Jump to ContentJump to Main Navigation
Turkish Myth and Muslim SymbolThe Battle of Manzikert$

Carole Hillenbrand

Print publication date: 2007

Print ISBN-13: 9780748625727

Published to Edinburgh Scholarship Online: September 2012

DOI: 10.3366/edinburgh/9780748625727.001.0001

Show Summary Details

The twelfth-century accounts of the battle of Manzikert

The twelfth-century accounts of the battle of Manzikert

Chapter:
(p.26) Chapter 2 The twelfth-century accounts of the battle of Manzikert
Source:
Turkish Myth and Muslim Symbol
Author(s):

Carole Hillenbrand

Publisher:
Edinburgh University Press
DOI:10.3366/edinburgh/9780748625727.003.0002

Abstract and Keywords

This chapter gives translations of the earliest surviving Muslim accounts of the battle of Manzikert dating from the twelfth century (five in Arabic and one in Persian). These narratives, which are written by authors originating from Spain, Syria, eastern Turkey and Iran, are presented in chronological order. Each of these translations is accompanied by a detailed commentary, analysing their content, style and the specific context in which the author is writing.

Keywords:   Muslim historiography, Arabic, Persian, Al-Turtushi, Ibn al-Qalanisi, Al-'Azimi, Ibn al-Azraq, Nishapuri, Ibn al-Jawzi

Alp Arslan was the first of the kings of the Turks to cross the Euphrates.1

The account of al-Turtushi (d. 520/1126) in Sirāj al-mulūk

Introduction to the text

Al-Turtushi was a leading religious and intellectual figure of his time. In 476/1084 he went east, as was frequently the custom with aspiring scholars from al-Andalus, and after performing the pilgrimage, he travelled widely in the Levant before finally settling in Alexandria. In the course of his travels, he met Ibn Tumart, al-Ghazali and other famous Muslim scholars and leaders.2 It is interesting to note that he had contact with teachers at the famous Nizamiyya madrasa in Baghdad where he established himself in 478/1085,3 and he might even have encountered Alp Arslan's vizier, Nizam al-Mulk, who by then was working for Malikshah, Alp Arslan's son and heir to the Seljuq sultanate. Even if the two men never met each other, al-Turtushi would have had access to stories and information from circles close to the Seljuqs and he would have heard about the battle of Manzikert, possibly even from eye-witnesses quite soon after the battle, when memories were fresh. It is clear that al-Turtushi admired Nizam al-Mulk, eulogising at length in Chapter 48 his remarkable skills in governing, and he talks in particular about his achievements in setting up the network of Nizamiyya madrasas throughout the Seljuq realm.4

The account of al-Turtushi is apparently the earliest extant narrative about the battle of Manzikert in the Islamic sources. It is therefore of key interest. Yet it has been overlooked by all scholars so far who have worked on the Arabic and Persian accounts of the battle, such as Cahen, Vryonis, Zakkar, and Sevim and Sümer. Perhaps the cause of its (p.27)

The twelfth-century accounts of the battle of Manzikert

Figure 2.1 Modern Turkish depiction of Alp Arslan

neglect is that it lies buried in an unusual place – the major work of al-Turtushi, Sirāj al-mulūk,5 completed in Fustat in 512/1122 and dedicated to the Fatimid vizier al-Maʿmun b. al-Bataʿihi. This is a very long Mirror for Princes, and not a town chronicle, dynastic or universalhistory. In this book of sixty-four chapters addressed to kings and rulers, al-Turtushi includes many moralising anecdotes. Chapter 61, entitled An account of the management, stratagems and rules of war, is devoted to a discussion of the stratagems of war and advice on how to conduct it well, and it is in this context that he provides an account of the battle of Manzikert.6

The translation

By this strategy,7 Alp Arslan, the king of the Turks, conquered and subdued the king of Byzantium, killed his men and destroyed his troops. The Byzantines had assembled armies the like of which were seldom gathered for anyone after him. The total of their number was six hundred thousand warriors – self-contained battalions, successive troops and squadrons following one after the other, [so numerous] that (p.28) the eye could not perceive them and their number could not be quantified. They had prepared an innumerable amount of animals, weapons and mangonels8 and pieces of equipment made ready for conquering citadels in war. They had divided up the countries of the Muslims – Syria, Egypt, Iraq, Khurasan and Diyar Bakr – and they did not doubt that the wheel [of fortune]9 had turned for them and that the stars of good fortune were in their favour.

Then they turned towards the lands of the Muslims. News of them kept coming and the countries of Islam were disturbed because of that. Alp Arslan the Turk mobilised troops to meet them; he is the one who is called the just prince (al-malik al-ʿādil). He assembled his troops in the city of Isfahan and he prepared what he could. Then their day [of battle] came. The two armies kept on coming closer to each other until the vanguards of the Muslims returned to the Muslims and said to Alp Arslan: ‘Tomorrow the two armies will [be able to] see each other.’ So the Muslims passed the night of Friday [i.e. Thursday] whilst the Byzantines were in a number which nobody except He who had created them could enumerate and the Muslims had nothing with them except gnawing hunger.10 The Muslims remained silent with fear about what had befallen them.

When they got up on the Friday morning they looked at each other. What the Muslims saw of the great number, strength and equipment of the enemy terrified them. Alp Arslan ordered that the Muslims should be counted and they amounted to twelve thousand Turks. There they were like the mark on the leg of the ass [i.e. a tiny speck].11 So he [Alp Arslan] assembled those possessed of judgement from amongst the men of war, of administration and care for the Muslims and insight into the consequences, and he sought their counsel about how to achieve correctness of judgement. They consulted amongst themselves for a short while. Then their opinion was agreed on meeting [the enemy in battle].

They made peace with each other, swore oaths to each other and showed sincere intentions towards Islam and its people. Then they made preparations for battle and they said to Alp Arslan: ‘We will invoke the name of God Most High12 and we will attack the people.’ Alp Arslan said: ‘O assembly of the people of Islam! Tarry [a while], for this is Friday and the Muslims are delivering the sermon on the pulpits, and praying for us in the east and the west of the lands. When the sun has set and the [evening] shadows have returned and we know that the Muslims have performed the prayer, and prayed for us and we ourselves have prayed, we will carry out our affair.’

(p.29) They waited patiently until the sun had set, then they performed the prayer and prayed to God Most High that He would make His religion triumph, that He would reinforce their hearts with patience, that He would enfeeble their enemy and that He would cast fear into their hearts. Alp Arslan had verified [the position of] the tent, banner, horse and outward appearance of the king of Byzantium. Then he said to his men: ‘Let each of you not fail to do as I do and strike with his sword and shoot with his arrow where I strike with my sword and shoot with my arrow.’ Then all of them launched an attack as one man on the tent of the king of Byzantium. They killed those who were in front of it and they reached him. Those round him were killed and the king of Byzantium was taken captive. They began shouting in the language of Byzantium: ‘The king has been killed! The king has been killed!’ The Byzantines heard that their king had been killed and they scattered and were totally torn to pieces. The sword was active amongst them for days, and the Muslims took their possessions and their spoils.

The king of Byzantium was brought into the presence of Alp Arslan with a rope round his neck. Alp Arslan said to him: ‘What would you have done with me if you had captured me?’ He said: ‘Do you doubt that I would have killed you?’ So Alp Arslan said to him: ‘You are too trivial in my view for me to kill you. Take him and sell him to the person who pays most.’ So he was led with the rope round his neck and a proclamation was made about him: ‘Who will buy the king of Byzantium?’ They went on like that going round with him to the tents and the Muslims' houses and the announcement for him was made in dirhams and fulūs [i.e. small coinage]. Nobody paid anything for him until they sold him to a man for a dog. The person, who was given the charge of brokering that on his [the sultan's ] behalf, took the dog and the king and brought them both to Alp Arslan and said: ‘I have been round the whole camp and made a proclamation about him and nobody spent anything on him except a single man who paid me a dog for him’ (pl. 3).

He [Alp Arslan] said: ‘That is just, because the dog is better than he is! Take the dog and give this dog [i.e. Romanus] to him.’ Then after that he ordered him to be released. He [Romanus] went to Constantinople. The Byzantines deposed him and blinded him with fire.

See what happens to kings when in wars they know about strategy and the intentional use of artifice.

(p.30) Commentary on the text

In comparison with the other accounts of the battle which follow, the version of al-Turtushi is highly idiosyncratic. Some of its narrative elements are not to be found elsewhere in the corpus.13 Other aspects of the account were to be taken up and elaborated by later writers. It is remarkable that there is no mention of the place of the battle. Nor is there any sense of the sequence of fighting. Everything is over quickly because of Alp Arslan's skill in capturing the emperor and the subsequent disarray amongst his demoralised troops. The account of the battle is telescoped; it mentions no preliminary skirmishes between the two armies and it has no distinct phases or evolution. It is over quickly because of Alp Arslan's skill in capturing the emperor.

The account of Ibn al-Qalanisi (d. 555/1160)14 in Dhayl tarʿīkh Dimashq

Introduction to the text

Ibn al-Qalanisi was an important figure in Damascus, serving there as mayor on two occasions. He wrote a well-known chronicle which focuses primarily on the history of his own city. The work follows a strictly annalistic format. The sources on which Ibn al-Qalanisi drew for his short account of the battle of Manzikert are unknown but it is likely that he had access to the work of the Baghdadi historian, Ghars al-Niʿma b. Hilal al-Sabiʾ [d. after 469/1077].15

Translation of the text

Alp Arslan left there [Aleppo] on 23 Rajab [463/26 April 1071], heading for the lands of Byzantium, seeking their king. Romanus16 had made his way to Manzikert,17 he reached it, fell upon it and vanquished it. His troops, according to what was related, amounted to six hundred thousand Byzantines and other additional contingents. The troops of Islam, according to what was mentioned, amounted to four hundred thousand from amongst the Turks and other contingents. Many of the Byzantine troops were killed, to such an extent that a valley there where the two sides had met was filled [with corpses]. The king fell prisoner into the hands of the Muslims. Hands were filled with their baggage and possessions, their equipment and their animals.18 Messages kept on going back and forth between the sultan Alp Arslan (p.31)

The twelfth-century accounts of the battle of Manzikert

Figure 2.2 Modern Turkish depiction of a mounted Alp Arslan leading his troops

and the captured king of Byzantium until it was established that he would be released and that he personally would be well treated, after [his] taking oaths and covenants that he would stop opposing any of the territories of Islam and that he would release prisoners. He was set free and sent to his country and the people of his kingdom. It is said that they seized him and handed him over [to his enemies within Byzantium].19 They appointed someone else to his position because of things for which they criticised him and accused him.

(p.32) Commentary on the text

The account mentions Alp Arslan as being in Syria just before the episode of Manzikert, besieging and taking Aleppo in 463/1071 and then appointing its Mirdasid ruler as his governor of the city.20

The version of Ibn al-Qalanisi has a second-hand feel to it. It is, however, significant that he includes it at all in his chronicle, which is so focused on his own city of Damascus. He gives a precise year for the events, but the number that he gives for the troops differs from those included in other accounts of the battle, which stress a much wider disparity between Romanus' and Alp Arslan's armies. The size of both armies is grossly inflated and there is mention of additional troops in Romanus' army. Ibn al-Qalanisi emphasises the enormous amount of booty taken by the Seljuq army, he mentions the drawing up of a treaty, and alludes rather cryptically to the final fate of Romanus when he arrived back in Byzantium. The account is rather flat and it lacks any triumphalist tone, although the Seljuq army is already called ‘the troops of Islam’. No attempt is made in this narrative to describe a military engagement, nor is there any reference to the battle having taken place on a Friday or to extract from the tale any other moral for the Muslim faithful. It is difficult to assess whether this account is fragmentary because the author lacked more detailed information or because the battle, although worthy of mention, was outside the focus of a local town chronicle. What is clear, however, is that the historiographical potential of the narrative, which will recur in later accounts, is already established here: the superior numbers and mixed nature of the Byzantine army, the victory of Alp Arslan, the capture and release of Romanus and his subsequent fate in Byzantium. It is noteworthy that the date of the battle is not mentioned. The style of the passage is unpretentious. The account, short as it is, focuses rather on the events before and after the encounter. Above all, the notion of the battle as a hinge of history is simply not there.

The account of al-ʿAzimi (d. c. 556 /1161), Taʾrīkh Ḥalab

Introduction to the text

Only one of the two known works of al-ʿAzimi, a rather neglected chronicler, is extant – a world history until 538/1143–4 seen from the perspective of Aleppo.21 The work is fragmentary but contains useful accounts, including occasional references to Anatolia.

(p.33) Translation of the text

Alp Arslan marched from there [Aleppo], aiming to meet Diogenes, the king of Byzantium, because he had ravaged the land. Alp Arslan met him in the environs of Manzikert. The sultan defeated him, took him prisoner and sold him for a dīnār. The sultan released him, sent him back to his country and Byzantium blinded him.

Commentary on the text

This narrative is laconic, even by the usual standards of the text of al-ʿAzimi. It is included here for the sake of full coverage of the Muslim historiographical tradition about Manzikert. That said, the account is not totally devoid of interest. No doubt because of Alp Arslan's presence at Aleppo during the time immediately preceding his journey to Manzikert, al-ʿAzimi had found reports about the battle; he is one of the few Muslim authors who reproduces correctly the second half of the Byzantine emperor's name – Diogenes – and he knows about the ultimate fate of the emperor after his return home. However, no date is given and no details of the battle are provided.22

The account of Ibn al-Azraq al-Fariqi (d. after 572/1176–7) in the Taʾrīkh Mayyāfāriqīn wa-Āmid23

Introduction to the text

Ibn al-Azraq worked as an administrator for Temürtash (ruled 518/1124–548/1154), the second Artuqid ruler of Mayyafariqin. His town chronicle has an annalistic structure, despite his often inaccurate dating. His account of Manzikert is especially interesting since it comes from an area geographically close to the battle. In an earlier passage, Ibn al-Azraq mentions that the Byzantine emperor had come to Manzikert in 463/1071. Unlike most Muslim chroniclers, who place Alp Arslan in Aleppo before the battle, Ibn al-Azraq notes that the sultan was in Iraq at the time and that he then made for Diyar Bakr.24

Translation of the text

It was reported:

Then the sultan heard that the king of Byzantium had come back.25 So Alp Arslan went down to Mosul. A large group of the people of (p.34) Akhlat26 and Manzikert went down after him, informing him that the king of Byzantium had come back to the country. So the sultan returned and went up to Arzan27 and Bitlis.28 With them was the qāḍī of Manzikert. Alp Arslan came to Akhlat, took possession of it and stayed there some days. Then the king of Byzantium came to the province of Manzikert. So the sultan went out and marched and encamped before the gate of Manzikert. Letters began to go back and forth between the two of them. The king of Byzantium was with an innumerable number of people. Ibn al-Muhallaban29 went from the sultan's presence to the king of Byzantium. Romanus asked him about the country and its condition and he said: ‘Tell me: which is better, Isfahan or Hamadhan?’ Ibn al-Muhallaban said: ‘Isfahan.’ Romanus said to him: ‘We have heard that Hamadhan is extremely cold.’ Ibn al-Muhallaban said: ‘It is so.’ The king said: ‘As for us, we will winter in Isfahan and the riding animals will be in Hamadhan.’ Ibn al-Muhallaban said to him: ‘As for the riding animals, it is true that they will winter in Hamadhan. As for you, I do not know.’ Then he left him and they met up to fight. The Byzantines set up their lines with three hundred thousand cavalry whilst the sultan had [only] a small troop. The time for fighting became pressing. It was Friday, towards the time when the sultan knew that the preacher would be on the pulpits. The time for his attack approached and he said to the people: ‘Charge’, so they all attacked and pronounced the takbīr.30 The sultan said: ‘This is the time of the prayer on all the pulpits for the armies of the Muslims, and the rest of the people are saying amen to their prayers. So perhaps God will answer [the prayer] of one of them.’ Then they charged and said the takbīr and God gave them the victory. The king of Byzantium was put to flight and a great number of his followers were killed. They [the victors] plundered their possessions to such an extent that they distributed amongst themselves the gold and silver in raṭls.31 The inhabitants of Akhlat and Manzikert plundered from their [the Byzantines'] possessions enough to keep them rich until now, for they went out, stayed with the army, fought and took most of the plunder. From that year the people of Akhlat were rich and became possessors of wealth. The sultan returned to Azarbayjan, having appointed a governor in Akhlat and Manzikert.

Commentary on the text

This is a relatively early account of the battle of Manzikert. It is fairly short and unvarnished. In some ways it is rather idiosyncratic. It too is inserted without explanation into the middle of a chronicle concerned (p.35) with the minutiae of the history of a city, this time Mayyafariqin, now called Silvan, not far from the site of the battle in eastern Turkey. Unlike other narratives, it emphasises plausibly the role played by local people in the battle, their participation in the fighting and their gaining vast wealth from the booty taken from the Byzantines. Their unexpected wealth had become the stuff of local folk legend by the time of Ibn al-Azraq a century or so later.

Already in this account there are signs of literary embellishments. Ibn al-Azraq reports an alleged conversation between Romanus and the envoy of Alp Arslan; in it Romanus makes veiled but arrogant threats to invade Seljuq territory in Iran. In the ensuing repartee Ibn al-Muhallaban has the last defiant word, which the readers of this account would relish with their retrospective knowledge of the outcome of the battle. Already, too, a religious dimension is given to this battle. It is shown to have taken place on a Friday, although no date for it is given, and God answers the prayers of His community by granting them victory. It should be noted that Ibn al-Azraq had access to the circles of religious scholars in Baghdad and he may well have been in contact with both caliphal and sultanal circles.32 Perhaps surprisingly, no ideological capital is made from the subsequent fate of the Byzantine emperor who has dared to attack the Muslim world. In fact there is no reference at all to what happened when he returned home.

The account of Nishapuri (d. c. 582/1186–7) in the Saljūqnāma33

Introduction to the text

Nishapuri was probably employed as a tutor to a Seljuq prince or princes. His short chronicle, a dynastic history, is the foundation for most of the subsequent histories of the Seljuqs written in Persian. It was written during the reign of the Seljuq sultan Tughril, the last ruler of the dynasty in Iran, and completed some time before 581/1186. In spite of its brevity,34 it is ‘the most important single Persian narrative source for the history of the Seljuqs’.35 Nishapuri's version of the battle of Manzikert is the earliest extant in Persian.

Translation of the text

He [Alp Arslan] went to wage ghazā36 against the king of Byzantium, Romanus. He [Romanus] left Amid with three hundred (p.36) thousand horsemen from Byzantium and he made for the lands of Islam. Alp Arslan came to him at Manzikert and he defeated him with twelve thousand men. They [the victorious Seljuq army] recited the verse: ‘How often a little company has overcome a numerous company by God's name.’37 Romanus was taken captive at the hand of a ghulām.38 It has been reported39 that the sultan Alp Arslan, at the time when he was waging ghazā against the king of Byzantium, requested an army inspection in Baghdad in front of him. The commander Sa‘d al-Dawla Gawhara’in40 was in his service and he inspected his troops. There was a ghulām, extremely puny, a Byzantine. He came on parade [but] the inspector41 did not write down his name. Sa'd al-Dawla said: ‘Write [it down]. It may be that he will capture the king of Byzantium himself’. It so happened that this ghulām recognised the king of Byzantium during the rout [of the Byzantine army] because he had seen him in Byzantium. He seized him and took him prisoner before the sultan. He [Alp Arslan] held him [Romanus] prisoner for a few days. After that he put a ring in both his ears42 and guaranteed him his life. He established with him [the amount of] a thousand dīnārs per day that he should pay as poll tax.43

Commentary on the text

This account is of value because it provides a very early version of the battle of Manzikert. It probably originates from Hamadhan in western Persia which was the centre of the Great Seljuq empire in its last phase. The narrative of Nishapuri is the oldest surviving account of the battle of Manzikert in Persian, the court language of the Seljuqs. Despite its unvarnished and staccato tone, the text of Nishapuri contains a number of narrative elements which were to become part of the received version of the battle in medieval Muslim sources – the discrepancy in the size of the two armies; a slave – whose physical appearance clearly did not pass muster – taking the emperor prisoner; and the sultan exacting an enormous payment from the emperor before setting him free. However, Nishapuri, despite the brevity of his narrative, points out that the slave was Byzantine, which is presumably the reason why he could recognise the emperor. Later sources produce a more elaborate version of the story which involves Alp Arslan seeking to verify the identity of the prisoner by consulting an envoy of his who had been sent on a mission to Romanus and could therefore recognise him.

(p.37) The account of Ibn al-Jawzi, (D. 597/1200) in Al-Muntaẓam fī taʿrīkh al-mulūk waʾl-umam44

Introduction to the text

Ibn al-Jawzi was a very famous Hanbalite scholar, preacher and prolific writer who lived in Baghdad. His work, the Muntaẓam, a valuable source for the history of the caliphate, often resembles a biographical dictionary more than a chronicle and it provides obituaries, mostly of scholars, under each year. His account of the battle of Manzikert is an unusually long excursus away from his usual focus on Baghdad. It became the authoritative source for many later chroniclers who wrote about the battle of Manzikert.

Translation of the text

Then the year 463/1071 began.

Amongst the events in it was that the sultan received the news about the king of Byzantium assembling numerous troops and marching towards the Islamic lands. The sultan was with the scattered remnants of the army45 since they had gone back from Syria to Khurasan in disarray because of the high prices which had depleted46 their wealth and they had set out to return to their centres. The sultan remained with around four thousand ghulāms. He did not think it wise, despite that [small number of troops], that he should return to his lands nor did he assemble [all] his troops, for that would be a calamity for Islam. He preferred to wage holy war (ghazā) and stand fast in it [even with only his small force of men].

So he sent the Khatun al-Safariyya47 and Nizam al-Mulk and the heavy baggage to Hamadhan and he ordered him to collect troops and to send them to him. And he said to him [Nizam al-Mulk] and the leaders of his army: ‘I am standing fast in this ghazā in the way that those seeking divine reward do;48 and [in that standing fast] I am becoming one of those who risk their lives in battle.49 If I am spared, that comes from my belief in God Most High. And if it is the other [outcome, i.e. death], then I enjoin you to hear and obey my son Malikshah and put him in my place and appoint him prince over you, for I have bequeathed this command to him and I have presented it to him.’ They responded to him with prayers and hearing and obeying. That was by the doing, organising and judgement of Nizam al-Mulk.

(p.38) The sultan remained with the above-mentioned section of the army as an isolated detachment.50 Each ghulām had a horse to ride and a horse to go by his side. He [Alp Arslan] marched, making for the king of Byzantium, and he waged war against them. He was victorious over them and he took the cross. They [the Byzantines] fled after they had been totally defeated, by killing and wounding. Their leader was taken to the sultan and he ordered that his nose should be cut off. He sent the cross, which was wood and on it were silver and pieces of turquoise, and a gospel51 in a silver casket, which he had with him, to Hamadhan. He wrote to Nizam al-Mulk about the victory and ordered that it [the cross] should be taken to the caliphal presence.52

The king of Byzantium arrived and they [the two sides] met at a place called al-Rahwa53 on Wednesday with five days remaining of Dhuʾl –Qaʿda [25 August 1071]. The army of Byzantium was numerous and the total of those with the sultan approached twenty thousand. As for the king of Byzantium, he had with him thirty-five thousand Franks54 and thirty-five thousand …55 with two hundred generals and commanders; each of them had between two thousand and five hundred horsemen. He [also] had with him fifteen thousand Ghuzz56 who were [living] beyond Constantinople; and one hundred thousand sappers and diggers and one hundred thousand siege engineers57 and four hundred carts58 on which were weapons, saddles, ballistas and mangonels, amongst which was a mangonel drawn by one thousand, two hundred men.

The sultan sent a message to the king of Byzantium, [saying] that he should return to his country and [saying]: ‘I myself will go home, and the peace treaty which the caliph brokered for us will be concluded between us.’ The king of Byzantium had [previously] sent his envoy asking the caliph to order the sultan to make peace and [conclude] a treaty. [But now] the answer of the king of Byzantium came back [to Alp Arslan]: ‘I have spent a lot of money and assembled many troops to come to the like of this situation. If I am victorious in it, how could I leave it? How preposterous! There will be no treaty except in Rayy and no going home except after I have done in the lands of Islam the like of what has been done in the lands of Byzantium.’

When it was the time of prayer on the Friday,59 the sultan performed the prayer with the army, he prayed to God Most High, he made invocations, wept and made humble supplications. He said to them [his army]: ‘We are with a depleted number of men. I want to throw myself on them [the Byzantines] at this hour when prayers are being said for us and for the Muslims on the pulpits. Either I will (p.39) achieve the goal or I will go as a martyr to Paradise. He amongst you who wants to follow me, let him follow me, and he who wants to leave, let him leave my company. Here is not a sultan commanding, nor an army being commanded, for today I am only one of you and a ghāzī with you. He who follows me and gives himself to God Most High, he will gain Paradise and booty. He who goes away, the Fire and ignominy are obligatory for him.’ They said to him: ‘O sultan! We are your servants and whatever you do we will follow you in it and help you in it. So do what you want.’

He threw down his bow and arrows60 and put on weapons, took the club, tied his mare's tail61 in a knot with his hand and mounted it. They did likewise. He advanced on the Byzantines and he shouted and they [his men] shouted. He launched an attack on them and the dust rose up and they fought against each other for an hour in which the situation far exceeded a mere rout of the infidels. They spent the day and the night killing in devastating fashion and they plundered and pillaged vast amounts of plunder and pillage. Then the sultan returned to his position and the khādim62 al-Kahraʿi63 came in to see him and he said: ‘O sultan! One of my ghulāms mentioned that the king of Byzantium is in captivity with him.’ This ghulām had been presented to Nizam al-Mulk with the whole army and he [Nizam al-Mulk] had looked down on him and rejected him. Overtures had been made about his case but he [Nizam al-Mulk] refused to appoint him and said scornfully: ‘Perhaps he will bring us the king of Byzantium as captive.’ So God Most High made the imprisonment of the king of Byzantium occur by his hand.

The sultan regarded that [story about the capture of Romanus] as unlikely and he summoned a ghulām named Shadhi who had gone several times with envoys to the king of Byzantium and he asked him to see him [the prisoner] and check him out. So he went and saw him, then he came back and said: ‘It is he.’ So he [Alp Arslan] gave orders for a tent to be set up for him and he brought him to it. He put him in chains and fettered his hand to his neck. [He also gave orders] that one hundred ghulāms should be put in charge of him. He put a robe of honour on the man who had captured him and secluded him [Romanus] and he gave him [the ghulām] what he suggested [as a reward] and he [the ghulām] explained the situation to him. He [the ghulām] said: ‘I attacked him, not recognising him, and around himwere ten young boys from amongst the servants. One of them said to me: “Do not kill him for he is the king”, so I took him captive and brought him [to the camp]” (pl.11).

(p.40) The sultan ordered him [Romanus] to be brought in and he was brought in front of him. He struck him three or four blows with his hand and he kicked him a similar number of times. He [Alp Arslan] said to him: ‘Did I not give permission to the envoys of the caliph that they should seek you out, draw up peace terms with you and respond therein to your request? Did I not now send a message to you and make an offer to you that I would withdraw from you, and you refused and so on. What thing caused you to infringe [the treaty]?’. He [Romanus] said: ‘I collected [troops], o sultan, and I had superior numbers and I had the upper hand, but64 the victory was yours. So do what you want and stop rebuking me.’ He [Alp Arslan] said: ‘ If I had fallen into your hands, what would you have done with me?’ He [Romanus] said: ‘Something vile.’ He [Alp Arslan] said: ‘He has spoken truthfully, by God! If he had said otherwise, he would be lying. This is an intelligent, strong man. It is not fitting that he should be killed.’ He [Alp Arslan] said [to Romanus]: ‘What do you think should be done with you now?’ He [Romanus] said: ‘ One of three things: the first is to kill me, the second is to parade me publicly in your country which I almost attacked and captured, and [as for] the third – there is no benefit in mentioning it, for you will not do it.’ He [Alp Arslan ] said: ‘Mention it.’ He [Romanus] said: ‘Pardoning me, accepting money, ransoming me, attaching me to your service and sending me back to my kingdom as a mamlūk of yours, as a deputy for you in the land of Byzantium.’ He [Alp Arslan] said: ‘ I have decided in respect of you only on that which suits your hopeless position. After thinking about it, bring enough money for your release65 and I will set you free.’ He [Romanus] said: ‘The sultan should say what he wants.’ He [Alp Arslan] said: ‘I want a million dīnārs.’ He [Romanus] said: ‘By God, you will deserve the kingdom of Byzantium, if you spare my life, but I have spent money and, since I have ruled over them [the Byzantines], I have used up from the wealth of Byzantium ten thousand66dīnārs on renewing the army and on the wars which I have fought with it [the money] until this present battle of mine, and I have made them poor by that. Were it not for that, I would not regard anything you suggest as excessive.’ And the conversation went on going to and fro until the matter was established on the basis of one and a half million dīnārs and for the treaty on the basis of three hundred and sixty thousand dīnārs each year and the release of every prisoner in Byzantium and the bringing of presents and gifts in addition to that, and that there should be brought at any time from the troops of Byzantium that had been sent away that which should be requested as the need demanded. So he (p.41)

The twelfth-century accounts of the battle of Manzikert

Figure 2.3 Modern Turkish depiction of Alp Arslan with sword and mace

[Romanus] said to him: ‘If you show favour to me, my being sent back will be speedy before the Byzantines appoint a king other than me; [if not] I will not be able to approach them and I will not be carry through anything of what I have offered.’

The sultan said: ‘I want you to return Anṭākiya,67 al-Ruhā68 and Manbij, for they were taken from the Muslims recently,69 and [I want you] to release the Muslim prisoners.’ He [Romanus] said: ‘If I return to my kingdom, I will send a contingent to each place in it and it will besiege it until I manage to make them surrender. As for my beginning with that, it is not possible for me [now]. As for the prisoners, I will release them all and I will behave kindly with them.’

The sultan ordered that his chains and collar be undone, then he said: ‘ Give him [Romanus] a drinking goblet.’ So it was given and he [Romanus] thought it was for him and so he wanted to drink from it, and he was prevented from doing so, was ordered to serve the sultan and to walk towards him and give it [the goblet] to him. So he nodded a little towards the ground according to the Byzantine custom and came towards him [Alp Arslan]. The sultan took the goblet, pulled70 his hair, put his face on the ground and said: ‘If you pay homage to kings, do it like this.’ And the reason for his demanding that was that the sultan had said in Rayy: ‘Here I am, going to fight the king of Byzantium, to take him prisoner and to appoint him at my head as a (p.42) cupbearer.’71 The king of Byzantium went away to his tent and raised a loan of ten thousand dīnārs and with it he settled his affair. He distributed it to his retinue, attendants and agents, and he sold a group of his generals and he gave away others.

The following day he [Alp Arslan] summoned him [Romanus]. He had set up for him his throne and chair which had been taken from him and he sat him on both of them. He removed his qabāʾ72 and qalansuwwa73 and dressed him with them [again] and he said to him: ‘I have attached you [to my service] and I am satisfied with what you have said. I will go with you to your country and will return you to your kingdom.’ He [Alp Arslan?] kissed the ground74 and said to him [Romanus]: ‘Did not the caliph of God Most High send you an envoy to take you to him, with the intention of sorting out your affair, and you ordered that he should uncover his head, tighten his belt,75 and kiss the ground in front of you?’76 He [Alp Arslan] had heard that he [Romanus] had done this with Ibn al-Muhallaban,77 so he [Alp Arslan] said: ‘Is not the matter as he [Ibn al-Muhallaban] says?’78 and he [Alp Arslan] appeared to change towards him,79 so he [Romanus] said; ‘O sultan, how have things come to this point?’80 He [Romanus] stood up, uncovered his head, made a gesture to the ground and said: ‘This is in exchange for what I did with his envoy.’ The sultan was happy with that and he ordered that a banner should be raised for him on which was written ‘There is no god but God. Muhammad is the Prophet of God.’ So he [Alp Arslan] raised it over his [Romanus'] head and sent two chamberlains and one hundred ghulāms to go with him to Constantinople and he [Alp Arslan] accompanied him for a farsakh.81 When he [Romanus] bade him [Alp Arslan] farewell, he wanted to dismount, so the sultan prevented him and they embraced each other and then parted company.

This victory in Islam was a wonder without peer, for people had assembled to destroy Islam and its people. The king of Byzantium had made up his mind to go to the sultan even as far as Rayy,82 and the generals had divided up the Islamic lands into fiefs.83 He [Romanus] said to the one to whom the fief of Baghdad had been given: ‘Do not interfere with that upright shaykh, for he is our friend,’ meaning the caliph. The generals were saying: ‘It is inevitable that we will winter in Rayy and summer in Iraq and that we will take the land of Syria on our return [home].’

When the victory happened and the news reached Baghdad, the drums and horns were played, the people gathered in the audience hall84 and the victory letters85 were read out.

(p.43) When [the people of] Byzantium heard what had happened they barred his [Romanus'] return to their country and they appointed someone else as king. He became an ascetic and put on woollen clothes and he sent to the sultan two hundred thousand dīnārs and a gold plate on which were jewels, the value of which was ninety thousand dīnārs, and he swore by the Gospel that he could not do more than that. He made for the king of Armenia seeking hospitality from him and he blinded him [Romanus] and sent a message to the sultan informing him about that.

Commentary on the text

Although there is a lack of historical precision in this long account, it is full of interesting details. Whilst Ibn al-Jawzi mentions that Alp Arslan had been in Syria with his army, he does not say what he was doing there. His narrative reiterates that Alp Arslan was with only his elite troops, numbering four thousand men, but his exact position when he heard about Romanus Diogenes' advance towards the Muslim world is not at all clear. Ibn al-Jawzi relates that Alp Arslan sent his wife and chief minister with the heavy baggage to the safety of Hamadhan but whether or not the latter sent the extra troops that he was asked by Alp Arslan to collect is not followed up in the account.

Some other Muslim narratives of Manzikert speak of a preliminary skirmish between the vanguards of the two armies; in it the Byzantine commander is captured and the sultan orders his nose to be cut off. The account of Ibn al-Jawzi appears to be confused in that it seems to suggest that Alp Arslan was directly involved in the preliminary fighting. A precise date for the battle – five days before the end of the month of Dhuʾl-Qaʿda – is provided, but the exact sequence of the battle and its military aspects are (predictably for a religious scholar in Baghdad) passed over in his hurry to talk triumphantly about the victory. He mentions the day of the week when the battle took place as Wednesday,86 but he presents Alp Arslan as preparing to advance on the Friday. What happened in the intervening period is not covered. A precise place for the battle – al-Rahwa – is given and the varied ethnic composition of the Byzantine army is mentioned. Once again much emphasis is placed on the impressive military impedimenta of the Byzantine emperor. It is interesting to note that Alp Arslan's elite troops had two horses each – a detail which is picked up in some later accounts. Ibn al-Jawzi is the first Muslim historian of Manzikert who gives a series of circumstantial details about Alp Arslan's preparations (p.44) for battle. He discards his bow and arrows, the standard nomad weaponry for long-range encounters, as a signal that he is about to engage in hand-to-hand combat. In the same spirit, he takes up an appropriate weapon for such combat, namely a club (together with other weapons that are not specified). Then he ties his mare's tail in a knot. Normally a horse's tail would be allowed to grow long, and this would enable the animal to disperse the flies that would plague it in a hot climate. But in battle, a cavalier could be disadvantaged by the long tail of his horse, which could be grabbed by enemy soldiers and thus result in a loss of his control over his mount – and perhaps even cause him to be unhorsed. In the same precautionary spirit the sultan, in other accounts, is described as tightening the girths of his horse just before the battle.87

Most, if not all, of the details of the treaty drawn up after the battle are probably apocryphal.

Nose-cutting has been known since antiquity as a brutal way of dishonouring a person, notably in times of war. Famous instances include Artaxerxes, who was punished in 329 BC by Alexander the Great according to ‘Persian law’ by the cutting off of his nose and ears before execution. The practice was still in existence amongst the Ottomans, who cut off the nose and ears of Marco Antonio Bragadino, the Venetian defender of Cyprus, in 1571.88 Such a widespread practice was intended as a shaming exemplary punishment; the Arabic term tashhīr means ‘to parade an offender as a public example’ (the offenderwas often mounted on an ass or camel).89 It is interesting to note that there is a parallel incident of this kind in Attaleiates' account of the battle:

When one of the soldiers was accused of having stolen a Turkish ass, he was brought in full view, bound, before the emperor but a punishment was decreed which surpassed the crime: the penalty was not set in money but in the cutting of the nose. Although the man pleaded a great deal and offered to give up all his property and although he put forward as a mediator the most holy icon of the All-Hymned Lady the Blachernitissa Theotokos which it was the custom for pious emperors to take on campaigns as an invincible weapon, the emperor did not feel pity nor even respect for the asylum granted by the holy icon. But with me and everyone looking on and with the icon itself being held, the wretch screeching loudly and groaning deeply had his nose cut off. It was indeed then that I had forebodings that great would be the nemesis which would befall us from God.90

(p.45) Notes

(1.) Ibn Khallikan, Wafayāt al-a‘yān, ed. I. Abbas, vol. 5, 69.

(2.) Cf. S. Dayf's Arabic introduction to his edition of al-Turtushi, 6–7. The account of the battle of Manzikert is to be found on pp. 694–7; cf. al-Turtushi, Sirāj al-muluūk, ed. S. Dayf, Cairo, 1991. The work has been translated into Spanish: cf. M. Alarcón, Lámpara de los principes, 2 vols, Madrid, 1930–1.

(3.) For an account of his stay in Baghdad, cf. Fierro's introduction to al-Turtushi, Kitāb al-ḥawādith waʾl- bidāʿ (El libro de las novedades y las innovaciones), tr. and ed. M. Fierro, Madrid, 1993, 34–9. Fierro lists histeachers in Iraq; ibid., 36–9.

(4.) Al-Turtushi, Sirāj, 513–18;

(5.) Cf. EI 2: s.v. al-Ṭurṭūshī (A. Ben Abdesselem).

(6.) For Alarcón's Spanish translation of the account of the battle, cf. Làmpara de los principes, 328–32.

(7.) Manzikert, cf. al-Turtushi, Sirāj, 694–6.

(8.) J. France, Western warfare in the age of the Crusades, New York, 1999, 124–43.

(9.) R. Macrides, ‘History-writing in the twelfth century’, in The perception of the past in twelfth-century Europe, ed. P. Magdalino, London, 1992, 124.

(10.) akla jāʾiʿ – literally ‘the meal of a hungry man’.

(11.) ka-al-raqma fī dhirāʿ al-ḕimār. This is an echo of the hadīth.: ‘You are amongst the nations no more than the mark on the leg of the riding-animal.’ raqma denotes the black spot on the rump of the ass; cf. Ibn Manzur, Lisān al-ʿarab al-muḥīṭ ed. Y. Khayyat, Beirut, n.d., 1210; E. W. Lane, An Arabic-English Lexicon, 1/3, Beirut, 1980, 1139.

(12.) By saying the bismillāh (‘In the name of God the Compassionate, the Merciful’).

(13.) For example, the mustering of Alp Arslan's troops in Isfahan.

(p.46)

(14.) Ibn al-Qalanisi, Dhayl tarʿīkh Dimashq, ed. H. F. Amedroz, Leiden, 1908, 99. This chronicle has been partly translated twice but in neither case was the account of Manzikert included; cf. H. A. R. Gibb, The Damascus Chronicle of the Crusades; London, 1932, and R. Le Tourneau, Damas de 1075 à 1154, Damascus, 1952. There is, however, a Turkish translation of the account of Ibn al-Qalanisi; cf. Sümer and Sevim, 1–3. The Arabic text is also included by Zakkar, Mukhtā rāt, 129. For the work of Ibn al-Qalanisi, cf. also C. Cahen, ‘Note d'historiographie syrienne. La première partie de l'histoire d'Ibn al-Qalanisi’, in Arabic and Islamic Studies in Honor of Hamilton A. R. Gibb, ed. G. Makdisi, Cambridge, MA., 1965, 156–68. For a biographical notice of Ibn al-Qalanisi, cf. al-Dhahabi, cited in Arabic by Amedroz in his edition of the text, Dhayl tarʿīkh Dimashq, 6.

(15.) ʿUyūn al-tawārīkhC. Cahen, ‘The historiography of the Seljuqid period’, in Historians of the Middle East, ed. B. Lewis and P. M. Holt, London,1962, 60–1.

(16.) Armānūs. The form Romanus will be used throughout the book.

(17.) Manazjird is the oldest Arabic form of the town now known in Turkey as Malazgirt. The form Manazjird is close to one of the Old Armenian names for it – Manazkert; cf. EI2 s.v. Malāzgird: The town (S. Faroqui). The translations in this book will use Manzikert, which is the best-known form of the name in the West.

(18.) Al-kirāʿ – a word meaning horses and donkeys.

(19.) The Turkish translation, although it corresponds to the version of events given in some later accounts, seems to be wrong here: ‘deniliğine göre, Bizanslilar onu yakalayıp gözlerine mil c¸ekmis¸ler – ‘it is said that the Byzantines put a collar round his neck and blinded him’. The Arabic is: ightalūhu wa-sallamūhu. Attractive as the translation of ightalūhu as ‘put a collar round him’ may be, it is not tenable grammatically. The root ghalla with the meaning being ‘to put a ring on the neck or hand of someone’ (cf. Lane, Lexicon, vol. 6, 2277–8) cannot be right here; cf. also Kazimirski, Dictionnaire, II, 487. Similarly sallamūhu does not mean ‘they blinded him’.

(20.) C. Hillenbrand, ‘Ibn al-ʿAdim's biography of the Seljuq sultan, Alp Arslan’, Actas XVI Congreso UEAI, ed. C. Vázquez de Benito and M. A. Manzano Rodriguez, Salamanca, 1995, 237–42.

(p.47)

(21.) There are two editions of the text: C. Cahen, ‘La chronique abrégée d'al-ʿAzīmī’, Journal asiatique 230 (1938), 335–448; I. Za‘rur, Ta’rīkh Ḥalab, Damascus, 1984. Cf. also C. Cahen, La Syrie du nord à l'époque des croisades (Paris, 1940), 42–3. Cahen's edition of the text has been usedfor the translation; cf. al-ʿAzimi, ‘La chronique abrégée’, 359.

(22.) The reference to the emperor being sold ‘for a dīnār’, described by Cahen as ‘unclear’ (‘La chronique abrégée’, 425, n. 2), is indeed a rather odd detail. A dīnār – a gold coin – does not represent a paltry sum. The use of fils or dirham would have been more appropriate here to underline the insult intended for Romanus.

(23.) For an account of the battle, cf. Ibn al-Azraq al-Fariqi, Tarʾīkh al-Fāriqī, ed. B. A. L. Awad, Cairo, 1959, 189–90. Cf. also the same text copied by Amedroz in the footnotes to his edition of Ibn al-Qalanisi, Dhayl, 99–100, n. 1. Cf. also Sümer and Sevim, Islâm kaynaklarina, 4–5. For information about Ibn al-Azraq, cf. C. Hillenbrand, A Muslim Principality in Crusader Times, Leiden, 1990, 5–14; H. F. Amedroz, ‘TheMarwānid dynasty at Mayyāfāriqīn in the tenth and eleventh centuries A.D.’, Journal of the Royal Asiatic Society (1903), 123–54.

(24.) Ibn al-Azraq, ed. Awad, 187.

(25.) Presumably a reference to this being the last of the expeditions of Romanus into eastern Anatolia.

(26.) Khilat is the form preferred by Ibn al-Azraq for the town Akhlat at the north-western corner of Lake Van; cf. EI2 s.v. Akhlāṭ (V. Minorsky).

(27.) Hillenbrand, Muslim principality, 35

(28.) A town to the south-west of Lake Van.

(29.) The vocalisation of this name is uncertain. Presumably he was the qāḍī of Manzikert mentioned above.

(30.) The phrase: ‘God is most great’ (Allāhu akbar).

(31.) W. Hinz, Islamische Masse und Gewichte, Leiden, 1955, 27–33.

(32.) Ibn al-Azraq, Tarʿīkh Mayyāfāriqīn, ed. C. Hillenbrand as A Muslim principality in Crusader times, Leiden, 1999, 98–9.

(33.) Nishapuri, The Saljuqnama of Zahir al-Din Nishapuri, ed. A. H. Morton, Chippenham, 2004, 22. This recent edition is based on the unique manuscript in the Library of the Royal Asiatic Society in London (Ms. Persian 22(b) (Morley no.138); cf. ibid., 6.

(34.) ibid

(p.48)

(35.) ibid

(36.) Raiding in the path of God.

(37.) Qurʾan 2: 249–50.

(38.) This term varies in meaning but it often denotes a slave servant.

(39.) Literally: ‘they (i.e. a number of sources) have reported’.

(40.) This Turkish slave commander worked for Alp Arslan and his son Malikshah. In the fratricidal struggle between the sons of Malikshah – Muhammad and Barkyaruq – he supported the latter and was killed on the battlefield in 493/1100; cf. Ibn al-Jawzi, Muntaẓam, IX, 115–16; Ibnal-Athir, Kāmil, X, 200–1.

(41.) ʿāriḍ – the inspecting officer. His task was to examine men, horses and weapons in the ʿāriḍ, the periodic inspection of the army; cf. Encycl. Ir., II/7, s.v. ʿarz, dīvan (-e) (C. E. Bosworth). If the slave's name had not been written down, he would not have been paid.

(42.) A sign of servitude since ancient times.

(43.) jizya – payable by the ‘People of the Book’ – usually Christians and Jews, but also adherents of other religions with established scriptures. In return, they were allowed to worship freely, under certain restrictions, and were promised protection from the Muslim state.

(44.) There are two editions of this text: ed. F. Krenkow, Hyderabad, 1938–40, and ed. M. A. Ata et al., Beirut, 1992–3. For the account of Manzikert, cf. ed. Krenkow, vol. 8, 260–5. Cf. also Sümer and Sevim, Islâm kaynaklarina 9–15; Zakkar, Mukhtārāt 119–23. There are occasional variations in the published text of Sümer and Sevim and that of Zakkar, and even omissions and alterations. When these are significant, they will be mentioned in the footnotes. In general, the starting point for the translation has been the Hyderabad edition.

(45.) fall – scattered remnants of an army; débris dʿune armée en déroute (Kazimirski, II, 626). fall is Zakkar's reading of the text; the edition of Sümer and Sevim has fī qalīl min al-Žaskar (‘with few troops’). 11. It is not clear where this version has come from.

(46.) Reading istanfada (Muntaẓam, 260; Sümer and Sevim, 11) rather than istanafadha (Zakkar, 119).

(47.) Presumably his wife. Khatun is a title used for royal Turkish women.

(48.) ṣabr al-mu ḥ taṣibīn

(49.) masīr al-mukhāṭirīn

(50.) jarīdaKazimirski, Dictionnaire, I, 277.

(51.) Presumably here this means a copy of the whole of the New Testament or at least all four Gospels.

(p.49)

(52.) Literally: ‘to the presence of the caliphate’. This fighting is a reference to a preliminary skirmish between the Turks and the Byzantines which took place before the battle itself.

(53.) The true form of this name seems uncertain. Al-Rahwa seems better than al-Zahwa since it is attested in Yaqut who says that it is a desert near Akhlat – cf. Yaqut, Muʿjam al-buldān, ed. F.Wüstenfeld, Leipzig, 1866–73, II, 880. Manzikert (in modern Turkish, Malazgirt) stands in the plain of the Murat Su (the lower Euphrates); Sinclair mentions the Rahva plateau; cf. T. A. Sinclair, Eastern Turkey: an architectural and archaeological survey, London, 1987–90, I, 177–8.

(54.) Western European mercenaries were commonly used in Byzantine armies.

(55.) The text has a gap here.

(56.) Called Ouzoi in Byzantine sources. The Ghuzz were to be found in southern Russia, the lower Danube area and Byzantium; cf. EI2 s.v. Ghuzz (C. Cahen).

(57.) Reading ruzdārī for ruzjārī. The translation ‘siege engineers’ is only tentative.

(58.) This detail is magnified to a thousand carts by Attaleiates; cf. tr. Macrides, Appendix A, p. 229.

(59.) This was the noon prayer.

(60.) For a detailed discussion of the importance of archery in the Turkish conquest of Anatolia, cf. W. E. Kaegi Jr, ‘The contribution of archery to the Turkish conquest of Anatolia’, Speculum, 39/1 (January 1964), 96–108. As Kaegi writes, ‘the Seljuks preferred the bow to other weapons’. ibid., 96. For the iconography of the bow as a symbol of power (and not only among the Turks), see J. Zick-Nissen, ‘The turquoise “jām” of King Jamshīd’, in The Art of the Saljuqs in Iran and Anatolia, ed. R. Hillenbrand, Costa Mesa, CA., 1994, 183–5, with references to the bow being depicted on Seljuq coins; and S. Redford, ‘A grammar of Rūm Seljuk ornament’, in Mésogeios Méditerranée 25–26 (2005), ed. G. Leiser, 294–5.

(61.) Turks; W. Ridgeway, ‘The origin of the Turkish crescent’, The Journal of the Royal Anthropological Institute of Great Britain and Ireland, 38 (Jul. -Dec. 1908), 241–58.

(62.) Household servant, often a eunuch, but the term had a wide range of meanings depending on place and period.

(63.) Probably a feeble attempt to produce the name Gawharaʾīn found in other texts.

(64.) changing ‘and’ in the text to ‘but’.

(65.) Literally: ‘money adequate for your neck’.

(p.50)

(66.) The text has ten thousand dīnārs. The editor writes that ‘perhaps it is ten million’, Krenkow, 263, n. 1. This suggestion is adopted in the editions both of Zakkar and of Sümer and Sevim.

(67.) Antioch. This city had actually been ruled by the Byzantines since 969; cf. EI2 s.v. Anṭākiya (M. Streck – H. A. R. Gibb).

(68.) Edessa. This important centre of Armenian Christianity had been under overall Byzantine rule since 1037; cf. EI2 s.v. al-Ruhā(E. Honigmann – C. E. Bosworth).

(69.) Romanus had taken the Syrian city of Manbij as late as 460/1068; cf. Cahen, Pre-Ottoman Turkey, tr. J. Jones-Williams (London, 1968), 28; Zakkar, Emirate, 174–5. Nicephorus Bryennius reports that Romanus was very proud of this conquest, which had given him false hope of being successful against the Turks again; cf. Nicephorus Bryennius, tr. Gautier, 106.

(70.) jarraZakkar, Mukhtārāt, 122;jazza

(71.) Presumably in retaliation for the humiliating way in which Romanus had treated the caliphal envoy.

(72.) A sleeved garment for men.

(73.) A pointed bonnet.

(74.) The phrase wa-qabbala al-arḍ is omitted by Sümer and Sevim in their Turkish translation (16) but it is there in their Arabic edition (15).

(75.) The belt was an important item of insignia.

(76.) These actions are intended to humiliate Romanus publicly.

(77.) The envoy of the caliph, Ibn al-Muhallaban, had been made to make proskynesis (a bow to the ground), an act of homage demanded by Roman emperors since the time of Constantine, or even earlier; Friendly, The dreadful day, 184.

(78.) Sümer and Sevim's edition gives: ʿalāmāyuqālu – according to what is said, 16.

(79.) Literally: ‘there appeared a change from him towards him’.

(80.) Literally: ‘in what thing am I placed until I am placed in this?’

(81.) A measurement of distance, roughly equivalent to a league or around three miles.

(82.) I.e. right into the heart of Seljuq territory. Rayy, near modern Tehran, was one of the favourite cities of the Seljuqs.

(83.) wa- aqṭaʾ al-baṭā riqa al-bilād al-islāmiyya

(84.) bayt al-nawba – audience hall or guardroom in a palace or a tent in a royal camp used for audiences.

(85.) The public reading of a fatḥnāma (a victory letter) was an important moment of celebration.

(p.51)

(86.) The first day of Dhuʿl-Hijja was a Tuesday. So the date given – five days remaining of Dhuʾl-Qaʿda – falls on a Thursday.

(87.) For the significance of the horse in the whole Turkic world, cf. J.-P. Roux, Etudes d'iconographie islamique, Leuven, 1982, 34. For the importance of bows and arrows, cf. ibid., 60–1.

(88.) J. W. Frembgen, ‘Honour, shame, and bodily mutilation. Cutting off the nose among tribal societies’, Journal of the Royal Asiatic Society 16/3 (November 2006), 243, 245 and 254

(89.) R. Peters, Crime and punishment in Islamic law, Cambridge, 2006, 34.

(90.) Tr. Macrides, Appendix A, pp. 229–37.

Notes:

(1.) Ibn Khallikan, Wafayāt al-a‘yān, ed. I. Abbas, vol. 5, 69.

(2.) Cf. S. Dayf's Arabic introduction to his edition of al-Turtushi, 6–7. The account of the battle of Manzikert is to be found on pp. 694–7; cf. al-Turtushi, Sirāj al-muluūk, ed. S. Dayf, Cairo, 1991. The work has been translated into Spanish: cf. M. Alarcón, Lámpara de los principes, 2 vols, Madrid, 1930–1.

(3.) For an account of his stay in Baghdad, cf. Fierro's introduction to al-Turtushi, Kitāb al-ḥawādith waʾl- bidāʿ (El libro de las novedades y las innovaciones), tr. and ed. M. Fierro, Madrid, 1993, 34–9. Fierro lists histeachers in Iraq; ibid., 36–9.

(4.) Al-Turtushi, Sirāj, 513–18;

(5.) Cf. EI 2: s.v. al-Ṭurṭūshī (A. Ben Abdesselem).

(6.) For Alarcón's Spanish translation of the account of the battle, cf. Làmpara de los principes, 328–32.

(7.) Manzikert, cf. al-Turtushi, Sirāj, 694–6.

(8.) J. France, Western warfare in the age of the Crusades, New York, 1999, 124–43.

(9.) R. Macrides, ‘History-writing in the twelfth century’, in The perception of the past in twelfth-century Europe, ed. P. Magdalino, London, 1992, 124.

(10.) akla jāʾiʿ – literally ‘the meal of a hungry man’.

(11.) ka-al-raqma fī dhirāʿ al-ḕimār. This is an echo of the hadīth.: ‘You are amongst the nations no more than the mark on the leg of the riding-animal.’ raqma denotes the black spot on the rump of the ass; cf. Ibn Manzur, Lisān al-ʿarab al-muḥīṭ ed. Y. Khayyat, Beirut, n.d., 1210; E. W. Lane, An Arabic-English Lexicon, 1/3, Beirut, 1980, 1139.

(12.) By saying the bismillāh (‘In the name of God the Compassionate, the Merciful’).

(13.) For example, the mustering of Alp Arslan's troops in Isfahan.

(14.) Ibn al-Qalanisi, Dhayl tarʿīkh Dimashq, ed. H. F. Amedroz, Leiden, 1908, 99. This chronicle has been partly translated twice but in neither case was the account of Manzikert included; cf. H. A. R. Gibb, The Damascus Chronicle of the Crusades; London, 1932, and R. Le Tourneau, Damas de 1075 à 1154, Damascus, 1952. There is, however, a Turkish translation of the account of Ibn al-Qalanisi; cf. Sümer and Sevim, 1–3. The Arabic text is also included by Zakkar, Mukhtā rāt, 129. For the work of Ibn al-Qalanisi, cf. also C. Cahen, ‘Note d'historiographie syrienne. La première partie de l'histoire d'Ibn al-Qalanisi’, in Arabic and Islamic Studies in Honor of Hamilton A. R. Gibb, ed. G. Makdisi, Cambridge, MA., 1965, 156–68. For a biographical notice of Ibn al-Qalanisi, cf. al-Dhahabi, cited in Arabic by Amedroz in his edition of the text, Dhayl tarʿīkh Dimashq, 6.

(15.) ʿUyūn al-tawārīkhC. Cahen, ‘The historiography of the Seljuqid period’, in Historians of the Middle East, ed. B. Lewis and P. M. Holt, London,1962, 60–1.

(16.) Armānūs. The form Romanus will be used throughout the book.

(17.) Manazjird is the oldest Arabic form of the town now known in Turkey as Malazgirt. The form Manazjird is close to one of the Old Armenian names for it – Manazkert; cf. EI2 s.v. Malāzgird: The town (S. Faroqui). The translations in this book will use Manzikert, which is the best-known form of the name in the West.

(18.) Al-kirāʿ – a word meaning horses and donkeys.

(19.) The Turkish translation, although it corresponds to the version of events given in some later accounts, seems to be wrong here: ‘deniliğine göre, Bizanslilar onu yakalayıp gözlerine mil c¸ekmis¸ler – ‘it is said that the Byzantines put a collar round his neck and blinded him’. The Arabic is: ightalūhu wa-sallamūhu. Attractive as the translation of ightalūhu as ‘put a collar round him’ may be, it is not tenable grammatically. The root ghalla with the meaning being ‘to put a ring on the neck or hand of someone’ (cf. Lane, Lexicon, vol. 6, 2277–8) cannot be right here; cf. also Kazimirski, Dictionnaire, II, 487. Similarly sallamūhu does not mean ‘they blinded him’.

(20.) C. Hillenbrand, ‘Ibn al-ʿAdim's biography of the Seljuq sultan, Alp Arslan’, Actas XVI Congreso UEAI, ed. C. Vázquez de Benito and M. A. Manzano Rodriguez, Salamanca, 1995, 237–42.

(21.) There are two editions of the text: C. Cahen, ‘La chronique abrégée d'al-ʿAzīmī’, Journal asiatique 230 (1938), 335–448; I. Za‘rur, Ta’rīkh Ḥalab, Damascus, 1984. Cf. also C. Cahen, La Syrie du nord à l'époque des croisades (Paris, 1940), 42–3. Cahen's edition of the text has been usedfor the translation; cf. al-ʿAzimi, ‘La chronique abrégée’, 359.

(22.) The reference to the emperor being sold ‘for a dīnār’, described by Cahen as ‘unclear’ (‘La chronique abrégée’, 425, n. 2), is indeed a rather odd detail. A dīnār – a gold coin – does not represent a paltry sum. The use of fils or dirham would have been more appropriate here to underline the insult intended for Romanus.

(23.) For an account of the battle, cf. Ibn al-Azraq al-Fariqi, Tarʾīkh al-Fāriqī, ed. B. A. L. Awad, Cairo, 1959, 189–90. Cf. also the same text copied by Amedroz in the footnotes to his edition of Ibn al-Qalanisi, Dhayl, 99–100, n. 1. Cf. also Sümer and Sevim, Islâm kaynaklarina, 4–5. For information about Ibn al-Azraq, cf. C. Hillenbrand, A Muslim Principality in Crusader Times, Leiden, 1990, 5–14; H. F. Amedroz, ‘TheMarwānid dynasty at Mayyāfāriqīn in the tenth and eleventh centuries A.D.’, Journal of the Royal Asiatic Society (1903), 123–54.

(24.) Ibn al-Azraq, ed. Awad, 187.

(25.) Presumably a reference to this being the last of the expeditions of Romanus into eastern Anatolia.

(26.) Khilat is the form preferred by Ibn al-Azraq for the town Akhlat at the north-western corner of Lake Van; cf. EI2 s.v. Akhlāṭ (V. Minorsky).

(27.) Hillenbrand, Muslim principality, 35

(28.) A town to the south-west of Lake Van.

(29.) The vocalisation of this name is uncertain. Presumably he was the qāḍī of Manzikert mentioned above.

(30.) The phrase: ‘God is most great’ (Allāhu akbar).

(31.) W. Hinz, Islamische Masse und Gewichte, Leiden, 1955, 27–33.

(32.) Ibn al-Azraq, Tarʿīkh Mayyāfāriqīn, ed. C. Hillenbrand as A Muslim principality in Crusader times, Leiden, 1999, 98–9.

(33.) Nishapuri, The Saljuqnama of Zahir al-Din Nishapuri, ed. A. H. Morton, Chippenham, 2004, 22. This recent edition is based on the unique manuscript in the Library of the Royal Asiatic Society in London (Ms. Persian 22(b) (Morley no.138); cf. ibid., 6.

(36.) Raiding in the path of God.

(37.) Qurʾan 2: 249–50.

(38.) This term varies in meaning but it often denotes a slave servant.

(39.) Literally: ‘they (i.e. a number of sources) have reported’.

(40.) This Turkish slave commander worked for Alp Arslan and his son Malikshah. In the fratricidal struggle between the sons of Malikshah – Muhammad and Barkyaruq – he supported the latter and was killed on the battlefield in 493/1100; cf. Ibn al-Jawzi, Muntaẓam, IX, 115–16; Ibnal-Athir, Kāmil, X, 200–1.

(41.) ʿāriḍ – the inspecting officer. His task was to examine men, horses and weapons in the ʿāriḍ, the periodic inspection of the army; cf. Encycl. Ir., II/7, s.v. ʿarz, dīvan (-e) (C. E. Bosworth). If the slave's name had not been written down, he would not have been paid.

(42.) A sign of servitude since ancient times.

(43.) jizya – payable by the ‘People of the Book’ – usually Christians and Jews, but also adherents of other religions with established scriptures. In return, they were allowed to worship freely, under certain restrictions, and were promised protection from the Muslim state.

(44.) There are two editions of this text: ed. F. Krenkow, Hyderabad, 1938–40, and ed. M. A. Ata et al., Beirut, 1992–3. For the account of Manzikert, cf. ed. Krenkow, vol. 8, 260–5. Cf. also Sümer and Sevim, Islâm kaynaklarina 9–15; Zakkar, Mukhtārāt 119–23. There are occasional variations in the published text of Sümer and Sevim and that of Zakkar, and even omissions and alterations. When these are significant, they will be mentioned in the footnotes. In general, the starting point for the translation has been the Hyderabad edition.

(45.) fall – scattered remnants of an army; débris dʿune armée en déroute (Kazimirski, II, 626). fall is Zakkar's reading of the text; the edition of Sümer and Sevim has fī qalīl min al-Žaskar (‘with few troops’). 11. It is not clear where this version has come from.

(46.) Reading istanfada (Muntaẓam, 260; Sümer and Sevim, 11) rather than istanafadha (Zakkar, 119).

(47.) Presumably his wife. Khatun is a title used for royal Turkish women.

(48.) ṣabr al-mu ḥ taṣibīn

(49.) masīr al-mukhāṭirīn

(50.) jarīdaKazimirski, Dictionnaire, I, 277.

(51.) Presumably here this means a copy of the whole of the New Testament or at least all four Gospels.

(52.) Literally: ‘to the presence of the caliphate’. This fighting is a reference to a preliminary skirmish between the Turks and the Byzantines which took place before the battle itself.

(53.) The true form of this name seems uncertain. Al-Rahwa seems better than al-Zahwa since it is attested in Yaqut who says that it is a desert near Akhlat – cf. Yaqut, Muʿjam al-buldān, ed. F.Wüstenfeld, Leipzig, 1866–73, II, 880. Manzikert (in modern Turkish, Malazgirt) stands in the plain of the Murat Su (the lower Euphrates); Sinclair mentions the Rahva plateau; cf. T. A. Sinclair, Eastern Turkey: an architectural and archaeological survey, London, 1987–90, I, 177–8.

(54.) Western European mercenaries were commonly used in Byzantine armies.

(55.) The text has a gap here.

(56.) Called Ouzoi in Byzantine sources. The Ghuzz were to be found in southern Russia, the lower Danube area and Byzantium; cf. EI2 s.v. Ghuzz (C. Cahen).

(57.) Reading ruzdārī for ruzjārī. The translation ‘siege engineers’ is only tentative.

(58.) This detail is magnified to a thousand carts by Attaleiates; cf. tr. Macrides, Appendix A, p. 229.

(59.) This was the noon prayer.

(60.) For a detailed discussion of the importance of archery in the Turkish conquest of Anatolia, cf. W. E. Kaegi Jr, ‘The contribution of archery to the Turkish conquest of Anatolia’, Speculum, 39/1 (January 1964), 96–108. As Kaegi writes, ‘the Seljuks preferred the bow to other weapons’. ibid., 96. For the iconography of the bow as a symbol of power (and not only among the Turks), see J. Zick-Nissen, ‘The turquoise “jām” of King Jamshīd’, in The Art of the Saljuqs in Iran and Anatolia, ed. R. Hillenbrand, Costa Mesa, CA., 1994, 183–5, with references to the bow being depicted on Seljuq coins; and S. Redford, ‘A grammar of Rūm Seljuk ornament’, in Mésogeios Méditerranée 25–26 (2005), ed. G. Leiser, 294–5.

(61.) Turks; W. Ridgeway, ‘The origin of the Turkish crescent’, The Journal of the Royal Anthropological Institute of Great Britain and Ireland, 38 (Jul. -Dec. 1908), 241–58.

(62.) Household servant, often a eunuch, but the term had a wide range of meanings depending on place and period.

(63.) Probably a feeble attempt to produce the name Gawharaʾīn found in other texts.

(64.) changing ‘and’ in the text to ‘but’.

(65.) Literally: ‘money adequate for your neck’.

(66.) The text has ten thousand dīnārs. The editor writes that ‘perhaps it is ten million’, Krenkow, 263, n. 1. This suggestion is adopted in the editions both of Zakkar and of Sümer and Sevim.

(67.) Antioch. This city had actually been ruled by the Byzantines since 969; cf. EI2 s.v. Anṭākiya (M. Streck – H. A. R. Gibb).

(68.) Edessa. This important centre of Armenian Christianity had been under overall Byzantine rule since 1037; cf. EI2 s.v. al-Ruhā(E. Honigmann – C. E. Bosworth).

(69.) Romanus had taken the Syrian city of Manbij as late as 460/1068; cf. Cahen, Pre-Ottoman Turkey, tr. J. Jones-Williams (London, 1968), 28; Zakkar, Emirate, 174–5. Nicephorus Bryennius reports that Romanus was very proud of this conquest, which had given him false hope of being successful against the Turks again; cf. Nicephorus Bryennius, tr. Gautier, 106.

(70.) jarraZakkar, Mukhtārāt, 122;jazza

(71.) Presumably in retaliation for the humiliating way in which Romanus had treated the caliphal envoy.

(72.) A sleeved garment for men.

(73.) A pointed bonnet.

(74.) The phrase wa-qabbala al-arḍ is omitted by Sümer and Sevim in their Turkish translation (16) but it is there in their Arabic edition (15).

(75.) The belt was an important item of insignia.

(76.) These actions are intended to humiliate Romanus publicly.

(77.) The envoy of the caliph, Ibn al-Muhallaban, had been made to make proskynesis (a bow to the ground), an act of homage demanded by Roman emperors since the time of Constantine, or even earlier; Friendly, The dreadful day, 184.

(78.) Sümer and Sevim's edition gives: ʿalāmāyuqālu – according to what is said, 16.

(79.) Literally: ‘there appeared a change from him towards him’.

(80.) Literally: ‘in what thing am I placed until I am placed in this?’

(81.) A measurement of distance, roughly equivalent to a league or around three miles.

(82.) I.e. right into the heart of Seljuq territory. Rayy, near modern Tehran, was one of the favourite cities of the Seljuqs.

(83.) wa- aqṭaʾ al-baṭā riqa al-bilād al-islāmiyya

(84.) bayt al-nawba – audience hall or guardroom in a palace or a tent in a royal camp used for audiences.

(85.) The public reading of a fatḥnāma (a victory letter) was an important moment of celebration.

(86.) The first day of Dhuʿl-Hijja was a Tuesday. So the date given – five days remaining of Dhuʾl-Qaʿda – falls on a Thursday.

(87.) For the significance of the horse in the whole Turkic world, cf. J.-P. Roux, Etudes d'iconographie islamique, Leuven, 1982, 34. For the importance of bows and arrows, cf. ibid., 60–1.

(88.) J. W. Frembgen, ‘Honour, shame, and bodily mutilation. Cutting off the nose among tribal societies’, Journal of the Royal Asiatic Society 16/3 (November 2006), 243, 245 and 254

(89.) R. Peters, Crime and punishment in Islamic law, Cambridge, 2006, 34.

(90.) Tr. Macrides, Appendix A, pp. 229–37.