国立情報学研究所 - ディジタル・シルクロード・プロジェクト
『東洋文庫所蔵』貴重書デジタルアーカイブ

taka and Cosmas' tupha), and LAUFER, Sino-Iranica, 564-565. As a rule, the tuγ was made with
the tail of a yak, and I think that the same word occurs in Tib. thug-rña, « tury-tail », and thug-
chom, « tury-bunch », both applying to a flag made of a yak's tail. For the use of the word in the
Uighur legend of Uyuz-khan (or Oγuz-khan), cf. TP, 1930, 291. In modern times, horse-tails were
substituted for yak-tails in the Ottoman empire, and the number of tails varied according to the
rank of officials, from one for the mīr-līwā to 7 or 9 for the sultan when he went to war (cf. « Tūgh »
in EI); this, to a certain extent, goes back to an ancient tradition, as the qayan, according to Kāšyarī
(BROCKELMANN, 216), had 9 tuy. But the Mongol custom was somewhat different. The word tuq
(= Turk. tuγ) occurs several times in the Secret History of 1240 (§ 181, 193, 202, 232, 278), where
we find even (§ 73) a verb tuqla-, « to raise the tuγ ». It is the standard of Chinghiz-khan, and it
is described (§ 202) as a « white standard with nine tails » (yäsün költü čaqa'an tuq). Muqali,
Chinghiz-khan's lieutenant-general in Northern China, had been granted the privilege of a standard
with nine tails, on which a « black moon » (a black crescent) was figured (cf. TP, 1930, 32). Accor-
ding to Polo (in RAMUSIO), Qubilai's standard wore the figures of the sun and the moon. Before
starting for a war, the tuγ was raised and « sprinkled » (saču; Secret History, § 193). The flag and
the drum were ancient regalia; as 旗 ch'i and 鼓 ku, they are often associated as gifts of investiture
from Chinese emperors to nomad princes, and this was the meaning of the high distinction granted
by Tamerlane when he promised a tuγ and a drum to every general who would conquer a kingdom
(cf. Y, I, 263).
It cannot well be doubted that the word tuγ, as ABEL-RÉMUSAT already maintained, is identical
with the Chinese 纛 tu (*d'âu, *d'uk, *d'uok) of the same meaning, which occurs already in Chinese
before the Christian era; and LAUFER is probably right in saying that it was borrowed from the Chinese
by the nomads at an early date.
But although the tuγ was raised in time of war and was the standard of the Emperor, and
sometimes of the commander-in-chief, no text gives it in the sense of « 100,000 » attributed to it
by Polo. I wonder whether some confusion of words did not occur in the traveller's mind. There
is in Turkish and in Mongolian an expression tük tümān; it has been met with half a dozen times in
the « Turfan » Uigur texts, and occurs also in the Sino-Uigur Mss. vocabulary belonging to the School
of Oriental Languages; in this last work, it is translated by wan-wan, « ten thousand times ten thou-
sand », which would be 100,000,000. Such is also the translation of tük tümān in the Sino-Mon-
golian vocabulary Hua-i i-yü of the end of the 14th cent. But KOVALEVSKIĬ (p. 1926) registers
it as the Mongolian equivalent of Skr. lakṣa, Tib. 'bum, and translates it by « 100,000 ». I suspect
that in RADLOV's « köp-tümān », « tüm tümān », « tüp tümān » (III, 1602), the forms — at least those
taken from the Qutaδγu bilig — are misread for tük tümān. In any case, the tük of tük tümān
cannot be an « intensive » of the type of qap qara, « quite black », ap aq, « quite white », as we have
tük ming, « a tük of thousand », in Qutaδγu bilig (wrongly read tök ming in RADLOV, III, 1241).
Although tük, by itself, cannot have been a definite number, it may be that Polo, who knew of the
tuγ of the whole army, but had to give a name for « 100,000 » along with that of tümān for
« 10,000 », adopted « tuc » for « 100,000 » owing to the frequent association of tük with tümān in
Turkish and Mongolian.