al-jamāli to seek the recovery of Palestine
and Damascus. At-siz, fearing the issue of the conflict he had provoked,
invited Tāj-ad-Daulah Tutush to his aid. The result might have been
expected. Tutush took possession of Damascus and put At-sīz to death
(September 1079)". Badr-al-jamālī withdrew his forces from Palestine.
The emirs of the coast-towns, for the most part, paid tribute to Tutush
rather than submit to their ancient rival, the governor of Egypt.
Finding himself secure in Damascus, Tutush at once sent most of his
army back into northern Syria. “Afshīn, his general, laid waste the
country from Baalbek to Aleppo and ravaged the territory of Antioch.
In consequence of this attack Sabiq and the citizens of Aleppo surrendered
the town to Sharaf-ad-Daulah Muslim of Mosul (June 1080). Sābiq re-
tired to Raḥabah, and Muslim and Tutush stood opposed as well-matched
antagonists.
As matters turned out, there was little actual fighting between the
rivals. For two or three years Muslim strengthened his position in
northern Syria and Upper Mesopotamia, held communications with Badr-
al-jamālī, and sought to divert the tribute of Antioch from the sultan to
himself.
