1. Hasan Umur described about 230 Ottoman documents, of which about 150 are dated tothe period of decentralization (roughly from 1688/1100 to 1834/1250). His first book (1951) is the result of research in the Başbakanlıık Arşiv Dairesi in Istanbul. It transliterates or describes 129 imperial edicts (ferman) that he located in the Trabzon complaints (şikâyat) and judgments (ahkâm) registers, dating from 1575/983 to 1875/1292. He culled these particular documents in the course of a search through hundreds of volumes for any material that related to the district of Of. Umur notes that he chose not to discuss twenty such documents because he felt it inappropriate to publish them. He also set aside many more documents relating to Of because they did not mention village or family names or they seemed uninteresting and uncharacteristic. His second book (1956) describes 100 documents from the Trabzon Şer'i ءcourt registers, which he located in the Topkapıı Saray Museum. Of these, he examined Nos. 1815 to 2024, which covered the years from 1557/965 to 1880/1297. He reports that he examined 209 separate registers comprising about 30,000 pages, culling 100 documents from more than 500 relating to Of. [BACK]

2. Fontanier 1834, 292–94. When travelers report groups of crypto-Christians in Of, they are usually referring to the Kurumlis or other groups who were not resident in Of. Defner, whose assistant visited the district in 1876, does not refer to any crypto-Christians in Of. Among the Oflus, the residents of Ogene [Köknar, Karaçam], located in the upper section of the western valley of the district, have the reputation of having been stubborn holdouts against conversion. However, the Trabzon yearbook (salname) for 1888/1305 locates an official religious academy in this very village, and it is attributed with one of the highest enrollments of religious students in the entire district of Of. [BACK]

3. Brant 1836, 191. The Rizelis and Sürmenelis have themselves been alternatively called Laz, Lazi, Greeks, Byzantines, Armenians (Bijişkyan 1969 [1817–19], 60–64), Çepni Turks (Şakir Şevket 1877/1294, 95–96; Sümer 1992, 83ff.), and Akkoyunlu Turks. [BACK]

4. PRO 526/8 "On the Lazistan Coast . . . ," Jan. 29, 1873, Palgrave. [BACK]

5. Defner 1877. [BACK]

6. Poutouridou (1997–98) cites a passage from the memoirs of Chrysanthos of Trebizond. Also see Asan's translation (1996, 44, 46) of the memoirs of an Oflu Orthodox priest. [BACK]

7. Sümer 1992, 83ff. [BACK]

8. Here the present tense refers to the period before the acceleration of rural to urban migration during the 1960s. [BACK]

9. The Trabzon yearbook (salname) for 1869/1286 records 98 percent of the total population as Muslim (Emiroğlu 1993, 1: 141). Palgrave reports 5,300 Muslim and 83 Orthodox household in Of (PRO 526/8, "On the Lazistan Coast . . . ," Jan 29, 1873, Palgrave). Defner (1877) counts 10,000 to 20,000 Muslim and 150 Christian households. [BACK]

10. Şakir Şevket (1867/1284, 98) dates the conversion of the district to two hundred years after Ottoman incorporation, that is, to about 1655/1065. He mentions that most of the priests accepted Islam. Defner (1877) dates the conversion of the Greeks to 180 years preceding the visit of his assistant, that is, the close of the seventeenth century. [BACK]

11. This situation, the east being Turkish-speaking and the west being Greek-speaking, prevailed during the visit of Defner's assistant in 1876 (Defner 1877). [BACK]

12. Günay (1978, 28–29) describes ııkizdere, adjacent to the eastern valley-system of Of, as the focus of the dominant Turkic dialect in the province and proposes that this was an area of early Turkic settlement. Otherwise, the eastern valley-system may have consistently received settlers from Anatolia since ancient times (Bryer and Winfield 1985, 11, 54-55). [BACK]

13. Bryer and Winfield (1985, 11, 55) do not believe that the contemporary road that connects Of with Bayburt, which was constructed during the Russian occupation (1916–18), was previously a trade route of any importance. In Of and Sürmene, it is asserted that it was previously a trade route. [BACK]

14. When I asked individuals from leading families about their family history, they commonly told me stories of a migration from the interior highlands, or even from as far away as Iraq or Syria. For example, one of the Muradoğlu said the founder of his family came from Van during the time of Mahmut II, while his two brothers settled in Hopa and Vakfııkebir. One of the Tellioğlu said his ascendants came from Baghdad out of Arabistan, one branch settling in Of and another branch settling in Giresun. A Selimoğlu says that the family came from Tercan (about halfway between Erzincan and Erzurum). [BACK]

15. Brendemoen (1987, 1990) has not discovered traces of a Çepni influence on coastal speech patterns east of Trabzon, although such an influence is clearly apparent west of Trabzon. However, like Günay (1978, 26), he says that the Turkish dialect in Rize is similar to the Turkish dialect in Erzurum. My Oflu contacts are also in agreement with this opinion. They say that the Turkish spoken in the eastern half of their district (which borders on Rize) is similar to the Turkish spoken in Erzurum. [BACK]

16. See, for example, the comments of Umur (1949, 89–90). The dialect of Pontic Greek spoken by the Oflus is apparently indigenous to the valley (Defner 1877). [BACK]

18. Brendemoen (1990, 49, 57) also notes that the same archaic Turkish usages are also found in the district of Tonya, another area where Muslim Greek-speakers were common. [BACK]

19. Asan 1996 and Poutouridou 1997-98. [BACK]

20. The three registers have not yet been exhaustively studied. See Gökbilgin (1962) for some general conclusions about the registers, Lowry (1977) for a study of that portion relating to the town of Trabzon, Umur (1951) for some of the details regarding the district of Of, and Bilgin (1990) for some of the details regarding the district of Sürmene. [BACK]

21. Umur (1951) records the total number of households for each village locale and the number of households that were recorded as Muslim rather than non-Muslim. I have grouped the villages according to their location in a particular sector of the district. In those instances where I was unable to determine the location of a village, the household total is listed under "unlocated." These tabulations of Umur's data were first published in Meeker 1971. [BACK]

22. The increase is about 15 percent from register to register. [BACK]

23. Umur (1951, 20) reached a similar conclusion. The registers demonstrated to him that the lower reaches of the Baltacıı and Solaklıı rivers were being Turkicized and Islamized during the sixteenth century. He also acknowledged that Orthodox Greeks would have been converting to Islam and therefore joining the new Turkic settlers. In mentioning this he asked his readers to understand that no nation is completely pure but all are composed of a mixture of peoples. [BACK]

24. Günay 1978, 26. [BACK]

25. The name ııskender, indicating a convert or a child of a convert (Lowry 1977), commonly occurs in the lists of Muslim residents of the villages of Of in each of the three registers, increasingly in the second and third. Since Umur gives only a few examples of the names for each village, the frequency of the name ııskender cannot be determined from his study. [BACK]

26. Lowry 1977. [BACK]

27. Lowry (1977, 243–44) found that the population of the town of Trabzon, still largely Christian at the time of the Ottoman register of 1554/961, was more than 50 percent Muslim by 1583/991, and yet it remained about 70 percent Greek-speaking. [BACK]

28. Since Rize was more like Trabzon than Of in this respect, it is possible that immigration and conversion were also already important there during the sixteenth century. [BACK]

29. Poutouridou (1997–98, 57) has concluded that church organization in the district of Of probably collapsed sometime during the second century of Ottoman rule. Citing Vryonis and Bryer, she notes that the bishopric of Of was one of three remaining in Anatolia during the late fifteenth century, but had disappeared from the episcopal lists of the Patriarchate of Constantinople by 1645. [BACK]

30. Umur 1951, No. 3 1615/1024. This document describes the desperate situation of the villagers in the district who may be forced by famine to abandon their villages. The writer observes that the Christian villagers were suffering even more because they were obliged to pay the special tax on non-Muslims, the haraç. Citing this document, Umur (1951, 20–24) concluded that the district would have been entirely Islamized and Turkicized at this date as a consequence of conversion or desertion of Christian villagers. However, the document only indicates the existence of hardship rather than actual desertions or conversions. [BACK]

31. Umur 1956, No. 8 1631/1040 refers to 441 cizye-paying households of zimmis in Of; No. 25 1673/1083 refers to 90 such households. The cizye was a head tax on non-Muslims. The term zimmi refers to a non-Muslim subject. Since it is possible that the counts are partial rather than total, they can only be taken as the minimum number of Christian households remaining. [BACK]

32. Bryer 1988. Cf. Janin 1912. [BACK]

33. Umur 1956, No. 38 1695/1106. The numbers for the districts add up to 7,650, not 7,700. [BACK]

34. According to Uğur (1986, lx, lxxvi), the Ottoman religious academies were already in decline by the seventeenth century, troubled by "nepotism, simony, and favoritism." [BACK]

35. Umur 1951, No. 20 1699/1110. [BACK]

36. The relevant passage (1787, 70–71) is as follows: "il y a sur-tout un nombre infini de Gens de Loi renommés pour leur érudition." Peysonnel (fils) did not publish his treatise on commerce in the Black Sea until the year 1787; however, he explains in his preface that his information was based largely on material collected from Turkish, Armenian, and Greek businessmen in Izmir during 1750, supplemented by more information gathered when he served as a consul in the Crimea after 1753, then finally completed in 1762 while a resident of the kingdom of Candie. Acknowledging that the situation in the Black Sea had changed, he affirms that he has presented his information exactly as he gathered it years before. For more on the Peysonnels, père and fils, both of whom served as consul in Izmir, see Veinstein 1975. [BACK]

37. Other, more indirect evidence is consistent with the existence of a learned tradition in the district of Of from the early eighteenth century. Oflu Bilal Efendi of Paçan [Maraşlıı] village is known as the "first poet of Of." Although his precise birth and death dates are unknown, he is thought to have lived in the first half of the eighteenth century, and one of his works is published in 1764. Hasan Umur (1951) transliterated a number of eighteenth-century court documents referring to disputes over the right to act as an official imam (imam, hatip) in this or that village in the district of Of (No. 22 1702/1114, No. 44 1719/1131, No. 73 1733/1146, No. 93 1770/1184, No. 94 1770/1184, No. 101 1780/1194, No. 109 1789/1203). Architectural evidence of religious buildings is fragmentary, since they were built of wood, and so most would have been reconstructed over the years. When a state offi-cial organized local brigands to raid the western valley of Of in 1711/1123, they are also said to have burned seven mosques in fifteen villages (Umur 1951, No. 35). A wooden mosque in Sürmene has the date 1785/1200 on its door (Bilgin 1990, 673). According to Umur (1956, 16–17), the death of the first preacher (hatip) in Yiğa [Yarlıı] village is dated to 1798/1212, as deduced from a document dated in 1836/1251. Umur estimates from this (by the average life of a man) that the mosque was built 45 years before, that is, in 1737/1150. According to Karpuz (1989), there is a date of 1767/1181 on the door of the mosque in Alanomakot [Ağaçlıı] village in the district of Of. He also mentions an eighteenth-century mosque in Çaykara, rebuilt in 1809. [BACK]

38. The descendants of Greek-speaking Christians appear to have made an important contribution to the tradition of religious study in these villages. The Trabzon yearbook (salname) for 1888/1305 (pp. 127–30, 313–16) states that lessons in the religious academies of Of were sometimes given in Greek, although most of the population in the district was Turkish-speaking. In the 1960s, by the reports of my interlocutors, much of the population in the district of Çaykara still spoke Pontic Greek in their families and villages. Although the male population always spoke Turkish (many of the women did not), mosque sermons and religious instruction were said to be sometimes conducted in Pontic Greek. Also see Asan (1996, 121) and Poutouridou (1997–98, 62) for further documentation of the use of Pontic Greek by Muslim teachers and preachers. Paçan [Maraşlıı] village, one of the most famous centers of religious study in the district and the site of the attack on teachers and students in 1737/1150, was among the Greek-speaking villages. [BACK]

39. Orthodox Greek writers of the nineteenth century date this event to the mid- or late seventeenth century (Poutouridou 1997-98). [BACK]

40. Umur (1951, 21-22), who knew the Ottoman documents relating to this question better than anyone, reaches this conclusion. So far as I am aware, there is no mention of mass conversion in any source earlier than the nineteenth century. Some such sources are of Muslim and some are of Christian origin (see chap. 8). [BACK]

41. There is convincing evidence of a flight of Christians from Trabzon to Canıık during the late seventeenth century. Biliotti encountered groups of Christian Armenians and Greeks in the latter province who had migrated there from Hemşin and Gümüşhane during the first part of the eighteenth century (PRO FO 195/1329, at Fatsa, No. 30, Aug. 1880; at Ünye, No. 32 and No. 33, Aug. 1880; at Çarşamba, No. 38 and No. 45, Oct. 1880; at Görele, No. 48, Oct. 1880). Janin (1912, 497–98) writes, without attribution, that the Christians of Rize and Of either fled to the Crimea and Moldavia or converted sometime around 1665. Also see Bryer (1970). [BACK]

42. Reporting on his travels during the spring of 1835, James Brant (1836, 192) wrote of the district of Of: "The people are a hardy laborious and bold race, they are skilled in the use of a short rifle, which every man carries slung at his back wherever and on whatever occasion he moves, and they enjoy a high reputation as Soldiers. A demand is always made on this country by the Porte, to supply a certain number of men for the Arsenal at Constantinople." [BACK]

43. More and more janissary appointments were improperly granted by bribes during the eighteenth century (Özkaya 1977, 52–53). Janissary appointments were bought and sold like mortgage contracts (Bilgin 1990, 277). [BACK]

44. For an early example, see Umur 1951, No. 79 1740/1153 and No. 81 1745/1158, which reports on a charge of murder where the victim is named as "dergâhıı muallâm yeniçerileri serdengeçti ağaleri emektarlerinden Piri Çavuş zade Mustafa nam ihtiyar kendi halinde iken Of kazasıı sakinlerinden." For a late example, see Umur 1956, No. 75 1843/1259, which reports on a charge of land theft where the plaintiff is named as "Of kazasıında Baltacıı deresinde vaki Konu karyesinden Tophane-i amire çavuşlarıından Ömer Çavuş bin Mehmed." [BACK]

45. Umur 1951, No. 20 1699/1110. Also see the analysis of the learned class (ulema) in the district of Of in chap. 8. [BACK]

46. At the time of my fieldwork, I had considered Umur's studies to be useless because it was impossible to establish his criteria for selecting documents. When reconsidering his books in the 1980s, I was surprised to discover that his documents regarding Of referred to events that correlated closely with the latest studies of provincial social history in the core Ottoman provinces. This is evidence of the reliability of his survey of documents since he would have been unaware of the results of the later studies. [BACK]

47. Şakir Şevket (1877/1294, p. 249ff.) comments that the ascendants of the Şatııroğlu family of Trabzon (prominent during the period of decentralization) first came to Trabzon at the time of the Ottoman conquest. Goloğlu (1975, xxxxi [sic]) writes that the ascendant of the Selimoğlu family of Of is believed to have come to Trabzon as an akııncıı (light cavalry raider) during the second half of the sixteenth century. According to Bryer (1970, 45), the names of the district and provincial lords (derebey) in the coastal region cannot be traced earlier than the seventeenth century. However, this may not be significant since it would not be possible to trace names until a family line had been established, that is, after the period of decentralization had already begun. [BACK]

48. The Çapoğlu and the Ayazoğlu are said to have set down family lines very early. The Muradoğlu most certainly set down a family line very late. [BACK]

49. Umur 1951, No. 9 1689/1100, No. 10 1689/1100, and No. 12 1689/1100. The documents all address the same incidents, which occurred in 1679/1090. [BACK]

50. One of my interlocutors, familiar with Umur's work, claimed that the Çap-oğlu were the first aghas and agha-families in the district of Of, the Ayazoğlu the second. [BACK]

51. Umur 1951, No. 29 1708/1120 and No. 32 1710/1122. [BACK]

52. Umur 1951, No. 81 1745/1158, No. 85 1752/1165, No. 86 1753/1166, No. 87 1763/1176, and No. 88 1764/1178 refers to the assaults on members of the Piri Çavuşoğlu family. Umur 1951, No. 83 1748/1161 refers to the assaults on the religious teachers and students (ulema ve talebe) in Paçan village. All six documents refer to events that took place in 1737/1150. [BACK]

53. Other patronymics listed in the documents are identical to family lines that were closely allied with the aghas and agha-families (e.g., Kalyoncu). Also see Umur 1951, No. 70 1732/1144 and No. 74 1735/1147, where incidents of brigandage are attributed to individuals with other patronymics, such as Ayazoğlu, Nuhoğlu, and Bektaşoğlu, patronyms associated with the Five and Twenty-five parties in the contemporary period. [BACK]

54. The individuals with patronymics are among those accused as principals during the incidents. Given the character of the crimes of which they are accused, the combination of patronymics, and the composition of their followers, it is unlikely that they were renegades of their family lines. One of the accused, Hasan Agha Fettahoğlu, is also mentioned in the "Fettahoğullarıınıın Tarihi." The writer of that document accepts that this man was a prominent ascendant and situates him in the family genealogy. However, see the following note. [BACK]

55. The author of the "Fettahoğullarıınıın Tarihi" claims that an enemy of his family, whom he names as "a hodja from the Kaltabanoğullarıı," assembled a group of false witnesses to bring trumped up charges against his ascendant. [BACK]

56. Umur 1951, No. 85. [BACK]

57. Peysonnel 1787, 73. [BACK]

58. Bryer (1970, 44) observes that the Twenty-fifth Regiment of Janissaries was headquartered in Trabzon. So it is possible that the Twenty-five Party was actually close to the level of a janissary regiment, while the Five Party was a regiment of private soldiers. Bilgin (1990, 312) associates the Five Party with the "Cemaat-iBeşluyan," janissaries appointed to the castle force or town police of Trabzon, and the Twenty-five Party with the "Yermibeşli," janissaries who were granted a tıımar and assigned to the Twenty-fifth Regiment. Also see the discussion of the troop call-outs later in the text. [BACK]

59. The Hacııhasanoğlu and the Keleşoğlu are associated with the Fifth Regiment of Janissaries and then later, consistent with this, the Five Party. [BACK]

60. See chaps. 6 and 7 for further discussion of this issue. [BACK]

61. Umur 1951, No. 35 1711/1123. The villages from which the brigands were recruited are located all along the lower and higher reaches of the western valley-system, that is, the Greek-speaking areas. [BACK]

62. Barkey (1994) has recently called attention to these kinds of incidents, which were especially common during the seventeenth century. At that time, they were a direct result of the need for a large reservoir of officers and troops on the occasion of imperial campaigns. With demobilization, officers who had no appointments and troops without employment were obliged to extract their living by raiding and looting villages. When their activities became intolerable, state officials would take steps to suppress them, providing they had the necessary troops and resources for doing so. [BACK]

63. The other three documents involve the following incidents: A man who had served as the district governor of Of (mütesellim) two years previously is accused of making war on the people of Of with the excuse that he holds an imperial decree (ferman) permitting him to do so (Umur 1951, No. 7 1689/1100). A man who had served as the district governor of Of is trying to recover a debt from certain Oflus (ibid., No. 6 1688/1099). A janissary and his associates are accused of attacking villages and committing all kinds of injustices (Umur 1951, No. 4 1665/1076). [BACK]

64. Umur 1951, No. 43 1709/1121 [date corrected in Umur 1956, 13, n. 1]; Umur 1956, No. 45 1709/1121, No. 46 1710/1121, andNo. 47 1710/1122. The seven villages are located in the western valley-system. They include some of the same villages from which the head steward had recruited soldiers for raiding villages in Of; see Umur 1951, No. 35 1711/1123. The deportation did not alter the prevalence of Greek as the language spoken in this part of the district. It is possible that not all the villagers were taken and removed, that some eventually returned from exile as was common in such cases, or that these seven villages were resettled by people from neighboring villages. [BACK]

65. Umur 1956, 57. [BACK]

66. Of 120 documents relating to incidents occurring between 1666/1076 and 1748/1161, about 40 refer to various kinds of social disorders that involve a challenge to the authority of district state officials (Umur 1951, 1955). About a third of the 40 cite individuals bearing patronyms that are the same as some of the agha-families that came to make up the Five and Twenty-five parties during the early nineteenth century. The 40 do not include about a dozen other documents that refer to "simple" incidents of brigandage involving various acts of destruction or spoliation. It should be remembered, however, that Umur specifically transliterated documents that mentioned well-known families in Of. [BACK]

67. Umur 1951, No. 26 1705/1117, No. 29 1708/1120, No. 30 1708/1120, No. 65 1730/1142, No. 72 1732/1144, No. 78 1740/1153, No. 79 1740/1153, No. 81 1745/1158, No. 85 1752/1165, No. 86 1753/1166, No. 87 1762/1176, and No. 88 1764/1178. [BACK]

68. Ibid., No. 52 1724/1136. [BACK]

69. Ibid., No. 21 1701/1113, No. 29 1708/1120, No. 30 1708/1120, No. 39 1716/1128, No. 42 1717/1129, and No. 56 1726/1138. [BACK]

70. Ibid., No. 55 1725/1137. [BACK]

71. Ibid., No. 54 1725/1137. [BACK]

72. Umur 1956, No. 48 1723/1135. [BACK]

73. Umur 1951, No. 72 1732/1144. [BACK]

74. Umur 1956, No. 71 1774/1188. The nine districts are Trabzon, Rize, Of, Sürmene, Pulathane, Vakfııkebir, Tirebolu, Keşap, and Giresun. [BACK]

75. Ibid., No. 65. Also see Sümer (1992, 104–5). See Aksan (1999a, 1999b) for a general study of military recruitment in the Ottoman Empire at this time. [BACK]

76. Two are personal names without patronymics, as in "Mehmet Bey of Viçe [Fıındııklıı]" and "Mamoli Mustafa of Hopa." [BACK]

77. Some of these correspondences that are known to me are as follows: Hacıı Salihoğlu at Tonya, Kalcııoğlu at Trabzon, Hacııhasanoğlu at Pulathane, Eyyuboğlu at Maçka, Tuzcuoğlu at Rize, Ekşioğlu at Rize, Canoğlu at Of, Kııraçoğlu at Of, and Selimoğlu at Of. [BACK]

78. Cevdet Asker 40224, dated 1789/1204. I am grateful to Mehmet Bilgin for providing me with a photocopy of this document. [BACK]

79. The fifth section, devoted to the western districts, does not give the names of individuals. The places and quotas listed in the fifth section are as follows: Amasya (1,500), Ünye and Niksar (500), Tirebolu (500), Giresun (200), Karahisar Şarkıı (3,000), and Canıık (5,000). [BACK]

80. The document mentions "Hüseyin Agha Selimoğlu of the 5th aghas." It does not mention an agha from the Muradoğlu in agreement with the family tradition that its founder arrived in Of sometime after 1800. [BACK]

81. Bryer 1975 and Lowry 1977. Also cf. Poutouridou 1997–98 and Vryonis 1968. [BACK]