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

the easy phonetic interchange between -g- and -v- (-w-). As to the vocalic differences, the alternation
between Arabic Lawāk, Ch. Lâ-yuk (> Lo-hu), Siamese Lavo (> Luvo) and Lopburi, gives ample
justification for Polo's « Lochac » (? *« Logac »).

But, if I agree with YULE as to the onomastic explanation of « Lochac », I do not think he was
right in thinking of Ligor, no more than Pe, LVII, or RR, 425, in speaking of the Malay Peninsu.la
or B¹, 444, in proposing Trengganu. I consider Polo's notices on Java, on « Lochac » and on
« Malaiur » as digressions which he introduces when his itinerary passes points from which visitors
can go to these countries; it is, in fact, what he has done so many times over in the account of his
outward journey to China. He had heard of Siam as a country producing an abundance of brazil-
wood, ebony, and cowries; he adds that it is a wild region, and that few travellers go there; it seems
plain to me that he does not pretend to have been one of those few. From the « Sondur » and
« Condur » islands (see « Condur »), Polo must have steered a straight course, as was usual, to
« Pentan » (Bintan), so that any question as to the extent of the Siamese domination over part of
the Malay Peninsula at the end of the 13th cent. has no bearing on the position of « Lochac ». In
the present case, I incline to the following view. The chapter on Java represents a digression,
after which the itinerary ought to start again from Champa, but all the Mss. make it start from Java
to go to the islands of « Condur » and « Sondur ». This has been retained in RR, 280 and in B¹,
291, although BENEDETTO admits in a note that it involves grave difficulties and even absurdities
« almeno per noi moderni ». I think they were just as grave in the 13th cent., and that we should
read Champa for Java, as has been done by MARSDEN, PAUTHIER and YULE. In my opinion, either
Rustichello or the early copyists were responsible for the change, because they were unaware of
the digression, and wished to give a connected account of the itinerary from place to place.
The same thing has occurred, in a somewhat different way, about «Malaiur», and two or three
times in the itinerary from Cho-chou to « Quinsai ». In the present case, Polo places his notice
on « Lochac » at the mention he makes of the islands from which ships would turn off towards
Siam, and afterwards his account starts again from « Condur ». But the « intelligent » redactor
or copyist seizing this opportunity, must have given as the distance from « Lochac » to Bintan what
was in reality given by Polo as the distance from « Condur ». As to the direction south-east from
«Condur» to «Lochac», it is a mistake whatever solution one may adopt for « Lochac »; but errors of
direction often occur in Polo's text.

In the description of « Lochac » (Vol. I, 369), F states that there is in that country plenty of
« be[r]çi domesce », and Z also speaks of « berci domesticum ». « Berçi » is brazil-wood (q. v.)
or sappan, a product known in Siam from early days. To brazil-wood VB alone adds musk and
ebony. RR and B¹ have accepted ebony in their text, but have left out musk. In a note (B¹,
433) BENEDETTO explains that the copyist of VB may have misinterpreted « domesce », perhaps
wrongly divided « do mesce », and misread it into his « muschio ». There has certainly been a
contamination between « domesce » and « muschio », but at first I was not certain in which way
it had taken place. There is no musk in Siam, but that country traded at an early date in musk
obtained from Yün-nan (cf. LAUFER, Sino-Iranica, 469). Civet-musk might also be meant here
(cf. Fe, 545), or, though less probably, the Indo-Chinese « musk-tree » (麝香木 shê-hsiang-mu),
on which cf. FUJITA, 17 a. On the other hand, BENEDETTO's literal translation does not explain