155. CIN 265
«Sères» (HERRMANN, « Seres » in PAULY-WISSOWA ; Das Land der Seide, 25); but it does not seem certain that the use of anpocev is not here due to Strabo himself. In the 2nd cent. A.D., Pausanias (vii, 26, 6) speaks of silk as produced « by a small animal which the Greeks call arip, although the Seres themselves give it another name ». KLAPROTH'S suggestion (in 1822) that arip renders the Ch. a ssû (*si), «silk », especially « silkthread », has generally been accepted, and HERRMANN (Das Land der Seide, 26) considers it is « so evident that no doubt ought to exist any longer on the point ». Yet there are certain difficulties. LOKOTSCH (Etymol. Wörterbuch, No. 1878) derives from Ch. ssû both Med. Lat. sèta (> It. seta, Fr. soie, Germ. Seide), and, with the addition of the Q, êrh suffix of Northern Chinese (ssû-êrh), «Sères» (> Lat. sericus, Fr. Engl. serge, Engl. silk, Russ. sëlk). I do not for one moment believe that the Ch. ssû could have developed (through what channels?) into a Med. Lat. séta, «silk », which must merely be the outcome, with a change of meaning, of Lat. saeta > sèta, «bristle », « coarse hair » (cf. Fr. « soie de port », hog-bristles). On the other hand, although the use of Q, êrh as a suffix is fairly early, going back at least to the 9th cent. (cf. LAUFER, Sino-Iranica, 538), êrh is an ancient *nzie, the initial of which is transcribed z- in mediaeval foreign scripts; ssû-êrh ought never to have been adduced to explain arip. More important are the Corean form sir of ssû, and Mong. sirkäg (<_ *sirkäg > Kalm. sirkac and sirgac), Manchu sirge, « raw silk », « silk thread ». I do not know of any form similar to sirkäg or sirge in mediaeval texts or vocabularies, but that does not prevent the words from being possibly ancient. LAUFER (IOC. tit., 538-539) was, in my opinion, mistaken when he denied a connection between Cor. sir and Ch. ssû. Cor. sir stands to Ch. ssû (*si) in the same relation in which Cor. mar does to Ch. ,, .1 ma (*ma), « horse ». Although the Ch. *ma shows no final consonant c. A.D. 600, and must not have had any for some centuries before that date, it is extremely probable that the -r of Cor. mar is etymological, and that the word is fundamentally connected with Mong. morin, « horse ». So both ssû and ma may originally have ended with an -r, which was dropped in the first centuries of our era; other words, such as nffi shih and rff chiu, seem to be in the same case. But even then the connection between ssû and arip is far from established. Pausanias gives grip as the Greek name of the
silkworm, different from the Chinese name. As a matter of fact, the Chinese name is ts'an
(*dz'âm), and there is no apparent reason why the silkworm should have been known in Greek by the Chinese name of the silk thread. Another difficulty is that air/10 occurs only in the 2nd cent. A.D., two centuries at least after the appearance of the derivative forms «Sères» and « arl ptx6v » in Latin and Greek texts. Lastly, there is no other Greek or Latin word which can be traced directly to a Chinese original. Even if it be alleged that the case is exceptional, silk being in ancient times the Chinese product par excellence, the word would not have reached the Greeks directly, and we are at a loss to understand how it could have passed and been so well preserved phonetically. This explains why LAUFER rejected the derivation of an'p from Ch. ssû. But while he thought « ar' p » and « Seres » to be of Iranian origin and connected them with Pers. säräh, « breadth of white silk » (=' Ar. saraq, « silk », « white silk ») in Sino-Iranica, 539 (cf. also Y', I, 20), he has given up this theory in the index of the sanie work (p. 612) : there he says that he thought he had meanwhile found what he believed to be the « correct derivation » of the name Sères. However, he does not seem to have ever published it.
IIIF Color HighRes
