An Indian Trader in Ancient Bali?

Article excerpt

Contacts between India and Bali

A little over a decade ago, Ardika and Bellwood (1991) reported their discovery of the first securely stratified evidence of Indian trade contact with Indonesia estimated to have occurred approximately 2000 years ago. This evidence consisted of 79 pottery sherds of rouletted and closely related ware of Arikamedu type, from excavations at Sembiran, a site on the north coast of Bali. Neutron activation analysis showed that the Sembiran specimens have identical fabrics to samples from Arikamedu (Pondicherry) and other Indian sites (Ardika et al. 1993). Subsequent excavations at Sembiran provided additional evidence for trading contact with India. The total quantity of known Arikamedu types is now at least 120 sherds, concentrated within an area about 130 x 100m. One sherd contained a graffito of three letters in early Indian script, Kharoshthi or Brahmi; another sherd from a large black-slipped storage jar was tempered with rice husks that were dated by AMS radiocarbon to 2660 [+ or -] 100 BP.

These discoveries significantly altered our picture of cultural contact between India and insular Southeast Asia. Before this, the earliest direct evidence for such contact consisted of stone and metal inscriptions dating from the fourth and fifth centuries AD, found in West Java and Kalimantan. However, indirect evidence, summarised by Glover (1990) and Ray (1989), suggested that contact might have begun much earlier. Miller (1969) provided detailed lists of spices and other products of presumed Indonesian origin mentioned in both Indian and Classical sources; cloves and cinnamon were known to Pliny in AD 70. Ardika and Bellwood (1991) assessed the date range for the Sembiran materials (Rouletted Ware, rice husk and Kharoshthi graffito) in order of likelihood: (1) 800 BC to AD 450 (outer possible range), (2) 150 BC to AD 450 (intermediate range), and (3) AD 1-200 (most likely date range in terms of the chronological overlap between use of the Kharaoshthi script and Rouletted Ware).

The site of Sembiran itself was located at the head of a small sheltered bay that no longer exists. Several inscriptions in the Old Balinese and Old Javanese languages were discovered in the vicinity. These inscriptions, written nearly a thousand years later (AD 896-1181), refer to long-distance or seafaring merchants (banyaga; banyaga saking sabrang); a merchant guild (banigrama; Sanskrit vanigrama); a market officer (ser pasar), and other aspects of seaborne trade. Ardika and Bellwood (1991:132, 148, 265) observed that in contemporary East Java the term banigrama is associated with foreign traders, and further that inscription Sembiran C (#621, Old Javanese, 1181 AD) mentions that the term juru kling may be a specific term for Indians or the descendants of Indians. Ardika and Bellwood (1991:230) interpreted these inscriptional finds to indicate that this region of north-eastern Bali was the scene of intense maritime trading activity about 1000 years ago, with archaeological evidence pushing this activity back perhaps a millennium further. At that time, the Sembiran site likely consisted of a settlement located inside a small and shallow bay in the coastline, peopled by native Balinese who were presumably in contact with visiting traders who were able to bring in large amounts of Indian trade pottery sometime between 200 BC and AD 200 (Ardika & Bellwood 1991). However, as concluded by Ardika and Bellwood (1991: 195), "whether these traders were ethnic Indians as opposed to Indonesians we may never know."

A foreign tooth?

Here we report the results of our analysis of a human tooth found in Ardika's original excavation, in the same strata in which the largest amounts of Rouletted Ware sherds were discovered (below spit 3.5 in the Pacung trench PCN III), in association with glass beads also thought to be of Indian origin (Ardika & Bellwood 1991:226). We subjected this tooth to three analyses: AMS radiocarbon dating, stable carbon isotope analysis and analysis of mitochondrial DNA. …