Langkasuka and Tambralinga: some archaeological notes
H.G. Quaritch Wales published this article in 1974, reporting on trial excavations and artifact examinations at three key sites in southern Peninsular Siam—Yarang, Satingphra, and Nakhon Si Thammarat—conducted during a short stay in 1973. His overarching thesis is that the ancient city of Langkasuka at Yarang was a Dvaravati-parallel civilization of the late sixth to early ninth century rather than a Srivijayan foundation, and that the subsequent urban history of the region involved a sequence of replacements: Langkasuka giving way to Satingphra as a Srivijayan seaport in the eleventh to thirteenth centuries, which in turn was superseded by Nakhon Si Thammarat as the capital of independent Tambralinga from approximately 1230.
Summary
Wales undertook his fieldwork with the specific aim of resolving long-standing problems of chronology and cultural identification at sites whose periods of occupation had been debated or conflated in the literature. At Yarang, the accepted location of Langkasuka’s capital, he declined to dig in the overgrown exterior where previous finds had been made and instead insisted on trenching within the city enclosure. The first trench proved sterile, but the impasse was resolved when the local education officer, Mr. Anan Vathanikorn, showed him a collection of artifacts he was safeguarding at his home, supplemented by objects from the principal wat in Patani town. These included terracotta votive stupas with Dvaravati-generic features, a stone dharmacakra, a standing Buddha in vitarka mudra, a bas-relief of the Buddha beneath the Bo tree flanked by the Bodhisattvas Maitreya and Lokesvara, a polished granite linga, and a stone door sill. Wales judged the assemblage absolutely homogeneous, dating from the late sixth to early ninth century, and of a style paralleling Dvaravati rather than Srivijaya. The single exception—the Mahayana bas-relief—indicated a degree of Srivijayan influence consistent with what was also appearing in Dvaravati at that time. The rounded corners of the ramparts and the complete absence of Srivijayan-style bronze Bodhisattvas (common at Chaiya and Satingphra) corroborated this reading.
At Satingphra, Wales dug three trenches within the moated enclosure, all of which encountered the same single stratum of pottery deposits extending to natural sand at a depth of about three feet. The assemblage comprised Sung-type glazed sherds, fine Lung-chuan celadons, greenish moulded wares possibly of Tonkin origin, white porcelains, and homogeneous local wares, but no blue and white and no Sawankalok. This led him to date the town’s foundation to the eleventh century with flourishing in the twelfth and thirteenth, a century or more earlier than A. Lamb’s previous estimate based on museum collections. He reinterpreted Satingphra not as an outpost of Phathalung but as a Srivijayan seaport on the Inland Sea, possibly the successor to Langkasuka after a transfer of the capital around the beginning of the eleventh century. At Nakhon, he examined the absence of Srivijayan-style sculpture in the new museum, the results of his earlier trenches at the Siva temple (Sung wares to the lowest level), and a new trench at Tha Rua (the remembered old port), which yielded Sung celadons and stamped domestic pottery but nothing earlier. Together with the comparative evidence from Buribhand and Griswold on post-Srivijayan Buddhist sculpture—showing a local “school of Chaiya” preserving older influences but no equivalent at Nakhon—this cumulative evidence led him to conclude that Nakhon was not established as a Srivijayan centre of importance before the twelfth century.
Wales also investigated Old Songkhla at the tip of the Satingphra peninsula, where Stargardt had recently claimed a Srivijayan town. His trenches produced only blue and white sherds mixed with local stamped pottery, no Buddhist or Hindu images, and no Sung celadons, leading him to date the settlement to the fifteenth century at earliest and to reject the Srivijayan identification. He further examined isolated shrine sites at Tha Sala and Na Khom, where seventh- to eighth-century Hindu sculptures (long-robed Visnus, lingas) had been found associated with brick foundations, arguing that the entire fertile region between Chaiya and Nakhon was dotted with Indianized communities by the fifth or sixth century, well before the Srivijayan arrival.
Key Findings
- The Yarang assemblage (terracotta votive stupas, stone dharmacakra 5½ inches high, standing Buddha 2 feet in vitarka mudra, Mahayana bas-relief c. 1 foot, polished granite linga 20 inches, stone door sill 3 ft. 11 in.) is homogeneous, dating from the late sixth to early ninth century, and is of Dvaravati-parallel style rather than Srivijayan (pp. 18–19).
- The Satingphra trenches (5 × 2 yards each) revealed a single pottery stratum to a depth of approximately 3 ft. 6 in. containing Sung celadons, Lung-chuan, white porcelains, and local wares, but no blue and white and no Sawankalok, supporting a foundation in the eleventh century and flourishing in the twelfth and thirteenth (pp. 30–31).
- The Old Songkhla trenches (5 × 2 yards) yielded only blue and white sherds and local stamped pottery to a depth of two feet, with no Buddhist or Hindu images and no Sung celadons, dating the settlement to the fifteenth century at earliest and refuting its identification as a Srivijayan town (p. 28).
- The Tha Rua trench (1 × 3 yards) at Nakhon produced sterile sand to 20 inches, then humus with Sung celadons, white porcelain, and stamped domestic pottery to just under a yard, with nothing suggesting a date earlier than the twelfth century (p. 35).
- The Takuapa triad’s central figure, approximately eight feet high, was recovered almost complete from beneath the ground at the knees, with the original headdress found at its feet; the attendant figures were largely destroyed (p. 34).
- The linga from Yarang, with its pronounced oval swelling, raised gland and frenum contours, and short octagonal section, is dated to the sixth or seventh century on the basis of Malleret’s comparative criteria for relative antiquity (p. 19).
Conclusion
Wales’s definitive historical takeaway is that the urban history of southern Peninsular Siam from the sixth to the thirteenth century was one of sequential replacement rather than continuous occupation at any single site. Langkasuka at Yarang, a Dvaravati-parallel polity emerging from the collapse of Fu-nan, was succeeded by Satingphra as a Srivijayan maritime port, which was in turn superseded by Nakhon Si Thammarat as the capital of independent Tambralinga under Chandrabhanu from approximately 1230. The broader region, however, had been continuously Indianized since at least the fifth or sixth century, with isolated shrines and village communities dotting the landscape independent of any single urban centre.
Context
- The article draws on fieldwork conducted with the Thai Fine Arts Department (Ninth Administrative Region, Songkhla; Eighth Region, Nakhon Si Thammarat), examining collections at the Songkhla and Nakhon museums, private collections, and objects safeguarded by local officials, in addition to the trial trenches described.
- Historiographically, the article represents Wales’s mature synthesis of his long career in Southeast Asian archaeology, integrating his own earlier excavations (Nakhon Siva temple, 1935; Satingphra, 1964) with new fieldwork and responding critically to the work of Wheatley, Stargardt, Lamb, and O’Connor. It advances the argument that the Dvaravati and Srivijayan cultural spheres in the Peninsula were distinct and sequential rather than overlapping, a position that remained influential in subsequent scholarship on the region.