For over a millennium, about 850 to 1900, Turks dominated the military and political life of the central Muslim lands. Although few in number, they consistently managed to conquer and rule the other peoples of this area. They not only held power in regions adjacent to their homeland, the Central Asian steppe, but even in distant countries. That is, they not only ruled northern India, Iran, and Anatolia for long periods, but also such remote areas as southern India, the Yemen, and Egypt. Turks founded many of the great dynasties of Islam, such as the Seljuks, Mamluks, Mughals, Safavids, and foremost of all, the Ottomans. Under the Ottomans, Turks ruled such distant areas as Algeria and the Sudan.
Although the reasons for the Turkish military and political predominance over other Muslims are too complex to discuss here, we can trace the Turks' first rise to power in an Islamic setting. Turkish domination did not begin during the very first years of Islam; nor did it suddenly occur in the reign of al-Mu'tasim (218-27/ 833-42), as is commonly thought. Rather, Turks played a small but growing role from the 50/670's on. The following pages trace this development from its beginnings to its florescence under al-Mu'tasim.
The history of Turks in early Muslim service divides at al-Ma'mun's accession in the year 198/813. Before then, Turks had only a moderate role in Muslim governments and armies; from that date on, they acquired enormous significance.
In reference to the first two centuries of Islam, the term Turk as used by Arabic and Persian sources presents difficulties. The Muslim authors mean different things by the term, depending on their era, proximity to Inner Asia, and knowledge of that region. It can overlap in meaning with other ethnic names (e.g. Soghdian, Khazar, Farghanian). Modern scholars have concluded that the term had a very wide meaning in the early centuries: "all the non-Persian peoples of the East";[1] "any people north-east of the Oxus";[2] "a person who spoke Turkish";[3] "nomads of the Central Asian Steppes."[4]
Despite the vague and wide meaning of the word "Turk," the following pages discuss only persons specifically identified as Turks, since it was soldiers by that name who eventually acquired control of the central Abbasid government shortly after the death of al-Mu'tasim.
By "Muslim service," I mean the armies, courts, and governments of the Muslims.
A number of secondary works deal with the Turks from the time of al-Mu'tasim;[5] little work has been done on their earlier role in Muslim countries.[6]
The First Turks, 54-198/674-813
Although the earliest certain date for a Turk in Muslim service is 60/680, it appears likely that Turks fought in numbers for the Muslims already a few years earlier. In 54/674 'Ubaydallah b. Ziyad, the Umayyad governor of Khurasan, vanquished the city of Bukhara and its Turkish allies.[7] A year later, on being appointed governor of Basra, he took with him 2,000 or 4,000 of the Bukharan captives as slaves.[8] These captives formed a military unit known as the Bukhariya which was famed for the skill of its archers.[9]
It is unclear whether the Bukhariya were slaves or free men;[10] it is also unclear how many of them were Turks. The fact that Turks fought with Bukhara against the Muslims implies that they made up a portion of the captives; the excellence of the Bukhariya at archery confirms this likelihood, since Turks have always been renowned for their skill at shooting arrows. Further support for this idea comes from the first specific mention of a Turk in Muslim service, for he was a freed slave of 'Ubaydallah b. Ziyad; most likely he belonged to 'Ubaydallah's Bukhariya corps.
This first Turk was a mawla (freed slave) named Rashid; in 60/680 he executed an ally of al-Husayn b. 'Ali b. Abi Talib, an enemy of the Umayyads'.[11] Six years later comes the first specific mention of a Turk fighting for the Muslims; he was a slave (a ghulam) who participated in the tribal warfare of Khurasan on the side of Tamim.[12]
When Qutayba b. Muslim faced a revolt by his Arabian soldiers in 96/715, he threatened them with his non-Arabian troops. The following passage comes from his speech to the rebellious soldiers:
"Who are the archers, those who with their arrows shot by sure hands can hit the eye of the enemy and sew his eyelids to his socket? "
Then the princes and nobles of Persia, the Turkish nobles, as well as those of Soghdia, Badeghi, Tokharistan, and Khurasan got up. They numbered more than 10,000 men, all of them capable archers who never missed the mark.
Pointing to them, Qutayba said " ... they are braver than the Arabians."[13]
Although highly respected as soldiers, this passage indicates that Turks had not distinguished themselves from other Central Asian groups by 96/715.
Just two years later the Muslims had an opportunity to acquire Turkish troops but passed it up. When they reached a truce with the pagan ruler of Tabaristan one of the conditions stipulated that "he deliver to them 500 Turks who had killed many Muslims and who then had taken refuge with him." Rather than incorporate these Turks into the Umayyad armies, the Muslims executed them.[14]
Some 35 years went by before the next mention of Turks in Muslim service; in 123/741 a Soghdian leader from Khurasan, Muqatil b. 'Ali traveled to Damascus to tell Caliph Hisham about conditions in Khurasan. On this trip, Muqatil b. 'Ali commanded a force of 150 Turks.[15] Two years later, when Muqatil b. 'Ali was appointed sub-governor of Amul, he and several other new appointees received instructions from Nasr b. Sayyar, the governor of Khurasan; in the event of Nasr's expulsion from Merv, they should import Turks and cross into Transoxiana.[16] Although this passage is not entirely clear, some Turkish military role can be adduced.
Rebellious Umayyads killed Caliph al-Walid II and his family in 126/744; a mawla "said to be Turkish" executed a grandson of al-Walid's.[17]
What role did Turks have in the Abbasid takeover? An historian of Turkish soldiers writes that "it may be assumed with certainty that Turkish units fought in Abu Muslim's army,"[18] though the sources (including the recently available Akhbar ad-Dawla al-'Abbasiya) do not substantiate this statement.[19] Further, I have found no mention of Turks in the service of the first Abbasid caliph, Abu'l-'Abbas.
The first mention of a Turk in Abbasid service comes only in the reign of the second caliph, al-Mansur; in 137/754, Zuhayr b. at-Turki was governor of both Hamadhan and Mosul.[20] This made him the first person of Turkish origins to hold a high position in Muslim service. Pseudo-Deyonosios Telmahriya', a Syriac chronicler, informs us of Turks in al-Mansur's army.[21] Another source states that at his death al-Mansur left 40,000 mawlas[22] and it is safe to assume that a good many of them were Turks.
Perhaps more important than their numbers, Turks under al-Mansur became trusted servants. Hammad at-Turki typified this new aspect. He had an important role in the building of Baghdad ;[23] he fought the 'Alid rebel al-Husayn b. 'Ali in 169/785 (one account reports that Hammad himself killed al-Husayn b. 'Ail) ;[24] he served al-Mansur personally as head of his bodyguard, chamberlain, and aide-de-camp (even taking care of such menial duties as saddling al-Mansur's horse);[25] finally, Hammad took charge of taxation in the Sawad region (ta'dil as-sawad).[26] Hammad appears to have been the first Turk with all-around importance in a caliph's entourage. Muslim writers noted both al-Mansur's heavy use of Turks and Hammad's special importance:
Al-Mansur was the first caliph to acquire Turks. He acquired Hammad and al-Mahdi [his successor; on him, more below] had Mubarak. Subsequent caliphs and everyone else followed them in this practice. [27]
The Turks did not stop increasing in numbers and eventually they overpowered the Abbasids.[28]
All evidence points to al-Mansur as the first Muslim ruler to bring substantial numbers of Turks into his service. A clue to al-Mansur's relations with his Turks may be gleaned from his statement about a Turkish slave who later served as governor of Egypt: "He is a man who fears me, not God!"[29]
Turks continue to appear in al-Mahdi's reign (158-69/775-85). The man who feared al-Mansur more than God served as governor of Egypt in the years 162-64/779-80; it is worth noting that not only was he a slave, but so were his brother and father; and his mother was the aunt of the king of Tabaristan.[30] A Turk led a rebellion in Fars; this man was related to a khadim (eunuch?) in the caliph's service, Faraj al-Khadim at-Turki (and they aided each other).[31] Turks fought in large numbers for the Muslims probably for the first time in 160/777 when they fought the Khariji rebel 'Abd as-Salam al-Yashkari.[32]
When al-Mahdi's son al-Hadi came to power in 169/785, he had to spend the first ten months of his reign fighting the rebel al-Husayn b. 'Ali in Arabia. The mawla Mubarak at-Turki was one of several Abbasid commanders;[33] al-Hadi thought he did not show enough vigor against the rebel and confiscated some of Mubarak's property.[34] As noted above, Hammad at-Turki also fought for the caliph in this campaign and is reported by one source to have killed al-Husayn b. 'Ali.[35]
Occasional mention of Turks in service continues through Harun al-Rashid's reign (170-93/786-809). Faraj al-Khadim at-Turki governed Tarsus in the first year of his rule.[36] Since Turks were ever associated with warfare, it comes as a surprise to hear of the Turkish mawla who was a religious scholar; he lived in Egypt and died there in 181/797.[37] The first mention of Turks in ceremonial occasions comes at about this time; when ambassadors arrived from India bearing gifts:
Harun al-Rashid ordered the Turks into two rows and armed them so heavily that only the pupils of their eyes showed. Then he let the ambassadors enter.[38]
A Turk relates an anecdote about the caliph and his vizier in 187/803;[39] as Töllner points out, this man must have been close to either the caliph or the vizier to be able make this report.[40] The first Turkish support for a sectarian movement came in 193/809 when they helped the rebel Rafi' b. Layth; this support was short-lived, however, for the Abbasid commander split the Turks from Rafi', so they fled and weakened him.[41] Finally, Harun al-Rashid's received 1,000 or 4,000 (presumably Turkish) slaves from Khurasan as kharaj (general tax or land tax).[42]
Al-Ma'mun and Al-Mu'tasim, 198-227/813-842
A gold dinar struck during the reign of al-Ma'mun. |
al-'Uyun wa'l-Hada'iq: "When [al-Ma'mun] reached Baghdad [in 204/819], he bought some of his soldiers in the market"[44]
al-Qazwini: "It is said that [Yahya b. Aktham, a high official under al-Ma'mun] made efforts to collect handsome youths so serve as slaves (mamluks) of the caliph [al-Ma'mun]. He said to them: 'If not for you, we would not be believers' [i.e. your defense of us keeps our religion safe]."[45]
an-Nuwayri: al-Ma'mun "was the first to take Turks into his service. [As a result] their price increased to the point that it cost 100,000 or 200,000 dirhams to purchase one of them."[46]
al-Maqrizi: al-Ma'mun "increased buying Turks so that their price rose to the point that he purchased a single slave (mamluk) for 200,000 dirhams."[47]
It is worth noting that some sources (such as the first two quoted above) refer to the purchase of slaves in general and others (such as the latter two) specifically to Turkish slaves. It would appear that Turks made up the most important group of slaves, though without predominating.
From the first, al-Ma'mun's younger brother, the future Caliph al-Mu'tasim, took an active part in the acquisition of military slaves, serving as al-Ma'mun's deputy.
Ibn Qutayba: "Al-Ma'mun ordered his brother Abu Ishaq [al-Mu'tasim] to acquire Turks, so he imported them [as slaves]."[48]
So great was al-Mu'tasim's role that some accounts lost sight of all the rulers who employed Turks before him:
Bal'ami: "He was the first Abbasid caliph to take Turks into his service."[49]
al-Maqrizi, as-Suyuti, an-Nahrawali: "He was the first to enter Turks on the Military Register."[50]
Al-Mu'tasim acquired this reputation by making a concerted effort to recruit Turks. Samarqand, a renowned slave trade center,[51] served as his main source.[52] The slaves had probably been captured first by local traders and rulers, then sold to agents of the Abbasids.[53] One of those agents, a man called Ja'far al-Khushshaki, recounted his activities:
Al-Mu'tasim sent me during the reign of al-Ma'mun to Samarqand, to Nuh b. Asad (the Samanid ruler there, 204-27/819-42] to purchase Turks. I presented him each year with a number of them.[54]
Al-Mu'tasim already began acquiring Turks, with al-Ma'mun's encouragement, right after the end of the civil war between al-Amin and al-Ma'mun. In 199/815 he purchased Itakh, possibly his first Turkish slave.[55] Al-Ma'mun received slaves from Samarqand already by 200/816.[56] In 204/819[57] al-Ma'mun "bought some his soldiers in the market" of Baghdad.[58] Al-Mu'tasim purchased Itakh, Ashnas, Wasif, and Sima ad-Dimashqi, all prominent figures in his own reign, in Baghdad.[59] He acquired Ashnas and Bugha al-Kabir already during al-Ma'mun's caliphate.[60] Yet, despite these early purchases, one account states that al-Mu'tasim began acquiring Turks in 220/835.[61]
Besides the organized purchase of Turks from Samarqand, the caliphs captured Inner Asian steppe peoples. In 211-12/826-27 al-Ma'mun received 2,000 Ghuzz Turks as captives from Kabul.[62] The early Tahirid governors of Khurasan owned numerous slaves which they presumably had captured (since the caliphs would surely not tolerate their purchasing slaves). Talha b. Tahir (governor 207-13/822-28) could give away 80 mamluks on a single occasion;[63] 'Abdallah b. Tahir (governor 213-30/828-45) sent as yearly tribute over 44 million dirhams and either 1012 slaves (raqiq) or 2,000 Ghuzz Turks captured in Khurasan.[64] The governor in Samarqand Nuh b. Asad, also sent slaves as part of his annual tribute to the caliph.[65]
As a result of these many efforts, the caliphs al-Ma'mun and al-Mu'tasim collected 1arge numbers of slaves. Al-Ma'mun had a sufficient number at his disposal to sent 400 handsome mamluks to ride in the service of a favorite cadi and high official;[66] one source says he had in all 70,000 slaves.[67] While al-Ma'mun still lived, al-Mu'tasim had three or four thousand slaves;[68] after his death al-Mu'tasim had 4,000, 8,000, 18,000, 38,000, 50,000 or 70,000 slaves.[69] Whatever the exaggeration of these figures, they point to the great efforts which the two caliphs made to collect Turkish slaves.
Turks did not achieve a predominant military role yet by al-Ma'mun's reign. They fought for him in the civil war against his brother al-Amin, usually under the command of Tahir b. al-Husayn.[70] Al-Amin lamented his loses and claimed that if only Tahir joined his side, the Turks and Daylamis could plot against him but they would fail;[71] this both confirms the Turkish presence and pays tribute to their military abilities. Rashid at-Turki led troops to Upper Egypt in 216/831.[72] Al-Jahiz recounts seeing a hundred Turkish cavalrymen line either side of a road for al-Ma'mun while on campaign. He was especially impressed by the fact that nearly all of them withstood an intense midday heat to remain on horseback.[73]
Besides soldiers, we also hear of two Turks serving the Abbasids as cooks during al-Ma'mun's caliphate; one worked for the father a vizier and the other for al-Mu'tasim.[74] When he died, al-Ma'mun was buried in a house which had formerly belonged to one of his Turkish eunuchs.[75]
Al-Mu'tasim's close association with Turks began already in al-Ma'mun's reign. As early as 202/818, when fighting a Khariji rebel, he had a bodyguard of Turkish slaves and one of them saved his life.[76] When he went to Egypt in 213/828, 4,000 Turks accompanied him; they then left with him two years later.[77]
Caliph al-Mu'tasim, the main figure associated with the import of Turks to Iraq, is remembered by a street name in Amman, Jordan. |
It was Ashnas[79] who defended al-Mu'tasim in the incident mentioned above; he saved the future caliph from an assailant's knife.[80] Ashnas undertook two expeditions for al-Ma'mun; in 215/830 he led Abbasid troops against the Byzantines and two years later he led them in Egypt.[81] When al-Mu'tasim came to power, he appointed Ashnas governor of Egypt.[82] Ashnas retained this title for some years without actually living in or presiding over Egypt. Indeed, the chroniclers ignore him almost completely when recounting the history of Egypt during his tenure. It appears that al-Mu'tasim wished to honor Ashnas with the governorship without losing his presence at the court. In 223/838 Ashnas led a variety of units on the way both to the battle of Amorium in Anatolia and on the return.[83] Again honoring Ashnas, al-Mu'tasim allowed him to sit on a throne (kursi) in 225/840.[84] A year later Ashnas went on the pilgrimage to Mecca and al-Mu'tasim paid him a yet greater honor by giving him control of every region through which be passed between Samarra and Mecca.[85] Accordingly, he is sometimes known as the governor of Syria, al-Jazira, and Egypt,[86] though again he never ruled those provinces. Ashnas died in 230/845.[87]
Itakh,[88] originally a cook, came to al-Mu'tasim in 199/815.[89] In 222/837 he joined the long campaign (220-23/835-38) against the Iranian rebel Babak.[90] A year later he led the right wing to Amorium and then the Turks and Farghanians at the battle there.[91] In 225/840 he became governor of the Yemen, but that same year we also have a report that he stood guard at al-Mu'tasim's doorway,[92] so this too was probably an honorary appointment. Two years later he fought a rebel near Mosul.[93] Soon after al-Mu'tasim's death he acquired the important position of governor of Khurasan;[94] al-Mutawakkil had him executed in 234/849.[95]
First mention of Bugha al-Kabir[96] comes in 210/825 when he took possession of someone else's lands.[97] He brought relief to the Abbasid troops fighting Babak in 220/835 and a year later led troops on his own.[98] Both to and from Amorium he led the rearguard;[99] and he served al-Mu'tasim as chamberlain (hajib).[100] After al-Mu'tasim's reign, Bugha al-Kabir filled several important positions and died in 248/862.[101] One account says he loved warfare and died at over 90 (lunar) years;[102] if true, then Bugha al-Kabir was already an adult when acquired by al-Mu'tasim.
Wasif[103] had a role at Amorium and served al-Mu'tasim as chamberlain;[104] however, like Bugha al-Kabir, his most important positions came after al-Mu'tasim's death. Wasif died in 253/867.[105]
Al-Afshin,[106] who was al-Mu'tasim's most prominent general, already led military campaigns for al-Ma'mun.[107] Under al-Mu'tasim he directed the battle against Babak and fought in the conquest of Amorium.[108] We shall not dwell on his important career and fascinating downfall, however, for although two sources call him a Turk,[109] he came from Farghana, an Iranian cultural region, and was not usually considered Turkish.[110]
Several other lesser-known Turks also had leading positions; Sima ad-Dimashqi, Sima ash-Sharabi, and Muhammad b. Hammad b. Danfash served al-Mu'tasim as chamberlains;[111] Bashir at-Turki led Farghanian troops in an ambush against Babak in 222/837.[112]
Anonymous Turks filled a variety of military roles: some became bodyguards, either for the caliph or for others;[113] Turks guarded Ibrahim b. al-Mahdi in 210/825 and they fought at Amorium.[114] They played a major part in the conspiracy of al-Ma'mun's son al-'Abbas against al-Mu'tasim.[115] For example, Ashnas was to be assassinated by a fellow Turk, his drinking companion.[116] When the conspiracy failed, it was again Turks who executed one of its leaders in 223/838.[117] Al-Afshin sent a Turkish mawla to kill one of Babak's men in 222/837.[118]
Turks provided personal services for al-Mu'tasim. On one occasion, he called for Itakh to bring him dates;[119] a Turkish slave eunuch swatted flies from al-Mu'tasim's head.[120] Turks served other persons too in non-military ways.[121]
Perhaps most indicative of the favor bestowed on the Turks was the fact that al-Mu'tasim raised one of them, al-Fath b. Khaqan, with his own son Ja'far, the future Caliph al-Mutawakkil.[122] The close relationship between these two lasted for decades and al-Fath played a vital role in al-Mutawakkil's reign.[123]
This survey should establish two points: (1) Turks served Muslim rulers, governments, and armies before al-Mu'tasim's reign while (2) they gained far greater importance during his reign. In light of the Turkish usurpation of Abbasid power soon after al-Mu'tasim's death, it is worth noting that the Turks showed no signs of disobedience or ambition during al-Mu'tasim's lifetime; yet he granted them so much power that within a few years of his death they effectively controlled the Abbasid caliphate.[124]
Jan. 1, 1980 update: For a companion study to this, see my "Black Soldiers in Early Muslim Armies," International Journal of African Historical Studies 13(1980): 87-04.
[1] H.A.R. Gibb, The Arab Conquests in Central Asia (London, 1923), p. 10.
[2] P.K. Hitti, History of the Arabs (9th ed rev. :New York, 1967), p. 210.
[3] R.N. Frye and A.M. Sayılı, "Turks in the Middle East before the Saljuqs," Journal of the American Oriental Society, 63(1943), p. 196.
[4] O S.A. Ismail, "Mu'tasim and the Turks," Bulletin of the School of Oriental and African Studies, 29(1966), p. 13. V. Minorsky, Sharaf al-Zaman Tahir Marvazil on China, the Turks and India (London, 1942), p. 92 notes that "Turks" include "Finno-Ugarian and Slavonic Peoples of Eastern Europe."
[5] Notably:
(1) P. G. Forand, "The development of military slavery under the Abbasid caliphs of the ninth century A. D. (third century A. H.), with special reference to the reigns of Mu'tasim and Mu'tadid." (Ph. D., Princeton University, 1962).
(2) S. Hamdi, "Die Entstehung und Entwicklung des turkishen Einflusses im 'Abba· sidenreich bis Mutawakkil." (Ph.D., University of Tubingen, 1954).
(3) Ismail.
(4) H. Töllner, Die türkischen Garden am Kalifenhof uon Samarra (Bonn, 1971).
Despite its title, D. Ayalon's "Military reforms of the Caliph al-Mu'tasim—their background and consequences", (Jerusalem, 1963. Mimeographed) hardly mentions the Turks.
I have not seen the first part of H. D. Yıldız, "Abbasiler devrinde Türk kumandanları," Istanbul Tarih Enstitüsü Dergisi, 1(1973). Part II is in idem., 2(1974), pp. 51·58 and deals with Itakh, a Turkish leader who began his career in al-Mu'tasim's time. I also know nothing more about his unpublished Ph. D. thesis than its title, "Mu'tasim devrinde Abbasī Ìmparatorluğu."
[6] Some works which deal with Turks in early Muslim service include:
(1) Ş. Günaltay, "Abbas oğuları imparatorluğunun kuruluş ve yükselişinde Türklerin rolü," Belleten, 6(1942), pp. 177-205.
(2) J. Karabacek, "Erstes urkundliches Aufreten von Türken," Mitteilungen aus der Sammlung der Papyrus Erzherzog Rainer, 1(1887), pp. 93-108.
(3) Z. al-Kitabchi, "at-Turk fi mu'allifat al-Jahiz wa makanathum fi't-tarikh al-Islami ila awasit al-qarn ath-thalith al-hijri." (Ph. D., University of Karachi [date not available]).
(4) E. Mainz, "Die Türken in der klassichen arabischen Literatur," Der Islam, 21 (1973), 279-285.
[7] al-Baladhuri, Futuh al-Buldan, ed. M. J. de Goeje (Leiden 1866), p. 411; at-Tabari, Tarikh ar-Rusul wa'l-Muluk, ed. M. J. de Goeje et al. (Leiden, 1879-1901), 2. 169.
[8] al-Baladhuri, p. 410; Ibn al-Faqih al-Hamadhani, Kitab al-Buldan, ed. M. J. Goeje (Leiden, 1885), p. 191; Narshakhi, Tarikh-I Buldan, tr. R. N. Frye, The History of Bukhara (Cambridge, Mass., 1954), p. 37; Yaqut, Mu'jam al-Buldan, ed. F. Wustenfeld (Leipzig, 1866-72), 1. 522.
[9] at-Tabari, 2. 170.
[10] al-Baladhuri, p. 376.
[11] at-Tarabi, 2. 268.
[12] at-Tabari, 2. 698.
[13] Bal'ami, Tarjama-yi Tarikh-i Tabari, tr. H. Zotenberg, Chronique de Tabari (Nogent-le-Rotrou, 1867-74), 4. 209. So far as I can tell, this passage is not found in at-Tabari. See al-Baladhuri 424.
[14] Ibn A'tham, Kitab al-Futuh, ed. M. 'Abd al-Mu'id Khan et al. (Hyderabad, 1388/1968-), 7. 293.
[15] at-Tabari, 2. 1719.
[16] at-Tabari 2. 1767.
[17] at-Tabari , 1805.
[18] Tollner 14.
[19] Günaltay's article on this subject turns up very little.
[20] at-Tabari, 3. 118.
[21] Makhtebanut, tr. J. B. Chabot, Chronique de Denys de Tell-Mahre (Paris 1895), 4. 72.
[22] al-Rashid b. az-Zubayr, adh-Dhakha'ir wa't-Tuhaf, ed. M. Hamidullah (Kuwait, 1959), p. 213.
[23] at-Tabari, 3. 118.
[24] Abu 'l-Faraj al-Isfahani, Maqatil at-Talibiyin, ed. S. A. Saqr (Cairo 1368/1949), p. 451.
[25] at-Tabari, 3. 280, 309, 392; al-Ma'udi, Muruj adh-Dhahab, ed. M. M. 'Abd al-Hamid (5th ed.: Cairo, 1393/1973).
[26] al-Jahshiyari, Kitab al-Wuzara' wa'l-Kuttab, ed. M. Saqa (Cairo, 1357/1938), p. 134.
[27] ath-Tha'alibi, Lata'if al-Ma'arif, ed. I. al-Abyari and H. K. as-Sayrafi (Cairo, 1960), p. 20.
[28] Ibn Badrun, Sharh Qasidat Ibn 'Abdun, ed. M.S. Al-Kurdi (Cairo 1340), p. 285.
[29] al-Kindi, Kitab al-Wulah wa Kitab al-Qudah, ed. R. Guest (Leiden, 1912), pp. 122-23.
[30] Ibid.
[31] al-Jahshiyari, p. 151; at-Tabari, 3. 604 calls him at-Turki.
[32] Khalifa b. Khayyat, Kitab at-Tarikh, ed. S. Zakkar (Damascus 1967-68), p. 701.
[33] at-Tabari, 3. 562.
[34] Ibid. 3. 563. That property may have been a land grant (qati'a) Mubarak owned in Baghdad mentioned in al-Ya'qubi, Kitab al-Buldan, ed. M. J. de Goeje (Leiden, 1892), p. 253. Mubarak must have been very rich, for he could grant a million dirham loan. (at-Tabari, 3. 981, al-Jahshiyari, p. 100)
For al-Haid's anger, see Abu'l Faraj al-Isfahani, p. 452. Yaqut's report on 3. 852 that Mubarak killed al-Husayn b. 'Ali appears to have confused Mubarak with Hammad.
[35] Abu'l Faraj al-Isfahani, p. 451.
[36] at-Tabari, 3. 604.
[37] Ibn Taghri Birdi, an-Nujum az-Zahira, ed. A. Z. al-'Adawi et al. (Cairo 1348-92/1929-72), 2. 103. A freed slave from Farghana became an expert on hadith and a poet (Ibn Qutayba, Kitab al-Ma'arif, ed. F. Wustenfeld [Gottingen, 1850], p. 270).
[38] Ibn 'Abd Rabbih, al-'Iqd al-Farid, ed. A. Amin et al. (Cairo 1940-53), 2. 203.
[39] at-Tabari, 3. 683.
[40] Töllner, p. 18.
[41] at-Tabari 3. 775.
[42] Ibn Hamdun, at-Tadhkira, 2. 234, cited by Hamdi, p. 9, fn 3.
[43] "From Mawla to Mamluk: the Origins of Islamic Military Slavery" (Ph. D., Harvard University, 1978), pp. 184-88.
[44] al-'Uyun w'al-Hada'iq, M. J. de Goeje (Leiden, 1869-1871) in Fragmenta Historicorum Arabicorum, p. 379.
[45] al-Qazwini, Athar al-Bilad wa khbar al-'Ibad (Beirut, 1389/1969), p. 318.
[46] Quoted on p. 255 in an editor's footnote to al-Ya'qubi, Kitab al-Buldan.
[47] al-Maqrizi, an-Niza' wa't-Takhasum, ed. G. Vos (Leiden, 1888), p. 63.
[48] Ibn Qutayba, Kitab al-Ma'arif, ed. T. 'Ukasha (Cairo 1969), p. 391. This quote is missing from Wustenfeld's edition.
[49] Bal'ami, 4. 524.
[50] al-Maqrizi, Kitab as-Suluk, ed. M. Ziyada (Cairo, 1935-1973), 1. 16; as-Suyuti, Tarikh al-Khulafa', ed. M. M. 'Abd al-Hamid (3rd ed.: Cairo, 1383/1964), 24; an-Nahrawali, al-A'lam bi-A'lam Bayt Allah al-Haram, ed. F. Wustenfeld (Gottingen, 1272/1857), p. 123.
Another indication of this comes from Yaqut 4. 454; Mubarak, who was al-Mahdi's favorite, is incorrectly called mawla of either al-Mu'tasim or al-Ma'mun.
[51] al-Muqaddasi, Ahsan at-Taqasim fi Ma'rifat al-Aqalim, ed. M. J. Goeje (Leiden, 1877), p. 278.
[52] al-Ya'qubi , p. 255; Ibn Hawqal, Kitab Surat al-Ard, ed. J. H. Kramers (Leiden 1939), p. 468; Ibn Khaldun, Kitab al'Ibar (Bulaq, 1284), 3. 257; Ibn Taghri Birdi, an-Nujum az-Zahira (Cairo 1929-72), 2. 233; an-Nahrawali, p. 123.
[53] al-Balawi, Sirat Ahmad b. Tulun, ed. M. K. 'Ali (Damascus 1358/1939), p. 34 calls Tulun a captive (sabi).
[54] al-Ya'qubi, pp. 255-56.
[55] at-Tarabi, 3. 1383.
[56] al-Balawi, p. 34; Ibn Jawzi, al-Muntazam (Hyperbad, 1357-59), 5. 71; Ibn Khallikan, Wafiyat al-A'yan, ed. I. 'Abbas (Beirut, 1968-72), I. 173.
[57] "When he reached Baghdad," i.e., in 204/819.
[58] al-'Uyun wa'l-Hada'iq, p. 379.
[59] al-Ya'qubi, p. 256.
[60] Bal'ami, 4. 534.
[61] Ibn Taghri Birdi, 2. 233.
[62] Ibn Khurdadhbih, al-Masalik wa'l-Mamalik, ed. M. J. de Goeje (Leiden, 1889), p. 37.
[63] Tayfur, Kitab Baghdad, ed. H. Keller (Leipzig, 1908), p. 172.
[64] al-Muqaddasi, p. 340 and Ibn Khurdadhbih, p. 39.
[65] al-Balawi, p. 33; Ibn al-Jawzi, 5. 71; Ibn Khallikan, 1. 173; Ibn Taghri Birdi, 3. 1.
[66] al-Qazwini, p. 318.
[67] Ibn Dihya al-Kalbi, an-Nibras fi Tarikh Khulafa al-'Abbas, ed. 'A. al-'Azzawi (Baghdad, 1365/1946), p. 65.
[68] al-Ya'qubi, pp. 255-56; al-Kindi, p. 188-89; Ibn Taghri Birdi, 2. 208-09.
[69] al-Mas'udi, 4. 53; Ibn Taghri Birdi, 2. 233; Ibn Kathir, al-Bidaya wa'n-Nihaya (Cairo, 1348-58), 10. 296; Bar Hebraeus, Makhtebanut Zabhne, ed. and tr. E. W. Budge, The Chronography of Gregorus Abu'l-Faraj (London 1932), 1. 140; al-Khatib al-Baghdadi, Tarikh Baghdad (Cairo, 1931), 3. 346; ar-Rashid b. az-Zubayr, p. 214; al-Qazwini, p. 385; Yaqut, 3. 16.
[70] at-Tabari, 3. 799, 891.
[71] al-Mas'udi, 3. 419.
[72] al-Kindi, p. 192.
[73] al-Jahiz, Manaqib al-Atrak in Majmu'at ar-Rasa'il, ed. M. as-Sasi al-Maghribi (Cairo, 1324), p. 37.
[74] Tayfur, p. 217, 268; at-Tabari, 3. 1234, 1383.
[75] at-Tabari, 3. 1140; al-Mas'udi, at-Tanbih wa'l-Ashraf, ed. M. J. Goeje (Leiden 1894), p. 190 calls him a Turk.
[76] at-Tabari, 3. 1107; Miskawayh, Tajarub al-Umam, ed. M. J. Goeje (Leiden, 1869-71) in Fragmenta Historicorum Arabicorum, I. 438; Abu Zakariya Yazid al-Azdi, Tarikh al-Mawsil, ed. A. Habiba (Cairo, 1967), p. 352.
[77] al-Kindi, p. 188-89; Ibn Taghri Birdi, 2. 208-209.
[78] al-Mu'tasim himself considered these men (naming them all but Bugha al-Khabir) as his most important agents (at-Tabari, 3. 1327).
[79] Ashnas is called a Turk: at-Tabari, 3. 1017, 1306, 1338; al-Mas'udi, Muruj adh-Dhahab, 4. 55, 60; al-Ya'qubi, Kitab al-Buldan, p. 259; al-Ya'qubi, at-Tarikh (Beirut, 1960), 2. 475, 479, 481; Bal'ami, 4. 524; Ibn Taghri Birdi, 2. 243, 245, 255-56, 274; Miskawayh, 438; Ibn Khallikan, 3. 89; Abu Zakariya Yazid al-Azdi, p. 352; Ibn al-'Adim, Zubat al-Halab min Tarikh al-Halab, ed. S. ad-Dahhan (Damascus, 1951-1968), 1. 69.
Ghulam: all the references in note 76.
Mamluk: al-Ya'qubi, Kitab al-Buldan, p. 256.
Mawla: Yaqut, 3. 16.
Mawla amir al-mu'minin (on this term, see "From Mawla to Mamluk," pp. 128-33): P. Balog, Umayyad, 'Abbasid and Tulunid Glass Weights and Vessel Stamps (New York, 1976), pp. 240-243 (five examples).
[80] See the references at note 76.
[81] at-Tabari, 3. 1103 and al-Kindi, p. 192.
[82] Al-Kindi, p. 194; Ibn Taghri Birdi, 2. 229; Abu Zakariya Yazid al-Azdi, p. 416.
[83] To Amorium: (Vanguard) at-Tabari, 3. 1236; al-Ya'qubi, at-Takrikh, 2. 475; (cavalry) at-Tabari, 3. 1241: (infantry) al-'Uyun wa'l-Hada'iq, p. 393; (left wing) at-Tabari, 3. 1244. From Amorium: (right wing) at-Tabari, 3. 1260; (rearguard) at-Tabari, 3. 1261-62.
[84] at-Tabari, 3. 1302; al-'Uyun wa'l-Hada-iq, p. 404.
[85] at-Tabari, 3. 1318.
[86] Ibn al-'Adim, 1. 69.
[87] at-Tabari, 3. 1338.
[88] Itakh is called a Turk: at-Tabari, 3. 1306, 1327; al-Mas'udi, Muruj adh-Dhahab, 4. 60; al-Ya'qubi, at-Tarikh, 2. 479, 481, 485; Bal'ami, 4. 524; Ibn Taghri Birdi, 2. 243, 255, 245, 274.
Khazar: at-Tabari, 3. 1383; Ibn Taghri Birdi, 2. 276.
Ghulam: at-Tabari, 3. 1383.
Mamluk: al-Ya'qubi, Kitab al-Buldan, p. 256; Ibn Taghri Birdi, 2. 276.
Mawla amir al-mu'minin: Balog, pp. 246, 248. On his career, see Yudez.
[89] at-Tabari, 3. 1234, 1383; Abu Zakariya Yazid al-Azdi, p. 424.
[90] at-Tabari, 3. 1195.
[91] at-Tabari, 3. 1236, 1250.
[92] at-Tabari, 3. 1303, 1307, 1327.
[93] at-Tabari 3. 1322.
[94] al-Ya'qubi, at-Tarikh, 2. 479.
[95] at-Tabari, 3. 1383-87; Ibn Taghri Birdi. 2. 276.
[96] Bugha al-Kabir is called a Turk: at-Tabari, 3. 1313; al-Ya'qubi, at-Tarikh, 2. 478; Ibn Taghri Birdi, 2. 218, 327.
[97] at-Tabari, 3. 1085.
[98] at-Tabari, 3. 1174, 1186-93.
[99] Abu Zakariya Yazid al-Azidi, p. 427; at-Tarabi, 3. 1261.
[100] al-Mas'usi, at-Tanbih, p. 356; Eutychius, Tarikh al-Majmu'a, ed. L. Cheicho, et al. (Beirut, 1906-09), 2. 61 (but not on page 284).
[101] at-Tabari, 3. 1506; Ibn Taghri Birdi, 2. 327.
[102] Ibn Taghri Birdi, 2. 327.
[103] Wasif is called a Turk: at-Tabari, 3. 1351, 1479, 1531, 1559, 1687; Ibn Habib, Kitab al-Muhabbar, ed. E. Lichtenstadter (Hyderabad, 1361/1942), p. 260; al-Ya'qubi, at-Tarikh, 2. 478 Ibn Taghri Birdi, 2. 327, 338, 340; Eutychius, 2. 61-62.
Mamluk: al-Ya'qubi, Kitab al-Buldan, p. 256; Ibn Taghri Birdi, 2. 340.
Mawla: al-'Uyun wa'l-Hada'iq, pp. 409-10; Ibn 'Abd Rabbih, 5. 121; at-Tabari, 3. 1481.
Malwa amir al-mu'minin: at-Tabari, 3. 1484-85; al-Baladhuri, p. 235.
[104] A role at Amorium: at-Tabari, 3. 1237.
As hajib: al-Ya'qubi, at-Tarikh, 2. 478; al-'Uyun wa'l-Hadah'iq, pp. 409-10; Ibn Habib, p. 260; Ibn 'Abd Rabbih, 5. 121; Eutychius, 2. 61.
[105] al-Y'qubi, at-Tarikh, 2. 502; Ibn Taghri Birdi, 2. 338, 240.
[106] Al-Afshin is called a Turk: Masalik al-Mamalik, ed. M. J. de Goeje (Leiden, 1870), p. 292; J. Saint Martin, Memoires historiques et geographiques sur l'Armenie (Paris, 1818-19), 1. 344 relying on the Armenian historians.
Mawla: ad-Dinawari, al-Akhbar at-Tiwal, ed. A. M. 'Amir and G. ash-Shayyal (Cairo, 1960), p. 403.
Mawla amir al-mu'muinin: al-Qalqashandi, Subh al-A'sha (Cairo, 1913-22), 6. 404.
[107] at-Tabari, 3. 1105, 1106.
[108] at-Tabari, 3. 1170-1234, 1236-56.
[109] See note 106 above.
[110] Al-Afshin's trial, at-Tabari, 3. 1303-18, is permeated with the fact of his Iranian culture; not especially 3. 1312 and 1315.
[111] al-Ya'qubi, at-Tarikh, 2. 478; Ibn 'Abd Rabbih, 5. 121. Muhammad b. Hammad b. Danfash's name seems to indicate that his father was already a Muslim.
[112] at-Tabari, 3. 1215-16.
[113] at-Tabari, 3. 1076, 1289.
[114] at-Tabari, 3. 1076, 1250; al-'Uyun wa'l-Hada'iq, p. 394.
[115] at-Tarabi, 3. 1267; Miskawayh, pp. 501-12.
[116] at-Tabari, 3. 1257, 1266.
[117] at-Tarbi, 3. 1265.
[118] at-Tarabi, 3. 1194.
[119] at-Tarabi, 3. 1325.
[120] al-Mas'udi, Muruj adh-Dhahab, 4. 50.
[121] Abu'l-Faraj al-Istahani, al-Aghani, (Bulaq, 1284-85), 7. 155.
[122] Ibn Taghri Birdi, 2. 325.
[123] O. Pinto, "Al-Fath b. Haqan, favorite di al-Mutawakkil," Revista degli studi orientali, 13 (1931-32), pp. 133-49.
[124] Erratum: After submitting this article to press, the author became aware of a mistranslation. Please ignore the quote from al-'Uyun wa'l-Hada'iq above notes 44 and 58.