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

tive names with the equivalent of a French « ou », meaning « or », and not with a Persian copula u,
« and ». In Rustichello's French (at least as we have it in F), that « ou » ought to be written « ou »
or simply « o » (cf. for instance B, 20⁵, 20²¹⁻²², and the notes), but we do not have a correct arche-
type, and the whole work shows uncertainties of o and u in proper names. Whether by Rustichello's
fault or by that of a very early copyist, that o (or ou) was misunderstood and the result is the pseudo-
compound name. I may add that, while all Mss. give u here, Polo seems to have still heard as o
what is now pronounced as u in Persian; so he writes « Cobinan », and also « Tunocain » for « Tun-u-
Qāïn ». This last name has precisely the Persian copula u which he heard as o, not as u in the would-
be « Gheluchelan ».
Up to this point I am pretty certain of the solution, but I am not so positive about the true
forms used by Polo for the alternative names. In spite of the Mss., and taking into account the
constant confusion between c and g, I would myself have little hesitation in writing the second name
« Ghelan » = Gīlān. But it is not clear that the first name should be spelt « Ghel » = Gīl. Most
of the Mss. suggest « Gel » or « Giel », which would transcribe the Arabic form Jīl; and although
Polo's nomenclature is mainly Persian, he may have quoted here both an Arabic form, and a Persian
one; « Ghel » remains nevertheless possible. The Catalan Map gives « Gellam », « Cilam » and
« Cillam »; the name is « Zilan » on the Bianco Map, « Zila » on the Genoese Map; Schiltberger names
« Gilan » (HALLBERG, 217-218, 571); Clavijo, « Guilan »; I read « Gillā » on Fra Mauro's map (not
given by HALLBERG).
From the country named Jīl or Gīlān comes, according to Polo, a certain kind of silk, the name
of which is written « gelle » in F, « guele » in FB, « grelle » in FA and FA², « ghellie » in R.
Gīlān was famous for its silk, and the name is of course derived from that of the country (« Ghel,
the Caspian » is a slip in Pe, 167). I find « seta guieli » in a Genoese document of 1274 (BRĂTIANU,
Le commerce génois, 306). The seta ghella, as YULE has said, is mentioned by Pegolotti (EVANS,
208, 297, 300); Clavijo mentions the silk of « Guilan »; as late as the 17th cent., « gilam », as the
name of the Gīlān silk, appears in P. VAN DAM, Beschryvinge van de Oostind. Compagnie, 2nd
part, Index, I, 695. But which was the form used by Polo? BENEDETTO has printed « g[h]elle »
in F, and he is probably right. Yet it is also possible that RAMUSIO corrected the name according
to a form which had obtained in Italy after Polo, and which is represented by Pegolotti's « ghella »;
Polo's Mss. seem to be almost as much in favour of « giele » or « gielle ». If I incline to BENEDETTO's
opinion, it is because the name in Polo is more likely to be Persian than Arabic, consequently Gīlī
rather than Jīlī. I have only to observe that Ya'qūt gives, as an adjective of origin, the two forms
jīlānī and jīlī, but adds that jīlānī has been reserved for the products of the country, and jīlī fo
the men born there (cf. BARBIER DE MEYNARD, Dict. Géogr., 187). The Western use is not in
agreement with this.
YULE added (Y, I, 59) that « seta ghella » appeared also in Uzzano with what YULE calls
« an odd transposition », to wit « seta leggi ». Since this was not corrected by CORDIER, it may
be worth remarking that HEYD (II, 671) has already given the true solution : « seta leggi » is the
silk not of Gīlān, but of لاهجان Lāhājān (also Lāhījān; cf. Mi, 497). This silk is praised by
Abū-'l-Fidā (II, II, 173) and by Ya'qūt (BARBIER DE MEYNARD, Dict. hist., 503); the adjective of
origin from Lāhājān was probably lāhājī, of which leggi is a fair transcription. Apart from