ii. 78, says that the Turkish tribes remained in these countries after

the Mongol conquest. Is this the Organa or Urgheuz of Marco Polo?

[14] Hussein Hezarfenn, ii. 287, and Chalcocondylas (_Patr. Graec._, Migne, vol. clix), 21, call the father of Ertogrul Oguzalp. For critical discussion see Appendix A.

[15] This title is invariably given by Neshri to every ruler in the direct line of Osman, just as he calls the Christian opponents of the Osmanlis unbelievers.

[16] Probably Sultan Inoenu, anticipating the later name of this district.

[17] Sagredo, the Italian historian, whose work was greatly esteemed by Gibbon, makes the curious error of calling Alaeddin ‘Lord of Aleppo and Damascus’.

[18] ‘A great mountain situated between Kutayia and Brusa’: Hadji Khalfa, _Djihannuma_, fol. 1975; ‘The paths up this mountain are so difficult that one on foot has a thousand pains to reach the top’: ibid., fol. 1850.

[19] Rasmussen, _Annales Islamici_, p. 41, confuses this city with Kutayia, and gives its capture by Ertogrul under date of 1285.

[20] Thus in Ali and Neshri. Seadeddin attributes this dream to Ertogrul. But the confusion between Ertogrul and Osman is marked in all the Ottoman historians.

[21] The Ottoman historians give as reason for the refusal the social difference between his daughter and the ‘young prince’. This is an excellent illustration of how, writing in the zenith of Ottoman prosperity, the historians lost their sense of proportion or were actually compelled to write in flattering terms of the founder of their royal house.

[22] Hammer, i. 67, in relating this dream, has transcribed with fidelity and felicity the Persian poetry of Idris.

[23] Leunclavius, _Pandectes_, p. 113, following Ali, attributes the moon dream to Ertogrul, and places it at Konia. Boecler, _Commentarius de rebus turcicis_, pp. 104-5, following Chalcocondylas, does likewise, but relates the Koran dream of Osman. Seadeddin, p. 11, makes the dream distinctly religious, and while not mentioning the love story or Malkhatun by name, infers that Osman receives intimation of his marriage with Edebali’s daughter only through Edebali’s interpretation of the dream. This failure to mention Malkhatun is all the more significant when we see later how much attention Seadeddin gives to Nilufer. Evliya effendi, ii. 19, says that through the marriage of Osman to Malkhatun, the Ottoman sultans became descendants of the Prophet!

[24] I should except from this statement Rambaud, who, in _Hist. générale_, iii. 822-4, states that the conversion of the Osmanlis to Islam took place during the chieftainship of Osman. The general character of the work to which he was contributing, and the limits of space, did not allow him to give any reasons in support of this position. Vanell, _Histoire de l’Empire ottoman_, p. 357, says that Ertogrul was a pagan until he became converted through reading the Koran.

[25] From personal acquaintance with them, I can testify that these nomads (Yuruks) have remained up to the twentieth century with only the most vague idea of Mohammed and with no idea at all of the Koran and the ritual observances of Islam.

[26] See Shehabeddin, MS. Paris, Bibl. Nat., fonds arabe 2325, fol. 69 vº-70 rº, citing Mesoudi and earlier writers for the propagation of Islam among the Bulgarians.

[27] Cf. Cahun’s masterly contribution to _Hist. générale_, ii. 887.

[28] Abul Faradj, _Chronicon Syr._, pp. 606-8.

[29] The Ottoman historians mention none, either of friendship or enmity, during the entire life of Osman.

[30] The improbable connexion between Ertogrul and Osman and the Seljuk sovereigns of Konia has been accepted without question by European historians, on the strength of the assertions of the Ottoman historians. This is curious, because the evidence against this connexion is overwhelming. The Seljuk Empire of Rum lost its independence at the battle of Erzindjian, 1244 (cf. Heyd, _Histoire du commerce dans le Levant_, i. 534). Neshri himself confesses that after this date ‘now remained only the bare name of the Seljuk Kings’: _ZDMG._, xiii. 195. In view of the established facts of history, it is astonishing that European historians should have up to this time perpetuated, and given their sanction to, a fiction which was invented for the purpose of helping Mohammed II to incorporate Karamania in his empire! The limits of a footnote forbidding the adequate discussion of this question and the citation of the authorities, I must refer my readers to Appendix A.

[31] Neshri, _ZDMG._, xiii. 196, says seventy years. But in his reckoning he constantly contradicts himself. _Sheïr_ means city, _eski_ old, and _yeni_ new.

[32] All the Ottoman historians agree upon this number.

[33] ‘The unbelievers and believers of that land honoured Ertogrul and his son’: Neshri, p. 197. That Christians lived everywhere without molestation in the midst of non-converted Turkish tribes is asserted by Heyd, ii. 65.

[34] It is altogether likely that Osman received his name at the time of his conversion. Is it not significant that his father, his brothers, his son even, as well as most of his warriors, had purely pagan Turkish names?

[35] _Tableau de l’Empire ottoman_, iv. 373.

[36] See Appendix B.

[37] During the late war with the Balkan allies, the newspapers of the world spoke of ‘driving the Turks back to Asia, where they belong’, and of the re-establishment of the Ottoman capital at Brusa or Konia!

[38] See Armain’s translation of the _Djihannuma_ (Mirror of the World), a universal geography by Hadji Khalfa, in the Bibl. Nat., Paris, MS., fonds français, nouv. ac., nos. 888-9. The section on Asia Minor, although written in some detail, does not contain many of the names which we find in the Ottoman historians. I wish to register a protest against inflicting on students and readers of history lists of names that can have no possible meaning to them. I have omitted from this work the names of places and persons upon which I can get no light.

[39] Hadji Khalfa, op. cit., fol. 1917, makes an error in giving the distance from Brusa to Yeni Sheïr as two days. I have driven from Brusa to Nicaea in one day of not fast going. Yeni Sheïr is on the main road between these cities, six hours from Brusa and four hours from Nicaea.

[40] The early European historians make the wildest statements about Osman’s field of action. Many of them call Ottomanjik, a place four days or five north-east of Eski Sheïr, his first conquest: Cuspianus (Antwerp ed., 1541), p. 6; Spandugino, in Sansovino, p. 143; Egnatius, p. 28. Cf. Hadji Khalfa, op. cit., fol. 1789. But this place was not captured by the Osmanlis until the reign of Bayezid: Evliya, op. cit., ii. 95. Paulo Giovio, an Italian historian greatly esteemed in his day, puts among the notable conquests of Osman the city and district of Sivas, as does also Rabbi Joseph, in his famous _Chronicles_, Eng. trans. of Biallobotzky, ii. 505. Donado da Lezze, _Historia Turchesca_, Rumanian edition of Ursu, pp. 4 and 5, makes him conqueror of Rum, province of Sivas, Phoenicia, ‘et altri luoghi’! Cuspianus, _De Turcarum Origine_, quotes Donado da Lezze almost literally. Richer, _De Rebus Turcarum_, written for the information of Francis I of France, says, p. 11: ‘Circiter 1300, Ottomannus impune invitis omnibus _summam imperii_, quod ante partitum tenebant factiosi magistratus, _occupavit_, seseque Asiae minoris sive Anatoliae _imperatorem_ nominare sit aggressus. _Syvam_, quae eadem cum Sebaste est, _expugnavit_, et oppida ad Euxinum posita _non pauca_ cepit.’ (The italics are mine.) Hussein Hezarfenn, one of the Ottoman historians whose work has been most widely read and quoted in Europe, says of Ertogrul, _who never saw the sea_, ‘He equipped several ships, with which he made a raid into the Aegaean Sea, pillaged the islands, descended upon Greece, penetrated up to the Peloponnesus, and returned to his home (_the little village of Sugut!_) laden down with wealth and followed by a great army composed of experienced warriors of all sorts of nations whom the renown of his bravery and his good fortune attracted to his service: which increased so greatly his reputation in Asia that Sultan Alaeddin even found it to his advantage to cultivate him’: trans. of _Petits de la Croix_, ii. 288-9.

[41] I am not sure that I am justified in using the expression ‘undisputed sway’ even for this small territory. Pachymeres, IV. 30, pp. 345-7, speaks of a certain Soleiman pasha, who was threatening Nicomedia in 1303; and V. 23, p. 427, of Alisur retiring to the Sangarius after Roger had relieved Philadelphia in 1307.

[42] Probably the first conquest of Osman. This city, on the Kara Su, is still a thriving place. Its situation is most picturesque. The author of the _Arabic History of the Kurds_ (Bibl. Nat., Paris, MS. of Ducaurroy, fol. 151 rº, 152 rº) makes Biledjik the city granted to Ertogrul by Alaeddin, and declares that he captured Sugut (Sukidjeh) from the ‘infidels of Tekkur’.

[43] Angelcoma of the Byzantines.

[44] The only conquest of Osman not in the direction of Byzantium. Hadji Khalfa, op. cit., fol. 1851.

[45] ‘Situated between Yeni Sheïr, Brusa, and Aïnegoel. They count one day from Yeni Sheïr to Yar Hissar by the road which goes to Kutayia’: Hadji Khalfa, fol. 1917.

[46] The Ottomans name this place Kuyun Hissar. See Schéfer edition of Spandugino, p. 16 _n._

[47] Pachymeres, IV. 25, p. 327, says the battle was fought July 27. Jorga, _Geschichte des Osmanischen Reiches_, i. 157, is in error in placing date June 27; Hammer, i. 190, and Jorga both give year 1301. Muralt, _Chronographie Byzantine_, ii. 480, has this battle under 1302.

[48] Pach., IV. 25, p. 335.

[49] Cantemir, Rumanian ed., i. 20, seems to infer that Osman attacked Nicomedia after this battle. He is certainly wrong in stating that Osman captured Kutayia. See pp. 274, 292-3.

[50] Pach., V. 9; Gregoras, VII, i, p. 214.

[51] Pach., in Stritter, _Memoriae Populorum_, iii. 1086-7; D’Ohsson, _Histoire des Mongols_, iv. 315. Andronicus made a second appeal in 1308, and gave his own sister, Marie, who is known to later Mongol historians as ‘Despina Khatun’, to Mohammed Khodabendah Khan, after Khodabendah’s conversion to Islam: ibid., iv. 536; Hertzberg, _Geschichte der Byzantiner und des Osmanischen Reiches_, p. 461.

[52] I can find no justification for Howorth’s statement, ‘This alliance seems to have had a restraining influence upon the Turks’, in his _History of the Mongols_, iii. 464.

[53] See _Bibliothèque de l’École des Chartes_, vi. 318, where the date of this momentous event is given as ‘vers 1305’.

[54] Pach., V. 14, pp. 399-400; 21, pp. 410, 417.

[55] Pach., V. 23, pp. 426-8; Greg., VII. 3, p. 221.

[56] Pach., V. 21, p. 423; Greg., loc. cit.

[57] Greg., loc. cit. Cf. Muralt, after Latin authorities, ii. 487.

[58] Pachymeres, Books V, VI, and VII; Gregoras, Book VII, _passim_, and Phrantzes, Book I; Moncada, _Expedicion de los Catalanes_; Muntaner, in _Bibliothek des lit. Vereins zu Stuttgart_, vol. viii. For their later adventures there is an excellent account in Finlay, _History of Greece_, iv. 146-56.

[59] Andronicus wrote to his empress, urging her not to try to return to Constantinople from Salonika by land: Pach., VII. 12, p. 586; Chalcocondylas (ed. Bonn), I, p. 19.

[60] Greg. VII. 8, pp. 254-8; Chalc., I, p. 19; Jorga, op. cit., i. 160, speaks of ‘die schöne mit Perlen und Edelsteinen geschmückte _Krone_’ of Michael. Was it not rather a turban? See Hammer, i. 364, note x.

[61] ‘The emperor of Constantinople fears the anger of the Khan of Kapdjak and is eager to disarm him by protestations of submission and efforts to obtain a continuance of the truce. Things have always been on this footing since the children of Djenghiz Khan began to reign in this country’: Shebabeddin, Paris MS., fol. 70 rº.

[62] Ducange, _Hist. de Constantinople sous les Emp. Français_, map section, p. 46.

[63] Ducange, _Hist. de Constantinople sous les Emp. Français_, map section, p. 54.

[64] The Venetians were jealous of the growing power of Genoa and the hostility shown to Venetian merchants at Constantinople. See Appendix B. Also Heyd, _Handelsgeschichte des Mittelalters_, i. 366.

[65] Ducange, ibid., p. 57; Buchon, _Collection des chroniques nat. fr._, p. lv.

[66] Muralt, _Chronographie Byzantine_, ii. 493, no. 21, _n._

[67] A rabble without arms actually arrived at Marseilles. The ships were prevented from leaving Brindisi by a storm. Cf. Iacomo Bosio, _Della Historia della Religione_, ii. 1. At the very moment this effort to start a crusade was ending in dismal failure, the two kings on whose behalf it was planned were engaged in a bitter quarrel! Clement V, _Epistola Comm._ vii. 773-4, 787.

[68] Les Giustiniani, _Dynastes de Chios_, Vlasto’s French translation of Hopf’s great monograph, p. 8.

[69] Mas-Latrie, _Histoire de Chypre_, ii. 602.

[70] Mas-Latrie, op. et loc. cit.; Heyd, French edition, i. 537.

[71] A splendid field for historical research, which, as far as I know, has never yet been touched, is the compilation, from the Vatican records, of the dates for the extinction of the dioceses of the early Christian world in Africa and Asia. When did the bishops of these dioceses begin to be appointed and consecrated _in partibus_?

[72] Bosio, op. cit., ii. 37; Abbé Vertot, _Histoire des Chevaliers de Malte_, i. 106.

[73] See Bosio, ii. 37 f., and Vertot, i. 101 f. With a view to glorifying the Order, and also the Duke of Savoy, this fiction has been fabricated and perpetuated. Even such a serious work as that of Muralt gives, upon the strength of Raynaldus, who merely quotes Bosio, Osman as leader of this attack upon Rhodes: see _Chronographie Byzantine_, ii. 507. During the recent war between Italy and Turkey, when it was a question of Rhodes, more than one leading Italian newspaper revived this story of the founder of the Italian royal house defeating the founder of the Ottoman royal house. There is, of course, no foundation whatever for the statement.

[74] So Clement V evidently believed. See his letter to the Genoese in _Epistola Comm._ vii. 10.

[75] That the Sangarius used to run into the Gulf of Nicomedia instead of into the Black Sea is the opinion of many geographers, ancient as well as modern. There have been a number of projects to connect the Sangarius, Lake Sabandja, and the Gulf of Nicomedia by canals that would give a deep waterway across the plain and prevent the frequent overflooding which has always been a source of loss to cultivators in that region.

[76] Idris, quoted by Hammer, i. 192.

[77] Brusa is three hours by carriage from its port on the southern side of the Gulf of Mudania, or one hour by narrow-gauge railway. One can reach Nicaea either from the Gulf of Mudania or that of Nicomedia.

[78] Pach., VII. 18, pp. 597-9.

[79] Pach., VII. 25, p. 620. The Turks call this castle Hodjahissar.

[80] Ibid., loc. cit. But Pachymeres puts the number of these Tartars as 30,000, which must be at least a tenfold exaggeration.

[81] Seadeddin, translation Brattuti, p. 27. Bratutti, whose transcription of Turkish names is often unintelligible to me, calls Karadja Hissar ‘Codgia’.

[82] Ibn Batutah, _Voyages_, ii. 320, speaks of buildings which must have been erected at these baths by Orkhan within the decade following the capture of Brusa. Earlier buildings, according to him, were constructed ‘by a Turcoman king’: ibid., p. 318. Tchekirdje is still a favourite resort for foreigners as well as for natives.

[83] Cantacuzenos and Gregoras.

[84] Greg., IX. 2, p. 401.

[85] Cant., I. 42, pp. 204-6, 208; Greg., VIII. 15, p. 384; Greg., IX, c. 1, pp. 390-2, says it was the young Andronicus who first planned to break again with his grandfather. However that may be, the impression among the Greeks in Asia Minor who were endeavouring to hold back the enemies of the empire must have been the same!

[86] Greg., IX. 1, p. 392.

[87] In the volume on ‘L’Ancien Régime’ in Taine’s _Origines de la France contemporaine_, pp. 3-6, there is a wonderful analysis of the effect of early Latin Christianity upon the pagan mind. The Greek Church of the fourteenth century could produce no such impression.

[88] From the earliest Ottoman times to the present day religion and nationality have not been divorced. Osmanli and Moslem were synonymous terms, just as to-day in the Balkan peninsula, where the Ottoman Empire was really founded, Turk and Moslem are synonymous terms. When once this is understood, the student and traveller is freed from his preconceived notion that the ‘Turks’, as that expression is to-day understood in Turkey, are an Asiatic race, who have held the country as conquering invaders.

[89] Jorga, i. 162, is mistaken in saying, ‘überall wurden die Goldmünzen Osmans gern angenommen.’ Hadji Khalfa says that Osman struck no money. Also Colonel Djevad bey, _Histoire militaire de l’Empire ottoman_, i. 95. Save several silver pieces, which are not proven genuine, of the collection of Abbé Sestini (Salaberry, _Hist. de l’Emp. ott._, iv. 193), I can find record in numismatic collections of no money of Osman. For discussion of this question see Hammer, i. 117, who cites several Ottoman historians against coinage before Orkhan, and Toderini, _Historia della letteratura ottomana_, French trans., iii. 183.

[90] Appendix B, on the Emirates of Asia Minor during the Fourteenth Century, contains the identification and description of these neighbours.

[91] See Shehabeddin, Paris MS., 139 vº, which is cited in part on p. 70.

[92] The chieftainship among the Turks was elective rather than hereditary. The Armenian Haython, who had excellent opportunities for observing their customs at this period, wrote: ‘Puisque les Turcs pristrent la seigneurie de Turquie, ilz ordonnerent un seigneur entre eulx, lequel ilz appelerent le Soudan’: MS. Bibl. Nat., Paris, fonds français, 2810, fol. 230 vº. Hussein Hezarfenn says (ii. 287-9) that Ertogrul succeeded his father by election and, in turn, manœuvred to secure the election of Osman. Evliya effendi, i. 27, declares that Osman was elected chief. This is also stated by Barletius, in Lonicerus, vol. ii, fol. 231-2; Spandugino; Cantemir (Rumanian ed.), i. 19; and Vanell, p. 359. Cf. Chalcocondylas (ed. Migne), col. 24.

[93] For dates see Bibliography.

[94] Nöldeke’s translation, in _Zeitschrift der deutschen morgenlândischen Gesellschaft_, xiii. 214-17.

[95] Gregoras, IX. 1, pp. 390-2. But Cantacuzenos, I. 42, pp. 208-15, maintains that young Andronicus heard that his grandfather was preparing a _coup_ before he thought of taking any action himself.

[96] Cant., I. 44, pp. 215-16; Greg., IX. 1, p. 392; Phrantzes, I. 6, p. 35.

[97] Cant., I. 4-5, pp. 216-23; Greg., IX. 1, p. 396.

[98] Cant., I. 50, pp. 248, 252; Greg., IX. 3, pp. 405-7.

[99] Cant., ibid.; Greg., IX. 3, pp. 407-9.

[100] Cant., I. 52, pp. 260-2; Greg., IX. 4, pp. 409-10; Cant., I. 53, pp. 267-70.

[101] Cant., I. 55, pp. 277, 281-2; Greg., IX. 4, p. 414.

[102] Cant., I. 55-II. 1, pp. 277-312; Greg., IX. 4-8, pp. 411-32; Phr. I. 6, p. 35.

[103] IX. 8, p. 431.

[104] Cant., II. 28, p. 473; Greg., IX. 14, p. 461, and X. 1, p. 474.

[105] II. 3, p. 324.

[106] Cantacuzenos uses this same expression concerning the collecting of the army with which Andronicus III repelled an invasion of seventy Turkish vessels in the autumn of the same year. Cf. II. 13, p. 390.

[107] I have gathered the account of this battle from Cant., II. 6-8, pp. 341-60; Greg., IX. 9, pp. 433-5; Phr., I. 7, pp. 36-7; Chalcocondylas (ed. Migne), I. 11, col. 32. It is interesting to note how much space Cantacuzenos gives in contrast to the brevity of the other writers.

[108] II, c. 8, 363. Seadeddin, Neshri, and Idris agree with Gregoras, IX. 13, p. 458, in putting the fall of Nicaea in 1330 or 1331. Gregoras euphemistically says the city was ‘pillaged by the Turks’. But Leunclavius, on the authority of Ali, gives A.H. 734, which would be 1333 or 1334.

[109] Phr., I. 7.

[110] In _Djihannuma_, Paris MS., fol. 1934.

[111] When I was in Nicaea in 1913, the imam of the Yeshil Djami told me that there were seventy thousand houses at the time of the Ottoman conquest. This is the local tradition.

[112] Hammer, i. 146, makes this claim.

[113] Ibn Batutah, ii. 322-3. For discussion of the value of Ibn Batutah’s testimony see Appendix B and Bibliography.

[114] Miklositch-Müller, Act. LXXXII, anno 1339, and Act. XCII, anno 1340.

[115] There is no way of establishing the date of the fall of Nicomedia. The Ottoman historians report that it was added to the dominions of Orkhan in 1326, the year of his accession and of the fall of Brusa. It is best here to follow the unanimous testimony of the Byzantine sources, which is in accord with the natural inference that Nicomedia fell some time after Nicaea: Greg., XI. 6, p. 545; Phr., I. 8, p. 38. Hammer cannot disregard the testimony of Gregoras here. He ingenuously suggests that the city might have been lost by the Osmanlis, and recaptured. Cantacuzenos (II. 24, p. 446, and 26, p. 459) says that Andronicus III went twice to the aid of Nicomedia in 1331, but he does not record the loss of either Brusa or Nicomedia. In the collection of Feridun, Bibl. Nat., Paris, MS. anc. fonds turc 79, there is a diploma appointing Soleiman governor of Nicomedia in 1332, but the authenticity of the earlier pieces in this collection is open to grave suspicion (cf. Bibliography).

[116] Howorth, iii. 613.

[117] Canale, i. 215.

[118] Not an actual defensive alliance against Orkhan, as Schlumberger, _Numismatique de l’Orient latin_, p. 480, supposes. See Cant., II. 13, pp. 388-90; Phr., I. 8, p. 37.

[119] Cant., II. 28, pp. 470-3.

[120] Ibid., 22, p. 435.

[121] Ibid., 25, pp. 455-6.

[122] Cant. II., 29-30, pp. 480-4; Greg., XI. 2, p. 530.

[123] Hammer, quoting Ashikpashazadé, i. 150-1.

[124] Mordtmann, in _ZDMG_. (1911), lxv. 105, basing his statement, like Hammer, on Ashikpashazadé, Vatican MS., fol. 33, gives A.H. 735, 737, or 740. The earliest of these dates is precluded by the testimony of Ibn Batutah, who found these places still independent about A.H. 735. A.H. 737 might be possible, if we decide that Orkhan accomplished everything during the one expedition against Pergama. Mordtmann, still quoting Ashikpashazadé, says that these three cities were held by relatives of the Palaeologi. If this be true, it goes to prove that there must have existed all along in the reigns of Osman and Orkhan quasi-friendly relations between Moslem and Christian. There was certainly no religious fanaticism during this period of Ottoman history.

[125] ‘Les Osmanlis avaient étendu leur domination en Asie Mineure et absorbé les états dont l’indépendance avait jusqu’alors empêché l’unité politique de l’Empire musulman!’ Delaville-Leroulx, _France en Orient au XIVe siècle_, i. 118. ‘Osmans Sohn Orkhan Kleinasien unterworfen hatte’: Wüstenfeld, _Geschichte der Türken_, p. 16. ‘Orkan s’impadroni di quasi tutta la Natolia’: Alberi, in preface (viii) to series III, vol. i, of _Relazione Ven. Amb._ One of the earliest western historians gives Orkhan’s ambition as ‘solus cupiens in minore Asia regnare’: Cervarius, p. 5. Even Hammer, i. 150, is considerably ahead of time in saying, in one of his chapters on Orkhan, ‘Les hordes ottomanes se précipitèrent du haut de l’Olympe comme une avalanche, franchissant montagnes et vallées, ajoutant à leurs possessions les neuf royaumes nés des débris de l’Empire seljukide, inondant Asie Mineure depuis l’Olympe jusqu’au Taurus.’ Hammer does not mean to give this wrong impression, but one has to read very closely not to get it. See discussion of this error in Appendix B.

[126] Cant., IV. 37, p. 284. Is it on the strength of this evident error of a Greek writer that Evliya effendi, ii. 229, says ‘Orkhan captured Angora from the Prince of Kutayia of the Kermian family’? Hussein Hezarfenn, following Chalcocondylas, is an example of an Ottoman historian basing his statements on a Greek authority.

[127] For the time of Ibn Batutah and Shehabeddin see Appendix B, p. 279. Mas-Latrie, _Trésor de Chronologie_, col. 1796, after careful collation of Shehabeddin and Ibn Batutah, comes to the conclusion that Orkhan added the emirates of Balikesri, Marmara, Akbara, Kaouïa, Keredek, Kul Hissar, and Thingizlu to his state between 1349 and 1360. This, too, is discussed in Appendix B.

[128] Marmara, for example, is given by the Ottoman historians as a conquest made by Osman. See Hammer, i. 89. But it is mentioned as an independent principality by Shehabeddin, in _Notices et Extraits des MSS. de la Bibl. du roi_, xiii. 358, 366.

[129] Ibn Batutah, ii. 321-2.

[130] Shehabeddin, Paris MS., fonds arabe 2325, fol. 139 vº-140 rº.

[131] Ibid., fol. 125 vº.

[132] Hammer, i. 110-11, says that Alaeddin, ‘stranger to the profession of arms, occupied himself solely with the cares of state’, but on p. 133 he has Alaeddin commanding the troops in battle while Orkhan watches from the top of a hill!

[133] For the derivation of vizier, with the double meaning of burden-bearer and the one who aids, see Ibn Khaldun, _Prolegomena_, in _Notices et Extraits_, xx. 4.

[134] Gen. xxviii. 11-18.

[135] Sale’s translation, c. 20, verse 30, p. 234.

[136] Col. Djevad bey, p. 20, _n._ 2. Col. Djevad claims that von Hammer’s derivation of the word ‘pasha’ from the Persian is wrong. But he gives no reason which would satisfy the philologist when he asserts that this word is essentially Turkish. Nor does he attempt to explain its original meaning. ‘Pasha’ is probably a shortened form of ‘padishah’. See _Century Dictionary_, v. 4228.

[137] According to the biographer of Brusa, cited by Hammer, i. 146, _n._ 4.

[138] I do not understand what Hammer means when he says, i. 116, that the _Kanunnamé_ must be taken in the sense of political rather than ecclesiastical law. The two cannot be separated in Islam. Or, perhaps, it is better to say that there is no political law. The very word _Kanun_ was taken from the Greeks, was used by them for ecclesiastical law, and its adoption by the Osmanlis (at a much later period than Orkhan) serves to emphasize the fact that there was no other land of law conceivable than the law of the Church. The word _Kanun_ had of course other meanings, but in its collective legal sense it seems to have stood only for rules or laws that had to do with things ecclesiastical or religious. See the various meanings of this word in A. Souter’s _Text and Canon of the New Testament_ (London, 1913), pp. 154-5.

[139] This petition is in the Litany of the Prayer Book of Edward VI. Cf. Schaff, _Church History_, iv. 151.

[140] I do not mean to assert that religious feeling has played no part in the massacres of our own day. But these massacres were arranged by the government, who incited the Moslems to attack their Christian neighbours, inflaming the ignorant mind more by an appeal to racial hatred, to loot, to lust, than to defence of the sacred faith. In the Armenian massacres it was represented to the ignorant village Moslem that the Armenians were plotting to set up an independent government or to betray the fatherland to some European power. I was in Adana during the terrible massacre of 1909, and make this statement from personal experience and observation.

[141] Michail Koëzé, Marco, and Evrenos were Greeks. Cf. Leunclavius, _Pandectes_, p. 125.

[142] Up to the time of the Tanzimat, in 1849, Christians were called _raïas_. The original meaning of _raïa_ was a flock, and was not a term of contempt, but a recognition of the fact that Christians were a taxable asset to the nation, at so much per head.

[143] In western Asia Minor, in Macedonia and Thrace, up to the present day the convert to Islam, no matter of what race, is immediately classified before the law as a Turk. When the Sublime Porte, after the reoccupation of Adrianople in the summer of 1913, laid a memorial before the Powers, it was claimed that the large majority of the population of the vilayet of Thrace was ‘Turkish’. This word had absolutely no racial significance. Every Mohammedan in Thrace, no matter what his race or language, would be considered a Turk. The Young Turks, when they established the Constitution in 1908, tried to revive the word ‘Osmanli’ as a term including all Ottoman subjects. But they not only failed to convince the nation--they failed to convince themselves--that a Christian could really be an Osmanli, with the full rights and privileges enjoyed by the Moslems.

[144] Ricaut, ed. 1682, p. 148. For confusion of the name ‘Turk’ with ‘Saracen’ by early western writers, see _Chronique latine de Guillaume de Nangis_, Géraud ed., i. 46, 86-8; _Mémoires d’Olivier de la Marche_, Beaune and d’Arbaumont ed., i. 22-5, iv. 83; _Gilles le Muisit_, Lemaître ed., p. 196. The mistake of Ricaut is common with many of the fifteenth-to seventeenth-century writers on the Crusades.

[145] Matthew of Edessa (Urfa), fol. 8 of MS. Bibl. Nat., Paris, fonds arménien, No. 95, quoted in _Notices et Extraits_, ix, 1^[ère] partie, p. 281, speaks of ‘les calamités que des peuples barbares et corrompus, tels que les Turcs et les Grecs, LEURS SEMBLABLES, ont causées’.

[146] This was true even of the conquest of Constantinople, which caused much more dismay and regret in Europe than among the Greeks. See the remarkable letter of Francis Fielphus to Mohammed II in _Bibl. de l’École des langues vivantes orientales_, série 3, xii. 63-6, 211-14.

[147] Cf. Rambaud in _Hist. Générale_, ii. 816.

[148] In Constantinople, Smyrna, Salonika, and the lesser coast cities of the Ottoman Empire, as well as in many of the cities of the interior, one feels the atmosphere of Sabbath rest much more on a Sunday than on a Friday.

[149] Evliya effendi, ii. 241.

[150] In the _Djihannuma_, p. 951.

[151] In a popular Anatolian love-song, there is the line, ‘Benim sevdijimie din var iman yok’, ‘She whom I love has religion, but not a bit of faith’, which illustrates the lack of deep religious feeling in the Osmanli. In this he is like the Greek, and different from the Slav, the Persian and Arab. See Kúnos, _Zeitschrift der deutschen morgenländischen Gesellschaft_, liii. 237.

[152] At Balikesri the sultan Dambur told Ibn Batutah that ‘the men follow the religion of their king’: ii. 317. Here was the principle of _cuius regio eius religio_ two centuries before Augsburg!

[153] Col. Djevad bey, pp. 18-19.

[154] Edward III of England had created a sort of obligatory military service. His organized infantry took part in the Battle of Crécy, 1346. Lavisse-Rambaud, _Hist. générale_, iii. 76.

[155] Halil Ganem, i. 39.

[156] This still holds. In October 1912, on the Seraskerat Square in Constantinople, I saw Sultan Mehmed V give over the command of the army for the Balkan War to Nazim pasha.

[157] Col. Djevad bey, p. 18.

[158] Bertrandon de la Broquière, Schéfer ed., pp. 220-1.

[159] This statement needs especial emphasis, as many historians have followed Chalcocondylas and Bosio in attributing the corsair fleets to Osman and Orkhan. An instance of a careful modern historian making this error is found in Romanin, _Historia documentata di Venezia_, iii. 147, where he says, ‘La lega ... per raffrenare l’ognor erescente potenza _ottomana_.’

[160] In Bongars, _Gesta Dei per Francos_, ii. 313.

[161] This letter, from the manuscript in the Bibliothèque Nationale, Paris, is published in _Bibl. de l’École des Chartes_ (1906), lxvii. 587. Other documents on this mission, ibid. (1892), liii. 254-7.

[162] See papers of H. Lot in _Bibl. de l’École des Chartes_, 4e série (1859), v. 503-9, and (1875) xxxvi. 588-600. Also Bosio, ii. 58.

[163] Raynaldus, Ann. 1334, pp. 17-19. As the repetition of all the negotiations in connexion with papal attempts for crusades cannot be included in the text of my book, I refer the reader to the section on papal negotiations in the Chronological Tables.

[164] Deliberation of Senate, November 18, 1333, in _Misti_, XVI, fol. 40.

[165] Raynaldus, Ann. 1344, p. 11; Stella (in Muratori), col. 1080; Dandolo, p. 418; Greg., II, p. 686; Cant., III, p. 192; _Mon. Hist. Patr._ x. 757; _Misti_ for 1344, fol. 30; Rymer, _Acta Publica_, vol. ii, part IV, p. 172; _Commemorialia_, iv. 80.

[166] For relations of Rhodes with Smyrna from 1347 onwards, see Bosio, _passim_, but especially ii. 80 and 118-19.

[167] Serbian chronicles, quoted by von Kállay, _Geschichte der Serben_, i. 66.

[168] In the fratricidal war of July 1913, the ignorant Serbian peasants really believed that they were fighting to take from the Bulgarians ‘the sacred soil of the fatherland’, as their newspapers and addresses to the soldiers called Macedonia. The name of St. Stephen was invoked when they went into battle.

[169] Orbini, _Il Regno degli Slavi_, p. 259, gives a circumstantial account of the assassination. He says that Stephen gave the order to men who strangled the old king in his cell at midnight. This does not prevent Orbini from saying later of Stephen ‘fu huomo molto pio’! Borschgrave, p. 266, is not certain of Stephen’s connivance.

[170] J. Schafarik, _Elenchus actorum spectantium ad historiam Serborum_, XXV-XXVII.

[171] I find no documentary authority for the often repeated statement that this coronation took place at Skoplje (Uskub or Scopia). At the time of the recent Balkan War, the Serbians, in order to preserve their friendly relations with Greece, supported the Uskub theory. But see Ljubić, _Monumenta spectantia ad hist. Slavorum meridionalium_, ii. 278, 279, 326; _Commemorialia_, IV; _Secreta Rog._, A. 33.

[172] ‘Stephanus, D. G. Serviae ... Albaniae, maritimae regionis rex, Bulgariae imperii princeps et fere totius imperii Romaniae dominus’: Ljubić, ii. 278.

[173] Ibid., ii. 326.

[174] Ibid., loc. cit.

[175] _Secr. Rog._, A. 33.

[176] _Misti_, xxiv. 12.

[177] Ibid., xxiv. 110.

[178] _Secr. Rog._, II, B. 4; _Misti_, xxiv. 103.

[179] Cf. _Misti_, xxv. 7, 10. Fiorinsky, _The South Slavs and Byzantium in the second quarter of the Fourteenth Century_, quoted by Borchgrave in _Bulletin de l’Académie royale de Belgique_ for 1884, 8e série, iv. 429-30.

[180] _Commem._ iv. 172.

[181] _Misti_, xxvi. 16-22; _Commem._ iv. 157.

[182] MS. Vatican 3765, quoted by Raynaldus, ann. 1347, XXX.

[183] Fiorinsky, p. 207.

[184] Engel, _Geschichte von Serbien_, 285-6; Müller, _Beiträge Byz. Chron._, p. 406 _n._

[185] Cant., IV. 43, p. 315; Greg., XXVII. 50, p. 557; von Kállay, i. 69.

[186] Cant., II. 9, pp. 363-70; Greg., XII. 3, p. 582; Ducas, p. 6.

[187] Cant., II. 1, pp. 14-18; 40, p. 560; and III. 4, p. 91; Greg., IX. 11, pp. 560-8; XII. 2, p. 576.

[188] Cant., II. 24-7, pp. 145-67; Greg., XII. 11-16, pp. 608-26; Phr. I. 9, p. 40; Ducas, 6, p. 24, to 7, p. 26.

[189] Cantacuzenos tries to make out that this was a justifiable arrangement, as this district had already been conquered by Stephen Dushan. But Ducas, 6, p. 26, and 8, p. 30, declares that Cantacuzenos sacrificed the empire to the Serbians.

[190] Cant., III. 57, pp. 347-8; Greg., XIII. 4, pp. 648-52.

[191] _Misti_, xxi. 35.

[192] Greg., XVI. 6, pp. 834-5; Ducas, 7, p. 29; Clement VI, _Epp. Secr._ vii. 99. ’Άμυρ is either ‘Emir’ or ‘Omar’.

[193] Cant., III. 31, p. 498; Ducas 9, pp. 33-4; Chalc., I, p. 24.

[194] Cant., III. 81, pp. 501-2; 84, pp. 518-19; 85, pp. 525-9.

[195] Cant., III. 95, pp. 585-9; Greg., XV. 5, pp. 762-3; Ducas, 9, p. 35.

[196] Greg., XV. 2, p. 749.

[197] For the action against Barlaam spoken of here, see Muralt, ii. 575, No. 17; p. 576, No. 22; p. 578, No. 37.

[198] Cant., III. 98, p. 604, to IV. 4, p. 29; Greg., XV. 9, p. 781, to 11, p. 791; Ducas, 9, p. 37, to 10, p. 38.

[199] Cant., IV. 1, p. 12, to 2, p. 19.

[200] Cant., IV. 4, p. 30; 5, p. 32; 20, p. 147.

[201] Cant., IV. 9, pp. 53-7.

[202] Raynaldus, ann. 1349, XXXI.

[203] Clement VI, _Epp. Secr._ viii. 248-50.

[204] Cant., IV. 13, p. 85.

[205] Marco Guazzo, _Cronica_, p. 269; Stella, _Annales Genuenses_, in Muratori, xvii, col. 1090.

[206] MS. Vatican 2040, cited by Muralt, ii. 618: Petrarch, _Epp. fam._ vii. 7. For historical and medical importance of the black death, see Hecker, _Der schwarze Tod im 14ten Jahrhundert_ (Berlin, 1832). MSS. Bibl. Nat., Paris, fonds latin 8369-70, contain an interesting contemporary account, mostly in hexameter verse, by Symon de Cavino, a Paris physician.

[207] _Breve Chronicon_ at end of Ducas, cited by Finlay, _History of Greece_, iv. 409 _n._

[208] In 1340 Venice had refused a loan of ships and money to Edward III of England on the ground that she needed all her resources ‘to guard against the Turkish danger about to become universal’: Wiel, p. 204.

[209] On March 17, 1351, Petrarch addressed from Padua to Doge Andrea Dandolo a letter of remonstrance and warning against engaging in a war with Genoa. This letter is quoted in Hazlitt, iii. 122.

[210] The Genoese archives contain a treaty between the Byzantine Empire and Genoa, dated May 6, 1352, which says: ‘debbono eziandio ritenersi per valide e ferme le convenzioni e la pace stipulata dai genovesi con Orcan bey.’ Belgrano, _Atti della Società Ligure di Storia Patria_, xiii. 124.

[211] The Signory of Genoa, writing to the Podesta of Pera, March 21, 1356, said: ‘Nobis, vobis ac omnibus ianuenibus est notorium et manifestum quantum bonum et gratias habuimus a domino Orchano amirato Turchie ad destructionem et mortem tam venetorum quam grecorum tempore guerre nostre’: ibid., p. 127.

[212] In the treaty of 1387 with Murad, the Genoese said: ‘quam inter recolendam memoriam magnifici domini Orchani patris sui ex una parte et illustrem Commune Ianue ex altera’: ibid., p. 147.

[213] Cant., IV. 11, pp. 68-77; Greg., XVI. 6, p. 835, to XVII. 7, p. 865.

[214] Cant., IV. 16-17, pp. 104-5, 108-11, 114-30; 19, pp. 133-5; 22, p. 156; Greg., XVI. 1, p. 795; XVIII. 2, p. 876. Phr., I. 9, p. 40, gives this as the time Cantacuzenos married his daughter to Orkhan.

[215] Cant., IV. 30, pp. 218-20; Greg., XXVI. 19, p. 86, and 22, p. 88. For explanation of action of Venetian admiral, Pisani, see histories of Daru and Romanin.

[216] Villani, _Historia Venetiana_ (Muratori), xiv. 200; Canale, _Nuova istoria di Genova_, i. 222.

[217] Cant., IV. 33, pp. 246-7; 36, p. 266. Cantacuzenos had tried to get the Bulgarians to attack Stephen Dushan in 1351. Cf. Cant., IV. 22, pp. 162-6.

[218] Greg., XXVII. 30, pp. 150-1.

[219] Cant., IV. 36, pp. 265-6; Greg., XXVII. 55, p. 171, and XXVIII. 3, pp. 177-8; Cant., IV. 34, pp. 247-50; Greg., XXVIII. 7, pp. 181-2.

[220] Cant., IV. 34, pp. 250-3; 36, p. 266; Greg., XXVIII. 19, p. 188.

[221] About two hours on horse from Gallipoli.

[222] Seadeddin, i. 58-63.

[223] Gilbert Cousin, _Opera_, i. 390 (evidently copying Drechsler), and Egnatius, _de Origine Turcarum_ (Paris, 1539), p. 29, give date A.D. 1363. But do they not follow Phr., I. 26, p. 80?

[224] Donado de Lezze, p. 7, and Paolo Giovio, both ardent Venetians, and Rabbi Joseph, i. 245, give the names of these vessels, though differently. Nicolas de Nicolay, who passed through the Hellespont in 1551, says that this story of the Genoese was a tradition of the locality. He locates the castle of Tzympe a few miles from the Aegaean end of the strait! _Les quatre livres des navigations_ (1587 ed.), p. 58. Sauli, _Della Colonia Genovese in Galata_, ii. 44-5, vigorously defends the Genoese against this calumny.

[225] There is no room for doubt about this date. Cf. Cant., IV. 38, pp. 277-80; Greg., XXXIII. 67, p. 220, and XXVIII. 40-2, pp. 202-4; Villani, p. 105; _Byz. Annalen_, ed. Müller, in _Sitzungs-Berichte der Wiener Akademie_, ix. 392; Muralt, _Chronographie Byz._, ii. 643.

[226] This place figured in the recent Balkan War. It was here that the Osmanlis stationed their army for the defence of the Dardanelles.

[227] Greg., XXIX. 26, p. 241.

[228] Greg., XXVIII. 30, pp. 195-201.

[229] Cant., IV. 37, pp. 270-2; 38, p. 276; Greg., XXIX. 17-18, pp. 234-6; 49, p. 257.

[230] At least, Cantacuzenos, IV. 38, p. 276, claims that he ransomed Tzympe.

[231] Cant., IV. 38, p. 283.

[232] Rumanian Chronicle, cited by Gregorović, _Relations of Serbia with her Neighbouring States, principally in the Fourteenth and Fifteenth Centuries_, Kazan, 1859, in an appendix.

[233] Cant., IV. 39-43, pp. 284-307; Greg., XXIX. 27-30, pp. 242-3.

[234] Cant., IV. 49, pp. 358-60.

[235] Tchorlu was the head-quarters of the Ottoman General Staff during the first month of the Balkan War. After the battle of Lulé Burgas, it became the head-quarters of the Bulgarians. From here the attack upon the defences of Constantinople was directed.

[236] Muralt, ii. 640, No. 10, _n._

[237] Greg., XXIX. 34, pp. 224-6.

[238] During the five years following the proclamation of the Constitution in 1908, I lived, and travelled extensively, in the Ottoman Empire. Rarely did I meet a foreigner engaged in business there who had the slightest sympathy with the Osmanlis in their aspirations or in their successive crushing misfortunes. This is not a criticism, but merely the record of a fact.

[239] Schafarik, CVII.

[240] The expression ‘la terre que les Turcs tiennent’ is always used to designate Asia Minor in the opinion which the council of the French King Philippe de Valois gave concerning the route to be followed in the abortive crusade of 1332. See Archives Nationales, Paris, P. 2289, pp. 711-12.

[241] See p. 97, and notes 3 and 4 on that page.

[242] Quoted from the Cancelleria Secreta by Romanin, iv. 232.

[243] This letter is reproduced by Jireček, _Geschichte der Bulgaren_, p. 309.

[244] Greg., XXXVII. 52, p. 558; 59-63, pp. 561-3; 67-9, pp. 565-6; XXXVI. 6-8, pp. 504-9; Cant., IV. 44, p. 320.

[245] The generally accepted date of Orkhan’s death is 1359 or 1360, following Ottoman sources. But Jireček, a careful and able scholar, p. 321, n. 10, is inclined to accept March 1362. There is great confusion about this period. I think that the Ottoman date is undoubtedly correct here.

[246] ‘Der eigentliche Begründer der osmanischen Macht war Orchan’, Fessler, _Geschichte von Ungarn_, ii. 151.

[247] Col. Djevad bey, p. 254.

[248] Seadeddin, i. 80.

[249] Seadeddin, i. 82; Hadji Khalfa, _Rumeli_, p. 19.

[250] But Matteo Villani, in Muratori, xiv. 672, who is followed by Leunclavius, says that Demotika was abandoned to Orkhan in November 1361.

[251] Cf. marginal note in Barberini MS. of Pachymeres, cited by Muralt, ii. 663, No. 9.

[252] Seadeddin, i. 84-5; Hadji Khalfa, _Rumeli_, p. 22.

[253] All the Ottoman historians.

[254] MS. Bibl. Nat., Paris, fonds turc, No. 79, p. 25; Leunclavius, _Annales_, p. 30; Seadeddin, i. 85.

[255] Muklis Abderrahman Efendy, quoted by Schéfer, in his edition of Bertrandon de la Broquière, p. 170, _n._ 3.

[256] Seadeddin, i. 89; Hadji Khalfa, _Rumeli_, p. 52.

[257] Villani tells of its terrible ravages in 1360 ‘ricominciata in diversi paesi del mondo’, Muratori, xiv. 653, 688-90, 727.

[258] Ibid., pp. 649-50. He declares that Murad had been ‘molte volte tentato di vincere Constantinopoli’.

[259] Cf. Finlay, iv. 45, 169.

[260] Seadeddin, i. 42. Hammer, i. 384-5, n. viii, says that Ottoman historians are unanimous in this assertion as against Byzantine sources. Col. Djevad bey, the modern Ottoman authority on military history, is disappointing and unconvincing in his discussion of this question. On p. 25 he gives 726 (1326) for the date, and on p. 78 730 (1329). He cites no sources, for there are none, and has to admit, p. 54, that Murad I made the laws for the janissaries. Among early European historians there is much divergency. Spandugino, p. 185, attributes their origin to Osman, and the name from the village of Sar: they are ‘the young men of Sar’. Ricaut, ed. 1682, p. 357, also attributes to Osman. Reineccius, influencing the Latin editor of Chalcocondylas (see ed. Migne, p. 26, _n._ 11), makes Osman the founder, and derives the name from ‘Januae’: they are the _custodes corporis_. Leuncl., _Pandectes_, p. 129, discusses these theories without coming to any conclusion. Giovio, Geuffraeus, and Nicolay, p. 83, attribute origin to Murad II. Certainly it was not earlier than his day that the janissaries attracted attention in Europe. D’Ohsson, vii. 311, asserts that there was no definite organization until Mohammed II. Mignot, i. 119-20, is in favour of the theory that Murad I created this corps.

[261] Seignobos, in _Hist. générale_, ii. 334.

[262] Col. Djevad bey, p. 251, says that Anatolian Christians were exempt to give time to recuperate ‘after the exhausting struggles of generations’. But exhausting struggles had been no less frequent and no less severe in the Balkan peninsula. Gibbon’s suggestion, that the levies were made in Europe because Moslem and Christian Anatolians were not apt for war, shows how completely the great English historian missed the _raison d’être_ of the janissaries.

[263] Hammer, i. 126.

[264] Col. Djevad bey, p. 90.

[265] Ibid.

[266] Ibid., pp. 55-6; Ducas, p. 16; Leunel., _Annales_, p. 34; Ricaut, pp. 358-9.

[267] Lavallée, i. 190-1.

[268] Phr., I. 26, p. 86; Chalc., I, p. 25. Cf. Michaud, _Hist. des Croisades_, v. 275.

[269] Seadeddin, i. 91.

[270] This colony was at Bigha. See Appendix B, p. 301.

[271] Phr., I. 26, p. 80.

[272] Katona, x. 393.

[273] Chale., I, p. 30, and the chronicle of Rabbi Joseph, i. 240, confuse this battle with that of Cernomen, near the same place and with the same result, in 1370. But there were certainly two distinct battles. Louis of Hungary took part in the first, as is shown by the date recorded at Mariazell and by a diploma in Fejér, _Cod. Dipl. Hung._, 9e partie, vii. 212. Cf. Aschbach, _Geschichte Kaiser Sigmunds_, I. 87. The account in Vambéry’s _Hungary_, Story of Nations Series, p. 171, is wholly wrong.

[274] Seadeddin, i. 94.

[275] Miltitz, ii. I^{ère} partie, 166.

[276] Col. Djevad bey, p. 97, _n._ 1; Engel, _Geschichte Rag._, p. 141; Hammer, i. 231, 405. But this was also Timur’s ordinary method of signing ordinances: cf. Shereffeddin, iv. 55. The document, with the marks of Murad’s hand, is preserved in the museum of the Communal Palace at Ragusa.

[277] Villani, x. 30.

[278] Cf. Hazlitt, iii. 216.

[279] Urban V, _Epp. secr._ iv. 114.

[280] ‘Il le print por prisonnyer, et le destint a cause de ce que le roy de Bourgarye sy sestoit accorde et alyez secrettement avecques le turc’: _Chronicques de Savoye_, col. 300.

[281] Cf. Jireček, _Geschichte der Bulgaren_, p. 325.

[282] Cibrario, _Storia di Savoya_, iii. 193. But I have followed closely the account of the expedition as given in the anonymous French chronicle, cols. 299-319, in _Monumenta Historiae Patriae_, Turin, 1840, vol. i. There is a modern book by Datta. Cf. also Delaville le Roulx, i. 148 f.

[283] Urban V, _Epp. secr._ iv. 124.

[284] Ibid., iv. 240.

[285] Greg., XXV. 17, p. 41.

[286] Urban V, _Epp. secr._ ii. 230; Petrarch, _Senilia_, iv. 2.

[287] ‘Nescio enim an peius sit amisisse Hierusalem an ita Bizantion possidere. Ibi enim non agnoscitur Christus, hic neglegitur dum sic colitur. _Illi_ (Turcae) _hostes, hi scismatici peiores hostibus_: illi aperte nostrum Imperium detractant: hi verbo Romanam ecclesiam matrem dicunt: cui quam devoti filii sint, quam humiliter Romani pontificis iussa suscipiant, tuus a te ille datus patriarcha testabitur. _Illi minus nos oderunt quam minus metuunt. Isti autem totis nos visceribus et metuunt et oderunt._’ _Senilia_, vol. vii.

[288] In the Church of the Holy Sepulchre, Jerusalem, and in the Church of the Holy Nativity, Bethlehem, anarchy--even bloodshed--is prevented only by the constant vigilance of the Ottoman military authorities. If one asks the Latin and Greek priests in Jerusalem, they will admit without shame that this statement is true.

[289] Miklositch-Müller, _Acta et diplomata graeca_, CLXXXIV.

[290] _Epistolae secretae_, vi. 1-10.

[291] Ibid., vi. 3.

[292] The date of this visit is certain from the formal act of abjuration, which is given in full in Raynaldus, ann. 1369, XI. Ducas, c. 11, and Chalc., I, p. 25, are in error in placing this voyage later. Berger de Xivrey, _Mém. de l’Acad. des Inscriptions_, xix, 2e partie, p. 35, suggests that the Byzantine historians have confused this voyage with that of Manuel, thirty years later.

[293] _Epp. secr._, viii. 37, 38, 80.

[294] By an encyclical: _Epp. secr._, viii. 4. Cf. also his letters to the doges of Venice and Genoa, ibid., p. 24.

[295] Ibid., viii. 55.

[296] Phr., I. 22, pp. 52-3; Chalc., I, pp. 50-1; Morosini, p. 13.

[297] Phr., I. 11, p. 46.

[298] Gregory XI, _Epp. secr._ iii. 36, 58.

[299] Chalc., I, pp. 51-2.

[300] Raynaldus, ann. 1371, VIII.

[301] _Epp. secr._, ii. 32, 87. Similar letter to Louis in December 1375, ibid. v. 46. Other letters reprinted in Fejér, 9e partie, iv. 583 4; v. 54-6; vi. 155-6.

[302] Bernino, pp. 15-20.

[303] Fejér, 9e partie, iv. 427-8.

[304] Ibid., v. 52-3.

[305] Rymer, _Acta Publica_, III, part 3, pp. 38-40.

[306] On December 12, 1374, Gregory XI wrote to John from Avignon, predicting that his ‘alliance with Murad’ would bring about the destruction of the empire: _Epistolae secretae_, iv. 68.

[307] Raynaldus, ann. 1378, XIX.

[308] Jireček, _Geschichte der Bulgaren_, p. 317.

[309] Cant. IV, 50, pp. 362-3.

[310] Although Engel says 1353, others 1356, and the Rumanian chronicle 1371, there can be no question that 1365 is the correct date; for both Byzantine and Ottoman historians speak of Alexander as Bulgarian Czar in 1364, and do not mention him later, while Sisman and his brothers come immediately into prominence.

[311] Schiltberger, Neumann ed., p. 93.

[312] Orbini, pp. 472-3.

[313] Bonfinius, II. 10.

[314] Fessler, _Geschichte von Ungarn_, ii. 152.

[315] Wadding, _Annales minorum_, ann. 1369, XI.

[316] _Epp. secr._, VI. 131, 136.

[317] Called Ishebol by the Ottoman historians.

[318] By the second division of the Ottoman army under Timurtash. Murad himself had captured Sozopolis. Cf. Jireček, p. 326.

[319] Seadeddin, i. 104. He does not give the name of the Serbian kral.

[320] The peasantry around Samakov will point out to you the ridge, south-east of the modern town, over which he vanished. They believe that Sisman haunts the foothills of the Rhodope mountains, and rides headless in the night down into the plain. This tradition, and the statement of Ducange, viii. 289, that Sisman died in 1373 in Naples, makes possible the theory that there were three successive Sismans connected with the Ottoman conquest of Bulgaria.

[321] Hadji Khalfa, _Rumeli_, p. 38.

[322] von Kállay, _Geschichte der Serben_, i. 152.

[323] Ibid., i. 152-9; Jireček, op. cit., 319-20; Ljubić, _Monumenta spect. ad hist. Slav. merid._, iv. 189.

[324] Cant. IV., 50, pp. 360-2; Müller, _Chron. Byz._, under 1364.

[325] Miklositch, _Acta Serbica_, CLIII.

[326] Ibid., CLX.

[327] Sons of a poor Dalmatian nobleman: Ducange, _Familiae Byz._ viii. 294.

[328] At Ipek, with an independent patriarch: Engel, _Geschichte von Serbien_, p. 279.

[329] Miklositch-Müller, _Acta gr._, CLXII; MS. Wiener Bibl., Gesch. gr., No. 47, fol. 290.

[330] Ibid., CLX; ibid., fol. 286.

[331] Orbini, p. 275.

[332] Engel, op. cit., pp. 321 f. For documented details, Müller, ed. _Byz. Analekten_, pp. 359-64, 405-6, based on Vienna MS. referred to above.

[333] Now called Cermen or Tchirman.

[334] Svilengrad, now the frontier station of Bulgaria, was known from 1361 to 1913 as Mustapha Pasha. Before the recent Balkan war, it was the frontier railway station of Turkey.

[335] But there were certainly two distinct battles here, in 1363 and in 1371. See p. 124, _n._ 2, above.

[336] Ducange, op. cit., p. 294; Bialloblotszky’s translation of Rabbi Joseph, i. 240; Klaić, p. 199; Jireček, pp. 329-30. Zinkeisen, i. 224, confuses this battle with the one fought in 1363.

[337] In Miklositch, _Chrestomathia palaeoslav._, p. 77.

[338] Phr., I. 26, p. 80, gives the capture of these cities in the same campaign as that in which Monastir was acquired, with 1386 as date. But the Serbian chronicles are so explicit here that we can follow them without hesitation, especially as they are seconded by the Ottoman historians. Cf. Hammer, i. 241, and Zinkeisen, i. 229.

[339] Pope Gregory XI, writing to Louis of Hungary, May 14, 1372, informed him that the Osmanlis had conquered some parts of Greece, ‘subactis quibusdam magnatibus Rasciae, tum in eis dominantibus’. Rascia was Servia. Theiner, _Monumenta Hungarica_, ii. 115.

[340] Gregory XI, _Epp. secr._ ii. 32-3.

[341] According to Amilhau and Jireček, who rely on Reynaldus, ann. 1364, XXVIII, this first invasion of the Greek peninsula took place in 1363. But the Turks referred to in that year, probably of the perennial corsair type, could not have been Osmanlis. They were from Aïdin or Sarukhan.

[342] Klaić, _Geschichte Bosniens_, p. 200.

[343] Hammer, i. 242, 409, places the first relations of Lazar with Murad after the fall of Nish, which he erroneously puts in 1376. See below, p. 161, _n._ 3.

[344] Gregory XI, _Epp. secr._ iii. 42.

[345] June 15, 1373: Andrea Gataro, in Muratori, xvii, col. 176.

[346] Ducas, 12, pp. 43-4; Phr., I. 11, pp. 49-50.

[347] Chalc., I, pp. 42-3. But Murad, according to the Collection of Feridun, when he wrote to the Prince of Karamania, stated that Saoudji had been conquered in a pitched battle: MS. Bibl. Nat., Paris, fonds turc, No. 79, p. 30.

[348] Letter just cited; Chalc., I, pp. 44-5; Phr., I. 12, p. 51. Saoudji is called Kontouz by Ducas, Mosis by Phrantzes, and Saouzis by Chalcocondylas. I cannot find the reading Siaous which Hammer, i. 412, and _n._ lix, attributes to Chalcocondylas.

[349] Chalc., I, p. 46; II, p. 69; Phr., I. 12, p. 51; Duc., 12, p. 44.

[350] Canale, ii. 16.

[351] Clavijo de Gonzáles, 15 vº and 16 rº.

[352] So Phrantzes thinks, I. 12, p. 51: ταύτην ὠμότητα καὶ ἀπανθρωπίαν ὁ ’Αμουράτης ἐποίησεν ἀεὶ εἰς τὰ πάντα καλῶς πολιτευόμενος.

[353] Chalc., I, pp. 46-7; Phr., I. 11, pp. 47-9.

[354] Romanin, iii. 255. This project, according to Cicogna, _Istoria di Venezia_, vi. 95, was first broached to John at the time of his visit to Venice in 1370.

[355] Raynaldus, ann. 1376, XXIII.

[356] _Epp. secr._, vi. 236.

[357] Ducas, 12, p. 45.

[358] Caresino, in Muratori, xii.

[359] Ducas, 12, p. 45; Chalc., II, p. 63; and Phr., I. 13, p. 54, say that Bayezid had given him 1,000 men, and had often advised him to have his father and brothers assassinated. Cf. Muralt, ii. 706.

[360] Sauli, ii. 57.

[361] Quirino, _Vita di Zeno_, cited by Muralt, ii. 707, Nos. 6-9.

[362] Ducas, 12, p. 45.

[363] The fortunes of Salonika at this period are obscure. See p. 231, below.

[364] Chalc., II, p. 63; Phr., I. 13, pp. 55-6.

[365] Chalc., II, p. 64. But Ducas, 4, p. 19, says that Bayezid captured this city.

[366] Bonlinius, II. 10; Sanudo, _Vite de’ Duchi_, in Muratori, xxii, col. 680.

[367] An excellent brief account of this war is found in Wiel’s _Story of Venice_, pp. 227-37.

[368] The Genoese forced John V to make peace with Andronicus in November 1382: Sauli, ii. 260.

[369] Cicogna, op. cit., vi. 97; Romanin, iii. 301.

[370] Hadji Khalfa, _Djihannuma_, fol. 1852; Evliya effendi, ii. 229.

[371] The testimony of Ibn Batutah, who travelled extensively among the Turks in Anatolia, southern Russia, and elsewhere between 1325 and 1340, is conclusive on this point. ‘Whenever we stopped in a house of this country (Anatolia), our neighbours of both sexes took care of us: _the women were not veiled_ ...’: ii. 256. ‘I was witness of a remarkable thing, that is, of the consideration which the women enjoy among the Turks: they hold, in fact, a rank more elevated than that of the men.... As for the women of the lower classes, I have seen them also. One of them will be, for example, in a cart drawn by horses. Near her will be three or four young girls.... The windows of the cart will be open and you can see the women’s faces: _for the women of the Turks are not veiled_.... Often the woman is accompanied by her husband, whom whoever sees him takes for one of her servants’: ii. 377-9. No student can have any doubt whatever upon the position of Turkish women during the fourteenth century. As among all vigorous peoples, the women of the Osmanlis held a high place, and were never secluded. It was not until Murad II that even the sovereign had a harem. The Moslem conception of the inferiority of women was not prevalent among the Osmanlis until after the reign of Soleiman the Magnificent. Immediately it became prevalent, the race began its decline.

So universal did veiling become in the seventeenth century that it was adopted by Christian and Jewish women in Turkey as well. See Père Febre, _Théâtre de la Turquie_ (1682), pp. 164-5. Père Febre spoke from personal experience ‘dans la plupart des lieux de la Turquie’.

[372] Hadji Khalfa, _Rumeli_, p. 96.

[373] _Historia epirotica_, Bonn ed., p. 228.

[374] Ibid., pp. 230-1.

[375] Ducange, viii. 292.

[376] Jireček, op. cit., 340.

[377] _Misti_, XL. 154.

[378] See below, p. 203.

[379] Silvestre de Sacy, in _Mém. de l’Acad. des Inscript._, vii. 327-34. But the commandant could hardly have been carried by his falconer in such a fashion as far as Philippopolis. The Ottoman historians probably forgot that Ishtiman, at the mouth of the pass, on the road to Philippopolis from Sofia, contained an Ottoman garrison.

[380] According to the anonymous _Ein gantz neu Reysebuch von Prag auss biss gen Constantinopel_, Nürnberg, 1622, p. 33, Sofia was captured in 1362. Hadji Khalfa, _Rumeli_, p. 51, with whom Schéfer, ed. Bertrandon de la Broquière, p. 202 n., seems to agree by citing, says Sofia capitulated in 780 (1378). Seadeddin, i. 125, is followed by Hammer, i. 250, Klaić, p. 237, and others in fixing the date as 1382. But these same authorities give 1375 and 1376 for Nish, which is altogether impossible. Phr., I. 26, p. 80, seems to place the capture of Sofia for 1385. This is the most reasonable date. It is consistent both with the topography of the places in question and with Murad’s methods of campaigning, as exemplified by all his conquests, to place the taking of Sofia close to the end of his reign, and within a year or two before the capture of Nish. To corroborate this date, letters in the collection of Feridun can be cited. Indje Balaban’s letter to Murad, announcing the acquisition of Sofia, is not dated. But immediately after it is the response of Murad, in which he gives to Indje Balaban for life the government of the new province, and states that he is sending him a fine horse and robes of honour because of his success. This letter is dated from Adrianople in the middle of the month of Redjeb, 788, which corresponds to 1386 in our era. These letters are in MS. Bibl. Nat., Paris, fonds turc, No. 79, pp. 31-2.

[381] Nish, from its geographical position, could not have fallen in 1375, as Chalcocondylas says. Hammer, i. 241, and Zinkeisen, i. 230, show an amazing nonchalance in transporting the Osmanlis from Kavalla, Drama, and Serres in the course of this one year, 1375. Engel, _Geschichte von Serbien_, p. 341, who, according to Hammer, ‘deceives himself by thirteen years in placing the capture of Nish in 1388’, is eleven years nearer the truth than Hammer! Strumnitza, from a diploma delivered in the name of the Serbian empress Eudoxia (Müller, _Acta Serbica_, CXXXI), was independent in 1379. Sofia did not fall before 1382. How, then, could Nish have been an Ottoman fortress from 1375?

[382] Von Kállay, i. 166.

[383] For distances between cities in the Balkan peninsula, see Jireček’s important and interesting work, _Die Heerstrasse von Belgrad nach Konstantinopel und die Balkanpässe_, p. 122. Jireček, for time of transit, depends upon Hadji Khalfa.

[384] Text in Sauli, ii. 260-8.

[385] _armiratus_ or _amiratus_, then _amiralus_, of which we have made admiral, originally had nothing whatever to do with the sea. It is a corruption of _emir_.

[386] ‘Magnificus et potens dominus, dominus Moratibei, magnus armiratus et dominus armiratorum Turchie’: the whole text is reproduced from the Genoese archives by Belgrano, in _Atti della Società Ligure di Storia_, xiii. 146-9, and by Silvestre de Sacy, in _Notices et Extraits_, xi. 58-61. Cf. Canale, ii. 59.

[387] ‘Contra illum Turcum filium iniquitatis et nequiciae, ac Sancte Crucis inimicum, Moratum bey et eius sectam, cristianum genus sic graviter invadere conantes.’ The text of this treaty is also in Belgrano, ibid., xiii. 953-65.

[388] Text in Romanin, iii. 386-9. There was an earlier law of similar nature enacted in 1334.

[389] Cf. Delaville Leroulx, i. 159-60.

[390] Romanin, iii. 331.

[391] _Bullarum_, III, part 2, pp. 4, 92, 338; Urban V, _Epp. secr._ iii. 25, iv. 256; Gregory XI, _Epp. secr._ ii. 32-3, v. 88-9, 311; Philippe de Mézeray, p. 19; Raynaldus, ann. 1372, XXIX. In 1425 Martin V repeated the anathema against those who sold Christian slaves to the Turks: _Bullarum_, III, part 2, p. 454.

[392] MS., Bibl. de Bâle, A 1, 28, fols. 232-54, cited by Delaville Leroulx, i. 70, n. 2. Adam’s project was a revival of Sanudo’s attempt to ruin Moslem trading.

[393] _Monumenta historiae patriae_, i. 320; iii. 336, 371.

[394] In 1432 Bertrandon de la Broquière met at Damascus one of these Genoese of Kaffa, who sold slaves to the Sultan of Egypt: _Voyage_, Schéfer ed., p. 68.

[395] Chalc., I, p. 53; Phr., I. 26, p. 81. Cf. Hertzberg, p. 503.

[396] Seadeddin, i. 130-2, draws here upon Idris and Neshri, and has been followed by all the Ottoman historians down to the present day.

[397] Col. Djevad, pp. 62-3. He speaks of Alaeddin bey ‘ayant levé l’étendard de la révolte’, and calls the punishment of the Serbians in this campaign the chief cause of Kossova.

[398] Chalc., I, p. 53; Phr., I. 26, p. 81.

[399] Up to 1383, in outlining the career of Tvrtko, I have followed Klaić, pp. 201-3.

[400] Schaffarik, _Acta archivii Veneti_, &c., CXLI.

[401] In a letter of April 1, published in Ljubić, iv. 185-6.

[402] _Misti_, xxxix. 113.

[403] Klaić, p. 237; Jireček, p. 341. But von Kállay, i. 166, attributes this victory to George Kastriota of Albania.

[404] Orbini, p. 361.

[405] Hopf, in Ersch-Gruber, _Allgemeine Encycl._, lxxxvi. 49.

[406] _Chronique de Morée_, p. 516. Evrenos is called Branezis. This is not the Evrenos heretofore mentioned, but another Christian renegade, of Macedonia. Cf. Finlay, iv. 233 _n._

[407] Jireček, _Die Heerstrasse_, &c., p. 147.

[408] Leunclavius (1611 Frankfort ed.), pp. 268-76. Jireček, _Geschichte der Bulgaren_, pp. 341-2, points out that Seadeddin and Leunclavius, whom Zinkeisen, i. 252-5, follows, are in error in representing the Bulgarians as wholly subdued in 1388.

[409] Mijatovitch, from Serbian sources, p. 13.

[410] Ibid., pp. 16-17.

[411] The railway between Mitrovitza and Skoplje (before the Balkan War Uskub) passes through the plain of Kossova. When this railway is connected through the former Sandjak of Novi Bazar with the Austrian (?) railways in Bosnia, Kossovapol will be on one of the great transcontinental routes.

[412] The date June 15 is fixed by the Serbian chronicles and songs and by unbroken tradition. Also by Tvrtko’s letter to Florence. But Tvrtko, in another letter to the inhabitants of Trau in Dalmatia, gives June 20 (Pray, _Annales_, ii. 90). Seadeddin stands alone in placing the death of Murad on the 4th Ramazan (August 27). The other Ottoman historians, as well as Chalcocondylas, Ducas, and the anonymous _Hist. Epirot._, speak of these events occurring ‘in the springtime’.

[413] _Chron. of Abbey of Tronosha_, section 54, p. 84, and _Chron. of Pek_, p. 53: cited by Mijatovitch, p. 12 n.

[414]

‘Sans arrêter, pendant quinze jours pleins, J’ai cheminé le long des hordes turques, Sans en trouver ni la fin ni le nombre.’--A. d’Avril, p. 36.

[415] Orbini, pp. 314-15. See also the Serbian songs about Kossova, which are accessible in the form of a continuous narrative in French by Adolphe d’Avril, and in English by Mme Mijatovitch, based on the composite poems of Stoyan Novakovich and A. Pavich.

[416] Solakzadé, cited by Col. Djevad bey, p. 196. The bow was used as an offensive arm by the Osmanlis until the middle of Murad II’s reign.

[417] Seadeddin, i. 147-52; Chalc., I, p. 53; Ducas, 3, pp. 15-16; _Hist. Epir._, p. 234; the Serbian chants; Bonincontrius, col. 52; and the modern writers, Hertzberg, pp. 503-7; Jireček, pp. 342-4; Fessler, ii. 254; von Kállay, i. 166; Klaić, pp. 236-40. Most illuminating of all is Rački, in Croatian, in _Jugoslav. Akademie_, iii. 92 f.

[418] Clavijo de Gonzáles, fol. 27 rº.

[419] _Annales_, ii. 186.

[420] This speech, from the chronicle of Monk Pahomye, is given in Mijatovitch, p. 17.

[421] Busbecq, English ed., i. 153; cf. Ricaut, ed. of 1682, p. 159.

[422] Const. Porphyr., i. 394, 396, 405.

[423] Howorth, ii. 796, commenting on Stoddard’s audience with the Emir of Bukhara.

[424] Text in _Mon. spect. hist. Slav. Merid._, i. 528-9.

[425] _Chronique du Religieux de St.-Denis_ (ed. Bellaguet), ii. 391.

[426] MS., Wiener Bibl., Gesch. gr., 48.

[427] As far as such records are accessible in the great collection of Miklositch and Müller. The statement of Ducas, 23, p. 137, about the persecutions of Christians by Murad, is without any foundation.

[428] Phr., I. 26, p. 82; Chalc., I, p. 59; Duc., 3, p. 16; also the Ottoman historians.

[429] Sura IV, verse 94 (Sale trans., p. 64).

[430] Sura V, verse 53 (ibid., p. 77).

[431] Hammer, iii. 302-4. Rambaud, _Histoire générale_, iii. 831, is mistaken in attributing this law to Bayezid.

[432] Ruled 1350 to 1369.

[433] In 1330. Panaretos, p. 7.

[434] In 1320 at Salonika: Greg., VII. 13, p. 271.

[435] Month of Shaban, a.h. 791: MS. turc, Bibl. Nat., Paris, No. 79, pp. 35, 40. Cf. Langlès, in _Notices et Extraits_, v. 672.

[436] Froissart, IV. c. 47, in Kervyn ed., xv. 216-17. Froissart calls Bayezid ‘Amoruth-Baquin’, confusing him with Murad. See below, p. 213, _n._ 2.

[437] Abul Yussif ibn Taghry, _Elmanhal Essafy_, Bibl. Nat., Paris, fonds arabe, No. 748, ii, fol. 70.

[438] Vuk Brankovitch, as the reward of his treason, received half of Lazar’s inheritance, however, with Pristina as capital. His family continued as Ottoman vassals, with varying fortunes, for a hundred years.

[439] Ducas, 4, p. 6.

[440] Kantitz, _Serbien_, pp. 254 f.

[441] Busbequius was informed at Constantinople that marriage had been abolished in the Ottoman royal family because Bayezid took to heart the disgrace of Despina by Timur. But Ricaut, p. 296, thinks that it was because of dowry expense and the desire of the Ottoman sovereigns to keep free from family alliances. Naturally, the difference of religion in time prevented the Osmanlis from finding brides for their sovereigns among the European royal families. If they married among their subjects, there was always fear of intrigues in the wife’s family. At a time when family alliances meant so much in Europe, the Ottoman Empire suffered greatly from this disability.

[442] Seadeddin, i. 158.

[443] Klaić, p. 248. I think Romanin, iii. 331, has confused Stephen Bulcovitch with Stephen Tvrtko. For it is difficult to understand what he means by the ‘pace vergognosa’ with Venice.

[444] Old Servian chronicle, quoted by Klaić, p. 271: ‘quasi totaliter destruxerunt Bosniam et populum abduxerunt.’

[445] Klaić, pp. 324-5.

[446] Accounts differ as to the place. There is some doubt as to whether the independence of Aïdin was totally destroyed before the restoration of Isa’s sons by Timur. Cf. Schlumberger, p. 484; Mas-Latrie, _Trésor de Chronologie_, col. 1800. Hammer, i. 300, cites no authorities for his statements about this usurpation.

[447] Bosio, ii. 143.

[448] Ibid., ii. 148.

[449] There is the same dearth of information about the details of the destruction of the power of the emirs of Sarukhan and Menteshe as there is about Aïdin. Hammer says simply, ‘Les principautés de S. et M. furent incorporées à l’empire ottoman,’ i. 300. He gives no authorities.

[450] Ducas, 13, p. 47.

[451] _Dialogi XXVI cum Persa quodam de Christianae religionis veritate_, Bibl. Nat., Paris, fonds grec, No. 1253: partly printed in _Notices et Extraits_, vol. viii, 2e partie, and in Migne, 156, pp. 111-74. In _Notices et Extraits_, loc. cit., C. B. Hase has given an interesting critical account of the dialogues, and the circumstances under which they were written.

[452] Seadeddin, i. 163. In Hammer, i. 301, in the sentence ‘quoique, depuis la paix renouvelée avec lui par Orkhan, les deux nations eussent continuellement vécu dans des relations de sincère amitié’, is not Murad meant instead of Orkhan?

[453] Evliya effendi, ii. 21, tells how Bayezid passed _seven times in one year_ from Anatolia to Wallachia.

[454] In matters relating to the progress of Ottoman conquest in Asia Minor, French, German, and British writers have been content to repeat, without critical comment, what they have culled from Leunclavius or the translations of Seadeddin. In many cases, they have gone back no farther than Hammer, and have transcribed, often literally, Hammer’s words. Hammer himself, in this early period of Ottoman history, in spite of his attainments as an orientalist, has relied mainly on Leunclavius, and on Bratutti’s Italian translation of Seadeddin.

[455] ‘La principauté _fut pour toujours réunie à l’empire_,’ Hammer, i. 308. In speaking of this second campaign, Hammer starts by saying, ‘Le prince de Karamanie avait de nouveau _levé l’étendard de la révolte_’. This is hardly the expression to use for the action of an independent prince. Alaeddin had never made himself the vassal of the Ottoman emirs.

[456] Striking testimony to the later power of the Karamanlis is given by Bertrandon de la Broquière, who visited the court of Ibrahim with the Cypriote ambassador in 1443: cf. Schéfer’s edition of his voyage, pp. 108-20. See Appendix B below, where the relations of the Osmanlis with the emirates of Asia Minor during the fourteenth century are discussed in detail, with fuller citation of authorities.

[457] Howorth, iii. 749.

[458] Sherefeddin, iii. 256, who is the only contemporary authority, says that Bayezid put him to death. This was one of the charges made by Timur against Bayezid.

[459] The earliest possible date could be 1393. Perhaps the Osmanlis first appeared near Sivas at this time. But Bayezid would hardly have undertaken so long and perilous an expedition before his position was secure in Karamania. Sherefeddin gives the more likely date 1395, while Ibn-Hedjir places the death of Burhaneddin in 1396.

[460] So d’Ohsson fils, vii. 442, says, but gives date 1390. Hammer more correctly puts it in 1391. Xénopol, in his authoritative and carefully documented history, gives a little different account of Mircea’s early relations with Bayezid, and attributes to Mircea a larger influence in the calculations of Murad than he deserves. But the exposition of Mircea’s policy in relation to Poland, Hungary, and the Osmanlis, as given by Xénopol, cannot be overlooked or disregarded by the student of this period.

[461] ‘Pierre Aron fut le premier des hospodars qui paya un tribut aux Turcs’: Costin’s _Hist. de la Moldavie_, p. 367, in _Notices et Extraits_, vol. xi.

[462] Phr., I. 13-14, pp. 58-9, and 26, p. 82; Bonfinius, iii. 2; _Chron. Anon. de St.-Denis_; Chron. of Drechsler; Campana, fol. 8 (but gives date 1393). Leuncl., _Annales_, p. 51, following Ottoman sources, speaks only of Sigismund’s defeat. This earlier victory and the disastrous retreat are mentioned also in several of the French chronicles which relate the expedition of 1396.

[463] Engel, _Gesch. von Ungarn_, ii. 368, who draws on all the earlier Hungarian authorities.

[464] Russian source cited by Muralt, vol. ii, No. 10 n.

[465] Cf. Baedeker, _Konst. und Kleinasien_, 2. Aufl., p. 46.

[466] Jireček, _Gesch. der Bulgaren_, pp. 347-9, gives Slavic sources for this date, and quotes Camblak’s graphic description of the terrible sacking of the city, the massacre, and the destruction of the churches.

[467] In Czech, the word _jazyk_ signifies _language_ as well as _nation_ (cf. Lützow, _Life and Times of Master John Hus_, p. 239). This illustrates the Slavic conception of nationality, and explains in a nutshell the Austro-Hungarian and Balkan problems. To the Slav, there can be no other test of nationality. The Bulgarian propaganda in Macedonia, carried on through the church and the schools, has been the resurrection of the nation through the language. The Greeks have used the Orthodox Church to combat and stifle this movement. They claim as Greeks all members of the Orthodox Church, while the Bulgarians claim that Bulgarophones, even if not attached to the exarchate, belong to the Bulgarian nation.

[468] Schiltberger, Neumann ed., p. 65. On this question cf. Jireček, op. cit., pp. 350-2; Miller, p. 189; and illuminating note of Rambaud, in _Hist. générale_, iii. 832 _n._ Also p. 143 of this book and accompanying foot-note.

[469] Schiltberger, op. et loc. cit.

[470] These cities, or rather, their fortresses, were captured and evacuated several times by the Osmanlis, especially Widin.

[471] Hammer, at the beginning of the reign of Bayezid, i. 295-7, relates the history of the quarrel between Andronicus and his father and Manuel, the rescue from the Tower of Anemas, &c., as if these events happened in 1389 and 1390, and gives the capture of Philadelphia for 1391. He has been led astray here by the story of Ducas, and by the fact that the Byzantine historians speak of Bayezid instead of Murad in connexion with the negotiations for restoration. By the internal evidence in the Byzantine historians themselves, the chronology of this period cannot be decided. But, by reading Phrantzes and Chalcocondylas in the light of Quirino, the continuation of Dandolo, and the archives of the colony of Pera, and also by piecing out the length of time of these events and matching them with Bayezid’s occupations during the first two years of his reign, it is not difficult to decide to place the Andronicus _versus_ John and Manuel struggles just before the Chioggia war. At any rate, Andronicus died ten years before the date Hammer gives to these events!

[472] Poem cited by Muralt, ii. 738, No. 1.

[473] MS. Bibl. Nat., Paris, fonds grec, No. 1253, fol. 198 vº.

[474] John V Palaeologos was of those who, in the words of Bernino (p. 9), ‘consumavasi vanamente il tempo più in dolersi delle calamità che in repararle’.

[475] Ducas, 13, pp. 25-49 _passim_; Chalc., II, pp. 66, 81-2.

[476] Evliya effendi, i. 29-30; ii. 21, who repeats the persistent Ottoman tradition of his day, that is also found in Hadji Khalfa and Nazmi Zadé. Cf. the Genoese accounts of Pera in Jorga’s scholarly _Notes à servir_, &c. i. 42. According to Schéfer, in his edition of Bertrandon de la Broquière, p. 165, there was a provision that slaves escaping to Constantinople should be given back, but we cannot be sure that this stipulation was made under Bayezid I. The date of the installation of the cadi, &c., is open to question. Some authorities place it after Nicopolis.

[477] Shehabeddin, fol. 72 rº, writes _Istanbul_; Sherefeddin, iv. 37, is transcribed by Petits de la Croix _Istanbol_; Arabshah, p. 124, transcribed by Vattier _Estanbol_. Wylie, i. 156, _n._ 2, gives the time-worn popular derivation from εἰς τὴν πόλιν; also Telfer, in his edition of Schiltberger, p. 119. Why go so far afield? _Istambul_ is a natural contraction of Constantinople. As the Greeks pronounced this long word, the syllables _stan_ and _pol_ bore the stress, and were naturally put together for a shortened form. As for the initial _I_, which has troubled the philologists, its explanation is easy to one who knows the Osmanlis. They cannot to this day pronounce an initial _St_ without putting _I_ before it.

[478] Neshri, trans. Nöldeke, _ZDMG._, xv. 345; Seadeddin, i. 189; Saguntinus, p. 187; Drechsler, p. 228, says: ‘octo annos vexatur et obsidetur.’

[479] Duc., 13, p. 50.

[480] Muralt, ii. 753, No. 29.

[481] _Secr. Cons. Rog._, III, E 84.

[482] Chalc., II, pp. 80-1; Phr., I. 13, pp. 57-8; 26, p. 82.

[483] Miklositch, _Acta Serbica_, CCIV. Hammer, i. 341, calls this Constantine ‘fils de Twarko’, meaning Stephen Tvrtko, I suppose. But I cannot find that the Bosnian king had such a son, or any reason, if he had, why this son should have been at Serres.

[484] Ibid., CCXXIII. For the later kings of the dynasty which Vuk founded, see Picot’s careful article in _Columna lui Traianu_, new ser., Jan.-Feb. 1883, p. 64 f.

[485] _Epp. cur._, ii. 64.

[486] _Epp. cur._, ii. 103-4. Urban VI in 1387 had written a letter from Lucca inciting the Frankish princes to a war against ‘schismatics’ in Achaia.

[487] _Secr. Cons. Rog._, iii, E 74.

[488] Miklositch-Müller, _Acta Graeca_, CCCCXXXV.

[489] Chalc., II, p. 75; Duc., 13, p. 73.

[490] Chalc., loc. cit.; _Epp. cur._, ii. 300-1, 311; iii. 261.

[491] Cf. Jorga, in _Bibl. de l’École des Chartes_, 2e série, 110e fascic.; Molinier, MSS. de P. de Mézières, in _Arch. de l’Orient Latin_, i. 335-64; Del. Leroulx, i. 201-8.

[492] ‘Nostra dominatio audiverat de morte ipsius dom. Morati, de qua maximam displicentiam habuerat, quia semper eum habuimus in singularissimum amicum, et dileximus eum et statum suum. Similiter audivimus de felici creatione sua ad imperium et dominium ipsius patris sui, de quo nos fuerimus valde letati, quia sicut sincere dileximus patrem, ita diligimus et diligere dispositi sumus filium et suum dominium et habere ipsum in singularem amicum’ ... &c.: _Misti_, xli. 24, reprinted in full in Ljubić, iv. 269-70.

[493] Ibid., xlii. 58-9; the treaty is in _Commem._, viii. 150. Cf. Romanin, iii. 330.

[494] Euboea is called Negropont, the Peloponnesus Morea, Lesbos Mytilene, while Crete is frequently called Candia and Chios Scio, in mediaeval and modern times.

[495] _Misti_, xlii. 55.

[496] Ibid., xliii. 156.

[497] _Secr. Cons. Rog._, iii. E 81.

[498] ‘Ire contra dictos Turchos ad damnum et destructionem suam’: ibid., p. 94, cited in Ljubić, iv. 335-6.

[499] _Misti_, lxiv. 140.

[500] Ibid., lxiv. 156.

[501] We must reject the statement of Morosini, MS. Wiener Bibl., fol. 135 rº, that Bayezid ‘entered in arms in the Strait of Romania with so many galleys that one could not navigate in the strait’, and doubt the opinion that Monicego, with his forty-four Venetian and Genoese galleys, had to force the Bosphorus, and contributed powerfully ‘a la destrucion del dito Turcho’.

[502] _Misti_, xliii. 29.

[503] Ibid., xliii. 5: ‘confidasse in Dio, confidasse nei provedimenti che saprebbero à fare i principi christiani, scrivesse al Papa e a questi promovendo una lega generale’.

[504] Ibid., xliv. 108.

[505] Ibid., xliv. 128.

[506] Belgrano, pp. 152-3.

[507] _Lib. iurium_, ann. 1392, fol. 474, in Turin archives, printed in _Bibl. de l’École des Chartes_ (1857), 4e série, iii. 451-2.

[508] _Religieux de St.-Denis_, ed. Bellaguet, i. 319-21.

[509] _Chronicorum Karoli Sexti_, ed. Bellaguet, i. 709-11. The relations of the ambassadors of Sigismund with the Duke of Burgundy and with Charles VI are found in _Religieux de Saint-Denis_.

[510] On September 13, 1395, in the presence of ambassadors from all parts of Christendom, and also ‘del gran Turco, del Rè de’ Tartari, del gran Soldano, del gran Tamerlano e di molti altri Principi infedeli e ribelli alla Fede christiana’, who were treated like Christians and lodged at the expense of ‘il Signore di Milano’, Galeazzo was solemnly raised to ducal rank and invested with the Duchy of Milan by Wenceslaus: Andrea Gataro, in Muratori, xvii, col. 820.

[511] _Mémoires de Madame de Lussan_, iii. 5.

[512] The references to Froissart which follow are given from vol. xv of Kervyn de Lettenhove’s edition, and the references to Schiltberger from the English translation in the Hakluyt Society series, vol. lviii, unless otherwise specified.

[513] See the sources and references for Nicopolis grouped in the classified bibliography. Although the citations in the text of my narrative are mostly from Froissart and Schiltberger, all chronicles and contemporary sources available have been used in the preparation of this section, especially Bellaguet’s edition of _Religieux de Saint-Denis_, ii. 425-30, 483-522 (Bellaguet’s notes, however, on these sections are very disappointing).

[514] Froissart, pp. 218, 221, 223.

[515] Ibid., pp. 227-8, 230, 394-8. A complete list of the chevaliers, compiled from sources, is found in Buchon, and, in much more complete and accurate form, in Delaville Leroulx, ii. 78-86.

[516] Froissart, and other earlier writers, have several ways of designating Bayezid. Froissart calls him Amorath-Baquin (p. 216), Amorath (p. 226), le roy Basaach, dit l’Amourath-Bacquin(p. 230), l’Amourath-Bacquin many times, and l’Amourath three times in one paragraph (p. 311). Chroniclers and writers of the fifteenth century were continually confusing Bayezid with Murad (cf. Cuspianus, Secundinus, Sylvius Aeneas, Donado da Lezze, Paolo Giovio, _et al._). From the different ways Froissart designates Bayezid, it is very clear that he is not mixing him with Murad, but that by ‘dit l’Amourath-Bacquin’ he means ‘_l’émir-pacha_’. The fact that he uses the definite article so frequently and says several times ‘l’Amourath’ is proof positive of this. His transcription of the title _emir_, and that of many other western writers, led later historians to think the chroniclers meant Murad! It is merely a coincidence that the words are so similar. Froissart, however, would be capable of mistaking Murad for Bayezid. On p. 216 he calls Sigismund Henry, and on p. 334 Louis! Olivier de la Marche (éd. Beaune et d’Arbuthnot), i. 83-4, speaks twice of ‘Lamourath-bahy’. Here, too, there is not a confusion of Murad and Bayezid. He, like Froissart, means to say ‘l’amiral-pacha’. On ‘amiral’ for ‘emir’ see above, p. 163, _n._ 2.

[517] Froissart, pp. 230-1, 242.

[518] Donado da Lezze, p. 9.

[519] Leunclavius, _Hist. Musul. Turc._, preface, p. 14, speaks of how grateful Sigismund was later for the services rendered to him personally by the Burgrave in the Nicopolis campaign, and that the friendship formed then led to the later advancement of the house of Brandenburg.

[520] Wylie, i. 6, 158, quoting Ducas, 13, and _Venetian State Papers_ (Brown), i. 85. Ducas knew nothing of Nicopolis, while the Venetian reference is based on a misapprehension.

[521] Lavisse, _Histoire de France_, iv. 311: ‘on l’avait vu à la bataille de Nicopoli sur les bords de la Baltique avec les chevaliers _teutoniques_.’ Lavisse has evidently mixed up the Nicopolis expedition with the earlier Prussian one in which Henry did take part. His statement on the same page that Henry IV took part in the Boucicaut expedition is another error.

[522] Conclusive proof of the whereabouts of Henry in the summer of 1396 is found in the letter ‘escript ... le xxe jour d’augst’. This letter is in Arch. Nat., Paris, J. 644: 35^{11}. For the expeditions in which Henry _did_ take part, when he was Henry of Derby, see vol. lii of the Camden Society, edited by Lucy Toulmin Smith, London, 1894, 4to.

[523] Froissart, p. 244.

[524] Phr., I. 14, p. 59; Bontinius, III. 2.

[525] Engel, _Geschichte der Bulgaren_, p. 468. According to the authority who has made the most exhaustive study of the Nicopolis expedition, Sigismund disposed of 120,000 men in all, including the western allies: Kiss, in _À Nikápolyi ülkozet_, p. 266. Kiss’s estimate is corroborated by the _Cronica Dolfina_, which says that Sigismund had one hundred thousand men under arms in 1394. Sanuto quotes this in Muratori, xxii, col. 762. Cf. also Hungarian Nat. Archives, Dipl. 8201, 8212, 8214, 8493, 8541.

[526] Schiltberger, p. 2.

[527] Bruun, in his _Geographische Anmerkungen zum Reisebuch von Schiltberger_ (_Sitz.-Ber. k. Bay. Akademie_, 1869, ii. 271), tried to prove that the battle was fought, not at Nicopolis on the Danube, but near the ancient Nicopolis of Trajan’s foundation. But in his notes to the English translation of Schilt., Hakluyt, lviii. 108-9, he assents to the contention which Kanitz makes in _Donau-Bulgarien_, ii. 58-70, that the battle was near Nicopolis-on-the-Danube. An examination of the chronicles corroborates Kanitz’s hypothesis over against the ingenuous argument of Jireček. Some historians have been so unmindful of geographical considerations as to put the battle at the ancient Nicopolis _ad Haemum_, of which Ortellius, p. 225, speaks.

[528] Schiltberger, p. 2.

[529] Froissart, pp. 251, 262-3, 310, 329. ‘Miscreant’, of course, in its original sense.

[530] Ibid., p. 310.

[531] Ibid., pp. 311-17; _Relig. de St.-Denis_, pp. 490-7. Schiltberger, p. 3, attributes this initiative to Jean de Nevers, whom, like many other writers on Nicopolis, he calls, by anticipation, Duke of Burgundy. Cf. Donado da Lezze, p. 9, who says: ‘Il signor Carlo, _prima_ Duca di Borgogna.’ Also Morosini, p. 6. Sigismund is frequently spoken of as German emperor at the time of Nicopolis. Cf. Chalc., ed. Migne, col. 76: ἡγουμένου Σιγισυούνδου Ῥωμαίων βασιλέως τε καὶ αὐτοκράτορος.

[532] Rabbi Joseph, i. 252.

[533] Froissart, pp. 313-16; _Relig. de St.-Denis_, pp. 490 f.; Rabbi Joseph, p. 253; Schiltberger, p. 3; Seadeddin, i. 184; Neshri, in _ZDMG._, xv. 345-8. Cf. authorities cited in Bibliography.

[534] Froissart, p. 317. Hermann de Cilly and the Burgrave of Nürnberg are said by some authorities to have thrown themselves in front of Sigismund, and to have saved him and carried him off to the boat.

[535] The bitterness against and contempt for the Hungarians is expressed in the following verses:

‘Nichopoly, cité de payennie, Au temps là où li sièges fut grans, Fut delaissés par orgueil et folie; Car les Hongres qui furent sur les champs Avec leur roy, fuitis et récréans, Leur roy meisme enmainent par puissance, Sans assembler.’ _Œuvres inédites d’Eustache des Champs_, ed. Tarbé, 1849, i. 166.

[536] Schiltberger calls him ‘der hertzog auss der Sirifey, der genant despot’: _Bibl. des Lit. Vereins_ (Tübingen), clxxii. 4.

[537] Cf. Miller, in Story of Nations Series, pp. 290-1.

[538] Belonging to the grand master of Rhodes: Froissart, p. 317. But Morosini, p. 15, and others, say that he went directly on board Monicego’s galley. It is a pity that Hammer, in his description of the battle of Nicopolis, relied so much on such an unreliable third-hand authority as Abbé Vertot. Skentkláráy, _À dunai hajóhadak törtenéte_, says that Jean de Vienne commanded the galleys.

[539] Schiltberger, p. 6.

[540] Bonfinius, one of the earliest Hungarian historians, recorded that Sigismund had boasted that he would not only turn the Osmanlis out of Europe, but also that with the army under his command, if the sky fell, it could be held up on their lances: _Decades_, ii. 403.

[541] ‘Sigismund was cruel and sensual, perjured and frivolous, rapacious and dissolute, fierce and pusillanimous, a byword and object of horror to the Bohemians, hated and despised by the Germans, a warning to all rulers. His companion, John XXIII, lewd and murderous, a simonist and an infidel, was a true comrade for Sigismund in all evil deeds’: Dr. Flajshans, in _Mistr Jan Hus_: quotation translated by Count Lützow, _John Hus and his Times_, pp. 137-8.

[542] Froissart, pp. 330-1.

[543] But not until he ‘regracioit les dieux et les déesses selon la loy où il creoit et que les paiens croient’: Froissart, p. 321. The ignorance among the western chroniclers of everything pertaining to the Osmanlis--or the wider circle of Mohammedan peoples--was appalling.

[544] Schiltberger, p. 5. Cf. Froissart, pp. 322-8; _Relig. de St.-Denis_; _Chronique de Boucicaut_; _Chronique des 4 premiers Valois_, éd. Luce, p. 326; and the other chronicles and secondary authorities given in Bibliography.

[545] Xénopol, in _Hist, générale_, iii. 882, whose writings furnish the most reliable and most accessible data for Rumanian history, allows his patriotism to get the better of his judgement when he writes that this unimportant skirmish was a complete defeat inflicted upon Bayezid, and that ‘le Sultan court jusqu’à Adrinople’! Xénopol makes no attempt to explain the battle of Nicopolis, and Mircea’s action in and after the battle.

[546] Schiltberger, p. 6. Chalc., II, pp. 76-80, who exaggerates the raid to the point of saying that Bayezid reached the environs of Buda.

[547] _Secr. Cons. Rog._, iii. 134-5. _Mém. d’Olivier de la Marche_ (éd. Beaune et d’Arbuthnot), i. 199-200, reads as if Bayezid had actually taken possession of Hungary.

[548] MS. Bibl. Nat., Paris, fonds turc, No. 79, pp. 61 f. (collection of Feridun). For wrong date, see Langlès, in _Acad. des Inscriptions_, iv. 673-4.

[549] Schiltberger, p. 7, who would have been chosen himself for Egypt but for the fact that he had been wounded.

[550] Froissart, p. 341; Rabbi Joseph, p. 254.

[551] Froissart, p. 345. In xvi. 40, Froissart makes a mistake in saying that the body of the Comte d’Eu was ‘en ung sarcus rapporté en France et ensevely en l’eglise Saint-Laurent d’Eu, et là gist moult honnourablement’. The tomb in St. Laurent is merely a memorial. Philippe was buried in the chapel of a monastery in Galata, where, seven years later, Clavijo, fol. 17 vº, saw his burial-spot, but unmarked. His tomb is described by Bulladius, who saw it in 1647, in his notes to Bonn ed. of Ducas, p. 560. Cf. Mordtmann, _Beiträge zur osmanischen Epigraphik_, I, in _ZDMG._ (1911), lxv. 103.

[552] Froissart, xv. 329, 332, 342 f., 355-8; xvi. 16.

[553] Godefroy, _Hist. de Boucicaut_ (1620 ed.), i. 16; Ducas, p. 52.

[554] _Chronique d’Enguerran de Monstrelet_ (ed. Douet d’Areq), i. 332-3; Froissart, xvi. 57-9.

[555] Jean de Nevers, as Duc de Bourgogne and leader of the faction against the king’s brother, openly accepted the responsibility of the assassination of the Duc d’Orléans. This was the beginning of the Burgundy-Armagnac civil war, which delivered France to the English until Jeanne d’Arc appeared to awaken the French to a feeling of nationality.

[556] Froissart, xvi. 47. For ransom, ibid., pp. 37-8, and Rabbi Joseph, i. 254; also _Livre des faicts_ of Boucicaut, _passim_.

[557] Raynaldus, ann. 1364, No. XXVIII. Jireček, _Gesch. der Bulg._, p. 323, says that at this time ‘Osmanen erschienen in Attika’. He has mistaken roving Turkish corsairs of Sarukhan or Aïdin for the Osmanlis. It must have been these Turks who attacked Thebes.

[558] For the deliverance of the grand master of Rhodes, Jean Ferdinand d’Hérédia: Ducange, viii. 296.

[559] _Chron. Breve_ at end of Ducas, p. 516.

[560] Ibid. According to Finlay, iv. 233, he captured Akova. Cf. Muralt, ii. 741, citing Guichenon MS., and Ducange, viii. 39, 296.

[561] Phr., I. 16, p. 62; 26, p. 83; Chalc., II, pp. 67-9.

[562] Muralt, under 1395 and 1397, gives the same expedition. From internal evidence of Byzantine historians, one might put the Morean expeditions in either or both of these years. But cf. _Chron. Breve_, p. 516, and the silence of the Ottoman historians on an expedition in 1395.

[563] Chalc., II, p. 67; Seadeddin, i. 192.

[564] Chalc., II, p. 67; Seadeddin, i. 192.

[565] _Chron. Breve_, p. 516; Phr., I. 16, p. 62; Chalc., II, pp. 97-9.

[566] Seadeddin, i. 193.

[567] The Venetians seized Athens in 1395, and sent Antonio Contareno to act as governor.

[568] Hammer describes the capture of Athens in 1397 in i. 350, and again in 1456 in iii. 51.

[569] Gibbon and Hammer follow Chalcocondylas in this error. Cf. Berger de Xivrey, in _Mém. de l’Acad. des Inscr._, vol. xix, partie 2, pp. 29-30.

[570] Seadeddin, i. 180.

[571] Ducas, 13, p. 50; Chalc., II, p. 59; Idris.

[572] The land walls of Salonika, still standing, are eloquent proof of the difficulty which confronted their assailants before the days of cannon.

[573] Phr., I. 17, p. 64.

[574] See p. 199. There is serious difference of opinion as to just when these concessions were made.

[575] Feridun collection, letter from Adrianople, ordering kadis to prepare for siege of Constantinople: Bibl. Nat., Paris, fonds turc, No. 79, p. 60.

[576] Ducas, 14, p. 53; Canale, ii. 62. Leunclavius, _Annales_, p. 52, puts this in 1391/2.

[577] Karamzin (Russian ed. of 1819), v. 164.

[578] Miklositch-Müller, _Acta Graeca_, DCXXXVI.

[579] Ibid., DXIV, DXV, DXVI.

[580] Froissart, xvi. 132-3.

[581] _Religieux de Saint-Denis_, ii. 559-62, 564.

[582] _Secreta Consilii Rogatorum_, E iii. 138, 146, printed in Ljubić, iv. 404.

[583] Ibid., p. 137.

[584] _Misti_, xliv. 210, xlv. 443; Belgrano, _Arch. Gen._, 1396-1464, pp. 175 f.

[585] Ducas, 14, p. 53; Chalc., II, p. 80; Sherefeddin, iv. 38.

[586] ‘El Cuirol castello de Grecia está despoblado y destruydo y el dela Turquia está poblado’: Seville ed., 1582, fol. 17 v°. Busbecq, i. 131, wrote: ‘stand two castles opposite each other, one in Europe and the other in Asia.... The former was held by the Turks a long time before the attack on Constantinople.’ Busbecq was, of course, misinformed, as Rumeli Hissar was built in 1452. It is still standing in excellent preservation. Anatoli Hissar, of which only one tower remains intact, was built between 1392 and 1397. There is no way of determining the exact date. But Saladin, in _Manuel de l’Art Musulman_,