National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

> > > >
Color New!IIIF Color HighRes Gray HighRes PDF   Japanese English
0315 Notes on Marco Polo : vol.1
Notes on Marco Polo : vol.1 / Page 315 (Color Image)

New!Citation Information

doi: 10.20676/00000246
Citation Format: Chicago | APA | Harvard | IEEE

OCR Text

 

158. CINGHIS   299

paper which I cannot trace for the present, VLADIMIRCOV remarked that such a derivation was improbable, since both « Cinggis » and tänggis, « sea », occurred in the Secret History (§1). I do not think that this objection is of great moment. Double forms occur in the Secret History; for instance « Idil » and « Äj it », discussed further on. Moreover, if we are right, both « Cinggis » and tänggis were foreign words in Mongolian; tänggis entered the legend of Mongol origins, closely connected with the ancient legend of the early Turks, at an unknown date, but there is no reason

to date this contemporaneously with that at which «   » became in certain religious or political
circles the accepted term it must have been to have been adopted by Chinghiz-khan. If the two forms came to the Mongols at different periods and through different channels, and moreover acquired different uses, we need not be surprised to find both occurring in the same work.

Mong. tänggis, « sea », is an cz7raÿ Xsy6Ftsvov in the Secret History, and is of rare occurrence elsewhere. KOWALEWSKI does not give it. Late Sino-Mongolian lexicography knew it, however (cf. KLAPROTH, Asia Polyglotta, 263), and it is found in GOLSTUNSKIÏ. In the Mongol period, it occurs in YS, 63, 16 a, as t'ien-chi-ssû, tängis, in the name of the Caspian (cf. JA, 1920, I, 166). The word is current in Kalmuk, and pronounced terigis (RAMSTEDT, Kalm. Wörterbuch, 392). Tänggis is merely borrowed from the Turkish tängiz (täiaiz), but is written according to the rules of Mongolian spelling, where Turk. -ng- (-4-) is rendered as -ngg- (-rig), and where -z becomes -s since there is no z in Mongolian.

The case of Turk. tängiz > Mong. Cinggis would be somewhat different. The treatment of the final -z > -s is the same, but the initial 6. implies a mongolization. In Mongolian, all ancient *ti- became ci- at an early date, for instance in the pronoun of the second person ci z*ti. I have quoted above the case of Turk. tegin, tägin (<_ proto-Mong. *tegin) > tigin Mong. cigin (in ot-

('-igin).   This process of palatalization went on in Mongolian even after the 13th-14th cents.
We still find in the Secret History (§ 262) « Idil » alongside of « :0 1» » (§ 274) as a name of the Volga (« Aiil », for « Ajil », in the Ulan-Bator ms.; the , {~i p M Yeh-chih-li River [ho] of YS, 121, 2 b, and the A-chih-li Lake [hai-tzic] of the map of the Yuän shih lei-pien may represent other originals). Cf. Turk. Atil in Kas; ari (not «Ytit» as in BROCKELMANN, 244), Ca; Atil and Idii (Edil), Kirgh. Edil, Tel. Ädäl, Kaz. Idii (RADLOV, I, 842, 850, 857, 1509; Proben, II, transi., p. xii), Ethil in Plan Carpine (Wy, 136), Etilia and Ethilia in Rubrouck (Wy, 195, 205, 210, 216, 223, 315), Edil in the Catalan Map (BucxoN, in Not. et Extr., XIV, II, 131) ; Mong. Ajii (KOWALEWSKI, 227),

New Mong. Ijil (VLADIMIRcov, Doklady Ak. Nauk, 1929, 289), Kalm. I)il,   (VLADIMIRcov, ibid.;
RAMSTEDT, Kalm. Wörterbuch, 205). The same obtains in the case of the Irtysh, called in the Secret History Ärdig (§§ 207, 264) or Ärdis (§ 198; the latter form is a misreading of the transcribers; the ancient Mongolian script did not distinguish -s and -s), Yeh-êrh-ti-shih (Ardis) in YS, I, s. a. 1206 and 1208; 122, 1 b; Yeh-li-ti-shih in YS, 22, 1 a; Yeh-êrh-ti-shih in the Shêng-wu ch'inchêng lu (47 a); cf. Ärtis in Kagyari (misread «Artïs» in BROCKELMANN, 240); Ertic in Kirakos (PATKANOV, Istoriya Mongolov, u, 82). But the Mongol name is Ardis in « Sanang Setsen » (cf. SCHMIDT, Gesch. der Ost-Mongolen, 211, 412), or more probably Ärjis as in the Chinese translation of that work (Meng-ku yüan-liu chien-chêng, 6, 22 b).

A derivation Turk. tängiz > Mong. Cinggis would require either the existence of an intermediary form *tingiz, or the palatalization t- T cY before vowels other than i. Both are possible.