1. İİnalcıık (1973, 29) regards Mehmet II as the true founder of the Ottoman Empire. [BACK]

2. Machiavelli (1992), in manuscript from 1505, in print from 1515. [BACK]

3. The slave official that appears in western European commentary can be considered a rhetorical figure. He is a person whose being was reduced to absolute obedience by separation from his place and family of origin. Ottoman officials of slave status (kul), however, might parade their ethnic origins, dressing as Circassians or Bosnians, forming factions with those of the same ethnicity in the palace and speaking among themselves in their mother languages (Kunt 1974). Despite these complications, the rhetorical figure of the slave official does accurately indicate the strategy of Ottoman recruitment and training. [BACK]

4. Barkey (1994) points out that the Ottomans regularly adjusted the size of their armies by either expanding or contracting military appointments and prerogatives. Even during the classical period, the high point of imperial centralism and exclusivity, they added military officials and formations when they required them, then shed them once they no longer needed them. The increasing numbers of soldiers and preachers were at least in part the intended result of state policies. [BACK]

5. İİnalcıık 1977. [BACK]

6. Ibid.; Nagata 1976; Özkaya 1977; Sakaoğlu 1984; and Veinstein 1975. These authors are not in agreement on the timing of the period of decentralization, setting its beginning variously between the middle of the seventeenth and the middle of the eighteenth century. [BACK]

7. ıınalcıık 1973, 103-18. [BACK]

8. Only a fraction of these military officers would have been raised and trained in the palace. [BACK]

9. For a more detailed summary of this complex system, see Barkey (1994, chap. 3). [BACK]

10. Barkey (1994, chap. 4) describes the position of the tıımar holder in relationship to his charges during the classical period. [BACK]

11. Bilgin (1990, 240–46) has examined fifteenth- and sixteenth-century tıımar lists for Trabzon. These lists include the name of sipahi to whom a tıımar was assigned, the reason it was granted, and, more exceptionally, the reason it was revoked. Bilgin gives some examples of complaints against sipahi that led to revocation of their appointments, such as 1) taking a wife without proper registration of marriage, 2) killing a man, 3) getting drunk and drawing a weapon on another sipahi, 4) insulting the sultan, and 5) a complaint lodged by an ordinary individual (reaya). [BACK]

12. The dates were 1486/892, 1515/921, 1554/961, and 1583/991. Three of the registers are discussed in more detail in the next chapter. [BACK]

13. Bryer (1975, 132–33) notes that Çepni beys were given tıımar in Trabzon after its incorporation. [BACK]

14. Of 207 tıımar holders in Trabzon, as listed in a document dating from 1486, Bilgin (n.d. a) believes 20 were of Albanian origin. Elsewhere, citing Beldiceanu, Bilgin (1990, 136, 145) points to evidence of tıımar holders of Albanian, Bosnian, Serbian, and Hungarian background. [BACK]

15. The villagers would have largely been Lazi-, Greek-, and Armenian-speakers who were Christians. [BACK]

16. The rural societies of the coastal region had taken part in the military and religious institutions of the Greek Empire of Trebizond (Bryer 1975). The defense system of the latter had extended into the upper and outlying coastal valleys, requiring the support of the local residents, Greeks and non-Greeks, Christians and Muslims. Orthodox churches, monasteries, and endowments were also dispersed through the coastal valleys, and their staffs and tenants were drawn from a cross-section of the population. [BACK]

17. The Ottoman mosque complex and the palace complex were designed to "fashion" both a Muslim state personnel and Muslim state subjects where there had been few or none before. In this respect, they were without precedent among other Islamic dynasties (ıınalcıık 1973; Necipoğlu 1991). [BACK]

18. Necipoğlu 1991, 21. [BACK]

19. I use the term "ethical" advisedly to emphasize the interpersonal character of the ceremony, a quality that linked it with Islam. This is not to say that court ritual was essentially Islamic, even if the Ottomans might wish to claim it was so. [BACK]

20. The wall was not built to serve a military purpose, its watch towers and gun emplacements being of symbolic rather than practical significance (Necipoğlu 1991). [BACK]

21. Bentham's panopticon (1787) consists of a central point of observation surrounded by, but invisible to, a circle of isolated cells. The individuals in each of the cells have no contact with their neighbors but are exposed to surveillance by the central point of observation. Foucault (1975) explains the panopticon as an architectural arrangement designed to instill an individualized discipline of behavior. As such, it was a model that could be used for the construction of prisons, barracks, schools, factories, or hospitals. [BACK]

22. Necipoğlu 1991, 85. The verse is from Cafer Çelebi's "Heves-Name": "Nedür her câm bir çeşm-i cihan-bîn / Temaşa itmeğe tertib ü âyîn. Nedür her küngüre ser-tâ kadem dil / K'ider medh ü senâ-yi şâh-ıı âdil" (Levend 1958, 72–73). Necipoğlu's translation appears to me to be an excellent rendering. A more literal, hence inelegant, translation would be, "What is every window but an eye on the world / for observing ordered ceremonies. What is every tower but from head to foot a tongue / for fulsome praise of the just shah." [BACK]

23. The figure illustrates the ethical underpinnings of architecture and ceremony and is not otherwise an accurate representation of the layout of the palace complex. [BACK]

24. The Ottoman sultans were represented at the portal of each gate by inscriptions, military guards, and victory banners. Pierce (1993) mentions the symbolism of inside and outside in the palace, as well as its incompatibility with contemporary concepts of public and private. Lewis (1988), cited by Pierce, observes that the language of power in Islamic societies, as opposed to the courts of western Europe, turns on spatial separations that are horizontal rather than vertical. [BACK]

25. Necipoğlu (1991, 52, 54) confirms that the middle court was in existence early in the reign of Mehmet II. She also notes that its layout was identical to the administrative enclaves of the tent palaces of the Ottoman sultans (ibid., 53–54). Necipoğlu (1991, 84–85) concludes that the Tower of Justice was probably an early feature of Mehmet II's palace complex. She cites Cafer Çelebi's verses as evidence of this possibility and notes that royal buildings built before the New Imperial Palace featured similar structures. [BACK]

26. The Council Hall and public treasury were located a short distance from the Tower of Justice during the reign of Mehmet II. They were moved to the base of the tower early in the reign of Süleyman I, sometime between 1525 and 1529 (ibid., 23, 79–80). For simplification, I am anachronistically referring to the latter arrangement, which Necipoğlu views as a refinement of the architecture and ceremony of the middle court. [BACK]

27. Necipoğlu 1991, 56–58. In the time of Mehmet II, "Any male or female subject of the sultan, Muslim or non-Muslim, could petition the high court of justice to have his case heard and decided" (ibid., 76). [BACK]

28. Such an aperture, or window, was part of the old Council Hall of Mehmet II, as well as of the new Council Hall of Süleyman I (ibid., 79, 83). [BACK]

29. Ibid., 59, 62 (fig. 41), 63 (fig. 42), 80, 86. [BACK]

30. Ibid., 80. [BACK]

31. Observers report that Mehmet II regularly appeared before his soldiers to assure them that "he was still alive and that they were not threatened by a usurper" (ibid., 18). Such an explanation would be relevant to any dynast who removed himself from his officials and soldiers. Otherwise, it does not explain the elaborate symbolism devoted to the personal presence of the sovereign in the palace complex. [BACK]

32. Iacopo de Campis Promontorio, a Genoese merchant who served in the Ottoman court from 1430 to 1475, described the ceremony as it was held in 1475 (ibid., xii-xiii, 18–19). This was before the adoption of a new codification (kanunname) of court ceremony and protocol. [BACK]

33. By this observation, I do not mean that the device of a windowed overlook was absent from earlier dynastic structures, whether buildings or tents. I mean to say that the windowed overlook became a fundamental principle of the ruling institution during the time of Mehmet II. [BACK]

34. Kafadar (1995) examines the policies and followings of the early Ottoman dynasts. [BACK]

35. See, for example, the depiction of more or less Islamized pastoral nomadic chiefs and tribes in The Book of Dede Korkut (Lewis 1974). The Dede Korkut stories have their origins among the thirteenth- to fourteenth-century Oghuz peoples of northeastern Anatolia. Meeker (1992) analyzes these stories as ethical representations of self and society. [BACK]

36. Ibid. [BACK]

37. Necipoğlu (1991, 19) refers to the commentaries of Aşııkpaşazade and Mihailovic, contemporary observers of the classical period who were of the opinion that the Ottoman sultan had personally appeared at communal banquets from the time of Osman, founder of the dynasty. [BACK]

38. Ibid., 69. [BACK]

39. Ibid., 19. [BACK]

40. Ibid., 61. [BACK]

41. Ibid., 68. [BACK]

42. Freud 1958 [1919]. [BACK]

43. During the earlier classical period, the sovereign might also observe executions of delinquent officials in the middle court below. In doing so, he opened the lattice to signal his presence while the condemned entreated his forgiveness (Necipoğlu 1991, 59). [BACK]

44. Necipoğlu 1991, 64–65, Necipoğlu's translation. The original account is in Italian. I have consulted the French edition (Fresne-Canaye 1980 [1897], 62, 64). [BACK]

45. Ibid., 65. [BACK]

46. Ibid., 68. [BACK]

47. Foucault 1975. [BACK]

48. Cf. Necipoğlu 1991, 90. [BACK]

49. Cf. Pierce 1993, 39-45. [BACK]

50. Necipoğlu 1991, 98. [BACK]

51. Ibid., 90, 102 (quoting Miller). [BACK]

52. Fresne-Canaye described the appearance of the sovereign as he was presented to him. "He did not look at us in the face, but with a troubled eye, mean and alarming, he held his head turned toward the fireplace, as though not really aware of those who came before him so humbly" (1980 [1897], 70). [BACK]

53. See the plates representing the visits of European ambassadors in Necipoğlu (1991, 103–5). Fresne-Canaye described his ambassador being presented to the sovereign: "When the Ambassador had arrived at the door of the [Petition Room] where the Grand Turk awaited him, two aghas, superbly attired, took him by his two hands, and as soon as he had kissed the garment of the emperor, he was taken to a corner of the room" (1980 [1897], 68-69). [BACK]

54. Necipoğlu 1991, 107–8. Fresne-Canaye writes, "All around the [Petition Room] there were hidden I don't know how many mutes, among whom are found the most faithful and proven executors of the atrocious orders of this tyrant" (1980 [1897], 70). [BACK]

55. On the occasion of the religious holidays food was served from the kitchens, but apparently not while the sovereign appeared in the middle court. Uzunçarşıılıı (1984, 209) describes the celebration of religious holidays in the middle court during the early seventeenth century. [BACK]

56. Cf. Necipoğlu 1991, 50. [BACK]

57. Ibid. Before the completion of the surrounding wall, the palace complex consisted only of two gates and two courts. [BACK]

58. Ibid., 32, 50–51. The exterior of the middle gate, as a segment of a fortress wall, could be regarded as a synecdoche of the surrounding fortress within which it was situated, making it a symbol of a symbol of sovereign power. [BACK]

59. The two towers flanking the middle gate included a prison for those on trial. The court martials were held regularly following the ceremonies in the middle court (ibid., 76; Uzunçarşıılıı 1984, 21). [BACK]

60. Necipoğlu 1991, 53. [BACK]

61. Goodwin 1971, 132-3. [BACK]

62. Necipoğlu (1991, 72) affirms the importance of the chimneys as visible architectural symbols. Both the Tower of Justice and Imperial Kitchens had been built by Mehmet II but were later either remodeled or replaced several times, during the classical period itself and also afterwards. But despite successive changes, they remained prominent architectural features of the palace complex. For early representations of the Imperial Kitchens, see ibid., 70–73, 84–86, plates 30a–b, 31a–c, and 32a-b. [BACK]

63. Ibid., 71-72. [BACK]

64. My reference for this is late. Ferrières-Sauveboeuf (1790, 220) noted that the worst thing that could happen to a unit of the janissaries was to suffer the dishonor of losing their cauldrons. They therefore assigned two "batteries of the kitchen" to prevent such a disaster. [BACK]

65. Necipoğlu 1991, 72. [BACK]

66. Indeed, the ethical underpinnings of the old dynastic court tradition are still to be found in the Turkish Republic, having been once again disarticulated and rearticulated. Meeker (1997) analyzes Atatürk's tomb and the Kocatepe Mosque as national monuments. [BACK]

67. Returning military expeditions did sometimes enter the outer gate and parade in the outer court (Necipoğlu 1991, plates 33a-q). [BACK]

68. Ibid., 24 (fig. 12). Fresne-Canaye (1980 [1897], 128) observed, "In this company and order, [Selim II] went off to the mosque of Sultan Beyazit, where an immense crowd had gathered, and he remained there a little more than an hour, then returned with the same people by the same procession." [BACK]

69. One of my readers has criticized the use of the word "performance" in this paragraph as indicating that the ceremonies of the court and mosque were but superficial rituals, lacking in psychological depth. It is not my intention to suggest that this is so. I translate namaz kıılmak as "perform the ablutions and prayers." I assume that these rituals are part of a range of rigorous disciplines that served to forge thought and behavior. For analysis of the discipline of religious belief and practice in terms of its power to create a psychological reality among contemporary believers in Cairo, see Mahmood (forthcoming). The example also serves to point out that discipline of religious belief and practice in the Islamic tradition can be used for various ends. [BACK]

70. Fresne-Canaye 1980, 69. [BACK]

71. Ibid., 127. [BACK]

72. Necipoğlu 1991, 34, 36. [BACK]

73. Necipoğlu 1991, 36. The Koranic citation is 15: 45-48. [BACK]

74. Arberry's (1955) translation. [BACK]

75. Unlike their Seljuk predecessors, the "countrified" Ottomans lacked a distinguished genealogical heritage. They therefore placed a special emphasis on sultanic sponsorship and support of the learned class of Islam, and, more specifically, ruling in conformity with the sacred law of Sunni Islam (Zilfi 1988, 23–24, 27-28). [BACK]

76. Kuran 1968. [BACK]

77. Ibid., 198. [BACK]

78. See Meeker (1997) on the nationalist features of contemporary mosques built in the classical Ottoman style. [BACK]

79. See ıınalcıık (1973, chap. 15), for an account of the role of the mosque complex in the formation of Muslim towns and cities. See Ergin (1939) for an account of the activities that took place in the mosque complex. See Ayverdi (1973, 356–406) for an account of the buildings and endowments of Mehmet II's mosque complex. [BACK]

80. Kunt 1974. [BACK]

81. ıınalcıık 1977, 31; Sakaoğlu 1984, 10; Özkaya 1977, 67, 99; and Veinstein 1975, 1991. The local elites varied in their sociological character. This has led to disagreements among historians about the causes that brought them to prominence, the exact periods when they were dominant, and their most important social and political characteristics. Cf. Gould 1976; Nagata 1976; Skiotis 1971; and Uzunçarsıılıı 1975. [BACK]

82. Akarlıı 1988; Özkaya 1977, 8, 24–27, 67–68, 98, 111; Sakaoğlu 1984, 10; and Veinstein 1975, 1991. [BACK]

83. According to Sakaoğlu (1984, 5), the term "valley lord" (derebey) is a corruption of the phrase "recognized lord" (derre bey). [BACK]

84. Akarlıı (1988) discusses the rise of provincial magnates through the accumulation of agricultural lands. Sakaoğlu (1984, 10) remarks that Ottoman officials referred to them as simple landowners (çiftlik ağasıı), but they had in fact become provincial lords and rulers. See, however, Veinstein (1991) on the question of the extent to which the rise of local elites in the eighteenth century was linked with the control of land. [BACK]

85. See Fontanier (1829, 13, 22) for coffeehouses in Sürmene and in Trabzon; Brant (1836) for comments on coffeehouses in the markets at Sürmene, Rize, Atine, Hopa, and Batum; Guarracino (1845) for coffeehouses from Batum to Artvin; Koch (1846, 3) for comments on a coffeehouse at Sürmene that was similar to those in Istanbul; and Decourdemanche (1874, 358, 360) for coffeehouses as places for travelers to stay and places for association. Also see Umur (1956, No. 89 1859/1275), which shows that the estate of Memiş Agha Muradoğlu of Of included a coffeehouse. [BACK]

86. This claim is certainly true for the major mosques in the district of Of. I am assuming that the pattern holds for all the province of Trabzon. [BACK]

87. Fontanier 1829, 17–18, cited in chap. 1. [BACK]

88. Fontanier 1834, 39. [BACK]

89. In this chapter, Fontanier is recounting the revolt against the government in the district of Sürmene in 1832. [BACK]

90. Fontanier 1834, 321. [BACK]