The Term “Mongγol” Revisited (蒙古一詞再議) 一文已於近日出版。該文擬就歷史文獻和蒙漢對音兩方面切入，以說明「蒙古」一詞乃源自蒙古的發源地「望建」河。全文刊登於Central Asiatic Journal 60 (2017)1/2, pp.183-206，下面為文章的部分章節。
The Term “Mongγol” Revisited
Landscape at the confluence of the
River and the
. Photo by Wei-chi ren, 2005 Shilka River
V. MONG ΓOOL: GEOGRAPHICAL FEATURES
If my hypothesis is correct, there is good reason to assume that the term Mongγol, which is cognate with the hydronym Wang-chien, is derived from the indigenous Mong Γool, the river proposed by Banzarov. According to Kowalewski, the word mong means ‘riche, opulent; fougueux, impétueux’, the Mong Γool, as such, must have been a fast-flowing river. This interpretation is corroborated by the fact that the Ergüne River, especially the section from Chi-la-lin 吉拉林 (or Shih-wei) to Lian-yin 連崟, where the early Mongols are known to have settled,flows rapidly. The name Kiyan (Kiān)乞顏 (kǐət ŋan), one of the two legendary Mongol groups that migrated to the Ergüne Qun, the Mongols’ homeland, echoes this thesis. According to Rašīd-al Dīn, Kiyan means in Mongolian ‘a rushing torrent from the mountains’. Moreover, the geographical features of the Ergüne Qun, which, as pointed out by Rašīd-al Dīn, means ‘sloping cliff’, are faithfully reflected by the steep slopes that characterize this particular section of the Ergüne River （figs. 2 and 3）, as the present-day toponym Lian-yin, which reflects the hilly terrain of the region, also suggests.
The use of the word γool in the ethnonym may seem to contradict what I have noted earlier, i.e., it first appears only in sources published after the 13th century, such as the Mongγol-un ni'uča tobča'an and the various Sino-Mongolian glossaries. Nonetheless, the naming by the Tibetans of a river south of the Kokonor as Jima Gol~Khol (Ch. Ta-fei Ch'uan大非川 < Mo. Dabu(sun) γool) in the seventh century demonstrates beyond a doubt that the term γool was in use during T'ang times. This Mongol vocable must have been brought by the T'u-yü-hun吐谷渾, a Hsien-pi group, when they migrated from Mongolia to the Kokonor area in the fourth century. The fact that the Dagur language, which preserves the Middle Mongolian forms, has the same word γol, adds credence to our argument.
Another possible derivation is Möngke Γool, i.e., an ever-flowing river that provided the early Mongols with a reliable source of water. The change from Möngke Γool to Mong Γool can be deduced as follows. Owing to haplology, the syllable -ke in möngke was dropped when merged with the following γool, which was contracted simultaneously to become -γol. The ö in the first syllable of the merged word, because of vowel harmony with the resultant -γol, was further changed to o through regressive assimilation. Though a tantalizing alternative, Möngke Γool is a less likely derivation than Mong Γool because of the complicated process of linguistic change involved.
It is interesting to note that one of the tributaries draining into the Hei-lung-chiang is known as the Mo Ho 漠河 River (or Mu Ho 木河 in Ming times) and that the County named after it is known to have been inhabited by the Shih-wei people. [See Mo Ho hsien-chih 漠河縣志, ed. Wang Shu-ts'ai 王樹才 (Peking: Chung-kuo ta pai-k'e ch'üan shu ch'u-pan she, 1993), pp. 1, 57-6, 107, 657]. However, no reference is made to this river in T’ang sources, and its medieval reading mak γa (Kuo, Ku-yin, pp. 17, 26) does not quite match the term Mongol.
J. E. Kowalewski, Dictionnaire mongol-russe-français, vol. 3 (Kazan, 1844; reprint, Taipei: SMC Publishing Inc., 1993), p. 2029a. Kowalewski’s definition is obviously a translation of I. J. Schmidt, Mongolisch-Deutsch-Russisches Wörterbuch, nebst einem deutschen und einem russischen Wortregister (St. Petersburg: Kaiserliche Akademie der Wissenschaften, W. Graeff und Glasunow, and Leipzig：Leopold Voss, 1835), p. 217b: ‘reich, űberflűssig; trotzig, driest’. Though rarely seen, the word mong is not a hapax legomenon. Its use is attested in the Qorin naimatu tayilburi toli (Kőkeqota: Őbőr Monγol-un arad-un keblel-ün qoriy-a, 1994), p. 1205, where the phrase mong song ügei is glossed as ‘masi elbeg delbeg,’ and in the Mongol kelen-ü toli (Kőkeqota: Őbőr Monγol-un arad-un keblel-ün qoriy-a, 1999), p. 1870, where the phrase mong ügei is also glossed as ‘elbeg delbeg’. Its existence is furthermore mirrored by its Oyirad counterpart moη, which carries the same meanings of ‘keck, trotzig’ as Kowalewski’s and is used similarly to constitute a toponym (moη-χamγ̥ name eines sandberges bei Sarepta a. d. Wolga). See G. J. Ramstedt, Kalmückisches Wörterbuch (Helsinki: Suomalais-Ugrilainen Seura, 1935), p. 264b and H.A. Zwick, Handbuch der Westmongolischen Sprache (Villingen: Ferd. Forderer, 1853), p. 266b, on which the former is based.
 Mo Ho Hsien-chi, p. 105; Hsü Chan-chiang et al., eds. Hu-lun Hu chi, p. 37; Hei-lung-chiang chi-kao, chüan 3, pp. 320-321; chüan 4, p. 571.
 Recent archaeological discoveries provide evidence that the early Mongols lived in the Ergüne River basin. Their presence is confirmed by the tree-trunk coffins found in the vicinity of the confluence of the Shilka River and the Ergüne River (fig. 1). These coffins, which date from the 8th century to the 9th century, are hollowed out of a massive log, and are typical of those used by the Mongols who later nomadized the Eurasian steppes. The Russian archaeologists who discovered these coffins in the late 1980s call the culture they represent the Dabsun culture. Similar dugout coffins dated around the 10th century have been discovered in West Wu-chu-erh西烏珠爾 and Hsieh-erh-t’a-la謝爾塔拉north of Hu-lun Lake, showing the Mongols’ gradual migration southward. It is from the Hu-lun Lake region that the Mongols moved further west and settled in the Onon River basin since the 10th century. See Lin Mei-ts’un林梅村, Sung Mo Chih Chien 松漠之間（Peking: San-lien shu-tien, 2007）, pp. 256-57, Chung-kuo she-hui k’e-hsüeh yüan et al., Hai-la-erh Hsieh-erh-t’a-la mu-ti海拉爾謝爾塔拉墓地 (Peking: K’e-hsüeh ch’u-pan she, 2006).
 Rivers in this area flow swiftly. A tributary of the Ergüne now known as the Chi-liu River激流河, as its name suggests, is one such example.
 Shi Chi, vol.1 part 1, p.252. For the term Kiyan, see Cha-ch’i-ssu-ch’in, Meng-ku mi-shih hsin-yi ping chu-shih 蒙古秘史新譯並注釋 （Taipei: Lien-ching ch’u-pan shi-yeh kung-ssu, 1979）, p. 44, note 2.
 Kuo, Ku-yin, p.74, 199.
 Shi Chi, vol.1 part 1, p.252. The Mongols’ penchant for evoking a hydronym to name themselves is also evidenced in the term Činggis, which is derived from the Turkic word teŋiz, meaning ‘ocean’. See Gerard Clauson, An Etymological Dictinary of Pre-Thirteehth Century Turkish, p.527.
Christopher I. Beckwith, The Tibetan Empire in Central Asia (New Jersey: Princeton University Press, 1987), p. 33, note 109. Note that γool was not used by the Turks to mean river in the T'ang period as Beckwith claims. See Clauson, Etymological Dictionary of Pre-thirteenth Century Turkish, p. 715, where it is noted that kö:l (g-) is ‘never used for “sea”, or for “river”’.
Chou Wei-chou 周偉洲, ‘Ta-fei yü Mo-li 大非與墨離’, in Hsi-pei li-shih yen-chiu 西北歷史研究 (Shan-hsi: San Ts'ing ch'u-pan she, 1990), pp. 129-34.
On the linguistic affinity between the Hsien-pi and T'u-yü-hun languages, see Ch'en Chien 陳踐 and Wang Yao 王堯, Tung-huang T'u-po wen-hsien hsüan 敦煌吐蕃文獻選 (Peking: Min-tsu ch'u-pan she, 1983), p. 162 and Paul Pelliot, ‘Notes sur les T'ou-yu-houen et les Sou-p'i’, T'oung Pao, vol. 20 (1921), pp. 323-30.
Nicholas Poppe, Grammar of Written Mongolian, p.2; Na-shun ta-lai那順達來, Niakan Daor Bulku biteg (Hu-ho-hao-t’e: Nei Meng-ku ta-hsüeh ch’u-pan-she, 2001), p. 128.