¹ The fullest account of the life and works of Ahmed b. Suleymān b. Kemāl Paaa (b. 873/1468–9, d. 940/1534), generally known as Ibn Kemāl or Kemālpaazāde, is the article by İsmet Parmaksizoǧlu in Islam ansiklopedisi, fasc. 62, s.v. ‘Kemāl Paşa-zâde’. The MSS of Kpz.'s ‘History’ in European libraries are listed in Fr. Babinger's GOW, 61 ff.; the scope of the ‘History’ seemed then to be A.H. 886–933. The Istanbul MSS are described by Forrer, L., Der Islam, XXVI, 1942, 184–7Google Scholar, and in Istanbul kütüpaneleri tarih-coǧrafya yazmalari kataloglai, 1, 2, 1944, 120 ff.Google Scholar The first eight books were written at the command of Bāyezīd II, the eighth reaching to the year A.H. 916; the ninth and tenth books were written for Suleymān, the ninth beginning with the events of the last two years of Bāyezīd II's reign. The text published by de Courteille, Pavet as Histoire de la campagne de Mohacz par Kemal Pachazadeh (Paris, 1859)Google Scholar has now been shown to be not a separate work but the tenth book of the ‘History’ (cf. Forrer, L., loc. cit.Google Scholar, and Yurdaydin, Hüseyin G., ‘Kemal Paşazâde'nin Tevârih-i Âl-i Osman'mm onuncu cildi hakkinda’, Vakiflar Dergisi, III, 1956, 107–15).Google Scholar

² Fac. = Şerafettin Turan, İbn Kemal: Tevârih-i Âl-i Osman, VII. defter (TTK, I. Seri, No. 5), Ankara, 1954.

Tr. = Ṣerafettin Turan, İbn Kemal: Tevārih-i Âl-i Osman, VII. defter (tenkidli transkripsiyon) (TTK, III. Seri, No. 5), Ankara, 1957; in a long introduction Dr. Turan reviews Kpz.'s life (making some corrections to İ. Parmaksizoǧlu's conclusions) and discusses the sources of this book of the ‘History’ and its importance. In what follows, I use the transcription system of Tr. for all quotations from Fac. and Tr.

³ Tr., p. xciv, n. 210.

⁴ The MSS Ali Emiri 30 (books I–IV, ‘Omān-Bāyezīd I), 31 (book VII, Meḥemmed II), and 32 (book VIII, Bāyezīd II) are all of the same dimensions, uniformly bound, and written in the same careful ta'līq hand. On the evidence of the paper and the orthography Dr. Turan dates M to the end of the eighteenth century (Fac., introd., p. 6). However, its companion-volume 32 ends with the date of copying, A.H. 1099 (beg. 7 November 1687). Unless the paper has a watermark showing that this date is false, it seems that M also was copied in or shortly before this year.

¹ For the conclusive proof cf. Fac., introd., p. 7.

² In preparing Tr. he used also an incomplete MS (Paris, supp. tures 157), which is discussed below (p. 263 f.).

³ The book has been trimmed, but long marginalia were first cut round and folded in to save them from damage. Fac. does not show the numerous erasures which also are apparent in the original. All the headings and the triple dots marking off sections of the saj' are written in red ink. The writer first wrote out the text in black ink, leaving two lines blank for each chapter-heading; when he came to fill in the headings in red he sometimes had to cramp the heading as two lines were insufficient (e.g. at Fac. 503, 508). Shorter headings, āyets, etc., he had noted in the margin in black ink, and then inserted them in the text in red ink (e.g. at Fac. 346, 347), except in a few cases where he overlooked them (e.g. at Fac. 367 = Tr. 337: here the facsimile does not reproduce the heading which is in the margin of the MS). At Fac. 96, where a chapter ends at the bottom of the page, he first wrote in black ink the catchword the first word of the new chapter; then, when filling in the headings, he crossed out in red ink and wrote the new catchword , also in red.

⁴ Thus the copyist of M transposes at Tr. 417 = Fac. 459, 1 (read düpdüz buz), Tr. 430 = Fac. 475,10 (read vüs'at ve füshatle eyvān-i…), Tr. 457 = Fac. 508,10 (read hemīn İskenderiyye kal'asi). Other necessary transpositions (for which I have not compared M): Tr. 282 = Fac. 298, 4 (read geldiler, kenār-i selāmete), Tr. 302 = Fac. 324, 7 (read kendüleri kovan maǧrürlarun zenbūr gibi), Tr. 325 = Fac. 351, 8 (read kişver-i Karaman), Tr. 363 = Fac. 399 (read seylāb … sōyündürelüm, Diyārbekr'i … dōndürelüm). Dr. T. has occasionally inserted an addition at the wrong point, e.g. Tr. 188 = Fac. 195, rūşen … tedbīrdi (which in the MS is written horizontally against 1. 18) should follow mezkūr emīr, whose original dative suffix has been altered to the caret (this is not the only place where a marginale has been moved in the ‘facsimile’ from its correct place in order to fit it into the page of the reproduction); Tr. 382 = Fac. 419, müzāheme … aridi should follow karidi. Lines of verse are often wrongly transcribed and the metre disrupted: e.g. Tr. 28 read Kasr-i ‘asr içre gōr ne manzardur; Tr. 195 read Haber şüd be-aksā-i ān merzubūm/Ki big’ zeşt …; Tr. 209 read Gök tenūrinda olur (P. tenūr is misread passim as A. tenevvür); Tr. 258 read ‘Ve lā tulkū bi-eydīküm’ le ‘āmil, i.e. ‘who act on the precept “and cast not yourselves with your own hands (to destruction)”’ (Qur'ān II, 191). At Tr. 136 = Fac. 139, 7 and Tr. 403 = Fac. 441, 12 and 13, Dr. T. transcribes as dizine, gōǧsine, başina, three examples of the now obsolete directive, dizire, gōǧsire, başira (Deny, Gr., § 896, Phil. Turc, fund., p. 170Google Scholar, § 32129).

¹ Dr. Turan's attitude to the marginalia is not clear, for whereas he incorporates many into Tr., others, apparently equally acceptable, he ignores. Some, but not all, of the latter were accepted by the copyist of M.

² Dr. Turan gives the reference for this important note by ‘Alī Emīrī, which has lain buried in his now very rare periodical ‘O mānli Ta'rī ve Edebīyāti Mejmū'asi. In the course of his article ‘Omānli mefāir-i ta'rīkhīyesine eref-baā bir izdivāj-i müteyemmin’ (i.e. the marriage of Sabiha, the daughter of Mehmed VI, to Ömer Faruk Ef.), OTEM no. 21, 1335, 487–92, ‘Ali Emīrī quotes Kpz.'s story of the feast given for the circumcision of the princes in A.H. 885 (Fac. 588–91), ‘taken from a MS of 622 pages’, i.e. F. His note on the MS reads (p. 488, n. 1):

³ Tr., p. xxx, n. 71.

⁴ This identification, first made by ‘Alī Emīrī (see n. 2, above), is confirmed by Dr. Turan (Tr., p. xxix, n. 70). The impression at Fac. 3 may be compared with the reproductions in TOEM, I (part 7, 1326), p. 401, and on the cover and in the text (no. 4819) of Topkapi Sarayi Müzesi mühürler seksiyonu rehberi, compiled by İ. H. Uzunçarşili (Topkapi Sarayi Müzesi Yayinlari, 8), Istanbul, 1959. The ‘garland’ round the impression in F is in gold ink: it seems to be later than the waqf-dedication of Maḥmūd I, for it is interrupted at the place where they overlap.

⁵ MS Aya Sofya 2612, a copy of the 932/1525 recension of Pīrī Re'īs's Baḥrīye (cf. the facsimile Kitabi Bahriye, Istanbul, 1935, text, p. 1); and MS Ahmed III K3592, a emā'il-i Āl-i 'O mān, written for Murād III by the (later) āhnāmeji Ta'līqī-zāde Subḥī Čelebi (cf. the guidebook Kanunī Sultan Süleyman sergisi (Topkapi Sarayi Müzesi Yayinlari, 6), Istanbul, 1958, p. 8, no. 23). This seal of Selīm I was used not only for the door of the Private Treasury (Mühürler, p. 12)Google Scholar but also for some of the books kept in it (cf. Tr., p. xxix, n. 70); thus its impression in F proves only that F was at some time in the Palace (as is clear anyway from its bearing the waqf-dedication of Maḥmūd I).

¹ Tr., p. xxx, n. 71.

² ‘Ilmīye sālnāmesi, Istanbul, 1334, 347–54. Needless to say the ‘questions’ in these fetwās are written, in a variety of hands, by the clerks who cast each case into the requisite ‘Zeyd-'Amr’ form for presentation to the Mufti for his verdict: the only part in Kpz.'s hand is the ‘answers’. I believe that it is possible to show that the hand in which these ‘answers’ are written is at least not dissimilar from the hand of F, but the necessity to demonstrate this does not arise (see below).

³ Turan, Şerafettin, ‘S Şbn-i Kemâl'in Kanûnî Süleymau'a bir mektubu’, Tarih Vesikalari, Yeni Seri I, Sayi 2 (17), 1958, 221–3 and plate IX.Google Scholar

⁴ Brockelmann, , GAL, Suppl., II, 301Google Scholar, no. 3f, where this MS is not noted; it is included (as an autograph) in the catalogue İstanbul Umumi Kütüphaneleri yazmalari sergisi, 15–22.ix.1951, p. 24Google Scholar, no. 15. It consists of 485 ff. (22·5 x 11·5 cm., the written area averaging 13 X 8 cm.) with 10 to 17 lines to the page. The colophon begins (with a transposition!):

¹ This MS also is included as an autograph in the catalogue of the 1951 exhibition (p. 25, no. 32). The first of the Hadith-collections is that which has been printed (erḥ-i Hadī -i erba'īn, Istanbul, 1316); I have not seen the printed text, but the first lines of the MS agree with the lines quoted from it by Abdülkadir Karahan (İslam-Türk edebiyatinda kirk Hadis, Istanbul, 1954, 176), and the MS, like the printed text, contains only 15 of the 40 Hadiths which appear in ‘Āiq Čelebi's so-called ‘translation’ (Hadī -i erba'īn terjümesi, Istanbul, n.d.). Dr. Karahan thinks that ‘Āiq Čelebi omitted Hadiths which had stood in Kpz.'s original version, but it seems rather that Kpz., having made two or more collections, selected a definitive collection of 40, and that it is this final collection which ‘Ā. Č. translated (cf. Terjüme, p. 4).Google Scholar Brockelmann (GAL, II, 2nd ed., 598) lists two collections of 40 Hadiths by Kpz., one of 30 (? the incomplete set in this MS) and one of 24.

² I take this opportunity of thanking the librarians of the Süleymaniye, Millet, and Murad Molla Libraries for the help they gave me, and the Libraries Directorate of the Turkish Government for granting permission to have photographs taken.

³ The points where these words appear are (page and line of Fac. and of MS Murad Molla 377): (1) 276, 9—16v., 3; (2) 320, 12—14r., 12; (3) 476, 20—4r., 4; (4) 70, 18—14r., 6; (5) 604, 14—3r., 3; (6) 344, 19—9v., 4; (7) (ḥsn) 360, 16—14r., 1; (8) 611, 10—13r., 8; (9) 72, 10—19r., 8; (10) 96, 17—7v., 1; (11) 1—14r., 3; (12) 1—4v., 6; (13) 195, 14—4v., 7; (14) ('ly) 141, margin—2v., 11; (15) 511, 9—23v., 4; (16) 245, 21—3r., 5; (17) 242, 1—11v., 1; (18) 85, 8—15v., 4; (19) 405, margin—17r., 1; (20) 320, 12—19v., 11; (21) 405, 21—4r., 11; (22) (ta'ālā) 26, 3—7r., 11; (23) 319, 4—18v., 9; (24) 39, 8—9v., 14; (25) 181, 11—23r., 6; (26) 195, 13—5v., 3; (27) 419, margin—7v., 8; (28) 225, 15—3r., 9; (29) 26, 2—2v., 4; (30) 20, 21—26r., 6; (31) 476, 21—26r., 4; (32) 322, 20—29r., 11; (33) 95, margin—3v., 12; (34) 38, 3—3v., 11.

¹ e.g. (all references not otherwise identified are to page [and line where necessary] of Fac.): 9 iki şehriyār-i kāmkārdan (+ yādigār), 27 lisān-i sinān-i la'l-bār ve (+ zebān-i) husām-i gevher-dār, 103 mergzārlar ve çeşmesārlarla ( + zeyn) oldi, 542, 10 (fig. 2) ki (+ nihāl-i) bustān.

² Other examples: 124 esrār-i ahbā( + r-i küffār-i bed-kirdā) run, 187 (+ -i bed-kirdāi), 203 (+ ol gümrāhi), 210 (+ ol bed-kirdārdan), etc.

³ Other examples: 29 (+ -i le'īm), 65 (+ -i revān), 76 (+ ve takrīr), 95 (+ -i hākānī), 138 ( + ki nār-i kārzārun yaliniydi), 145 ( + -i cennet-nişān), 234 hayl-i seyl-cū( + ş ve pīl-tū)ş, i, etc.

⁴ Other examples: 100 (+ fermān-i sultānīyle), 126 (+ olub ihzār), 133 (+ geçmiş), 246 (+ sāir yerler gibi), 261 (+ tekrār), 542, 12 (fig. 2) mezbūr (+ Yūnus) voyvodaya, etc. The two remaining additions visible in fig. 2 (lines 10 and 13) seem intended to improve the rhythm of this very elaborate passage: ne bitürürse (+ yanindaǧilar) ani yirdi, … ol iki şīr hancer ü şemşīr gibi (+ biri birine) baş koşdi. Dr. Turan reads şīri (Tr. 485), but şīr is the subject: ‘those two lions “put their heads together”’ (cf. TTS, iv, s.v. baş koşmak, definition 1).

¹ On the third of these Dr. Turan comments (Tr. 409, n. 2) that the heading should be Nazm or Kit'a and not Beyt; but perhaps the writer intended the couplet in the margin to replace the one in the text.

² 226. As it appears in Fac. this marginale seems to be a classic example of the restoration of an omission by a copyist, whose eye has jumped from the rüzgār with which it begins to the rüzgār at the beginning of 227, 1, so that the marginale should be read after the last line of 226. But in the MS the marginale is written 2 cm. lower down, and its rüzgār is linked by a line of dots to Ungurūs at 1. 11; it is to be read at this point (as it is in Tr. 218), and is an addition to the text.

³ In this chapter Kpz. is closely following Tursun Bey (TOEM, ‘Ilāve to parts 26–38, pp. 85–9), who, however, does not speak of the building of Ḥavāle in the reign of Murād II. This and the following example are in effect cross-references to earlier books of the ‘History’.

⁴ The writer uses the sign to mean not only ‘add’ or ‘insert’ but also, even when he fails to delete the word(s) superseded, ‘substitute’ (in which case the sign is usually written over the body of the superseded word); 249, 14 is a particularly clear proof of this.

⁵ Other examples: 201 gibi → mānend-i, 201 and 266 gibi → -vār, 350, 5 kiş eyyāmi → eyyām-i şitā, etc.

¹ Other examples: 59, 8 cerrārla → mansūrla, 152, 4 zevādesi harmanin → harman-i zādin, 267 nahcīri → saydi, 295 teftīş → istifsār, 327 şīr-i delīrün hücūin → şīr-i civānun nā-gehān hücūmin, 567, 5 perr ū bālleri → perr-i ferr ü bāl-i ikbāalleri, etc.

² cf. also 124 Türk → hism (Kpz. using the word Türk only for the frontier-warriors and never for the Ottomans in general), 139 ol çinari→ şöyle çaldi ki ol çinar (çinari is altered to make it the subject: the verb then becomes passive [bölündi] and supplies an eye-rhyme to bulundi above). At two points there are double corrections, clearly showing the writer's hesitancy: at 292 ol is changed to mezkūr, which is then deleted and ol restored; at 296 tutdurdi is changed to tutdurub, which is deleted and adam gönderüb bendle kapuya substituted. A similar hesitancy is visible at 434, 2–3 (where the two sonras are deleted in red ink).

³ 192 biri issi güneş → birisi issi gün; 199 ki etdi … yolini → ki oldi … yolina; 245 the margin has kosa yerde alin tutarlar elin, to replace the text's el veren kimsenün alurlar elin; 309 the margin has gördi çun kan āsumān, to replace kan göricek nāgehān; 317 the writer begins öninde eyledi …, changes this to nedāmet eyledi öninde bülbül, and then writes in the margin the final version nedīm oldi önine geldi bülbül; 377 etdüǧi → takdīre; 499 halki → ehli; 526 beli → nire; 262 is a puzzle: I read Yüz kişi (→ kemīyi or → bahädur) atiyila en kemi ‘the meanest of them (would swallow) a hundred warriors with their horses’.

⁴ TOEM, ‘Ilāve, 91, 1. 13.Google Scholar

¹ op. cit., 118, 1. 12.

² Menzel, Codex (ed. Fr. Taeschner), 199, 1. 6.Google Scholar

³ Other examples of such factual changes: 158, 12 iki kerre deleted; 177, 10 Vidin → Vize; 259 Germeye → ol ile güne; 267 ol diyārdan → hisār üstinden; 308, 18 bir iki bin → birkaç bin; 323 iki üç bin → birkaç bin; 523 and 524 Geliboli → Avlonya.

⁴ The first passage originally read: Hazret-i Hudāvendigar Mora diyārina gazāya vardukda (+ ol bed-rey) firsat buldi, deniz kenārinda bir bāzirgān gemisi eline girdi, anunla Firenge kaçub mürted oidi; all but the last four words is deleted, and the passage is made to read, with marginal additions: Hengām-i firsatda bir sūrette Firenge kaçub mürted oldi, ol bed-sīret-i fāsid-serīretün eǧininden hil'at-i Islām soyuldi. The second passage, which follows immediately the account of Gedik Ahmed Paa's capture of the fortress of Azak, originally read: Azak asilda bir halīce derler ki Bahr-i Muhītden münşe'ib olur, Kara Denize koyulub anun vāsitasiyle ol derya Muhīte ittisāl bulur. Kenārinda bir hisār var … (but he has already mentioned the hisār!); by deletions and additions the passage is shortened to: Azak asilda bir bahra derler ki mezkür hisār …

⁵ Though Dr. Turan follows the revised version for the two modified passages (Tr. 266 and 386), for this passage he includes the deleted matter in his text (Tr. 154).

⁶ TOEM, 'Ilāve, 121.Google Scholar

⁷ Codex Menzel, 199, 11. 3–4.Google Scholar

¹ The noting of sub-headings, etc., in the margin and their later incorporation into the text (see p. 251, n. 3, above) is not necessarily an indication that the MS is an autograph: I have noticed it in several other MSS, and it seems to have been the usual practice of copyists, in order to avoid repeatedly changing from black to red ink. The cramping of the chapter-headings, however, does seem to be significant, for a copyist would have seen from his exemplar how much space he should leave; but Kpz., composing them later, at the stage of revision, had to squeeze them into the uniform spaces he had left. The transpositions too are so numerous that they cannot all arise from accidental omissions; they reveal rather Kpz. in the throes of composition: he realizes that the phrase he has just written could be heightened by an additional word or two, or that a whole sentence should be inserted at an earlier point, writes in his afterthought at the point he has reached, and indicates how the sequence has to be rearranged (cf. 139, 9, 159, 4–7, 199, 1–3, 389, 12–15, 434, 13–14, 550, 21, etc.).

We have seen (p. 252, above) that no conclusions about F can be drawn merely from the presence in it of the seal of Selīm I; but in the light of all this, there can be little doubt that F (not a conspicuous monument of calligraphy) was lodged in the Treasury, probably as soon as it was written, as being the draft of a work commissioned by the Palace (as was the draft of Idrīs Bidlisī's He t Behi t, also a commissioned work, cf. the letter of Idrīs published by Faik Reşit Unat in Belleten, VII, 1943, 199Google Scholar, and the facsimile 1. 34:.

¹ MSS Fatih 4221, ff. 169–89, and Şehid Ali Paşa 2720/14, ff. 199–221. The latter is described in İstanbul kütüpaneleri tarih-coǧrafya yazmalari kataloglari, 1, 2, no. 48, where it is treated as an independent work; Agah Sirri Levend recognized that it is part of book VII of the ‘History’ (Gazavāt-nāmeler, Ankara, 1956, 17 and 170).Google Scholar The scope of this MS (= Fac. 26–75) is clear from these descriptions; the scope of Fatih 4221 seems to be the same.

² Supp. turcs 157, which contains four portions of Kpz.'s ‘History’: (1) the last part of book VI (Murād II, 847–55), for which no other MS is known; (2) (ff. 48r.–76v.) the chapters of book VII found in Fac. 4–75 and 88–94, = P; (3) (ff. 77v.–99r.) the chapters of book VII found in Fac. 26–75 (same scope as MS Şehid Ali Paşa, cf. n. 1 above), = P₁; (4) the first part of book IX (last years of Bāyezīd II and first years of Selīm I).

³ Omissions: Tr. 5c (i.e. p. 5, apparatus note c), 6h, 11a–12a, etc.; anecdotes: 20f, 31f, and a long rhetorical passage at 44a; extra lines: 4c, 4g, 5f, 6d, etc.; verses ‘by the author’: 7b.

⁴ Tr., pp. xcv–xcvi.

⁵ cf. p. 264, n. 3, below.

⁶ P₁ is shorter than P, but might be only a fragment of a longer text from which P could stem.

¹ P does not derive from P₁ for (i) words lacking in P₁ stand in P (Tr. 38f, 63a), (ii) extra words in P₁ are lacking in P (38g, 39d, 76g); nor P₁ from P for (i) words lacking in P stand in P₁ (29i–l, 48f), (ii) extra words in P are lacking in P₁ (44a, 45a, 47b–c), (iii) P re-casts while P₁ supports F (31c–d, 32a and c).

² If X derives from F independently of *T, the variants of P and P₁ are worthless.

³ In three cases P and P₁ lack words which have evidently been lost through homoeoteleuton: Tr. 59c ricāl dropped after abtāl, 64a ehl … dāri after girdāri, 69b tāk after çak (which itself has been dropped in Tr.—read çakaçaki ki çak tāk-i 'ayyūca). Others of their variants are stylistically inferior to the readings of F, e.g. 30f ( ? meaning), 45b (the addition spoils the balance of the saj' yetüremez … getüremez), 52a and c, 56a and e (the headings are misplaced, coming after the mention in the text of the chain and of the appeal for help), 67c (the omission weakens the saj'), 78a (the added sentence has less saj' than we would expect of Kpz. and is inelegant [repetition of içinde, āyīn]).

⁴ Namely Tr. 59a, the addition (+ mutāba'at ve) muvāfakat; 70a, the extra miṣrā'; 75b, the addition to the saj'. Other variants seem to me to be stylistically neutral (e.g. 46a, 58f, 74a).

⁵ P in fact (like the other three MSS, P₁, Fatih 4221, and Şehid Ali Paşa 2720) is not a fragment surviving from a complete copy of book VII, but a ‘gazā-nāme’ for the capture of Constantinople, copied out from book VII (P at somewhat greater length than the others) to form an independent whole; but in P (unlike the others) the text has been freely tampered with. This ‘überarbeitung’ could have occurred anywhere in the line from X to P, but the many marginalia in P (e.g. Tr. 51d, 61a, 65a, 66a and d, etc.) suggest that it is the copyist of P who is responsible (cf. DrTuran, 's just comments at Tr., pp. xxxi–xxxiiGoogle Scholar and n. 75).