A new Muslim source on the Hungarians in the second half of tenth century CHRONICA (SZEGED) 4: 22-31 (2004)

A New Muslim Source on the Hungarians in the Second Half of Tenth Century ISTVÁN ZIMONYI The tenth century was the golden age of the Muslim culture. The political centers of the Islamic civilizations such as Buchara in the East, Baghdad and Cairo in the central regions, and Cordova in the West had developed high scientific and liter- ary levels. The neighboring and even remote non-Muslim lands were also well known if they were parts of the worldwide commercial system. Europe was in close contact with Muslim Spain, while Eastern Europe stood in the middle of the area of interest of the Samanids and Baghdad. The Hungarians living in the Car- pathian Basin attracted attention from both ends of the Muslim world. The Samanid wazir, al-Űayhání, preserved a discourse on the Hungarians before they conquered the Carpathian Basin.1 In the tenth century the cartographer al-Balhi and his followers al-Istahri and Ibn Hauqal 2 and the traveler al-Mas'üdl 3 gave ac- counts of the Hungarians. Most of this information reached Andalusia, as is at- tested in the Andalusian author al-Bakri's geographical chapters on the Hungari- ans based on the books of al-Gayhani and al-Mas'üdl. The work of al-Bakri, enti- tled The Book of Routes and Kingdoms, was published in parts until recently. In the 1970s Károly Czeglédy discovered a new Andalusian Muslim source concerning 1 T. Lewicki, trôdla arabskie do dziejôw slowianszczyzny. [Arabic sources on the history of the Slavs] T. 2/2. Wroclaw-Warszawa-Kraköw 1977, 32-35, 94-107; H. Göckenjan- I. Zimonyi, Orientalische Berichte über die Völker Osteuropas und Zentralasiens im Mittelalter. Die Öayhäni-Tradition (Ibn Rusta, Gardïzï, Hudüd al-'Alam, al-Bakri und al-Marwazï). Ver- öffentlichung der Societas Uralo-Altaica Band 54, Wiesbaden 2001, 64-75,172-178, 210, 228, 252. 2 D. M. Dunlop, The History of the Jewish Khazars. Princeton 1954, 98; Ibn Hauqal, Configu- ration de la terre (Kitab surat al-ard), par J. H. Kramers, G. Wiet, Beyrouth-Paris 1964. 3 Mas 'ûdï, Les Prairies d'or. Traduction française de B. de Meynard et P. de Courteille re- vue et corrigée par Ch. Pellat. Tome I, Paris 1962,177-179. 22  A NEW MUSLIM SOURCE ON THE HUNGARIANS ... the Hungarian raid against Andalusia in 942, which also contained a description of their country.4 It is the work of Ibn Hayyán called Muqtabis.5 The existence in Cordova of a good knowledge of the Hungarians forecast more sources. As mentioned above, al-Bakri's geographical compendium was not been com- pletely edited until recently. In 1878 two Russian orientalists, Kunik and Rosen, published nine excerpts from the Istanbul manuscript with reference to Eastern Europe.6 Besides al-Gayhani and al-Mas'üdl he quoted the Jewish diplomat and traveler Ibrahim ibn Ya'qüb, who visited Emperor Otto I in Merseburg in 965 and spent some time in Prague. His report provided an account of the Slavs, Bohemia, Poland, and Bulgaria, including a short reference to the Hungarians, whom he called Turks. In 1946, Kowalski issued a critical edition with Polish and Latin translations and commentary.7 The Hungarian orientalist Mihály Kmoskó, who translated Muslim geographical literature into Hungarian and German, recom- mended the study of all the literary remains of al-Bakrl because they could hold new data on Hungarian history.8 In 1968, Hajji edited the chapters on Andalusia and Europe from the geographical al-Bakri's book. 9 In this edition I discovered a parallel account of Hárün ibn Yahyá, known from the work of Ibn Rusta, con- taining a description of the Byzantine bodyguard of Khazars and Turks. 10 Ac- cording to the Byzantine usage "Khazars" meant Hungarians in the service of the Byzantine emperor. In 1992, al-Bakri's The Book of Routes and Kingdoms appeared 4 K. Czeglédy, "Új arab forrás a magyarok 942. évi spanyolországi kalandozásáról." [New Arabic source on the raid of the Magyars in Spain in 942] Magyar Nyelv 75 (1979), 273-285. 5 Ibn Hayyán, Al-Muqtabas. V. ed. P. Chalmeta, E. Corriente, M. Subh, Madrid 1979,481- 483; The Hungarian translation with commentary by I. Eltér, "A magyar kalandozás- kor arab forrásai." [Arabic sources on the period of Magyar raids] in A honfoglaláskor írott forrásai, ed. L. Kovács, L. Veszprémy, Budapest 1996,174-179. 6 A. Kunik, V. Rozen, Izvestija al-Bakri i drugyh avtorov o Rusi i Slavjanah. [Reports of al- Bakri and other authors on Rus and the Slavs] Sanktpetersburg 1878. 7 Relacja Ibráhima ibn Ja 'küba z podrózy do krajów slowianskich w przekazie al-Bekriego. [Ibra- him ibn Ja'küb's description of the country of the Slavs in the work of al-Bekri] ed. T. Kowalski, Kraków 1946. English translation: D. Mishin, "Ibrahim ibn-Ya'qub at- Turtushi's Account of the Slavs from the Middle of the Tenth Century," in Annual of Medieval Studies et the CEU1994-1995. Budapest 1996,184-199. 8 M. Kmoskó, Mohamedán írók a steppe népeiről. Földrajzi irodalom. [Muslim writers on the steppe peoples. Geographical literature] Vol. 1/1. ed. I. Zimonyi, Magyar őstörténeti Könyvtár 10, Budapest 1997, 80. Kmoskó translated and commented on the nine frag- ments published by Kunik and Rosen: M. Kmoskó, Mohamedán írók a steppe népeiről. Földrajzi irodalom. [Muslim writers on the steppe peoples. Geographical literature] 1/2. ed. I. Zimonyi, Magyar Őstörténeti Könyvtár 13, Budapest 2000,229-258. 9 The Geography of al-Andalus and Europe from the Book "al-masalik zval-mamalik" by Abú 'Ubayd Al-Bakri. ed. Abdurrahman Ali El-Hajji, Beirut 1968. 10 I. Zimonyi, "Why were the Hungarians Referred to as Turks in the Early Muslim Sources?" in Néptörténet - Nyelvtörténet. A 70 éves Róna-Tas András köszöntése. L. Károly, É. Kincses Nagy, Szeged 2001,202-203. 23  ISTVÁN ZIMONYI in a critical edition based on ten manuscripts. 1 1 1 have compared the chapters on Slavic and n o m a d i c peoples in Eurasia to b e found in the earlier publications with those of the n e w edition. T h e Rus' chapter of the a l - G a y h a n i - t r a d i t i o n I f o u n d in the description of Southeastern Europe, i n c l u d the f o l l o w i n g geo- graphical accounts: Thrace, Thessaly, Macedonia, Anqilis and Rus'. Reading these chapters it b e c a m e evident that the n a m e Anqilis meant Hungarians. The account preserved b y al-Bakrl is the following: Discourse on the country of al-Unqulus 1 2 They are Turkic people 1 3 w h o live adjacent to the Slavs. 1 4 T h e border of their country is the country of B u w a y r a 1 5 and the country of B ü y a s l á w 1 6 in the west; north 1 7 of them are the Russians; 1 8 east of them are the Pechenegs 1 9 and the unin- 11 Kitáb al-Masálik wa-1-Mamálik d'Abu Ubayd al-Bakri. Edition critique avec introduc- tion et indices A. P. Van Leeuwen et A. Ferre, Qartág 1992. 12 The form of the ethnic name is based on the contemporary Latin sources in Spain: Un- guli cf. R. Szántó, "Spanyolországi források a kalandozó magyarok 942. évi hadjáratá- ról." [Spanish sources on the campaign of the raiding Magyars in 942] Acta Universitas Szegediensis de Attila József Nominatae. Acta Histórica 103 (1996), 43-48. 13 The expression gins min al-atrák 'a people from the Turks (plural)' corresponds with that of the al-Cayhárű-tradition gins min at-turk 'a people from the Turks (singular)' Zimonyi, "Why were the Hungarians," 203 note 17. The Muslim authors of the ninth and tenth centuries called Turks the nomadic peoples inhabiting the steppe between Mongolia and the Lower Volga and those moving westward from east of the Volga such as the Pechenegs and Hungarians. 14 The term as-Saqáliba and its interpretations: A. Nazmi, Commercial Relations between Ar- abs and Slavs (9th-llth centuries). Warsawa 1998,86; D. Misin, Sakaliba (slavjane) v islam- skom mire v rannee srednevekov'e. [The Saqaliba (Slavs) in the Islamic world in the early Middle Ages] Moskva 2002,50-60. 15 Buwayra was identified with Paris by Hajji on the basis of the following passage on the Frankish Empire: "The Franks (al-ifranga) were the descendants of Jafet just as the Galls (al-galáliqa), Slavs, al-Isban, Turks, Khazars, burgán, Alans, Gog and Magog. The Franks are Christians namely Melkits. Their capital is Bariza (MSS: B.wyra). It is an enormous town." (Hajji, The Geography of al-Andalus, 137-138; Leeuwen-Ferre, Kitáb al- Masálik, 340; Kowalski, Relacja Ibráhima, 127, note 184). The description was copied from the work of al-Mas'üdl (Kmoskó, Mohamedán írók, 1/2, 201). The identification of 'Bayern' with Bavarians can be preferred from philological and geographical point of view, as the next neighbor is Bohemia. 16 Ibráhlm ibn Ya'qüb used the same name, the country of Boleslav, for the Bohemian kingdom (Kowalski, Relacja Ibráhima, 48-51, 60, note 13; Mishin, Ibrahim ibn-Ya'qub, 185-187). There were two kings under the same name: Boleslaw I (929-967 or 973) and Boleslaw II (967/9737-999) R. Turek, "Boleslav I," II. in Lexikon des Mittelalters, Vol. 2, 357-359. 17 The Arabic gawf 'north' cf. Kowalski, Relacja Ibráhima, 56-58, note 5. 18 The Kievan Rus' became a neighbouring power after Vladimir took the cities Cerven and Peremysl in Eastern Galicia in 981 (The Russian Primary Chronicle. Laurentian Text. tr. and ed. S. H. Cross, O. P. Sherbowitz-Wetzor, Cambridge 1953, 85; cf. C. Goehrke, Frühzeit des Ostslaventums. Unter Mitwirkung von U. Kalin. Darmstadt 1992,38-47. 24  A NEW MUSLIM SOURCE ON THE HUNGARIANS . . . habited deserts.20 These are between the country of the Pechenegs and the coun- try of the Bulghars21 belonging to the Slavs. To the south are some parts of the country of the Bulghars and a strip of the uninhabited deserts.22 The geographical coordinates refer to Hungarians living in the Carpathian Ba- sin. The account was published by Hajji in 1968. He read Inqilish and identified the ethnonym with English, remarking that there is a contradiction between the title and the contents, as the context of the description fits a country between Eastern and Western Europe, i.e. Hungary. How can this inconsistency be ex- plained? Hajji suggested that there is a gap in the text. The excerptor wrote the title, then he copied a text from another account. The quoted passage has been preserved in two manuscripts in Morocco (the Qarawayyin Library of Fez and the National Library of Rabat). The manuscript of Rabat is a copy of the Fez manuscript.23 This complicated argument can be omitted if the Arabic letters are read as Ungulus and it is identified with Ungarus, the Western name of the Hun- garians. Mention must be made of an ethnonym recorded by Ibrahim ibn Ya'qub in the following context: "Many Northern tribes speak Slavic languages, for they are mixed with the Slavs. Among them are the Tadaskiyin (Germans), Anqaliyin (Hungarians), the Pechenegs, the Russians, and the Chazars." 24 According to Ku- nik and Lewicki the term Anqaliyin means Hungarians.25 Marquart preferred the Slavic tribe Uglic,26 while Kmosko did not accept the identification with the Hun- 19 Constantine Porphyrogenitus mentioned Kato Gyla among the Pecheneg tribes as the neighbours of the Hungarians in his De adminstrando imperio in the middle of the tenth century (Gy. Moravcsik Gy. Az Árpád-kori magyar történet bizánci forrásai. [Byzantine sources of the Hungarian history in the Arpadian period] Budapest 1984,41). 20 The Arabic word qafr means here uninhabited border, which was applied by the Hun- garinas as a defense against the neighboring kingdoms: cf. A. Miquel, La géographie humaine du monde musulman jusq'au milieu du 11e siècle. Vol. 3, Paris-La-Haye 1980, 73- 74; G. Vékony, "A gyepű szerepe az etnikai és politikai átalakulásokban." [The role of the 'gyepű' in the ethnic and political transformation] in Nomád társadalmak és államala- kulatok, ed. F. Tőkei, Budapest 1983,215-236. 21 Bulqàrïn. The same is recorded by Ibrahim ibn Ya'qüb. He gave a detailed description on the country of the Danubian Bulghars: cf. Kowalski, Relacja Ibrähima, 48, 51, 59, note 12; Mishin, Ibrahim ibn-Ya'qub, 184,189; Kmoskó, Mohamedán írók, 1/2,240,245-246. 22 Hajji, The Geography of al-Andalus, 150-151. The editor used the MS of Qarawiyyïn de Fès N. L. 390/80 (pages: 123a-b) (Q), and it was supplemented by the MS of Rabat (T) Hajji, The Geography of al-Andalus, 150, note 1. 23 Hajji, The Geography of al-Andalus, 150, note 1. 24 Hajji, The Geography of al-Andalus, 181, Leeuwen-Ferre, Kitäb al-Masälik, 1992, 336; Kowalski, Relacja Ibráhima, 111-115; Mishin, Ibrahim ibn-Ya'qub, 190. 25 Kunik-Rozen, Izvestija al-Bakri, 107; T. Lewicki, '"Kitab ar Raud al-mi'tar' Ibn A b d al- Mun'ima al-Himjari (XV v.) kak istocnik svedenij o Vostocnoj, Central'noj i Severnoj Evrope," [Kitab ar-Raud al-mi'tar' Ibn A b d al-Mun'ima al-Himjari (15th century) as the source of information about Eastern, Central and Northern Europe] in Problemy Vostokovedenija, T. 3, Moskva 1960,129-136. 26 J. Marquart, Osteuropäische und ostasiatische Streifzüge. Berlin 1903,510. 25  ISTVÁN ZIMONYI garians, as they were mentioned earlier as Turks.27 Elter reads al-Unqaliyin as, the plural of al-Unqali, the name of the Hungarians in the Carpathian Basin, as con- trary to the term al-Maggariya, denoting the nomadic Hungarians north of the Black Sea before the conquest of the Carpathian Basin. 28 Lewicki points out that the two forms, al-Unqaliyin and al-Unqalus, were variants of an original Ungarus. The name Unqalus seems to be a Spanish form and Unqaliyin could be the cor- ruption of the Muslim author based on the misinterpretation of the final Arabic letter. Based on Levi-Proven?al,29 Lewicki revised his earlier identification with English, concluding that al-Unqalus was a designation for the Hungarians because in the work of al-Bakrl the name for England is Bartaruya and the geographical position of the country of al-Unqalus precludes another possibility than Hungary. Lewicki emphasized that al-Bakri could not have used the report of Ibrahim ibn Ya'qub for the discourse on the Hungarians because Ibrahim ibn Ya'qub called the Hungarians Turks following the practice of the Byzantines. 30 Hungarian and European historians have not devoted attention to the notes of Hajji and Lewicki. The latter author has taken into consideration the parallel data preserved in the book of al-Himyarl (d. 1494), who lived in the Muslim west and composed a geographical dictionary in 1464. Al-Himyari used three geographical compendia as his sources: the geographical treatises of al-Bakri, al-ldrisl and a rewritten form of al-Bakri.31 Al-Himyari has an entry on al-Unqulus including additional information. Lewicki proved that the source of these new data was the book of al-Bakri. The surplus was published in parentheses in the critical edition of al-Bakrf s geographical work. The text of al-Himyari reads: The Country of al-Unqulus They are Turkic people, who live adjacent to the Slavs. The border of their town is the town of Buwayra in the west; north of them are the Russians. They are a sort of people who worship only God [Allah], may He be exalted! They believe in the Lord of the Sky,32 he is alone the Almighty. They avoid eating 27 Kmoskó, Mohamedán írók, 1/2,246, note 163. 28 I. Elter, "A magyarok elnevezései az arab forrásokban," [Designations of the Magyars in Arabic sources] in Honfoglalás és nyelvészet. ed. L. Kovács, L. Veszprémy, Budapest 1997,100. 29 La Péninsula Ibérique au moyen âge d'après le Kitäb ar-Rawd al-mi 'târ fi habar al-aktâr d'Ibn 'Abd al-Mun 'im al-Himyarï. Text arabe des notices relatives à l'Espagne, au Portugal et au Sud-Ouest de la France publié avec un introduction, un répertoire anaytique, une traduction annotée, un glossaire et une carte par E. Lévi-Provençal, Leiden 1938, 33. 30 Lewicki, Kitab ar Raud, 131-132. 31 I. Ju. Krackovskij, "Arabskaja geograficeskaja literatura," [Arabie geographical litera- ture] in Izbrannye socinenija, T. IV, Moskva-Leningrad 1957, 441^445. T. Lewicki, "Ibn 'Abd al-Mun'im al-Himyarï," in The Encyclopedia of Islam. Vol. 3, New Edition, 675-676. 32 The "Sky, Heaven" cult among the Hungarians is a typical religious phenomenon at- tested among the Turkic and Mongolian speaking peoples of the steppe known as Tengri cf. Ibn Fadlän in Z. V. Togan, Ibn Fadlän's Reisebericht. Leipzig 1939, 20; J. P. Roux, La religion des Turcs et des Mongols. Paris 1984,110-124; K. Lech, Das mongolische Weltreich. Al- 'Umarî's Darstellung der mongolischen Reiche in seinem Werk Masâlik al-Absâr 26  A NEW MUSLIM SOURCE ON THE HUNGARIANS ... pork33 and present offerings (to God). If one of them acquires some food, he lights a fire and takes the best part of his bread and food and throws it into the fire, calling on his most beloved patron, because they believe firmly [are convinced] that the smoke ascends to the sky and it is stored away for the dead in front of God, to Him belongs glory and power, in order to gain the grace of God.34 They are immigrants from Chorasan. 35 Islam is widespread there. 36 These Turks redeem the Muslims and Jews from captivity, if they were captured in one of the neighboring regions.37 fi Mamälik al-Amsar. Wiesbaden 1968,95,192, note 37; J. Gießauf, Die Mongolengeschichte des Johannes von Piano Carpine. Graz 1995,133, note 359.1. Fodor, "Über die vorchristli- che Religion der Altungarn," Acta Ethnographica Hungarica 48 (2003), 327-351. 33 The prohibition of eating pork concerned the Muslim and Jewish communities living among the Hungarians in the Carpathian basin. Cf. F Vire, "Khinzir," in The Encyclope- dia of Islam. Vol. 5, New Edition, 7-9. 34 The offering of food was an integral part of nomadic life cf. P. Ratchnevsky, "Über den mongolischen Kult am Hofe der Großkhane in China," in Mongolian Studies, ed. L. Li- geti, Budapest 1970, 429-430; Roux, La religion, 237-245; Gießauf, Die Mongolenge- schichte, 134, note 364; G. Mészöly, "Az ugor kori sámánosság magyar szókincsbeli emlékei." [Reminiscences of shamanism of the Ugor period in the Hungary vocabu- lary] Magyar Nyelv 48 (1952), 46-61. The cult of fire among the Hungarians was men- tioned in the al-Gayhänl-tradition, (Ibn Rusta and Gardizi, Göckenjan-Zimonyi, Orien- talische Berichte, 73,177. Gardizi recorded an interesting explanation of the fire, which purified the dead from their sins in the passage of the Qirgiz, Ibrahim ibn Wasif de- scribed a shamanistic rite of the Khazar ruler, where the colors of fire gave indications of the future. Göckenjan-Zimonyi, Orientalishe Berichte, 126, 239-240. Cf. N. N. Poppe, "Zum Feuerkultus bei den Mongolen." Asia Maior 2 (1925), 130-145; Roux, La religion, 103-105,222-226, 237-238; Gießauf, Die Mongolengeschichte, 137, note 374. 35 hum näqila min Huräsän. Al-Mas'üdi noted on the Muslim body guards in the court of the Khazar Khaqan: hum näqila min nahw biläd Hwärizm, Murüg II, 10 (Mas'üdi, Murüg = Murüj al-dhahab wa ma'ädin al-jawhar. ed. B. de Meynard-P. de Courteille, I-XI, Paris 1861-1877). "they are immigrants from the environs of Khwärazm," V. Minorsky, A History of Sharvän and Darband in the 10th-llth Centuries. Cambridge 1958, 147; "Originates des environs du Kharezm," Pellat, Les Prairies d'or, 162. The Balhi-tradi- tion used the same word in connection with the origin of the inhabitants of Buhara. Cf. Istahri BGA I, 315: "It is said that the original inhabitants of Buhara had wandered (näqila) from Istahr in old days," Cf. Ibn Hauqal BGA II2, 491, French translation: Kramers-Wiet, Ibn Hauqal, 471. According to the Muslim author the Hungarians originated from Khuräsän, the eastern province of Persia south from the River Amu-darya (C. E. Bosworth, "Khuräsän," in The Encyclopedia of Islam, Vol. 4, 55-59). Accordingly, the categorization of the Hungari- ans as Turkic peoples reflects the same idea; they came from Central Asia, east of the Volga River. It is well-known that the Hungarian tribal confederation included ele- ments from Khwärazm, who were called káliz/caliz in the medieval Hungarian king- dom (K. Czeglédy, "Az Árpád-kori mohammedánokról és neveikről." [On Muslims and their names in the Arpadian period] Nyelvtudományi Értekezések 70 (1970), 254-259). As Khwärazm belonged to the Samanides, whose rule included Transoxania and Khuräsän, significant groups among the Hungarians took part in their ethnogenesis 27  ISTVÁN ZIMONYI They treat their guests well [hospitably]. Their morals 38 are satisfying except that they leave their women with their slaves and guests and those who want them [the women] to be alone. In this respect they take the rank of dogs.39 If the account of al-Bakri is compared with that of al-Himyari it can be ascer- tained that al-Himyari omitted some parts of the geographical data, but he who wandered from the provinces of eastern Persia. As for the inhabitants of Khwä- razm, they played a predominant role in Eastern Europe in the ninth and tenth centu- ries. They were the bodyguards of the Khazar ruler, they were mentioned in Kiev and the name of the Caspian Sea was Hvaliskoje more in the Russian annals. They were re- corded as the agents of Islam among the Volga Bulghars and the Pechenegs (Kmoskó, Mohamedán írók, 1/2, 252-253; Togan, Ibn Fadlän, 217-220; Göckenjan-Zimonyi, Orien- talische Berichte, 223, note 15). 36 Muslim merchants among the Hungarians were recorded by al-Mas'üdi. Cf. Mas'üdí, Murüg II, 58-64; Pellat, Les Prairies d'or, 178 Marquart, Streifzüge, 61-63; Kmoskó, Mohamedán írók, 1/2, 183-184. Ibrahim ibn Ya'qüb also mentioned Muslim merchants from Hungary (Mishin, Ibrahim ibn-Ya'qub, 186). Until the thirteenth century Muslim communities had a dominant role in the finance of the Hungarian court and the popes wrote letters to the Hungarian kings to impose restrictions on them (Czeglédy, Az Ar- pád-kori, 254, T. Lewicki, "Madjaristan," in The Encyclopedia of Islam, Vol. 5,1010-1022). The traveller Abü Hamid al-Andalusi al-Garnatí visited Hungary in 1150, where he found a Muslim community which was rich and numerous (I. Hrbek, "Ein arabischer Bericht über Ungarn (Abü Hamid al-Andalusí al-Garnati, 1080-1170)." Acta Orientalia Academiae Scientiarum Hungaricae 5 (1955), 205-230; O. G. Bolsakov-A. L. Mongajt, Putesestvie Abu Hamida al-Garnati v vostocnuju i central'nuju Evropu (1131-1153 gg.). [The travel of Abü Hämid al-Garnati in Eastern and Central Europe (1131-1153)] Moskva 1971). Yäqüt met Muslims from Hungary in Aleppo in 1220 and mentions that they had thirty settlements in Hungary (Jacut's Geographisches Wörterbuch aus den Hand- schriften zu Berlin, St. Petersburg und Paris, ed. F. Wüstenfeld, 6 vols., Leipzig 1866-1873, Vol. 1, 468^470; Hungarian translation with commentary: A honfoglalás korának írott for- rásai. [Written sources of the age of the Conquest] ed. Gy. Kristó, T. Olajos, I. H. Tóth, and I. Zimonyi, Szeged 1995, 71-72. Cf. Czeglédy, Az Árpád-kori, 254-259; A. Márton, "Izmaeliták," [Ismaelites] in Korai magyar történeti lexikon (9-14. század), ed. Gy. Kristó, Budapest 1994,298. 37 Al-Mas'üdí reports on the Pecheneg-Hungarian campaign in 934 against Byzantines that the Muslim merchants of the nomads offered to help the former Muslims of the Byzantine troops convert to Islam again and in that case help them to return to the land of Islam (Murüg II, 61; Pellat, Les praires d'or, 178). Similar history on the redeeming of Muslim captives was noted by al-Bakri in connection with the Pechenegs (Göckenjan, Zimonyi, Orientalische Berichte, 223-224). 38 Ibn Fadlän described the protection of hosts in detail among the Oghuz. Visiting mer- chants made a special contract with one of them protecting and profiting both of them (Togan, Ibn Fadlän, 23-24, 133-134). The dog symbolized promiscuity among Muslim authors. The Muslims visiting the nomads of the steppe noticed that the women of the nomadic societies played a significant role in public life and that sexual regulations dif- fered in some respects. 39 Ibn 'Abd al-Mun 'im al-Himyari, Kitäb al-Rawd al-Mi 'tar ft Khobar al-Aqtär. ed. Ihsän 'Ab- bas, Beyrut 1975, 39. The text is published by the editors of al-Bakri in parentheses Leeuwen-Ferre, Kitäb al-Masälik, 490. 28  A NEW MUSLIM SOURCE ON THE HUNGARIANS ... quoted new material on the Hungarian religion, food-offerings, origin, and rights of hospitality, which deserve a separate study. Only two questions must be an- swered in this paper: when were these data gathered and how did they reach Andalusia? As the description mentions the Hungarians in the Carpathian Basin, it must be dated post 895, i.e. after its Conquest. Another decisive date can be the fact that they were Sky worshippers. This means the pagan period before conversion to Christianity, which took place in 1000. A narrower time span can be deter- mined, since Bohemia is called the country of Büyasláw, which is confirmed by Ibrahim ibn Ya'qüb. Boleslaw I ruled Bohemia between 935 and 967. But his son ruled under the same name, that is Boleslaw II from 967 to 999. It can be con- cluded that this Hungarian passage was recorded in the second half of the tenth century. The Andalusian Muslims could have had different motives in collecting in- formation on the Hungarians and Central and Eastern Europe. One of the main reasons was a military operation of the Hungarians in 942, when they laid siege to Lérida and some other towns.40 Recording the raid, Ibn Hayyán gathered some information about their homeland: Those who know their affairs mentioned that their country lies in the far East. The Pechenegs live east of them and they are their neighbors. The land of Rome lies south of them. The town of Constantinople lies deviating a little from the di- rection to the east. The town of Muráwa (Moravia) and the rest of the Slavic countries lie north of them. The Saxons and the Francs are west of them. They covered a long distance to the land of Andalusia. In the desert... [blank space] the kings from them. Their way during their march crossed Lombardia, which bor- ders them. There is a distance of eight days between them and it [Lombardia]. Their dwelling places are on the Danube River and they are nomads as the Arabs without towns and houses living in felt tents in scattered halting-places.41 According to al-Himyari the Hungarians redeemed not only the Muslims but also the Jews from captivity. This information was worth mentioning for some- one who was interested in the fate of the Jewish community or was himself a Jew. It is well attested that the Jews played a predominant role in the caliphate of Cor- dova, among others the diplomat Ibrahim ibn Ya'qüb, who met Emperor Otto I in 962. His report on the Slavs is an important source. He recorded only marginal notes on the Hungarians, whom he called Turks: 40 On the raid against Andalusia, see: Czeglédy, "Új arab forrás," 273-285; K. Czeglédy, "Még egyszer a magyarok 942. évi spanyolországi kalandozásáról." [Once again on the raid of the Magyars in Spain in 942] Magyar Nyelv 77 (1981), 419-423; I. Eltér, "Néhány megjegyzés Ibn Hayyánnak a magyarok 942. évi spanyolországi kalandozásáról." [Some remarks of Ibn Hayyán on the raid of the Magyars in Spain in 942] Magyar Nyelv 77 (1981), 413-419; Gy. Györffy, "Dual kingship and the seven chieftains of the Hun- garians in the era of the conquest and the raids." Acta Orientalia Hungarica 47 (1994), 96-100. 41 Ibn Hayyán, Al-Muqtabas, V, 482; Hungarian translation with commentary in Elter, A magyar kalandozáskor, 178-179. 29  ISTVÁN ZIMONYI "As for the country of Buyaslaw (Tawisla=Buyaslaw/Bohemia), its extension from F.raghah (Faraga, Prague) to Karakwa (Karakwa, Cracow) equals three weeks of travel. On its length, it is limited by the country of the Turks (At- rak=Hungarians). The city of F.raghah is built of stones and limestone. It is the richest place in goods. Russians and Slavs come from Karakwa with goods. Mos- lems, Jews, and Turks come there from the country of the Turks and bring goods and trade balances. Flour, [slaves], tin, and various kinds of furs are exported from there." 42 Another reason for the keen interest of Cordova on Eastern Europe was the slave trade. The slaves were transported from east of the Elbe River. During the rule of 'Abd al-Rahman III (912-961) three censuses were recorded: first, 3,750 Slavic slaves, then, 6,087 and finally 13,750. The number rose ten times higher within half a century.43 The slave trade was the business of Frankish and Jewish merchants. The Slavic slaves were transported first to Prague, which was a sig- nificant castrating-center. From here the way led to Regensburg and Venice. The slaves from the Elbe River were transported via Verdun, Lyon, Aries, and Nar- bonne to Andalusia. 44 Jewish communities, with strong political influence in Cordova, were inter- ested in the slave trade. They had two synagogues in Cordova and in Toledo and a Talmudic school in Cordova where they learned not only Hebrew but classical Arabic. Hasday ibn Saprut, a Jewish physician, was the wazir of the Caliph 'Abd al-Rahman III (912-961). He sent a letter to the Khazar ruler, Joseph, as the news of conversion of the Khazars to Judaism had reached the court of Cordova. The famous Khazar correspondence includes this letter and Joseph's reply. Hasday ibn Saprut had some difficulties getting in touch with the Khazar ruler. The Byzantine emperor hindered the Jewish envoys from visiting the Khazar country. Another possibility was the route through Syria, Iraq and Armenia. But before realizing the plan the embassy of the king of Giblim, that is, Otto I of the Ger- mans, arrived in Cordova. It was accompanied by two Jews, Mar Saul and Mar Joseph. Hearing about the difficulties in making contact with the Khazars, they offered their services. They suggested giving them the letter and they would hand over it to the king of Giblim, i.e. Otto I, who would send it to the Jewish community in the country of Hungrin, i.e. of Hungary. The letter might also have arrived via Rus' and Bulghar at the hands of the Khazar king.45 It is well known that the Jewish communities of Eastern Europe were in close contact with one another, so it is no wonder that the Jews in Hungary knew the way to the Khazar 42 Kowalski, Relacja Ibrähima, 49 Arabic text: 2-3; Mishin, Ibrahim ibn-Ya'qub, 185-186. 43 M. Lombard, Blütezeit des Islam. Eine Wirtschafts- und Kulturgeschichte 8.-11. Jahrhundert. Frankfurt 1991, 200. 44 Lombard, Blütezeit des Islam, 88-89,199-200. 45 S. J. Spitzer-G. Komoróczy, Héber kútforrások Magyarország és a magyarországi zsidóság történetéhez a kezdetektől 1686-ig. [Hebrew Sources Relating to the History of Hungary and Hungarian Jewry in the Middle Ages (from the Beginnings until 1686)] Budapest 2003,83-84; Dunlop, The History, 136-137. 30  A NEW MUSLIM SOURCE ON THE HUNGARIANS ... court. It is worth mentioning that the author of Hasday ibn Saprut's letter applied the Hebrew form of the same Slavic and European name Ungarus as was used in the source of al-Bakri. The recently published Hebrew Khazar letter from Kiev has information on the contact between the Jews of Kiev and Hungary. 46 In conclusion, the source on the Hungarians preserved by al-Bakri can be dated to the second half of the tenth century. The court of Cordova had a keen interest in obtaining valid information about Eastern Europe. It was motivated on the one hand by the Hungarian raid against northern Andalusia in 942 and on the other hand by political contacts with the Franks, Germans, Hungarians, and Slavs, and, of course, by international trade. The Jewish community in the caliph- ate of Cordova was interested in the flourishing slave trade in Eastern Europe and the news of the conversion of a great power to Judaism in Eastern Europe opened up new vistas. Thus they played a decisive role in getting intensive con- tacts and through them a better knowledge of Eastern Europe. Al-Bakri's dis- course on the people of Ungarus corroborates the existence of adequate knowl- edge of Eastern Europe and it encourages further study of the Muslim literature of Andalusia. 46 A. Róna-Tas, "A keleti magyarok egy kairói geniza-levélben." [The Eastern Hungarians in a Geniza letter of Cairo] in Évkönyv 1983-1984. ed. S. Scheiber, Magyar Izraeliták Or- szágos Képviselete, Budapest 1984,293-296. 31