5. Southeast Asia maritime exchange

Over the past decade prolific archaeological missions in coastal and mainland Southeast Asia (Burma, Thailand, Malaysia, Indonesia, Vietnam, Laos, Kampuchea) have brought to light artefactual evidence of early trans-oceanic contact with India, the Mediterranean and China. The results from the prospections have pushed back the chronology for early Southeast Asian contact with the Indian civilization to mid-1 st millennium BC. The central data in this regard has emerged from the cemetery-site of Ban Don Ta Phet (BDTP) in Thailand (Figs. 20, 38) Excavations conducted here jointly by the Institute of Archaeology (London) and the Thai Fine Arts Department has revealed Buddhic funerary deposits. The Indian material from the cemetery complex comprises semi-precious and glass beads, green glass cut into shape of beryl, bronze vessels with a central protusion or 'knob' at the inner base and votive stupas made of clay. Five radiocarbon dates provide a chronology of 390-360 B.C. for the funerary deposit at BDTP (Glover 1990; Glover 1996:138). The 4 th century B.C. context provided by BDTP is the earliest chronology available for Southeast Asia - India contact. No other site in Southeast Asia has

200 600 1000 LOMETRES BIHAR PRET CHINA 37 VIETNAM BAY of BENGAL SRI LANKA. MALAYA 30121 20 27 SUMATRA PALAWAN BORNEO ROULETTED WARE . KNOBBED VESSEL D HIGH TIN BRONZE X ETCHED BEADS 2 LUZON PHILIPPINES MINDANAO TALAUD ISLANDS SULAWESI HALMAHERA MOLUCCAS 0 CERAM .13 JAVA BALI 0092 TIMOR • South and Southeast Asia locating the more important archaeological sites referred to in the text and the distribution of Rouletted Ware, knobbed-base vessels, high-tin bronzes. Only the etched beads found in Southeast Asia are located. Key to the numbered sites 1. Arikamedu, 2, Kok Charoen, 3, Ban Kao, 4, Ban Na Di, 5, Tha Khae, 6, Ban Chiang. 7, Non Nok Tha, 8, Phung Nguyen, 9, Pong Tuk, 10, Oc-eo, 11, Ban Don Ta Phet, 12, U-Thong, 13, Buni, 14, Chansen, 15. Beikthano, 16, Khlong Thom, 17, Saaburi, 18, Chaiya, 19, Taxila, 20, Khao Mogul. 21, Ban Plai Nam, 22, Bhita, 23, Likiang, 24, Khao Sam Kao, 25, Kuala Selinsing, 26, Shi Zhai Shan & Lijiashen, 27. Sa-Huynh, 28, Xuan An, 29, Dongson, 30, Tham Ongbah, 31, Chombung, 32. Adichanallur, 33, Coimbatore, 34, Kok Khon, 35, Sisulpalgarh, 36, Mantai, 37, Tra Kieu. Fig. 38. Location of important archaeological sites in South and Southeast Asia. (after Glover 1996) 239

240 yielded such early dates for the Indian connection, though Glover (1996:136) is of the opinion that etched stone beads from India at other sites in the region may come from 2 nd century B.C. contexts. Of course, towards the beginning of the Christian Era there is greater frequency of occurrence of Indian artefacts, particularly the Rouletted Ware (RW). From the 1 st century A.D. we find widespread deposition of Indian sea borne objects about the Malacca Straits and further east. On the Indonesian coast the RW dated to 1 st century A.D. has been excavated in substantial quantity from the coastal sites of Sembiran (Ardika and Bellwood 1991:221-232; Ardika et al. 1993:101-109) Indian artefacts of early-mid 1 st millennium A.D also been excavated from Beikthano, Chansen, Gilimanuk, Buni Complex in Java, Khlong Thom, Kuala Selinsing, Tra Kieu and Oc-eo (Fig. 20). Interestingly, objects of Mediterranean origin also appear in early centuries A.D. contexts in many coastal sites of South-East Asia. These settlements are UThong (Roman coin), Khlong Thom (Roman intaglios and glass). Pong Tuk (Roman lamp) and Oc-eo (Roman coin) (Chapter II for details of Roman artefacts in Southeast Asia, also see Fig. 3). The 'exotic' deposition briefly summarized above corroborates the textual records for Indian and Roman sea borne trade with Southeast Asia in the transition towards the 1 st millennium A.D. Southeast Asia was the fabled suvarnabhumi of ancient Indian literature (Ray 1991:357-365). References to ancient Indian voyaging to Southeast Asia in search of spices are to be found in the Arthasastra, Jatakas, Raghuvamsa and Milindapanho (Fig. 21). The textual references to contact and trade between India and Southeast Asia are congruous with the chronological range of the Indian and Roman artefactual deposition in the latter region. Broadly, three stages in the growth of India-Southeast Asia contacts has been envisaged. The first stage, represented by Indian finds in the mid-1 st millennium B.C. contexts at BDTP, has been perceived to be essentially non-commercial, brought about more by Buddhist proselytizing missions from eastern India (Ray 1991: 357-359). The second stage, in the 1 st-2 nd century A.D., witnessed according to Glover (1996:131), the integration of *the previously rather separate Southeast Asian exchange systems into a vast network stretching from Western Europe, via the Mediterranean basin, the Persian Gulf and the Red Sea, to India, Southeast Asia and China' The third stage arrives with the fullfledged 'Indianization' of Southeast Asia when Brahmanical and Buddhist doctrines become predominant from mid-1 st millennium A.D. and result in creation of powerful

241 architectural symbolisms such as the Vishnu temple of Angkor Vat and the Buddhist monument of Borobodur In this section, the focus is on the two early stages which created the conditions for large-scale diffusion of Indian culture and polity into Southeast Asia from the mid- 1 st millennium A.D. The repetoire of Indian artefacts and material cultural elements (iron metallurgy, lapidiary techniques, glass crafting) appearing in Late Prehistoric (equivalent to the Indian Early Historic) levels in Southeast Asia have generated an imperitive for provenance studies. The direction of research is strongly oriented towards precising the resource-areas of commodities and technologies reaching Southeast Asia from India and vice versa. The glass beads from BDTP and Khlong Thom have been analysed in terms of morphology, manufacturing technique and chemical make-up. The results indicate that both glass beads as well as glass crafting technology was exported from India, the latter probably taken by expatriate artisans (Basa, Glover and Henderson 1991:351-365; Basa 1993:93-100). Ray (1991:357- 365) traces Indian seals and intaglios found in Southeast Asia to Gangetic origins. In a series of detailed studies Glover (1990; 1996: 129-158, also Glover and Syme 1993:119-127) searches for parallels for stone beads and 'knobbed' vessels from BDTP in consonant Indian contexts. Recently, scientific analysis on RW from Sembiran (Indonesia), Anuradhapura (Sri Lanka), Arikamedu and Karaikadu (India) reveals a common production source for the ware (Ardika and Bellwood 1991:221- 232, Ardika et al. 1993:101-109). The range of Indian artefacts being recovered from coastal Southeast Asia can be traced to resource-areas spread across the entire eastern Indian seaboard as well as the interior of peninsular and northern India (see under Eastern India in Chapter V) However, while the exotic material has been broadly attributed to likely resource-areas in India, it is now necessary to attempt more precise connections. Take the case of semi-precious stone beads excavated at BDTP and associated sites in Southeast Asia Glover (1996:139) reports some 600 stone beads (mostly agate, carnelian, crystal) from the site. Distinctive among the stone beads from BDTP are a group of etched carnelian and agate beads to which Glover attributes a Gangetic origin (Glover 1996:fig.2; pl. IV b). A detailed study of decorative patterns on ancient Indian etched stone beads undertaken by Dikshit (1949) is particularly useful for delimiting likely provenances of the etched beads from BDTP. According to Dikshit (1949:14) 'Repeated occurrence of the same (etched) pattern on beads from more than one site, shows the trend of fashion current in the locality at certain periods... Following from this observation Dikshit demarcates the etched Indian beads in terms

242 of the Northern and Southern Indian groups on the basis of specific motifs (Dikshit 1949 Pl V). In this regard, the beads from BDTP having chevrons within marginal bands (3 out of 4 displayed by Glover 1996 Fig 2) would be attributed a likely south Indian origin provenance in Dikshit's scheme To quote Dikshit (1949:28) 'A chevron within marginal bands is a very common decorative motif in the Southern Group Beads mostly occur in the megalithic burials but the type continues in the Satavahana period also Those from the former are generally barrel-shaped but plain cylinders seem to be preferred at a later stage In this regard it is to be noted that two of the etched beads with chevrons from BDTP are plain cylinders and one barrel-shaped In his Northern Group Dikshit also shows the presence of chevron decoration However here the chevron motif appears on beads with rectangular cross-section (Dikshit 1949 Pl. V; Pl. XX.b). Apart from beads with chevron motif, the rest of the etched beads published by Glover (1996: Pl. IVb ) are common in the Gangetic zone and are well represented in the Northern Group of Dikshit (1949 PLV, nos.6, 9) My study of historical period beads (surface finds) from Kausambi in the Allahabad Museum revealed parallels for most of the designs on etched stone beads published by Glover (see Pl. XX.b.d.e). ་ Besides the etched beads of likely Indian origin, Glover (1996 PLV ) has published a photograph of a lion pendant made of carnelian excavated from BDTP The photograph shows the couchant lion pendant in situ together with spherical and tubular carnelian beads. Glover (1996:140) is certain of a northern Indian origin of the lion pendant. He compares the BDTP pendant to a similar one of crystal recovered from the Dharmarajika stupa at Taxila (Peshawar, Pakistan) A similar couchant lion pendant made out of soapstone has also been recovered from the Early Historic/Historic site of Kausambi in the Middle Ganga Valley (Pl. XX.a, Allahabad Museum Collection, on lion pendants from other sites see Margabandhu 1985:222) Glover (1996:139) reports a number of green glass pendants from BDTP which are cut in the hexagonal shape of beryls. These 'imitation beryls' have been found also at Oc-Eo and Ban Chiang (Glover 1996:139). On the Indian side, the imitation beryls have been recovered from the site of Kausambi, Ahichchatra and Narhan in the Gangetic Valley (Basa, Glover and Henderson 1991:374) and the coastal trading station of Arikamedu on the Tamil coast. Two specimens from Kausambi have been catalogued by Postel (1989) and one piece was observed by me in the Allahabad Museum Collection (Pl. XX.c) From Arikamedu, Francis, Jr (1987) reports a number of imitation beryls and even unfinished pieces broken in course of crafting An unfinished (fragmented) green glass close to colour of beryl was picked up by me from

243 the 'bead factory' area at the site of Arikamedu (Pl. XII.c) It is significant that Arikamedu happens to be the only site yielding evidence of imitation beryl crafting This fact should be seen in the light of Roman techniques of making glass imitations of various stones practised at the site (Francis, Jr. 1987). In this context, the imitation 'beryls' represented, in all likelihood, Mediterranean glass crafting techniques assimilated in South India. The imitation 'beryls' found in North India and Southeast Asia may have been exported from southern manufacturing centres such as Arikamedu The consolidated evidence for semi-precious stone beads and glass beads/pendants of Indian origin in Southeast Asia suggests acquisition from resourceregions in northern and southern India. In the north, the ancient cities of the Ganga Valley must have been active as mercantile centres both producing as well as facilitating flow of export-commodities to the eastern seaboard and beyond In particular, the ancient settlement of Kausambi seems to have been a major supply point, for this site yields a number of Indian artefacts (etched beads, couchant lion pendant, intaglios, seals) recovered in coastal Southeast Asia. The export of glass beads and glass cut in the shape of semi-precious stone seemed to have been primarily initiated from southern India. For instance, the material remains of glass crafting at Arikamedu, BDTP and Khlong Thom show striking similarity small monochrome beads, glass cullet, attempts to recreate agate recreate agate from glass and presence of Mediterranean glassware fragments in the residue are present at all three sites. Parallels for the BDTP 'type' high tin bronze vessels with central protrusion surrounded with concentric circles have been found in bowls discovered in the Nilgiri area (Glover 1996:142). However the Nilgiri vessels are an isolated cache and may be imported from Southeast Asia. The 'knobbed' vessels, in fact, are copiously found as pottery in Early Historic levels of important Kalinga sites: Sisupalgarh, Jaugada and Salihundam At Sisupalgarh and Jaugada these 'knobbed' vessels are recovered from 3 rd century B.C. contexts and therefore are chronologically consonant with the BDTP 'knobbed' vessels (for dates of 'knobbed' vessels at Jaugada/Sisupalgarh see respectively Indian Archaeology - A Review Luders 1956-57:30-31; Lal 1949: 72,89). At both Sisupalgarh and Jaugada the 'knobbed' pottery comprised the dominant wares of the early levels (discussion under Eastern India in Chp. V). The 'knobbed' motif was probably transmitted from Kalinga to Southeast Asia where it was incorporated in high-tin bronze vessels. Some of the latter may have been exported to India, as the Nilgiri cache indicates. (further discussion on Orissa-Southeast Asia connection see under Eastern India in Chapter V). The Indian (and Indo-Roman) connections with Southeast Asia are represented by the increasing incidence of Indian and Roman objects appearing in the latter region.

244 Though there is a virtual absence of early Southeast Asian artefacts in India, recent findings of pottery with impressed designs on the eastern Indian coastal sites of Kottapatnam (Rao 1994 a: 829-831) and Manikpatna ? (Pradhan et al. 1996 120-123) reveals them to be ceramics of Southeast Asian affinity on the Indian coast. According to Rao (1994 a:829-831), the impressed pottery of probable Southeast Asian origin was earlier excavated by Wheeler at Arikamedu who published them as 'Local Wares' of the 'pre-Arretine' period dated by Wheeler et al. (1946: 24) to the beginning of the 1 st century A.D. Systematic dating and precise provenancing of the impressed pottery in India may lead us to delineate sharper patterns of early exchange between India and Southeast Asia. To recapitulate, the archaeological evidence for India-Southeast Asia maritime exchange indicates regular exchanges taking place across the Bay of Bengal in the early centuries of the Christian Era. The Roman involvement in this segment of trade may not have been marginal The concentration of Roman-contact' sites on the Tamil coast and Lower Krishna region cannot be wholly explained in terms of coastal trade on the eastern Indian seaboard. These bases of Mediterranean traders on the east coast must have been established, among other reasons, to take advantage of the traditional India-Southeast Asia exchange mechanism in place since the mid-late 1 st millennium B.C. In particular, the tradition of fine glasscutting to make imitations of precious and semi-precious stones observed as trans-oceanic practice at Arikamedu, BDTP and Khlong Thom integrated Mediterranean techniques and may have even involved expatriate Hellenic beadmakers together with Indian artisans in Southeast Asia.