Glass beads in Malaya: a reassessment
Peter Francis Jr. (1991) reassesses the glass bead record of the Malay Peninsula, arguing that Malaya was not a passive recipient of Roman trade goods but rather a major production centre for the ubiquitous Indo-Pacific drawn glass beads, manufactured by Tamil Indian beadmakers using the distinctive Lada process from approximately the 2nd century B.C. until the industry’s disappearance around 1200 A.D.
Summary
The article opens by dismantling the long-standing claim, initiated by Gardner (1937) and amplified by subsequent writers, that Roman glass beads at Johore Lama and Kota Tinggi evidence direct Roman-Malay trade. Francis re-examines the original Beck identifications and shows that of 59 beads, only eight were tentatively assigned to Roman or Mediterranean origin, and even those are problematic—some are Bohemian, one is a steatite bead misidentified as Hittite, and the famous “Roman” chevron bead is now known not to have been produced before c. 1480. The weight of evidence supports Gibson-Hill’s earlier scepticism that the presence of a few early beads does not make the site ancient (pp. 98–99).
The core of the article is the reconstruction of the Indo-Pacific beadmaking industry. Francis identifies the Lada process—a unique tube-drawing technique still practised at Papanaidupet in southern India—as the manufacturing method behind the most common trade bead in the ancient world. By matching diagnostic wasters (gedda paru flakes, twisted tubes, flares, collapsed tubes) between Papanaidupet, Arikamedu, and Southeast Asian sites, he traces the industry’s migration: from Arikamedu (c. 250 B.C.–A.D. 250) to Klong Thom (Thailand) and Oc-eo (Vietnam) in the early centuries, then to Kuala Selinsing and Sating Pra (7th–10th centuries), and finally to Takua Pa and Sungai Mas (9th–13th centuries). All four early sites shared Roman emporia, strong South Indian (Pallava/Tamil) associations, and cross-site exchange of beads and stone beadmaking, leading Francis to propose a loose trading federation—the “Arikamedu League”—that may have provided the germinal idea for Srivijaya (pp. 101–103).
The article also documents the shift in bead trade after c. 1000, when direct Muslim-world imports (segmented beads, fused mosaic cane beads, folded beads) began arriving in large quantities at Kedah and Sungai Mas, and the eventual replacement of Indo-Pacific beads by Chinese coil beads from the 13th century onward. Glass composition analyses confirm that Indo-Pacific beads were made from locally produced glass, not recycled Western material (pp. 106, 109–111).
Key Findings
- Of the 59 beads Gardner (1937) claimed were 20% Roman from the Johore River, Beck’s original 1936 notes identified only eight as Roman or Mediterranean; re-examination shows several are Bohemian, one is steatite (misidentified as Hittite), and the chevron bead type is not produced before c. 1480 (pp. 98–99).
- The Lada process, still practised at Papanaidupet (Andhra Pradesh), involves rolling molten glass on a long iron tube (lada), piercing it with a chetleak rod, and hand-pulling the tube over approximately three hours; its diagnostic wasters match those at Arikamedu exactly, with the sole exception that the Arikamedu master used a pincher-like tool rather than an iron hook (pp. 99–100).
- Glass composition analyses of 21 early Indo-Pacific beads (Arikamedu, Oc-eo, Klong Thom) and 29 medieval beads (India, Thailand, Malaysia, Sarawak, Flores) show distinct differences from contemporary Roman, Egyptian, Islamic, Byzantine, and European glasses, confirming local glass production (p. 106).
- The degree of roundedness (R-scale) of beads at Takua Pa (49% R1, 59.2% R2, 6.1% R3) differs markedly from Mantai (18.3% R1, 54.9% R2, 26.8% R3) and Sungai Mas (18.5% R1, 51.9% R2, 29.6% R3), providing a potential fingerprint for site attribution (p. 107).
- In the Philippines, Indo-Pacific beads constitute 61.5% of Developed Metal Age assemblages but only 1.1% in the period corresponding to the Southern Song (from A.D. 1127), corroborating a terminal date of c. 1200 for the industry in Southeast Asia (p. 104).
- The 1988–89 Fort Canning Hill excavations (Singapore) uncovered a 14th-century Chinese coil bead workshop with a small fire, glass sherds, melted pieces, and finished beads, confirming local manufacture of the beads that replaced Indo-Pacific beads across the region (p. 111).
Conclusion
Francis concludes that Malaya’s significance in the ancient bead trade was as a production centre rather than merely a consumer, and that the Indo-Pacific beadmakers—almost certainly Tamil Indians—developed a complex suite of subsidiary bead types (pinched beads, striped drawn beads, square-sectioned beads, false gold-glass beads) from the basic Lada-drawn tube. The disappearance of the industry around 1200, coinciding with the Chola invasions of Srivijaya and the rerouting of trade through the Strait of Malacca, created a vacuum filled by Chinese coil beads for the next three centuries. The “Arikamedu League” hypothesis, while speculative, opens a productive line of inquiry into the political and economic structures that preceded Srivijaya (pp. 112–113).
Context
- Francis drew on his own fieldwork at Papanaidupet and Arikamedu (cataloguing 50,000 beads and wasters), personal examination of assemblages at Kuala Selinsing, Sungai Mas, Takua Pa, and Sating Pra, and the Fort Canning History Project excavations in Singapore.
- The article represents a major historiographical intervention: it corrects the Gardner-Beck “Roman bead” narrative that had dominated Malayan archaeology since 1937, and reframes Malaya from a peripheral trading post to a central node in a pan-Indian Ocean beadmaking network.