Yazijioghlu 'Ali on the Christian Turks of the Dobruja
Paul Wittek (Bulletin of the School of Oriental and African Studies, University of London, Vol. 14, No. 3,
Studies Presented to Vladimir Minorsky by His Colleagues and Friends (1952), 639-668.)


The steppe which stretches between the Lower Danube and the Black Sea, from the Delta southward as far as the foothills of the Emine Dagh, [1] and which since the middle of the 14th century has been called, after the Bulgarian prince Dobrotitsa, the Dobruja, is the homeland of a small Turkish-speaking people, the Gagauz. It is because of their religion that they appear as a distinct group among the Turks : they are Christians belonging to the Greek Orthodox Church. In the past the Gagauz may have constituted, among the various ethnic elements of the region, a group of considerable importance, especially in the southern and middle Dobruja, from Varna and Kaliakra towards Silistria on the Danube. Besides, small isolated groups of them are to be found also in the Balkans (where they are more commonly known by the name of Sorguch): in Eastern Thrace, round Hafsa, to the south-east of Adrianople, and in Macedonia, to the east and west of Salonica, round Zikhna (near Serres) and round Karaferia (Verria). In modern times the Gagauz of the Dobruja have shrunk to a feeble minority chiefly as a result of a prolonged and massive emigration into Bessarabia. To-day even this remnant is rapidly dwindling.

The fact that they are Christians makes of the Gagauz an intriguing historical problem. Certainly they cannot be Anatolian Turks who had immigrated into the Dobruja under the Ottomans and been subsequently christianized there, say under the influence qf the surrounding population; even such an unostentatious, gradual apostasy from Islam is something inconceivable under the sultan's sway. Their conversion must therefore have been completed before the Dobruja became Ottoman, i.e. before the end of the 14th century. Could the Gagauz, then, be regarded as Greek, Bulgarian, or Wallachian Christians who under the Ottoman domination adopted the Turkish language? This, too, is a priori most unlikely since in the Balkans conditions favoured on the contrary acceptance of Islam combined with retention of the native language — witness the Muslim Bulgarians (the Pomak), the Bosnian Muslims speaking Serbo-Croat, the Muslim Albanians. A Turkish immigration from the north, from the South-Russian steppe across the Danube, is certainly the first thing which will come to the mind as historically probable. Indeed, the Gagauz have been identified with one or the other of the 'Northern' Turkish peoples — Petcheneg, Uz, Kuman — who are known to have passed through the Dobruja in the 10th and 11th centuries. However, very little and only doubtful evidence has been produced for these various identifications; on the other hand, the late T. Kowalski's careful analysis of the Gagauz Turkish has firmly established


that, in spite of some 'Northern' elements, it is essentially of 'Southern', i.e. 'Anatolian' character. [2]

An Anatolian origin is precisely what we have to accept if we are to believe the account which a very early Ottoman text gives of a Turkish immigration from Asia Minor into the Dobruja — not in Ottoman but in pre-Ottoman times. This account relates events which happened just after Michael VIII Palaeologos had recaptured Constantinople from the Franks, in 1261, and was being helped in his Balkan campaigns by the Seljuk troops who had joined their sultan 'Izzeddīn Kaikāūs II, then an exile at the Byzantine court. Our account tells us how the Turkish troops were allowed to bring their folk from Seljuk Anatolia and how these nomads, having entered Byzantine territory, crossed over to Europe and settled in the Dobruja which the grateful emperor had assigned to them and where they finally became Christians. [3]

A late and very incomplete redaction of our account, made in 1599 by the Ottoman court historiographer Seyyid Loqmān, [4] was first mentioned and used by J. von Hammer-Purgstall [5] and subsequently published by I. J. Lagus, together with a Latin translation. [6] Loqmān claims to have drawn his information from a book called Oghuz-name and, indeed, it comes from a work known by this name, but Loqmān has omitted the later and most revealing


part of the story. [7] As long as he was our sole authority and his source completely unknown, doubts were justified about the value of information which is obviously to a large extent legendary in character and makes its first appearance more than 300 years after the events concerned. Some scholars, indeed, thought it better to ignore the account. [8] Now, however, we not only know that it is 175 years older than Loqmān — in fact, it is only 30 years younger than the Ottoman occupation of the Dobruja, i.e. as old as an Ottoman reference to this region can be — but we also possess it in full and are thus enabled to determine its true character. [9] We may therefore dismiss Loqmān and turn to his source.


The Oghuz-nāme or, as it is more often called, Seljūq-nāme — both titles being justified by its contents — is a historical compilation, composed in Turkish by a certain Yazjoghlu 'Alī under the Ottoman Sultan Murād II (1421-51). [10] In the main it is a translation of Ibn Bībī's History of the Rūm Seljuks. The work of Ibn Bībī, finished in 1281, [11] was written for the famous historian Juwainī who, like Ibn Bībī himself, had by family tradition followed the dīwān career. Little wonder, therefore, that the book is written in the most involved and flowery Persian of which a high chancery official was capable. The only known manuscript, [12] a splendid volume written for one of the last Rūm Seljuks (Sultan Kaikhosrou III), is now preserved — as a waqf made by Maḥmūd I (1730-1754) [13] — in the library of the Aya Sofya mosque, having previously been, probably for a very long time, the property of the Ottoman Sultans. The work must always have been extremely rare; for soon after its appearance it was replaced, obviously on account of its bulk and the difficulty of its language, by an excellent and very readable abridgement. [14] Copies of this 'royal book' in its


full original form, as far as they existed at all, must have been highly treasured at the courts of the emirates which arose from the ruins of the Seljuk Sultanate. As long as the chanceries of those new states had not yet produced an insha literature of their own, the book was to them a priceless source of instruction and guidance. Ibn Bībī, after mentioning briefly the appearance of Sulaimān b. Qutlumush in Rūm, deliberately passes in silence over more than a century, 'for lack of information,' and begins his in general astonishingly well-founded account with the year 1192, carrying his narrative down to 1281, the very year in which he completed his work.

Yazjoghlu 'Alī translates from the full original version. Though he makes no attempt to fill the long gap at the beginning of Ibn Bībī’s history, he puts before it a short account of the Great Seljuks, taken from Rawandi, [15] down to 471/1079, the year which he accepts as the date when Sultan Malikshāh sent his 'nephew' Sulaimān b. Qutlumush [16] to Rūm. By means of a 'fore-runner', i.e. an interpolation in Rawandi’s text which foreshadows what will follow in Ibn Bībī, the welding is so cleverly done that the reader will not easily be aware of having passed into a quite different work. [17] Yazjoghlu has likewise


added at the end a continuation [18] which carries the narrative down to the first years of the 14th century, ending with a chapter on 'the situation in Rūm after the death of Ghāzān Khān' (1304). [19] Here, too, 'fore-runners' inserted at the chronologically appropriate places in the later parts of Ibn Bībī's text, ensure the continuity of the narrative. The main part of the account of our Dobruja Turks belongs to that continuation, but instalments of it appear already, long before, as 'fore-runners' carefully fitted into the body of Ibn Bībī's text, not always without causing some slight adaptations in the latter. Furthermore, right at the beginning of his work, Yazjoghlu 'Alī has put, in the guise of an introduction, the legendary history of Oghuz Khān and the Turkish tribes descended from him. This is entirely taken from Rashīdeddīn's Jāmi' et-tevārīkh, of 1310, which the translator must have used in a specially fine copy as we can see from the admirably executed tamghas of the 24 Oghuz tribes, missing, as it seems, in almost all the manuscripts of the Persian original. [20]


The 'Oghuzian theme', so forcefully put forward with this introduction, permeates the entire work right to the end in the form of prose and verse interpolations on 'oghuzian' matters scattered throughout the text: begs of the Oghuz are introduced wherever possible and their deeds, their feasts and their customs exalted in prose and in verse; the great 'Alāeddīn Kaiqobād is said (in addition to what Ibn Bībī ascribes to him) to have had full knowledge of the Oghuz-nāme and the Oghuz lore (türe). [21] Towards the close of the work the 'Oghuzian theme' becomes once more supreme, when the begs of the Oghuz according to their türe elect 'Osmān b. Ertoghrul, a descendant of Qayn. [22] All these interpolations are, as one can easily recognize, mere inventions [23] but the emphasis accorded to the 'Oghuzian theme' reflects, as we shall see, a very real purpose. Apart from a parallel 'Ghāzī theme' which can here be left aside, [24] there runs throughout the work yet another set of arbitrary interpolations in prose and in verse extolling the chancery official, the yazj, and his art and stressing the importance of his position. [25] Furthermore, here and there passages are inserted in praise of


the author's own sultan, Murād II (1421-1451), by which Yazjoghlu's work can be roughly dated. [26] Finally, there is an epilogue in verse: here the author gives his name and a rather enigmatic chronogram concerning the completion of the work. [27]

The chronogram has been interpreted as indicating the hijra year 827, i.e. 1424. [28] In fact, there is a passage which seems to support this date. At the end of the chapter on the conquest of Antalia by Kaikhosrou I in 1207, we find an especially fervent prayer that 'with the help of God Sultan Murād may be victorious over his enemies and crush the tyrants, infidels, and rebels of the time.' [29] In 1424 Antalia was a remote and desperately defended outpost towards which all the thoughts of a devoted Ottoman were naturally turned. We shall encounter still stronger evidence that the work was really written in the first years of Murād II's reign. [30]

As to the author, his work both in its character and in numerous points of detail gives the strong impression that he was a high official in the Ottoman chancery. His very name of Yazjoghlu, 'Clarkson', suggests that, like Ibn Bībī and Juwainī, he had entered the dīwān career by family tradition. Little wonder therefore that he endeavours in his work, as we have seen, to enhance the prestige of the yazj and that he knew of, and chose to translate, a work like that of Ibn Bībī. In his translation he shows special care and skill in the rendering of the administrative formulae and technical terms as if one of his aims were to provide the Ottoman chancery with a model of style. That he was an official of importance can be inferred from the fact that the 'Oghuzian theme' clearly serves aims of high policy. Interpolated in Rashīdeddīn's 'Testament of Oghuz Khān' [31] is to be found the audacious declaration that


'Sultan Murād is by origin and süngük [32] superior to all the khān families of the other Oghuz as well as to the various branches of the house of Jingizkhān; therefore shar' and 'urf demand that the Turkish and Tatar khāns come to his Porte for salutation and service'. [33] In justification of this Ottoman claim there follows at once the vaticination of the sage Qorqud Ata to the effect that 'the khānship shall in the end return to the Qay from whose hands no one shall take it away' [34] - which, of course, implies that 'Qay' stands for 'Ottoman', though this becomes clear only towards the end of the book, in the chapter on 'Osmān's election. [35] Here a political programme is laid down in which tendencies, timidly appearing as early as in the time of Murād I, [36] are developed into a precise and most ambitious ideology, destined at first for the ' innermost circle' only. To be the instigator of such a programme, or even only the formulator of its 'scientific' and literary expression, Yazjoghlu must surely have been a great official in close touch with the sultan.

Having formed an idea of the author [37] and his work, we can now turn to his account of the Dobruja Turks. I summarize it as briefly as possible and divide it into paragraphs in order to facilitate the commentary.


§ 1. 'Izzeddīn Kaikāūs II (who by decision of the Mongol overlord rules over the western half of the Sultanate whereas the eastern half obeys his brother Rukneddīn), feeling himself threatened by his brother and the latter's Mongol protectors, flees with his family and household to Antalia and from there by ship to Constantinople (Istanbūl). His army makes a fighting retreat to Sivrihisar and the border region, crosses into Byzantine territory, and finally joins the sultan in Constantinople. The sultan and his warriors find favour with the basileus (fāsilyevs) whom they valiantly help against his enemies.

§ 2. [38] One day the sultan and one of his generals complain to the basileus that being Turks they cannot endure town life for ever; if they were given a dwelling in the country-side, they could summon their nomad families from Anatolia. The basileus gives them the Dobruja (Dobruja-éli) as abode and they send word secretly to the nomad clans to which they belong. Whereupon their kinsfolk descend in large numbers from the mountains to Iznik and then cross over at Üsküdar. 'Ṣar Ṣaltq of blessed memory too crossed over with them.' Soon there exist in the Dobruja two or three Moslim towns and 30-40 oba (clans) of Turkish nomads. These Turks ward off the enemies of the basileus and destroy them.

§ 3. During a banquet (in Constantinople) the sultan is urged by his friends to profit by the strength of his followers and overthrow the basileus. This being reported to the latter, he orders one of the two Turkish army chiefs to be killed and the other blinded, and pardons only those of the Turkish soldiers who accept baptism. The sultan and his two 'older' sons, Mas'ūd and Kayūmerth, are imprisoned in a fortress.

§ 4. The sultan's mother, 'a sister of the basileus,' together with two younger sons of Kaikāūs, is kept in the palace of the basileus. Later she is sent to Karaferia (Qara-Vérya) where she is granted the tolls which are levied at the Anaqaps — i.e. the 'Mother Gate' as it is therefore called still nowadays. The two young princes receive the governorship (subashlq) of the town.

§ 5. The sultan is liberated from his fortress prison by the Tatars of the khān of the Golden Horde, Berke khān, who gives him hospitality in the Crimea.

§ 6. The sultan's mother, on hearing the false Rūmour that her son has perished on his flight, throws herself from the tower which flanks the 'Mother Gate'.

§ 7. After this the basileus gives Karaferia to the elder of the two princes and takes the younger one into his palace.

§ 8. Kaikāūs feels deeply grieved on the death of his mother and the captivity of his sons in the hands of the basileus.

§ 9. [39] Berke khān transfers the Turks of the Dobruja, 'and with them Ṣar Ṣaltq,' into the steppe (desht).
At this point the story seems to finish and, indeed, Loqmān's version stops here. Yazjoghlu, however, resumes the story, much later, with a short interpolation in Ibn Bībī's last chapter, where the latter tells how after 'Izzeddīn Kaikāūs' death in the Crimea, in 679/1280, Mas'ūd is acclaimed as his successor (for he, like his brother Kayūmerth, is present at his father's deathbed) and prepares to return to Anatolia by ship.
§ 10. When Mas'ūd asks for and obtains permission to cross over to Rūm, Ṣar Ṣaltq, on the order of Berke Khān, leads the nomad folk (göcher él) with all their cattle overland back to ' their abode', to Dobruja éli. [40] 'Their story shall be told in detail at its proper place.'
The next instalment follows immediately the last chapter of Ibn Bībī whose history ends with Mas'ūd's arrival in Rūm, his journey to the Mongol Court where he is recognized as ruler of the eastern half of the Sultanate (the western half being left to his cousin Kaikhosrou III b. Rukneddīn). [41] In this chapter Yazjoghlu has made a number of interpolations; inter alia he adds to the territories now subject to Mas'ūd 'all the lands until the frontier region of Izniq', since in view of what follows he feels obliged to represent him as an immediate neighbour of the Byzantines.
§ 11. In order to learn about his brothers and the Turks in Rūmeli, Mas'ūd sends ambassadors to the basileus Palseologos (fāsilyevs Balālōoghōs [42]) who replies as follows: one of your brothers stays with me, the other at Karaferia where he is invested with the government (beylik) of that country. As to the Rumeli Turks, some of them have joined him, the others have remained in the Dobruja. [43] This news reassures the sultan. Of the tribute which the basileus pays from of old, he (now) sends one-third to the Mongol khān, one-third to Mas'ūd, and one-third to Ghiyatheddīn (Kaikhosrou III).

§ 12. Matters continue like this for a fairly long time. The Turks in the Dobruja remain there with Ṣar Ṣaltq. Then Mas'ūd's brother who was with the emperor (tekvur) tries, together with some Turks, to escape. However, he is arrested and imprisoned. The Patriarch, 'that is to say the caliph of

the Infidels', [44] asks the basileus (fāsilyevs) to grant him the prince. He obtains him. baptizes him, and makes him a monk. The prince is for some time at the Hagia Sophia (Āyā Sōfya) in the service of the Patriarch. Then Ṣar Ṣaltq asks the Patriarch for him, and as the Patriarch knows Ṣar Ṣaltq to be a holy man, he sends the prince to him. There, after a while, the prince returns to Islam and becomes a dervish in the service of Ṣar Ṣaltq. [45] One day the supernatural power, which Ṣar Ṣaltq as a shepherd had received from the holy Maḥmūd Ḥayrān of Aqshehir, is transmitted to him and the name of Baraq ('dog') bestowed on him. He is sent to Sultanīye where 'still nowadays' his sanctuary exists. The Baraqī are his disciples.

§ 13. As to the Muslims at Karaferia, being tired of life among the Infidels, they migrate to Anatolia across the sea. The prince and his son live and die at Karaferia as Muslims, but the children of the latter are baptized on the order of the basileus in the year of his coming to Salonica. It is from one of their descendants, a certain Līzaqōs, that the town is taken (by the Ottomans) in the time of the grandfather of our Sultan. This Līzaqōs and his brothers, all valiant infidel warriors, are transferred to Zikhna and Līzaqōs, the eldest of them, is made governor (subash) of that place. In Sultan Bāyezīd's campaign against Malatia and Erzinjan, Līzaqōs and his brothers are with the army. Līzaqōs, having suffered many hardships and difficulties in these two campaigns, on his return renounces his office and asks for a diploma of exemption (müsellemlik ḥükmü) for himself and for his brothers. When Sultan Bāyezīd learns that they are of Seljuk origin, he grants them the privilege. Līzaqōs dies at Zikhna as a monk. 'His brothers and their sons are still nowadays at Zikhna and pay neither kharāj (here: poll-tax) nor onda (tithe). Recently they have secured the renewal of their diploma. One of them is called Dīmitrī Sulān, the other Mīkhō Sulān. That's all (wa's-salam).' [46]


Again, some pages later, in the chapter headed 'Rest of the story of 'Alāeddīn Kaiqobād b. Ferāmurz, nephew of Sultan Mas'ūd, and conquest of Bilejik by 'Osmān b. Ertoghrul, King of the Ghazis', the Dobruja Turks reappear. Following the story of 'Osmān's election as khān and the conquest of Bilejik in 699/1300, Yazjoghlu speaks of the increasing immigration of Turks from the Anatolian interior into the lands of 'Osmān, into Aydn and Qaras (i.e. the Troas), adding with regard to the last-named:

§ 14. Those in Rūmeli, the Muslims dwelling in the Dobruja, tired of the Infidels, emigrate to Qaras.
This is repeated and amplified in Yazjoghlu's last chapter, 'The events after the death of Ghāzān Khān', following a short account of the situation in Anatolia after the appearance of Timurtash b. Emīr Chōban in 1318.
§ 15. At this time, the Muslims dwelling in Rumeli, in the Dobruja, join Khalīl Eje and go over by ship to Qaras, for in the troubled state of Anatolia no more news comes through from there. And in Rumeli the Bulgarian princes rise up and get the better of the basileus. They occupy the major part of Rumeli. That is why those (Muslims), tired of them, emigrate and go over to Anatolia. 'As to those who remained in Rumeli, after the death of Ṣar Ṣaltq their race renounced the faith and forgot it.' [47]
A number of general observations must be made before entering a detailed analysis of the account. It covers a period of much more than a century (from about 1261 to about 1395), which made it necessary to give it in instalments at the chronologically appropriate places. As a whole, the account fills in the MS. Topkapi Sarayi, Revan K., No. 1391, eleven pages (out of more than 900), of which seven pages, containing the §§ 1-9, occupy as a solid block ff. 373b-376b, and three pages, §§ 11-13, as another block, ff. 415a-416a, whereas the rest, the §§ 10, 14, and 15, appear as small passages on ff. 41 Ib, (about) 430, and 444a respectively.


The account falls visibly into four different stories: that of (1) 'Izzeddīn Kaikāūs, (2) the Dobruja Turks, (3) the Karaferia family, (4) Ṣar Ṣaltq and Baraq. Of these four stories that of 'Izzeddīn Kaikāūs derives from Ibn Bībī: in §§ 1 and 5 the original is rendered exactly and in §§ 3, 4, 6, and 8 with some adaptations which can easily be established by comparison with the Persian text and will later be noted and discussed. For the other three stories we have no source at hand, nevertheless, we can say that these, too, must have undergone similar adjustments, since Yazjoghlu had to knit all these stories together into a tolerably consistent narrative. The story of Ṣar Ṣaltq and Baraq (§ 12) stands out as a unity, though it is announced by small 'fore-runners' in the preceding §§ 2, 9, and 10 ('and Ṣar Ṣaltq with them'), preparing the reader for the holy man's presence in the Dobruja, and other 'fore-runners' in §§ 4, 7, 8, and 11, introducing the future Baraq, and still finds an echo in § 15 (Ṣar Ṣaltq's death). This use of 'fore-runners' and the isolated occurrence of tekvur instead of fāsilyevs, let alone its strictly hagiographic character, indicate for this story a separate origin. Similarly distinct is the Karaferia story, represented by § 13, with 'fore-runners' in §§ 4,6, 7, 8, and 11.

The Karaferia story is of pre-eminent importance because it provides, itself, the means to understand how Yazjoghlu got hold of it and to assess its true character, and also because it links the whole account with the author's own time. The story was obviously told to him by the two brothers from Zikhna when, after the accession of Murād II in 1421, they came to the capital, i.e. Adrianople, in order to have their privilege renewed at the chancery. Such a renewal was due at the beginning of each new reign. Yazjoghlu says that it was made 'recently', which gives very strong support to the acceptance of 1424 as the date of his work [48]; it also conveys the impression that he had personally dealt with the matter. One thing is fairly certain: the Seljuk origin of the family was mentioned in their diploma since, as we are expressly told, it was for this reason that Bāyezīd I had granted the privilege. Without claiming literal accuracy, we may imagine how Yazjoghlu met Dīmitrī Sulān and Mīkhō Sulān: when the request for the renewal of this certainly exceptional privilege was laid before the chief of the chancery, i.e. as we assume, Yazjoghlu himself, the translator of Ibn Bībī would hardly have missed such an occasion to converse with descendants of the Seljuks; surely he summoned these interesting visitors to his presence in order to hear from them as much as possible of their story. After this we can have no doubt as to the character of the Karaferia story; it is oral tradition, handed down from generation to generation in a family claiming royal origin — indeed, the two brothers put 'sultan' after their names — a story certainly not to be entirely believed but neither to be entirely rejected. There is one point in it which strikes us at once: the relation in which it sets the Turks of Karaferia and Zikhna with those


of the Dobruja (§ 11) — let us recall that Karaferia and Zikhna are two of the three places where Gagauz are found outside the Dobruja.

Lastly, the Dobruja story (§§ 2, 9, 10, 14, and 15) provides us likewise with a clue concerning its provenance and value: it is obviously a tradition kept alive in those families of Qaras which descended from the Dobruja Turks. As we shall see, such families can be assumed to have existed and even to have played no small role in early Ottoman history. Some connexion must have subsisted with their former abode, for they call it 'Dobruja-éli', i.e. by the name it received only half a century after they had left it. [49] Again, one point strikes us at once: the mention of Turks who stayed on in the Dobruja and there lost their Muslim faith, i.e. became Christians. This refers beyond any doubt to our Gagauz. The last paragraph (§ 15) would by itself already make it sufficiently clear that the Dobruja Turks had come from Anatolia — for, when they arrived, they were Muslims, and so were also the Dobruja Turks who joined the Seljuk prince at Karaferia (§ 11). From both places 'Muslims' are said to have returned to Anatolia (§ 13 and §§ 14-15), evidently in both cases to Qaras across the Dardanelles; this gives to understand that some of their kinsfolk had renounced Islam and stayed behind, as indeed the §§ 13 and 15 clearly show. There can be no doubt that the essential contents of § 2 come from the Qaras tradition, which must have recorded explicitly the Anatolian origin of the Turks of the Dobruja and the main facts concerning their immigration there, above all the name of the Sultan whose exile in Constantinople had given rise to the whole adventure.

In fact, 'Izzeddīn Kaikāūs must somehow have appeared also in the two other stories, as Baraq's father in the one and as ancestor of the Karaferia family in the other — without mention of his stay in the Byzantine empire and his subsequent escape both stories would remain incomprehensible. Thus, Yazjoghlu had a good reason to work the three stories into Ibn Bībī's account of 'Izzeddīn and to knit them together as closely as possible. How he proceeded will be seen by the separate analysis of each paragraph.

§ 1. As already stated, this is a faithful rendering of Ibn Bībī. [50] There is no need here for a discussion of the facts since they are well known from the Byzantine historians, [51] who on the whole are in agreement with Ibn Bībī. However, the latter shows 'Izzeddīn as sailing directly to Istanbul, whereas


from Pachymeres it becomes quite clear that the sultan reached the imperial court (probably at Nymphaeum) before Constantinople was retaken from the Franks (25th July, 1261) and only entered the capital afterwards (probably in the retinue of the emperor who always kept him at his side). This mistake of the Seljuk author, who in general is well informed, must not astonish us: once the fugitive Sultan was beyond the frontiers genuine information about him was no longer available at Konia. Of course, also in exile he remained an object of interest — but news was rare, vague, belated — sometimes mere rumour. Ibn Bībī's entire account of the 'Izzeddīn episode is mixed with romance [52] and may well be due to the tales of the late sultan's servants who had returned with Mas'ūd. Therefore the somewhat later Byzantines are here not only more explicit but also more trustworthy than the strictly contemporary Seljuk author.

§ 2. No trace of this is found in Ibn Bībī, a fact which was used to discredit the entire information given in this paragraph [53] — as if the chancery in Konia either would or could care about nomad movements on a distant frontier. As we have said, Yazjoghlu relies here on the genuine, oral Qaras tradition, but in order to link it closely with the preceding passage he introduces a meeting of the basileus with the sultan and one of his generals. It is in the form of their complaint to Michael VIII that we learn about the general situation. They describe it admirably: how understandable, indeed, that the Turkish soldiers who now fought for the emperor did not like the idea of spending their life, far from their families, in barracks. Recruited among nomads, they naturally desired that their clans should be near them. As Michael VIII's campaigns were fought at that time exclusively in the Balkans, the kinsfolk of these Turkish troops had to be brought to Europe though, if possible, not into a Byzantine province. To assign to them the Dobruja was the ideal solution: this 'corridor' through which the Tatars of the Golden Horde again and again swept down for deep incursions into the Balkans, was nominally part of Bulgaria but in reality more or less a no-man's-land. Besides, Michael VIII was on bad terms with the Bulgarian's and as 'restorer of the empire' would not have hesitated to dispose of a territory once Byzantine, though actually not in his possession. Immediately after the reconquest of Constantinople he had re-established Byzantine control of the Danube delta, where Vicina was an outlying possession, communicating with the empire only by sea. [54] To back this outpost by filling its hinterland, the Dobruja, with warlike allies and to erect there an obstacle against the Tatar incursions was excellent policy. The nomads, on their part, would slip into the Bulgarian Dobruja as inconsiderately as they had crossed the Byzantine-Seljuk frontier — and again, of course, no


(this time Byzantine) source would take notice of their movement. [55] Our text says expressly that the immigrants, in their new abode, had to fight with the emperor's enemies and overwhelmed them. The figures it ascribes to them are no more than a way of saying: numerous. As already stated, the mention of Ṣar Ṣaltq is here but a 'fore-runner' of the Ṣar Ṣaltq-Baraq story. On the other hand, the latter as well as the Karaferia story must also have contained something about the Dobruja Turks whose existence they presuppose.

§ 3. Here the account returns to Ibn Bībī but with the addition: '(two) older (sons)' and the omission of the 'mother'. [56] The addition is made in view of the Karaferia as well as the Baraq story, each of which presupposes a prince left behind for good. Since Ibn Bībī later shows Mas'ūd and Kayūmerth as present in the Crimea at their father's death (though he does not say how they came there), Yazjoghlu had to present them as the 'older' princes (leaving the reader to infer their escape with 'Izzeddīn) and to invent the two younger sons of § 4. The mother, on the other hand, had to be omitted here since she was needed for the Karaferia story (§ 4).

§ 4. First instalment of the Karaferia story: here we find the 'mother' left out in the preceding paragraph; she is described as 'a sister of the basileus' which does not come from Ibn Bībī and is certainly not true; we know, however, that she was a Christian. [57] One of the two princes is evidently only a 'fore-runner' of the Baraq story.

The linking of Ibn Bībī's account with the Karaferia story was obviously facilitated, and even perhaps suggested, by the occurrence in the latter of the Anaqaps, the 'Mother-Gate', thus named because the tolls collected there had been assigned to 'Izzeddīn's mother. But how, in a Byzantine town, should a gate bear a Turkish name? Clearly, anaqaps must represent something Greek. In a text of 1219 we see tenants (in this case: of vineyards) liable to make to


their landlord an annual payment called anacapsi. [58] No doubt, the gate in Karaferia, where 'tolls' were levied, was the place where the peasants had to deliver to the landlord (residing in the town) their annual due, the anacapsi, and was therefore called 'Anacapsi-Gate', a name which inevitably had to become in Turkish Anaqaps. In this Turkish form the gate assumed an important role in the Karaferia story — it is even possible that the figure of the 'mother', who is the 'sister of the basileus', is entirely derived from Anakaps.

§ 5. Ibn Bībī unchanged. [59] As to the facts told by Ibn Bībī in § 3 and in this passage, they are again in general confirmed by the Byzantine and some other sources though concerning the details these sources are at variance with Ibn Bībī as well as with each other. [60] 'Izzeddīn's liberation by Berke Khān's Tatars appears to have taken place towards 1265.

§ 6. Ibn Bībī, [61] but adapted to the Karaferia story: the tower from which the mother throws herself is in the original the tower of the fortress where she shares the captivity of her grandsons Mas'ūd and Kayūmerth. To locate the suicide at the Anakaps is little short of explaining the name of the gate by this dramatic event. This, however, Yazjoghlu does not do — obviously out of respect for the Karaferia tradition, where the name Anakaps already had an explanation (§ 4), though a much less romantic one.

§ 7. The Karaferia and Baraq stories combined.

§ 8. Ibn Bībī. [62] Only whereas he speaks of the captivity of two sons (i.e. Mas'ūd and Kayūmerth), Yazjoghlu tacitly understands by the 'two sons' the two younger princes of his invention.

§§ 9 and 10. This emigration from and re-emigration into the Dobruja may seem at first to be nothing but an invention of Yazjoghlu intended to keep the Dobruja story in step with that of 'Izzeddīn. If it be an invention, how admirably does it fit into the context! For 'Izzeddīn, as Ibn Bībī shows him, resided in the Crimea with many followers, all endowed with fiefs by


Berke Khān, which presupposes that many of the Turkish soldiers had been able to make their way to him. Since their escape route led through the Dobruja their kinsfolk there were informed of what was happening. Soon there was again a Turkish army — this time in the Crimea — wanting to have their clans near them — a desire which, of course, needed Berke Khān's agreement to be fulfilled. However, certainly not all the nomads will have left the Dobruja since, according to Gregoras, [63] many of their young men were still fighting as Tourkopouloi under the emperor's banners (having accepted baptism). That after 'Izzeddīn's death in 1280 it should have been Berke Khān who ordered the nomads back to the Dobruja, is of course an anachronism — he had died in 1266. The mention of Ṣar Ṣaltq in both paragraphs is nothing but a 'forerunner' of the Baraq story.

That such a migration to and from the Crimea did take place finds strong support in Pachymeres [64] who, dealing with the Tourkopouloi in the years soon after 1300, describes them as Christians of only recent date and only recently arrived in the empire 'from the Northern regions'. Furthermore, when mentioning their defection to the Catalans in 1307 and the possible reasons thereof, he speaks of their fear that the emperor might yield to the demand of the khān of the Golden Horde who wanted them back as his subjects. [65] The khān must therefore have had a real claim upon them, and one which was of fairly recent date. On the other hand, Gregoras says that the Tourkopouloi were those Turkish soldiers who after the sultan's flight stayed on in the empire; they were baptized and enrolled in the army, [66] their numbers being maintained by their own offspring. [67] To a certain degree he may be right: undoubtedly there were isolated groups of Turkish soldiers who being posted to various duties were unable to leave the empire with the sultan — but the bulk of the 'Seljuk army in exile' must, at the critical moment, have been in winter quarters (the Tatars had come over the frozen Danube!), i.e. with or near their families — as we assume in the Dobruja — and could therefore join their sultan. They may have remained with the Tatars for some time after 'Izzeddīn's death; when at length they returned, for whatever reason, the emperor accepted them no longer as a 'Seljuk army in exile' but as a regular corps of his army, i.e. as Byzantine, Christian soldiers. Significantly enough this corps is not mentioned before the events of 1307 in which, as we shall see, it played a considerable role.


§ 11. Karaferia and Baraq stories combined. To link this paragraph to the account of Mas'ūd's accession in Rūm, Yazjoghlu makes the latter send ambassadors to the basileus. Thus the narrative is given in the form of the emperor's reply to Mas'ūd's inquiry. Though intended as a mere bridge leading to the Baraq and Karaferia stories, the passage contains none the less the very important information that some of the Dobruja Turks had joined the prince in Karaferia.

§ 12. The Baraq story. It would need, and deserve, a separate study. My commentary (like my summary) is confined to what is strictly relevant to our purpose. Chronologically the story is remarkably sound: Sheikh Maḥmūd Ḥayrān died at Aqshehir in 1268-9 [68], so that Ṣar Ṣaltq can easily have been his disciple before going to the Dobruja shortly after 1261; after something like fifteen years spent in the desht (-i Qpchaq), Ṣar Ṣaltq returns to the Dobruja in about 1280 (§ 10) and stays there until his death, soon after 1300 (§ 15). As to Baraq, he is supposed to be at the time of 'Izzeddīn's escape, in 1265, old enough to become, conjointly with his brother, ‘subash’ of Karaferia; on the other hand, he is then still under his grandmother's tutelage (§ 4) — all this, however, has to be cut out as a product of Yazjoghlu's desire to co-ordinate the separate stories; once he had introduced the ‘two younger sons’ he had to keep them together as long as possible. Surely, the prince of the original Baraq story had nothing to do with Karaferia but was in the emperor's palace from the beginning until his attempt to escape. According to Yazjoghlu this attempt was made some time after Mas'ūd's inquiry. This leads to the of course completely fictitious date: some time after 1280 — which, however, fits well into the frame of the story; after 1280 Ṣar Ṣaltq is indeed again in the Dobruja, so that the prince can join him there to become the ecstatic dervish Baraq. As such, he is known to have played an important role at the Mongol court in Sultānīye under Öljaitu and to have perished in Gīlān in 1307-8,


leaving behind him disciples [69]. The name of Baraq is just about 1300, and still later, not uncommon both for men and women [70]. The account of how our Baraq received supernatural powers [71] may be suspected to be nothing more than an explanation of the name: one day Ṣar Ṣaltq vomits a lump which once Sheikh Ḥayrān had spat into his mouth, and the prince, in a fit of ecstasy, swallows it. Ṣar Ṣaltq caresses him and calls him baraghm ('my dog').

Most interesting for our present study is the fact that Baraq presents a Christian as well as a Muslim aspect: born a Muslim prince, he is baptized and becomes a monk in the patriarch's retinue, only to end as the founder of a mystic dervish order. The same is true also for Ṣar Ṣaltq: he appears on the one hand as the spiritual leader of the Muslim nomad Turks and on the other hand is regarded by the patriarch as a saintly man to whom unhesitatingly he entrusts the newly converted prince. This Christian aspect of Ṣar Ṣaltq is clearly recognized in a fetwa of Abu 's-Su'ud, which has just come to light [72]. This outstanding scholar and sheikhulislam of the 16th century describes Ṣar Ṣaltq as 'a Christian monk (keshīsh) who by ascetism has become a skeleton'.

This wavering between the two religions is characteristic of the entire account which time and again, though with much reticence, records cases of conversions to Christianity, or rather apostasy from Islam, some of them only temporary. Historically, the principal figure of the account, Sultan 'Izzeddīn himself, appears in the same ambiguous light. His mother is described by Pachymeres, as we have seen, as a pious Christian woman [73]. To the same author we owe a detailed account of the trial instigated in 1266 (shortly after 'Izzeddīn's flight) by Michael VIII against the inflexible Patriarch Arsenios, a trial in which the Patriarch's indulgence towards 'Izzeddīn, his sons and followers, played a great role. Pachymeres records, among other things, as one point of


the accusation, that Arsenios had ordered ‘his own monk’ to admit the Seljuk princes to the holy communion [74], and as the main argument of the Patriarch's defence, that in regarding the sultan and his sons as Christians he had acted upon the testimony of the bishop of Pisidia [75]. While Pachymeres reproduces the procès-verbal of the trial, Gregoras reflects a shorter popular version, which as such is for our purpose perhaps even more valuable. According to Gregoras; the Patriarch was accused of having admitted the sultan to the holy ceremonies and of having conversed with him inside the house of God — although, says the author, the emperor and the clergy knew very well what 'Izzeddīn had declared: that he was the son of Christian parents and had himself received the holy baptism, that he had become sultan of the Turks by the whim of fortune only, but even then had always cherished in secret the essentials of the faith, and that now in Constantinople he was openly adoring the sacred icons and celebrating all the rites of the Christians [76]. As a matter of fact, we shall soon meet a son of 'Izzeddīn, Melik Konstantinos, who is described as a perfect Christian, Byzantine gentleman.

§ 13. Main part of the Karaferia story, introduced already by §§ 4, 6, 7, and 11. In addition to what Yazjoghlu reproduces it must have contained its own version of 'Izzeddīn's stay with and escape from the Byzantines. The ‘mother’ who is the sister of the basileus and is given the tolls levied at the ‘Mother-Gate’ belongs to the Karaferia story — perhaps, as we have seen (§ 4), nothing but an invention derived from the ‘Anacapsi-Gate’ changed into Ana-qaps. Her suicide clearly comes from Ibn Bībī though it has been transferred to Karaferia (§ 6). That she is the mother of 'Izzeddīn, and thus the grandmother of the prince, is probably a further concession to Ibn Bībī: in a story of this kind and in view of the ‘Mother-Gate’ one would rather expect her to be the prince's mother — a Byzantine princess married to the sultan and left behind with her son (one son, since the other, Baraq, has to be dismissed). However this may be, the Karaferia family appears as claiming descent not only from the Seljuk sultans but also from the Palæologi.

The family has its own following: Turks from the Dobruja had joined them at Karaferia but the ‘Muslims’ are said to have returned at a certain moment to Anatolia, the ‘non-Muslims’, as it is implied, remaining at Karaferia — obviously they had become Christians. Astonishingly, the conversion of their leaders, the prince's family, is said to have taken place only later, in the generation of the prince's grandsons and ‘in the year when the basileus came to Salonica’.

This last indication must mean Andronikos III's entry into Thessalonica in January 1328, when the citadel, loyal to the old emperor Andronikos II, held out until the defenders themselves, seeing that all hope had gone, forced their


commandant to accept the amnesty offered by the victor and to yield. That stubborn commandant was a certain Georgios Lyzikos of Berrhoia (Verria), i.e. of our (Kara-)Feria [77]. As to his family name, it is undoubtedly identical with the name of Līzaqōs borne by the chief of the Karaferia family who surrenders his town, some sixty years later, to Bāyezīd's Ottomans. In the winter 1350-51 we meet Georgios Lyzikos again, this time as the commandant of Edessa, a strong place near Berrhoia, which he defends against the Serbians, the enemies of his emperor Kantakuzenos. The town is taken and destroyed, the valiant Lyzikos is sent in chains to Scopia (Üsküb) for punishment but dies on the way [78].

The Karaferia story has certainly mentioned some of these heroic events, in any case, as our text shows, that of 1328 when a Lyzikos was forced to submit to the emperor. For Yazjoghlu, however, this ‘Infidel matter’ was of no interest save in so far as he could use it to explain, in a manner calculated to win the reader's forbearance, such a lamentable fact as the apostasy of these valiant descendants of the Seljuks. We have seen how Yazjoghlu shrinks from saying outright that some of the Turks in Karaferia had become Christians. In the case of the Turks in the Dobruja, where he cannot help mentioning that those who remained there abandoned Islam (§ 15), he seeks at least to give the impression that their numbers were insignificant. The future Baraq has to suffer imprisonment before he accepts baptism. The Lyzikos of 1328 bears clearly a Christian name: Georgios. The family must have been Christians already when they came to Karaferia and so also the Turks who were with them — as we shall see, also those of them who returned to Anatolia were in reality not Muslims but Christians.

As to the Ottoman part of the story: the occupation of Karaferia is known [79] to have occurred already under Murād I, in 1387; its attribution to Bāyezīd I suggests that the town was again abandoned and then re-occupied. In all these events what counted most for the Lyzikos family was, of course, their transfer to Zikhna, and this probably happened under Bāyezīd I. It goes without saying that part of the Karaferia Turks, too, were directed to Zikhna, a most illuminating fact for our study. ‘Līzaqōs’ became subash of Zikhna because he was the chief of these Turks. It may appear almost incredible that a Christian should have been appointed subash of an Ottoman district and in this capacity, as an officer in the feudal cavalry, should have taken part in the Sultan's


campaigns; that in the early Ottoman state such a thing was in fact possible will soon be clear beyond doubt after the imminent publication of the defter of Albania, dating from the reign of Murād II [80].

A final question: is it possible that the Lyzikos family was descended from 'Izzeddīn Kaikāūs? Certainly, since 'Izzeddīn had left with the Byzantines a son Melik Konstantinos, whom we have already met and shall meet again. When this prince in 1307 disappears from the Byzantine scene he must have been about 50 years old and very probably the father of children. The Lyzakos of 1328 may well be a son of his. It is not excluded that Melik Konstantinos himself had for a certain time been connected with Karaferia. There is, however, no need to take the claim of the Lyzikos family so seriously. We shall see that all the Turks connected with 'Izzeddīn continued to call themselves ‘the people of Kaikāūs’. For the noble families among them it was but a small and tempting step to change this into ‘the lineage of Kaikāūs’. There was yet another family in Rumeli which claimed descent from Kaikāūs: that of the famous Sheikh Bedreddln [81].

§§ 14 and 15. End of the story of the Dobruja Turks: except for the few who stay behind and are doomed to become Christians, especially since their spiritual leader Ṣar Ṣaltq is dead or nearing death, they return to Anatolia (just as those of them who had joined the prince in Karaferia are said in § 13 to have done). Their return is connected with the name of Khalīl Eje and with the ascendancy the Bulgarian princes had won over the emperor. This enables us to perceive which events are meant.

Khalīl Eje is doubtless identical with the Khalīl (Χαλήλ) in Gregoras' account [82] of the happenings after the chief of the overbearing Catalan mercenaries, Roger de Flor, had been assassinated in the palace of the junior emperor Michael IX at Adrianople, in April 1305. The Catalans transform Gallipoli where they are stationed into a stronghold and wage open war on the emperor. To strengthen their ranks ‘they send envoys to the Turks of the opposite littoral (i.e. the Troas, Qaras) inviting them to fight on their side and take into their service 500 of them being themselves 3,000 strong’ [83]. Together they devastate the neighbouring country so that Michael IX has to march against them with an army which included the corps of the Tourkopouloi.


‘These Tourkopouloi, numbering a 1,000 men, were Turks who had followed Sultan 'Izzeddīn when he came over into the empire and had remained there when the sultan was carried off by the Tatars of the Golden Horde — they had adopted the Byzantine way of life and accepted the Christian faith and baptism; from then onward they were enrolled in the Byzantine army.’ [84] In the battle of Aproi (1307), in which the Catalans and their Turkish allies defeat the army of Michael IX, the Tourkopouloi behave ambiguously and ‘a few days later they go over to the Catalans and are gladly accepted; being of the same race they are joined to Khalīl’s Turks — for the leader of the Turks was called Khalīl’ [85]. The Catalans, emboldened by their victory and the adherence of the Tourkopouloi, ravage Southern Thrace for two whole years, after which, crossing the Rhodope mountains, they establish themselves in Kassandreia (the westernmost of the three ‘fingers’ of the Chalkidike) where they create a stronghold from which to plunder Macedonia. Foiled in their hopes by the emperor's effective defence measures they leave for Thessaly (1309), still accompanied by their Turks, then numbering 3,000 men — 1,100 of them those who on 'Izzeddīn's flight had been left behind together with Melik (Μελήκ), been baptized and enrolled in the Byzantine army, and who had increased in numbers by their offspring, whereas the majority were the Turks who had crossed over from Asia with Khalīl as hired auxiliaries of the Catalans [86]. On the march to Thessaly the Turks want to separate from the Catalans: their leaders Melik and Khalīl arrange with the Catalans for an amicable separation [87] and the division of the prisoners and the booty. ‘After their separation from the Catalans the Turks themselves divide into two groups, the one following Khalīl, the other Melik.’ Melik, having forfeited for good the friendship of the Byzantines — indeed, in spite of his baptism and the emperor's largesse he had gone over to the enemy — leads his 1,000 riders and 500 footmen into Serbia and submits to the King (Urosh II Milutin). Khalīl, with his 1,300 horsemen and 800 soldiers, returns to Macedonia and enters into negotiations with the emperor in order to obtain free passage to Gallipoli and ships to take him across the straits. Watched by strong Byzantine forces these Turks are conducted to the Dardanelles, but at the sight of their horses, their money, ahd their other booty the Byzantines decide to annihilate them [88]. Aware of the danger, the Turks


entrench themselves on the peninsula. Reinforced by Turks whom they summon from Asia Minor, they repulse and defeat their aggressors [89]. Once more for two years they devastate the neighbouring districts of Thrace [90] until the Byzantines are compelled to make a supreme effort [91]. This is the end of Khalīl’s Turks: in a desperate effort to break out and to find ships for the crossing to Qaras, they perish (1311) [92].

Gregoras’ account is supplemented by that of Pachymeres, whose history, however, stops at 1308. According to him the Tourkopouloi served under their own officers, [93] were Christians of recent date, and had come only lately to the emperor — from the ‘Northern regions’ [94]. After the battle of Aproi they move to Gallipoli [95] to join the Catalans [96]. Their wives and children follow them but fall into the hands of the Alan mercenaries who send a small number of them to the emperor [97]. He uses them to draw the Tourkopouloi once more to his side [98] but has no success. In these negotiations a certain Isḥāq (Ἰσαάκ) had offered his good services. He is a mighty Turkish leader (probably from Qaras); though in the camp of the Catalans, he is ready to abandon them [99]. He must have nourished high ambitions since he gives himself the title of ‘Melik’ and asks in marriage the daughter of Sultan Mas'ūd, 'Izzeddīn's grand-daughter [100]. The princess had been left in Constantinople when Mas'ūd, after his return to Rūm, was compelled to take refuge momentarily with the Byzantines, and there grew up generously provided for by the emperor [101]. She has also an uncle there,


Melik Konstantinos (Μελὴκ Κονσταντῖνος), the second of 'Izzeddīn's two sons, who, unlike his brother Mas'ūd, had not left Constantinople, had readily accepted baptism and become a perfect Rhomaios [102]. Ishaq, now returned to Anatolia, asked that he should be sent to him and proposed nothing less than to make him sultan of Rūm. Though Andronikos II found this latter project inopportune, he nevertheless sent Melik Konstantinos with the princess, his niece, to Pegai (Bigha), in the Troas, still a Byzantine possession, to be governor there and at the same time arrange with Isḥāq the matter of the marriage [103]. The dubious game which Isḥāq played with the emperor, the Catalans, and the Tourkopouloi, ended in his death at the hand of the Catalans [104]. The leader of the Tourkopouloi, Taghachar [105], barely escaped the same fate but was eventually spared and left in command of his troops — only to desert at the first opportunity with a number of his men to the Byzantines [106].

Pachymeres does not tell us what became of Melik Konstantinos but it is clear that he is the Melik whom Gregoras shows, two years later, as chief of the Tourkopouloi when they separate from the Catalans and who leads them afterwards into Serbia (where he eventually perishes [107]). Residing at Pegai so near to the events [108] and being in contact with Ishaq over the marriage of his niece, he must have become involved in Isḥāq's intrigues to find himself one day, whether he wished it or not, drawn into the Catalan camp, where after Taghachar's flight to the Byzantines he was the obvious man to take the command of the Tourkopouloi.

Combining the two accounts we learn that the Tourkopouloi, who had deserted at, or after, Aproi and then were most intimately connected with Khalīl’s Qaras Turks, spent the two years 1307-8 in Gallipoli; there was ample time for their kinsfolk to be sent over to Qaras. If their women and children had, after Aproi, fallen into the hands of the Alan mercenary corps — though probably not all of them — there was still a possibility of recovering them. We do not know if those few (probably the wives and children of officers) whom the Alans handed over to the emperor and whom he subsequently used as a means to separate the Tourkopouloi from the Catalans, were eventually restored to them. Those who remained in the hands of the Alans may have been recaptured when the Tourkopouloi attacked and annihilated the Alans in the Balkan passes while the latter were crossing into Bulgaria for service with the


tsar — indeed, the whole action was undertaken with this aim in view [109]. The far-reaching raids the Tourkopouloi undertook throughout Thrace offered them other opportunities to recover dispersed groups of their families. At the end of Khalīl's adventure there were again two years, 1310-11, spent in the peninsula of Gallipoli so that, if some families had followed their men to Thessaly and back, there was time and opportunity to pass them over to Qaras. Passing twice close to Karaferia, Khalīl could easily have been joined also by Turks of that region. Dobruja Turks would have joined Khalīl rather during his first stay in Gallipoli, in 1307-8. Yazjoghlu's account establishes a connexion between their exodus and the advance of the Bulgarians — it was, indeed, just in 1307 that Andronikos II had to make peace with the Bulgarian tsar Svetoslav, leaving him in possession of all his conquests including the two important Black Sea ports of Anchialos and Mesembria [110]. This meant that the Dobruja Turks were now completely cut off from the empire and from their kinsfolk there. It is only too understandable that some of them should have tried to join the Tourkopouloi at Gallipoli. Finally, what Gregoras describes as the complete annihilation of all the Turks of Gallipoli in 1311, may well have been nothing more than the last phase in a prolonged evacuation to Qaras. Above all, Khalīl himself seems to have escaped to safety, for Gregoras would not fail to mention his death if he had felt able to do so.

As we see, these closing paragraphs of the account, like the earlier ones, prove to contain information which is founded in fact or is at least historically possible, indeed, even probable. The story of the Dobruja Turks therefore appears to represent a genuine tradition. Where should it live on if not in Qaras — among people descended from the Dobruja Turks (the Tourkopouloi and their kinsfolk) and in the tales commemorating the great adventure of Khalīl Eje. When the Ottomans repeated, a generation later, the Gallipoli adventure, destined this time to initiate an era of great conquests, they certainly made good use of the help and advice which the experienced Turks of Qaras were able to offer — hence the appearance in the Ottoman historical legend of Eje Bey. This Eje Bey, an old and tried warrior, who shows Orkhān's son Sulaimān Pasha how to cross over to Gallipoli, is obviously a man of Qaras, perhaps conceived, though hardly as Khalīl Eje himself [111], yet at least as representing the veterans of 1307-1311. There is one figure, unmentioned before in Ottoman history, who with the first Ottoman conquests in Rumeli suddenly appears on the scene as a brilliant leader; he may well be one of those Qaras Turks who only then entered into the Ottoman community (and therefore into their history) and were predestined to be in the forefront of the Rumeli adventure. This figure bears the name of Evrenos — a name which seems to


point to an origin from Varna, the chief town of the Dobruja (cf. the Avren Dagh, to the south of the town). These are, of course, and may never be more than, conjectures.

We have endeavoured to show that some of the Rumeli Turks could and probably did return with Khalīl Eje to Anatolia ; we have to stress, however, that their numbers must have been small. The bulk of the Tourkopouloi had gone with Melik Konstantinos to Serbia; another, smaller group, led by Taghachar, had rejoined the Byzantines; other Tourkopouloi, individuals or isolated groups, may never have deserted the emperor. As to those of their kinsfolk who had managed to reach Gallipoli in 1307-8, they may have been evacuated to Qaras before the march into Thessaly began — but their numbers were certainly limited. In the Qaras tradition, of course, only those counted who actually reached Anatolia, and their arrival was easily magnified into a mass immigration. Since as we know for certain the Tourkopouloi were Christians and this must also be true for their nomad kinsfolk with whom they maintained the closest contact, the newcomers arrived in Qaras not as Muslims but as Christians. They had, however, to turn Muslim at once — if only in appearance, and their Christian past fell into oblivion or was at least not openly spoken of. When Yazjoghlu presents them as Muslims who had left Rumeli because of their distaste for life among the Infidels, he is no doubt to a large extent following the Qaras tradition. In two other points, however, it is rather his reluctance to mention apostasy from Islam, which must have led him into misrepresentation of the facts: the Turks who remained in Rumeli after the events of 1307-11 were by far the majority and they were by then no longer Muslims, having accepted baptism perhaps a generation before. These christianized Turks in the Dobruja and their splinter-groups in Karaferia and Zikhna, explicitly attested by our account (§13), are beyond all possible doubt identical with the Gagauz, those Christian Turks who speak a Turkish of Anatolian character and until modern times had their main abode in the Dobruja, with small isolated colonies at Karaferia and Zikhna.

To conclude: the account as a whole shows remarkable consistency and chronological soundness, but these merits are no doubt to a great extent the result of Yazjoghlu's skill in blending the four stories into a ‘Destan of 'Izzeddīn's and his people's exile in the Dar ul-Harb’. By the sultan's flight to the Byzantines he himself and all those who had followed him — his family, his warriors, and the nomads — became involved ever more deeply in a situation most perilous for Muslims. The destan carries the adventure of each of the dramatis personæ to a point where for a Muslim it has to end — in death, in return to the Dār ul-Islām, or in apostasy. Throughout, Yazjoghlu has endeavoured to spare the feelings of the Muslim reader: ‘Izzeddīn dies in exile but on Muslim soil and one of his sons returns to the throne of Rūm; another son ends as a Muslim saint; the third one dies as a faithful Muslim among the Infidels — it is true, alas, his descendants become Christians, nevertheless they are valiant soldiers fighting for the Ottoman sultan, the champion of Islam;


the warriors and the nomads eventually return to Anatolia, to the Dār ul-Islām — alas, some of them remain in Rumeli and turn Christian. Even in this form the account must have shocked pious Muslims, for Loqmān thought it wise to suppress the later part of the account and thus to avoid all mention of apostasy; for him the story ends with the emigration to the Desht-i Qpchaq, i.e. to the safety of the Dār ul-Islām.

Already Ibn Bībī's chapter on 'Izzeddīn's flight is in itself a destan but strictly limited to the sultan's own adventure. To expand this destan into a much richer one comprising also the adventures of 'Izzeddīn's people, was certainly suggested to Yazjoghlu by the dominant role 'Izzeddīn played in a number of other stories which had come to his knowledge. Indeed, the Dobruja Turks not only enter the empire at 'Izzeddīn's summons, they also follow him into the Desht-i Qpchaq; a number of them join his son at Karaferia while his other son, the future Baraq, naturally finds shelter with his father's people in the Dobruja. The intimate connexion in which these stories show the sultan and his sons on the one hand and his people on the other, must be the reflection of an underlying historical reality, all the more since in the Byzantine accounts of events which took place 40 years after 'Izzeddīn's disappearance from the scene, the sultan's name is still very much alive and his son appears as the natural leader of the Tourkopouloi. It would certainly not be surprising if a Turkish community were to take their name from the man who had shaped their destiny. In this light Balaschev's ingenious derivation of ‘Gagauz’ from ‘Kaikāūs’ becomes self-evident [112]. The Gagauz are, indeed, the people of Kaikāūs.

It lies beyond the scope of this article to deal with all the problems involved [113] and to attempt a reconstruction of the historical reality. I shall be satisfied if my study restores the reputation of a source whose value and importance the Russian scholars Bruun and Smirnov had rightly perceived. To their illustrious compatriot Vladimir Minorsky, whose work has thrown so much light on the history of the Turkish peoples, I offer these pages as an expression of gratitude and homage.


1. The easternmost chain of the Balkan range which here, near to the coast where Greek survived the longest, retained its classical name of Haemus : Αἵμον (acc.) > Emine, with an intermediate form *Ἔμμωνα, Emona, found in documents of the early 14th century; see C. Jirecek, Das Füerstenthum Bulgarien, Vienna, 1891, pp. 4 and 527, n. 1.

2. T. Kowalski, Les Turcs et la lange turque de la Bulgarie du Nord-Est. Polska Akademja Umiejetnosci, Memoires de la Commission Orientaliste No. 16, Cracow, 1933, 28 p. In the introductory pages of this study is to be found an excellent survey of the rich literature on the subject. Kowalski has supplemented this paper by two shorter ones: 'Compte-rendu de l'excursion dialectologique en Dobroudja, faite du 10 septembre au 1 octobre 1937,' in : Bulletin de l’Academie Polonaise des Sciences et des Lettres, Cracow, 1938, pp. 7-12, and 'Les elements ethniques turcs de la Dobroudja', in : Rocznik Orjentalistyczny, xiv, 1938, pp. 66-80.

3. I have dealt briefly with this account in my article 'La descendance chretienne de la dynastie Seldjouk en Macedoine', in: Echos d'Orient, xxx, 1934, pp. 409-12, and more fully in my study 'Les Gagaouzes = les gens de Kaykaus', written for the Tadeusz Kowalski Memorial volume in 1948. I still hope it will one day appear in print, since it is by no means superseded by this present article; on the contrary, both are complementary one to the other.

4. The opuscule is entitled Ijmal-i ahval-i al-i seljuq ber mujib-i naql-i Oghuz-name. On Loqman see F. Babinger, Die Geschichtsschreiber der Osmanen und ihre Werke, Leipzig, 1927, pp. 164 seqq., where, however, p. 167, n. 1, Loqman's indisputable authorship of the Ijmal is, without any justification, denied. To the information given there can be added the firman in Ahmed Refiq, Istanbul hayati, i, Istanbul, 1333h. = 1917-18, p. 52, No. 5 (German translation by G. Jacob in Der Islam, ix, 1919, p. 251), and the accompanying note, both of which give interesting information. For Loqman's famous Hünername (by no means missing from the Topkapi Sarayi!) see also J. Karabacek, Zur orientalischen Altertumskunde IV = SB. Akad. d. W. Wien, cxxii, 1 (1913). Of Loqman's Qiyafet ül-insaniye a new MS. has in the meantime been brought to notice and described in Fehmi Edhem and Ivan Stchoukine, Les manuscrits orientaux illustres de la Bibliotheque de l’Universite de Stamboul, Paris, 1933 (Mem. de l'Institut Francais d'Archeologie de Stamboul I), p. 1 and pl. I.

5. In his Geschichte des osmanischen Reiches, i, p. 122 ; ii, p. 143 ; iii, p. 202 ; viii, p. 354, and also in his Geschichte der Goldenen Horde, p. 174. Hammer found Loqman's opuscule in a MS. of his own between two works of Lutfi Pasha wherefore he ascribed it to the latter, quoting it as 'Lutfi's Oghusname oder Geschichte der Seldschuken', Hammer's copy, apparently still the only known one, is now in the Nationalbibliothek of Vienna, H.O. 17b = Flüegel, No. 1001, 2.

6. I. J. Lagus, Seid Locmani ex libro turcico qui Oghuzname inscribitur excerpta. Helsingfors, 1854.

7. Of the 15 paragraphs which constitute my resume of the whole account (see below, pp. 648-51) only the first nine are found in Loqman. On the other hand he has made some additions — above all the date which he gives for the immigration into the Dobruja, right at the beginning, in the form of a distich : —

This date 662h. = 1263-4 is clearly the result of calculation, nevertheless, it cannot be far from the truth.

Another addition is his statement (ed. Lagus, Turk, text, p. 7) that the story of Sultan 'Izzeddin's flight to the Byzantines is to be found in the 'Destan of the blessed Sari Saltiq'; in his source the chapter where Sari Saltiq appears is in fact headed 'Flight of Sultan 'Izzeddin to the Byzantines'; clearly Loqman' s interest was so exclusively concentrated on Sari Saltiq that he felt the rest of the account to be merely a subordinate framework.

8. As Th. Menzel did in his article 'Gagauz' in the Encyclopaedia of Islam. The account was mercilessly analysed and rejected by the Bulgarian historian P. Mutafciev, Die angebliche Einwanderung von Seldschuk-Türken in die Dobrudscha im XIII. Jahrhundert, Sofia, 1943, 129 p.; (, lxvi, 1). He found the support of an Orientalist: H. W. Duda, Zeitgenössische islamische Quellen und das Oguzname des Jazygyoglu 'Ali zur angeblichen turkischen Besiedlung der Dobrudscha im 13. Jhd. n. Chr., ib. (Bulg. Ac., lxvi, 2), pp. 131-145. Other scholars who had doubts about the value of the information found in Loqman are listed in Mutafciev, p. 13, n. 1 (for those who accepted it see ib., p. 12, n. 2).

The late Mutafchiev was obviously inspired by his patriotic zeal to show the Dobruja Turks as Bulgarians who had adopted the Turkish language. In defence of his hopeless case he displays an enormous erudition and his book is certainly the most comprehensive study of the subject ever made: he reviews at length the various theories of his predecessors and at the same time he brings together a rich information from the sources, subjecting it to a thorough though strongly biased commentary. I wish to state that I am much indebted to Mutafchiev's work.

In the main the book is a bitter attack on a pamphlet by the Bulgarian scholar G. D. Balaschev:
Sofia, 1930, 26 p., where Loqman, reproduced in Greek translation, is accepted almost without criticism and used to reach rather rash and exaggerated conclusions. Nevertheless, Balaschev, too, has his merits, above all, his explanation of the name of 'Gagauz' (see below, p. 668) is a brilliant and important discovery. For Mutafchiev, of course, Balaschev has only the negative merit of discrediting, by carrying it ad absurdum, the identification of the Gagauz with 'Izzeddin's Turks, a theory which had been advanced by Bruun, , ii, Odessa, 1880, p. 333, and accepted by Smirnov, , St. Petersburg, 1887, p. 17 (see Mutafciev, p. 82, n. 1, and p. 84, n. 2).

9. The first to state Loqman's dependence on Yazijioghlu was M. Th. Houtsma in his Recueil des textes relatifs a l’histoire des Seldjoucides, iii, p. x; he shows him, however, not as dependent on Yazijioghlu himself but on an abridged version of his work — indeed, an abridgement, MS Paris, Bibl. Nat., Suppl. turc 1182 (see E. Blochet, Catalogue des manuscrits turcs, Paris, 1932-33, ii, p. 190) was all that Houtsma had for comparison.

Mutafchiev (p. 16, n. 1) knew from my remarks in Der Islam, xx, 1932, p. 202 seq., that Loqman comes from a much older source but took no account of it. The intervention of the orientalist Prof. Duda reassured him, indeed, that Loqman's information, though taken from Yazijioghlu, did not occur in the contemporary Seljuq sources — and therefore was worthless. Both these scholars ignored my article in Echos d'Orient (see above, p. 640, n. 2) which would have shown them that Loqman has omitted not only the larger but, indeed, the essential part of Yazijioghlu's account.

10. Complete manuscripts: two in Istanbul, Topkapi Sarayi, Revan Kosku, 1390 and 1391, one in Berlin, Staatsbibliothek, Orient Quart 1823 (from these three MSS. my notes are taken), and one in Paris, Bibliotheque Nationale, Supplement Turc 737 (Blochet, Cat., ii, p. 47).

Incomplete manuscripts: Leyden, Warner 419, Paris, Bibl. Nat., Ancien fonds turc 62 (Blochet, Cat., i, p. 24) and Suppl. Turc 1185 (Blochet, ii, p. 191).

Edition : M. Th. Houtsma, Recueil III: Histoire des Seldjoucides d'Asie Mineure d'apres Ibn Bibi, Texte turc, Leyden, 1902, based on the two incomplete MSS. Leyden and Paris, A.f. 62, reproduces only that portion of the text which is in the main a translation from Ibn Bibi and this only as far as the then available MSS. permitted, i.e. less than the half. The end of Rec. III corresponds to Rec. IV (Ibn Bibi), p. 159, ult.

An edition of the complete work is, as I was glad to hear, being prepared by the Turkish Historical Society (Türk Tarih Kurumu).

11. H. W. Duda in ZDMG., N.F. 14 (89), 1935, p. *19* seq.

12. Istanbul, Library of the 'Aya Sofya, No. 2985 : Al-awamir al-'alaniya fi'l-umur al-'ala'iya, by Husain b. Muhammad b. 'All al-Ja'fari ar-Rughadi () al-mushtahir bi-Ibn Bibi al-munajjima (MS., p. 10, 1. 7), 744 pages; cf. F. Tauer, ' Les manuscrits persans historiques des bibliotheques de Stamboul, iv,' in: Archiv Orientalni, iv, 1932, p. 92.

The Turkish Historical Society have announced their intention to publish a facsimile of the manuscript.

13. Notice on the donation, accompanied by the sultan's seal, on the frontispiece.

14. Edited by M. Th. Houtsma, Recueil IV: Histoire des Seldjoucides d'Asie Mineure d'apres l’abrege du Seldjouknameh d'Ibn Bibi, Texte persan, Leyden, 1902.

Prof. Duda has claimed (loc. cit.) that the abridgement was made in the lifetime of Ibn Bibi, assuming that the Malik-i (or: amir-i) diwan at-tughra Amir Nasir ad-din Yahya al-ma'ruf bi-Ibn al-Bibi (Rec., iv, p. 2,1. 3, and p. 196,1. 2) is the author of the original — which he is most certainly not since from the original we know (see above, n. 3) that its author was named Husain (al-mushtahir bi-Ibn al-Bibi); there is no mention in the original of an Emir Yafrya nor of the office which according to the abridgement he held. The chapter heading in Rec., iv, p. 196, appears in the original MS., p. 442, simply as . Prof. M. Fuad Köprülü (in Belleten, vii, 1943, p. 388 seq.) is certainly correct in regarding the Emir Yahya as the brother of Husain but errs in believing that he is the author of the abridgement which explicitly attributes to him the authorship of the original (). I see no solution of this puzzle other than the assumption that the author of the abridgement, who remains anonymous, has attributed to Yahya Ibn Bibi what belonged to Husain Ibn Bibi, probably after the latter's death and in order to ingratiate himself with his new superior in the divan. Though we do not know what office Husain had held, it is probable that he was his brother's predecessor as malik-i diwan at-tughra. It is the very office which their father Majd ad-din Muhammad Tarjuman had held until his death in 1272. He is described as chief of the insha office, which is probably identical with the diwan at-tughra (see W. Bjorkman, Beiträge zur Geschichte der Staatskanzlei im islamischen Aegypten, Hamburg, 1928, p. 44, n. 4). (According to the original he was appointed to this post after having been head of the 'tent and carpet depot' firash-khane-i khass [MS., p. 443, 1. ii : ].)

At least it is certain that the abridgement was, like the original, written in the reign of Mas'ud; Rec., iv, p. 334, 1. 20 seq.:  belongs to the epitomist since in the original, MS., p. 736, 1. 2, the corresponding passage reads: .

15. Muhammad ar-Rawandi, Rahat as-sudur (composed in 1202-3, retouched between 1207 and 1210), ed. Muh. Iqbal, London, 1921 (Gibb Mem. Ser., N.S. ii).

16. I am not dissuaded from transcribing Qutlumush by Prof. Fuad Köprülü's argument in Tarih Dergisi, i, 1950, pp. 227-230; the passage from 'Ayni's 'Iqd al-Juman quoted in his postscript (ib., p. 236) even gives welcome support to my reading.

17. The last passage taken from Rawandi (p. 128, 1. 5):

reads in Yazijioghlu as follows:

(with this heading Rec. III begins).

In the 'fore-runner' (between the asterisks) Yazijioghlu shows himself influenced by Hamdullah Mustaufi's Ta'rikh-i Guzide (Gibb Mem. Ser., xiv, 1), which reports immediately before the date 471 h. Sulaiman's dispatch by Malikshah, not against Rum, but against Antioch; ib., p. 441, the name of the Byzantine emperor is given as , to be read Urumanos, i.e. 'Romanes'; cf. p. 481, where it is corrupted into . On the other hand, Yazijioghlu's , i.e. fasilyevs, is a first indication of his turning to Ibn Bibi where this is the usual term for the Byzantine emperor (Rec., iv, p. 14, 11. 2 and 19, p. 15, 11. 9 and 13, etc.); indeed, Sulaiman's dispatch to Rum by Malikshah is mentioned in Ibn Bibi, in Qilij Arslan's speech to his son (MS., p. 18, 11. 14-16 ; not in Rec., iv, p. 3): 
(= Rec. iii, p. 9, 11. 6-9).

18. Houtsma states (Rec. iii, p. ix) that in the abridged version of the Oghuzname (see above, p. 641, n. 3) the 'continuation' refers here and there to sources, e.g. to the Ta'rikh-i Guzide. As far as my reading of the original work goes I have come across no such reference.

19. I have reproduced, translated, and analysed the beginning of this closing chapter in my Das Füerstentum Mentesche, Istanbul, 1934, p. 32 seqq.

20. The MSS. used by Berezin for his edition of Rashideddin's Jami' et-tevarikh (, vii, St. Petersburg, 1861, pp. 32-8), contain the tamghas though probably in a form far inferior to the fine tables found in the MSS. of Yazijioghlu's Oghuzname. In H. Vambery, Das Türkenvolk, Leipzig, 1885, pp. 4-6, and (incomparably better) in L. A. Mayer, Saracenic Heraldry, Oxford, 1933, pls. l and li, the tamghas (in the last named work the entire tables) are reproduced from the Leyden MS. — they are still more finely executed in the Berlin MS.

Occasionally Yazijioghlu makes additions to Rashideddin's text, some of which are not without interest, e.g. when he expands Rashideddin's (ed. Berezin, p. 5)

in the following way (MS. Berlin, f. 2b):

This addition (from the asterisk onward) can be regarded as an observation of his own despite the fact that he refers to raviler. Some lines later he mentions an Oghuzname in Uighur characters (); we read further on (f. 3a) that information on the Oghuz is to be found 'in the Jami' et-tevarikh and in the Oghuzname'. It is possible that here Yazijioghlu has in mind the famous Uighur Oghuzname made known by Radloff and subsequently studied by Riza Nour, Pelliot, Bang, and Arat — in his own time uigurica were, indeed, in fashion at the Ottoman court — but as far as my notes go there is no passage which could be traced back to that text; what I have said on this in Der Islam, xxx, p. 202, has therefore to be corrected.

21. Numerous examples in Rec. III, e.g. p. 74, 1. 21 (Oghuz boylari); p. 80, 1. 20 (beys) ; p. 99, 1.11 (deeds); p. 204, 1.3, p. 205, 1.17 (feast, verses); p. 217, 1.12 (Sultan 'Alaeddin, the Oghuzname and the türe). It goes without saying that all these passages are missing not only in the abridgement but also in the original of Ibn Bibi.

22. The passage on 'Osman's election' has been reproduced, in a very much shortened form, by Münejjimbashi, Jami' ad-duval, iii, p. 278, and before him, at greater length, by Lutfi Pasha, Ta'rikh, p. 21, and by Ruhi Edrenevi (see J. H. Mordtmann in Mitteilungen zur osmanischen Geschichte, ii, p. 136). The influence of Yazijioghlu's work appears indeed very early in Ottoman historical writing.

23. Though most of the elements of which these interpolations are made up come clearly from Rashideddin, some of them as well as the vivid way in which they are presented may well be due to Yazijioghlu's personal knowledge of oghuz traditions; cf. Abdülkadir in Türk Hukuk ve Iktisat Tarihi Mecmuasi, i, 1931, p. 123 seq.

24. It would be unthinkable, of course, that there should be no Ghazi theme in this work. It appears frequently, often linked with the oghuzian theme, and is introduced right at the beginning of the work (in the MS. Berlin, 1. 6 of the first page) :

(The subject of verdi is, of course, God.)

25. From the numerous examples found in Rec. III, I quote p. 10, 1. 15, where the yazijilar are added to Ibn Bibi's text in order to make them appear among the grandees, and above all the long poem, pp. 253-7.

26. Rec., iii, pp. 87, 372, and 382.

27. H. W. Duda, Zeitgenossische islamische Quellen, etc. (see above, p. 641, n. 2), p. 138, n. 4 and 6, quotes from MS. Berlin  and (I add the first hemistich):  as well as the distich which reveals, or rather hides, the date: 

Prof. Duda has succeeded in wresting from this chronogram an acceptable date, understanding that the reader is invited to add to  = 540 the number value of  = 287, which gives as date 827 h./beg. 5 December 1423.

28. See in the preceding note Prof. Duda's interpretation, which can, however, just as well yield as date 833/beg. 30 September 1429, if we take into account the tashdid of  and reckon the waw twice.

29. Rec., iii, p. 87, 11. 15-19.

30. See below, p. 652.

31. The 'Testament', fully vocalized at the beginning to give it the aspect of a sacred text, and headed (MS. Berlin, f. 18a)
follows immediately the tables containing the tamghas; though it replaces Rashideddin's account (ed. Berezin, vii, p. 38 seq.), it is obviously very much influenced by it.

32. The yük 'bone' is a certain part of a certain animal which at the festivities is reserved for a certain tribe. See M. Th. Houtsma, 'Die Ghusenstämme,' in WZKM., ii, 1888, p. 229.

33. MS. Berlin, f. 19a:

34. Following immediately the preceding quotation:

This passage has been placed at the beginning of the Kitab Dede Qorqud (ed. Killisli Mu'allim Rif'at, Istanbul, 1332, p. 3, 11. 1-6); it is alien to the rest of the book and visibly a later addition.

35. Foreshadowed by the 'fore-runner' Rec., iii, p. 218, 1. 1.

36. See the remarks on the introduction of the khan title in my 'Notes sur la tughra ottomane [II]' in Byzantion, xx, 1950, pp. 279-282.

37. In Der Islam, xxx, p. 203, I have presented Yazijioghlu 'Ali and the famous mystic writer Yazijioghlu Mehmed (and implicitly also the latter's brother Yazijioghlu Ahmed Bijan) as 'brothers', in the belief (1) that Yazijioghlu 'Ali wrote in the later years of Murad II's reign, and (2) that Yazijioghlu Mehmed was, as Evliya, Seyahatname, iii, p. 366, says, the author of a risale on Sari Saltiq, which seemed to me to explain the interest his 'brother' 'Ali took in this holy man. Both assumptions have proved to be wrong. It is obvious that Evliya has mistakenly attributed our 'Ali's account, which he like Loqman (see above, p. 641, n. 1) calls a 'Saltiq-name', to the much better known Yazijioghlu, i.e. Mehmed. It remains nevertheless true that all three have in common a family name of great distinction which was probably reserved for the members of one family, that all were literary people though in very different fields, and that they were contemporaries, for Mehmed and Ahmed were well advanced in age when they began writing about 1450, almost a generation after 'Ali had composed his Oghuzname. Still another Yazijioghlu is known: Neshri (Gihannüma, ed. F. Taeschner, i, p. 65 ; edd. F. R. Unat and M. A. Köymen, i, p. 239) mentions a Yazijioghlu as ambassador to Egypt in the later years of Murad I's reign — a mission often entrusted to a high divan official. This ambassador could well be our 'Ali himself or his father, just as the two mystic writers could be his brothers or his sons (that Katib Salaheddin has been regarded as their father [cf. Gibb, History of Ottoman Poetry, i, p. 390 seq.] is obviously nothing but an inference from katib). As matters stand one can at best speak of a certain probability that the four Yazijioghlus belong all to one and the same family.

38. Text (MS. Berlin) and German translation in H. W. Duda, Zeitgenossische islamische Quellen, etc., p. 143 seq.

39. Text (MS. Berlin) and German translation in Duda, op. cit., p. 144.

40. MS. Revan K. 1391, f. 411b:

41. Cf. the last chapter in Rec. IV.

42. ; this name was known to Yazijioghlu from Ibn Bibi who, however, writes it ; missing in Rec., iv, p. 334, 1. 15, where the original, MS., p. 735,1. 6, reads as follows:
Could the change into  (bela 'calamity') be meant as a pun?

43. MS. Revan K. 1391, f. 415a:

44. For this comparison see my 'Islam und Kalifat' in: Archiv für Sozialwissenschaften und Sozialpolitik, liii, 1925, p. 412, especially n. 82. The anonymous history of the Rum Seljuqs there referred to (MS. Paris, suppl. pers. 1553) has ; see Feridun Nafiz Uzluk, Anadolu Selguklari tarihi, Ankara, 1952, facsimile, p. 42, 1. 6.

45. MS. Revan K. 1391, f. 415a:

46. MS. Revan K. 1391, f. 415b:

47. MS. Revan K. 1391, f. 444a:

( The passage has been quoted in transcription by Aurel Decei in his article 'Dobruca' in Islam Ansiklopedisi, iii, p. 632b.)

48. See above, p. 646.

49. Dobruja-eli is one of the numerous designations of countries formed by a name + el-i, the possessive suffix indicating that the country is regarded as belonging (or having belonged) to the person or people named in the first element; thus it means 'Land of Dobruja (Dobrotitsa)', as Chalkokondylas, ii, p. 98, 1. 15 Darko, says: .

That 'Dobruja' = 'Dobrotitsa' becomes quite clear from Neshri (ed. Taeschner, p. 68,1. 13; edd. Unat-Koymen, p. 249, 1. 15:  where Dobrotitsa's son and successor Ivanko appears under the name of 'Dobruja oghlu'.

50. Rec., iv, p. 296, 1. 1-p. 297, 1. 12.

51. Georgius Pachymeres, i, p. 130, 1. 17-p. 132, 1. 16, and ii, p. 609, 1. 12-p. 611, 1. 15, Bonn. Nicephorus Gregoras, i, p. 82, Bonn.

52. Rec., iv, p. 298, 1. 11, the epitomist speaks of the episode as an efsane but this expression is not to be found in the original.

53. H. W. Duda, Zeitgenossische islamische Quellen, etc. (see above, p. 641, n. 2), p. 145.

54. See V. Laurent, 'La domination byzantine aux Bouches du Danube sous Michel VIII Paleologue,' in Revue du Sud-Est europeen, xxii, 1945, pp. 184-198.

55. Pachymeres, i, p. 133, 11. 3-15, is, however, explicit at least about the fact that at the time 'of 'Izzeddin's stay in the empire Turkish nomads () appeared on the Eastern frontier: hating all discipline, loath to submit to the Tatars, and anxious to be left alone, they infiltrated into the Byzantine defences, proclaiming themselves to be allies of the emperor but none the less plundering under the cover of darkness, though the frontier people were able to hold them in check. Finally the emperor bound them closely to his service, i.e. used them for military purposes. This measure, which meant the settlement of these nomads in the Eastern frontier zone, was designed to strengthen that frontier against the Tatars who, in spite of the prevailing good relations, had nevertheless to be deterred from all aggressive intentions. The dangers inherent in the measure may soon have become apparent, since there was no guarantee that at a given moment these Turks might not make common cause with nomads beyond the frontier. Besides the European campaigns drained off from Anatolia all available reserves for service in the Balkans. The soldiers levied among the newly arrived nomads may, sooner or later, have also been sent there — and their kinsfolk with them. It is not impossible, therefore, that Pachymeres' nomads and those of Yazijioghlu's account are one and the same.

56. Cf. Rec., iv, pp. 297, 1. 12—298, 1. 2. The original, MS. p. 639, 1. 2, is still more explicit:

57. Pachymeres, I, 131, 1.2: . See also EI., s.v. ' Kaika'us II', where on the testimony of Frater Simon in Vincentius Bellovacensis, lib. xxxi, cap. 26, she is said to have been the daughter of a Greek priest.

58. Tafel and Thomas, Urkunden zur älteren Handels- und Staatsgeschichte der Republik Venedig, vol. ii, p. 208, in the Tributa Lampsacenorum, of 1219: de CXX plinthis de vineis quas receperunt pro anacapsi pp. (= perpera, hyperpers) VIII annuatim. The term anacapsi of this Latin text can be regarded as a perfect rendering of the vernacular form for , the annual payment due by a special category of tenants, the . It was good luck, indeed, to come across Prof. D. A. Zakythinos' brilliant study on 'La Societe dans le Despotat de Moree' in L’Hellenisme Contemporain, 2nd ser., vols. iv and v (Athens, 1950 and 1951), and to find there, at the very last moment, just what was needed to solve the anaqapisi problem with which I had been struggling in vain for years.

59. Cf. Rec., iv, p. 298,11. 2-6.

60. The story of 'Izzeddin's projects, imprisonment and liberation in Pachymeres, i, pp. 229, 1. 3-240, 1. 22, and in Gregoras, i, pp. 82, 1. 10-83, 1. 2 ; 99, 1. 21-101, 1. 19, also in Aqsarayi, ed. O. Turan, pp. 75, 1. 5-76, 1. 12. His liberation, in Maqrizi, Suluk, i, Cairo, 1934, p. 522 (transl. Quatremere, Histoire des Sultans Mamlouks, I 2, p. 57 seq.); other Arabic sources in W. de Tiesenhausen, Recueil de materiaux relatifs a l’histoire de la Horde d'Or, i, St. Petersburg, 1884, pp. 81 (Rukneddin Baibars), 133 (Nuwairi), 179 (Mufaddal), 200 (Dhahabi), 482 ('Aini).

61. Cf. Rec., iv, p. 298, 11. 7-9.

62. Cf. Rec., iv, p. 298, 11. 9-11.

63. Gregoras, i, p. 101, 11. 16—19:
Also p. 229, 11. 11-17, and p. 248, 11. 6-10.

64. Pachymeres, ii, p. 574, 11. 5—7:

65. Ib., p. 550, 11. 15—19:

66. See above, n. 1.

67. Gregoras, i, p. 248, 1. 9 seq.: 

68. The türbe of the saint still exists at Aqshehir (see F. Sarre, Reise in Kleinasien, Berlin, 1896, p. 22). Of its two inscriptions, one, bearing the date 1224, comes from a mosque and is here re-employed as an ornament (Cl. Huart, Epigraphie d'Asie Mineure, Paris, 1895, no. 15); the other, above the door, mentions the restoration of the turbe by the saint's great-grandson Seyyidi Muhyi ed-din in 812 h. = 1409-1410. I give here its full reading to replace the incomplete copy in Huart, no. 16:

The three wooden coffins, wonderfully carved (F. Sarre, Seldschukische Kleinkunst, Leipzig, 1909, ii, pi. 14, shows them still in situ), are now in the Museum of Turkish and Islamic Antiquities (the former Evqaf Müzesi); in the guide book of the museum (Türk ve Islam eserleri rehberi, Istanbul, 1939, p. 13, nos. 193, 191, and 194) are fairly correctly reproduced the inscriptions of the three sarcophagi, i.e. of our saint Mahmud b. Mas'ud, died in 667 h. = 1268-9, of his brother Ahmed b. Mas'ud, died 649 h. = 1251-2, and of his grandson 'Ali b. Mehmed (i.e. Muhyi ed-din) b. Mahmud er-Rufa'i (the father of the restorer of the turbe), without date.

Mahmud al-Hayrani appears as a contemporary of Jelaleddin Rumi (died in 1273) in Eflaki ; see Cl. Huart, Les Saints des Derviches Tourneurs, Paris, 1918-1922, ii, p. 108. E. Gross, Das Vilajet-name des Haggi Bektasch, Leipzig, 1927 (Turkische Bibliothek xxv), p. 80 seq., shows our saint as claimed by the Bektashi, just as it is the case with Sari Saltiq (ib., p. 73).

69. See Köprülüzade M. Fu'ad in Darülfünun Edebiyat Fakültesi Mejmu'asi, ii, 1922, p. 292 seq., and more fully in his Influence du Chamanisme Turco-Mongol sur les Ordres Mystiques Musulmans, Istanbul, 1929, pp. 14-17; cf. also the same in Belleten, vi, 1943, p. 431, n. 1, where Prof. F. Köprülü promises a monograph on Baraq and also a study on Sari Saltiq which will make use of a newly discovered Saltiq-name of the late 15th century.

70. G. Moravcsik, Byzantinoturcica, Budapest, 1942-3, ii, records s.v. Parak two baptized Tatar women who died in 1280 and 1308 respectively, and s.v. BarakoV an Ottoman army chief of the early 15th century as well as  in an Athos document of 1292. For Mongol and Turkish rulers of the name of Baraq see EI. s.v., and Khalil Edhem, Düvel-i islamiye (index) or E. de Zambaur, Manuel de genealogie (index, s.v. Boraq). For the central asiatic Baraq Khan, who in the years 1422-7 dominated the events in the realm of the Golden Horde, see B. Spuler, Die Goldene Horde, Leipzig, 1943, p. 156 seq.

71. For this is doubtless its meaning. In classical antiquity the saliva was considered as a means of conferring spiritual power; see e.g. J. Davreux, La legende de la prophetesse Cassandre, Paris, 1943, p. 69, and Prof. R. Goossens' remarks thereon in L'Antiquite Classique, xiii, 1944, p. 178 seq.

72. M. Tayyib Okic, ‘San Saltuk'a ait bir fetva’ in: Ankara Üniversitesi Ilahiyat Fakültesi Dergisi, i, 1952, pp. 3-13, giving (p. 10) the text of the fetva: the question submitted by Sultan Sulaiman I:  ‘Is the person known by the name of Sari Saltiq a saint?’ and Abu's-Su'ud's reply: 

73. See above, p. 655, n. 3.

74. Pachymeres, i, p. 258, 1. 7:

75. Ib., p. 259, 11. 11-13:

76. Gregoras, i, p. 94, 11. 10-19.

77. Cantacuzenus, i, p. 269, 1. 21, Bonn:  and pp. 271, 1. 12-272, 1. 22.

78. Ib., iii, pp. 161,1. 7-163,1. 3. His appointment as commandant of Edessa, ib., pp. 129 ult.-130, 2: . Probably Lyzikos was one of those notables of Berrhoia who had fled before the Servian occupation to the emperor, with whose army they returned to the reconquest of their town; he was certainly among the many men of Berrhoia who then went with the emperor to recapture also the neighbouring Edessa from the Servians (ib., p. 123,11. 1-5, and p. 127, 1. 2).

79. Sp. Lampros, , ed. K. I. Amantos, Athens, 1932, no. 29, 1. 7 ; no. 42, 1. 29; no. 49,1. 46: 8 May 6895 A.M. = 1387 ( is obviously a corruption of ).

80. I owe this information to my friend and colleague Prof. Halil Inalcik, of Ankara University, who is engaged in preparing the edition of this defter; for the present see the brief information he gives on this matter in Belleten, xv, 1951, p. 650.

81. It seems to me very probable that in the important movement connected with the sheikh's name the ‘people of Kaikaus’ played a foremost role, politically as well as ideologically, and that Bedreddin was involved in it because of his descent from Kaikaus. (Since Prof. F. Babinger's inspiring monograph in Der Islam, xi, 1921, pp. 1-106, the Bedreddin movement has been the subject of many studies of which the most recent is H. J. Kissling, ‘Das Menaqybname Scheich Bedr ed-Din's, des Sohnes des Richters von Samavna,’ in ZDMG., c, 1950, pp. 112-176.)

82. Gregoras, i, pp. 227,1. 4-233,1. 13; pp. 244,1. 16-249,1. 2; pp. 254,1. 2-258,1. 14; pp. 262, 1. 20-269, 1. 23, Bonn.

83. Ib., p. 228, 11. 22-4:

84. Ib., p. 229, 11. 11-17 ; cf. above, p. 657, n. 1.

85. Ib., p. 232, 11. 11—13:

86. Ib., p. 248, 11. 5-15.

87. Ib., p. 248, 11. 18—20:

88. Gregoras, i, pp. 254, 1. 2-255, 1. 14:

89. Ib., pp. 255, 1. 14—256, 1. 3:

Ib., pp.; 257, 1. 11-258, 1.8: Michael IX's defeat. Here we read (p. 257, 11. 14-17) that the Turks have their women with them. Ib., p. 258, 11. 11-14: Khalil's triumph

90. Ib., p. 265, 11. 15—18:

91. Ib., pp. 267, 1. 23-268, 1. 14. The action starts with a measure to prevent Khalil from getting reinforcements from Anatolia:

92. Ib., pp. 268, 1. 15-269, 1. 23: Only those are spared who fall into the hands of the Genoese of Galata who, commanded by their podesta, are present with their ships as allies of the emperor: (p. 269, 1. 21)

93. Pachymeres, ii, pp. 523, 1. 18-524, 1. 2:

94. Ib., p. 574, 1. 5 ; see above, p. 657, n. 2.

95. Ib., p. 590, 1. 10 seq.: 

96. And their Turks, whose presence in the Catalan camp Pachymeres had mentioned before the battle of Aproi, p. 550, 1. 2: 

97. Ib., p. 590, 11. 11-14.

98. Ib., p. 591, 1. 17.

99. 1 Ib., p. 591, 11. 1-7:

100. Ib., pp. 591, 1. 12 ; 608, 1. 18 ; 612, 1. 16.

101. Ib., p. 612, 11. 11-13.

102. Ib., p. 612, 11. 13-16.

103. Ib., pp. 612, 1. 18-613, 1.7.

104. Ib., pp. 631, 1. 13-632, 1.17.

105. Ib., p. 632, 1. 11: . For the variants of the name  (probably the most correct form), , etc., see G. Moravcsik, Byzantinoturcica, ii, s.v.  in Aqsarayi, ed. O. Turan, passim (see index), is obviously the same name though not denoting the same person.

106. Ib., p. 633, 11. 2-16.

107. C. Jirecek, Staat und Gesellschaft im Mittelalterlichen Serbien, i, Vienna, 1912 (Denkschriften Ak, d. W. Wien, lvi), p. 78 seq.

108. Ib., p. 631, 1. 13: , shows that the bride was already in Pegai.

109. Ib., pp. 601, 1. 11—603, 1. 11:

110. G. Ostrogorsky, Geschichte des byzantinischen Staates, München, 1940, p. 354.

111. He appears in the sources occasionally as Ya'qub Eje Bey.

112.  etc. (see above, p. 641, n. 2), p. 19, n. **. Balaschev seems to have been afraid of his derivation since he chose to hide it in a note. In support of it he quotes toponyms of the Dobruja where an original k has changed into g, e.g. , which appears in a document of 1320 as  (for this he quotes, p. 15, Miklosich-Müller, Acta Patr., i, p. 95). Indeed, it is the ‘Gelaghra’ (Evliya, ii, p. 133) and even ‘Gülgrad’ (Hajji Khalfa; Hammer, Rumeli und Bosna, p. 27) of the Ottomans and the ‘Gelare’ (Mutafciev, op. cit., p. 37, n.) of to-day. As to the transition ay > ā (> a) instead of ay > ī, which would be the normal (Kowalski, Les Turcs et la langue turque de la Bulgarie du Nord-Est, p. 19) it is sufficiently explained by the influence of the back vowels which follow.

113. E.g. I had to abstain from dealing with the Gadjal of the Deli-Orman, the neighbours of the Gagauz to whom linguistically they are so closely related that they must be of the same origin; they are distinct from the Gagauz only in that they are Muslims — though Muslims of a very unorthodox kind.