1 Sobranie sochineniy, v pyali lomakh, editor–in–chief Margulan, A. Kh., Ata, Alma, I 1961, 289–300; I, 421 f. Valikhanov died in April 1865. The text may have been recorded on 26 or 27 May 1856, when Valikhanov was visited by a bard ⊂irdi⊃ who sang manas. V. was then on the Santash Pass, east of Lake issyk Kul…, among the Bugu tribe; I, 249 f.Google Scholar

2 Narechiya tyurkskish plemen zhivushchikh v Yuzhnoy Sibiri i Dzungarskoy Stepi. Obraztsy narodnoy literatury severnykh tyurkskikh plemen. Karakyrgyz, v., St.Petersburg, , 1885, pp. 140 ff.Google Scholar

3 cf.Jackson, K., ‘Incremental repetition in the early Welsh Englyn’,Speculum, XVI, 1941, 304 ff.; A. T. Hatto (ed.), Eos, the Hague, 1965, index sub ‘parallelistic songs’. Valikhanov's commentators unaccountably refer to the original recorded in 1856 as in prose (I, 670). Even if the two couplets cited by V. were not extant, we could be sure the Kirgiz orginal set down by V.was in verse, See p. 346, below.Google Scholar

4 op. cit., I 421.

5 Die Sprachen der nordlichen turkischen Stamme. Proben der Volkslitteratur der nordlichen turkischen Stamme. V Der Dialekt der Kara–Kirgisen, St. Petersburg, 1885, pp.142 ff.Google Scholar

6 Proben, v, p. ii. (References to Radlov's introduction will be to the German edition, as the more essily accessible).

7 op. cit., I, 367.

8 I, 299, 29 f.

9 In the second part of this study, to appear in BSOAS, XXXII, 3, 1969.Google Scholar

10 I, 291, 21 f.

11 For another example see p. 357, below.

12 As Professor Bernard Lewis observed.

13 I, 290, 36.

14 See the Kirgiz edition, Frunze, 1956, and P. Boratav (tr.), Er–TōshtÜk (Collection Unesco d'Oeuvres Representatives), Paris, 1965. With commentary by P. Boratav and L. Bazin.

15 Radlov's Er Tōštϋk does not know Ay–salkun. In it there is the parallel and contrasting figure of Bek Toro, a peri disguised in rags, who never enjoys Tōštϋk's love, though the commentators may well be right in thinking Bek Toro and Ay–salkin to be variants of one and the same original figure.

16 Recherches sur l'épopée st le barde au tibet (Bibliothéque de I'Institut des Hautes Études Chinoisee, XIII), Paris, 1959, 543 ff. I am grateful to Sir Harold Bailey and Dr. C. R. Bawden for this reference.

17 BM, 317 ff.

18 On the ‘black bone’ cf. Krader, L., Social organization of the Mongol–Turkic pastoral nomads, The Hague, 1963, 212,259,277.Google Scholar

19 Kökčö has a piece of his own with Manas at v, I, (3) and with ALman Bet at v, I, (2).

20 Papers written on by this holy man and clapped on to the withers of a horse help it to win: but Ay–khodzha runs his own horse, Ayban (298, 40ff.).

21 Line 167 Mis-Burčak; 1298 Mus Burčdak: Radlov, translation 167 alys Burtschak; 1298 Mus Buetschak ! cf. v, I(5), 599 Mus Burčak.

22 173 Orongu; 747 Orongu; 1372ff. regularly Orongo; translation Orangi; Orongy; Orongo! Further edition will moddify this list 1–22;, which in tentative.

23 1366 ff.

24 In the Kazakh Er KōkšÜ the genealogy is: Uak_Kambar_ KÖkšÜ_Kosay (Košoy), and KškšÜ is lord of the Ten Nogay, Radlov, III, 1870, viii, 10.

25 Radlov, v,I,(2), 23 ff., etc.; K 291,37.

26 v, II. In this poem, in Which he is presented positively, ĵoloy is the son of Nogay-bay of the Ten Nogay and is at odds with the Kalmak Khan and his henchman Korjur-bay. He is therefore presented here as a tribal hero of the Kiegiz Solto, possibly in a mocpic sense.

27 1554 ff.

28 See the discussion of these epithets in part II, BSO AS, XXXII, 3, 1969.

29 I, 421. For initial ‘M’ cf. further Radlov, v, II, 4740 ff. Mis Kara (Joloi–kan, Soltu tradition).

30 The harmonized version of Manas, Frunze, 1958, II, 131 ff.

31 cf. BM, 1403 ff.

32 Thus the 1958 ed.,Izbrannye Proizvedeniya, 356, 28: the current ed., I, 297, 13 reads Urunkha. Similarly 1958 ed. Irku, ourrent ed. Urkhu.

33 See n. 32 above.

34 Ěrigis, MG. U., Istoricheskie predaniya i rasskazy Yakutov, Moscow, Leningrad, I,1960, 282.Google Scholar

35 Radlov omits to translate aišig.

36 p.348.

37 Studies in frontiee history: collected papers, 1929–1958, London, 1962, e.g. pt. I ‘ The Inner Asian frontier’,

38 See below, p.372.

39 cf. Harva, U., Die religiōsen Vorstellungen der altaischen Vōlker (FFC, 125), 1938, 57 ff.Google Scholar

40 See below, p.372.

41 In a sense KÖkÖtÖy's ‘ Behests ’ dominate K. In view of the tone of some of the Turkic runic inscriptions it is to be considered whether perhaps the Behest or Bequest is not a fundamental attitude od rhetorical or poetic utterance in early Turkic which has been preserved in K.

42 I, 292.

43 Malov, S.E., Pamyatniki drevnetyurkskoy Pis'mennosti, Moscow, Leningrad, 1951, 61and 64 f.Google Scholar

44 Czeglédy, K., ‘ ā Coγay–quzī, Qara–qum, KÖk–Öng ‖, Acta orientalia Acad. Sci. Hungaricae, XV, 1–3, 1962, 55 ff.Google Scholar

45 If so, it would not imply the only displaced couplet or quatrain in K. See p. 357, below.

46 Czeglédy, art. cit., 55.

47 i.e. ‘Dzhulay’ (ǰoloy), see p. 376, n. 183, below.

48 saz= ‘ marsh ’.

49 i.e ‘ Er Dzhulay’, see n. 47 above, and p. 376, n. 183, below.

50 The 1958 ed. reads Katan; the current ed., I, 293, 13 reads katay, whether by conjecture or by improved understanding of the manuscript one can never tell. Katay does not appear in the geographical index, but is listed without discussion under Kitay. Since, stylistically, Katan/Katay is a ‘Parallelistic’ variation of Altay, the reading ‘Katay/Kitay’ would require justification. If the text actually read ‘K(h)angay’ must would be explained.

51 See the discussion on pp. 374 ff., below.

52 293, 24. See the discussion on pp. 374 ff., below.

53 The learned institutions of this country are very ill–provided with large-scale maps of Kirgizia and its neighbours. Such identifications of places as are offered in this were the fruit of three to four weeks of intensive research in the full sense of that term.

54 I, 670 ff.

55 I, 280. There is a ‘Kozu[!]baŝt’ (translation:‘ Kosy Basch’) at Radlov v, I, (5), 239 which I have failed to tie down.

56 I, 671.

57 ibid, Aktam is not indicated at 1:250,000. VoL.XXXII. PART 2.

58 I, 244.

59 II, 61S.

60 K, 292, 41 Dzhalanache reke. ch(č)and sh(ŝ) tend to alternate in the names and Place names of this region owing largely to the correspondence Kirgiz (č): Kazakh (ŝ).

61 Thus the current ed. 292, 44. The 1958 ed., however, reads: Ak–Tersken. There is no discussion. Neither form occurs elsewhere in the three current volumes of Valikhanov. Terskey occurs variously. One suspects that Tersken may have arisen through H for H.

62 I, 671.

63 II, 554.

64 II, 227; 514.

65 See above, p.355.

66 I, 671.

67 Are we to understand that it is still the custom of the Kirgiz to take rice here?

68 I, 671.

69 ibid. ‘Beš–Teräk’ is named in the disordered list of places in Young Manas's Plan for his future at Radlov, v, I, (1), 101.

70 Similarly, archaeologists needed warning that a famous passage on swords in Beowulf was not an ironmonger's list till it had been divested of its stylistic wrappings.

71 I, 293, 17 ff.

72 See pp. 374 ff., below.

73 Kōkōtōy ūlu Bok Murun (181).

74 čorolorun.

75 Radlov 198 ačamin (= ‘I shall open’) (Radlov, translation: ‘Willich…ϋberschreiten’): read ašamin, cf. 194, 208 ašamin (R. ‘Will…ϋberschreiten’).

76 maldi subai kilamin.

77 i.e. ‘release the camels’ ? (R. 212 ‘entlass’ ich die Kameele’).

78 koi kϋzōn (ǃ) alamin.

79 basamin (227).

80 Tūluktan ašamin (228).

81 The easten Sairam and Lake Sairam are exculded by stages IV, V, VI, etc.

82 Radlo's translation ‘ der Sxhiba Stadt’ suggests that he was thinking of the Tungusspeaking ‘Sxhibä’ in the Manchu service in the Ili Province, who interested him greatly. Did Radlov mentally place the beginning of the itinerary in the easten Sairam ?

83 Streams are not visible on the map at 1:250,000.

84 Shibbe, far to the south below Fergana, is to be excluded on grounds of phonology, distance, and absence of –kent.

85 See p. 356, n. 60, above.

86 See Batmanov, I.A., Souremennyy Kirgizskiy yazyk, I, fourth ed., 1963, 80Google Scholar

87 cf. Valikhanov, II, 530, map.

88 ibid.

89 The second i in Šibi would seem to be Radlov's responsibility.

90 The Pass is c. 60 km. south of the present town of Karabalty.

91 BM, 207 Kairdt.

92 Perhaps other formidable obstacles were avoided, too.

93 I, 316. For kidirip with a river as object see Čϋidϋ kidirip below on this page.

94 R.'s translation at 221 ‘Kopuy’ suggests an incomplete correction from Kopu to Kopy which the printer had not understood. The lack of vowel–harmony in Kopi, however, is disturbin in Kirgiz.

95 I, 410.

96 ‘Cooking salt of/from the Karkara’.

97 I, 245.

98 40–60 km. west of the river Sharyn.

99 At the head of the river Tyup, east of Lake Issyk Kul'.

100 I, 246.

101 ibid.

102 ibid.

103 Not identified by the commentators of K.

104 See p. 376, n. 183, below. From this point on it becomes pÓssible to detect line–divisions of the original. Some well–balanced parallelistic quatrains emerge.

105 See p. 376, n. 183, below.

106 A ‘Kalmak’ over–garment.

107 A sign of high rank rank in the Chinese hierarchy, sometimes also conferred on Kirgiz manap and on princes of the lesser dependents of China, ef. the commentary, I, 671.

108 e.g. BM,1725; 1802; 2100 and v, I, (5), 1635.

109 Lattimore, op. cit., 184.

110 I, 235.

111 II, 106.

112 See above, p. 362

113 Maps at 1:250,000 show a route through the valley to Kulja also from the southern Kush Murun Presumably ending with a ferry.

114 Radlov, , Aue Sibirien,second ed.,II, 1893, 305.Google Scholar

115 To which, however, V.V.Bartol’d is ready to concede very little antiquity—see his article ‘Kuldia’ in the Encyclopaedia of Islam, first ed.Google Scholar

116 Valikhanov, II, 105 f.

117 Bartol’d, loc. cit.

118 II, 106.

119 op. cit., II, 305.

120 In Radlov’s day the Bugu nomadized as far east as the river Tekes, proben, v, p. i.

121 See p. 357, above.

122 Meticulous in other matters, Radlov does not always specify the source of his Poems, so that one has to resort to inference.

123 presumably, R. collected Joloi–kan and Er Tōštūk (both still independent of the Manascycle) from the Solto.Google Scholar

124 See my article‘The birth of Manas: a confronatation of two branches of heroic epic poetry in kirgiz’, Asia Major, NS, xiv, 2,1969.Google Scholar

125 proben, v, p. xiii.

126 See p. 364, above.

127 Radlov, , Aus Sibirien, II, 403–12Google Scholar. The last battle the victorious Taranchi had to fight was with the Kalmak. after the Kalmak defeat, the latter withdrew to Lake Issyk Kul’ and submitted partly to the Russians, partly to the Taranchi (August 1866). Under 1868, R. speaks speaks of the plight of the Kalmak in the Semirech′ skaya Oblast‵ (II, 373).

128 Bartol‽, Zwōlf Vorlesungen, second ed., Hildesheim, 1962, 178.

129 op. cit., 179.

130 ‘ Belaya orda’ in the Bol′shaya sovetskaya ėntsiklopediya, second ed., iv, 1950, 409 f.Google Scholar

131 Spuler, , Die golden Horde, Leipzig, 1943, 25Google Scholar: ‘…wo sie das Reich der“Wissen Horde” bildeten, von dessen Geschichte nur weing bekannt ist und dessen Schicksale im folgenden n i c h t [Spuler's emphasis] ausfÜhrlicher behandelt werden’.

132 Bartol'd, op. cit.,172, citing Mu'īn al-Dīn Natanzī.

133 Bol'sh. sov. énts., loc. cit.

134 Encyclopaedia of Islam, second ed., under ‘Djanāza’ (A.S.Tritton).

135 Roux, J.-P., La mort chez les peuples altaiques anciens et médiévaux, d'aprés les documents écrits, Paris, 1963, 81; 158.Google Scholar

136 Russia: or a complete historical account of all the nations which comprise that Empire, II, 1780, 289 f. (Russian original in Brit. Mus. incomplete.)

137 I am grateful to Dr. G. F. Cushing for this elucidation.

138 Radlov, Aus Sibirien, II, 112; 126; 130.

139 op. cit., II, 121.

140 Levshin, A., Description des hordes et des steppes des Kirghiz–Kazaks ou Kirghiz–Kaissaks, trans. F. de piguy, Paris, 1840, 365 (Russian orignal, 1832).Google Scholar

141 K, 291, 30.

142 ‘Djanāza’, see p. 368, n. 134.

143 Radlov, op. cit., II, ch. vii, passim.

144 op. cit., 167.

145 ‘ Volkskundliches aus Altturkestan’, Asia major, II, I, 1925, 122.Google Scholar

146 Levshin, op. cit., 366.

147 Radlov, op. cit., II,77, ‘Meine Grab–Oeffnungen in der Kirgisen–Steppe haben mir aufs Deutlichste bewiesen, dass die Gräber der Kirgisen–Steppe ebenso grÜhlt sind wie die Graäber am Ob, im Altai und am Jenissei’.

148 SeeMau-tsai, Liu, Die chinesischen Nachrichten zur Geschichte der Ost–TÜrken (T'u–kÜe), 1958, I, [text in German translation] 179; II, [commentary] p. 620, n. 999.Google Scholar

149 Malov, S.E., Pamyatniki drevnetyurkskoy piṡmennosti Mongolii i Kirgizii, Mosvow, Leningrad, 1959, xb (14), pp.19and 24.Google Scholar

150 cf. Pelliot, P., ‘Neuf notes sur des questions d'Asie Centrale’, T'oung Pao, XXVI, 1929, 229 ff.Google Scholar

151 Gumilev, L.N, Drevnie Tyurki, Moscow, 1967, 329 ff.; L. Jisl, Mongolian journey, London, 1960, ref. illustrations nos. 8–10.Google Scholar

152 op.cit., 367.

153 op.cit., I, 489.

154 ibid.

155 op.cit., 321 ff.

156 pominka (the pl. pominki is normal) andtriznaare not used consistently in K to make a distinction between smaller and larger feasts. treba ‘religious rite’ is also used. aš may well have stood in the original in all cases.

157 cf. BM, 609 f.; barsin ko·ur salkin kÜs–minän / kelsin čkan ala čalbir jas–minän.

158 See Roux, op. cit., 163 ff. on aγit.

159 Roux, op. cit., 185. M.Roux gives no precise reference.

160 P.352.

161 Roux, op. cit., 118 ff.

162 Roux, op. cit., 105.

163 op. cit., 374.

164 under kan II.

165 Travels in the regions of the Upper and Lower and Lower Amoor, London, 1861, 62 ff.

166 See p. 354, above.

167 I, 672.

168 kan is the more authentic form in Kirgiz.

169 I, 672.

170 See below, p. 375.

171 Harva, op. cit., 58. On the Turkic reverence for sacred mountains see Roux, op. cit., 97 ff. Cf. further the role of the Burhan Khaldun Mountains, in which, according to the ‘Secret history’, Činggis's ancestor was born of a wolf and a hind.

172 Malov, op. cit., p. 18 (Mog. 37) and 94b.

173 Il ‘Kitāb–i Dede Qorqut’ (Studi e Testi, 159), Cittá del Vaticano, 1952, 49, quoting Ziya Gok Alp, Tϋrk tϋresi, Istanbul, 1339/1923–4, 79.

174 Narody sredney Azii i Kazakhstana (Narody Mira), II, 1963, 12 (quoting Abu ‚I Ghāzḹ).

175 op. cit., 58.

176 Twentieth–century material is excluded in principle, see p. 350, above.

177 I, 416.

178 ibid.

179 Urumchi is reached from eastern Kirgizia by entirely different routes.

180 Occasional reports of Kirgiz (and Kazakh) nomadizing in the Altay or further east in the nineteenth century are naturally met with, e.g. Radlov, op. cit., I, 219.

181 My italics.

182 I, 357.

183 I, 421. On the basis of the Résumé: Dzhulay, and the parallelism at K, 293,4: nevernomu khanu khrabromu Dzhuzayu and 293, 6 k étomu–to nevernomu khanu Ir–Cholanu, one can con fidently emend Dzhuzayu to Dzhulayu, and Cholany to Dzholayu (another H for И? cf. p. 356, n. 61, above). The whole passage concerns Dzhulay/Dzholay (i.e.Ĵoloy) and only after this emendation does he clearly become a Kalmak, cf. the Résumeé: ‘mongol′skogo khana Dzhulaya’ (I, 421,34). On the basis of the discrepancies between (i) K, (ii)the Resume, and (iii)V.’s account of the ‘Manas saga’ as given at I, 357, it would be possible to argue that V. knew three different versions of Kukotay: yet I suspect he was quoting rashly from memory in (ii)and (iii), in which case, however, an explanation is needed why ‘Dzhulay’ should be correct in (iii)and wrong in (i). The chronology of the documents—(i) translated 1861–5; (ii)1860; (iii)post 1856—does not of itself suggest a solution.

184 Reproduced in the commentary without discussion as Buruntal, I, 671.

185 II, 105 ff.

186 I, 270.

187 Inscriptions de l’Orkhon déchiffrées, Helsingfors, 1896, p. 152, n. 32.

188 art. cit., 212 ff.

189 art. cit., 55 ff.

190 op. cit., II, p. 500, n. 58.

191 I, 270.

192 ibid.

193 II, 124.

194 Bichurin, op. cit., I, 332.

195 D. Banzarov, Chernaya vera ili shamanstvo u Mongolov, St. Petersburg, 1891, 20 f.; cf. Harva, op. cit., 244.

196 A continuation of this article will offer, among other topics, a stylistic comparison of K and BM, and will attempt a description of their common source in outline. N. I.Veselovskiy's edition of Valikhanov's Kukotay, to make confusion over the text of belugy ordu worse confounded, reads: (a)belugy Ordu,(b)beluyu ordu (‘Sochineniya Chokana chingisovicha Valikhanova’, Zapiski Imperat. Russk. Geograf. Obshchestva, Otdel Étnograf., XXIX, 1904, 209 and 213 respectively, cf. pp. 365 ff and 355, above; on p. 213, Veselovskiy reads Ak Tersken, cf. p. 357 and n. 61, above, and Katay, cf. p. 355, n. 50, above). I am grateful to Sir Harold Bailey, Dr. C. R. Bawden, and Dr. T. O. Gandjeț, all colleagues in the London seminar on Epic, for their helpful suggestions when I was completing my manuscript. They of course bear no responsibility for the views set forth here. It goes without saying that I have learned much from Professor V. M. Zhirmunskiy' authoritative ‘Vuedenie v izuchenie éposa “Manas”’ in the symposium volume entitled Manas, Moscow, 1961, 85–196. My thanks are also due to those in charge of the map rooms at the Bodleian Library, the British Museum, the Royal Geographical Society, and University College, London.