^{page 485 note 1} Brandes, J., “Nâgara Krětâgama” (Verhandelingen van het Bataviaasch, Genootschap van Kunsten en Wetenschappen, deel liv, Batavia, 1902).Google Scholar

^{page 485 note 2} “En Ond - Javaansch geschiedkundig gedicht uit het bloeitijdperk van Madjapahit,” in Indische Gids, March, 1903, pp. 341–360.

^{page 485 note 3} Mājapāhit was founded some time between 1278 and 1292, probably nearer the latter date.

^{page 486 note 1} Dr. Brandes' speech in the compte rendu of the “Premier Congrès International des Études d'Extrême Orient, Hanoi, 1902,” Hanoi, 1903. Also, Bulletin de l'École Française d'Extrême Orient, t. iv, pp. 344, 345, 475.

^{page 486 note 2} This name makes its appearance in Cham inscriptions in a.d. 1159, but occurs nearly two centuries earlier on the Khmër inscription of P‘hūm Mīen (province of Thbōng Khmum, eastern Kamboja), which, under a date corresponding to a.d. 987, mentions Yvan (Yūen or Yavan) settlers, trading, among other things, in slaves. It might, in this instance, be a question of Arabs or Moorish merchants (cf. Mahāvaṁsa, ch. 76, v. 268, date about 1180), though this is made somewhat doubtful from the fact that Annam, in a.d. 968, had regained independence, which event naturally led to a revival of trade with foreign countries. We are told, in fact, that not long afterwards, in 1140, she opened her ports to ships of all nations.

^{page 487 note 1} This designation dates back from at least the thirteenth century, and applies then more particularly to the territory of C‘hīeng Sn further to the north.

^{page 487 note 2} “Cāmadevī-vaṁsa,” by Bodhiraṁsi-Mahāthēra, composed about the end of the fifteenth century, ch. xii, under date corresponding to a.d. 924 : “Tadā ēko Sujito nāma rājā Siridhammanagarē kāretvā,” etc. The “Jinakāla Mālinī,” composed in Pāli at C‘hīeng Mãi in 1516, by Ratanapaññañāṇa Thēra, alludes to the same circumstances. The older form of the city's name thus appears to be Śri Dharmanagara.

^{page 487 note 3} Among which the too often unreliable Balfour's “Cyclopædia of India,” 3rd ed., vol. ii, p. 711; Professor Keane's “Geography of the Malay Peninsula,” etc., 2nd ed., London, 1892, p. 17 ; and so forth.

^{page 487 note 4} “Description du Royaume Thai ou Siam,” Paris, 1854, t. i, pp. 26, 27. If I am led to go into such particulars in order to refute an obviously absurd statement, it is because error dies hard, as experience has taught me in my turn. I have, for instance, years ago pointed out, among other matters, that the term Śyām (Siām) has existed as the name of a country and people for at least nineteen centuries, and that Chām inscriptions of the first half of the eleventh century testify to the presence in Indo-China of such a country and people at that date. Yet I have seen in recent publications by writers whom one would expect to know better, the absurd and worn-out statement repeated, that the term Siam was invented by the Portuguese in the sixteenth century !

^{page 488 note 1} Journal Asiatique, 1903, pp. 228 et seqq.; and “Le Cambodge,” t. iii, Paris, 1904, pp. 659 and 724–733.

^{page 488 note 2} I have, a few days since writing the above, confuted it in full, by simply availing myself of such documents as are in print, and therefore accessible to everyone (though ignored by M. Aymonier), at a meeting of the Siam Society, Bangkok, on the 1st March, 1905. (See Bangkok Times of March 2nd for a summary, and forthcoming number of the Journal of the Siam Society for a fuller account.)

^{page 488 note 3} Cf. Groeneveldt, , in “Miscellaneous Papers relating to Indo-China,” 2nd series, vol. i, London, 1887, pp. 194, 195.Google Scholar

^{page 489 note 1} So (Siṅgapurī) in the Kaṭa Mandirapāla, and in the law on the status of provincial governors of circâ 1454: see Laws of Siām, vol. ii, p. 93, and vol. i, p. 203 (Siāmese ed.).

^{page 489 note 2} It is met at times in old records with the spelling Swankhaburī (Svarga-purī), which is incorrect.

^{page 489 note 3} See Anderson's “English Intercourse with Siam,” p. 41, according to which Baldæus says: “betwixt Tanassery [Tenasserim] and Oceeda (Quedah), towards Malacca, are the harbours of Tanangar, Sencaza, and Perach, opposite to Achem.” In order to clear Baldæus of blame, it would be necessary to demonstrate that Perach is a misprint, or lapsus calami, for Perlis, in which case Sencaza would have to be looked for between Perlis and Trang. There is a little stream named Kacha (Khlōng Kacha) debouching on that tract of coast through the Lawang estuary (70° 9′ N. lat.). This may have of old borne the name Sungei Kacha, of which Sungi-kacha, Sencaza, would be possible contractions. In default, there is nothing left but the Kesang River below Malacca and immediately above the Muār, which appears in old European accounts as the Gaza, Jyga, Kroisant, and Krisarant (Dutch), Cação (Portuguese), etc. Nieuhoff was wrong in thinking it to be the Muār, and Dennys in not rectifying that blunder in his “Descriptive Dictionary of British Malaya,” London, 1894, p. 208. But whichever of the two here proposed be the correct location of Sencaza, this evidently cannot be the Syangka of the Nāgarakretāgama.

^{page 491 note 1} So the annals of Pegu, Siām. transl., p. 203.

^{page 491 note 2} In locating these States in Siām, I am of course aware of the existence of several similarly named cities in India, e.g. Ayodhyā = Oude, still alluded to as Ayujjha in about a.d. 1052 in Mahāvaṁsa, ch. 56 ; Sīhapura of both Dīpavaṁsa and Mahavaṁsa; Rājapuri = Rajaurī of the Rājataraṅgiṇī (viii, 617, a.d. 1118); Raipur in the Central Provinces, and so forth. But it will be evident to everyone that, owing partly to the great distance and partly to the non-existence of some of such cities at the period we are concerned with, they are entirely out of question. Invasions of Malays (called Javaku in the Siṅhalese chronicles) occurred, it is true, on the coasts of Ceylon and Southern India in 1251 (cf. Mahāvaṁsa, ch. 84) and earlier ; but these freebooters came, I think, from Sumatra, and as the range of their exploits did not extend beyond the Coromandel coast, it is unlikely that relations could be established by them with the States further to the north.

^{page 492 note 1} Annals of Pegu, Siām. transl. (“Rājādhirāj”), p. 10.

^{page 492 note 2} It goes without saying that this is the State of Ma-li-yü-êrh, which Sinologists have placed on the territory of Palembang, east coast of Sumatra; as well as the hitherto vainly sought for Maliur or Malavir of Marco Polo. I cannot go here into the long discussion that the subject would entail, especially as I have fully made it elsewhere in a work now being passed through the press. Suffice to point out, as some of my witnesses, the river Malāyu (Sungei Malāyu), still so called, and the village Bentan (probably connected with Marco Polo's Pentam), both lying there (ignored by all my learned predecessors), on the northern shore of the Old Singapore Strait.

^{page 492 note 3} “Sejarah Malāyu,” Leyden's transl., p. 73; and Marre's, “Histoire des Rois de Pasey,” Paris, 1874, pp. 48–50Google Scholar, which, however, takes a far more rosy view of the matter.

^{page 493 note 1} Defrémery, & Sanguinetti's, “Voyages d'Ibn Batoutah,” t. iv, Paris, 1858, pp. 230 and 306.Google Scholar

^{page 493 note 2} Groeneveldt, in “Miscellaneous Papers relating to Indo-China,” 2nd series, vol. i, pp. 243, 245, 248, etc., etc.

^{page 494 note 1} Cf. China Review, vol. xxiii, p. 256; and Asiatic Quarterly Review for January, 1900, p. 135, where 1436 is doubtless a misprint for 1406.

^{page 494 note 2} Cf. Marre, op. cit., p. 97.

^{page 495 note 1} Now pronounced chrēi by the Khmërs, and srôa, or soa, by the Mōñs. Kedah is down to the present day called Srai (Müang Srai); officially, Sai-burī (Srai-purī) by the Siāmese.

^{page 496 note 1} See Journal of the China Branch R.A.S., vol. xxi, map, and p. 38, No. 7.

^{page 496 note 2} Bulletin de l'École Française d'Extrême Orient, t. iv, pp. 328, 345, etc.

^{page 496 note 3} Journal R.A.S., 1896, p. 478.

^{page 496 note 4} T‘oung-Pao, 1901, p. 129.

^{page 497 note 1} T‘oung-Pao, 1901, pp. 125–134 ; and 1898 (vol. ix), pp. 402–406.

^{page 499 note 1} As regards Tan-ma-ling, there is a river Tambilang on the east coast of Sumatra in 2° S. lat. From the position described for Fo-lo-an in relation to neighbouring countries (T‘oung-Pao, ix, p. 404), it would appear that Berūan, on the north coast of Sumatra, is the most likely place, and the sailing distance from it to San-fo-ch‘i may be merely meant to the northern borders of this State.

^{page 499 note 2} Equally impossible is the rapprochement made by both M. Pelliot and M. Huber, in the same number of the Bulletin (pp. 407 and 475), of Ling-ya-ssŭ-ka with Lang-ya-hsiu, . In the Asiatic Quarterly Review for January, 1901, pp. 157, 158, I have conclusively shown that the latter stood on the east coast of the Malay Peninsula and on the territory of the present C‘hump‘hōn (about 10° 30′ N. lat.), where the name survives to this day in the two islets of Langkachu, nearly in front of C‘hump‘hōn Bay. These, I have now no doubt, are the very ‘mountains’ () of Lang-ya-hsiu sighted in a.d. 607 by the Chinese embassy to Ch‘ih-t‘u (= Sukda, Sukhada, later on Sukhodaya, in Central Siām); while C‘hump‘hōn harbour and district is I-tsing's , Lang-ka-hsü, as well as the , Lang-ya-hsiu of Liang history. It is amusing to see Sinologists go on suggesting imaginary locations for place-names which have already been identified with absolute certainty and shown to correspond to actually existing places.

^{page 500 note 1} Journal Str. Br. R.A.S., No. 42, p. 200.

^{page 500 note 2} Cf. Leyden's, “Malay Annals,” London, 1821, pp. 42–44.Google Scholar

^{page 501 note 1} See Journal R. Asiatic Society for July, 1897, table x, at foot and on the right - hand side. There I suggested the Tong - si - tiok of the Chinese as a probable equivalent, relying on Groeneveldt, who (op. cit., pp. 258–9) identified it with Singapore Island. But when I began to feel out the way for myself, I at once recognized Tamasak in the Tan-ma-hsi above referred to, and corrected the mistake in a new monograph still in the press. M. Pelliot, who, I am glad to notice, proceeds far more cautiously and with more critical acumen than his predecessors in his new inquiries on these subjects, recently suggested, in his turn (op. cit., p. 345, n. 4), the probable identity of the Tumasik of the “Nāgarakretāgama” with the Tan-ma-hsi of the Chinese map published by Phillips, from which latter he argued its location to be about the site of the present Johore. He may now see, however, that it is more precisely Singapore Island, the hill represented on that map being unquestionably Būkit Tīmah. The Old Singapore Strait is not shown there, as scarcely any longer used by Chinese junks at that time. It appears that the Chinese discovered the new passage on or about the end of the fifteenth century, and therefore at least two centuries before the Hispano-Portuguese. The new channel is, in fact, duly marked in the map in question, the date assigned to which by Phillips I see no reason to dispute.

^{page 501 note 2} In Khmër a rock is also called t‘mō or thmō; but we cannot explain the name by the Khmër language except by admitting a form Bā-T‘mō, meaning ‘excellent rock,’ ‘sacred rock,’ which may have been the name applied to the great mass of unhewn coarse red silicious sandstone above alluded to. I do not positively assert that Bī- Thmôḥ was the name of Singapore River, or, for that matter, of the Kallang or Rochor streams flowing close by. It is a mere conjecture, though, as may be seen, not altogether unfounded. I do not, in fact, say that Bī-Thmôḥ cannot have been used as the name of the Old Strait; for bī in Mōñ is a rather elastic term, it being used to denote, besides a large river, an arm of the sea, and the sea itself (termed bī-c‘hnôk = great river). The Khmër name for ‘tin’ is samnō; also pahang, whence the name of Pahang may have been derived (do those who talk about Panggang tribes in those parts know this?), though the reverse may, after all, be the case. But I do not think that samnō is the prototype of either Tuma or Tamara. Many toponyms on the west coast of the Malay Peninsula, and even on North Sumatra, are unmistakably Moñ-derived ; hence my conjectural etymology for Be-Tūmah = Bī-T‘môḥ. This is further supported by the fact that the ancient pronunciation of Bī seems to have been Bē, for Ptolemy spells with Bη all the toponyms on the west coast of the Malay Peninsula of which the Mōñ Bī forms the initial syllable; even to-day the sound of ī in Bī somewhat inclines towards a closed ē. It ia not improbable that the old Bη or Bē still survives in the Straits and neighbouring islands under the somewhat modified form Wē (meaning ‘water,’ and perhaps ‘rivulet’), which occurs in several toponyms, e.g. Pulo Way, etc.

^{page 502 note 1} Or Tamarum, see Strabo, lib. xi, 7. It is the Promontorium Samara of the planisphere of the fourteenth century in Hereford Cathedral, which bears at this point the explanation: “India quæ finem facit.” Santarem has noticed (“Essai sur l'histoire de la Cosmographie,” etc., t. ii, p. 343) that the change in nomenclature from Tamos, Tamus, or Tamarum into Samara took place on mediæval maps in the fifth century.

In this connection it is interesting to observe that, according to the Chinese annals of the Liang dynasty, during tho first quarter of the third century a.d. Fu-nan (Kamboja) conquered a number of places on and about the southern part of the Malay Peninsula. Among such were:

(1) , Ko-ying, which I take to be Krian in Pêrak, nowadays denoted by the Chinese in the Straits as , Kau-yen or Ko-yin (see Journal Str. Br. R.A.S., No. 42, p. 187); and

(2) , , or , Tien-sun or Tun-sun, a State situated at over 3,000 li (circâ 500–600 miles) from the southern borders of Fu-nan, and which therefore Sinologists have identified with Tenasserim ! If not in name connected with Tamsak or Tumasik, this State, the territory of which is said to be only 1,000 li (say, 180–200 miles) in extent, and to project in a curviform direction into the sea, cannot have been far from it on the Malay peninsula. The chief city is said to have stood at 10 li (about two miles) from the sea, and to have been a great emporium—a gathering-place, in fact, for traders from east and west, just as Singapore is described afterwards by De Barros. The name recalls the Malay term Dūsun, meaning an ‘orchard,’ but also a ‘village,’ or the ‘country’ as distinguished from the town, and is common to several places on the Malay Peninsula, besides being applied to certain tribes in North Borneo calling themselves Kadasan.

Now, in view of the conquests of Fu-nan in the south of the Malay Peninsula, it is not altogether impossible that the name Tamara (= ‘tin’) of Singapore Island was changed into Samnō, the corresponding Khmër word, whence the Samara of European cartographers was afterwards derived (in the fifth century or earlier). Of course, this is a mere suggestion; the change, if it ever occurred, did not last a long time locally, for the Arab navigators of the ninth century again employ the form Tuma. But among our geographers the variant Samara would naturally persist longer; and to this circumstance is perhaps due the fact that Marco Polo, when speaking of the petty State of Samudra on the north coast of Sumatra, spells the name Samara.

The Sanskrit inscription on the Bn Thāt (Dhātu) stele near Bassac (Campāsak, Upper Kamboja), erected by the warlike king Sūryavarman II (a.d. 1112–1152 circâ), and published by Professor Kern (Annales de l'Extrême Orient, t. iii, pp. 65–76), mentions an expedition undertaken by that famous potentate to the “Land of Elephants and Copper,” Dvipatāmra-dēśa, by which “he eclipsed the glory of victorious Rāghava (Rāma)”:—

“So' yaṁ prayāya Dvipatāmrade[śaṁ]

Raghuñ jayantaṁ laghayañcakāra” (v. 35).

Professor Kern thinks the island of Ceylon is meant, which is not altogether unlikely in view of the allusion to Rāma's exploit in the above lines, and also of the fact that a few years afterwards (circâ 1170–80) the Ceylon king Parakkama Bāhu sent a princess as a gift (or tribute?) to the ruler of Kamboja, the son or other successor of Sūryavarman II (cf. “Mahāvaṁsa,” ch. 76, v. 35).

I would point out, nevertheless, that it is not impossible that Singapore Island be meant, in which case Tāmra should be taken as a lapsus, whether intentional or not, for Tamara. Singapore Island is much nearer to Kamboja than Ceylon, and has doubtless been at some time or other under Kambojan sway; whereas, in respect to Ceylon, no such expedition is recorded in local chronicles, and no such name as Dvipatāmra, the nearest one to it being Tāmra-parṇī or Tamba-paṇṇī, unless we take the term Nāgadvīpa, applied to one portion of that island, to mean ‘Elephant Isle’ (or District; Ptolemy mentions, by the way, feeding-grounds for elephants on its territory).

I am, notwithstanding this, under the impression that the Dvipatāmra-dēśa of the inscription above cited may, after all, mean Lān-c‘hạng (‘Elephant plains’), i.e. Eastern Lāos, which, besides being the traditional land of elephants, is also that of copper.

^{page 504 note 1} The “Tung-hsi-yang-k‘ao”(publ. 1618) still mentions, as M. Pelliot observes (op. cit., p. 345, n. 4), the Strait of Tan-ma-hsi, (Tan-ma-hsi Mên), as being passed by junks at that time. If this information is taken from old records, the Old Strait may be the one meant; but if gleaned from contemporary sources or accounts not earlier than the fourteenth century, the new passage would be intended, in which case the existence of the term Tamasak or Tumasik might be traceable to a yet more recent date than could be argued from the evidence we have examined above.

^{page 505 note 1} It is in the Journal of the Straits Branch R.A.S., No. 42, p. 153, that I have noticed for the first time the use of the character instead of the one that has so far obtained in Chinese publications.

In his study of an itinerary through the Straits recorded by Chia Tan in circâ a.d. 785–805, M. Pelliot (op. cit., p. 231), following Chavannes, takes the Strait of , Chih (or Chöt, Chit), mentioned therein, to be the Strait of Malacca; but it appears to me that either the new Singapore passage or the Old Strait are more likely meant, in which case we would have in a pretty old prototype of the present and ; if not, possibly an evidence as to the existence, at such an early period, of the suffix sik or sak attached to the name of Singapore Island.

^{page 506 note 1} Or, as some Malay scholars would have it, the Malay name of Singgah-pura, meaning ‘a place of call,’ from , singgah, ‘to visit,’ ‘to call in.’ But this term is certainly not Malay: cf. siṅghāṭaka = a market-place, in “Questions of King Milinda,” Sacred Books of the East, xxxv, pp. 2, 53, and xxxvi, p. 279, n. 1. I should think, moreover, that Siṁha-pura is the really correct form of the toponym. The derivation given in “Hobson-Jobson” (2nd ed., p. 839), from singah + pora-pora, is inadmissible.

^{page 506 note 2} Remains of an earthen wall and other relics were also discovered, including an inscription in characters resembling those of ancient Java, on a rock since blown to pieces.

^{page 507 note 1} This I take to be a veiled hint to the fact that the king of Bintang had probably been taken prisoner to Siām, whence he appears never to have returned. At all events, he must have gone to Siām in order to pay homage, or to arrange matters that the interference of that country had made somewhat critical for him.

^{page 508 note 1} Friar Odoric has Malamasmi in Ramusio (“Navigationi et Viaggi,” vol. ii, 1583, fol. 247 verso), which may be compared to the Malanir, Malavir, and similar variœ lectiones in the texts of Marco Polo.

^{page 509 note 1} Right opposite the mouth of the Sungei Selitar, on the northern shore of Singapore Island.

^{page 509 note 2} Colonel Yule's genius, which has elucidated so much of Marco Polo's text, seems to have grown dim in the course of his treatment of the Venetian traveller's route in the southern seas (especially for the portion comprised between the south borders of China and the north coast of Sumatra, which I consider the least satisfactory portion of that monumental work). Nor has Cordier, who, in my opinion, misunderstood that part also of Friar Odoric's itinerary, succeeded in throwing any further light on the subject in his recent edition of Yule's “Marco Polo.” Both scholars have been misled by De Barros’ and Valentijn's mention of a river Malāyu in the interior of Palembang, which these writers believed to have been the cradle of the Malay race; as well as by those Sinologists who located I-tsing's Mo-lo-yu (lying, according to this author, at fifteen days’ sail from Palembang) in the valley of that very stream Malāyu in the interior of the country, or else in all sorts of other impossible places which have naturally been adopted also as the site for the Ma-li-yü-êrh of later Chinese historians.

I have neither space nor leisure to go here into Marco Polo's and Friar Odoric's itineraries in the Southern Seas, and must accordingly defer the treatment of them to another occasion. All I can add for the present is this:—

1. Marco Polo's channel, where “there is but four paces’ depth of water,” so that great ships, in passing it, “have to lift their rudders” (Yule's, “Marco Polo”, 3rd ed., vol. ii, p. 280)Google Scholar, is unmistakably the Old Singapore Strait. There is no channel so shallow throughout all those parts except among reefs.

2. The island of Pentam cannot be either Batang or Bitang, the latter of which is likewise mentioned by Marco Polo under the same name of Pentam, but 60 + 30 = 90 miles before reaching the former. Batang, girt all round by dangerous reefs, is inaccessible except to small boats. So is Bintang, with the exception of its south-western side, where is now Riāu, and where, a little further towards the north, was the settlement, as we have seen, at which the chief of the island resided in the fourteenth century. There was no reason for Marco Polo's junk to take that roundabout way in order to call at such, doubtlessly insignificant place. And the channel (i.e. Rhio Strait) has far more than four paces’ depth of water, whereas there are no more than two fathoms at the western entrance to the Old Singapore Strait.

^{page 510 note 1} See also “Cingaporla, che è il capo,” as well as a city, in Pigafetta, 1522 (Ramusio, op. cit., vol. i, ed. 1563, f. 369 recto).