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

originally meant « desert », like *qobī* > *γobi*, our Gobi, and the two words are connected; the « hollow of a tree » is a late invention. This legend of the birth from the « hollow of a tree » is not known elsewhere in connection with the Uighurs. Yet Polo is not so wrong in his statements. Both Juwaini and the *YS* (122,1 *a*) relate how at night a stream of light fell upon a tree; where-upon the tree began to swell and in due time gave birth to five boys, the youngest of whom was the first king of the Uighurs (cf. *Br*, 1, 247; MARQUART, in *SPAW*, 1912, 490). I have already expressed the view (*TP*, 1930, 22) that the two legends must have a common origin.

**222. ESCEQE**

*esceqe* F *antixi* VA *vechi* V
*sech* Z *chadi* VB *veschovi* TA¹
*siechi* R *ueschoui* TA³ *vieillars* FB
*viellars* FA

Esceqe, not «esceque» as in *B*, 205 (CXCII²), represents Arabic *šaiḫ*, meaning « old man », « elder », as Polo says, and is our sheikh; it occurs as « seic » in Joinville. On the various forms of the word in European languages, cf. LOKOTSCH, No. 1775; and it is *šeg* in Tibetan (LAUFER, *Sino-Iranica*, 595). « Sech » in Z may be for *Scech or *Siech (cf. R's Siechi), and it may be that, owing to Polo's explanation « old man », it is this word which has been altered (and not at first translated) into *vechi* in V and *veschovi* in TA¹. In FA and FB, «vieillars» does not of course represent esceqe, which is omitted in all Mss. except F, Z, but its meaning of «old man» as given by Polo (in FC¹ it has become « vilains »).

VB offers a « learned » correction when its substitutes « chadi » for esceqe. « Chadi » is Arabic قاضي *qāḍī*, a Mussulman « judge », but the word does not occur in Polo (see « Casses »). On the forms in European languages, cf. LOKOTSCH, No. 984, particularly Span. and Port. *cadi*, Span. *alcalde*. In India, the Arabic *ḍwād* sounds *z*, so that the Anglo-Indian forms are based on *qāzī* (cf. *Hobson-Jobson*², 177-180). In Malay, it is pronounced *-ll-*, so that *qāḍī* becomes *qāllī*; this accounts for the form *kali* in Cham (cf. CABATON, *Dictionnaire čam-français*, 67). Modern Chinese transcriptions in Central Asia are based on *qāzī*, as is shown in the second half of the 18th cent. by the transcriptions 哈 子 *ha-tzŭ* and 海 子 *hai-tzŭ* in ch. 8 of the *Hsi-yü wên-chien lu* (COURANT, *Catalogue*, No. 1831). But the mediaeval transcriptions suppose *qāḍī*. Such is the case with the « cadi » repeatedly mentioned by Odoric in the region of Bombay (*Wy*, 424-434) and with the « cadinus » of Jourdain Cathala (*JRAS*, 1928, 349-376) or the « cadini » (plural) of Pascual de Vittoria (*Wy*, 505). It was also the pronunciation *qāḍī* which the Chinese transcribed in the Mongol period as 哈 的 *ha-ti*. The *ha-ti* had official authority over the Mussulmans, but their offices were suppressed in 1311 (cf. *YS*, 24, 4 *a*; *T'ung-chih t'iao-ko*, 29, 5 *a-b*). So we may practically be certain that, if Polo had mentioned the *qāḍī*, it would have been as *cadi* or *chadi*; but this is not enough to make us accept VB's « chadi » as really due to Polo.