Copper-base metallurgy in Late Iron Age Cambodia: Evidence from Lovea

https://doi.org/10.1016/j.jasrep.2017.04.005Get rights and content

Highlights

  • Fourteen copper-base artefacts were studied from Late Iron Age (c. AD 100-400) Lovea in north-western Cambodia.

  • Elemental and isotopic data for these bracelets, rings and earrings were compared with existing regional databases.

  • Results indicate the presence of bronze, leaded bronze, and possibly high-tin bronze alloys on-site.

  • Unleaded bronzes are consistent with the lead isotope signature of a prehistoric production centre in central Laos.

  • The link to a single copper source, when others were available, requires further investigation.

Abstract

In this paper we present the study of the copper-base assemblage of the Late Iron Age (c. 100–400 BC/AD) moated site of Lovea in north-western Cambodia. Fourteen artefacts, approximately 20% of the metals assemblage, including bracelets, rings and earrings, were analysed for their elemental and lead isotopic composition, and these data compared with existing Southeast Asian archaeometallurgical databases. Results indicate the presence of bronze, leaded bronze, and possibly high-tin bronze alloys on-site. The unleaded-bronzes exhibit a strong consistency with the lead isotope signature of Sepon in central Laos, a major prehistoric copper production centre. We propose that the apparent presence of a sole copper source, when others were available, may be linked to economic and/or cultural factors during a period of increasing social complexity.

Introduction

Whilst a substantial number of moated Iron Age sites have been excavated in Thailand, Lovea, the focus of this paper, is the only one to have been excavated in Cambodia (Fig. 1, O'Reilly and Shewan, 2016, O'Reilly and Shewan, 2015). As Thai examples of these moats are thought to relate to agriculture-related water management and defence against population-growth driven violence, they offer a rich window on socio-cultural changes occurring during the regional Iron Age, c. 500 BCE to c. 500 CE (O'Reilly, 2008). An 8 × 8 m unit was placed in a vacant lot towards the centre of the modern village (O'Reilly and Shewan, 2015: Fig. 2). Three main stratigraphic layers were encountered: recent historic/modern settlement; Angkorian settlement, as presented by pottery and an inscribed lintel fragment; and a prehistoric cemetery (O'Reilly and Shewan, 2015). A series of thirteen radiocarbon dates, of which nine from layer three, constrain the cemetery to the Late Iron Age, c. 100–400 CE (O'Reilly and Shewan, 2015: Table 1).

The excavated unit exposed twelve Late Iron Age mortuary contexts (fourteen individuals), at depths ranging from 0.9 m to 1.4 m. Six of the fourteen were oriented South and two Southeast, the remaining seven being undeterminable due to poor preservation. Likewise, seven of the burials were sexed as male and the rest cannot be identified. Ten of the burials were adult, three were elderly and there was one young adult but no subadults and no children. All burials had a selection of grave goods from eight main classes: pottery, glass and stone beads, iron implements, marble bangles, spindle whorls, clay pellets, and copper-base artefacts (Table 1).

The glass1 and the stone beads can certainly be thought of representing longer distance exchange systems (O'Reilly and Shewan, 2016), either materials and/or technologies from the Indian sub-continent or reproductions at regional Southeast Asian industrial centres, e.g. Khao Sam Kaeo on the Thai-Malay Peninsula (Bellina et al., 2014, Bellina, 2014). The other material classes (e.g. pottery and iron) could, potentially, be acquired more locally but there is as yet no scientific proof for their provenance.

Focussing on the copper-base assemblage from Lovea, a total of 75 artefacts or fragments thereof were found, averaging 5.4 per burial but with a large standard deviation of 3.4. Bracelets and fragments thereof form the largest group (47), followed by finger/toe rings (21), and earrings (6), along with a single coin. This latter has been stylistically identified as a ban liang of the Wang Mang interregnum period (7–23 CE), which separated the Western (206 BCE–9 CE) and Eastern Han (25–220 CE) dynasties (O'Reilly and Shewan, 2015, Loewe, 2000: 536–545). It may be surmised based upon the presence of copper-base grave goods in eleven of the fourteen burials that copper alloys were readily available to the local population but there was significant variation between individuals.

Archaeometallurgical research over the last decade has demonstrated that although Mainland Southeast Asian primary copper production was initiated very soon (perhaps as little as 50 years) after the appearance of copper-base artefacts in regional contexts c. 1000 BCE , the intensity of that production increased dramatically from the early Iron Age (Pryce et al., 2014). This tendency is production behaviour is particularly well evidenced in the Khao Wong Prachan Valley of central Thailand (Pryce et al., 2010) but it also mirrored by the evidence for mid/late 1st millennium BCE mining and smelting at Phu Lon in northeast Thailand (Pigott and Weisgerber, 1998, Pryce et al., 2011) and at the Sepon complex in central Laos (Pryce et al., 2011, Tucci et al., 2014). Regional copper production, supply, was naturally following regional consumption, demand, which at the juncture of Mainland Southeast Asian Iron Age can be seen in the much increased quantity of metal artefacts used in funerary rituals, especially bangles. This increasingly common and occasionally ostentatious consumption of copper/bronze has been suggested to relate to growing concern with social competition and display (Higham & Higham, 2009); of which only the funerary behaviour is well-preserved by the archaeological record. With its chronology lying relatively late in the regional Iron Age sequence, the remainder of this paper concerns the elemental and isotopic analysis of fourteen of the Lovea copper-base artefacts and the comparison of those data with the available prehistoric Cambodian dataset. This exercise allows us to make preliminary interpretations of copper-base metal exchange networks during the final centuries prior to state formation in northwest Cambodia.

Our funding allowed for the analysis of fourteen of the seventy five available samples, or c. 20%; a proportion which compares favourably with many of the assemblages studied by the Southeast Asian Lead Isotope Project (Pryce et al., 2014). A selection was made on the basis of representing the different burials and morphologies, as well as expected corrosion levels (Fig. 3). Given the tropical climate, Southeast Asian copper-base assemblages frequently suffer from a degree of corrosion but cut samples are infinitely preferable as matrices are often only partially corroded and extant metal is often available for more secure elemental and isotopic determinations. By way of contrast, surface analyses can only ever be compromised. Samples were taken in Cambodia using a 0.2 mm jeweller's saw blade after photographic recording by the excavators. The cut samples were sent to the Curt-Engelhorn Centre for Archaeometry (CEZA) in Mannheim (Germany) for elemental analysis using energy-dispersive X-ray Fluorescence Spectrometry (XRF) and lead isotope ratios with Multi-Collector Inductively-Coupled Plasma Mass Spectrometry (MC-ICP-MS) using the established instruments and protocols of the CEZA laboratories (Niederschlag et al., 2003, Lutz and Pernicka, 1996).

Section snippets

Results

The laboratory analyses confirm field observations regarding the degree of corrosion, which can be seen in Table 2′s ‘corrosion’ column divided into ‘metal’, ‘partly corroded’ and ‘corroded’. This is corroborated by the elemental analyses, with depletion of copper content and, generally, enrichment of tin, as well as the presence of iron, which is immiscible in uncorroded copper/bronze. Those samples containing some uncorroded metal indicate elemental compositions ranging from 72 to 96 wt% Cu,

Discussion

The identification of Sepon as a potential copper source for most of Lovea's metal consumption assemblage is entirely in line with recent regional lead isotope studies that show central Laos to have supplied a huge area of continental Southeast Asia from the turn of the 2nd/1st millennium BC (Pryce et al., 2014, Pryce, n.d), including assemblages as far afield as southern Thailand and central Myanmar (Pryce et al., n.d, Pryce and Bellina, n.d). Lovea is, of course, much closer to Sepon than

Conclusion

This study has contributed to the growing body of late prehistoric metallurgical data for Southeast Asia in general and Cambodia in particular. The tight range of Lovea's lead isotope values are comparable to those of contemporary Cambodian sites, in that inland sites tend to have less variability than those nearer the sea and potentially more exposed to exotic imports. Lovea data also indicate a strong consistency with a major central Lao copper production area, Sepon. The absence of any

Acknowledgements

The excavation and study of Lovea was financed by a Discovery Project grant from the Australian Research Council (DP110101997). We would like to gratefully acknowledge the APSARA Authority, Robert Christie Research Centre, the participation and understanding of the people of Lovea, and our many colleagues, students and volunteers in the field.

The Southeast Asian Lead Isotope Project was initiated, and this study performed, during Pryce's Leverhulme Trust Early Career Fellowship at the

References (26)

  • M. Loewe

    A Biographical Dictionary of the Qin, Former Han and Xin Periods (221 BC–AD 24)

    (2000)
  • J. Lutz et al.

    Energy dispersive X-ray fluorescence analysis of ancient copper alloys: empirical values for precision and accuracy

    Archaeometry

    (1996)
  • S.A. Murphy et al.

    Introduction: Transitions from late prehistory to early historic periods in mainland Southeast Asia, c. early to mid-first millennium CE

    Southeast Asian Stud.

    (2016)
  • View full text